Elephants & Big Trees: Mitigation Methods

Summary Document of the Thornybush Workshop – 08

November 2018

Compiled by: Dr Michelle Henley (Elephants Alive Director & Principal Researcher)

Mr Robin Cook (Elephants Alive Big Trees Project Manager)

1 | P a g e

Protecting both big trees and elephants

What is the individual importance of big trees in ecological terms? (Kerley and

Landman 2006; Ludwig et al. 2008; Treydte et al. 2008; Vogel et al. 2014; Edge et al.

2017)

buffering against climatic imbalances

Important nutrient pumps

Crucial to prevent flooding and erosion

Important as a food source

Aesthetic and iconic value

Home and/nesting site to many faunal components

What is the individual importance of elephants in ecological terms?

Key stone species – some ecosystem functions cannot happen without elephants (digging

of water wells, dispersing mega-faunal fruits of certain tree species)

Umbrella species – because of their large space and nutritional requirements, when you

conserve elephants you will automatically be creating suitable habitat for many other

smaller animals that share the landscape with elephants

Path finders and makers – elephants play a very important role in promoting landscape

connectivity and with South Africa’s mandate to try and increase land under protected

area status, elephants and their travelling paths can play a vital role in defining corridors

to link conservation areas

Mega-composting agents – elephants play a critical role in nutrient recycling as they can

carry up to 140kg of dung against the gradient which is important as the ancient soils of

the APNR have been leached of minerals over many centuries

Aesthetic and iconic value

Mega-pruning agents – elephants’ lighter feeding habits play a critical role in lowering the

canopy to make more browse available to other animals, also to promote micro-climates

under felled trees (when Acacias then a protective thorn cage is formed which enables

valuable grass swards to be ‘banked’ during the dry season) that promote the growth of

particular grass species.

2 | P a g e

What is the interactive importance of big trees and elephants?

Big trees provide shelter and shade to elephants, which is vital to their thermoregulation.

Big trees offer elephants security in places where they are persecuted i.e. out tracking

data shows that elephants hide out in the ironwood forests of Mozambique during the

day to avoid people and only leave the forests at night when people are sleeping.

Big trees directly offer a valuable food source to elephants (browse primarily during the

dry season).

Big trees indirectly also offer a valuable food source to elephants as they function as

nutrient pumps, you get certain grass species growing under large trees that you wouldn’t

find in the open veld.

Elephants are valuable fertilising agents, making allowance for many large trees to

germinate in the compost they provide and disperse widely over the landscape.

Elephants are critical mega-faunal fruit dispersers with an ability to carry seeds up to

65km from their source (bulls) and deposit the seed in a nutrient rich package (Bunney et

al. 2017).

Elephants are vegetation modifying agents with their feeding habits often promoting the

foraging habits of other species as moderate to light feeding habits can either promote

new shoot growth or change the chemical composition of the vegetation to make it more

nutrient rich (Makhabu et al. 2006, Kohi et al. 2011).

What is the similarities between big trees and elephants, which make them both

vulnerable ecosystem components?

Both large is size

Both long lived

Both have slow recruitment into older age classes

Both of economic value to man (trees for furniture and fuel; elephants for tusks, tourism

and trophies)

Both are vulnerable to over-exploitation

3 | P a g e

Why are we worried about big trees and elephants when looking into the future

and outside our borders?

Big trees (Geist and Lambin 2001; Parrikka 2004)

Wood fuel (charcoal and firewood) demands have led to 68% of deforestation in Africa

Wood fuel constitutes up to 85% of household energy consumption in Sub-Saharan

Africa

With a 1% increase in urbanisation there is a concurrent 14% increase in charcoal

demands

By 2025 more than 50% of Africans will be urbanised

Charcoal consumers use 4-6 X more wood than firewood users

Charcoal demands causes the greatest loss of natural forests & woodlands because of

the selection for living old growth species

Elephants (Wittemyer et al. 2014; Chase et al. 2016)

Continental scale

More than 30 years ago the Southern states (including SA) held 21% of the continental

population. Due to excessive poaching in Central and East Africa, the southern states now

hold 55% of the continental population. Poachers will turn their efforts south for the best

catch per unit effort.

