nicholas j. gotelli department of biology university of vermont burlington, vt 05405 u.s.a

Nitrogen deposition and extinction risk in carnivorous plants: ecological challenges

for the next century

Nicholas J. GotelliDepartment of BiologyUniversity of VermontBurlington, VT 05405

U.S.A.

0

50

100

150

200

1900 1920 1940 1960 1980 2000

Year

Ter

agra

ms

of

Nit

rog

en

Fertilizer NOx

Total anthropogenic N fixed

Natural range

Effects of N Deposition

• IndividualAltered morphologyChanges in reproduction, survivorship

Effects of N Deposition

• IndividualAltered morphologyChanges in reproduction, survivorship

• PopulationIncreased long-term extinction riskChanges in short-term dynamics

Effects of N Deposition• Individual

Altered morphologyChanges in reproduction, survivorship

• Population Increased long-term extinction riskChanges in short-term dynamics

• CommunityChanges in abundance and compositionAltered nutrient transfer and storage

Effects of N Deposition on Carnivorous Plants

• Life History

• Effects on Individuals

• Effects on Populations

• Effects on Communities

• The Role of Ecologists

Carnivorous plants: well-known, but poorly studied

Carnivory in plants

• Phylogenetically diverse

• Morphological, chemical adaptations for attracting, capturing, digesting arthropods

• Common in low N habitats

• Poor competitors for light, nutrients

Family Sarraceniaceae

Genus Common Name

Number of Species

Distribution

Darlingtonia Cobra Lilly 1 Northwest USA

Heliamphora Sun Pitchers 5 North-central South America

Sarracenia Pitcher Plants 8 Eastern USA, Canada

Genus Sarracenia

• 8 described species

• Center of diversity in southeastern US

• Many subvarieties

• Extensive hybridization

• Sarracenia purpurea (New Jersey- Canada)

The Northern Pitcher Plant Sarracenia purpurea

• Perennial plant of low-N peatlands

• Lifespan 30-50 y

• Arthropod prey capture in water-filled pitchers

• Diverse inquiline community in pitchers

The Inquilines

Wyeomyia smithii

Metriocnemus knabiHabrotrocha rosa

Blaesoxipha fletcheri

Sarraceniopus gibsoni

Inquiline food web

Phyllodia

• Flat leaves

• No prey capture

• High concentration of chlorophyll, stomates

• Photosynthetically more efficient than pitchers

Flowering Stalks

• Single stalk per rosette

• Flowering after3 to 5 years

• Bumblebee, fly pollinated

• Short-distancedispersal of seeds

Leaf Senescence

• End-of-season die off

• Production of new leaves in following spring

• Annual increase in rosette diameter


• Life History





Nutrient Treatments

• Distilled H20

• Micronutrients• Low N (0.1 mg/L)• High N (1.0 mg/L)• Low P (0.025 mg/L)• High P (0.25 mg/L)

• N:P(1) Low N + Low P• N:P(2) Low N + High P• N:P(3) High N + Low P

Nutrient Source:

Micronutrients: Hoaglands

N: NH4Cl

P: NaH2PO4

Anthropogenic N additions alter growth and morphology


Increasing N

Effects of Anthropogenic N additions

• Increased production of phyllodiaPhenotypic shift from carnivory to

photosynthesis

• Increased probability of flowering

Contrasting effects of anthropogenic N vs. N derived

from prey

Food Addition Experiment

• Ecological “press” experiment

• Food supplemented with house flies• Treatments: 0, 2, 4 ,6, 8,10,12, 14 flies/week

• Plants harvested after one field season

Food additions do not alter growth and morphology

Increasing prey

N uptake increases with food level

P uptake increases with food level

N:P ratio decreases with increasing food supply

Altered N:P ratios suggest P limitation under ambient conditions

P limitation (Koerselman & Meuleman 1996, Olde Venternik et al. in press)

Ambient

Food additions do not alter growth and morphology

Increasing prey


Increasing N

Contrasting effects of anthropogenic and natural sources of N

Anthropogenic NAltered N:P ratiosMorphological shiftReduction in prey uptake

Prey NUptake, storage of N & PNo morphological shiftsContinued prey uptake


• Life History





Study Sites

Demography survey

• 100 adult, juvenile plants tagged at each site

• Plants censused and measured each year

• Seed plantings to estimate recruitment functions

Recruits Juveniles Adults Flowering Adults

Sarracenia matrix model


Hawley Bog Transitions

0.10

4.00

0.04

0.09

0.18

0.83

0.95 0.70 0.17


Molly Bog Transitions

0.10

4.00

0.13

0.17

0.10

0.66

0.85 0.71 0.31

Matrix Transition Model(stationary)

nt+1 = Ant

Population vector at time (t + 1)

