the biology and control of arrowhead - riverina weeds · i the biology and control of arrowhead...
TRANSCRIPT
i
The Biology and Control of Arrowhead
March 2004
Aquatic Plant Services Goulburn-Murray Water
ii
Contents
List of Figures ........................................................................................................................................vi
List of Tables..........................................................................................................................................xi
1. Executive Summary .....................................................................................................................xii
2. Introduction - Current Knowledge on Arrowhead .......................................................................1
2.1 THE BACKGROUND OF THE ARROWHEAD PROBLEM IN VICTORIA ...........................................1 2.2 TAXONOMY, FORM AND ORIGIN OF ARROWHEAD ...................................................................4
2.2.1 Taxonomy...............................................................................................................................4 2.2.2 Form of arrowhead................................................................................................................5 2.2.3 Origins and spread of arrowhead..........................................................................................9
2.3 ALISMATACEAE SPECIES IN AUSTRALIA ...............................................................................10 2.4 CONTROL OF AQUATIC WEEDS..............................................................................................11 2.5 THE AQUATIC ECOSYSTEM ...................................................................................................19
2.5.1 General principles ...............................................................................................................19 2.5.2 The environment for arrowhead in Goulburn-Murray Water .............................................20 i) irrigation channels ....................................................................................................................20 ii) drains........................................................................................................................................21 iii) natural waterways ...................................................................................................................22 2.5.3 Case study – arrowhead in the River Murray......................................................................24
2.6 THE BIOLOGY OF AQUATIC PLANTS, WITH PARTICULAR REFERENCE TO ARROWHEAD...........26 2.6.1 Forms resulting from phenotypic plasticity in arrowhead...................................................26 i) rosette form................................................................................................................................27 ii) emergent forms – broad leafed.................................................................................................28 iii) emergent forms – narrow leafed .............................................................................................29 iv) seed production, dispersal and survival ..................................................................................33 v) other methods of reproduction..................................................................................................36
2.7 THE CONTROL OF ARROWHEAD – THE GOULBURN-MURRAY WATER PERSPECTIVE .............41 2.7.1 Current knowledge...............................................................................................................41 2.7.2 Mechanical control ..............................................................................................................42 2.7.3 Natural waterways...............................................................................................................44 2.7.4 Eradication versus control...................................................................................................45
2.8 INTEGRATED WEED MANAGEMENT (IWM) .........................................................................46 i) herbicides ..................................................................................................................................46 ii) mechanical control ...................................................................................................................46 iii) biological control ....................................................................................................................47 iv) control based on plant biology and ecology............................................................................49
iii
2.9 LIST OF QUESTIONS/HYPOTHESES .........................................................................................50 i) herbicide....................................................................................................................................50 ii) biology......................................................................................................................................52
3. Summary of arrowhead research facets from current project ...................................................58
3.1 HIGH PRIORITY WITH CLEAR OPERATIONAL GAINS FROM RESEARCH..................................58 3.2 MEDIUM PRIORITY WITH UNCLEAR OPERATIONAL GAINS.....................................................59 3.3 LOW PRIORITY WITH LITTLE CHANCE OF RESEARCH RESULTING IN OPERATIONAL GAINS .59
4. Details of findings from experimental and survey work.............................................................61
4.1 FACETS OF HIGH PRIORITY WITH CLEAR OPERATIONAL GAINS ..............................................61 4.1.1 Glyphosate......................................................................................................................61 4.1.2 2,4-D concentrations in water ........................................................................................65 4.1.3 Casoron G ......................................................................................................................69 4.1.4 Water management – Physiological response of arrowhead to water depth. ................71 4.1.5 Channel design ...............................................................................................................74
4.2 FACETS OF MEDIUM PRIORITY - OPERATIONAL GAINS CURRENTLY UNCLEAR .......................76 4.2.1 Amitrole T.......................................................................................................................76 4.2.2 Seed germination ............................................................................................................77 4.2.3 Seed dispersal and establishment ...................................................................................79
4.3 FACETS OF LOW PRIORITY WITH LITTLE CHANCE OF OPERATIONAL GAIN..............................83 4.3.1 Channel profile – aspect.................................................................................................83 4.3.2 2,4-D...............................................................................................................................83 4.3.3 Seedlings – establishment and development...................................................................86 4.3.4 Control of seedlings with herbicides ..............................................................................87 4.3.5 Corms – development, production and propagation ......................................................89 4.3.6 Corm control using herbicides .......................................................................................90 4.3.7 Rhizomes – production and movement. ..........................................................................92 4.3.8 Other aspects of propagation .........................................................................................94 4.3.9 Forms of arrowhead – broad-leaf and narrow-leaf .......................................................95
5. Implications of results for future research and project direction ..............................................99
5.1 FACETS OF HIGH PRIORITY WITH CLEAR OPERATIONAL GAINS FROM RESEARCH ...................99 5.2 FACETS OF MEDIUM PRIORITY WITH UNCLEAR OPERATIONAL GAINS ....................................99 5.3 FACETS OF LOW PRIORITY WITH LITTLE CHANCE OF FURTHER RESEARCH RESULTING IN
OPERATIONAL GAINS ........................................................................................................................100 5.4 SUMMARY OF FUTURE RESEARCH AND PROJECT DIRECTION ...............................................101
6. Arrowhead management plan based on findings from current research ................................102
6.1 ASPECTS CONTRIBUTING TO BROAD MANAGEMENT PLAN...................................................103 6.1.1 Control to reduce movement into system......................................................................103
iv
6.1.2 Preventative measures..................................................................................................104 6.1.3 Predictive measures......................................................................................................105 6.1.4 Load management ........................................................................................................106 6.1.5 Reactive management ...................................................................................................106
6.2 DRAFT MANAGEMENT PLAN ...............................................................................................108
References............................................................................................................................................113
Appendix 1 - Herbicide Experiments ..................................................................................................118
EXPERIMENT 01 – GLYPHOSATE CONCENTRATIONS.........................................................................118 EXPERIMENT 02 – 2,4-D FORMULATIONS AND TIMINGS....................................................................121 EXPERIMENT 03 – 2,4-D FORMULATIONS, GLYPHOSATE & CONCENTRATIONS & TIMING .................124 EXPERIMENT 04 – GLYPHOSATE FORMULATIONS AND CONCENTRATIONS ........................................126 EXPERIMENT 05 – 2,4-DS, GLYPHOSATE & TIME OF DAY..................................................................128 EXPERIMENT 06 – GLYPHOSATE CONCENTRATIONS & TIME OF YEAR ...............................................130 EXPERIMENT 07 – AMICIDE625, GLYPHOSATE & REPEATS WITHIN A SEASON ..................................133 EXPERIMENT 08 – AMICIDE625, GLYPHOSATE & REPEATS WITHIN A SEASON ..................................137 EXPERIMENT 09 – RESIDUAL HERBICIDES FOR ARROWHEAD CONTROL.............................................140 EXPERIMENT 10 – A COMPARISON OF THE EFFECT OF AMITROLE VERSUS AMITROLE AND
GLYPHOSATE WITH AND WITHOUT FOLLOW-UP ................................................................................144 EXPERIMENT 11 – A COMPARISON OF THE EFFECT OF AMITROLE VERSUS AMITROLE AND
GLYPHOSATE WITH AND WITHOUT FOLLOW-UP ................................................................................146 EXPERIMENT 12 - COMPARISON OF EFFECTS OF FOUR HERBICIDES ON ARROWHEAD CONTROL ........148 EXPERIMENT 13 – THE EFFECT OF DIFFERENT HERBICIDES, RATES AND AMBIENT TEMPERATURE AT
TIME OF APPLICATION .......................................................................................................................150 EXPERIMENT 14 – THE EFFECT OF GLYPHOSATE FORMULATIONS, CONCENTRATIONS AND TIMINGS .152 EXPERIMENT 15 – THE EFFECT OF AN ADJUVANT ON EFFECTIVENESS OF AN AMICIDE/GLYPHOSATE
MIX...................................................................................................................................................154 EXPERIMENT 16 – GLYPHOSATE CONCENTRATIONS.........................................................................156 EXPERIMENT 17 – GLYPHOSATE AND 2,4-D EFFICACY ON ARROWHEAD PLANTS ON CHANNEL BERM
.........................................................................................................................................................158 EXPERIMENT 18 – INFORMAL INVESTIGATION OF CASORON G APPLIED TO ARROWHEAD .................160 EXPERIMENT 19 – INFORMAL INVESTIGATION OF CASORON G APPLIED TO ARROWHEAD .................161 EXPERIMENT 20 – THE EFFECT OF CHANNEL WATER HEIGHT ON THE EFFICACY OF HERBICIDE ON
ARROWHEAD ....................................................................................................................................163 EXPERIMENT 21 – THE EFFECT OF WATER HEIGHT ON ARROWHEAD CONTROL .................................165 EXPERIMENT 22 – CONTROL OF ARROWHEAD SEEDLINGS ...............................................................168 EXPERIMENT 23 – CASORON G EFFICACY ON ARROWHEAD AND RIBBONWEED IN EXCAVATED VS
UNEXCAVATED CHANNELS ...............................................................................................................170 EXPERIMENT 24 – CONTROL OF ARROWHEAD SEEDLINGS ...............................................................171
v
Appendix 2 - Biology Trials ................................................................................................................173
TEMPERATURE INVESTIGATIONS.......................................................................................................173 EFFECT OF WATER DEPTH ON ARROWHEAD ROSETTE BEHAVIOUR ....................................................175 EFFECT OF WATER DEPTH ON ARROWHEAD CORM RESPROUTING......................................................180 EFFECT OF TEMPERATURE AND WATER SOURCE ON ARROWHEAD SEED GERMINATION .....................184 EFFECT OF DARK ON SEED GERMINATION AND SURVIVAL OF SEEDLINGS ..........................................187 EFFECT OF MANUAL CUTTING OF ARROWHEAD SURVIVAL ................................................................190 EFFECT OF MANUAL CUTTING OF ROSETTE LEAVES ON FORMATION OF UPRIGHT STEMS ...................191 EFFECT OF WATER DEPTH ON EMERGENT LEAF FORM .......................................................................192 EFFECT OF 2,4-D CONCENTRATIONS IN WATER ON ARROWHEAD ROSETTE CONTROL........................193
Appendix 3 - Cross-section Surveys ....................................................................................................195
vi
List of Figures
Figure 2.1 Map of Goulburn-Murray Water Irrigation Areas ..................................................................2
Figure 2.2 the erect, ovate leaves of arrowhead .......................................................................................5
Figure 2.3 Narrow-leafed form of arrowhead growing in a channel near Mooroopna. Submersed form
just visible under surrounding water...............................................................................................6
Figure 2.4 The three different forms of arrowhead, clockwise from top, left: broad-leafed emergent:
narrow-leafed emergent; rosette (submersed form); rosette in situ.................................................7
Figure 2.5 the white, three-petalled flowers of arrowhead.......................................................................8
Figure 2.6 Seed of arrowhead floating on the surface of water in an irrigation channel north of
Numurkah .......................................................................................................................................9
Figure 2.7 Rosette form of arrowhead, completely submersed ..............................................................12
Figure 2.8 broad leafed arrowhead growing in Goulburn-Murray Water’s Ardmona Drain II .............22
Figure 2.9 healthy arrowhead growing in sheltered natural waterway of Broken Creek .......................23
Figure 2.10 arrowhead infestation in wetland off Ovens River .............................................................24
Figure 2.11 Small, thread like seedling of arrowhead, growing in channel sediment............................26
Figure 2.12 Rosette form of arrowhead, showing rhizomes produced from single plant, leading to new
stems .............................................................................................................................................28
Figure 2.13 Boat-like appearance of a stand of arrowhead, produced by the presence of broad-leafed
plants at the extremities ................................................................................................................30
Figure 2.14 Re-growth of narrow-leafed form of arrowhead following application of 2,4-D herbicide31
Figure 2.15 Depletion of rhizome resources following application of 2,4-D herbicide.........................32
Figure 2.16 The seeds (achenes) of arrowhead, released from the seed capsule, may spill on the
ground, as here, or on the water surface .......................................................................................33
Figure 2.17 Cross section of a drained channel, showing the spread of an arrowhead population via the
emergence of new plants from rhizomes sent out by mature plants .............................................37
Figure 2.18 Rhizome formation at the base of a young arrowhead plant, with new white rhizomes
emerging to the right and older, soil-stained rhizome coming in from the left.............................39
vii
Figure 2.19 Corm formed from the incoming rhizome of a mature arrowhead plant gives rise to new
arrowhead plant ............................................................................................................................40
Figure 2.20 juvenile plants grown over a period of two weeks, following planting of dried propagules
in soil under 10 cm of water. L to R: from large corm; from small corm; from seed (coin is 5¢
piece).............................................................................................................................................40
Figure 2.21 Stump of arrowhead stem, attached to underground biomass, left following abscission of
main part of stem after 2,4-D treatment. .......................................................................................42
Figure 2.22 Excavation of arrowhead in a channel, showing dislodged stems that may float
downstream and establish .............................................................................................................44
Figure 2.23 Healthy growth of arrowhead encouraged by shading under bridge over Goulburn-Murray
Water Drain 13 north of Numurkah..............................................................................................48
Figure 2.24a diagrammatic representation of leaf modification by glyphosate application...................51
Figure 2.24b diagrammatic representation of the effect of elevated rates of glyphosate on arrowhead
control and re-growth ...................................................................................................................51
Figure 2.25 possible effect of depth on germination of arrowhead seed (red dots represent seeds) ......52
Figure 2.26 possible effect of depth on form of arrowhead – submersed vs. emergent .........................53
Figure 2.27a possible formation of emergent form with lowering of water level ..................................54
Figure 2.27b possible formation of rosette form with raising water level..............................................55
Figure 2.28a active growth of rhizome towards shallower water...........................................................56
Figure 2.28b active growth of rhizome towards deeper water ...............................................................57
Figure 4.1 percentage arrowhead cover in plots sprayed in April 2002 and 2003 with varying rates of
glyphosate (Experiment 01, McCracken Rd, Shepparton) (Average +/- Standard Error) ............61
Figure 4.2 percentage arrowhead cover in plots sprayed in April 2003 with varying rates of glyphosate
(Experiment 16, Drain 13, Numurkah) (Average +/- Standard Error)..........................................62
Figure 4.3 percentage arrowhead cover in plots with varying rates of glyphosate at various times of
year (Experiment 06, Main No. 6 channel, north of Numurkah) (Average +/- Standard Error) ..63
Figure 4.4a Arrowhead treated with glyphosate at 36 L/ha, with water level kept at delivery level
(Experiment 21, Fuzzard’s Rd, near Waaia, Vic.) ........................................................................64
viii
Figure 4.4b Arrowhead treated with glyphosate at 36 L/ha, with water level lowered to about 15 cm
depth. ............................................................................................................................................64
Figure 4.5 Arrowhead rosette leaf-bases, showing elongation of aerenchyma cells (top), which causes
weakness and eventual abscission of the leaf, compared with healthy leaf (bottom). Elongation
caused by application of 2,4-D. ....................................................................................................66
Figure 4.6 The effect of increased concentrations of 2,4-D in water on arrowhead control (initial bin
trial)...............................................................................................................................................66
Figure 4.7 The effect of several herbicides on arrowhead cover, measured at various intervals
following herbicide application in June 2002 (Kerang) ...............................................................69
Figure 4.8a Experimental plot treated with Casoron G, causing suppression of arrowhead emergence70
Figure 4.8b Untreated control plot, showing unaffected arrowhead emergence....................................70
Figure 4.9 Channel cross section – Mulwala main channel – showing presence of emergent plants
(squares) and rosette plants (shaded circles) around the depth cut-off of 50 cm..........................74
Figure 4.10 Effect of Amitrole T treatments on arrowhead cover over time (Drain 13, north of
Numurkah) ....................................................................................................................................76
Figure 4.11a The effect of temperature on the germination of arrowhead seed in water in a controlled-
temperature environment ..............................................................................................................77
Figure 4.11b Arrowhead seedlings floating in a vial of water, having germinated under controlled
conditions......................................................................................................................................78
Figure 4.12 Cross-section of berm of Yarrawonga Main Channel, showing green “carpet” of
arrowhead seedlings growing in mid-August 2003. .....................................................................78
Figure 4.13 Small arrowhead seedlings growing in exposed, saturated soil. Seedlings around 1-3 cm
tall. ................................................................................................................................................79
Figure 4.14 Number of seeds left floating over time, after 50 seeds were dropped onto the water
surface in troughs..........................................................................................................................80
Figure 4.15a Arrowhead growing around an inlet in the River Murray, where slower-moving water
has allowed seed deposition..........................................................................................................81
Figure 4.15b Arrowhead rosettes (bottom right) growing on a newly-exposed sandbar in the River
Murray ..........................................................................................................................................82
ix
Figure 4.16 Arrowhead “stump”, attached to healthy root system, resulting from the removal of top
growth with 2,4-D.........................................................................................................................84
Figure 4.17 Percentage cover of arrowhead following application of several treatments with different
2,4-D formulations, over time. Shows re-infestation following removal of top growth by
herbicide application the previous December (2,4-D formulations are AF300, Amicide 625 and
Surpass 300)..................................................................................................................................84
Figure 4.18 The effect of multiple applications of 2,4-D and glyphosate at label rates on arrowhead
cover in trial plots (Experiment 07 north of Numurkah) ..............................................................85
Figure 4.19 The effect of multiple applications of 2,4-D (10 L/ha of Amicide 625) and glyphosate (9
L/ha of glyphosate 360) on arrowhead cover in trial plots (Experiment 08 west of Katamatite) .86
Figure 4.20 Healthy untreated arrowhead seedlings, Yarrawonga Main Channel berm.......................87
Figure 4.21 Mortality of arrowhead seedlings treated with glyphosate at 4.5 L/ha. .............................88
Figure 4.22 Mortality of arrowhead seedlings treated with Casoron G (230 kg/ha). ............................88
Figure 4.23 Plots clear of arrowhead following application of Casoron G at 23 kg/ha. Some arrowhead
plants in adjacent untreated area can be seen at the top of the photo............................................89
Figure 4.24 Reduction in corm biomass associated with an increase in rate of glyphosate application.
......................................................................................................................................................91
Figure 4.25 Arrowhead rosette plant, showing 5 rhizomes that have formed and are ready to produce
further plants. ................................................................................................................................92
Figure 4.26 Cross section of berm on Yarrawonga main channel, showing positions across gradient of
seedling rosette plants (pink circles), rosette plants arising from rhizomes (green circles) and
erect plants arising from rhizomes (green square). Green line represents elevation gradient,
brown line is depth through sediment to clay base. ......................................................................93
Figure 4.27 Excavation of arrowhead in channel near Corop, showing rhizomes of arrowhead running
shallowly under the surface of the sediment, visible to the right of the photograph. ...................94
Figure 4.28 Excavation of arrowhead near Cobram with water in channel at a high level, showing
plant material floating away from site of excavation – may contain corms, rhizomes and other
propagules.....................................................................................................................................95
Figure 4.29 Broad-leafed arrowhead growing in the Broken Creek, near Numurkah. .........................96
Figure 4.30 Broad-leafed arrowhead growing in a drain near Ardmona...............................................97
x
Figure 4.31 Narrow-leafed arrowhead in a channel, Shepparton North................................................98
Figure 6.1 Steps in approaching arrowhead management in an irrigation system ...............................103
Figure 6.2 Channel cross section – Mulwala main channel – showing presence of emergent plants
(squares) and rosette plants (shaded circles) around the depth cut-off of 50 cm in existing
channel profile (black line) and a theoretical channel profile (red line) that would reduce the
width of the zone in which emergent plants could grow. ...........................................................105
Observing an arrowhead infestation in the days before the species became a major problem to
Goulburn-Murray Water .............................................................................................................207
xi
List of Tables
Table 2.1 – levels of infestation of arrowhead in Goulburn-Murray Water Irrigation Areas and natural
systems............................................................................................................................................3
Table 2.2 – Alismataceae species present in Australia...........................................................................10
Table 2.3 Physical and toxicological properties of selected herbicides used for control of aquatic weeds
......................................................................................................................................................17
Table 6.1 – Management options in natural waterways .......................................................................108
Table 6.2 – Management options in irrigation channels ......................................................................109
Table 6.3 – Management options in irrigation drains...........................................................................111
xii
1. Executive Summary
Having first been reported in the northern Victorian irrigation areas in 1962, arrowhead
(Sagittaria graminea Michx.) has since gone on to become the most troublesome aquatic
weed in irrigation infrastructure and some natural waterways in the Murray Valley and
Shepparton Irrigation Areas, and in parts of southern New South Wales.
The research project was established in response to the variability of success with arrowhead
management and concern about the spread of arrowhead into previously un-infested irrigation
systems and natural waterways, in particular its proliferation in the River Murray. A lack of
understanding of the plant’s responses to and interaction with the environment and control
methods was identified as one of the key shortcomings of the established management
program. In response to this, the following objectives for the project were established:
• To obtain a greater knowledge and understanding of the biology and ecology of
arrowhead, its propagation and dispersal.
• To investigate and develop management and control strategies for aquatic environments
where arrowhead exists.
Investigations were conducted into aspects of arrowhead biology and control, in the following
categories:
• Herbicide Experiments – investigating the effects of differing application rates,
formulations, timings and techniques for use of existing herbicides, with the aim of
developing practices to improve the efficacy of those herbicides. The efficacy of
alternative herbicides not currently used for arrowhead control was also investigated
• Biology and ecology trials – in the field and in controlled conditions, investigating
aspects such as the effect of environmental variables on germination, re-establishment
from corms and growth
• Surveys – to investigate the effect of water management and channel structure on
arrowhead establishment and growth
The results of the trials allow the formulation of a draft management plan for arrowhead
which broadly covers measures to take in situations ranging from an environment that
arrowhead is yet to infest to one where arrowhead is established, dense and prolific. There are
five levels to managing arrowhead in the range that covers these extremes:
xiii
• Control to reduce the number of reproductive propagules entering the system – in
systems where arrowhead is not established, measures should be taken to ensure it does
not enter the system. For example, the plant should be removed from areas that feed the
system.
• Preventative measures – taking steps to establish an environment that is not conducive to
arrowhead establishment and perpetuation, if it does enter the system.
• Predictive measures – if unable to prevent arrowhead from entering the system,
predicting where and how strongly arrowhead may first occur, with reference to amount
of propagule input and water use patterns.
• Load management – establishing a management program that uses the knowledge gained
from research, to manage an established arrowhead problem and reduce its impact and
spread.
• Reactive management – measures taken to restore irrigation capacity when an arrowhead
population has established and is having a negative impact on operations.
Within that framework of the extent of the arrowhead problem in a system, the following
draft management plan has been proposed for arrowhead, with reference to the results of trials
and surveys obtained during the course of the current project:
1. Management options in natural waterways
• Control of arrowhead emergence with the herbicide dichlobenil (Casoron G).
• Control of existing arrowhead growth with glyphosate, in accordance with the APVMA
permits (http://permits.apvma.gov.au/PER6875.PDF).
• Manual removal of small infestations.
2. Management options in irrigation channels
• Prevention of infiltration into the system.
• Re-design channel cross-sections to minimise growth of obstructive forms of arrowhead.
• Minimise sediment build-up in channels through excavation.
• Predict where arrowhead may grow, with reference to arrowhead response to the
environment, and water use and channel structure variables.
• Control emergence of arrowhead using dichlobenil.
xiv
• Control small seedlings that emerge prior to the irrigation season with low doses of
glyphosate or dichlobenil.
• Control of arrowhead plants in “focus months” of March – June. One option for control in
these months may be the use of glyphosate at 40L/ha, in accordance with APVMA
permits (http://permits.apvma.gov.au/PER6999.PDF).
• Control outside focus months, to restore capacity, with 2,4-D in accordance with
APVMA permits (http://permits.apvma.gov.au/PER6341.PDF).
• Removal using excavation.
3. Management options in irrigation drains
• Prevention of infiltration into the system.
• Predict where arrowhead may grow, with reference to arrowhead response to the
environment, and drain structure and use variables.
• Control of standing arrowhead with glyphosate or with a tank-mix of Amitrole T and
glyphosate.
• Removal using excavation.
A draft control and management program has been formulated, using the practices and
principles described here and elicited by the current research.
Future research into the management of arrowhead will focus on expanding on areas of the
current research where more information is needed to take full benefit of the variables
governing arrowhead management, particularly those areas where the greatest operational
gain can be made from further research. The emphasis will also be on establishing a
management program for arrowhead on a large scale, based on the information now available
and monitoring the success and failings of such a program, in order to advance it.
1
2. Introduction - Current Knowledge on Arrowhead
2.1 The background of the arrowhead problem in Victoria
Arrowhead (Sagittaria graminea Michx.) was first reported in the northern Victorian
irrigation areas in 1962, and had probably been present for some time before then.
The first reported populations in Victoria occurred in a drain at Katandra West and in
the Nine Mile Creek at Wunghnu, part of the Broken Creek system (Aston, 1973).
Prior to this its first Australian record was in the Ekibin Creek, near Brisbane, in 1959
(Aston, 1973). Arrowhead has been mapped to three distinct zones along the
Queensland coast (Stephens and Dowling, 2002) and in the Canning River, south of
Perth (Sage et al., 2000).
Arrowhead was not initially treated as a major threat to Victorian irrigation, until the
early 1980s, when the distribution of the plant increased rapidly. The reasons for this
dramatic increase are unclear, though one theory may be that the number of
propagules produced constantly since the 1960s by smaller populations reached a
critical level by the 1980s, that allowed the plant to spread beyond established
populations. This is in accordance with established principles of aquatic weed
infestation (Arthington and Mitchell, 1986) that invasion by aquatic species is
followed by a period of establishment before dispersal.
Arrowhead now infests drains and channels across all of Goulburn-Murray Water’s
Irrigation Areas (see Figure 2.1 for a map of the areas), and natural systems. These
include the Goulburn River, Broken Creek and associated Nine-Mile and Boosey
Creeks, the Ovens River, particularly at its confluence with the River Murray and the
River Murray itself. By the end of 2002 it was the most widespread emergent aquatic
plant on the River Murray between Echuca and Torrumbarry Weir.
Annual expenditure on the control of arrowhead by Goulburn-Murray Water alone is
estimated at $250 000, depending on seasonal variables that govern the growth of the
weed.
The weed is managed by Goulburn-Murray Water because it blocks channels and
drains, causing increased water levels that lead to inefficiencies in delivery and
damage to infrastructure. It may also cause flooding in drains where water flows are
2
retarded during rain and periods of high drain flow (Gunasekera and Krake, 2001). It
also has a negative impact on native species of both flora and fauna and on the
integrity of natural waterways, such as the waterways of the Barmah-Millewa Forest,
where it has also been recorded.
Arrowhead is considered to be the greatest weed threat to efficient operation and
management of Goulburn-Murray Water’s open, earthen channel supply systems
(Gunasekera and Krake, 2001).
Estimates of the level of infestations throughout the Goulburn-Murray Water
Irrigation Areas and natural systems are set out in Table 2.1. Representatives from
each area supplied these estimates in late 2002, and their format therefore varies.
Figure 2.1 Map of Goulburn-Murray Water Irrigation Areas
3
Table 2.1 – levels of infestation of arrowhead in Goulburn-Murray Water
Irrigation Areas and natural systems
Area Description of infestation level
Central Goulburn Irrigation Area (IA) 2 major outbreaks:
• Ardmona Drain 11 - around 400 metres in length at its very top end
• Channel CG 2/9/3 with around 600 metres of it south of Pogue Rd Toolamba
• Another small trouble spot that contained scattered patches was the Mosquito Main drain between McEwen Rd and the Cooma-Kyabram Rd but that seems to be under control at the moment.
Murray Valley IA An estimated 30% of irrigation channels in the area are infested with arrowhead. The area of these infestations is estimated at 48ha, with a further 12ha of known arrowhead growth in the Broken Creek in this area. Infestations in the Shepparton and Murray Valley Irrigation Areas are often very thick, with many irrigation channels covered in dense infestations all the way across.
Pyramid-Boort IA In the Bullock Creek – approx 13 km, patchy
Rochester-Campaspe IA About 1 hectare around Corop
Shepparton IA An estimate of the infestation levels in the Shepparton Irrigation Area was unavailable, though the levels are high, and probably of similar proportions to those in the Murray Valley Irrigation Area (see above)
Torrumbarry IA Gunbower - 8 sites = 50 sq metres Pyramid Creek - 4 sites = 10 sq metres On Farm - 3 sites = 1000 sq metres Pyramid Drain No2 - 1 site = 20 sq metres
Natural systems present in the following Victorian creeks and rivers
• Broken River • Goulburn River • Broken Creek (many hectares) • Boosey Creek • Nine Mile Creek • Ovens River (several hectares in river and
associated wetlands, near confluence with River Murray)
• River Murray (up to 800 separate sites between Corowa and Torrumbarry, ranging from one or two plants to up to a hectare)
4
Murray Irrigation Ltd (NSW – data supplied by Russell Webb, MIL)
• 100% of East Berriquin supply channels (1260 km)
• 40 – 60% of West Berriquin supply channels (from total of 295 km of channels)
• 40 – 65% of Deniboota supply channels (from 500 km)
• 0% (zero) of Wakool supply channels (from 670 km)
• 85% of East and West Berriquin drainage channels (from approx 760 km)
These estimates demonstrate the dramatic spread of arrowhead infestations, from two
small populations in 1962, through large sections of the Goulburn-Murray Water
channel and drainage systems and natural waterways and into New South Wales. It is
now the most troublesome aquatic weed in Goulburn-Murray Water’s irrigation
systems and is a major problem in several natural waterways.