4 | P a g e

In 7 years ranging from 2007-2014 we have lost 30% or 144 000 elephants

Continue at present rate, no elephants in places in 20 years’ time

Figure 1: Continental decline in the African elephant from 5 million at the turn of the century to

below 450 000 at present.

Our neighbour

50 000 in 1970 now only 19 600

Poaching on an industrialised scale

1500-1800 poached per year

Niassa, their biggest reserve, 63% decline in 3 years, second highest carcass ratio in

Africa

48% decline in 5 years

In South Africa

In the Kruger National Park close to 200 elephants have been poached since September 2015,

these are the highest poaching incidents in their history and they are indicative that we are

not immune to the potential risks.

In the APNR

The APNR density has increased over time and the individual reserve densities have reflected

the expansion of the protected areas westwards. In comparison, the APNR densities have

historically always been higher than those of the Kruger National Park (Figure 2). Within a

national context, the APNR plays an important role (Figure 3).

5 | P a g e

Figure 2: Changes in elephant densities together with the expansion of the Protected Area over time.

Figure 3: Elephant distribution in South Africa.

0,59 1,41 0,95

0,69

APNR density in 2017: 1.11 elephants/km2 KNP density in 2017: 1.05 elephants/km2

KNP density when culled: 0.32 elephants/km2 from 1967 to 1995/1996

6 | P a g e

Will reducing elephant numbers save large trees?

We are dealing with complex systems with many factors that influence large tree

survival rate.

Fence line contrasts are often used as illustration why elephant numbers need to be

reduced.

Below the roan antelope enclosure has a healthy age structure for trees inside the

enclosure while the surrounding landscape has very few large trees and almost no

younger trees of particular species.

The enclosure was erected at the time when culling was implemented and yet

reducing elephant numbers did not save the big trees.

Rather than illustrate the efficacy of culling as a management option, the enclosure

clearly illustrates just how complex the system is and that controlling elephant

numbers does not ensure the survival of big trees.

Inside the roan enclosure no hot fires were lit and there was also an absence of any

other smaller browsers other than elephants, save for a few steenbuck and duiker.

Already from just this example we can conclude that fire and herbivory (from a number

of species) are playing a very important role in tree population dynamics.

Fence line contrast, which always illustrate the extreme points along an ecosystem continuum. You

can never expect to achieve the same outcome for one side of the fence with a particular suite of

ecosystem drivers by comparing it with the other side, lacking those same ecosystem drivers. You

would first need to know which drivers are contributing towards the observed difference and then

attempt to remove all those same drivers. Within complex systems such as African savannahs, we are

only beginning to identify the potential drivers and are very far off from fully understanding how they

tie into each other.

7 | P a g e

Asner et al. (2016) used an airborne Light Detection and Ranging technique known as

LiDAR, to monitor 10.4 million woody plant species canopies in Kruger to determine

what environmental and management factors were influencing tree-fall rates.

Tree-fall rates were found to vary over space and time and have been occurring since

the 1940s, irrespective of the elephant management intervention that has been

applied.

In chronological order: elevation, soil type, elephants, fire were found to be the best

predictor of tree-fall rates.

The scale of this monitoring technique is unmatched but is still limited as additional

factors that are only obtainable by on-the ground census techniques are still required

to determine whether insect attack, herbivory (especially of younger woody species),

natural senescence and/or wind-fall could have affected tree-fall rates.

A conceptual model clearly outlines the numerous factors that could play a role in the

survival rates of large trees (Henley and Cook, in review) and once the complexity of

the possible causal factors are grasped, it becomes apparent why a number of

scientists have failed to find a linear relationship between elephant impact and

elephant density.