Transition matrixPopulation vector at time (t)

Population Projections

Site r individuals/individual•year Doubling Time

Hawley Bog 0.00456 152 y

Molly Bog 0.00554 125 y

Deterministic Model: Results

• Growth, survivorship, and reproduction are closely balanced in both sites

• Doubling times > 100 y

• Juvenile, adult persistence contribute most to population growth rate

• Sexual reproduction, recruitment relatively unimportant

Matrix Transition Model(stochastic)

nt+1 = Atnt


Random transition matrix at time (t)

Population vector at time (t)

Hawley Bog

-0.30-0.20-0.100.000.100.200.30

Deter

min

istic

ME (m

ild)

ME (s

ever

e)

ES (mild

)

ES (sev

ere)

r

Hawley Bog

200

400

600

800

1000

0 5 10 15 20 25Time Step

n

DeterministicES (severe)ES (mild)

Stochastic Model: Results

• Cannot reject H0 (r = 0.0)

• Environmental variation can lead to a substantial risk of long-term extinction (0.3 < p(ext) < 0.6)

How do N and P concentrations affect population growth of

Sarracenia?

Nutrient Addition Experiment

• 10 juveniles, 10 adults/treatment

• Nutrients added to leaves twice/month

• Nutrient concentrations bracket observed field values

• Nutrient treatments maintained 1998, 1999

• “Press” experiment

Nutrient Treatments

• Distilled H20

• Micronutrients• Low N (0.1 mg/L)• High N (1.0 mg/L)• Low P (0.025 mg/L)• High P (0.25 mg/L)

• N:P(1) Low N + Low P• N:P(2) Low N + High P• N:P(3) High N + Low P

Nutrient Source:

Micronutrients: Hoaglands

N: NH4Cl

P: NaH2PO4

Effects of N additions

• Increased production of phyllodia


Effects of N additions

• Increased production of phyllodia


• Decreased juvenile survivorship

Population Growth Rate(Deterministic)

-0.15

-0.10

-0.05

0.00

0.05

r

Distilled Micros Low N

High N Low P High P

NP (2) NP (1) NP (3)

LLM

H

H

Population Growth Rate(Measurement Error)

-0.40

-0.35

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

Distil

led

Mic

ros

Low N

High

N

Low P

High

P

NP (2)

NP (1)

NP (3)

r

H

ML

H

L

Effects of Nitrogen on Demography: Results

• Population growth rates respond to different N and P regimes

• Population growth rate decreases in response to increasing N

• Population growth rate decreases in responses to increasing N:P

Modeling Long-term Environmental Change

Observed N Deposition

Long-termForecast

N(t)

Transition Matrix (t)

PopulationStructure (t)

Time Series Modeling

Transition Function

Population Time SeriesExtinction RiskTime to Extinction

MatrixMultiplication

N monitoring

• National Atmospheric Deposition Program

• NH4, NO3 measured as mg/l/yr

• Annual data 1984-1998

• Monitoring sitesShelburne, VTQuabbin, MA

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

1984

1986

1988

1990

1992

1994

1996

1998

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

1984

1986

1988

1990

1992

1994

1996

1998

0.0

0.5

1.0

1.5

2.0

2.5

1984

1986

1988

1990

1992

1994

1996

1998

0.0

0.5

1.0

1.5

2.0

2.5

1984

1986

1988

1990

1992

1994

1996

1998

Quabbin, MA Shelburne, VT

NH4

N03

Regression Models

Ordinary Least Squares (OLS)

Nt = a + bt + e

First-order auto-regressive (AR-1)

Nt = a +bNt-1 + e

Quabbin (AR-1)

0.01

0.1

1

10

1 10 19 28 37 46

Year

N (m

g/l/y

r) b = 0.947

b = 1.000

b = 1.053

Quabbin (OLS)

00.10.20.30.40.50.60.7

1 10 19 28 37 46

Year

N (m

g/l/y

) b = -0.004

b = 0.000

b = 0.004

Shelburne (AR-1)

0.01

0.1

1

10

1 10 19 28 37 46

Year

N (m

g/l/y

)

b = 0.978

b = 1.000

b = 1.022

Shelburne (OLS)