2.2 Taxonomy, form and origin of arrowhead
2.2.1 Taxonomy
Arrowhead is an emergent aquatic macrophyte, known in New South Wales as
“sagittaria”, to distinguish it from another species (Sagittaria montevidensis Cham. et
Schlecht.) that grows in rice bays and is known in that industry as arrowhead, due to
its “broad-arrow” shaped leaves. In this report, however, the common name,
“arrowhead”, will refer to S. graminea.
Arrowhead is a member of the family, Alismataceae, which includes many weeds of
the rice industry, including starfruit (Damasonium minus (R.Br.) Buchenau), water
plantain (Alisma plantago-aquatica L.), alisma (Alisma lanceolatum With.) and S.
montevidensis. The biology of some of these species has been studied as it relates to
the rice industry (Flower, 2003; Graham, 1999; Heylin, in press; Pollock, 1992),
though information on the biology of these species may be applicable in other
systems.
Regarded as one of the most primitive families of monocotyledons (Argue, 1976;
Pandey, 1992), the family Alismataceae is part of the order Alismatales, in which
5
there are three families (Alismataceae, Scheuchzeriaceae and Petrosaviaceae) and
which is part of the division Calyciferae. The Alismataceae comprises 11 genera and
just over 100 species (Turner, 1981), of which around 25 (Sage et al., 2000) belong to
the genus Sagittaria.
Sagittaria is considered amongst the more advanced genera in the family
Alismataceae, with Alisma representing the more primitive types (Brown, 1946).
2.2.2 Form of arrowhead
Aston (1973) describes arrowhead as an erect, emergent, attached perennial with
radical leaves and stolons, corms and/or rhizomes. The emergent leaves are linear to
ovate, with acuminate (tapering gradually to a point) blades (see Figure 2.2), 10-25
cm long and 2-8 cm broad, whilst submersed leaves are strap-shaped without
expanded blades. The leaves and flower stems emerge basally, and the leaves have
long petioles.
Figure 2.2 the erect, ovate leaves of arrowhead
The variation in leaf shape, described above, is one of the important features of
arrowhead in Victoria. It has been described as being very plastic in leaf size and
shape (Sainty and Jacobs, 1981), a trait that is evident in the irrigation systems of
6
south-eastern Australia. In these systems, the leaf shape of some arrowhead
populations ranges from very thin and grass-like to the large, ovate or lanceolate
leaves described by Aston (1973). Evidence from Goulburn-Murray Water systems
indicates that the narrow-leafed form (Figure 2.3) grows as a response to herbicide
application, such that a population of broad-leafed plants, when sprayed, will re-
emerge as the narrow-leafed form. An explanation of this response in an ecological
sense may be that a population, when damaged or broken off by flooding, grazing or
other factors, can respond by producing a narrow-leafed form that does not use as
much resource from the rhizome system to produce as the broad-leafed form. These
narrow leaves can contribute resources to the rhizome through photosynthesis until
the rhizome is healthy enough to contribute to new broad-leafed plants.
Narrow leafed plants also have less drag and therefore are more resistant to damage
by flooding. They would be produced after removal of erect stems by floodwaters.
This action is analogous to the effect of 2,4-D, which acts as a “chemical mower”,
breaking stems off at their base.
Figure 2.3 Narrow-leafed form of arrowhead growing in a channel near
Mooroopna. Submersed form just visible under surrounding water
In addition to these emergent forms, the submersed form of arrowhead has a much
more prominent role than the submersed forms of other species in the family
Alismataceae. In other species, the submersed form is the juvenile plant, but in
7
arrowhead, the submersed form can grow to a large size and exist for a long period
without producing erect, emergent stems. The submersed form produces no flowers or
seeds. The length of time that the plant can spend in this form is unknown but, given
stable conditions, that period may extend over years.
Observations of arrowhead in a field situation reveal that the plant occurs in all three
of these forms in Australia. The forms are illustrated in Figure 2.4. This sort of
phenotypic plasticity is not uncommon in aquatic plants (Haraguchi, 1993) and is
used to survive varying environmental conditions, such as water level, light or
competition.
Figure 2.4 The three different forms of arrowhead, clockwise from top,
left: broad-leafed emergent: narrow-leafed emergent; rosette (submersed
form); rosette in situ
Arrowhead flowers for a large proportion of the irrigation season, from soon after
establishment in August-September to around May or June when the plants begin to
be affected by cold weather. The flowers bear three white petals around a bright
8
yellow centre (see Figure 2.5) and are borne on separate flowering stems, below the
height of the leaves. The flowers mature into seed capsules, to which are attached the
achenes (a dry fruit containing one seed). Each capsule can contain up to 1000
achenes, and each plant may produce up to 20 capsules. After these seeds are shed,
they are able to float for an extended period of time and are often observed floating on
water many kilometres from a mature arrowhead stand (see Figure 2.6)
Figure 2.5 the white, three-petalled flowers of arrowhead
9
Figure 2.6 Seed of arrowhead floating on the surface of water in an
irrigation channel north of Numurkah
2.2.3 Origins and spread of arrowhead
The natural geographic range of arrowhead is believed to be the southern part of
North America, from an eastern limit in the state of Missouri, west to Kansas and
south to Texas and Alabama (Rataj, 1972a). This distribution relates to the variety,
Sagittaria graminea var. platyphylla Engelm. Although Rataj (1972a) suggests that
the arrowhead found in Australia is a different variety (he suggests S. graminea var.
weatherbiana (Fernald) Bogin) to that found in America, Sainty and Jacobs (1981)
state that it would be more appropriately included as variety platyphylla. The
recurved pistillate pedicels found in Australian specimens would support Sainty and
Jacobs (1981), particularly with reference to published keys to identification (Rataj,
1972a, b) that suggest platyphylla is the only variety with this feature.
There is no evidence in the literature that arrowhead is a weed in its natural range.
Unlike other alismataceous species, such as alisma and S. montevidensis, arrowhead
has not spread into agricultural systems from its natural environment in America.
Recently, however, an outbreak in a lake in Washington has increased enough to
10
cause concern to local authorities in that state (K. Hamel, Washington Department of
Ecology, pers. comm., 13 November 2003).
2.3 Alismataceae species in Australia
A number of alismataceous species occur in Australia, both in natural systems and as
weeds in irrigation farming and infrastructure. Table 2.2 lists the species from the
family Alismataceae that are present in Australia. This information is taken from
Aston (1973) but more recent literature has not added to the list.
Table 2.2 – Alismataceae species present in Australia
Species Details
Alisma lanceolatum With. Common name “alisma”. Native to Europe, North Africa and Western Asia. A weed in rice in NSW. Also occasionally found in Victoria
Alisma plantago-aquatica L. Common name “water plantain”. Widespread in temperate Europe, West Asia and North and Central Africa. A weed in rice in NSW, particularly the Murray Valley. Occurs in natural waters, such as Lake Nagambie, in Victoria.
Caldesia oligococca (F. Muell.) Buch. no common name. Restricted to tropical areas of Australia and also found in Timor.
Caldesia parnassifolia (Bassi ex L.) no common name. Restricted to north-east Queensland. Also found in Africa, Madagascar, to south and central Europe, as well as through South-East Asia to China, Japan, Celebes, Moluccas and eastern New Guinea.
Damasonium minus (R.Br.) Buchenau Common name “starfruit”. Endemic and widespread across Australia. A weed in rice in NSW. Occurs in natural waterways and some irrigation channels in Victoria
Sagittaria engelmanniana J.G. Sm. no common name. Collected only once in Australia, in Victoria on the Goulburn River near Nagambie.
Sagittaria graminea Michx. Common name “sagittaria” in NSW and “arrowhead” in Victoria. Native to North America. Distribution described earlier in chapter
Sagittaria montevidensis Cham. Et Schlecht. Common name “arrowhead” in NSW. Not recorded in Victoria. Two subspecies recorded in Australia, one perennial (ssp. montevidensis) and one annual (ssp. calycina (Engelm.) Bogin)
Sagittaria sagittifolia L. Common name also “arrowhead”, reported once in NSW in 1947. Later reports were unfounded due to confusion with S. montevidensis and S. engelmanniana. Native to Europe and Asia.
11
As mentioned above, most of the research on alismataceous species in Australia has
centred on their status as weeds of the rice industry. The species of importance in that
industry are S. montevidensis, Alisma lanceolatum, A. plantago-aquatica and
Damasonium minus. The situation in which these weeds grow can, however, be quite
different from that in which arrowhead usually grows.
The growth of arrowhead in irrigation systems is usually restricted to channels, drains
and associated infrastructure, as well as in natural systems, such as creeks, rivers and
wetlands. Alismataceous weeds associated with rice cultivation, however, tend to be
restricted to the rice bay, with some growth occurring in channels leading into farms,
but usually not in larger supply channels
2.4 Control of aquatic weeds
The conditions in which aquatic weeds grow, along with aspects of their biology,
present many challenges to their control.
Variation in water level may reduce control. When water levels rise to cover the plant,
successful herbicidal control is reduced because herbicide contact with exposed tissue
is limited. This problem may be overcome by applying herbicide when the water level
is low, to expose more of the plant to herbicide contact, although this may be difficult
in irrigation channels or rivers during the irrigation season. This is also offset by the
fact that many aquatic plants require free-standing water or saturated soil to remain
healthy. If lower water levels reduce the health of plant, then herbicide may not be as
effective as when the plant is healthy.
Another aspect of aquatic weeds that make them difficult to control is their biology.
Many aquatic weeds reproduce via several methods. They can be prolific seeders,
with seed that remains viable for many years, so that control of a standing crop of
plants may provide an opportunity for new plants to be recruited from the soil seed
bank.
As well as reproduction via seed, many aquatic plants reproduce vegetatively,
whether through underground structures, such as rhizomes, corms or tubers, or by the
spread of stem or root fragments. Goulburn-Murray Water observations of arrowhead
suggest that glyphosate and 2,4-D are poorly translocated into underground structures
12
and regrowth of plants is common. There has been some success in rice fields with
the breaking up of corms of water plantain by cultivation (Heylin, in press), but this
approach is not suitable for most other aquatic situations, particularly in irrigation
infrastructure.
The case of arrowhead highlights many of the aspects of aquatic plant biology that
make control difficult. As well as producing large amounts of seed that can perpetuate
populations, arrowhead populations are interconnected by rhizomes and produce
corms. These structures not only help the species to survive over winter, but also
allow it to recover following the removal of top growth by foliar-applied herbicides,
such as glyphosate and 2,4-D.
In addition to these structures, foliar herbicides are unable to come into contact with
submersed rosettes of arrowhead (Figure 2.7). As water height cannot always be
manipulated in accordance with an arrowhead control program that would benefit
from the exposure of rosettes, it is unlikely that contact herbicides could be applied to
rosettes during the active growing season. These rosettes, along with rhizomes, corms
and seeds, ensure that the success of foliar-applied herbicides is greatly reduced.
Figure 2.7 Rosette form of arrowhead, completely submersed
Commonly, the control of emergent aquatic weeds is achieved by the use of
glyphosate formulations (Ailstock et al., 2001), however there are other herbicides
13
registered for use to control aquatic weeds. Goulburn-Murray Water has been using
2,4-D amine (commercially available as “Amicide”) in channels and drains and
amitrole in drains, as well as glyphosate, for control of arrowhead.
These are all foliar-applied herbicides, however, and their effectiveness is reduced by
the factors mentioned above. Metsulfuron (Brush-off®) is another example of a foliar-
applied herbicide. Soil-applied herbicides work in different ways, residing in the soil
or channel bed for extended periods of time and acting on plants that may grow in that
soil, rather than relying on direct initial contact with existing plants. Herbicides that
work in this way include Casoron and bensulfuron (Londax®, used in the rice industry
to control alismataceous weeds).
Control of other alismataceous weeds in rice systems in Australia relies on three main
herbicides. The first of these is bensulfuron, marketed as Londax® DF herbicide, a dry
flowable formulation (Parsons, 1995). It is usually applied from the air onto rice bays.
MCPA is usually applied to weeds that are more mature than those controlled by
bensulfuron and benzofenap. In a rice growing situation, it is applied for further
control of weeds that were missed or not controlled by earlier applications of other
herbicides. Benzofenap is a newer herbicide to the rice industry, registered in
Australia to control S. montevidensis and seedlings of alisma, starfruit and water
plantain. It is marketed as Taipan®, a liquid herbicide of 300g/L benzofenap.
These herbicides are not registered for use in irrigation water, however, and control of
arrowhead by Goulburn-Murray Water has been restricted to the use of only three
herbicides, 2,4-D, glyphosate and, in drains, amitrole. The control of arrowhead using
2,4-D is now not listed on the label for this product, so Goulburn-Murray Water has
obtained an off-label permit from the Australian Pesticides and Veterinary Medicines
Authority (APVMA).
As a result of the litigious nature of society, potential for off-target damage and
diminishing research and development funding in the agricultural chemical industry,
very few new products are registered for use in waterways. The application of
herbicides in irrigation systems to control aquatic weeds is therefore restricted by this
lack of available herbicides and by legislation that limits herbicide use in irrigation
water and in natural systems into which drainage water may flow. These limits are
14
documented in the Australian and New Zealand Guidelines for Fresh and Marine
Water Quality (Australian and New Zealand Environment and Conservation Council
and Agriculture and Resource Management Council of Australia and New Zealand,
2000)
All herbicides used by Goulburn-Murray Water are applied according to product
labels or under a minor use permit issued by the APVMA. The amount of product
applied is limited to that specified by the ANZECC guidelines or that specified by
biocide audits. Three herbicides are used to control arrowhead and include 2,4-D
amine, glyphosate and amitrole + ammonium thiocyanate. All of these herbicides are
more than 30 years old.
2,4-D amine is a systemic herbicide that belongs to the phenoxy group of herbicides.
It disrupts cell growth at multiple sites within the plant and is closely related to
MCPA. The most commonly used 2,4-D products used by Goulburn-Murray Water
have been Nufarm Amicide LO 500A and subsequently Amicide 625 Low.
Glyphosate is a systemic herbicide that inhibits the enzyme 5-enoyl-pyruvyl shikimic
acid 3-phosphate synthase. It is a non-selective herbicide that is readily translocated
within plants. There are several types of glyphosate. In aquatic situations the adjuvant
system in glyphosate has been modified to reduce toxicity to fauna.
The amitrole and ammonium thiocyanate mix is a member of the triazole group of
herbicides, and is used by Goulburn-Murray Water for the control of weeds in drains.
This herbicide inhibits biosynthesis of carotenoids, which play a role in
photosynthesis and protect the plant from sun damage.
Some information on the physical properties of the herbicides listed here and others is
included in Table 2.3
Because of the disadvantages associated with the control of arrowhead using foliar-
applied herbicides, residual or soil-applied herbicides may provide a more effective
control method.
Control with foliar applied herbicides is affected by:
• The existence of the plant in a submersed form, not contacted by herbicide
• The production of seed from September to May, which can germinate
immediately or contribute to the seed bank
15
• The presence of a network of rhizomes that is unaffected by herbicide, as
herbicide may not be translocated to these parts of the plant
• The vigorous regeneration of the plant from seeds, rhizomes and corms
A herbicide that can control arrowhead for a longer period of time (residual), and
before it germinates (soil-applied) may overcome these problems. However, there are
currently no such herbicides registered for use in the control of arrowhead in
irrigation systems.
Herbicides such as metsulfuron-methyl (e.g. Brushoff®) or dichlobenil (Casoron G®)
have potential to fill this role, but concerns about off-target damage, residual levels in
irrigation water and safety must be addressed.
Dichlobenil is a systemic herbicide that has been used for the selective control of
terrestrial and aquatic weeds in New Zealand (Hofstra and Clayton, 2001). Tests have
suggested that its control of aquatic weeds is not especially good when applied into
standing water (Hofstra and Clayton, 2001), however the label recommendations
suggest that it is best applied to moist, drained soil, for best residual effect.
Metsulfuron-methyl has not been used in the past for control of aquatic weeds in
irrigation channels, because of concerns about off-target damage to crops and
horticulture at extremely low rates, and its very slow breakdown in irrigation water.
Other methods of control may include physical cutting, burning, shading and water
level modification (Apfelbaum, 2001). These methods, along with herbicidal control
are all used for the control of Typha spp. in America, with burning, cutting and
herbicidal control also accepted methods of Typha control in Australia. Ailstock et al.
(2001) found that control of Phragmites australis was slightly improved with the
application of fire following glyphosate application. Not all of these methods will be
effective for arrowhead control. For instance, any shading would have to be intense to
have any success in controlling arrowhead, which grows particularly luxuriant in
sheltered situations (see Figure 2.23).
As mentioned above, the use of chemicals in natural waterways, such as the Broken
Creek, and irrigation waters is subject to strict guidelines. To better manage the use of
herbicides in these systems, it is important to build up knowledge of the ecology and
16
biology of weed invasions. Knowledge of these aspects can help in the development
of both economically and environmentally acceptable weed management systems
(Bhowmik, 1997), that may include some of the physical techniques mentioned here
along with herbicidal control or other biological aspects. Such an approach may be
classified as Integrated Weed Management.
17
Table 2.3 Physical and toxicological properties of selected herbicides used for control of aquatic weeds Active Ingredient
Product Use Environmental Fate
Acute Toxicity (mg/kg)
Chronic Toxicity (mg/kg)
Avian Toxicity (mg/kg)
Aquatic Toxicity (mg/kg)
Vap. Press. (MPa)
Melting Point (°C)
Solubility in Water (mg/L)
Amitrole Amitrole T Non-crop land for control of annual grasses and perennial and broadleaf weeds, poison ivy and aquatic weeds in marshes and drainage ditch.
Half-life 14 days, microbial breakdown takes 2-3 weeks in warm moist soil, some chemical breakdown may also occur, biodegradation is 40 days in water, degradation in open water may occur through oxidation by other chemicals.
>5000 rats Enlarged thyroid
>2000 mg/kg non toxic
Slightly toxic <1 157 280,000
Glyphosate Weedmaster Duo
Broad spectrum, non-selective systemic herbicide used for selective control of annual and perennial plants including grasses, sedges, broad-leaved weeds and woody plants.
Half-life 47 days in soil, range from 1 -174 days. Strongly adsorbed to most soils especially those with low OM and clay content. Does not leach appreciably, low potential for runoff except as adsorbed to colloidal matter, < 2% of applied chemical is lost to runoff. Microbes responsible for breakdown, volatilisation and photo degrad. are negligible.
>5600 rats >10000 mice, rabbits, goats
No effects >4000 mallards and quail
Non toxic to fish 86-280 for fish
<1 200 Melting pt.
12,000
Metsulfuron methyl
Brushoff, Ally
Systemic, selective residual pre- and post-emergent herbicide for broadleaf and some annual grasses in wheat, barley, rye and pasture. Inhibits cell division in shoots and roots and is active at low doses.
Half life 14-180 days, average 30. Rate of degradation depends on temp, moisture and pH. Fast in acid soil, high moisture content and temp. Highly mobile in alkaline soil c.f. in acidic, more soluble in alkaline soil. In surface water, DT 50 >84 d at high dose, and 29 in forestry. Stable to hydrolysis at neutral and alkaline pH, at pH 5 half life is 21 days @25°C, >30 d @15°C.
>5000 rats 25 mg/kg/day rats showed no observable effect in 2 y
>2510 ducks quail 5620
Very low 150 fish
17 21 pH dependent, pH 4.6- 270, pH 9- 213,000
18
Table 2.3 (continued) Physical and toxicological properties of selected herbicides used for control of aquatic weeds Active Ingredient
Product Use Environmental Fate
Acute Toxicity (mg/kg)
Chronic Toxicity (mg/kg)
Avian Toxicity (mg/kg)
Aquatic Toxicity (mg/kg)
Vap. Press. (MPa)
Melting Point (°C)
Solubility in Water (mg/L)
2,4-D Amicide Used in pastures, forestry & aquatics to control broadleaf weeds
Half life < 7 d, microbial degradation is the main route of degradation. Despite short half life in soil and in aquatic environments, it has been detected in groundwater supplies. Very low concentrations have also been detected in surface waters. Breakdown in water increases with nutrients, sediment load, and dissolved organic carbon. In oxygenated conditions half life = 7-21 days.
375-666 rats
50 mg/kg/day rats showed no observable effect in 2 y
Moderately toxic 272-1000 mallards, quail pheasants
Some formulations are highly toxic to fish. 1-100 fish
0.02 140 900
Dichlobenil
Casoron Selective herbicide, pre- & post-emergent controlling annual and perennial weeds at seedling and later stages of growth. For control of fruit and other crops at 2.5-10 kg ai /ha, for control of aquatic weeds at 4.5-12 kg/ha. For total weed control <20 kg/ha. Inhibits actively dividing meristem.
Stable in sunlight, rapidly hydrolysed by alkali >3160 20 mg/kg/day rats showed no observable effect over 2 years. (3 generations)
Non toxic >5000 quail, 1500 pheasant
18 guppies 9.8 daphnia
0.073 145 18
19
2.5 The aquatic ecosystem
2.5.1 General principles
The environment in an irrigation channel or drain, particularly when influenced by the
structures associated with irrigation water management, may be seen as analogous to
that of a wetland. It follows that the physical environment in which plants live in
irrigation systems is similar to those in wetlands. For example, the structure of soils in
the sediment is similar, as is the presence of wetting and drying cycles.
The wetland environment is very different from a terrestrial one. The flooding and
drying cycle associated with temporary wetlands can have a marked effect on plant
nutrients present in the soil, in particular nitrogen and phosphorus, decreasing their
availability to plants, whilst nitrogen can be removed from the water column via
denitrification (Moss, 1988). Oxygen is depleted quickly in a soil after it becomes
inundated. Aerobic activity initially increases until oxygen is fully depleted, then
anaerobic activity begins. The depletion of molecular oxygen by aerobic
microorganisms is quicker in the presence of higher levels of organic matter, due to
increased microbial activity associated with organic matter decomposition.
Another factor affected by anaerobic conditions is the production of ethylene, a gas
that affects plant growth and seed germination. Under conditions of complete
anaerobiosis, ethylene production is favoured (Zechmeister-Boltenstern and Nikodim,
1999), whereas in soils with higher water tensions, ethylene degradation rates are
high. This ethylene production is implicated in the success of germination of some
alismataceous species (Flower, 2003; Graham, 1999).
As well as these chemical changes in the environment in which plants grow, the soil
structure and ambient conditions in the soil changes. The porosity of the soil
decreases, reducing diffusion between the soil and the atmosphere. Percolation of
water through the soil is also decreased.
Free standing water above the soil surface changes the temperature of the soil,
buffering the soil from temperature changes in the atmosphere above it. Water is less
reflective than dry soil, absorbs more heat than dry soil and therefore heats up under
20
daylight conditions to a higher temperature than dry soil. Soil itself also becomes less
reflective under water, as it becomes darker and can absorb more heat.
The changes in oxygen and nutrient status in sediment and water, as well as
temperature and light availability all have an impact on plant growth and morphology
(Rea and Ganf, 1994).
As a general description of wetland-style environments, the above applies in irrigation
channels and drains, but there are some factors that are more characteristic of the
irrigation system.
2.5.2 The environment for arrowhead in Goulburn-Murray Water
i) irrigation channels
The physical aspects of irrigation channels, including temperature regime, nutrient
status, depth, flow velocity and turbidity, can vary considerably. The water level in
channels fluctuates with demand on water, with a peak period in the growing months
of the irrigation season. As well as this temporal variation, the size and capacity of
channels varies, depending on the purpose of the channel. Larger channels, closer to
storages and feeding smaller channels, are characterised by high flow rates, water up
to four metres deep and little water level fluctuation during the season. The higher
flow rates lead to greater turbulence around structures, such as bridges and road
crossings. These rates are less conducive to arrowhead establishment. For example,
floating seed is moved rapidly through these areas and the deep, fast-moving water
makes it difficult for seedlings to establish and grow.
In these large channels arrowhead establishes on berms. These shoulders along the
channel edges form a flat, shallowly inundated platform, which is suitable for
arrowhead establishment. Berms provide the ideal water depth for arrowhead, and
produce a slower movement of water, allowing seed to settle and plants to establish.
As the irrigation channels divide and become smaller than the main channels, the
environment becomes more suitable for arrowhead establishment and growth. Smaller
channels and spurs are slower moving and not as deep as larger channels and the
water becomes warmer, further favouring arrowhead growth. The slower-moving
21
water also means that seed deposited on the water surface does not move far before
sinking, while fluctuations in water level, due to changes in demand for water, mean
that more plants can be exposed to shallow water and encouraged to grow.
Sometimes, because they can be quite still, these smaller channels and spurs are less
turbid than larger channels, allowing light to reach germinating arrowhead seedlings,
although this is not always the case. Even with high turbidity, the shallow channels
allow more light to reach the sediment than deeper channels.
Arrowhead tends to germinate and grow on silty sediment, rather than the clay
bottoms of larger channels. These silts collect where water movement is slower, such
as in smaller channels and on the inside of bends in larger channels. Once arrowhead
is established in these silty areas, the problem is compounded by the entrapment of
further sediments by the existing population of weeds, increasing the amount of
sediment that can be colonised further as a result of rhizome production.
ii) drains
Irrigation drains produce a different environment to channels because flooding is rare
and these systems are characterised by frequent shallow water, often only a few
centimetres deep, and occasionally complete drying. A constant water height over 20-
30 cm deep is very rare in these systems. Water depth does, however, fluctuate with
irrigation run-off and with rainfall events, particularly at the beginning of irrigation
seasons.
Commonly, broad-leafed arrowhead occurs in drains (Figure 2.8). The low flows in
drains in Goulburn-Murray Water areas, particularly in the drought seasons of 2001-
2003, favour the settling of any seed input. Following this settling, the warm, shallow
and still water present in drains favours the germination of the seed and the growth of
healthy, broad-leafed arrowhead
22
Figure 2.8 broad leafed arrowhead growing in Goulburn-Murray Water’s
Ardmona Drain II
iii) natural waterways
Like irrigation channels, rivers and creeks provide a varying range of environments.
The River Murray, particularly in the irrigation season, is characterised by large flows
and deep water. Its natural meandering, like that of all natural waterways, provides
many microenvironments in which arrowhead can establish. Natural shallow areas
and sheltered bays provide slow-moving and shallower water, ideal for arrowhead
seed to settle, germinate and establish. Commonly these populations are the broad-
leafed form. Like the situation in channels, existing populations of arrowhead can
perpetuate through the entrapment of further sediment and vigorous rhizome
production.
Smaller waterways, slower moving and shallower, provide ample numbers of smaller
environments like those formed along the River Murray, in which arrowhead growth
is favoured. In these smaller waterways, the natural stands of tall trees that surround
them provide shelter from damaging frosts and cold conditions that can affect
arrowhead growth as winter approaches. This shelter can lead to stands of very
23
healthy plants growing along natural watercourses at a time when arrowhead in
irrigation channels may not look as healthy (Figure 2.9).
Figure 2.9 healthy arrowhead growing in sheltered natural waterway of
Broken Creek
Arrowhead also grows in lakes, dams and other wetland systems. Again, the
conditions in these systems that favour arrowhead growth are warm, slow-moving,
shallow water and shelter. An example of this occurs at the confluence of the Ovens
River and the River Murray, where there are numerous sheltered wetlands, some
infested with arrowhead (Figure 2.10) and the waterways of the Barmah-Millewa
Forest (Gunasekera and Krake, 2001).
Natural waterways, like irrigation systems, contain all three forms of arrowhead, with
a very dense population of the submersed form usually interspersed with the erect
form and extending to greater depths.
24
Figure 2.10 arrowhead infestation in wetland off Ovens River
2.5.3 Case study – arrowhead in the River Murray
Goulburn-Murray Water has conducted surveys of arrowhead distribution in the River
Murray downstream of Yarrawonga to Torrumbarry weir every season since February
2000. With the exception of a slight decrease during high river levels in the
2000/2001 irrigation season, arrowhead increased in the River in this time, with a
general trend towards a movement of arrowhead downstream.
A peculiarity of the River Murray is the distinct zones created by the presence of
weirs. The weir pool, directly upstream of the weir, is a very stable environment,
usually slightly deeper than the river average, with small fluctuations in water level
taking place over long periods of time. This contrasts with areas further away from
the weir, upstream, which are subject to more rapid fluctuations in water level,
sometimes changing day-to-day, often resulting in either full exposure of arrowhead
or inundating it under several metres of water. It is believed that the arrowhead
populations that grow closer to the weir pool are better adapted to the usually deeper,
25
less variable conditions there and, when water levels in the rest of the river increased
and remained deep for some time, these populations were able to withstand the
changes more readily than populations that had established further upstream. These
populations suffered and were greatly decreased when River Murray water levels
were high for an extended period during the 2000/2001 irrigation season.