Specific life history- and phenotypic characteristics of the plant species under

consideration affect survival rate (germination rate from seed, seedling protection

mechanisms, ability to recoppice, susceptibility to disease, natural life span length,

palatability, root structure and bark structure).

Various abiotic factors influence survival rates (fire, climatic factors such as drought,

floods, rainfall, increased carbon levels which favour woody species due to our actions

and wind, edaphic determining factors such as soil type, elevation and topography)

Various biotic factors influence tree survival rates and include the surrounding plant

community type and herbivory.

Herbivory includes all browsers of which elephants are but one and which are known

to target trees of particular size and stem diameters (Moe et al. 2009)

When it comes to elephant impact, two factors could directly influence big tree

survival rate and they include encounter rate and density of elephants.

Encounter rates are determined by the spatial distribution of elephants while densities

of elephants are influenced by their numbers.

8 | P a g e

Densities of elephants are more inclined to influence the frequency of encounter rates

while their spatial distribution will influence the duration of the encounter rate

(O’Connor et al. 2007).

Elephants do not spread their impact evenly across the landscape unless their

distribution is controlled or influenced by regularly spaced artificial water points.

Densities of elephants are not as important as their spatial distribution as a few

elephants in an area with a high density of their favoured species will have more of an

effect in that particular patch than many elephants over a large area surrounded by

staple food species.

There are various lethal (culling, poaching, disease, drought) and non-lethal ways

(contraception and translocation) to control elephant numbers.

The spatial distribution of elephants can be controlled by the closing of artificial

waterholes, by erecting barriers, by expanding their range, by attracting them away

from sensitive areas or by frightening them away (creating fear-landscapes)

Both management methods aimed at controlling elephant numbers or their special

distribution have their pros and cons, which could be evaluated against ethical

concerns (Henley and Cook, in review).

Alternatively, one can focus on the age and size class of trees that elephants are known

to target and directly protect the resource.

There are no ethical concerns involved if focussing on iconic tree specimens with

aesthetic appeal or on a spatial scale that only aims to protect various adults of specific

age and sex to ensure seed production throughout the landscape or the maintenance

of large tree nesting bird sites.

By so doing one could promote the interactive importance between various species

and not try to change important linkages between species, which we do not yet fully

understand.

Since 2004, Elephants Alive has investigated various mitigation strategies to protect

iconic big trees throughout the APNR.

Our big tree monitoring includes 3500 large trees spread throughout the APNR.

9 | P a g e

Big Tree Population Dynamics Woody cover is influenced by a number of factors (Osborne et al. 2018):

Land clearance

Herbivory

Fire

Climate change and rising carbon dioxide

A single marula tree population can experience pressures and bottlenecks at various

stages in its life history (Cook et al. 2017):

Seed predation

Seedling recruitment

Elephant impact

Elephants and Big Trees

Forms of monitored elephant impact

Bark-stripping

Branch breakage

Main stem snapping

Uprooting

Encounter rates

Strategy is reduce encounter rates between elephants and big trees

10 | P a g e

Encounter rates affected by artificial boundaries, proliferation of water points and a

weakened density-dependent regulation of elephant populations (O’Connor et al. 2007)

Landscape approach

Manage how and where elephants use the landscape (Ferreira et al. 2015)

Waterhole closures to promote landscape heterogeneity (Sianga et al. 2017; Dzinotizei

et al. 2018)

Small-scale approach

Individual tree protection = directly protecting the resource

Not a landscape solution

Ideal for small reserves that are unable to manipulate elephant movements through

waterhole closures

Ideal for lodges/management wishing to protect particular iconic trees

11 | P a g e

Big Tree Mitigation Methods

Wire-netting

Beehives

Rock packing/Pyramids

Creosote

Bee pheromones

Bioneem oil

Chilli oil

Dung paste

Wire-Netting

Purpose: protect tree against ring-barking

Ring barking

Prevents nutrient flow

Tree becomes vulnerable to insect invasions and fire damage

Tree may be hollowed out from the inside

Variety of methods/techniques used within the Associated Private Nature Reserves

12 | P a g e

Elephant Alive method (as shown by Bateleur Safari Camp)

Double-wrapping method

13 mm Chicken mesh: 1.8 m tall

Measurement of length of mesh required: (Tree circumference x 2) + 50 cm

Fold mesh in half and then wrap around tree’s main stem, starting 50 cm above the

ground

End of chicken mesh are stapled to tree using 150-200 mm U-staple nails

Ensure that tree has a small amount of room to grow, i.e. tree must not be strangled

Important to monitor wire-netting over time

Remember: Wire-netting is only effective against bark-stripping

Our method of wire-netting not an eye-sore for tourists (Edge et al. 2017)

Longevity Up to 5 years (but ensure that tree is never

strangled)

Cost ± R100.00 per tree

Results Henley (2013): increased tree survival over

3000 days versus unwired trees

Derham et al. (2016): Bark-Stripping: 1.7%

wired vs 25% unwired trees after 12 years

Cook et al. (2017): Bark-Stripping: 2% wired

(tusking) vs 26% unwired trees after 2 years

Scale Can be applied on a large scale across a

property

13 | P a g e

Beehives

African honeybees have been used to successfully protect crops (King et al. 2017)

and trees (Cook et al. 2018) from elephants

Information on setting up beehives in trees available in a manual on the Elephants

Alive website (www.elephantsalive.org)

Two types of hives have been tested:

Longevity Wooden: > 3 years

Beepak: > 10 years

Cost Wooden: R620 per tree

Beepak: R5000 per tree

Results (Cook et al. 2018) –

2 year study

(Jejane Private Nature

Reserve)

Impact-type

Bark-stripping

Stem snapping

Uprooting

Branch

breakage

Bees

0%

0%

2%

6%

Wire-netting

2%

12%

4%

40%

Control

26%

4%

4%

48%

Scale Small-scale: Handful of iconic trees around property

14 | P a g e

Feeding may be required during droughts

Risks: bee-stings (allergies); Drought; Insecticides

Honey production potential for tourists/staff

Rock packing/Pyramids Rings of rocks or pyramids stacked around the main stem of a tree

15 | P a g e

Problems identified in the Associated Private Nature Reserve: not enough

distance between tree’s main stem and end of pyramids/rocks

Noticeable trend that elephant impact decreases as pyramid radius around tree

increases

Calculating cost of method

Total cost = Single pyramid cost x Number of pyramids in a square metre x

𝛑(Desired radius)²

Rocks/pyramids need to be tightly stacked to prevent elephants manoeuvring

between gaps

Rocks/pyramids should try keep elephants 4-5 m away from tree’s main stem

Caution should be taken for the number of large natural rocks moved around

(micro-habitat disturbance)

Longevity 10-20 years (or more) if maintained

Cost R4.50 per pyramid made (concrete + labour costs)

= > R2000 per tree

Scale Small-scale application around particular iconic

trees

16 | P a g e

Creosote

Tins/jars of creosote nailed to the main stem of a tree

What is creosote? (Choudhary et al. 2002)

Coal-creosote is a residue produced during the distillation of coal tar, which is a

by-product of the carbonization of coal

Creosote is a thick, oily and flammable liquid, and does not easily dissolve in

water

Creosote has suspected carcinogenic properties and the breakdown of creosote

in soil can take months to years to occur

In the United States of America, spills of creosote into the environment which

exceed 0.454 kg need to be reported to the Environmental Protection Agency

Assessment on 100 trees creosote trees over 9-month period

Category Result

Number of trees still alive 96

Number of trees dead 4

Number of trees impacted since method went up 23

Number of creosote jars broken 25

Concern over the spillage of creosote into the environment and breakage of

tins/glass

Comparison to trees without creosote in the same property: No statistically significant

difference between the likelihood of a tree being impacted with or without creosote