Goulburn-Murray Water has undertaken to try and control arrowhead infestations in
the River Murray and, in 2002, Roger Baker began a program of arrowhead spraying
in the River. Spraying was conducted in December 2002 and February 2003, using
Weedmaster Duo (a glyphosate formulation suitable for use in aquatic situations). The
control program has been successful in preventing the downstream spread of
arrowhead infestations and minimising current infestations after the December and
February application of glyphosate. For subsequent treatments, glyphosate usage was
greatly reduced. Further growth of arrowhead following these applications may be
due to factors such as the height of the river at the time of spraying causing plants to
be missed and the re-growth of arrowhead from the submersed form or from seed.
This short case study indicates that:
a) arrowhead is spreading in the River Murray, according to the yearly surveys
b) arrowhead growth can be hindered by elevated water levels in situations where
arrowhead populations are not adapted to constant high water levels
c) the application of glyphosate can reduce the size of arrowhead infestations in
natural waterways, by reducing the occurrence of exposed erect stems
d) water level is an important factor when spraying, to ensure good coverage of
herbicide and to identify submersed arrowhead plants that may become emergent
plants in the right conditions.
26
2.6 The biology of aquatic plants, with particular reference to arrowhead
Research into the biology of weeds and the potential uses of biology in weed
management systems has been limited (Bhowmik, 1997). An understanding of the
morphology of the plant, the reasons for the morphology, the reproduction, spread and
growth of the plant can, amongst other information, provide valuable information for
the management of the species.
2.6.1 Forms resulting from phenotypic plasticity in arrowhead
After arrowhead seed is shed from the plant and germinates, it forms small, thread-
like seedlings (Figure 2.11). From these, arrowhead grows into a mature plant and can
ultimately take on any of the forms shown in Figure 2.4. These forms all play a part in
the life history of arrowhead in the field. This phenotypic plasticity is not uncommon
in the genus Sagittaria, with S. sagittifolia L. being a prime example of a species with
many leaf forms (Hroudová et al., 1988), up to 12 different forms being classified.
Figure 2.11 Small, thread like seedling of arrowhead, growing in channel
sediment
27
i) rosette form
The rosette is a submersed form that develops either from a rhizome or from seed. In
other alismataceous species, this form is the juvenile form, as mentioned previously,
whereas the rosette form of arrowhead can exist for a longer period of time.
The rosette form of arrowhead is still capable, however, of changing into emergent
forms, given the appropriate stimuli. Observations from pot trials suggest that the
rosette form does not produce erect leaves unless water height is suitable. This may
be due to depth-related factors, such as light attenuation.
The rosette often escapes contact with foliar applied herbicides. Often rosettes
growing in populations with erect plants are stimulated by the removal of those erect
plants lessening competition, following application of glyphosate or 2,4-D, to produce
emergent, narrow-leafed plants. Application of 2,4-D causes existing erect plants to
die and fall into the water, and may stimulate the transition from rosette plants to erect
plants. Although, in some cases, new erect plants arise from the living root mass left
behind following removal of top growth.
Unlike the juvenile submersed form of other alismataceous species, the rosette form
of arrowhead is able to produce rhizomes, once it has established (Figure 2.12). In
this way, the establishment of rosette forms from seeds can quickly lead to a
proliferation of the population in an area.
The rosette form of arrowhead is commonly found in deeper water or in between the
emergent plants in a dense stand of arrowhead. It is common in both natural systems
and irrigation channels and drains.
28
Figure 2.12 Rosette form of arrowhead, showing rhizomes produced from
single plant, leading to new stems
ii) emergent forms – broad leafed
The other two forms that arrowhead takes are both emergent. The broad-leafed
emergent form (see Figure 2.2, earlier) is the most obvious, and gives arrowhead its
common name, bearing an ovate or lance-shaped leaf. It is believed that this form
arises from an energy-rich rhizome system or as the initial emergent stems forming
from seedling establishment. As mentioned above, it is produced from the rosette
form when conditions, such as water depth, are suitable.
The emergent broad-leafed form tends to occur in slow-moving parts of channels and
streams, along river banks and at the extremities of populations. It is also the form
that occurs most commonly in drains, possibly because populations in drains tend to
be established from seed, rather than from existing subterranean biomass.
Populations made up of rosettes, emergent forms and tiny, thread-like seedlings tend
to form a zonation across the depth profile of a channel, with rosettes across the
whole profile and other forms towards the edges, in shallower water.
29
iii) emergent forms – narrow leafed
The third form of arrowhead is an emergent form with narrow leaves (see again
Figure 2.3), almost grass-like in appearance. It is these leaves that contribute the
species name, graminea, meaning grass-like. These leaves give the plants an
“unhealthy” or depleted appearance, compared to the broad-leafed form, and are
believed to arise from depleted rhizomes. In channels, re-growth following the
application of herbicides is usually of the narrow-leafed form. After a period of
growth, these narrow-leafed populations can recover enough resources to begin
spreading the population via rhizomes. This leads to the production of broad-leafed
stems at the extremities of the populations, resulting in a “boat-like” appearance to the
stand, with taller, broader plants at either end (Figure 2.13).
The depletion of rhizomes resulting from the application of herbicides (Figure 2.14
and 2.15) may pre-dispose the plant to the narrow-leafed form. Herbicide removes top
growth from the plant that has used some of the resources of the rhizome to grow. If
this above-ground biomass has not been able to contribute resources to that rhizome
through photosynthesis it may become depleted.
30
Figure 2.13 Boat-like appearance of a stand of arrowhead, produced by the presence of broad-leafed plants at the extremities
31
Figure 2.14 Re-growth of narrow-leafed form of arrowhead following
application of 2,4-D herbicide
32
Figure 2.15 Depletion of rhizome resources following application of 2,4-D
herbicide
33
iv) seed production, dispersal and survival
Like S. montevidensis, which flowers very early in the rice season, often before the
plant has reached its full size (Flower, 2003) and the closely related S. calycina,
which flowers in the northern hemisphere all season until frost kills it (Kaul, 1985),
arrowhead seems to flower very early in the irrigation season and continue to flower
for a large part of that season, producing large amounts of seed (see Figure 2.16).
Other alismataceous species produce large numbers of seeds on each plant. S. latifolia
produces up to 832 achenes (each containing one seed) per inflorescence, which
equates to around 12500 achenes per plant (Collon and Velasquez, 1989), whilst S.
montevidensis can produce between 800 and 2000 achenes per inflorescence and
between 15 and 32 inflorescences per plant, so that an average plant produces
between 20000 and 21000 achenes (Flower, 2003). Other species produce between
1200 and 1500 seeds per inflorescence (Kaul, 1985).
Figure 2.16 The seeds (achenes) of arrowhead, released from the seed
capsule, may spill on the ground, as here, or on the water surface
34
The achenes of alismataceous weeds may remain on the plant for an extended time, if
not disturbed. It has been suggested that this is due to their being crowded together on
the plant (Björkqvist, 1967), requiring some force to release them.
Achene dispersal in Sagittaria species may be via animals (usually birds) or flotation
(Turner, 1981). Hydrochory (dispersal of seeds and fruits through transport by water
currents) has long been recognised as a seed dispersal agent (Cellot et al., 1998).
The achenes produced by alismataceous species have a pericarp that provides
buoyancy for up to several months (Kaul, 1978). Björkqvist (1967) states that this is
due to air content between the pericarp and the testa, and intercellular spaces within
these tissues. Although said to have an unspecialised floral organisation, indicating a
primitive family, Sagittaria species have an advanced seed morphology (Kak and
Durani, 1989), demonstrated by a complex seed coat, consisting of a waxy layer and
the air in the mericarp (Hroudová et al., 1988).
By contrast with arrowhead, Nymphoides peltata, from the family Menyanthaceae,
achieves flotation of seeds through the seed possessing a hydrophobic surface (Cook,
1990).
Achenes of S. latifolia have been measured as floating on the surface of undisturbed
water for more than two months (Collon and Velasquez, 1989) and those of S.
sagittifolia for many months (Hroudová et al., 1988), while those of S. montevidensis
float for a little under two weeks (Flower, 2003). By comparison, water plantain seeds
have been measured as floating for 16 to 128 hours (Björkqvist, 1967), or just over
five days at most.
There is also some anecdotal evidence to suggest that animals play a role in the
dispersal of arrowhead seed. Populations of arrowhead have been seen to exist at
some distance from each other, with no visible sign of physical connection between
them, such as being connected by irrigation drains, channels or natural waterways.
Whilst they may be considered to be separate outbreaks in some cases, there is also
the possibility that one population may have been established from another, through
dispersal by animals, either attached to the animal (epizoochory) or after being
ingested and deposited by the animal (endozoochory). Although endozoochory can
reduce the viability of the ingested seed (Powers et al., 1978), there is still evidence
for the success of this method of spread.
35
Although exotic plant species may be less attractive to bird species than natives (Loyn
and French, 1991), they still can provide food (which may lead to endozoochory) and
shelter (which may lead to epizoochory). The movement of stock that has accessed
water at infested sites may also facilitate epizoochory.
Amongst the Sagittaria species, S. latifolia is an example of an alismataceous species
that disperses via both hydrochory and epizoochory (Gordon, 1996), whilst the seed
of S. montevidensis has, along with its floating abilities, been noted to have the ability
to stick to animals, both attached to mud and due to a slightly sticky outer surface
(Flower, 2003).
Once seed has been produced and dispersed, it may germinate, be removed by
predation or degradation or be retained in the soil seed bank (Buhler et al., 1997). The
existing soil seed bank, supplemented by new inputs, makes an important contribution
to the plant population in a particular area (Grime, 1989).
The longevity of seeds in this soil seedbank has an impact on how long those seeds
will continue to make a contribution to the plant population growing in a particular
location. The seeds of alismataceous weeds have been found to survive for a number
of years, whilst still retaining some viability. The viability of S. montevidensis seeds
does not reduce significantly after three years in the soil (Flower, 2003), suggesting
that they remain viable for many years after being deposited in the soil. Other
alismataceous species are similarly long-lived in the soil, starfruit seed remaining
highly viable after 7 years (Graham, 1999) and alisma remaining viable after more
than 4 years (Pollock, 1992), while Damasonium alisma seeds are viable for at least
ten years (Birkinshaw, 1994) and water plantain has been shown to remain viable
after 10 years of laboratory storage (Björkqvist, 1967).
Germination of seed may depend on many variables. In the Alismataceae, the main
factors that affect germination appear to be light and temperature. Exposure to light
breaks the dormancy of many plant species (Buhler et al., 1997), with even short daily
periods of light (six hours or less in a 24 hour period) causing S. montevidensis seeds
to germinate (Flower, 2003). Seeds of S. sagittifolia often don’t germinate until the
following spring, remaining dormant for a season (Hroudová et al., 1988).
Temperature is a particularly widespread factor in the germination of seeds, allowing
plants to establish at a time when the temperature is optimal for growth of the
36
particular species. S. latifolia germinates best at 25°C (Gordon, 1996), a temperature
that also corresponds with optimal root growth, emergence and cotyledon formation,
whilst S. sagittifolia germinates best in a broad range from 13°C to 35°C, and is
suppressed at temperatures around 40°C. S. montevidensis germinates well below
10°C, but a trigger temperature of around 11°C causes a marked increase in the
percentage germination of these seeds, corresponding with the lower temperatures
experienced at the beginning of the rice growing season in NSW (Flower, 2003).
Other factors that may affect germination of aquatic species include water depth
(Hroudová et al., 1988; Kaul, 1985), soil atmosphere (including nutrients, hormones,
compaction and other factors), organic matter (Apfelbaum, 2001; Flower, 2003), soil
type (Flower, in press; Hroudová et al., 1988), competition and allelopathy (Gopal
and Goel, 1993).
v) other methods of reproduction
Arthington (1986) states that there are four main factors that favour the adventive
spread of aquatic plants. These are:
• The prevalence of vegetative reproduction
• The role of humans in spread between and within continents
• The capability for rapid reproduction
• Sexually sterile plants that are capable of wide dispersal through small vegetative propagules
Like alisma and water plantain (Pollock, 1992), arrowhead reproduces via seed and
sometimes corms (a type of rootstock consisting of a swollen stem base that arises
underground, usually, in the case of arrowhead, from a rhizome), but also perpetuates
itself via rhizomes (Figure 2.17). This is not an uncommon phenomenon. Seeds of
Phragmites australis may be carried by wind, water or birds, but existing colonies
expand peripherally via rhizome growth (Ailstock et al., 2001).
37
Figure 2.17 Cross section of a drained channel, showing the spread of an arrowhead population via the emergence of new plants from
rhizomes sent out by mature plants
38
The underground parts of aquatic macrophytes form a large part of their biomass
(Kunii, 1993), and contribute to the maintenance of aquatic plant populations. Being
actively growing fleshy organs, rhizomes may not be as long-lived as seeds, but
research into other species (e.g. Kunii, 1993) have shown that, in some species,
rhizome life can be between one and five years.
When perpetuating via rhizomes, both S. cuneata and S. latifolia send rhizomes out
from existing plants, which then turn upright to produce a single daughter plant some
distance away (Lieu, 1979). Observations of arrowhead in the field suggest that it
behaves the same way. From each daughter plant, then, more rhizomes can emerge.
The rhizomes of arrowhead are produced (Figure 2.18) by all of the morphological
forms of arrowhead. The ability of the submersed form to produce rhizomes is one of
the keys to this form being able to survive and perpetuate without producing emergent
stems. In S. sagittifolia, underground biomass is formed very soon after the
production of above-ground parts (Hroudová et al., 1988). At the end of the season,
when above ground biomass dies, the plant survives through these rhizomes, or
through corms.
Overwintering buds, such as corms, turions or tubers, are a way to protect fragile
plants from freezing or decaying in unfavourable conditions (Adamec, 1999), and
usually consist of a swollen stem base, formed from modified shoot apices or at the
ends of rhizomes as a response to cooler, shorter days. This being the case, formation
of corms in arrowhead would be expected to begin in the latter half of the irrigation
season, into May or June.
Corms in S. pygmaea tend to grow under anaerobic condition (Ishizawa et al., 1999),
which would be experienced in irrigation channels for the duration of a season. Corms
of alismataceous species tend to form at the end of rhizomes (Hroudová et al., 1988)
and can be a mechanism to survive desiccated soils, as well as overwintering.
The corms of arrowhead are rounded, fleshy organs that contain starch and form at the
terminus of a rhizome (Figure 2.19). When dried, these organs become very hard.
Corms are produced by most perennial Sagittaria species, including S. sagittifolia
(Hroudová et al., 1988), S. cuneata (Lieu, 1979), S. brevirostra, S. latifolia (Kaul,
1985) and S. pygmaea (Ishizawa et al., 1999)
39
Observations suggest that these organs are commonly produced in populations that
have been subject to herbicide application. This is consistent with the production of
corms as a response to stress, whether that stress be associated with the onset of
winter or other aspects, such as damage due to perturbations or herbicide application,
although corms can be observed in most populations. The production of large corms
allows for a more rapid regeneration following periods of stress than allowed by
regeneration from seed (Figure 2.20).
When the corms re-sprout, this process is usually promoted by anaerobic conditions,
with shoot growth much slower in the presence of air (Ishizawa et al., 1999). Often,
after some growth, the shoots of alismataceous plants separate from the corm and the
corm degrades (Lieu, 1979).
The spread of regeneration function to more than one technique allows plants to
survive adverse conditions and resume growth when conditions are more favourable
(Spencer et al., 2000) and to allow the spread of regeneration over a longer period to
make better use of fluctuations in conditions over extended periods.
Figure 2.18 Rhizome formation at the base of a young arrowhead plant,
with new white rhizomes emerging to the right and older, soil-stained
rhizome coming in from the left
40
Figure 2.19 Corm formed from the incoming rhizome of a mature
arrowhead plant gives rise to new arrowhead plant
Figure 2.20 juvenile plants grown over a period of two weeks, following
planting of dried propagules in soil under 10 cm of water. L to R: from
large corm; from small corm; from seed (coin is 5¢ piece)
41
2.7 The control of arrowhead – the Goulburn-Murray Water perspective
2.7.1 Current knowledge
As mentioned previously, Goulburn-Murray Water has faced many issues and
challenges with the management of arrowhead. Foremost amongst these has been the
variability in control achieved through the use of available herbicides. Goulburn-
Murray Water has found that the same herbicide treatment will not always give the
same level of control in different places. Where arrowhead is managed, it often grows
back, due to the prolific underground biomass, as covered earlier in this chapter. The
regrowth stems from the fact that herbicides are not translocated into this
underground biomass, for example 2,4-D causes abscission of the stem before
translocation into the roots. This herbicide, therefore, acts like a chemical mower,
removing stems, but leaving underground biomass unaffected (Figure 2.21).
Translocation of substances, particularly amino acids, in arrowhead leaves has been
found to be quite good, especially in the presence of light (Schenk, 1972). Schenk
also found, however, that there was an accumulation of amino acids in some parts of
the leaves, particularly the base of the leaf. This tendency may suggest that the
abscission at the base of an arrowhead leaf treated with 2,4-D may be the result of
accumulation of the herbicide at that point.
The effects of glyphosate on arrowhead are probably not as well understood as those
of 2,4-D. Label rates of glyphosate burn the ends of the leaves and don’t offer much
control. Nor does this treatment allow greater hydraulic capacity, like the application
of 2,4-D at label rates. Higher rates of application for glyphosate have been successful
in trials on other aquatic species, such as senegal tea. The effects of glyphosate at
high rates, short and long-term, on arrowhead are unknown, however.
There has been little success in identifying the optimum timings, concentrations,
temperatures or other variables for successful arrowhead management with foliage-
applied herbicides. It is believed, however, that control is not optimal when spraying
at the end of the irrigation season, when arrowhead plants are beginning to
overwinter. It has also been observed that arrowhead grows very actively and rapidly
in autumn, around March and April. This information suggests that these months may
be optimum times for herbicide application. This would make it difficult to implement
a control program, as it falls in the middle of the irrigation season, when water levels
42
are high, submersing much of the plant, and herbicide residues in irrigation water are
particularly undesirable.
The depth of water is a critical factor in arrowhead control, as mentioned previously.
Foliar-applied herbicides are not well translocated within arrowhead plants. When
using these herbicides, therefore, it is important that the herbicide come into contact
with a large surface area of the plant. This is only possible when the water level is
lowered.
Figure 2.21 Stump of arrowhead stem, attached to underground biomass,
left following abscission of main part of stem after 2,4-D treatment.
2.7.2 Mechanical control
Mechanical control is used to manage arrowhead in Goulburn-Murray Water channels
and drains when hydraulic capacity needs to be restored quickly. The technique most
commonly involved by Goulburn-Murray Water involves excavation with machinery
(Figure 2.22). This allows water movement to be restored quickly in situations where
43
large infestations have blocked channels or drains. It is also a technique used in areas
where herbicide use is inappropriate, such as near sensitive crops or channels in
continual use that cannot be shut down during herbicide application.
There are, however, some problems associated with this form of mechanical control.
From a biological perspective, excavation can be undesirable, as it dislodges large
fragments of plants, including stems, roots, rhizomes or corms, that can then float
downstream and establish elsewhere. Unlike the similar movement of fragments
created by 2,4-D, the fragments dislodged by excavation are healthy sections of plant.
It is not known if there is an optimum size or weight of stem, root or rhizome
fragment that will establish new plants. Like the one-off application of contact
herbicides, excavation can leave healthy root and rhizome fragments in the soil, from
which new plants can grow.
Excavation for weed management is also seen as undesirable from an engineering
perspective, as it can damage or re-profile channels or drains. The gradients and
structures of channels and drains are designed to optimise water delivery and
disposal. When weeds are removed using excavation, soil is removed, increasing the
possibility of changing the gradient of the channel or drain and thereby reducing its
efficiency. Drains can become deeper than necessary and sections of channels
changed in such a way as to encourage ponding. There is also the possibility of
damaging channel structure, causing leakage and therefore increased water loss.
Excavation should, therefore, be restricted to removal of deposited sediments.
Excavation is also a less cost-effective means of weed control than herbicide
application. It takes more man-hours to treat a length of weed infested channel
through excavation than through herbicide application. Added to this are the extra
costs of transporting equipment and personnel to the site for excavation works, where
Goulburn-Murray Water herbicide equipment is self-contained on one vehicle.
44
Figure 2.22 Excavation of arrowhead in a channel, showing dislodged
stems that may float downstream and establish
2.7.3 Natural waterways
Control of arrowhead in river systems and other natural waterways is a different
challenge to that in irrigation systems. Natural waterways, although subject to flow
regulation, are more open systems, subject to fluctuations that cannot always be
predicted. Added to this is pressure to ensure any control methods are
environmentally sensitive and to be seen to be doing the right thing with a public
resource. Guidelines mentioned earlier are in place to ensure that control in these
systems is well regulated.
This means, however, that control of arrowhead in natural waterways may be more
difficult than in irrigation systems. It is important, however, to manage weeds in these
systems as much as in irrigation systems, as they may be the source of propagules (as
has probably been the case in the Torrumbarry Irrigation Area – most likely seeded
from the River Murray) and they will also be the destination of propagules emerging
from irrigation systems.
45
2.7.4 Eradication versus control
In many situations, the complete eradication of a weed species is not possible, and the
reduction of the species to a sub-economic level is a more feasible approach. This is
achieved when, despite being unable to eradicate the species, enough is known of its
biology and ecology, that management can be more predictive and effective. This can
result in a reduction in the infestation of or interference caused by the species to
levels that do not have an economic impact on operations or too negative an impact
on the natural value of waterways.
The success of any management program, be it geared towards eradication or control,
is the application of a sound knowledge of the weed’s response to herbicides, as well
as of its biology and ecology. This is the basis of Integrated Weed Management
(IWM).
46
2.8 Integrated Weed Management (IWM)
IWM can be defined as the integration of effective, environmentally safe and socially
acceptable control tactics that reduce weed interference below the economic injury
level (Elmore, 1996). In practical terms, this means the development of a management
plan that includes aspects of the target species' biology, along with targeted or
specific herbicide use and other management techniques, such as minimising the
spread of weed propagules. It may also include aspects of biological control, if
available.
i) herbicides
A description of the herbicides currently used for arrowhead management has been
given above. Given the strictures of the environments in which arrowhead grows,
particularly irrigation channels and drains, herbicides are likely to continue having a
role in arrowhead management.
With a sound knowledge of other possible approaches to arrowhead management,
however, a good IWM program may also be able to be implemented.
ii) mechanical control
The management of arrowhead infestations through excavation has been mentioned
earlier. As mentioned, there are disadvantages to this method in terms of cost and
control, but there are other methods of mechanical control, that may involve the
removal of biomass from existing stands, or the prevention of spread.
Cutting of Typha spp. is an accepted method of control, particularly when the plant is
cut below water level, allowing the plant to “drown” (Apfelbaum, 2001). This process
may not be effective against all species, as some species can respond to cutting by
actively growing.
Another method of control is burning. This, again, is an accepted method of control
for Typha spp., providing the fire is intense enough to destroy the plants completely,
and not just the above-ground biomass (Apfelbaum, 2001). It also provides a good
method of control of Phragmites australis, when combined with appropriate herbicide
47
use (Ailstock et al., 2001). Burning is possible in some plants, as they dry off over
winter. Arrowhead is affected by frost, leaving some dry, brown material that may
possibly be burnt, but it may prove difficult.
Shading is another method of mechanical control that is gaining in popularity.
Anecdotal evidence that aquatic weed growth in small, on-farm channels is reduced
by the presence of large shade trees (E. Hardie, pers. comm. April 2002), is backed up
by the more intense shade provided by the use of plastic sheeting (Carter et al., 1994).
This sort of control has its disadvantages, however, being prohibitively expensive for
large areas, such as Goulburn-Murray Water’s 7000 km of open channels, and being
more appropriate for submersed vegetation, over which the sheeting can sit. As well
as this, re-colonisation is rapid after removal of matting (Eichler et al., 1995), and
matting can become covered with sediment in a dynamic system, providing a fresh
substrate for weeds to colonise.
Equally, while shade trees may appear to contribute to a clean channel, this is only
anecdotal evidence and examples of areas where arrowhead thrives in shaded sections
of Goulburn-Murray Water infrastructure are plentiful (see Figure 2.23). Trees also
reduce access for channel and drain maintenance and create potential occupational
health and safety issues for workers.
iii) biological control
Biological control can be defined as “the use of living organisms to suppress a pest
population, making it less abundant and thus less damaging than it would otherwise
be” (Crump et al., 1999). It can be broadly divided into two categories, classical
biological control, where an organism is released into the environment to reproduce
and proliferate and, from there, to infect, compete with or consume the target
organism, and inundative biological control, where the controlling organism is
cultured and applied directly to the pest organism in large doses.
Classical biological control is hampered by the large amount of money required to
implement it (Chokder, 1967) and sometimes variable success rates. An example of
the successful implementation of classical biological control is the introduction of the
48
Cactoblastis moth into Australia to control prickly pear. Such success stories are
somewhat rare, however.
Figure 2.23 Healthy growth of arrowhead encouraged by shading under
bridge over Goulburn-Murray Water Drain 13 north of Numurkah
As well as examples, like prickly pear, of biological control using an organism that
eats or infects the pest species, introduced species may compete with the pest plant
for resources (allelospoly), or interfere with the pest species by releasing compounds
into the environment that act upon the pest species, a process known as allelopathy
(Szczepanski, 1977). Literature on allelopathy in aquatic plants is very limited,
however, and the effects of other processes, such as competition, are often mistakenly
attributed to allelopathy.
The most successful method of inundative biological control for alismataceous
species has been the mycoherbicide approach, where a mycoherbicide is defined as “a
fungal pathogen which, when applied inundatively, kills plants by causing a disease”
(Crump et al., 1999). In the Australian rice industry, most work has been done using
49
the fungus, Rhynchosporium alismatis (Cother, 1999), but other pathogens have been
investigated overseas (Chung et al., 1998).
iv) control based on plant biology and ecology
Morphology, seed dormancy and germination, physiology of growth, competitive
ability and reproductive biology are all examples of aspects that may be used for
management of weeds, if there is sufficient knowledge in these areas (Bhowmik,
1997). Information on seed banks, root reserves, dormancy and longevity of
propagules may be used to better predict infestations. Weed seed bank densities and
root reserves can be greatly reduced by eliminating seed production for a few years
(Buhler et al., 1997) or through interference with dormancy or germination
requirements (Bhowmik, 1997), or can increase rapidly if plants are allowed to
produce seed.
Generation of knowledge in any of these areas can make a positive contribution to
management of a pest species, through integration into a broad-based IWM program.
Such diverse knowledge has not been available for many pest plant species, including
arrowhead, in the past but is important in developing more successful management
programs. Through increasing the knowledge of arrowhead biology, ecology and
responses, such a program can be developed for the control of this species.
50
2.9 List of questions/hypotheses
All of the information that we already know about arrowhead, along with information
on other aquatic plants, can lead us to a list of issues with arrowhead that should be
addressed by the current study. These relate broadly to the application of herbicides to
arrowhead, the biology of arrowhead, and the response of arrowhead to its
environment. Examples of some of the questions that might be asked are as follows:
i) herbicide
• How effective are the products we currently use for the control of arrowhead?
• Are alternative herbicide chemistries effective in the control of arrowhead?
• What is the optimum timing for the application of herbicides for arrowhead
control?
• What are the optimum rates for the application of herbicides for arrowhead
control?
• The possible effect of 2,4-D on arrowhead has been described (see Figures 2.14
and 2.15). What is the possible effect that glyphosate has on arrowhead (Figures
2.24a and 2.24b) and how does it effect the form of re-growth, particularly at rates
higher than that on the label?
51
Figure 2.24a diagrammatic representation of leaf modification by
glyphosate application
Figure 2.24b diagrammatic representation of the effect of elevated rates
of glyphosate on arrowhead control and re-growth
• Is re-growth after herbicide application due to re-sprouting of rhizomes?
52
ii) biology
• How long does the seed of arrowhead float, allowing it to move with water
currents?
• What are the requirements for arrowhead germination and growth, with respect to
depth (for example, Figure 2.25), light, temperature and other environmental
variables?
Figure 2.25 possible effect of depth on germination of arrowhead seed
(red dots represent seeds)
53
• What factors influence the form arrowhead takes (rosette, broad leaf, narrow
leaf)? For example, is it affected by depth (Figure 2.26)?
Figure 2.26 possible effect of depth on form of arrowhead – submersed vs.
emergent
54
• Does lowering water level cause the rosette form to produce emergent stems
(Figure 2.27a)? Does raising the water level cause the emergent form to return to
a rosette form (Figure 2.27b)?
Figure 2.27a possible formation of emergent form with lowering of water
level
55
Figure 2.27b possible formation of rosette form with raising water level
56
• does the rosette form viable rhizomes toward optimum water level for emergent
growth (Figure 2.28a)? Does the emergent form produce rhizomes towards deeper
water to form more rosettes (Figure 2.28b)?
Figure 2.28a active growth of rhizome towards shallower water
57
Figure 2.28b active growth of rhizome towards deeper water
The diagrams presented in this section on questions and hypotheses show situations
that may arise in the field. The body of this report will attempt to answer these
questions and further aspects of the biology and control of arrowhead.
58
3. Summary of arrowhead research facets from current project
The Arrowhead Project has characterised plant response to herbicide treatment and the impact of channel & drain design and flow characteristics on its growth pattern. Each research facet has been ranked according to the potential for operational Area gains - high, medium and low priority. These have been collated into the tables below. Symbols indicate what we believe is the extent of our understanding ( ) and the extent of potential gains ( ) for each facet from current and future field (operational) trials and smaller trials.