Results suggest that creosote is not an effective method for protecting trees against

elephant impact and should be replaced with a more suitable method

17 | P a g e

Bee Attack Pheromone

Bee attack pheromone help bees recruit more bees from their colony to

handle large predators (Nouvian et al. 2016)

Bee attack pheromone is comprised of iso-amyl acetate and 2-heptatone

compounds which have been synthetically turned into a synthetically

produced Specialised Pheromone and Lure Application Technology

(SPLAT) paste

Developed by Prof. Mark Wright (University of Hawaii at Manoa) and Dr

Agenor Mafra-Neto (ISCA Technologies)

Research carried out in Balule with Transfrontier Africa and Elephants

Alive (Wright et al. 2018)

86% of elephants encountering smell in Jejane Private Nature Reserve

displayed avoidance behaviours (Jejane is the beehive project study site)

Results suggest that this method may be effective if elephants have prior

knowledge of bee attacks (learnt association)

Early research phase - not yet available to public

Can last up to 2 months when applied (weather dependant)

18 | P a g e

Bioneem Oil What is Bioneem? (Benmhend et al. 2012)

Bioneem is pressed directly from seeds of the Neem tree (Azadirachta indica),

which is a tropical evergreen tree native to India

Bioneem is an insecticide that controls insects which are harmful in agriculture,

forestry and horticulture, and can be used in place of chemical insecticides

Bioneem is effective against more than 400 pests of economic importance

Half-life of bioneem: 3-22 days (45 minutes in water exposure)

Method

Bioneem is sprayed over the branches and stem of a tree as an elephant

deterrent

Hypothesis: Elephants do not like the smell/taste of bioneem and will not impact

the tree

Bioneem effects (Benmhend et al. 2012)

Anti-feedant and repellent: deters insects from feeding on treated plant

Insect growth regulation: disrupts insect development. Insects do not reach

adulthood

Oviposition deterrent: insects are prevented from laying eggs on bioneem plants

Honeybees and other pollinators avoid plants with bioneem

Risks

What else in the environment is affected if an insecticide is used solely as an

elephant repellent?

Short half-life means that trees will need to be continuously sprayed

Far more research is required on this product before being used in a protected

area

19 | P a g e

Chilli Oil

Chillies have been used as elephant detergents in sprays (Osborn and Rasmussen

1995), chilli-grease covered fences (Osborn and Parker 2002) and dung balls

Hessian ropes soaked in chilli oil (capsicum mixed with oil) is now being tested on

trees in Olifants North (Balule)

Thornybush are directly paste chilli oil onto tree stem

Reapplication every 2-4 weeks (weather dependent)

Oil-based chilli lasts longer than water-based chilli

Study sites require evaluation to test the effectiveness of the hessian ropes

(Olifants North) and direct pasting (Thornybush) methods

More data is required at a manageable scale

Elephants Alive have a product in small quantity measuring 2 million Scoville Heat

Units (SHU)

20 | P a g e

Dung Paste

Tree healing properties

Alteration of the old cow dung and clay tree healing method

Anti-bacterial properties and adhesion constituents

Seals tree wounds and prevents pathogens to otherwise exposed surfaces

Suggested method: Clay + buffalo/cattle dung mixed with fermented elephant

dung (amounts can be reduced for small quantities of trees)

Fresh buffalo/cattle manure gathered in 100kg or ¾ 200Ltr drum (no oil or

chemicals inside)

Elephant manure 100kg in 200ltr (CLEAN NB) drum filled water ¾. Add 2kg sugar

press through to dissolve and keep doing for 3 days or so until fermentation

stops. Press plug forming at top down as much as needed

200kg clay – from destroyed anthills or ones not in use

On day of use, mix the fresh buffalo/cattle manure with the clay and add the

fermented elephant ‘tea’ to the mixture. Ensure the final mix is not too fluid-like

Use a brush to paint the tree’s main stem, covering as much as you can

NB: Primarily a tree healing method, however,

elephants have been observed avoiding dung pasted

trees in Botswana

Aesthetically pleasing: tree has a rubbing post

appearance

Longevity may be affected by rainfall

21 | P a g e

Conclusions

Summary table of methods

Identifying method of use

What is my purpose for protecting the trees? Vulture nest? Aesthetic/Ecological

purposes/Tree propagation purposes?