3.1 High Priority with Clear Operational Gains from Research
Research Facet Understanding Operational trial
Small plot &
glasshouse trials
Measures to maximise herbicide efficacy in channels Glyphosate • Determine how glyphosate rate, timing &
water height/plant exposure affect efficacy
2,4-D • Concentration of 2,4-D found in water
after application to arrowhead increases with decreasing water volume, increasing the efficacy of 2,4-D through direct contact with plants, especially submersed plants.
Casoron G • Test effectiveness in channels on
submerged rosettes, rhizomes, corms and seedling (6 weeks prior to season commencing)
Water Management • Determine if emergent (obstructive) forms
can be triggered by lowering water during peak growth period for improved foliar herbicides
Measures to reduce arrowhead establishment or impact on flow Manipulate arrowhead physiology with water management • Determine if variation in water height
increase number of obstructive plants • Submerged plants >50 cm depth don’t
become obstructive plants when a constant water depth is maintained
• At 0-50 cm depth, obstructive plants usually develop
Channel Design • Batter Slope • Channel Depth (de-
silting)
59
3.2 Medium Priority with unclear operational gains
Research Facet Understanding Operational trial
Small plot &
glasshouse trials
Amitrole T • Successful in drains
Seed Germination • in situ germination observed in winter
between -4°C and 10°C
Seed Establishment • Seeds deposit and establish mature plants
in slow-moving or static shallow water (e.g. farmers’ irrigation channels, or between logs and river banks in River Murray)
• Estimated <1% of seed establishes as mature plants
Seed Dispersal • Potential for long-distance dispersal, as
seed floats for up to 3 weeks
3.3 Low Priority with Little Chance of Research Resulting In Operational Gains
Research Facet Understanding Operational trial
Small plot &
glasshouse trials
Channel Profile • Aspect
2,4 D • Gives immediate channel capacity
2,4 D • Provides 6-12 week of channel clearance
2,4 D • Stops seed setting with correct timing
Seedling Establishment • Sensitive to frost and desiccation
Seedling Development • When seedling matures, it will always go
through the rosette stage before becoming an obstructive plant
Seedlings • Water level fluctuation (e.g. over a 3 week
interval) required to develop and ultimately survive
Seedlings • Sensitive to low doses of glyphosate
Seedlings • Casoron G at label rate controls seedlings
Seedlings • Don’t develop into a mature plant under
60
constant deep water. Corm development • Corms produce rosette or emergent plants
depending on water depth
Corm production • Literature suggests they are produced in
greater numbers just before winter
Corm propagation • Number of corms per plant related to
number of rhizomes per plant • Corms usually sink, but float after
disturbance (birds, excavation). • Viability is unknown.
Corm propagation • Corms are common in channel, drains and
natural waterways
Corm propagation • Glyphosate and 2,4-D will not control un-
emerged corms. Re-growth will occur where corms are present
• Glyphosate at high rates reduces below-ground biomass which reduces corm production
• Casoron G controls obstructive growth from corms
Rhizomes • 0-5 rhizomes produced per plant
Rhizomes • As water depth increases, rhizome production
decreases
Rhizomes • Where the channel slope is steep, rhizomes
move across slopes rather than down slopes
Other aspects of propagation • Excavation with machinery releases
broken-off stems and propagules of arrowhead into the water, which might re-establish.
Broad-leafed obstructive form • Dominant form in natural waterways
Broad-leafed obstructive form • Dominant form in drains
Narrow-leafed obstructive form • Develops from corm, seed and rhizomes
Narrow-leafed obstructive form • Appears after treatment with 2,4-D in
channels
Narrow-leafed obstructive form • Is the predominant form that causes
problems with channel flow in Shepparton and Murray Valley
61
4. Details of findings from experimental and survey work
In this section, the findings summarised in Section 3 will be expanded and explained, with
reference to particular experiments, surveys, trials and observations, in order to explain the
conclusions for each facet from Section 3.
4.1 Facets of high priority with clear operational gains
4.1.1 Glyphosate
Experiments that assessed glyphosate rates showed that the maximum label rates for
glyphosate (9 L/ha) did not give satisfactory results for the control of arrowhead. Rates of
glyphosate >9.0 L/ha have proven successful in the control of other aquatic species, and this
concept was extended to arrowhead, through three experiments. Experiment 01 (channel 2 m
width, 40 cm depth) (Figure 4.1), undertaken north of Shepparton and Experiment 16 (Figure
4.2) undertaken in Drain 13, near Numurkah, indicate that an increased application rate of
glyphosate improved initial control.
0
10
20
30
40
50
60
70
80
90
100
Control 9L/ha 18L/ha 36L/ha 72L/ha
glyphosate concentration applied
arro
whe
adco
ver (
%)
assessed 07/05/2003assessed 04/06/2003assessed 14/08/2003assessed 13/10/03
Figure 4.1 percentage arrowhead cover in plots sprayed in April 2002 and 2003
with varying rates of glyphosate (Experiment 01, McCracken Rd, Shepparton)
(Average +/- Standard Error)
62
0
10
20
30
40
50
60
70
80
90
100
Control 9L/ha 25L/ha 50L/ha
glyphosate concentration applied
arro
whe
adco
ver (
%)
assessed 25 days after treatment (DAT)48DAT76DAT
Figure 4.2 percentage arrowhead cover in plots sprayed in April 2003 with
varying rates of glyphosate (Experiment 16, Drain 13, Numurkah) (Average +/-
Standard Error)
Experiment 16 also illustrates the positive correlation between rate, control and the time taken
for glyphosate to control arrowhead. Anecdotal evidence from spray operators suggested that
glyphosate alone was not effective at controlling arrowhead. These assessments are generally
made a short time after application. Figure 4.2 indicates, however, that it can take up to ten
weeks for the symptoms of glyphosate damage to develop. For example, Figure 4.2 shows
that, when assessed after 25 days, arrowhead cover in plots sprayed with glyphosate at 25
L/ha was still around 60%. This decreased significantly after a further 23 days and again after
a further 28 days, to almost no cover at all. The trend is similar for other application rates.
Experiment 06 measured the differences between different rates of glyphosate and their
control of arrowhead, coupled with time of the herbicide application. This experiment
concluded that, as well as increased application rates contributing positively to arrowhead
control, the time of year at which the glyphosate was applied was also significant (Figure
4.3).
63
0
10
20
30
40
50
60
70
80
90
100
control Dec-01 Mar-02 Jun-02
month of application
cove
r Oct
02
(%)
9L/ha18L/ha36L/ha72L/ha
Figure 4.3 percentage arrowhead cover in plots with varying rates of glyphosate
at various times of year (Experiment 06, Main No. 6 channel, north of
Numurkah) (Average +/- Standard Error)
Figure 4.3 indicates that the optimum time for application of glyphosate for arrowhead
control is March, near the end of the irrigation season. Application of glyphosate in June,
while not as effective as March, was also significantly better than the December application
and the unsprayed control.
The greater efficacy at this time of year corresponds with observations in the field, which
indicate that arrowhead is growing vigorously at that time of year. This peak in arrowhead
vegetative growth, and therefore metabolic activity, implies that herbicides applied at this
time have the best opportunity to be translocated and metabolised by the plant.
The efficacy of glyphosate can also be increased through manipulation of water levels. With
water levels lowered, more of the plant is exposed to contact with the herbicide, resulting in
greater glyphosate control. This effect is shown in Figures 4.4a and 4.4b.
Knockdown assessments showed that lowering the water level increased efficacy of
glyphosate by 60% in Experiment 21. In this Experiment, brown out of arrowhead biomass in
Figure 4.4a is due to winter frosts, rather than herbicide application, but the amount of
standing biomass left to frost still indicates lower efficacy of glyphosate in that situation.
64
Figure 4.4a Arrowhead treated with glyphosate at 36 L/ha, with water level kept
at delivery level (Experiment 21, Fuzzard’s Rd, near Waaia, Vic.)
Figure 4.4b Arrowhead treated with glyphosate at 36 L/ha, with water level
lowered to about 15 cm depth.
65
Future Research Requirement
The composition of the arrowhead population (rosette, narrow or broad leaf, rhizome and
corms) and depth of water covering the plants were not considered important when these
experiments were initially conducted. These factors are now known to significantly affect the
efficacy of glyphosate. The results, although providing a guide, are therefore not definitive.
Determine how rate, timing & water height/plant exposure and plant growth stage and
type affect glyphosate efficacy in large scale trials that are representative of Area
situations.
4.1.2 2,4-D concentrations in water
Observations of an un-replicated field Experiment (Grinter Rd), 6 months after spray
application showed that 2,4-D controlled submerged rosettes in shallow, still water along the
berm of a larger channel (200-300 ML/day), whereas glyphosate (36 L/ha) had no effect. This
raised the possibility that 2,4-D may control submerged rosettes in situations where water
levels are low (<0.1 m depth) and water movement is slow.
Trials were established in 80 L pots to test this hypothesis. 2,4-D was injected into the water
containing transplanted rosettes in concentrations ranging 0-32 mg/L of 2,4-D (6.25 mg/L ≈
the field rate of 10 L/Ha when water 0.1 m deep). Results indicated that concentrations of
2,4-D (2 mg/L) produced the typical 2,4- D response, abscission of leaves (Figure 4.5). This
effect increases with increased concentration (Figure 4.6). The results presented here are from
initial trials and more investigation of this aspect is required.
Future Research Requirement
Determine the relationship between concentration of 2,4-D amine in water, time
of exposure and mortality in bin trials.
Verify results in large scale field experiments and manipulate water height to
optimise efficacy.
66
Figure 4.5 Arrowhead rosette leaf-bases, showing elongation of
aerenchyma cells (top), which causes weakness and eventual abscission of
the leaf, compared with healthy leaf (bottom). Elongation caused by
application of 2,4-D.
0
10
20
30
40
50
60
70
80
90
100
Control 0.5 1 2 4 8 16 32
Concentration of 2,4-D added to water (mg a.i./L)
leaf
loss
(%)
Figure 4.6 The effect of increased concentrations of 2,4-D in water on arrowhead
control (initial bin trial)
67
Comment
Frequently, variable efficacy of 2,4-D amine has been reported by spray operators. Although
not fully understood, we believe that this variation is due to the size and type of arrowhead
population, water depth and the characteristics of water flow. Several scenarios are presented
below that may explain the variation.
Infestations that cover whole channel profile
The amount of herbicide applied to the channel is largely proportional to the amount of
arrowhead in the channel when applied with a hand gun (typical practice). Therefore, in
channels with a dense arrowhead infestation, more herbicide will be added to the water
volume, resulting in a higher concentration than would be expected with intermittent or low
numbers of plants. If the population consists of some rosettes, then control will be greater in
shallow, slow moving or still water because 2,4-D will be at sufficient concentration for
control.
Figure – dense infestation of arrowhead (above, left) and a sparse infestation of arrowhead
(above, right)
Channel nearly empty
The concentration of herbicide for a treated area is inversely related to water depth i.e. as the
water depth decreases the concentration increases and hence arrowhead control of submerged
rosettes will increase if water is shallow (< 0.1 m depth).
68
Figure – channel flowing at supply level (above, left) and a close-up at the same site with water-
level lowered (above, right), exposing rosette plants and more of the emergent plants
Channel flowing
The concentration of herbicide will be diluted when fresh water is allowed to flow past the
plant. On berms with slow moving or static water movement, the concentration will be much
higher and more concentrated than in a moving channel toward the centre where water flow is
greatest. Therefore control of submerged rosettes will probably be reduced if channel is
flowing.
69
4.1.3 Casoron G
Experiments using Casoron G, applied in August prior to the commencement of the irrigation
season, showed exceptional results, with control of arrowhead regrowth extending at least 12
months following the application (Figure 4.7).
0
1
2
3
4
5
6
7
8
9
10
Control Amicide 625 Brushoff Casoron Londax Simazineherbicide used
cove
r rat
ing
(out
of t
en)
Oct-02Jan-03Jun-03
Figure 4.7 The effect of several herbicides on arrowhead cover, measured at
various intervals following herbicide application in June 2002 (Kerang)
Figure 4.7 shows the effectiveness of Casoron G (230 kg/Ha) in controlling arrowhead.
Subsequent large scale trials in other channels have shown similar results, with no re-growth
of arrowhead from corms, rhizomes or seed following a winter application, even when growth
of arrowhead in adjacent untreated areas has reached 30 cm in height. An unreplicated
Experiment at the “9-mile knife edge” on the Yarrawonga Main Channel (Figure 4.12)
showed the effectiveness of low doses of Casoron G (23, 50 and 160 kg/ha). Casoron G,
applied prior to filling with water, has significantly reduced arrowhead infestations. No
arrowhead plants have emerged from the treated area, whereas around 70 plants m-2 were
present in the untreated control (Figure 4.8a and 4.8b). The composition of the below ground
biomass was not determined and so effectiveness of Casoron G at 23 kg/ha on rhizomes and
corms is unknown.
70
Figure 4.8a Experimental plot treated with Casoron G, causing suppression of
arrowhead emergence
Figure 4.8b Untreated control plot, showing unaffected arrowhead emergence
Future Research Requirement
71
Determine dose of Casoron G that provides effective control in channels on
submerged rosettes, rhizomes, corms and seedling (6 weeks prior to season
commencing).
Determine potential for contamination of water and the risk of off-target damage.
Comment
Casoron G (dichlobenil) is a granular residual herbicide with registration in some aquatic
situations and is classed as a non-hazardous substance according to Worksafe Australia. It is a
systemic herbicide which inhibits cellulose synthesis in actively growing plant tissue such as
dividing meristems, germinating of seeds and rhizomes. Casoron G’s selectivity for annual
species (vs perennial spp.) is due to its strong adsorption to the soil matrix which limits it
movement to 5-10 cm soil depth. Consequently it has a low potential for ground water
leaching. It is very expensive; when applied 270 kg/ha, the cost per hectare for the product
alone is $3000 AUD, which is similar to hire of an excavator.
4.1.4 Water management – Physiological response of arrowhead to water depth.
Ten field surveys were conducted in drained channels in Victoria (Murray Valley) and
southern New South Wales (Deniliquin) to characterise the form of arrowhead (emergent,
rosettes, seedling and rhizome plant). Channel cross-sections were surveyed for depth using a
laser beacon and staff. Transects were sited across the depth gradient and at intervals the form
of arrowhead was noted (rosette or erect). In general the results showed that at > 50 cm water
depth only rosette plants occurred, with erect plants and rosette plant occurring 0-50 cm depth
(Figure 4.9).
Trials conducted in large pots were conducted outdoors (March-May) and in controlled
environment rooms (June-August) at Tatura to replicate the observation. The trials confirmed
that rosette plants did not produce emergent stems in water > 50 cm depth. Plants grown in 5
cm of water, however, produced 2-6 emergent stems per plant. When water was lowered from
50 to 5 cm depth, all plants produced emergent stems after a short period of time, ranging
from a 2-10 days. If these results extend to the field, lowering water levels could produce of
emergent arrowhead plants that could then be effectively controlled with glyphosate. The
results in Section 4.1.1 indicate that glyphosate efficacy increases when water depths are
lowered. As arrowhead suffers no ill-effects from periods of exposure between one and two
weeks, water level should be lowered as far as practicable to irrigation operations, in order to
maximise exposure to glyphosate.
72
An alternative use of this aspect of arrowhead biology is to maintain water height above the
trigger level for production of emergent stems. The emergent stems are the only form of the
plant that significantly reduces water flow. Maintenance of water heights >50 cm depth may
decrease of the percentage of obstructive arrowhead plants by maintaining them in their
rosette form. LWRDDC (2002) states that fluctuations in water level can prevent
establishment of beneficial vegetation that stabilises banks and the associated wetting/drying
cycle can destabilise the batter material if it has weak physical or chemical characteristics. A
constant water height for arrowhead management therefore also has advantages for batter
maintenance.
Future Research Requirement
Quantify the relationship between arrowhead exposure on erect plants (water depth)
and glyphosate efficacy.
Determine if regular variation in water height increases the number of obstructive
(emergent) plants and if so, is the efficacy of glyphosate greater.
The results from bin trials support the conclusions from the surveys that maintenance of water
>50 cm depth may decrease the numbers of emergent plants and so reduce the development of
obstructive plants. The maintenance of water at >50 cm depth in channels, either by removal
of silt or increased running height, may reduce the impact of the plant on water flow.
Modification of channel design to reduce the width of the zone in which emergent plants
grow (e.g. area of 0-50 cm depth) may further reduce the impact of obstructive forms.
73
Comment
Changes in channel design or use could reduce the load of emergent arrowhead plants in particular systems and favour the less obstructive submerged form of arrowhead.
Figure – Theoretical changes to channel profile to reduce emergent
arrowhead growth
A – unmodified shallow channel with shallow-sloping batters - lots of emergent arrowhead growth, deep sediment layer B – unmodified channel treated with Casoron G to suppress arrowhead re-growth C – unmodified channel, water depth kept high and constant, reduces emergent growth and favours submerged form D – removal of sediment produces a deeper channel, favouring submerged form E – Change in channel design, steep batters and deeper channel, reduces emergent growth
74
-130
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
0 0.5 1 1.5 2 2.5 3
distance from high water mark (m)
heig
ht (c
m)
Supply Height
Figure 4.9 Channel cross section – Mulwala main channel – showing presence of
emergent plants (squares) and rosette plants (shaded circles) around the depth
cut-off of 50 cm
4.1.5 Channel design
A reduction in establishment of obstructive forms of arrowhead relates to the responses of
arrowhead to water depth. If arrowhead remains in the submersed rosette form where water
depth is >50 cm depth, then emergent stems are not produced. Situations were arrowhead
infestation causes greatest disturbance to water flow (e.g. lower end of the channel network in
the Murray Valley) usually occur where water levels >50 cm depth are difficult to maintain
due to flat topography (i.e. lack of grade) and soil type (sandy). Further, deposition of silt (as
a result of slow water flow) contributes to the reduction in water depth and creates a
favourable environment for growth of obstructive arrowhead.
If removal of silt was possible from an engineering perspective (without causing channel
seepage), channel depths >50 cm depth and an unhospitable soil medium (as a result of
excavation removing sediments) would limit arrowhead re-establishment and growth. Where
channels have been de-silted in the past, for arrowhead removal, re-growth has been slow
(often up to 5-10 years), aided not only by the removal of arrowhead plant material but also
by the lack of sediments and the water depth available in the newly-excavated channels.
75
Field observations indicate that the degree and type of arrowhead infestation is affected by
channel batter slope. In channels where the slope of the batter is steep (e.g. 1:2) arrowhead
rhizomes tend to move along or up the channel batter, rather than down the slope, whereas in
channels with a shallow gradient (e.g. 1:7), the rhizomes move both across and along the
channel batter to a water depth <50 cm.
Modifying channels to increase batter slope to steeper than 1:3 may not be feasible because of
the engineering and soil type constraints. However, an increase in the batter slope, where
possible, will decrease the growth of arrowhead rhizomes into the deeper parts of the channel
and minimise the infestation of arrowhead.
LWRRDC (2002) reported on the factors that affect good design for earthen channels. Their
guidelines for channel cross-sections cover bed width, water depth, batter slopes, freeboard,
bank dimensions and operations and maintenance, with a view to constructing channels that
will cost the least to construct, operate, maintain and renew. Their recommendations include
the suggestion that batter slopes of 1:2 are the steepest that should be considered, with
shallower slopes for greater channel bank heights. When designing batter slopes,
consideration needs to be given to operating conditions, effects of water, soil type, shear
strength, soil shrinkage conditions, depth of cutting or height of bank, surcharge loading,
ground water and climatic conditions. Within these engineering constraints, consideration
could be made for optimal batter slope to discourage establishment of arrowhead.
Future Research Requirement
Verify in a large scale field situation that water levels >50 cm depth prevent
obstructive forms of arrowhead developing.
Determine practicality of maintaining water depths >50 cm depth in several
channel networks.
Determine constraints to modification of channel batter so slopes are greater
than 1:3.
76
4.2 Facets of medium priority - operational gains currently unclear
4.2.1 Amitrole T
Amitrole T is widely used by Goulburn-Murray Water for control of weeds in drains (especial
in the Central Goulburn and Rochester Areas) however, is not used in channels in spite of it
been registered in this situation. In the 1980’s, off-target damage (obvious bleaching of
foliage) was observed in a lucerne sward (L. Jackel, pers. comm.) prompting this embargo.
The half-life of Amitrole T in aquatic situations ranges from 23-26 days (aquatic aerobic
metabolism) and several years in anaerobic environments.
Like glyphosate, the symptoms of Amitrole T develop slowly, so that assessments of
bioefficacy should be conducted 6-8 weeks after application. Herbicide experiments 10 & 11
show the effectiveness of Amitrole T in controlling arrowhead in drains where plants are fully
exposed, and that a follow-up application, around 6 weeks after the initial application,
increases its efficacy slightly, as does the addition of glyphosate in a mix with Amitrole T
(Figure 4.10)
0
10
20
30
40
50
60
70
80
90
100
Unsprayed Control Amitrole T (one application) Amitrole T + WeedmasterDuo (one application)
Amitrole T (twoapplications)
Amitrole T + WeedmasterDuo (two applications)
herbicide treatment
% c
ontr
ol
% Control 13 DAT
% Control 37 DAT
% Control 65 DAT
Figure 4.10 Effect of Amitrole T treatments on arrowhead cover over time
(Drain 13, north of Numurkah)
Investigations into the re-growth of arrowhead following successful removal with Amitrole T
are inconclusive. Amitrole T will, however, remain one of the key herbicides used for
arrowhead management in drainage infrastructure.
77
4.2.2 Seed germination
Trials carried out in controlled-environment chambers in the laboratories at DPI, Frankston,
indicated that the trigger temperature for arrowhead seed germination was around 21°C
(Figure 4.11a). The trials indicated that the temperature range for germination was narrow,
centring on this temperature.
The vials were maintained in controlled-environment chambers, in a light regime of 12 hours
of light and 12 hours of dark, and germination occurred, in solution, within a week of trials
being established.
This regime does not replicate the situation in the field for a number of reasons. Germination
was achieved floating in water (Figure 4.11b). In a field situation, it is not uncommon for seed
to germinate whilst floating, but a large number of seeds are also present in the soil seed bank,
with soil being the medium in which they are more likely to germinate. Ambient conditions
with respect to daylength and light intensity may also have differed in the controlled
environment.
Figure 4.11a The effect of temperature on the germination of arrowhead seed in
water in a controlled-temperature environment
Left to right: two vials each at 11°C, 16°C, 21°C and 26°C
78
Figure 4.11b Arrowhead seedlings floating in a vial of water, having germinated
under controlled conditions
The differences in various factors between the controlled environment in which germination
trials were undertaken and conditions in the field may explain differences in germination
patterns between the two situations.
Whilst germination occurred at 21°C in the laboratory, a flush of arrowhead germination can
be seen in the field in late winter to early spring, when the ambient temperature can vary
between -4°C and 10°C (Figure 4.12).
Figure 4.12 Cross-section of berm of Yarrawonga Main Channel, showing green
“carpet” of arrowhead seedlings growing in mid-August 2003.
The situation shown in Figure 4.12 indicates that the arrowhead population in that location is
dependent on the germination of seed to proliferate. In situations such as this, early control of
seedlings that germinate on exposed substrate may be a good option for controlling the
79
population of arrowhead (see Section 4.3.4). The seedlings are very small (Figure 4.13) and
may be sensitive to small doses of herbicide. They also lack an extensive underground
biomass and are not submerged prior to the irrigation season, so contact with herbicide is
maximised, while contact between herbicide and irrigation water is reduced.
Comment
The transition from seedlings to mature plants is a point of weakness in the life cycle of
arrowhead. Arrowhead can produce up to 2 million seeds per square metre in a very dense
stand. However, only a fraction of these establish as mature plants initially, due to sensitivity
to inundation and desiccation. Seedlings do not develop if water covering them exceeds 15-
20cm for a period of more than 6-8 weeks. If water depth fluctuates, however, and the seed
bed does not dry out for longer than a week, seedlings survive.
Once the seedlings establish, they produce an extensive network of rhizomes, which results in
production of daughter plants.
Figure 4.13 Small arrowhead seedlings growing in exposed, saturated soil.
Seedlings around 1-3 cm tall.
4.2.3 Seed dispersal and establishment
Arrowhead seed, like the seed of many alismataceous species, is capable of floating for
extended periods of time. Experiments carried out in troughs indicate that around 3 weeks
80
were required for 100% of seeds being tested to sink. Care was taken to ensure that seed did
not adhere to the sides of the troughs, affecting the counts of floating seeds. Over that period,
there were some seeds that sank in a shorter time, but the majority of seeds took 3 weeks
(Figure 4.14).
This ability of seed to float for up to three weeks gives it the opportunity to move with water
currents and settle in new areas downstream. Depending on water velocity and obstructions,
seeds may be able to move large distances in that time.
0
5
10
15
20
25
30
35
40
45
50
0 5 10 15 20 25
days after placed in water
no. o
f flo
atin
g se
eds
(from
50)
Figure 4.14 Number of seeds left floating over time, after 50 seeds were dropped
onto the water surface in troughs
The distance that the seeds travel whilst floating is dependent on factors such as water
velocity and obstructions. This means that seeds will tend to settle in slow-moving or static
shallow water, such as farmers’ irrigation channels and delvers, dead-end spurs or between
obstructions and in slower-moving areas and inlets in river and stream systems. This is
particularly evident in the River Murray where potential sites for arrowhead growth can be
easily identified (Figure 4.15a and 4.15b).
Where seed is deposited, a rough estimate would be that around 1% of that seed germinates.
The mechanisms for this are unclear. In laboratory conditions around 90% of seed can
germinate soon after it is shed from the plant. In the field, however, this percentage is
moderated by environmental conditions that may include competition or allelopathy from
established arrowhead populations, physical parameters, such as substrate type or light
attenuation or by physiological factors that prevent the seed from germinating immediately
81
and thus ensure future generations by contributing to the soil seed bank. If the seed is as
persistent in the soil as other alismataceous species (3-10 years), this has the potential to
contribute strongly to future arrowhead populations.
Figure 4.15a Arrowhead growing around an inlet in the River Murray, where
slower-moving water has allowed seed deposition
82
Figure 4.15b Arrowhead rosettes (bottom right) growing on a newly-exposed
sandbar in the River Murray
83
4.3 Facets of low priority with little chance of operational gain
4.3.1 Channel profile – aspect
An inspection of Figure 4.12 shows that the northern bank of the channel (right hand side of
the picture) is in the shade, whereas the southern bank of the channel is in sunshine. In this
situation, a difference in the survival of arrowhead seedlings can be seen. Those seedlings that
emerge near the northern bank of the channel are subjected to more severe frost conditions
than those on the southern side of the channel. While frost occurs across the width of the
channel, the rising sun warms the ground that is exposed and alleviates the effect of that frost.
Where the seedlings have emerged in shade, the frost is more persistent and observations
suggest that this has a negative impact on the growth of seedlings. While this information can
guide control programs as to which channel bank is the best from which to approach
arrowhead seedling control, channel structure and obstacles do not always allow that choice
to be made and the benefits of aspect, as it relates to arrowhead growth, are therefore limited.
4.3.2 2,4-D
Unlike glyphosate and amitrole, 2,4-D is a quick-acting herbicide that restores channel
capacity almost immediately (3-7 days). Its application to arrowhead results in a “mowing”
effect, removing top-growth of arrowhead by causing cell elongation and subsequent
weakness at the base of the stem (Figure 4.5). The result is a “stump” left in the sediment,
attached to a root system that remains unaffected by the herbicide (Figure 4.16).
Because the underground biomass in not affected by 2,4-D, re-growth of emergent plants can
occur. This regrowth occurs within 6-12 weeks, depending on conditions and can result in the
channel becoming blocked again within 8-12 weeks (Figure 4.17). After 12 months, the
infestation may be as dense as the previous year, before herbicide application.
If timing of 2,4-D application is exact, however, this removal of top growth may result in
flowering and subsequent seed-set not occurring. Given arrowhead’s long flowering time
(most of the growing season if unchecked), this would be a difficult aspect to perfect, and
would have to combine precise timing to prevent the first flowering and sustained control, to
prevent re-growth and subsequent flowering.
84
Figure 4.16 Arrowhead “stump”, attached to healthy root system, resulting
from the removal of top growth with 2,4-D
0
10
20
30
40
50
60
70
80
90
100
Control AF300 early AF300midday
Amicide 625early
Amicide 625midday
Surpass 300early
Surpass 300midday
Glyphosateearly
Glyphosatemidday
Herbicide Treatment
Arr
owhe
adco
ver (
%)
Percent cover JanuaryPercent cover MarchPercentage Cover JulyPercentage Cover October
Figure 4.17 Percentage cover of arrowhead following application of several
treatments with different 2,4-D formulations, over time. Shows re-infestation
following removal of top growth by herbicide application the previous December
(2,4-D formulations are AF300, Amicide 625 and Surpass 300)
85
Timing of application of 2,4-D may be as important as with glyphosate. Two experiments
were conducted in irrigation channels between Numurkah and Katamatite, where 2,4-D and
glyphosate were applied to arrowhead plants once, twice or three times in a season. In
Experiment 07, those spray times were in January, April and June, whereas the times of
spraying in Experiment 08 were December, February and May.
Initial results from Experiment 07 suggested that one application of 2,4-D was not sufficient
to control arrowhead, but two or three applications both resulted in good control (Figure
4.18). The results from Experiment 08 were not as clear-cut (Figure 4.19), with two of the
three multiple treatments working well, while the third multiple treatment was no better than a
single treatment.
Further examination of the data and the combinations of months presented by the treatments
suggested that these effects were due to timing, with applications made between December
and February being less effective than those between April and June.
Coupled with the results from glyphosate trials, which indicated March to be an effective
month to spray, it was concluded that the “focus months” for spraying arrowhead are March
to early June. These results were replicated when Experiment 07 was repeated the following
season (see Appendix 1, Experiment 07).
0
10
20
30
40
50
60
70
80
90
100
control Jan, Apr & Jun Jan & Apr Jan Jan & Jun
treatment times
% c
over
Oct
ober
200
2
Amicide 625
glyphosate
Figure 4.18 The effect of multiple applications of 2,4-D and glyphosate at label
rates on arrowhead cover in trial plots (Experiment 07 north of Numurkah)
86
0
10
20
30
40
50
60
70
80
90
100
control Dec, Feb & May Dec & Feb Dec Dec & May
Time of herbicide application
% c
over
Oct
ober
200
2Amicide 625
glyphosate
Figure 4.19 The effect of multiple applications of 2,4-D (10 L/ha of Amicide 625)
and glyphosate (9 L/ha of glyphosate 360) on arrowhead cover in trial plots
(Experiment 08 west of Katamatite)
4.3.3 Seedlings – establishment and development
As mentioned in 4.3.1, arrowhead seedlings are sensitive to frosts and, on the northern,
sheltered side of a channel, can be killed by persistent frosts.
Soon after establishment, very small seedlings are vulnerable to sub-optimal growth
conditions, such as frosts and drought. Just as frosts can kill seedlings, so drought can
desiccate the seedlings and kill them. For Sagittaria montevidensis, one week of dry
conditions will reduce small seedling survival by 80% (Flower et al., 1999), and a similar
response is likely from arrowhead, though this has not been measured. Seedlings in this early,
vulnerable stage are easily controlled using herbicides (see 4.3.4).
As the seedling matures, in all cases it first becomes a rosette plant. The mature rosette plant
is more robust than the seedling, and therefore harder to control. It is from these plants that
erect, obstructive plants can develop (see 4.1.4 and 4.1.5).
If, however, seedlings are subjected to constantly deep water, their development is stalled.
Whilst deep water will not directly kill the seedling, the accompanying suppression of
development into the rosette form will eventually result in the death of the plant. Therefore a
period of fluctuating water, for example fluctuations over a three week period, must occur for
87
the seedling to develop into a mature plant. An extended period of deep water is required for
death of seedling plants and the length of this was not measured.
4.3.4 Control of seedlings with herbicides
The vulnerability of small arrowhead seedlings to external conditions such as frost and
drought (see “Comment” in Section 4.2.2 on seedling “weakness” and Section 4.3.3) also
means that they are more susceptible to application of herbicides. Seedlings on the berm of
the Yarrawonga Main Channel were treated with glyphosate at 4.5, 9.0 and 40 L/ha, and with
Casoron G at 23, 50 and 230 kg/ha.
Seedlings were killed by glyphosate at all rates (compare Figure 4.21 with Figure 4.20) within
4 weeks of application. Casoron G killed seedlings within four weeks at the upper label rate
of 230 kg/ha (Figure 4.22), while lower rates of Casoron G took 3-5 weeks longer to kill the
seedlings.
Figure 4.20 Healthy untreated arrowhead seedlings, Yarrawonga Main Channel
berm.
88
Figure 4.21 Mortality of arrowhead seedlings treated with glyphosate at 4.5
L/ha.
Figure 4.22 Mortality of arrowhead seedlings treated with Casoron G (230
kg/ha).
89
Casoron G also suppressed growth of arrowhead plants from underground biomass in the
treated plots, resulting in plots completely free of arrowhead after 6 weeks (Figure 4.23). This
corresponds with results for Casoron from 4.1.2.
Figure 4.23 Plots clear of arrowhead following application of Casoron G at 23
kg/ha. Some arrowhead plants in adjacent untreated area can be seen at the top
of the photo.
4.3.5 Corms – development, production and propagation
Corms are produced at the end of rhizomes that emerge from mature plants (Figure 2.19), so
the number of corms a plant can produce is proportional to the number of rhizomes it
produces. Instead of turning upward and emerging as a new plant, these rhizomes produce
corms under the influence of an external stimulus. In arrowhead and other species, corm
production is most prolific just before winter, allowing the plant to invest energy for winter
survival and a rapid recovery during the spring, when conditions are favourable. Arrowhead,
however, also produces corms throughout the entire growing season, another aspect in the
successful reproduction and proliferation of the species. The situation in which arrowhead
grows (channel, drain or natural waterway) has no effect on arrowhead corm production,
meaning that in all systems arrowhead can proliferate via corms or rhizomes or, if conditions
are conducive, via seed.
90
Unlike their seeds, the corms of arrowhead are not buoyant after production and are only
moved by disturbance to the soil in which they have been produced. This may be caused by
animals, scouring from flood events, or as a result of mechanical excavation. Occasionally,
this movement may result in corms appearing to float. This appearance may be due to currents
tossing the light corms around, or due to the corms being “spent.” In most cases, when the
corm has produced a mature plant, it breaks away from the root system and, depleted of its
resources, may float.
In trials using corms to grow arrowhead plants, 95% of corms were viable, producing mature
and healthy adult plants within 6 weeks. Upon establishment these plants behave like other
mature arrowhead plants, producing emergent plants when water depth is < 50 cm or
remaining as rosette plants where the water depth is > 50 cm. Like other forms of
propagation, the plants that establish from corms always go through a rosette stage,
independent of the final form of the plant.
4.3.6 Corm control using herbicides
In situations where removal of top growth of arrowhead does not affect underground biomass,
it is not believed that the corms of arrowhead are affected by the application of herbicides.
Examples of this effect are the application of 2,4-D, where it doesn’t kill the root system of
arrowhead, and glyphosate at label rates, where control of arrowhead can be minimal.
The application of glyphosate at higher than label rates and the use of Casoron are exceptions,
affecting corm biology in different ways.
(i) Glyphosate at higher than label rates
Glyphosate, when applied at rates between 36 and 72 L/ha, is effective for the control of
arrowhead (see 4.1.1). The complete removal of above-ground arrowhead biomass achieved
using glyphosate at these rates results in a reduction of the below-ground biomass, including
corms. Whether it is a direct effect of the herbicide or an indirect effect, brought about by the
removal of above-ground biomass, it results in a reduction in rhizomes and corms in treated
plots, proportional to the rate at which glyphosate was applied to those plots (Figure 4.24). In
trials (Experiments 01 and 06), this leads to a decrease in the amount of re-emergence of
arrowhead the following year in plots treated at higher rates, of which some can be attributed
to re-growth from surviving below-ground biomass, some to incursion from outside the plots
in small, adjacent-plot trials and some to re-colonisation from remote sources.
91
0
50
100
150
200
250
300
350
400
450
500
control 9 L/ha 18 L/ha 2002 36 L/ha 72 L/ha
glyphosate rate applied to plot
wet
wei
ght
corm
s (g
/sq.
m)
Figure 4.24 Reduction in corm biomass associated with an increase in rate of
glyphosate application.
The discrepancy in plots treated with 18 L/ha of glyphosate is due to an error in
sampling plots that were sprayed only in the 2001/2002 irrigation season, while
all other plots sampled were sprayed in the 2002/2003 season. There are,
however, still reduced numbers of corms present in the 18 L/ha plots, compared
with the unsprayed control plots. Plots are unreplicated.
(ii) Casoron G
Arrowhead re-growth from corms is controlled in a different manner with the use of Casoron
G. Casoron G is applied between irrigation seasons, and its action is to prevent emergence of
plant material through the topsoil of areas that have been treated. In this way, its action is not
to reduce the numbers of corms present in the soil, but to prevent plants that grow from
below-ground biomass from emerging and producing mature rosettes or erect plants. This
action of Casoron G has been recorded in field experiments where Casoron G prevented the
emergence of any arrowhead plants, whilst adjacent, untreated areas supported large
infestations of arrowhead (see 4.1.3). The success of Casoron G in arrowhead suppression at
these sites was despite the presence of underground resources, in the form of rhizomes and
corms, as indicated by prolific re-growth in untreated control plots and plots treated with
2,4-D.
92
4.3.7 Rhizomes – production and movement.
Rhizomes are produced by mature arrowhead plants very soon after establishment. They can
be produced by all mature forms of the plant. Each plant can produce 0-5 rhizomes (Figure
4.25), all of which have the ability to produce more plants, or to produce corms.
Figure 4.25 Arrowhead rosette plant, showing 5 rhizomes that have formed and
are ready to produce further plants.
Experiments in bins indicate that arrowhead rhizome production is suppressed in deeper
water. Arrowhead plants grown at 50 cm water depth grew no rhizomes, whereas plants
grown at 5, 20 and 35 cm produced rhizomes, from which new plants emerged.
The significance of this response in an environmental context stems from the role of rhizomes
in expanding arrowhead populations. Surveys of arrowhead infested channels conducted
during the current study indicate that seedling plants do not establish at depths > 50 cm (see
4.3.3). In this situation, rhizomes play a role in expanding the arrowhead population into
deeper water. The surveys showed that at > 50 cm depth, all arrowhead plants were produced
from rhizomes (Figure 4.26).
93
-160.0
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
20.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
Distance from high water mark (m)
elev
atio
n (c
m)
Figure 4.26 Cross section of berm on Yarrawonga main channel, showing
positions across gradient of seedling rosette plants (pink circles), rosette plants
arising from rhizomes (green circles) and erect plants arising from rhizomes
(green square). Green line represents elevation gradient, brown line is depth
through sediment to clay base.
To achieve this aim of expansion, rhizomes must grow from plants established in shallow
water and either move into deeper water or move along the channel to uncolonised sections at
the same depth. In this way, rhizomes move into areas where there is no competition from
established arrowhead plants. Plants in deeper water, in order to avoid moving back into areas
where arrowhead plants are already established, do so by not producing as many or any
rhizomes. In this way, expansion of the population is into new areas, rather than back into
areas already infested.
As mentioned in 4.1.5, arrowhead rhizomes don’t move down steep slopes, moving instead
along the channel in situations where the slope is too steep. The physiological reasons for this
are unclear, though it is known that rhizomes do not run very far under the soil surface. It may
be that rhizomes running shallowly under the surface (see Figure 4.27) are unable to cope
with sudden, steep changes in slope. The advantages of a steep batter slope in taking
advantage of this aspect of rhizome growth are set out in 4.1.5.
94
4.3.8 Other aspects of propagation
It was noted in 4.3.5 that excavation may be a cause of disturbance that moves corms from the
soil. While excavation can provide good control of arrowhead, its disadvantages include the
potential for damage to channel structure, as well as the release of arrowhead material which
may float downstream and cause new infestations (Figure 4.28). Experiments in troughs
indicate that stem sections, removed by breaking, cutting, or abscission by 2,4-D do not re-
establish into new plants. Uprooted plants, however, even with only a small amount of root
attached, can re-establish. Similarly, healthy root-masses, devoid of any top growth, can re-
establish in new areas. These plants, along with corms and seeds released during excavation
when channels are full, can float away. Ideally, channels should therefore be excavated
carefully when drained.
Figure 4.27 Excavation of arrowhead in channel near Corop, showing rhizomes
of arrowhead running shallowly under the surface of the sediment, visible to the
right of the photograph.
95
Figure 4.28 Excavation of arrowhead near Cobram with water in channel at a
high level, showing plant material floating away from site of excavation – may
contain corms, rhizomes and other propagules.
4.3.9 Forms of arrowhead – broad-leaf and narrow-leaf
(i) broad-leafed form
The broad-leafed form of arrowhead is the dominant form in natural waterways (Figure 4.29)
and in irrigation drains (Figure 4.30). It is unclear why this form dominates these areas,
though the factors affecting this are likely to be many and complex.
One theory that may explain the prevalence of these forms is the source of the infestations.
Surveys of drains indicate that the plants there grow mainly from corms and seed. As newly
established plants, these plants have not undergone the cycle of disturbance and re-growth
from rootstock experienced by plants in other systems. It has been suggested that the narrow-
leafed form of the plant grows from “old plants” that have been through this cycle and are
regenerating from a depleted resource.
96
Figure 4.29 Broad-leafed arrowhead growing in the Broken Creek, near
Numurkah.
Another theory may be that drains and, at least in the case of the Broken Creek, some natural
waterways are nutrient sinks, receiving runoff from irrigation and other land, containing
nutrients, topsoil and other materials. These plants are therefore not nutrient-limited and do
not rely so much on resources stored in their below-ground biomass for regeneration and
growth. This results in a fuller-formed, healthier-looking plant.
In reality, the reasons for this form being dominant in drains and natural waterways are
unknown and may be a combination of these two theories, or other factors.
97
Figure 4.30 Broad-leafed arrowhead growing in a drain near Ardmona.
(ii) narrow-leafed form
The narrow-leafed form of arrowhead is the dominant form in channels, particularly in the
Murray Valley and Shepparton Irrigation Areas (Figure 4.31). The fact that this form often
grows after the application of 2,4-D supports the suggestion that the narrow-leafed form
grows from an old or depleted root stock. When this root stock first re-grows, it is the narrow-
leafed form that grows. As the population expands via rhizomatous growth, it is able to
produce more resources for the rootstock through photosynthesis and subsequent plants may
take on the broad-leafed form. It is this process that produces the “boat-like” appearance of
some populations (Figure 2.13), as the broad-leafed form grows at the extremities of an
otherwise narrow-leafed population.
98
Figure 4.31 Narrow-leafed arrowhead in a channel, Shepparton North
In contrast to the situation in drains and channels, another reason for the presence of the
narrow-leafed form in channels may be a lack of available nutrients. If the root stock of the
plant is depleted of resources through having to produce vegetative biomass (new stems and
leaves) following disturbance, such as rapidly changing water levels or removal of top growth
by herbicides, then the form of the resulting plants may similarly be “depleted” by a lack of
available external nutrition. Rather than exhaust its available resources by producing a fully
formed leaf, the plant produces narrow leaves to conserve those resources.
Again, these explanations of the prevalence of the narrow-leafed form in some situations have
not been fully researched and may be only part of the picture of arrowhead form.
99
5. Implications of results for future research and project direction
The course of the current program has been to expand our knowledge of the biology and
control of arrowhead, in order to understand how the species can be managed. Previously, the
understanding of arrowhead biology, ecology and control had been limited, and these steps in
building our understanding are important for implementing management programs.
Having increased our knowledge through the course of this project, that knowledge will now
be applied through implementation of control and management programs utilising the latest
findings. In addition, Section 3 indicates where our understanding of arrowhead biology and
control is still lacking in some aspects and further research in the field and in smaller trials
may yield results that will lead to operational gains for arrowhead control.
5.1 Facets of high priority with clear operational gains from research
Of the facets that appear to offer clear operational gains (Section 3.1), most gains will be
made through conducting operational trials in the field. These include gaining further
knowledge of how some of the aspects of herbicidal control work on a large scale,
particularly with respect to the utility of Casoron G in an operational situation and the
utilisation of 2,4-D concentrations in water for better control of submerged and emergent
plants. The role of rate, timing and water height in glyphosate efficacy is well understood, but
fine-tuning of operational procedures to best exploit these variables can be achieved through
further operational trials.
Whilst knowledge of the responses of arrowhead plants to changes in water depth are now
better understood, this is another facet of arrowhead biology where further operational trials
will increase the ability to utilise that knowledge. Operational trials will help develop best
procedures and practices for manipulating water levels for arrowhead management and also
allow for development of channel design and maintenance programs to minimise arrowhead
infestations.
5.2 Facets of medium priority with unclear operational gains
As outlined in Section 3.2, some of the facets of arrowhead biology that have been identified
do not have such clear potential for operational gains as those discussed in 5.1. As most of the
facets in this category are well understood, the long-term benefit of investing further time into
their research is unclear.
Those facets relating to the dispersal of seeds and establishment of plants from those seeds
have a clear role in the understanding and prediction of arrowhead responses to environmental
100
variables and can therefore be used to direct management practices. As our understanding of
these processes is already clear, however, gains from further research will probably be small.
Gains to be made from these facets are more likely to come from their integration into
management programs.
Operational trials into the effects of Amitrole T may produce more significant results.
Amitrole T is a herbicide the use of which, for arrowhead control, is not well understood.
Trials in the current program have indicated that good control can be achieved through the use
of Amitrole T, especially in mix with glyphosate and when application is repeated 6 to 8
weeks after the initial application. Further trials, however, may reveal aspects that influence
the efficacy of Amitrole T and the pattern of re-growth following Amitrole T application, in
much the same way as trials have increased our understanding of glyphosate and 2,4-D
application.
5.3 Facets of low priority with little chance of further research resulting in
operational gains
This category includes facets of which the current understanding is complete, making further
research unnecessary or facets on which further research will not yield significant gains for
operations.
Whilst research to date on seedling development, corm biology and rhizome production has
produced results that have been important in fulfilling the objective of learning more about
the biology of arrowhead, future research will be aimed more at implementing management of
arrowhead, based on the current knowledge. This means that future research will be centred
on those facets that provide efficiencies in arrowhead management, such as herbicide efficacy
and management of current populations. The use of 2,4-D is one facet in this category that is
directly related to control of arrowhead and where knowledge is well developed. Further
research in this area is therefore not likely to yield any further understanding. As the use of
2,4-D for arrowhead control is an established aspect of the current program, further field or
operational trials are similarly not necessary.
101
5.4 Summary of future research and project direction
A summary of the current state of knowledge on arrowhead biology and control was
presented in Section 3 of the current report. The information gained from the current project
and summarised in Section 3 can be used to devise a draft management plan for arrowhead
(Section 6) and will provide direction for future research into arrowhead control. This future
research will be aimed at testing the draft management plan, fine tuning or re-configuring that
plan and ensuring its implementation.
The most promising aspects of arrowhead biology and control, as identified in Section 3, are
those aspects that can contribute to this process and are, therefore, the aspects that warrant
further research. The current project has contributed valuable knowledge with which future
research and implementation can go forward. Whilst new knowledge on arrowhead in general
will continue to be obtained, the focus of further work will be to put the knowledge gained
from this project into practice.
102
6. Arrowhead management plan based on findings from current
research
The objectives of the research project “Arrowhead – Biology and Control” were as follows
(Krake and Breewel, 2000):
• To obtain a greater knowledge and understanding of the biology and ecology of
arrowhead, its propagation and dispersal.
• To investigate and develop management and control strategies for aquatic environments
where arrowhead exists.
The knowledge gained in the course of this study and outlined in Section 4 fulfils the first of
these objectives and allows for the fulfilment of the second of these objectives through the
formulation of a draft plan for the management of arrowhead in areas where the plant
currently grows. Knowledge on the timing and application of herbicide has already proven
successful in the treatment of the plant in the River Murray (see 2.5.3) and is being applied to
the treatment of the plant in the Goulburn River. This knowledge, however, needs to be linked
with other findings to produce a more comprehensive plan for arrowhead management both in
these systems and in irrigation infrastructure.
The formulation of a management plan for arrowhead starts with a broad approach (Figure
6.1) and is built upon utilisation of knowledge gained through the course of this study, with
respect to optimal herbicide efficacy and aspects of the plant’s biology.
103
Predictive measures (where’s arrowheadgoing to occur?)• How much arrowhead at top of system?• How will water be used?• What changes in water height will occur?
Reactive Management• 2,4-D to give quick capacity• remove arrowhead by excavation
Load Management• Glyphosate program March - June• Seedling control with Glyphosate around August• Casoron control - off season• Excavation to reduce sediment
Control to reduceseed coming intoirrigation system
Preventative measures• control of sediment build-up• maintenance of water height• change in batter shape• deepen channels at bottom of system
Source (eg River Murray)
Figure 6.1 Steps in approaching arrowhead management in an irrigation system
The broad approach shown in Figure 6.1 spans several steps in the infiltration of a system by
arrowhead, from prevention of influx of propagules to a clean system, through to the reactive
approach of applying techniques to rapidly allow delivery of water in systems where severe
infestations of arrowhead are already established. The application of the knowledge gained
from the present study allows each of these steps to be expanded.
6.1 Aspects contributing to broad management plan
6.1.1 Control to reduce movement into system
This first step in arrowhead management is important in situations where arrowhead has not
yet established. In such situations, it is important to gauge the potential for arrowhead
infiltration through an investigation of potential sources. A good example of this approach is
the control program instigated for arrowhead in the River Murray (see 2.5.3).
This program was instigated not only to preserve the natural integrity of the River Murray,
but also to minimise the source of seeds and other propagules that may be transported into
irrigation areas fed from the Murray in which arrowhead had not yet become established. The
need to control arrowhead in the River Murray was identified following extensive surveys of
the river carried out in the years before the program.
104
Similar efforts to identify potential sources of arrowhead propagules in areas that may be in
danger of arrowhead infiltration should be undertaken and efforts made to stem the movement
of propagules into those areas.
6.1.2 Preventative measures
Where the possibility to remove the threat of incursion by arrowhead propagules is not
available, but arrowhead is not yet established, preventative measures should be employed.
These are steps, based on the current knowledge of arrowhead biology and ecology, that can
be used to promote an environment that is unsuitable for the proliferation of arrowhead.
The main area in which measures can be taken relates to the relationship between water depth
and arrowhead biology. Where water depth is above trigger levels for development, seedlings
that may arise are less likely to develop into adult plants. Equally, any rosette plants that may
form are unlikely to develop emergent stems and, if established in deep water, may not
produce rhizomes. Prevention of emergent plant formation not only reduces the amount of
obstruction to water supply, but also prevents flowering and subsequent seed production
which can perpetuate an invading population.
The water depths needed to restrict arrowhead establishment or the production of emergent
plants can be achieved through careful channel maintenance and use, or through changes to
channel design which discourage colonisation of shallow water yet do not compromise
structural integrity.
Control of sediment build-up or regular excavation of built-up sediment makes it easier to
maintain channel water levels above around 50cm, the cut-off depth for a number of aspects
of arrowhead biology. Coupled with changes in procedure, where possible, that restrict the
length of time when water levels in channels are low, these maintenance issues should lessen
the potential impact of arrowhead infestations.
Similarly, within the bounds of engineering possibility, changes could be made to channel
design to form channels of modified cross-section that reduce the area available for the
formation of emergent arrowhead plants (Figure 6.2 for example)
105
-130
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
0 0.5 1 1.5 2 2.5 3
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure 6.2 Channel cross section – Mulwala main channel – showing presence of
emergent plants (squares) and rosette plants (shaded circles) around the depth
cut-off of 50 cm in existing channel profile (black line) and a theoretical channel
profile (red line) that would reduce the width of the zone in which emergent
plants could grow.
6.1.3 Predictive measures
In some situations, modifications to channels may not be feasible, or changes to the
maintenance and use of those channels may be constrained by channel design, water
availability or other external factors. In such situations, if it is not possible to prevent
arrowhead proliferation, then a prediction needs to be made of where and under which
circumstances arrowhead may occur.
An estimate of how much arrowhead is feeding into the system or exists at the top of a system
will allow a judgement to be made as to how much of a problem arrowhead may become
further down the system. Further investigation of where arrowhead is likely to become
established in the system can be aided by incorporating current knowledge on arrowhead
spread and establishment. An estimate of how much water will be in the system, how that
water will be used and what effect that usage will have on water level fluctuations, coupled
with knowledge of possible inputs from further up the system, will help in predicting how
severe arrowhead may become in a particular part of the system.
106
6.1.4 Load management
Once a prediction is made of how severe a potential arrowhead problem may be, the best
approach is to investigate the possibilities for load management. Removal of sediment
through excavation, where possible, will reduce the availability of suitable habitat for
arrowhead infestation, whilst a more direct approach would be the use of herbicides.
Casoron G is registered for the control of arrowhead, if applied to channels prior to the
commencement of the irrigation season. Early results demonstrate that Casoron G prevents
the emergence of new plants at the beginning of the following season. Whilst not tested over
the long term, the suppression of arrowhead emergence attained through use of Casoron G
after 12 months is promising. As a tool for managing arrowhead load, the suppression of
emergence in turn reduces the number of arrowhead plants that gain full maturity and produce
reproductive structures, such as seeds and corms, and invasive rhizomes.
Load management is a means of removing the contribution that one or more of these means of
reproduction makes to future arrowhead populations. In situations where seedlings make a
vast contribution (e.g. Figure 4.12), removal of those seedlings can have a significant effect
on the size of the population that establishes and the subsequent propagule production by that
population. In cases such as these, a program of small seedling removal using low rates of
glyphosate will remove the contribution of seed to the population. In conjunction with
Casoron G application to reduce emergence of plants from other reproductive structures, this
will reduce the population in the subsequent season significantly.
The alternative is to effectively manage the contribution of plants that do establish. This can
be done through the efficient use of herbicides. Efficiencies can be gained through managing
the timetable of herbicide application to correspond with the best timing for arrowhead
removal or to prevent significant events, such as flowering and seed set. Other aspects of
herbicide application, such as the use of appropriate rates or techniques like follow-up or
repeat applications may also produce efficiency gains.
6.1.5 Reactive management
In many situations where arrowhead has become firmly established, often the only option is to
conduct reactive management. This is where arrowhead populations have reached such levels
that set load-management programs alone are not possible and immediate action must be
taken to restore water delivery capacity or to prevent serious infestation. Such actions,
sometimes referred to as “putting out fires”, do not necessarily fit in with best practice with
respect to aspects of load management such as timing of herbicide application.
In this area of management, excavation and 2,4-D application are the key practices. Unlike
excavation to construct environments detrimental to arrowhead establishment, excavation in
107
this case is to remove established populations of arrowhead. It can be coupled with a removal
of sediment with the plant infestations, to produce an environment that discourages further
growth, but its primary purpose in this case is to restore water flows.
Similarly, the application of 2,4-D in this situation is aimed at restoring capacity quickly. The
action of 2,4-D, in causing abscission and removal of upright arrowhead stems, is rapid in
comparison to other herbicides. This rapid removal restores capacity quickly but allows for
re-growth at a later stage.
The nature of reactive management means that the long term results of these practices may
not be as positive as the results from careful preventative and load management measures.
However, in many situations these works must be undertaken to maintain operations in
irrigation systems.
Ultimately, the best practices are those nearer the top left part of Figure 6.1, and that is where
most gains in efficiency can be made. In many situations, however, current practice is closer
to the bottom right part of Figure 6.1. Careful management, utilising current and future
knowledge gained on the biology, ecology and control of arrowhead, may be able to shift
arrowhead control measures towards more preventative, pro-active techniques. For this to
occur, a comprehensive management plan for control of arrowhead in the many situations in
which it grows is required.
108
6.2 Draft management plan
The management plan for arrowhead should be based on the approaches outlined in Figure
6.1 and Section 6.1, notably that the best management of arrowhead is to prevent its
infiltration where possible. The following tables outline options for arrowhead management
in natural waterways, channels and drains, from prevention through to removal.
Table 6.1 – Management options in natural waterways
Control option Best practice / Advantages Constraints / Shortcomings
Control of arrowhead emergence with Casoron G
Control is best achieved in areas where arrowhead will grow through application of Casoron G before the growing season
Casoron G acts by suppressing emergence of arrowhead from existing underground biomass
Casoron costs around $3000 AUD/ha
Manual removal of small infestations
Control is achieved cheaply and in an environmentally sensitive fashion very early in infestation process
Only effective with very small initial infestations. Care must be taken to ensure all propagules – seeds and underground biomass – are removed
Control of existing arrowhead with glyphosate – 10 L in 100 L mix as per APVMA permit (PER6875)
Control is best achieved when plant is actively growing. This is usually in the March to June period, but may occur at other times. Monitor arrowhead to establish best spray time
Control before flowering helps reduce the number of propagules
Lower water levels expose more of the plant and improve glyphosate efficacy
Glyphosate residues in waterways are subject to legislative controls, care should be taken to not exceed accepted levels
109
Table 6.2 – Management options in irrigation channels
Control option Best practice / Advantages Constraints / Shortcomings
Prevention of infiltration into system
Control of arrowhead in areas feeding into irrigation systems or areas at the top end of irrigation systems (see Table 6.1)
Constraints similar to those outlined in Table 6.1
Often, problem is not realised until arrowhead is already in the system
Channel structure
• Re-design channel cross-sections
• Minimise sediment build-up in channels through excavation
Channel cross-sections could be modified to minimise areas where water depth is <50 cm, to discourage emergent arrowhead growth
Removal of sediment build-up to maintain channels that are deeper will again discourage arrowhead emergent growth
Cross-sections of channels subject to engineering constraints (e.g. reduction of slumping, erosion etc.)