How many trees am I wanting to protect and what am I prepared to spend on a

mitigation method?

What species or size classes of trees am I wanting to protect?

What type of elephant impact am I trying to protect against?

Is my property exposed to intense fires?

How close are the trees to my property or lodge?

Tourist perceptions – what am I willing to let my tourists see in/on/around the

trees?

22 | P a g e

Summary

Tree populations are affected at various demographic stages

Landscape management approaches can manipulate elephant encounter rates

with big trees

Methods are available to protect individual trees against impact

Method of choice is dependent on your management requirements and

available resources

Certain methods are still undergoing testing and caution is required before large-

scale usage

No method is fool proof but survival rates of big trees can be increased if

methods are applied correctly

References

Asner, G. P., Vaughn, N., Smit, I. P., & Levick, S. (2016). Ecosystem‐scale effects of megafauna in African

savannas. Ecography, 39(2), 240-252.

Benmhend, D., Butnett, G., Hollis, L., Moore, J., Xue, M., Fuentes, C., & Jones, R. (2012). Cold pressed

neem oi. Biopesticides registration action document, United States Environmental Protection

Agency.

Bunney, K., Bond, W. J., & Henley, M. (2017). Seed dispersal kernel of the largest surviving

megaherbivore—the African savanna elephant. Biotropica, 49(3), 395-401.

Chase, M. J., Schlossberg, S., Griffin, C. R., Bouché, P. J., Djene, S. W., Elkan, P. W., et al. (2016)

Continent-wide survey reveals massive decline in African savannah elephants. PeerJ 4: e2354.

Choudhary, G., Citra, M. J., McDonald, A. R., & Quinones-Rivera, A. (2002). Toxicological profile for

wood creosote, coal tar creosote, coal tar, coal tar pitch, and coal tar pitch volatiles. U.S.

Department of Health And Human Services, Public Health Service, Agency for Toxic Substances

and Disease Registry.

Cook, R. M., Parrini, F., King, L. E., Witkowski, E. T. F., & Henley, M. D. (2018). African honeybees as a

mitigation method for elephant impact on trees. Biological Conservation, 217, 329-336.

Cook, R. M., Witkowski, E. T. F., Helm, C. V., Henley, M. D., & Parrini, F. (2017). Recent exposure to

African elephants after a century of exclusion: Rapid accumulation of marula tree impact and

mortality, and poor regeneration. Forest Ecology and Management, 401, 107-116.

23 | P a g e

Derham, K., Henley, M. D., & Schulte, B. A. (2016). Wire netting reduces African elephant (Loxodonta

africana) impact to selected trees in South Africa. Koedoe, 58(1), 1-7.

Dzinotizei, Z., Murwira, A., & Masocha, M. (2018). Elephant-induced landscape heterogeneity change

around artificial waterholes in a protected savanna woodland ecosystem. Remote Sensing

Applications: Society and Environment.

Edge, A., Henley, M., Daday, J., & Schulte, B. A. (2017). Examining human perception of elephants and

large trees for insights into conservation of an African savanna ecosystem. Human Dimensions of

Wildlife, 22(3), 231-245.

Ferreira, S. M., Greaver, C., Knight, G. A., Knight, M. H., Smit, I. P., & Pienaar, D. (2015). Disruption of

rhino demography by poachers may lead to population declines in Kruger National Park, South

Africa. PLoS One, 10(6), e0127783.

Geist, H.J., Lambin, E.F. (2001). What Drives Deforestation? A Meta-analysis and Underlying Causes

of Deforestation based on Subnational Case Study Evidence. Ciaco Printshop, Louvaine-la

Neuve.