Some channels and drains may not be deep enough to modify arrowhead growth, even after excavation
Some channels and drains (e.g. in some Murray Irrigation Ltd areas) are already over-excavated
Predict where arrowhead may grow –– and direct control effort to these areas
Direct effort towards shallow channels, slow moving channels, areas behind flow obstructions, where propagules may deposit
Direct efforts to areas fed from established populations upstream
Detailed knowledge of all parts of the system required
Unpredictability of arrowhead establishment makes it difficult to predict whether infestations will actually occur in areas identified with potential for establishment
Control of emergence with Casoron G
Control is best achieved in areas where arrowhead will grow, by application of Casoron G before the growing season
Casoron G acts by suppressing emergence of arrowhead from existing underground biomass
Casoron is not specifically registered for use against arrowhead
Casoron controls emergence from underground biomass that is already present
Probably best used for control at the top end of the system or on berms of high-flow channels
Control of seedlings that grow before irrigations season
Control of seedlings is available with low rates of glyphosate (4.5 L/ha) – inexpensive
Control of seedlings with label rates of Casoron G will also minimise regrowth from underground biomass
Location dependent, as seedlings may make a major contribution to the population in some areas, where other areas are populated with plants from vegetative reproduction
Casoron not yet registered for use against arrowhead
110
Table 6.2 (continued) – Management options in irrigation channels
Control option Best practice / Advantages Constraints / Shortcomings
Control of established arrowhead plants in “focus months” of March - June
Arrowhead control in the “focus months” is greatly improved, both with glyphosate and 2,4-D
For both herbicides, where possible, water level should be lowered when applied, as this improves efficacy
2,4-D efficacy improved if plants are covered with shallow water at time of application (see 4.1.2)
Glyphosate control is greatly improved at higher rates (40 L/ha, as per APVMA permit PER6999), while label rates of 2,4-D (APVMA permit PER6341) are effective
Period of most active arrowhead growth usually falls in these months, but environmental variables may affect this – active period may fall outside these months, so monitoring of arrowhead growth rates important
Regulation of water-level not always possible
The action of 2,4-D usually results in re-growth of arrowhead plants from underground biomass in 6-12 weeks
APVMA restricted use permits impose constraints on use of these herbicides and should be consulted
Control outside “focus months” with 40 L/ha of glyphosate or with 2,4-D, to restore delivery capacity
For both herbicides, where possible, water level should be lowered when applied, as this improves efficacy
2,4-D efficacy improved if plants are covered with shallow water at time of application
Glyphosate action can take up to 10 weeks to fully express itself, making this option less attractive for quick restoration of water delivery capacity
Although 2,4-D acts quickly to remove standing biomass and restore channel capacity, its action results in re-growth of arrowhead plants from underground biomass in 6-12 weeks
Control using excavation Excavation should preferably occur in channels that have been emptied, to minimise movement of propagules with water flow, and to improve ease of excavation
No chemicals required, so can be utilised near sensitive crops.
Control may extend for many years
Excavation can release propagules that may move in water flow to colonise new areas
Damage to channel structure, changes to channel cross-section and drop may occur
111
Table 6.3 – Management options in irrigation drains
Control option Best practice / Advantages Constraints / Shortcomings
Prevention of infiltration into system
Control of arrowhead in areas feeding into drainage systems or areas at the top end of systems where arrowhead is established
Often, problem is not realised until arrowhead is already in the system
Predict where arrowhead may grow and direct control effort to these areas
Direct effort towards drains where water levels are frequently or continually conducive to arrowhead growth (e.g. fluctuate at <50 cm deep)
Direct efforts to areas fed from established populations
Extensive knowledge of all parts of the system required
Unpredictability of arrowhead establishment make it difficult to predict where infestations will occur
Control of standing arrowhead with glyphosate or with Amitrole T and glyphosate in mix
Glyphosate works better at higher concentrations (e.g. 40 L/ha)
Amitrole T alone can be used, but it works slightly better in a mix with glyphosate
A follow-up application of Amitrole T, 6 to 8 weeks after initial application improves results
Both Amitrole T and glyphosate are slower-acting herbicides. Results may take up to 10 weeks to fully express
Control with these herbicides does not remove below-ground biomass and propagules. Invasion by seed is also common in these systems. This means that control efforts with herbicides will have to be ongoing
Control using excavation Excavation should preferably occur when drains are not carrying water, to prevent movement of propagules with water flow, and to improve ease of excavation
No chemicals required, so can be utilised near sensitive crops.
control may extend for many years
Excavation can release propagules that may move in water flow to colonise new areas
Damage to channel structure, changes to channel cross-section and drop may occur
112
The options suggested in these tables should be put into the context of Figure 6.1. Where
possible, efforts should be made to control arrowhead infiltration and have control efforts
remain in the upper portion of the flowchart represented in Figure 6.1. The tables of
management options are arranged with options relating to prevention of infiltration and
establishment near the top of the table, with options for more established problems located
further down the table for each situation. In areas where arrowhead is not yet a major
problem, such as the Torrumbarry Irrigation Area, management should start at the top of each
table. Areas where arrowhead is established, such as the Murray Valley Irrigation Area, will
move down the table, past those options regarding prevention of infiltration and
establishment, to those concerned more with management of existing infestations.
The steps outlined in this suggested management plan indicate that options for management in
channel systems are more numerous than in natural systems and in drains. While this is true, it
should also be noted that anecdotal evidence has suggested in the past that arrowhead is easier
to control in these systems than in channel systems.
It is hoped that the management options presented here will improve the ability to control
arrowhead in all systems. As these measures are implemented and further investigated as
everyday practice, they will be fine-tuned and, with the addition of findings from further
research, improved over time. The aim of the draft management plan is for areas to improve
arrowhead control and be able to move away from “reactive management” further upwards
towards “preventative measures” as represented in Figure 6.1.
113
References
Adamec, L. (1999). Turion overwintering of aquatic carnivorous plants. Carnivorous Plant Newsletter 28, 19-24.
Ailstock, M. S., Norman, C. M., and Bushmann, P. J. (2001). Common reed Phragmites australis: control and effects upon biodiversity in freshwater nontidal wetlands. Restoration Ecology 9, 49-59.
Apfelbaum, S. I. (2001). Cattail (Typha spp.) management. www.appliedeco.com/Projects/CattailManage.pdf, accessed 15th October, 2002.
Argue, C. L. (1976). Pollen studies in the Alismataceae with special reference to taxonomy. Pollen and Spores 18, 161-202.
Arthington, A. H., and Mitchell, D. S. (1986). Aquatic invading species. In "Ecology of Biological Invasions." (R. H. Groves and J. J. Burdon, eds.), pp. 34-53. Cambridge University Press, London.
Aston, H. I. (1973). "Aquatic Plants Of Australia," John Sands, Halstead Press Division, Melbourne, Victoria.
Australian and New Zealand Environment and Conservation Council and Agriculture and Resource Management Council of Australia and New Zealand (2000). "An Introduction to the Australian and New Zealand Guidelines for Fresh and Marine Water Quality (2000) - Paper No. 4a. " http://www.deh.gov.au/water/quality/nwqms/#quality, accessed 19th January, 2004
Bhowmik, P. C. (1997). Weed biology: importance to weed management. Weed Science. 45, 349-356.
Birkinshaw, C. R. (1994). Aspects of the ecology and conservation of Damasonium alisma Miller in Western Europe. Watsonia 20, 33-39.
Björkqvist, I. (1967). Studies in Alisma L. I. Distribution, variation and germination. Opera Botanica 17, 1-128.
Brown, W. V. (1946). Cytological studies in the Alismaceae. Botanical Gazette. 108, 262-267.
Buhler, D. D., Hartzler, R. G., and Forcella, F. (1997). Implications of weed seedbank dynamics to weed management. Weed Science. 45, 329-336.
Carter, D. R., Carter, S., and Allen, J. L. (1994). Submerged macrophyte control using plastic blankets. Water Science Technology 29, 119-126.
Cellot, B., Mouillot, F., and Henry, C. P. (1998). Flood drift and propagule bank of aquatic macrophytes in a riverine wetland. Journal of Vegetation Science. 9, 631-640.
Chokder, A. H. (1967). Biological control of aquatic vegetation. Agriculture Pakistan 18, 225-229.
114
Chung, Y. R., Koo, S. J., Kim, H. T., and Cho, K. Y. (1998). Potential of an indigenous fungus, Plectosporium tabacinum, as a mycoherbicide for control of arrowhead (Sagittaria trifolia). Plant Disease. 82, 657-660.
Collon, E. G., and Velasquez, J. (1989). Dispersion, germination and growth of seedlings of Sagittaria latifolia L. Folia Geobotanica et Phytotaxonomica. 24, 37-50.
Cook, C. D. K. (1990). Seed dispersal of Nymphoides peltata (S.G. Gmelin) O. Kuntze (Menyanthaceae). Aquatic Botany 37, 325-340.
Cother, E. J. (1999). Host range studies of the mycoherbistat fungus Rhynchosporium alismatis. Australasian Plant Pathology. 28, 149-155.
Crump, N. S., Cother, E. J., and Ash, G. J. (1999). Clarifying the nomenclature in microbial weed control. Biocontrol Science and Technology. 9, 89-97.
Eichler, L. W., Bombard, R. T., Sutherland, J. W., and Boylen, C. W. (1995). Recolonization of the littoral zone by macrophytes following the removal of benthic barrier material. Journal of Aquatic Plant Management. 33, 51-54.
Elmore, C. L. (1996). A reintroduction to integrated weed management. Weed Science. 44, 409-412.
Flower, G. E. (2003). The Biology and Ecology of Arrowhead (Sagittaria montevidensis Cham. et Schlecht), a Weed in Rice in NSW. PhD thesis, Charles Sturt University, Wagga Wagga.
Flower, G. E., Pratley, J. E., and Slater, P. D. (1999). Early desiccation as a tool for arrowhead control. In "12th Australian Weeds Conference. Papers and Proceedings." (A. C. Bishop, M. Boersma and C. D. Barnes, eds.), pp. 543-546. Tasmanian Weed Society, Hobart.
Flower, R. (in press). Developing Integrated Weed Management Strategies for Dirty Dora (Cyperus difformis L.). PhD thesis, Charles Sturt University, Wagga Wagga.
Gopal, B., and Goel, U. (1993). Competition and allelopathy in aquatic plant communities. The Botanical Review. 59, 155-210.
Gordon, E. (1996). Tipo de dispersión, germinación y crecimiento de plántulas de Sagittaria latifolia (Alismataceae). Fragmenta Floristica et Geobotanica 41, 657-668.
Graham, R. J. (1999). The Biology and Ecology of Damasonium minus (R.Br.) Buch. - The First Step in Developing an IWM Strategy for the Control of Herbicide Resistant Aquatic Weeds of New South Wales Rice Crops. PhD thesis, Charles Sturt University, Wagga Wagga.
Grime, J. P. (1989). Seed banks in ecological perspective. In "Ecology of Soil Seed banks." (R. L. Simpson, ed.), pp. xv-xxii. Academic Press, Inc., San Diego, California.
Gunasekera, L., and Krake, K. (2001). Arrowhead - a serious aquatic weed in northern Victoria. Victorian Landcare and Catchment Management. 19, 7.
Haraguchi, A. (1993). Phenotypic and phenological plasticity of an aquatic macrophyte Menyanthes trifoliata L. Journal of Plant Research. 106, 31-35.
115
Heylin, E. M. (in press). Developing Control Strategies for Water Plantain (Alisma plantago-aquatica L.). PhD thesis, Charles Sturt University, Wagga Wagga.
Hofstra, D. E., and Clayton, J. S. (2001). Evaluation of selected herbicides for the control of exotic submerged weeds in New Zealand: I. The use of endothall, triclopyr and dichlobenil. Journal of Aquatic Plant Management. 39.
Hroudová, Z., Hrouda, L., Zákravsky, P., and Ostry, I. (1988). Ecobiology and distribution of Sagittaria sagittifolia L. in Czechoslovakia. Folia Geobotanica et Phytotaxonomica. 23, 337-373.
Ishizawa, K., Murakami, S., Kawakami, Y., and Kuramochi, H. (1999). Growth and energy status of arrowhead tubers, pondweed turions and rice seedlings under anoxic conditions. Plant Cell and Environment. 22, 505-514.
Kak, A. M., and Durani, S. (1989). Seed morphology of the family Alismataceae. Journal of Economic and Taxonomic Botany. 13, 501-510.
Kaul, R. B. (1978). Morphology of germination and establishment of aquatic seedlings in Alismataceae and Hydrocharitaceae. Aquatic Botany. 5, 139-147.
Kaul, R. B. (1985). Reproductive phenology and biology in annual and perennial Alismataceae. Aquatic Botany 22, 153-164.
Krake, K., and Breewel, L. (2000). "Research and Development Proposal. “Arrowhead” – Biology and Control" Goulburn-Murray Water, Document No. 434361 – October 2000
Kunii, H. (1993). Rhizome longevity in two floating-leaved aquatic macrophytes, Nymphaea tetragona and Brasenia schreberi. Journal of Aquatic Plant Management. 31, 94-98.
Lieu, S. M. (1979). Growth forms in the Alismatales. I. Alisma triviale and species of Sagittaria with upright vegetative axes. Canadian Journal of Botany 57, 2325-2352.
Loyn, R. H., and French, K. (1991). Birds and environmental weeds in south-eastern Australia. Plant Protection Quarterly. 6, 137-149.
LWRRDC (2002). "Guidelines to Good Practice for the Construction and Refurbishment of Earthen Irrigation Channel Banks," Land and Water Resources Research and Development Corporation, Canberra.
Moss, B. (1988). "Ecology of Freshwaters. Man and Medium," Blackwell Scientific, Oxford.
Pandey, B. P. (1992). "Taxonomy of Angiosperms," 5th/Ed. S. Chand and Company, New Delhi.
Parsons, J. M. (1995). "Australian Weed Control Handbook," 10th/Ed. Inkata Press, Melbourne.
Pollock, D. C. (1992). The Biology and Control of Alisma lanceolatum (With.) in Rice in New South Wales. Master of Rural Science thesis, University of New England, Armidale.
Powers, K. D., Noble, R. E., and Chabreck, R. H. (1978). Seed distribution by waterfowl in southwestern Louisiana. Journal of Wildlife Management. 42, 598-605.
116
Rataj, K. (1972a). Revision of the genus Sagittaria. Part I. (Old World species). Annotationes Zoologicae et Botanicae. 76, 1-36.
Rataj, K. (1972b). Revision of the genus Sagittaria. Part II. (The species of West Indies, Central and South America). Annotationes Zoologicae et Botanicae. 78, 1-61.
Rea, N., and Ganf, G. G. (1994). How emergent plants experience water regime in a Mediterranean-type wetland. Aquatic Botany 49, 117-136.
Sage, L. W., Lloyd, S. G., and Pigott, J. P. (2000). Sagittaria platyphylla (Alismataceae), a new aquatic weed threat in Western Australia. Nuytsia 13, 403-405.
Sainty, G. R., and Jacobs, S. W. L. (1981). "Water Plants Of New South Wales," Water Resources Commission N.S.W., Sydney.
Schenk, E. (1972). Quantitative studies on translocations in Sagittaria graminea Michx. leaves: a new method of measurement. Acta Botanica Neerlandica 21, 231-234.
Spencer, D. F., Ksander, G. G., Madsen, J. D., and Owens, C. S. (2000). Emergence of vegetative propagules of Potamogeton nodosus, Potamogeton pectinatus, Vallisneria americana and Hydrilla verticillata based on accumulated degree-days. Aquatic Botany 67, 237-249.
Stephens, K. M., and Dowling, R. M. (2002). "Wetland Plants of Queensland," CSIRO Publishing, Melbourne.
Szczepanski, A. J. (1977). Allelopathy as a means of biological control of water weeds. Aquatic Botany 3, 193-197.
Turner, C. E. (1981). Reproductive Biology of Sagittaria montevidensis Cham. & Schlecht. ssp. calycina (Engelm.) Bogin (Alismataceae). PhD thesis, University of California, Berkeley.
Zechmeister-Boltenstern, S., and Nikodim, L. (1999). Effect of water tension on ethylene production and consumption in montane and lowland soils in Austria. European Journal of Soil Science 50, 425-432.
117
118
Appendices – detailed write-ups of experiments
Appendix 1 - Herbicide Experiments
Experiment 01 – Glyphosate Concentrations
Aim: To investigate the control of arrowhead using glyphosate at concentrations higher
than the label rates (9L/ha)
Location: McCracken’s Rd, North of Shepparton, GV Water drain.
Treatments:
• Four concentrations of glyphosate 360 - 9, 18, 36 and 72 L/ha – Season 1
• Four concentrations of glyphosate Duo - 9, 18, 36 and 72 L/ha – Season 2
• Four replicates
• Four control plots
• Total 20 plots
Dates / Notes:
Started in 2001/2002 season
Season 1:
First treatments applied 18/01/2002
Assessment made 26/06/2002, and 22/10/2002, beginning next season
Season 2:
Second season treatment applied 04/04/2003
- broadleafed plants, 5cm tall, 10cm water
- 20ºC, wind 0-5km/h SE
Assessment made 07/05/2003
119
Data and other notes:
• Data indicate that there were significant differences in the level of control achieved by
different concentrations of Weedmaster 360 in the first year.
• Although 9L/ha gave some control (significantly different from the Unsprayed Control
plots), the results were not satisfactory.
• Significant differences between the 18, 36 and 72 L/ha plots were not apparent until the
beginning of the following season (22/10/2002), when 72L/ha plots showed significantly
less cover of arrowhead than other plots.
• A large amount of variation in the first year data hinders interpretation of the data
♦ At the time of the Year 2 application, 100% cover of arrowhead was present in all plots
again.
♦ Assessments made of the success of the Year 2 application, using Weedmaster Duo, after
33 days, indicate that there are significant differences again, with 36 and 72 L/ha causing
far more browning than the other treatments.
♦ 9 and 18L/ha both performed poorly, with less than 10% browning in each. 36L/ha
produced significantly more browning and 72L/ha significantly more again.
♦ Variation was much less, although this will probably increase with further assessments at
a later time.
25th September 2003 – differences in plots still visible, but with new growth beginning in
all.
Conclusion – 36 and 72L/ha give good initial control of arrowhead, with regrowth in the
following season. 9 and 18L/ha were not satisfactory. The effective concentrations are,
however, much higher than the label rate.
Control in the second year of application was significantly better at higher concentrations
(36 and 72 L/ha) than the previous year, but significantly weaker in the second year at
lower concentrations (9 and 18L/ha).
120
0
10
20
30
40
50
60
70
80
90
100
Control 9L/ha 18L/ha 36L/ha 72L/ha
Glyphosate rate applied
perc
enta
ge b
row
ning
26-Jun-02
07-May-03
121
Experiment 02 – 2,4-D formulations and timings
Aim: To investigate the effect of time of year on control of arrowhead using three 2,4-D
formulations.
Location: Channel 34/12 – Hick’s Rd, north of Shepparton
Treatments:
• 3 herbicides - AF300, Amicide 625 and Surpass 300 (2,4-D formulations)
• Six times of year
• November
• December
• February
• April
• May
• June
• Four replicates
• Four control plots
• Total of 76 plots
Dates / Notes:
Started in the 2001/2002 season
November spray on 19/11/2001
December spray on 19/12/2001
February spray on 11/02/2002
April spray on 16/04/2002
May spray on 17/05/2002
June spray on 20/06/2002
Assessments were made 26/06/2002 and 22/10/2002
Not continued in 2002/2003 season
Data and other notes:
122
• The data generated from this experiment have high variability, making conclusions on the
data difficult. Data analyses indicated significant differences between herbicides and
between timing, but a significant interaction term also occurred, making interpretation
difficult.
• The data collected in June 2002 indicate that there are significant differences between a
number of the treatments and the unsprayed control plots, with those sprayed in
December and April appearing the most successful.
• The data collected in October 2002 show a very similar pattern, albeit with more
variability.
• Neither set of data shows the effectiveness of 2,4-D formulations very well. Although the
data show significant differences between the unsprayed control plots and the treatments,
the variability means the differences are not very great.
♦ In October 2003, an assessment was made of the plots. All were 100% covered in
arrowhead (reasonably sparse as it was the beginning of the season), except for two plots.
These had both been sprayed with Amicide 625, one in November (10% cover) and one
in December (25% cover), although the other replicates for both treatments were 100%
covered, so it is difficult to draw conclusions from these plots. They are probably just
aberrations.
Conclusion – 2,4-D formulations at label rates do reduce standing biomass of arrowhead
in the year of application, with higher success rates in December and April
123
0
10
20
30
40
50
60
70
80
90
100
Nov-01 Dec-01 Jan-02 Feb-02 Mar-02 Apr-02 May-02 Jun-02
Month of Application
Perc
enta
ge C
over
late
Jun
e 02
Amicide 625AF300Surpass300control
0
10
20
30
40
50
60
70
80
90
100
Nov-01 Dec-01 Jan-02 Feb-02 Mar-02 Apr-02 May-02 Jun-02
Month of Application
Perc
enta
ge C
over
22
OC
tobe
r 03
Amicide 625AF300Surpass300control
124
Experiment 03 – 2,4-D formulations, glyphosate & concentrations & timing
Aim: To investigate the effect on arrowhead control of applying 3 herbicides at 3 different
concentrations at 2 different times in the irrigations season
Location: Channel 6/4/8/6 below wheel 6242, Shinnicks Rd, West of Numurkah
Treatments:
• Three herbicides – AF300, Amicide 625 and Weedmaster 360 (glyphosate)
• Three concentrations – 0.5, 1.0 and 2.0 times the recommended rate for each herbicide.
Rec. rates – AF300 (21 L/ha)
Amicide 625 (10L/ha)
Weedmaster 360 (9L/ha)
• Two times of year (November and February)
• Four replicates
• Four control plots
• Total of 76 plots
Dates / Notes:
Started in the 2001/2002 season
Season 1:
November spray on 28/11/2001
February spray on 11/02/2002
Season 2:
Blanket application of 2,4-D occurred in December 2002 for 2002/03 season
125
Data and other notes:
• Data indicate that initially all treatments had significantly less arrowhead cover than the
unsprayed control plots.
• Amicide 625 February application was initially the most effective. Overall, glyphosate
was the least effective treatment.
• By October 2002, the variability in data meant that no significant differences could be
found between treated plots and the unsprayed control plots.
• This variability arises as arrowhead populations recover, with re-growth from
underground biomass and re-invasion by seed.
• The lack of good initial control with 2,4-D in this trial does not concur with G-MW
experience, however the re-growth of arrowhead in the following season is common.
• Blanket application with 2,4-D in following season gave good control, still no re-growth
by October 2003.
Conclusion – 2,4-D formulations reduce standing biomass of arrowhead, but populations
recover.
Comparing this with other trials, it becomes apparent that the two timings used for this
trial are not optimal for the control of arrowhead.
0
10
20
30
40
50
60
70
80
90
100
control 0.5 1 2
rate (x recommended rate)
perc
enta
ge c
over
Oct
200
2
AF300 DecAmicide 625 DecGlyphosate 360 DecAF300 FebAmicide 625 FebGlyphosate 360 Feb
126
Experiment 04 – Glyphosate formulations and concentrations
Aim: To investigate the effect on arrowhead control of two different glyphosate
formulations applied at rates higher than the recommended rate (9L/ha)
Location: Drain 20, North of Numurkah
Treatments:
• Two herbicides – Weedmaster 360 and Weedmaster Duo
• Four concentrations (9, 18, 36 and 72 L/ha)
• Four replicates
• Four control plots
• Total of 36 plots
Dates / Notes:
Started in 2001/2002 season
Season 1:
Sprayed on 27/05/2002
Season 2:
Blanket application of Amitrole T and Glyphosate Duo occurred November 2002 for 2002/03
season
Another blanket spray occurred 01/04/2003, using (20L Amicide 625, 15L Amitrole T and
10L Weedmaster Duo) in 1300L of water
Data and other notes:
• Initially, all plots were showing potential for control
• After initial spray, there was an unknown influence on the drain, and virtually all plants
disappeared, even in unsprayed control plots.
• This is one danger of experiments in drains, where it is not always known what will
happen next.
127
• By November 2002, the arrowhead was thick and dense again, and Murray Valley IA
contractors sprayed the drain. It was all broad-leafed arrowhead
• Again by March 2003, broad-leafed arrowhead regrowth necessitated an application of
herbicide.
• No regrowth October 13th 2003
Conclusion – regrowth of arrowhead in drains is rapid and almost always of the broad-
leafed variety
Arrowhead re-growth in drains more likely to be from seed – hence rapid re-growth of
broad-leafed form
Drain 20, near Numurkah - recorded July 26 2002
0
10
20
30
40
50
60
70
80
90
100
9L/ha 18L/ha 36L/ha 72L/ha control
rate
perc
enta
ge b
row
ning
Glyphosate 360Weedmaster Duo
128
Experiment 05 – 2,4-Ds, glyphosate & time of day
Aim: To investigate the effect on arrowhead control of the application of 4 herbicides at
different times of day
Location: Channel 6/6, below wheel 6073, Naring Hall Rd north of Numurkah
Treatments:
• Four herbicides – AF300, Amicide 625, Surpass 300 and glyphosate 360
• Two times of day – early in the morning and at 2 o’clock in the afternoon
• Four replicates each
• Four control plots
• Total of 36 plots
Dates / Notes:
Started in 2001/2002 season
Afternoon spray completed 20/12/2001, 3pm
Morning spray completed 21/12/2001, 7am
Assessed 23/01/2002, 27/03/2002, 26/07/2002 and 2/10/2002
Not continued in 2002/2003 season
Data and other notes:
• Data indicate that 2,4-D formulations give excellent initial arrowhead control,
significantly better than glyphosate 360
• Control with 2,4-D formulations in this trial was significantly better than others
• March data begin to show the re-growth or re-infestation of arrowhead, with all but one
treatment (glyphosate 360 applied in afternoon) showing significantly greater arrowhead
cover by March.
129
• October data indicate that, a short time into the next season (2002/03 season), arrowhead
levels were back to almost 100% cover in all treatments.
♦ October 2003 – all plots 100% cover of arrowhead, although density is only 80% of that
in surrounding areas that were not sprayed at all.
Conclusion – 2,4-D formulations can provide excellent initial arrowhead control, to
maintain hydraulic capacity in the short term, although populations will recover given
time.
0
10
20
30
40
50
60
70
80
90
100
Control AF300 early AF300midday
Amicide 625early
Amicide 625midday
Surpass 300early
Surpass 300midday
Glyphosateearly
Glyphosatemidday
Herbicide Treatment
Arr
owhe
adco
ver (
%)
Percent cover JanuaryPercent cover MarchPercentage Cover JulyPercentage Cover October
130
Experiment 06 – glyphosate concentrations & time of year
Aim: To investigate the effect on arrowhead control of glyphosate applied at higher than
the recommended rate (9L/ha) and at three different times of year
Location: Main number 5 channel, between Wheels 5165 and 5166, upstream of Berry’s Rd,
North of Numurkah
Treatments:
• Four rates of glyphosate 360 (9, 18, 36 and 72 L/ha) year One
• Four rates of glyphosate Duo (9, 18, 36 and 72 L/ha) year Two
• Three times of year (December, March and June)
• Four replicates
• Four control plots
• Total of 52 plots
Dates / Notes:
Started in 2001/2002 season
Season 1:
December spray on 20/12/2001
March spray on 14/03/2002
June spray on 20/06/2002
Season 2:
Half of the December plots sprayed again on 13/12/2002
Half of the March plots sprayed again on 25/03/2003
Half of the June plots were sprayed again on 18/06/2003
Season 3:
This trial was sacrificed in August 2003, in order to set up Experiment 24
131
Data and other notes:
• First year’s data indicate that control of arrowhead with Weedmaster 360 increases with
concentration of application, particularly at high concentrations.
• Control was best in March, with plots having significantly less cover in these plots than in
the plots sprayed in December for all concentrations of glyphosate, and significantly less
cover than the plots sprayed in June, for the two highest concentrations of glyphosate.
• Reasonably good control with the June application of glyphosate was unexpected, as it
was expected the plants would not be metabolising as well at this time of year and would
thus not be as well controlled.
• However, one possibility to explain the apparent good result compared to other treatments
may be that, by October, the December and March treated plots may have had more time
to recover.
♦ 29th August 2003 – no regrowth yet. Small seedlings (around 10-20mm high) on bare
patches.
♦ 13th October 2003 – some regrowth, though reasonably sparse, covering 100% of all
plots, except 36L/ha plots and 72L/ha plots (all between 0 and 10% cover)
Conclusions - 36 and 72L/ha give good initial control of arrowhead, with regrowth
occurring in the following season. 9 and 18L/ha did not perform as well. The effective
concentrations are, however, much higher than the label rate.
Effective control was best in March. This concurs with field observations that suggest
that arrowhead is growing most vigorously in March-April.
Assessment at beginning of following season suggested June spray was also effective at
higher rates (36 and 72 L/ha)
Similar results beginning to show in 2nd season
132
0
10
20
30
40
50
60
70
80
90
100
control Dec-01 Mar-02 Jun-02
month of application
cove
r Oct
02
(%)
9L/ha18L/ha36L/ha72L/ha
133
Experiment 07 – Amicide625, Glyphosate & repeats within a season
Aim: To investigate the effect on arrowhead control of the application of two herbicides
with or without repeats during the season.