Henley, M. D. (2013). Vegetation and questionnaire report. Unpublished report to the Associated

Private Nature Reserves). Hoedspruit, South Africa: Elephants Alive.

Henley, M. D. & Cook, R. M. (in review). The management dilemma: removing elephants to save large

trees.

Kerley, G. I., & Landman, M. (2006). The impacts of elephants on biodiversity in the Eastern Cape

Subtropical Thickets: elephant conservation. South African Journal of Science, 102(9-10), 395-402.

Kohi, E.M., de Boer, W.F., Peel, M.J.S., Slotow, R., Van der Waal, C., Heitkoenig, I.M.A., Skidmore, A. &

Prins, H.H.T. 2011. African elephants Loxodonta africana browse heterogeneity in African savanna

Biotropica. 43: 711-721.

Ludwig, F., De Kroon, H., & Prins, H. H. (2008). Impacts of savanna trees on forage quality for a large

African herbivore. Oecologia, 155(3), 487-496.

Makhabu, S., Skarpe, C., Hytteborn, H. 2006. Elephant impact on shoot distribution on trees and on

rebrowsing by smaller browsers Acta Oecologica. 30: 136-146.

Moe, S. R., Rutina, L. P., Hytteborn, H., & Du Toit, J. T. (2009). What controls woodland regeneration

after elephants have killed the big trees?. Journal of Applied Ecology, 46(1), 223-230.

Nouvian, M., Reinhard, J., & Giurfa, M. (2016). The defensive response of the honeybee Apis

mellifera. Journal of Experimental Biology, 219(22), 3505-3517.

24 | P a g e

O’Connor, T. G., Goodman, P. S., & Clegg, B. (2007). A functional hypothesis of the threat of local

extirpation of woody plant species by elephant in Africa. Biological conservation, 136(3), 329-345.

Osborn, F. V., & Parker, G. E. (2002). Community-based methods to reduce crop loss to elephants:

experiments in the communal lands of Zimbabwe. Pachyderm, 33(32), e38.

Sianga, K., van Telgen, M., Vrooman, J., Fynn, R. W., & van Langevelde, F. (2017). Spatial refuges buffer

landscapes against homogenisation and degradation by large herbivore populations and facilitate

vegetation heterogeneity. koedoe, 59(2), 1-13.

Treydte, A.C., Heitkonig, I.M.A., Prins, H.H.T. & Ludwig F. 2007. Trees improve grass quality for

herbivores in African savannas. Perspectives in plant ecology, evolution and systematics 8: 197–

205.

Osborn, F. V., & Rasmussen, L. E. L. (1995). Evidence for the effectiveness of an oleo-resin capsicum

aerosol as a repellent against wild elephants in Zimbabwe. Pachyderm, 20, 55-64.

Osborne, C. P., Charles‐Dominique, T., Stevens, N., Bond, W. J., Midgley, G., & Lehmann, C. E. (2018).

Human impacts in African savannas are mediated by plant functional traits. New Phytologist.

Parikka, M. (2004). Global biomass fuel resources. Biomass and Bioenergy 27, 613–620. Vogel, S. M., Henley, M. D., Rode, S. C., van de Vyver, D., Meares, K. F., Simmons, G., & de Boer, W. F.

(2014). Elephant (Loxodonta africana) impact on trees used by nesting vultures and raptors in

South Africa. African journal of ecology, 52(4), 458-465.

Wittemyer, G., J. M. Northrup, J. Blanc, I. Douglas-Hamilton, P. Omondi, and K. P. Burnham. 2014.

Illegal killing for ivory drives global decline in African elephants. Proceedings of the National

Academy of Sciences 111:13117–13121.

Wright, M. G., Spencer, C., Cook, R. M., Henley, M. D., North, W., & Mafra-Neto, A. (2018). African

bush elephants respond to a honeybee alarm pheromone blend. Current Biology, 28(14), R778-

R780.

__________________________________________________________________________________