Location: Channel 2A/2/5, above wheel 5101, cnr. of Lorenz and Pinnucks Rds, North of
Numurkah
Treatments:
• Two herbicides – Amicide 625 and Weedmaster 360 – label rates in 2001/02 season (10
and 20 L/ha in 2002/03 season)
• Four repetition treatments – Season 1:
1. January, April and June
2. January and April
3. January and June
4. January alone
• Four repetition treatments – Season 2:
5. February, April and May
6. February and April
7. February and May
8. February alone
• Four replicates
• Four control plots
• Total of 36 plots
134
Dates / Notes:
Started in the 2001/2002 season
Season 1:
January spray on 18/01/2002
April spray on 17/04/2002
June spray on 21/06/2002
Assessed – 26/06/2002 and 22/10/2002
Season 2:
February spray 13/02/2003 – Temperature 35°C, little wind
April Spray 04/04/2003 - 18°C, 5-10km/h wind
May Spray 26/05/2003 - 14°C, little wind
Assessed - 10/06/2003
Data and other notes:
• Data from Year One indicate that control with Amicide 625 is significantly better than
control with glyphosate 360 in most cases. This is particularly marked when data
collected at the beginning of the following season are compared.
• October data also indicate that there is little significant difference between the three
treatments that were applied more than once.
• Treatments applied in January, however, had a percentage arrowhead cover not
significantly different from the unsprayed control plots, indicating that the April and June
sprays were the timings that significantly effected the success of control
• This is in accordance with the control achieved with elevated rates of glyphosate in
March and June, and again, the April spray corresponds with the time that arrowhead is
growing most aggressively.
♦ Year Two data - assessed 10th June 2003
♦ 13th October 2003 – channel still dry, no visible regrowth of arrowhead in dry conditions
135
♦ 30th October 2003 – water level in channel very high. Some arrowhead plants beginning
to show above water level. These plants occur mainly in those plots that were sprayed
once, particularly with glyphosate, and in the unsprayed control plots.
Conclusion – a single application of Amicide in January does not give sustained control,
however, when combined with a second application, at an appropriate time of year,
control is much improved.
Comparing this with other 2,4-D trials, where initial control was followed by intense re-
growth such that infestations were back at high levels by the following October, the
success of multiple applications, even when assessed as late as the following October,
indicate that they achieve better control than a single application in the “focus” months of
March-June.
Three applications did not give significantly better control than two applications and is
therefore not justified financially.
0
10
20
30
40
50
60
70
80
90
100
control Jan, Apr & Jun Jan & Apr Jan Jan & Jun
treatment times
% c
over
Oct
ober
200
2
Amicide 625
glyphosate
136
0
10
20
30
40
50
60
70
80
Control February February April andMay
February and April February and May
application times
% s
urvi
val
Oct
ober
200
3Amicide 625
Glyphosate Duo
137
Experiment 08 – Amicide625, Glyphosate & repeats within a season
Aim: To investigate the effect on arrowhead control of the application of two herbicides
with or without repeats during the season.
Location: Channel 3/7/3, above wheel 3135, cnr. of Learmont and Lukies Rds, near
Katamatite
Treatments:
• Two herbicides (Amicide 625 and Weedmaster 360)
• Four repetition treatments
9. December, February and May
10. December and February
11. December and May
12. December alone
• Four replicates
• Four control plots
• Total of 36 plots
Dates / Notes:
Started in the 2001/2002 season
Season 1:
December spray on 20/12/2001
February spray on 12/02/2002
May Spray on 14/05/2002
Season 2:
2002/2003 season – blanket spray with Weedmaster Duo @ 25L/ha on 13/02/2003 –
Temperature 31°C, little wind
138
Data and other notes:
• The data from this trial show a similar trend to those from Experiment 07.
• Control with Amicide is significantly better than with glyphosate.
• Multiple applications of Amicide tend to give more satisfactory results, although by the
following October, the differences are more difficult to identify.
• The months of application in this trial were slightly different to those in Experiment 07.
Data taken the following October indicate the best control was achieved in treatments
including a May application. These had significantly lower coverage of arrowhead by
October than the December and “December and February” treatments.
• When compared to the data from Experiment 07, this suggests that February is too early,
with respect to the “focus” months of arrowhead growth, but the May application did fall
into this period.
♦ Data from Year Two, when a blanket application of glyphosate occurred in order to see if
any differences between treatments could be focussed, are as yet unavailable.
♦ 13th October 2003 – water level high, re-growth not visible.
Conclusions – As for Experiment 07, multiple applications of Amicide result in better
arrowhead control.
Three applications do not give significantly better results than two applications, however
the timing must be correct.
Results highlight the “focus” months of arrowhead growth.
The results also highlight the fact that glyphosate is a much slower working herbicide
than 2,4-D, with better results with time occurring with glyphosate, where 2,4-D results
remained stable.
139
0
10
20
30
40
50
60
70
80
90
100
control Dec, Feb & May Dec & Feb Dec Dec & May
Time of herbicide application
% c
over
Oct
ober
200
2Amicide 625
glyphosate
140
Experiment 09 – Residual herbicides for arrowhead control
Aim: To investigate the effect on arrowhead control of five herbicides, most with residual
control
Location: Farmer’s channel, near Kerang
Treatments:
• Five herbicides (Amicide 625, Casoron G, Londax, Simazine and Brushoff)
• Four replicates
• Four control plots
• Total of 24 plots
Dates / Notes:
Started between 2001/02 and 2002/03 seasons
Set up on 13/06/2002
Assessed 24/10/2002, January 2003 and 23/06/2003
Data and other notes:
• The data indicate that Brushoff, Casoron G and Londax give excellent control of
arrowhead at the beginning of the following season (all 100% control), when applied in
the winter, through a suppression of emergence that lasts for at least 12 months.
• Amicide 625 also gives good control (still 95% control by October 2002).
• Simazine gave disappointing control, with an average of 64% by October, ranging from
15 to 100 percent between plots.
• Further assessment later in the irrigation season indicated control with Casoron G was
still very good, whilst the pattern of regrowth after Amicide application is repeated.
• Following application of a blanket cover of Casoron G at approximately 300kg/ha
(22/07/2003), an inspection of the channel on 14/10/2003 revealed one small arrowhead
plant had emerged in the length of the channel.
141
Conclusion – Residual herbicides provide good, lasting control of arrowhead, when
applied over winter
Londax suppresses weeds in rice bays when applied pre- or early post-emergence. In rice
systems, water is required to be held for 4 days before release.
Casoron G shows most potential
Registration for these herbicides a hurdle in irrigation systems, Casoron G probably has
the most chance.
• 26/02/2003 – individual plants in the Control plots were treated with herbicide to assess
the effect of herbicides on different forms. No data were produced.
Peg/Plant
No.
Form Herbicide
Used
Peg No. Form Herbicide
Used
1 Submerged
Rosette
Weedmaster
Duo 20L/ha
28 Submerged
Rosette
Weedmaster
Duo 40L/ha
2 29 (continued)
3 30
4 Submerged
Stalky
31 Submerged
Stalky
5 32
6 33
7 Broadleaf
34 Broadleaf
8 35
9 36
10 Submerged
Rosette
37 Submerged
Rosette
Amicide 500
12L/ha
11 38
12 39
13 Submerged 40 Submerged
142
Stalky Stalky
14 41
15 42
16 Broadleaf
43 Broadleaf
17 44
18 45
19 Submerged
Rosette
Weedmaster
Duo 40L/ha
46 Submerged
Rosette
20 47
21 48
22 Submerged
Stalky
49 Submerged
Stalky
23 50
24 51
25 Broadleaf
52 Broadleaf
26 53
27 54
143
0
1
2
3
4
5
6
7
8
9
10
Control Amicide 625 Brushoff Casoron Londax Simazineherbicide used
cove
r rat
ing
(out
of t
en)
Oct-02Jan-03Jun-03
144
Experiment 10 – A comparison of the effect of Amitrole versus Amitrole and Glyphosate
with and without follow-up
Aim: To investigate the effect on arrowhead control of Amitrole T and Amitrole T +
Glyphosate Duo when followed by a second treatment 6-8 weeks later, versus no follow-up
Location: Ardmona Main Drain II, east of Tatura
Treatments:
• Two herbicide mixes – Amitrole T and Amitrole T + Weedmaster Duo
• Two follow-up treatments - follow-up after 13 weeks versus no follow-up
• Four replicates
• Four control plots
• Total of 20 plots
Dates / Notes:
Started in the 2002/2003 season
Sprayed 11/02/2003 afternoon – Temperature 34°C, very little wind
Sprayed again 16/05/2003 (94 Days after first application = 13 weeks and three days) –
Temperature 20°C, no wind, some water in drain from recent rainfall
The Amitrole label recommends the following:
A further application 6-8 weeks after the initial application, for the control of
water couch
Application during flowering period between January and May for the control
of cumbungi, phragmites and nutgrass
Data and other notes:
• After the first treatment, the second treatment was initially abandoned, as there appeared
to be no arrowhead growing in the drain after 6 weeks
• After 10 weeks, an inspection identified that regrowth of arrowhead had occurred
• After 13 weeks from the initial application, it was decided to apply a second treatment on
the appropriate plots
145
• This was done in the middle of May and also should give some indication of the efficacy
of an Amitrole/Weedmaster mix at this time of year.
Conclusion – this trial again shows the problems of conducting experiments in drains (see
Experiment 04), as variable results appear to have resulted from some external influence
Amitrole worked well where results could be seen, but a large reduction in arrowhead in
the control plots makes conclusions difficult.
0
10
20
30
40
50
60
70
80
90
100
Control Amitrole T (one appl) Amitrole T +Weedmaster Duo(one
appl)
Amitrole T (two appl) Amitrole T +Weedmaster Duo (two
appl)
herbicide treatment
perc
enta
ge c
ontr
ol
13 DAT117 DAT2
146
Experiment 11 – A comparison of the effect of Amitrole versus Amitrole and Glyphosate
with and without follow-up
Aim: To investigate the effect on arrowhead control of Amitrole T and Amitrole T +
Glyphosate Duo when followed by a second treatment 6-8 weeks later, versus no follow-up
Location: Drain 13, off Centre Rd. and towards Boothroyd’s Rd., north-west of Numurkah
Treatments:
• Two herbicide mixes – Amitrole T and Amitrole T + Weedmaster Duo
• Two follow-up treatments - follow-up after 6 to 8 weeks versus no follow-up
• Four replicates
• Four control plots
• Total of 20 plots
Dates / Notes:
Started in the 2002/2003 season
Sprayed 11/02/2003 late morning – Temperature 34°C, very little wind
Sprayed again 26/03/2003 – (6 weeks and 1 day) - Temperature 18°C, 5km/h wind
Assessed 14/04/2003 and 07/05/2003
Data and other notes:
• Data analyses highlight the obvious success of both herbicide treatments in the initial
weeks following their application
• Two weeks after the applications, data suggest no significant difference in control
between one application of herbicide and two applications.
• They do suggest, however, that one application of Amitrole T + Weedmaster Duo
achieves better control than Amitrole alone, applied once or twice.
• 37 days after treatment, however, the differences between treatments have begun to
disappear. Although Amitrole T appears to work better in mixture with Weedmaster Duo,
the differences are not statistically different.
• 29th August 2003 (86 DAT) – lots of grass growth in plots and some re-growth from
corms in unsprayed areas
Conclusion – In the first few weeks after application, Amitrole T by itself and in mixture
with Weedmaster Duo, provide good control of arrowhead.
147
Visible signs of control seem to take longer to appear than control using 2,4-D
formulations, but initial control is good.
0
10
20
30
40
50
60
70
80
90
100
Unsprayed Control Amitrole T (one application) Amitrole T + WeedmasterDuo (one application)
Amitrole T (twoapplications)
Amitrole T + WeedmasterDuo (two applications)
herbicide treatment
% c
ontr
ol
% Control 13 DAT
% Control 37 DAT
% Control 65 DAT
148
Experiment 12 - Comparison of effects of four herbicides on arrowhead control
Aim: To investigate the effect on arrowhead control of 3 herbicides and one herbicide mix.
Location: 16/6 spur, between wheels 6724 and 6725, cnr. of Gardiner’s Rd. and
Nathalia/Waaia Rd., west of Waaia
Treatments:
• Four herbicide treatments all at recommended rates
• Amicide 625
• Surpass 300
• Weedmaster Duo
• Weedmaster Duo + Surpass 300
• Four replicates
• Four control plots
• Total of 20 plots
Dates / Notes:
Started in 2002/2003 season
Sprayed on 05/03/2003
Assessed 14/04/2003, 42 days after treatment (water level high, plants 10-30cm out of water)
Data and other notes:
• 42 days after treatment, all single herbicide treatments gave adequate arrowhead control
(between 74 and 88 percent browning).
• The Mixture of Surpass 300 and Weedmaster Duo gave less satisfactory result (12.5
percent browning).
• 14th August 2003 – unlike some other trial sites, no seedling plants present
Conclusions – 2,4-D formulations give good initial control.
149
In this situation, label rates of Weedmaster Duo have worked well. This contrasts with
results from label rates of glyphosate in other trials.
On the evidence here, perhaps there is some antagonism between Surpass 300 and
Weedmaster Duo.
Regrowth of arrowhead from these treatments will be followed.
0
10
20
30
40
50
60
70
80
90
100
Control Amicide 625 Surpass 300 Weedmaster Duo Weedmaster Duo +Surpass
Herbicide Treatment (all at label rates)
Perc
enta
ge C
ontr
ol
42 DAT93 DAT164 DAT
150
Experiment 13 – The effect of different herbicides, rates and ambient temperature at
time of application
Aim: To investigate the effect on arrowhead control of three different herbicide mixes,
applied at label rates and higher rates, when applied at high temperature and a lower
temperature
Location: Main No. 6 Channel, above wheels 6715 and 6716, adjacent to western side of
Waaia – Bearii Rd, just south of Waaia at Tiger Ranch.
Treatments:
• Three herbicide treatments
• Weedmaster Duo alone
• Weedmaster Duo + Amicide 625
• Weedmaster Duo + Surpass 300
• Two rates – label rates versus elevated rates
• Weedmaster Duo – 9 and 40 L/ha
• Amicide 625 – 10 and 20 L/ha
• Surpass 300 – 20 and 40L/ha
• Two ambient temperature regimes, High and Low
• Four replicates
• Four control plots
• Total of 52 plots
Dates / Notes:
Started in 2002/2003 season
High temperature plots sprayed 14/02/2003 – Temperature 37°C, very little wind
Lower temperature plots sprayed 19/02/2003 - Temperature 22°C, some wind
151
Data and other notes:
• 54 Days After Treatments - Data indicate that control is quite good with all herbicide
applications except a low dose of Weedmaster Duo
• Data from most treatments had large degrees of variability, making it difficult to interpret
data.
• For all but the Duo and Surpass low dose treatment, there were no significant differences
between data for plots treated at high temperature and those treated at low temperature
• Best treatments were a high dose of Weedmaster Duo (40L/ha) at high temperature and a
low dose of a mix of Weedmaster Duo and Surpass 300 at a low temperature
• 85 Days After Treatment, Weedmaster Duo by itself, without 2,4-D in the mix, starting to
look more effective. Possible that the fast action of 2,4-D prevents Weedmaster from
working properly, as it’s action is slower
• 2nd October 2003 (225DAT) – observation: new arrowhead coming through for this
season, difficult to see any differences between plots. All seem 100% covered in
arrowhead growth
Conclusion – No differences between effectiveness with different temperatures
2,4-D gives effective initial control, glyphosate gives control, but takes longer to show
0
10
20
30
40
50
60
70
80
90
100
Control Weedmaster DuoLow Dose
Weedmaster DuoHigh Dose
Duo and Amicide625 Low Dose
Duo and Amicide625 High Dose
Duo and Surpass300 Low Dose
Duo and Surpass300 High Dose
herbicide and dose
perc
enta
ge c
ontr
ol54
DA
T
High TempLow temp
152
Experiment 14 – The effect of glyphosate formulations, concentrations and timings
Aim: To investigate the effect on arrowhead control of two different glyphosate
formulations applied at rates higher than the recommended rate (9L/ha) and at two times of
year
Location: Channel 3/4/8/6, between wheels 6234 and 6235, adjacent Saxton St West
(formerly Hodge’s Rd), West of Numurkah
Treatments:
• Two herbicides – Weedmaster 360 and Weedmaster Duo
• Three concentrations - 9, 25 and 50 L/ha)
• Two times of year – January and March
• Four replicates
• Four control plots
• Total of 52 plots
Dates / Notes:
Started in 2002/2003 season
January spray on 17/01/2003
March spray on 24/03/2003
Assessed 14/04/2003
Data and other notes:
• Initial indications were that all the treatments had been unsuccessful and a spray crew
from Murray Valley IA was allowed to spray with 2,4-D, in order to clear the channel for
water movement
• Around this time, control with glyphosate had begun to work slightly, however the
variation between replicates was too great to make any conclusions
• Some treatments had more arrowhead cover than unsprayed control plots
153
♦ Trial was assessed after the application of 2,4-D. Control was better in all plots at this
time, as would be expected.
♦ Control with 2,4-D was not as good with 2,4-D in the plots that had been sprayed in
March. Two possible explanations for this:
♦ The data taken 21DAT indicated that the January treatment with glyphosate had worked
better, with an application of 2,4-D on top of that giving better results
♦ Control with glyphosate takes up to seven weeks. The results taken 3 weeks after the
application of the March glyphosate treatment may not have reflected the success of that
treatment. Other trials indicate March is a good month for control with glyphosate. By
the time the blanket spray of 2,4-D was applied, the March treatment may have been
showing better results. If control in March had been better than control in January, there
may have been less contact of 2,4-D on arrowhead plants, so that re-growth of plants
sometime after the application of 2,4-D may have been stronger in those plots – that is,
the “January” plots may have received two doses of (different) herbicides, where the
“March” plots may have missed out on the 2,4-D treatment.
0
10
20
30
40
50
60
70
80
90
100
Control 9L/haWeedmaster
360
25L/haWeedmaster
360
50L/haWeedmaster
360
9L/haWeedmaster
Duo
25L/haWeedmaster
Duo
50L/haWeedmaster
Duo
herbicide treatment
perc
enta
ge c
ontr
ol
January (21DAT)
March (21 DAT)
January (101DAT)
March (101DAT)
154
Experiment 15 – The effect of an adjuvant on effectiveness of an Amicide/Glyphosate
mix
Aim: To investigate the effect on arrowhead control of a mix of Amicide 625 and
Weedmaster Duo with or without the adjuvant Liaise
Location: Main No. 6 Channel, above wheels 6715 and 6716, adjacent to western side of
Waaia – Bearii Rd, just south of Waaia at Tiger Ranch.
Treatments:
• 2 herbicide treatments – Weedmaster Duo versus Weedmaster Duo + Liaise
• Four replicates
• Four control plots
Dates / Notes:
Started in 2002/2003 season
Sprayed on 03/03/2003 - Amicide 625 was not added to the mix as it was decided that the
2,4-D induced abscission of the plants would hinder the ability to measure differences in
brown-out with and without Liaise.
Assessed 14/04/2003, 04/06/2003 and 03/07/2003
Data and other notes:
• The assessment on 14/04/2003 suggested that the treatment hadn’t worked at all.
Subsequent assessments suggested some efficacy.
Conclusions – there was no effect of the addition of the adjuvant, Liaise, on effectiveness
of Weedmaster Duo for arrowhead control
Data from 03/07/2003 indicate both treatments achieved significantly more browning
than the control, and no significant difference between the “Weedmaster Duo + Liaise”
and the “Weedmaster Duo alone” treatment (see graph).
155
Weedmaster Duo seemed to have given very poor control, but control had improved by
May, suggesting that glyphosate control takes a long period of time, in contrast to control
with 2,4-D
Despite significance, differences do not show promise, as control was very limited. This
confirms the inadequacy of glyphosate at 9L/ha for arrowhead control.
0
5
10
15
20
25
Control Weedmaster Duo + Liaise Weedmaster Duo alone
herbicide treatment
perc
enta
ge b
row
ning
91 DAT120DAT
156
Experiment 16 – Glyphosate Concentrations
Aim: To investigate the control of arrowhead using Weedmaster Duo at concentrations
higher than the label rates (9L/ha)
Location: Drain 13, off Centre Rd. and towards Boothroyd’s Rd., north-west of Numurkah
Treatments:
• Three concentrations of Weedmaster Duo– 9, 25 and 50 L/ha
• Four replicates
• Four control plots
• Total 16 plots
Dates / Notes:
Started 2002/2003 season
Sprayed 20/03/2003
Data and other notes:
• Control with 50L/ha of Weedmaster Duo is significantly better than control with 25L/ha,
which is significantly better than 9L/ha
• Control takes some time to manifest itself, with control data significantly improved 48
Days After Treatment (DAT) over 25 DAT
• 29/08/2003 (162DAT) – some arrowhead surviving winter frosts in control plots where
there was no control. Sprayed plots contain no arrowhead. Arrowhead regrowth from
corms occurring in unsprayed control plots, between frosted stems.
Conclusion – control of arrowhead with Weedmaster Duo (glyphosate) is greater at
higher doses, although control at lower doses improves with time
Successful control of arrowhead with glyphosate reduces re-growth from corms
157
0
10
20
30
40
50
60
70
80
90
100
Control 9L/ha 25L/ha 50L/ha
glyphosate concentration applied
arro
whe
adco
ver (
%)
assessed 25 days after treatment (DAT)48DAT76DAT
158
Experiment 17 – Glyphosate and 2,4-D efficacy on arrowhead plants on channel berm
Aim: To investigate the control of arrowhead plants in larger plots (c.f. small plots for
earlier trials) on the berm of a larger channel.
Location: cnr Katamatite to Yarrawonga Rd and Grinter’s Rd
Treatments:
• Amicide 500 LO (2,4-D) 12.5L/ha (plot 1)
• Weedmaster Duo (glyphosate) 9L/ha (plot 2)
• Weedmaster Duo 36L/ha (plot 3)
• No replication
Dates / Notes:
Started 2002/2003 season
Sprayed 01/04/2003
Low wind, 27°C, 10% cloud cover
Plot size – 4m across x 15m along channel.
Emergent plants exposed 20-30cm above water surface
Many submerged rosettes – no herbicide contact with these plants
159
Data and other notes:
• Some control with 2,4-D and higher rate of glyphosate
• Differences not visible with regrowth of plants
• Regrowth aided by the presence of so many rosettes that were not contacted with the
herbicide
• Regrowth also aided by extensive rhizomatous material.
Conclusion – although contact herbicides remove some of the erect plants, the presence of
regenerative material allows rapid re-colonisation
160
Experiment 18 – informal investigation of Casoron G applied to arrowhead
Aim: to investigate the effect of Casoron G on mature , emergent arrowhead plants
Location: Yarrawonga Main Channel, 10km East of Cobram, at flow measure shed.
Treatments:
• Casoron G (dichlobenil) at 170kg/ha (lower end of label recommendation)
• Casoron G at 230kg/ha (higher end of label recommendation)
• No replication
Dates / Notes:
Started 2002/2003 season
Casoron G applied 13/06/2003
Wind 5-10km/h across plots (from NW), 12°C, no cloud cover
Erect plants, all exposed
Data and other notes:
Control of mature plants not particularly successful with Casoron G
This may be due to the action of Casoron G, which is to suppress emergence, rather than
kill what’s already mature
161
Experiment 19 – informal investigation of Casoron G applied to arrowhead
Aim: to investigate the effect of Casoron G on mature, emergent and rosette arrowhead
plants
Location: cnr Katamatite to Yarrawonga Rd and Grinter’s Rd
Treatments:
• Casoron G (dichlobenil) at 170kg/ha (lower end of label recommendation)
• Casoron G at 230kg/ha (higher end of label recommendation)
• No replication
• Plot size: 4m wide x 25m along channel
Dates / Notes:
Started 2002/2003 season
Treatments applied 13/06/2003
No wind, no cloud cover, 12°C
Rosette and erect narrow-leafed plants present, exposed (not under water)
14/08/2003:
• Many frosted rosettes present
• No seedlings visible
• Rushes compete with arrowhead, not as many arrowhead in dense rush
infestations
02/10/2003 – no living arrowhead present in either Casoron G plot, many new plants arising
in untreated sections. Some up to 30cm high.
162
Data and other notes:
• The efficacy of Casoron G in controlling established plants is difficult to infer from this,
due to the frosting-off of established plants when exposed over winter
• October results indicate, however, that frosted off material or new plants don’t re-
establish in plots treated with Casoron G.
Conclusion – while the effect of frost means it is difficult to establish the effect of
Casoron G in controlling established plants in this trial, success in suppressing
establishment is clear.
163
Experiment 20 – The effect of channel water height on the efficacy of herbicide on
arrowhead
Aim: To determine if herbicide application when the channel is drawn down offers better
control than when the channel is at an operational level.
Location: “Summerfields”, on the Katamatite-Yarrawonga Rd, just west of Grinter’s Rd –
channel runs parallel to the Y-K road.
Treatments:
Plot number Herbicide Rate
1 Weedmaster Duo 9L/ha
2 Weedmaster Duo 36L/ha
3 Amicide 500 LO 12.5L/ha
4 Casoron G
5 control
6 Weedmaster Duo 9L/ha
7 Weedmaster Duo 36L/ha
8 Amicide 500 LO 12.5L/ha
9 Casoron G
10 control
164
• Different coloured pegs were placed next to individual plants in an attempt to follow their
progress
Peg colour Plant phenotype
White Mature
Green Juvenile rhizomatous plant
Red Thread-like seedling
Yellow Rosette
Dates / Notes:
Started 2002/2003 season
Herbicide applied 01/04/2003
Wind 0-5km/h from North, 26°C, 10% cloud cover
Data and other notes:
2,4-D suppression poor, as no water covering plants (see section on 2,4-D concentrations
in water)
165
Experiment 21 – The effect of water height on arrowhead control
Aim: To determine the effect of water height (supply level vs. empty channel) on
arrowhead control using glyphosate and 2,4-D
Location: Channel 15B/6, south of wheel 6709
Treatments:
Plot
No.
Herbicide used Herbicide rate Water
height
1 Weedmaster Duo (glyphosate) 9 L/ha High
2 Weedmaster Duo + Liaise
adjuvant
9 L/ha + 2% Liaise High
3 Weedmaster Duo 36 L/ha High
4 Weedmaster Duo + Liaise 36 L/ha + 2% High
5 Amicide 500 LO 12.5 L/ha High
6 Weedmaster Duo (glyphosate) 9 L/ha Low
7 Weedmaster Duo + Liaise
adjuvant
9 L/ha + 2% Low
8 Weedmaster Duo 36 L/ha Low
9 Weedmaster Duo + Liaise 36 L/ha + 2% Low
10 Control - -
11 Amicide 500 LO 12.5 L/ha Low
• Plot size was 15m x 5m
166
Dates / Notes:
Started 2002/2003 season
Low water plots were sprayed on 05/03/2003
Wind 2-7km/h S, 24°C, no cloud cover
High water plots were sprayed 20/03/2003
Wind 5-10km/h S, 17°C, 60% cloud cover
Data and other notes:
Plot No. % control (13/06/2003) % control (14/08/2003)
1 0 0
2 0 0
3 5 20
4 5 25
5 100 40 (green rosettes ready
to grow)
6 10 20
7 10 20 (frosted off, looks
dead)
8 70 90
9 70 100
10 0 0
11 100 100 (many small
seedlings)
Data indicate that 2,4-D treatment highly effective, initially
Lower water increases efficacy by between 20 and 60%, particularly of glyphosate (see
Figure 4.4)
• Many small seedlings had started to grow on bare patches by 14/08/2003
167
0
10
20
30
40
50
60
70
80
90
100
Control Weedmaster Duo -9L/ha
Weedmaster Duo -9L/ha + Liaise
Weedmaster Duo -36L/ha
Weedmaster Duo -36L/ha + Liaise
Amicide 500 LO -12.5L/ha
herbicide treatment
perc
enta
ge c
ontr
ol(1
4/08
/200
3)
water up
water down
168
Experiment 22 – Control of Arrowhead Seedlings
Aim: To investigate the control of young arrowhead seedlings using Weedmaster Duo and
Casoron G at 3 rates each
Location: Yarrawonga Main Channel (YMC)
Treatments:
• Three concentrations of Weedmaster Duo – 4.5, 9 and 40 L/ha
• Three concentrations of Casoron G – 23, 50 and 230 kg/ha
• No replication, simple trial run
• One control plots
• Total 7 plots
Plot Plan:
Plot No. Herbicide Applied Rate
1 Casoron G 23 kg/ha
2 Glyphosate 4.5 L/ha
3 Glyphosate 40 L/ha
4 Casoron G 50 kg/ha
5 Casoron G 230 kg/ha
6 Unsprayed Control USC
7 Glyphosate 9 L/ha
Dates / Notes:
• Started August 2003:
• Casoron G applied 18/08/2003 - 15ºC, windy and sunny, so glyphosate not sprayed
• Glyphosate applied 19/08/2003
• 12th September 2003:
169
• Damage to arrowhead seedlings in all plots except Unsprayed Control
• Good kill of seedlings at all glyphosate rates, Casoron G at 230kg/ha. Lower Casoron G
rates not quite as good result, but better than USC.
• Emergence of new arrowhead from sources other than seed (e.g. rhizomes, corms) seems
to be inhibited in Casoron G plots, at all rates.
• 30/10/2003 – all plots examined. Many broad-leafed arrowhead plant to about 30cm tall
growing in unsprayed control plot, and unsprayed area immediately east of trial plots, but
no plants visible in any of the plots, except some at the edges, where they may be the
result of being missed by herbicide, and along high point of bank at edge of berm (see
photos)
Background:
• Rosettes and some upright plants present initially. Control of these plants was achieved at
the end of the 2002/03 season.
• In August 2003, it was noted that a “carpet” of new arrowhead plants that had grown
from seed was present.
• Control of these seedlings, particularly if possible at low herbicide rates, may reduce the
arrowhead population load in the area at the beginning of the new irrigation season.
Conclusions:
• Control by end of October 2003 effective in all treated plots (see Figure 4.23)
Glyphosate, even at low doses, controls seedlings, while Casoron G controlled seedlings
at label rate, and suppressed regrowth from underground biomass
170
Experiment 23 – Casoron G Efficacy on Arrowhead and Ribbonweed in excavated vs
unexcavated channels
Aim: To investigate the control of arrowhead and ribbonweed using Casoron G at label
rates in excavated and unexcavated channel sections
Location: Channel near Corop
Treatments:
• Two physical channel properties – excavated vs unexcavated
• One concentration of Casoron G – around 170 kg/ha
• No replication, simple trial run
• One control plot
Dates / Notes:
Started 19/06/2003
Casoron G applied 19/06/2003 - 12ºC, wind 0-5km/h and sunny
Plants around 40cm high
Water depth about 20cm
No results at time of writing
171
Experiment 24 – Control of Arrowhead Seedlings
Aim: To investigate the control of young arrowhead seedlings using Weedmaster at 4 rates
Location: Main No. 5, on the site of old Experiment 06
Treatments:
• Four concentrations of Weedmaster Duo – 2.5, 5.0, 10.0 and 40.0 L/ha
• No replication, simple trial run, plots 40m long – plots contiguous.
• One control plot
• Total 5 plots
Plot Plan:
Plot No. Herbicide Applied Rate
1 Glyphosate 2.5 L/ha
2 Glyphosate 5.0 L/ha
3 Glyphosate 10.0 L/ha
4 Control -
5 Glyphosate 40.0 L/ha
Dates / Notes:
• Started September 2003
• Glyphosate applied 02/09/2003 at 2.5L/ha.
• On that day, spraying was abandoned due to a sudden increase in wind speed.
• The rest of the plots were sprayed on 04/09/03
• Will record percentage cover in future
• Will record an estimate of the percentage of plants that are seedlings
• 12th September 2003
172
• Some purpling of arrowhead seedlings. May be due to frosts or glyphosate. Only very
early after application yet.
Background:
• A few dead or almost dead rosette plants present initially. Control of these plants was
achieved at the end of the 2002/03 season.
• In August 2003, it was noted that a “carpet” of new arrowhead plants that had grown
from seed was present.
• Control of these seedlings, particularly if possible at low herbicide rates, may reduce the
arrowhead population load in the area at the beginning of the new irrigation season.
No results at time of writing
173
Appendix 2 - Biology Trials
Temperature investigations
Growth Cabinets at KTRI, Frankston 5th September 2002 12 hours day and 12 hours night 4 replicates for each combination Temperatures Water Samples Vessels set-up date 10ºC / 7ºC tap water Sample no. 24 glass vials 05/09/2002 15ºC / 12ºC channel water Sample no. 27 petri dishes 12/09/2002 20ºC / 17ºC 05/09/2002 25ºC / 22ºC 05/09/2002 20ºC / 17ºC (e.g.) refers to 20ºC during day cycle & 17ºC during night cycle. 25th September 2002 no germination in petri dishes vials - no germination at any temperature, except 20ºC / 17ºC
- germination at 20ºC / 17ºC : 90%, 90%, 100%, 90% - Tapwater - germination at 20ºC / 17ºC : 90%, 80%, 100%, 90% - Channel water
transferred some from ungerminated temperatures to 20ºC / 17ºC cabinet Set up the same again at 19ºC / 17ºC & 21ºC / 19ºC. This time with just channel water. 31st October 2002 Actual temperatures after set-up 18ºC, 20ºC and 21ºC Percentage germinations:
Temperature sample no. 21 sample no. 24 sample no. 27
18ºC 0%, 20%, 50% 5, 0, 0 0, 0, 5
20ºC 90, 60, 60 10, 50, 50 5, 5, 10
21ºC 95, 5, 95 50, 95, 95 95, 95, 0 (channel)
21ºC - - 80, 95, 90 (tapwater)
New temperatures set up - 20ºC, 21ºC, 25ºC
174
21st November 2002 Actual temperatures after set-up 19ºC, 21ºC and 24ºC
Temperature sample no. 21 sample no. 24 sample no. 27
19ºC 90%, 90%, 90% 5, 5, 5 0, 5, 5
21ºC 90, 90, 90 90, 90, 90 90, 90, 90
24ºC 5, 0, 10 10, 0, 15 0, 0, 0
In 21ºC cabinet, set up three replicates of a number of samples, to assess the variability between samples. Samples 28 (newly collected) 12 15 (alisma) 8 9
Trigger temperature for germination in laboratory conditions around 21°C
175
Effect of water depth on arrowhead rosette behaviour
Aims: to determine the effect of four different depths of water on whether arrowhead rosettes remain as rosettes or become erect plants. Location of trial: Temperature controlled room at APS office, Tatura, Victoria Trial set-up: i) Set up in plastic laundry baskets lined with orange plastic bin liners. Soil was added
to the bins, until the soil surface was roughly the required depth from the rim of each basket. Soil was saturated to allow settling before more soil was added to obtain the final depths required. Water was then filled to the rim of the baskets (23/07/2003)
ii) Four water depths:
• 5cm deep • 20cm deep • 35cm deep • 50cm deep
iii) Four replicates for each treatment iv) Arrowhead rosettes grown in the hoop house were transplanted into laundry baskets,
one plant per basket v) Temperature controlled room was maintained on a cycle of 14 hours of light and 10
hours of dark for the duration of the trial vi) The number of rosette and erect plants were counted for each treatment after 8 weeks,
then again after 16 weeks. The experiment concluded after 20 weeks. vii) October 8 2003, the water level in two of the four replicates was lowered to 5cm deep
for the three deeper depths viii) November 10 2003, all baskets moved out into open air. Placed on concrete pad
between two sheds, APS yard, Tatura. ix) November 17 2003, 2,4-D injection made into baskets 3,6,8,11,12,13,14,15 at
equivalent to 10L/ha.
This injection follows from observations in the field that suggest that a concentration effect from application of 2,4-D means that when 2,4-D is applied in channels that have a small amount of water in the bottom of them, then the chemical mixes with the water, producing a “solution” of concentration that varies with the depth (and therefore volume) of water in the channel to control the rosette form of arrowhead. Calculations as to how much solution to inject into baskets was made on the basis of the amount of herbicide that would be added to water surface at an application rate of 10L/ha in 1000L/ha of mix. The hypothesis would be that the deeper the water, the lower the concentration of 2,4-D and therefore control will be less successful.
176
Calculations: Surface area of tubs = (18.5)2 x ∏
= 1075cm2 = 0.1075m2
→ 10L/ha ⇒ 0.01mL herbicide per basket 1000L/ha mix ⇒ 10mL mix per basket
Table 1: October changes to water levels on established plants Basket number Water level Basket number Water level 1 Lowered 9 Lowered 2 Lowered 10 Unchanged 3 Unchanged 11 Unchanged 4 Lowered 12 Unchanged 5 Lowered 13 Unchanged 6 Unchanged 14 Unchanged 7 Unchanged 15 Unchanged 8 Unchanged 16 Lowered
177
Trial plan (inside):
Basket 1
50 cm deep
Basket 2
35 cm deep
Basket 3
5 cm deep
Basket 4
20 cm deep
Replicate
1
Basket 5
20 cm deep
Basket 6
50 cm deep
Basket 7
5 cm deep
Basket 8
35 cm deep
Replicate
2
Basket 9
50 cm deep
Basket 10
5 cm deep
Basket 11
35 cm deep
Replicate
3
178
Basket 12
20 cm deep
Basket 13
5 cm deep
Basket 14
20 cm deep
Basket 15
50 cm deep
Basket 16
35 cm deep
Replicate
4
179
Trial Plan (outside): Basket
15 Basket
16 Basket
13 Basket
14 Basket
11 Basket
12 Basket
9 Basket
10 Basket
7 Basket 8
Basket 5
Basket 6
Basket 3
Basket 4
Basket 1
Basket 2
← North
Conclusions:
After 8 weeks, only those arrowhead rosettes planted in 5cm deep water have produced upright plants. All of these plants have also produced rosette plants off the original plants
Following lowering of the water level, the rosette leaves desiccate very quickly on exposure to air. This is due to their very soft structure, which probably requires water as a medium to support the leaves. The degrading of the rosette leaves may be a trigger for the production of upright stems?
When placed in outside environment, all plants except those at 50 cm depth, produced
emergent plants
180
Effect of water depth on arrowhead corm resprouting
Aims: to determine the effect of four different depths of water on arrowhead corm resprouting. Location of trial: Temperature controlled room at APS office, Tatura, Victoria Trial set-up: i) Set up in plastic laundry baskets lined with orange plastic bin liners. Soil was added
to the bins, until the soil surface was roughly the required depth from the rim of each basket. Soil was saturated to allow settling before more soil was added to obtain the final depths required. Water was then filled to the rim of the baskets (23/07/2003)
ii) Four water depths:
• 50mm deep • 200mm deep • 350mm deep • 500mm deep
iii) Four replicates for each treatment iv) Four arrowhead corms were added to each of the baskets after they were allowed to
equilibrate and settle for a few days (13/08/2003) v) Temperature controlled room was maintained on a cycle of 14 hours of light and 10
hours of dark for the duration of the trial vi) Percentage resprout and phenotype were measured in each of the baskets after 2, 4
and 6 weeks (Table 2) vii) Height / length of leaves was measured at 4 and 6 weeks (Table 3) viii) October 8 2003, the water level in two of the four replicates was lowered to 5cm deep
for the three deeper depths (Table 1) ix) November 10 2003, all baskets put out in the open air (see trial plan below), between
two sheds in APS yard, Tatura.
181
Table 1: October changes to water levels on established plants Basket number Water level Basket number Water level 1 Unchanged 9 Unchanged 2 Unchanged 10 Unchanged 3 Unchanged 11 Unchanged 4 Lowered 12 Unchanged 5 Unchanged 13 Unchanged 6 Lowered 14 Lowered 7 Unchanged 15 Lowered 8 Lowered 16 Lowered
182
Trial plan (inside):
Basket 1
50 cm deep
Basket 2
5 cm deep
Basket 3
35 cm deep
Basket 4
20 cm deep
Replicate
1
Basket 5
20 cm deep
Basket 6
50 cm deep
Basket 7
5 cm deep
Basket 8
35 cm deep
Replicate
2
Basket 9
50 cm deep
Basket 10
5 cm deep
Basket 11
35 cm deep
Replicate
3
Basket 12
20 cm deep
Basket 13
5 cm deep
Rep
3
Rep
4
Basket 14
20 cm deep
Basket 15
50 cm deep
Basket 16
35 cm deep
Replicate
4
183
Trial Plan (outside): Basket
16 Basket
14 Basket
12 Basket
10 Basket
15 Basket
13 Basket
11 Basket
9
Basket 8
Basket 6
Basket 4
Basket 2
Basket 7
Basket 5
Basket 3
Basket 1
← North Conclusions:
After 6 weeks, the number of arrowhead corms that resprouted to produce plants was significantly lower at 500mm depth than at other depths. The number of corms resprouting was not significantly different between the other three depths
After 6 weeks, similarly to the rosette behaviour trials, only plants at 50mm depth had
produced upright stems. One of the replicates at this depth had also produced rhizomes and daughter plants, where none of the replicates for other depths had done so.
Following lowering of the water level, the rosette leaves desiccate very quickly on
exposure to air. This is due to their very soft structure, that probably requires water as a medium to support the leaves. The degrading of the rosette leaves may be a trigger for the production of upright stems?
All corms produced new plants. Form of plant followed the trend set in the previous trial
184
Effect of temperature and water source on arrowhead seed germination
Aims: i) to determine the effect of four different ambient temperatures on arrowhead seed
germination ii) to determine if there is a difference in the rate of arrowhead germination between two
water samples Location of trial: Temperature controlled growth cabinets at Keith Turnbull Research Institute, Frankston, Victoria Trial set-up: a) trialed in petri dishes i) Four separate temperatures:
• 10ºC started 05/09/2002 (this was a cycle of 10ºC day and 7ºC night temperature)
• 15ºC started 12/09/2002 (this was a cycle of 15ºC day and 12ºC night temperature)
• 20ºC started 12/09/2002 (this was a cycle of 20ºC day and 17ºC night temperature)
• 25ºC started 05/09/2002 (this was a cycle of 25ºC day and 22ºC night temperature)
Temperatures were put on a cycle of small differences between day and night temperature. This was to keep temperature relatively stable, but provide some variation to prevent the reduced germination that is often seen in situations of no temperature variation
ii) Two water sources:
• tapwater, Frankston • channel water, 4/8/6, G-MW, west of Numurkah
iii) Arrowhead seed from two different sources:
• Sample 24 – taken from Channel 6/4/8/6 below wheel 6242, Shinnicks Rd, West of Numurkah
• Sample 27 – taken from main no. 5 channel, Berrys Rd
iv) Four replicates for each treatment Petri dishes were filled with water and sealed with sticky tape to minimise evaporation of water. Light regime - 12 hours of light and 12 hours of dark
185
b) trialed in glass sample vials i) Four separate temperatures:
• 10ºC started 05/09/2002 (this was a cycle of 10ºC day and 7ºC night temperature)
• 15ºC started 12/09/2002 (this was a cycle of 15ºC day and 12ºC night temperature)
• 20ºC started 12/09/2002 (this was a cycle of 20ºC day and 17ºC night temperature)
• 25ºC started 05/09/2002 (this was a cycle of 25ºC day and 22ºC night temperature)
ii) Two water sources:
• tapwater, Frankston • channel water, 4/8/6, G-MW, west of Numurkah
iii) Arrowhead seed from only one source:
• Sample 27 – taken from main no. 5 channel, Berrys Rd
iv) Four replicates for each treatment Vials were capped with seeds suspended in water. Light regime - 12 hours of light and 12 hours of dark Notes: Petri dishes and vials were rated for percentage germination on 18/09/2002 Petri dishes – no germination was recorded for any of the petri dishes. This may be due to a number of factors: i) some of the petri dishes dried out during the course of the experiment, so germination
would have been severely hampered. ii) Petri dishes present a large surface area to contact with the air. Aquatic plants that
require inundation to germinate often have less successful germination in the presence of oxygen. Unless the petri dishes oxygen was completely absent from the petri dishes and they were well sealed against contact with external oxygen, this could have contributed to the lack of germination.
186
Glass vials – germination only at 20ºC, zero germination at 10ºC, 15ºC or 25ºC. Channel water tap water Temperature 10ºC 15ºC 20ºC 25ºC 10ºC 15ºC 20ºC 25ºC rep 1 - % germ. 0 0 90 0 0 0 90 0 rep 2 - % germ. 0 0 80 0 0 0 90 0 rep 3 - % germ. 0 0 100 0 0 0 100 0 rep 4 - % germ. 0 0 90 0 0 0 90 0 Vials produce an atmosphere more conducive to the germination of arrowhead, with a lower surface area to the water and a better ability to seal the container. Conclusions:
There is no effect of water source on arrowhead germination
Arrowhead germination is optimal at 20ºC, when compared to 10ºC, 15ºC or 25ºC
Further investigation indicated 21°C is the optimal temperature for arrowhead germination
187
Effect of dark on seed germination and survival of seedlings
Aims: i) to determine the effect of darkness on arrowhead seed germination ii) to determine the effect of darkness on the survival of young arrowhead seedlings Location of trial: Temperature controlled growth cabinets at Keith Turnbull Research Institute, Frankston, Victoria Trial set-up: a) trialed in plastic vials – effect of darkness on germination i) Temperature: 21ºC started 30/01/2003 ii) Water source: tapwater, Frankston iii) Arrowhead seed from only one source:
• Sample 27 – taken from main no. 5 channel, Berrys Rd
iv) Four replicates for each treatment v) Two treatments:
• Full light regime – 12 hours of light and 12 hours of dark
• Dark regime – no light, vials kept in dark containers
Vials were capped with seeds suspended in water. b) trialed in plastic vials – effect of darkness on survival of very young seedlings i) Temperature: 21ºC started 30/01/2003 ii) Water source: tapwater, Frankston iii) Arrowhead seed from only one source:
• Sample 27 – taken from main no. 5 channel, Berrys Rd
iv) Seeds were germinated and grown for two weeks in vials, then four vials were placed in each of the two treatments, light and dark
v) Four replicates for each treatment
188
vi) Two treatments: • Full light regime – 12 hours of light and 12 hours of dark
• Dark regime – no light, vials kept in dark containers Vials were capped with seeds suspended in water. Notes: a) trialed in plastic vials – effect of dark on germination Vials were rated for percentage germination on 15/04/2003 No germination was recorded for any of the vials in the dark.
Vials in the light regime returned the following percentage germination:
Rep 1 - 40%
Rep 2 - 60%
Rep 3 - 60%
Rep 4 - 80%
b) trialed in plastic vials – effect of dark on survival of very young seedlings Vials were rated for percentage survival on 15/04/2003-06-26
No survivors were recorded for the vials placed in the dark
100% survival was recorded for all vials kept in the light regime.
c) post-experiment assessment
All vials from parts (a) and (b) were brought back to constant temperature room at APS and kept under fluorescent lighting.
i) Ungerminated seeds from dark treatment of experiment (a) germinated at the following rates:
Rep 1 - 60%
Rep 2 - 60%
Rep 3 - 80%
Rep 4 - 80%
ii) Vials that showed no seedling survival in dark from experiment (b) did not recover
iii) Germinated seedlings from both experiments all survived
189
Conclusions:
Arrowhead seed does not germinate in complete darkness, though seed remains viable
Young arrowhead plants (small, thread-like seedlings) do not survive prolonged periods of complete darkness
190
Effect of manual cutting of arrowhead survival
Aims: To determine if physical removal of arrowhead leaves (both straplike submerged leaves and emergent leaves) kills arrowhead plants. Location of trial: Plastic-skinned hoophouse at APS office, Tatura, Victoria Trial set-up: i) Pots containing established arrowhead plants in black plastic stock troughs were used. ii) Three treatments:
• Cut just above soil level (pots marked with stakes) • Erect plants cut just above water level (pots marked with
silver painted stakes) • uncut
iii) Four replicates for each treatment, spread randomly through stock troughs. iv) Trial was set up on 29/10/2003 Notes: It was suggested that, if arrowhead plants did not survive cutting below water level (in much the same way as cumbungi can be controlled by cutting below water level), control could be achieved through the use of a weed cutting boat. The “cut just above water level” treatment was added to see if it is just the action of cutting that causes any effects that may occur, as opposed to the action of “drowning” caused by cutting plants under water. Conclusions:
Arrowhead is not killed by cutting above or below the water level
191
Effect of manual cutting of rosette leaves on formation of upright stems
Aims: To determine if physical removal of arrowhead rosette leaves (straplike submerged leaves) promotes the growth of upright arrowhead stems. Location of trial: Plastic-skinned hoophouse at APS office, Tatura, Victoria Trial set-up: i) Pots containing established arrowhead rosettes in black plastic stock troughs were
used. ii) Two treatments:
• Cut just above soil level • uncut
iii) Three replicates for each treatment iv) Treatments were established 29/10/2003 Notes: It was noted that, when the water level was reduced to only a few centimetres in trials run in the coolroom, the straplike leaves of arrowhead rosettes, because of their structure, break easily and desiccate when exposed. It was suggested that this degradation and subsequent removal of rosette leaves may trigger the formation of upright stems. This is simulated here by the removal of leaves by cutting. It should be noted that cutting does not allow for the movement of nutrients from degrading leaves back into the root system as leaves degrade Conclusions:
Arrowhead form is not influenced by cutting of the rosette plants
192
Effect of water depth on emergent leaf form
Aims: To determine if water depth has an effect on the type of emergent leaves produced from arrowhead rosettes. Location of trial: Plastic-skinned hoophouse at APS office, Tatura, Victoria Trial set-up: i) Pots containing established arrowhead rosettes were placed at different depths in
black plastic stock troughs ii) Two water depths:
• Saturated soil (water at 0cm deep) • Water at 20cm deep
iii) Three replicates for each treatment Notes: It was noted that all the plants growing under any depth of water in coolroom and hoophouse only produced narrow-leafed upright stems. Arrowhead in drains is almost 100% broad-leafed form, where arrowhead in channels are commonly (but not exclusively) of the narrow-leafed form. The theory was suggested that the broad-leafed form may arise when there is no standing water over the substrate in which the plant grows. Conclusions:
Arrowhead emergent leaf form is not affected by water depth. It is more likely to be due to the age and condition of the plant root system
193
Effect of 2,4-D concentrations in water on arrowhead rosette control
Aims: to determine if differing concentrations of 2,4-D in water surrounding arrowhead rosettes causes death of plants. Location of trial: Laundry baskets, on concrete pad between two sheds, APS office, Tatura, Victoria Trial set-up: i) Arrowhead rosette plants were collected from the field ii) Set up in plastic laundry baskets lined with orange plastic bin liners. Three shovels of
soil were added to the bins. Five arrowhead plants were then added to each basket. Water was then filled to the rim of the baskets (05/11/2003) and plants allowed to equilibrate in the water for ten days. In this time, some of the rosettes produced upright stems.
iii) 15/11/2003 – Water depth was lowered to 20 cm above the soil iv) 2,4-D was added to the baskets at several different concentrations (see table 1) and
treatments arranged randomly (see table 2):
Table 1- concentrations of 2,4-D added to baskets (bins)
Duplicate 1 Duplicate 2
mg/L mL of a.i./L mL product/L
mL diluted product/L
Closest to APS shed
Buffer Bins Bin 0 Bin 0
0 0 0.000 0 Bin 1 Bin 12 0.5 0.010752 0.017 17 Bin 5 Bin 10 1 0.021504 0.034 34 Bin 7 Bin 16 2 0.043008 0.069 69 Bin 2 Bin 14 4 0.086017 0.138 138 Bin 6 Bin 15 8 0.172033 0.275 275 Bin 4 Bin 11 16 0.344067 0.551 551 Bin 2 Bin 9 32 0.688134 1.101 1101 Bin 8 Bin 13
v) After 2 days the treated water was removed from each basket and the baskets were re-
filled to the rims.
194
Table 2 – basket (bin) layout
Bin 8 32mg/L Bin 16 1 Bin 7 1 Bin 15 4 Bin 6 4 Bin 14 2 Bin 5 0.5 Bin 13 32 APS shed Bin 4 8 Bin 12 Control Chemical store Bin 3 2 Bin 11 8 Bin 2 16 Bin 10 0.5 Bin 1 Control Bin 9 16 → North buffer buffer
Dependant on the growth rate of arrowhead plants at time of 2,4-D addition,
concentrations of the herbicide in water can kill above-ground biomass. More research
required.
195
Appendix 3 - Cross-section Surveys
Surveys were carried out in channels in the Murray Valley Irrigation Area and in southern New South Wales. At each location, a transect was placed across the channel, and the elevation gradient was measured using a laser beacon and staff. The growth form of arrowhead was then recorded at regular intervals along the transect. At some sites, the source of the plant (rhizome or seed) was recorded, while at some sites, the depth of sediment deposited on top of the hard clay base was recorded at regular intervals. Graphic representations of cross sections are presented here. Data suggest that emergent plants only form when water depth is < 50 cm, while rosette plants grow across the depth gradient. Data also suggest that the cut-off point for seedling establishment is between 40 and 50 cm below full supply level.
In all of the following representations of channel cross-section, abbreviations referred to in legends, where used, are as follows:
RR – Rosette (submerged) plant arising from a rhizome SR – Rosette plant arising from seed ERP – Emergent plant (broad or narrow-leafed) arising from a rhizome ESP – Emergent plant arising from seed Figures A3.1 to A3.14 are taken from sites in the Murray Valley Irrigation Area of northern Victoria, and Figures A3.15 to A3.23 are taken from sites in southern new South Wales
-120.0
-110.0
-100.0
-90.0
-80.0
-70.0
-60.0
-50.0
-40.0
-30.0
-20.0
-10.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
distance from high water mark (m)
elev
atio
n (c
m)
= erect rhizome plant
= erect non-rhizome plant
= rosette rhizome plant
= rosette seedling plant
High Water Mark
Depth to Clay
Depth to Soil Surface
seedling cut-off depth
North South Figure A3.1 (above) – transect in Main No. 5 Channel, at Trial Site No. 13
196
-200.0
-150.0
-100.0
-50.0
0.0
50.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationSRRRPoly. (elevation)
Figure A3.2 – transect in 8/6 channel, Walshs Bridge Rd, near Numurkah (1)
-200.0
-180.0
-160.0
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationRRERPPoly. (elevation)
Figure A3.3 – transect in 8/6 channel, Walshs Bridge Rd, near Numurkah (2)
197
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
20.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationSRRRESPdepth to clay (cm)Poly. (elevation)Poly. (depth to clay (cm))
West East
Figure A3.4 – Saxton St West, near Numurkah (1)
-150.0
-130.0
-110.0
-90.0
-70.0
-50.0
-30.0
-10.0
10.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationSRRRERPESPdepth to clay (cm)Poly. (elevation)Poly. (depth to clay (cm))
Figure A3.5 – Saxton St West, near Numurkah (2)
198
-60.0
-50.0
-40.0
-30.0
-20.0
-10.0
0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
distance from high water mark (m)
elev
atio
n (c
m)
elevation
RR
ERP
Poly.(elevation)
Figure A3.6 – Broken Creek, just downstream from Galt’s Bridge
-180.0
-160.0
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
20.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
distance from high water mark (m)
elev
atio
n (c
m)
elevation
SRRRESP
ERPPoly. (elevation)
Figure A3.7 – 9/6 Channel, cnr Goulburn Valley Highway and Centre Rd, north of Numurkah
199
-35.0
-30.0
-25.0
-20.0
-15.0
-10.0
-5.0
0.0
5.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationERPESPPoly. (elevation)
Figure A3.8 – Drain 13, North-West of Numurkah
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
distance from high water mark (m)
elev
atio
n (c
m)
elevation
SR
RR
ESP
depth to clay
Poly. (elevation)
Poly. (depth to clay)
East West
Figure A3.9 – corner of Boothroyds Rd and Katamatite to Nathalia Rd
200
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
20.0
0.0 2.0 4.0 6.0 8.0 10.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationSRRRERPESPdepth to clay (cm)Poly. (elevation)Poly. (depth to clay (cm))
West East
Figure A3.10 – upstream of herbicide trial site 12
-80.0
-70.0
-60.0
-50.0
-40.0
-30.0
-20.0
-10.0
0.0
10.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
distance from high water mark (m)
elev
atio
n (c
m)
elevationSeed RosetteRhiz RosetteSeed ErectPoly. (elevation)
Figure A3.11 – 100m downstream of herbicide trial site 6
201
-50.0
-45.0
-40.0
-35.0
-30.0
-25.0
-20.0
-15.0
-10.0
-5.0
0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationERPESPPoly. (elevation)
Figure A3.12 – Herbicide trial site 7
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0
distance from high water mark (m)
elev
atio
n (c
m)
elevationSRRRERPESPdepth to clay (cm)Poly. (elevation)Poly. (depth to clay (cm))
West East
Figure A3.13 – Herbicide trial site 17
202
-160.0
-140.0
-120.0
-100.0
-80.0
-60.0
-40.0
-20.0
0.0
20.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
Distance from high water mark (m)
elev
atio
n (c
m)
Figure A3.14 – Herbicide trial site 22
Moulemein Ext A
-40
-30
-20
-10
0
10
20
30
40
50
60
70
0 1 2 3 4 5 6
distance from high water mark (m)
heig
ht (c
m)
Height
Rosette
Erect
Supply Height
Figure A3.15 – Moulamein Extension A channel, NSW
203
Moulamein Ext B
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
40
50
60
0 1 2 3 4 5 6
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.16 – Moulamein Extension B channel, NSW
Moulamein Ext C
-80
-60
-40
-20
0
20
40
60
80
0 1 2 3 4 5 6
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.17 – Moulamein Extension C channel, NSW
204
Mulwala Channel a
-120
-100
-80
-60
-40
-20
0
20
0 0.5 1 1.5 2 2.5 3
distance from high water mark (m)
heig
ht (c
m)
Height
Rosette
Erect
Poly. (Height)
Figure A3.18 – Mulwala Channel A, NSW
Mulwala 28
-60
-50
-40
-30
-20
-10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.19 – Mulwala Channel 28, NSW
205
-130
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
0 0.5 1 1.5 2 2.5 3
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.20 – Mulwala Channel B, NSW
Mundy War
0
10
20
30
40
50
60
70
80
90
100
0 0.5 1 1.5 2 2.5
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.21 – Mundy War, NSW
206
-60
-50
-40
-30
-20
-10
0
10
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
distance from high water mark (m)
heig
ht (c
m)
HeightRosetteErectPoly. (Height)
Supply Height
Figure A3.22 – Birgib Bigil, NSW
Blighty
-70
-60
-50
-40
-30
-20
-10
0
0 0.5 1 1.5 2 2.5 3 3.5 4
distance from high water mark (m)
heig
ht (c
m)
Height
Rosette
Poly. (Height)
Supply Height
Figure A3.23 – Blighty, NSW
207
Observing an arrowhead infestation in the days before the species became a
major problem to Goulburn-Murray Water