0.2

0.3

0.4

0.5

0.6

0.7

1 10 19 28 37 46

Year

N (m

g/l/y

) b = -0.001

b = 0.000

b = 0.001



Long-termForecast

N(t)




Transition Function



Establishing Transition Functions

• Linear interpolation between observed data points (n = 3)

• Asymptotic transitions at extreme levels of nitrogen:pij = observed pij if [N] < 0.01 mg/l/yr

pij = 0.0 if [N] > 10.0 mg/l/yr

• Logarithmic response curve

0

0.2

0.4

0.6

0.8

1

-3 -2 -1 0 1 2

Log10 [N]

Ad

ult

s →

Ad

ult

s

ExtrapolatedObserved

Modeling Demographic Transitions as a Function of Nitrogen



Long-termForecast

N(t)




Transition Function



Matrix Transition Model(changing environment)

nt+1 = Atnt


Sequentially changing transition matrix at time (t)

Population vector at time (t)

Estimated population size at Hawley bog

Stage Number of individuals

Recruits 1500

Juveniles 23,500

Non-flowering Adults 1400

Flowering Adults 500

Quabbin Exponential Forecast Models (AR-1)

Scenario Annual % Change

P (ext) at 100 y

Time to ext (p = 0.95)

Best case -4.7% 0.00 > 10,000 y

No change 0.0% 0.038 650 y

Small increase

1% 0.378 290 y

Worst case 4.7% 0.996 70 y

Shelburne Exponential Forecast Models (AR-1)

Scenario Annual % Change

P (ext) at 100 y

Time to ext (p = 0.95)

Best case -2.2% 0.158 > 10,000 y

No change 0.0% 0.510 230 y

Small increase

1.0% 0.694 200 y

Worst case 2.2% 0.838 140 y

Shelburne Nitrogen Forecast Model

0

5000

10000

15000

20000

25000

30000

1 10 19 28 37 46

Year

Po

pu

lati

on

Siz

e

AR

OLS

Forecasting Models for Nitrogen Deposition: Results

• Increasing or stationary models of Nitrogen deposition drive Sarracenia populations to extinction

• Extinction risk declines with reduced nitrogen

• Correlated nitrogen series can induce cycles and complex population dynamics


• Life History





Sarracenia Nutrient Feedback Loop

PitcherNutrient

Pool [N,P]

InquilineCommunity

ArthropodPrey

PlantGrowth

AtmosphericDeposition

Nutrients → Inquilines

• Manipulate [N], [P] in leaves

• Orthogonal “regression” design

• Maintain [] in a “press” experiment

Response Surface Experimenal Design

[N]0 1 2 3 4 5 6 7

[P]

0

1

2

3

4

5

6

7

Effects of [N,P] on Inquiline Abundance

0

10

20

30

40

01

23

45

6

0123456

Rot

ifer

Abu

ndan

ce

[P]

[N]


PitcherNutrient

Pool [N,P]

InquilineCommunity

ArthropodPrey

PlantGrowth


Inquilines → Nutrients

• Manipulate [N], [P] in leaves

• Orthogonal “regression” design

• Establish initial [] in a “pulse” experiment

Response Surface Experimenal Design

[N]0 1 2 3 4 5 6 7

[P]

0

1

2

3

4

5

6

7

Null Hypothesis

[N]0 1 2 3 4 5 6 7

[P]

0

1

2

3

4

5

6

7

Community Regulation of Nutrients

[N]0 1 2 3 4 5 6 7

[P]

0

1

2

3

4

5

6

7


PitcherNutrient

Pool [N,P]

InquilineCommunity

ArthropodPrey

PlantGrowth


Nutrients ↔ Inquilines

),,(

),,(

tNIgdt

dI

tINfdt

dN


• Life History





0

50

100

150

200

1900 1920 1940 1960 1980 2000

Year

Ter

agra

ms

of

Nit

rog

en

Fertilizer NOx

Total anthropogenic N fixed

Natural range

Ecology ≠

Environmental Science

Reasons for Studying Ecology


• Natural History


• Natural History

• Field Studies & Experiments


• Natural History


• Statistics & Data Analysis

Population Growth Rate(Deterministic)

-0.15

-0.10

-0.05

0.00

0.05

r

Distilled Micros Low N

High N Low P High P

NP (2) NP (1) NP (3)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

1984

1986

1988

1990

1992

1994

1996

1998


• Natural History



• Modeling

),,(

),,(

tNIgdt

dI

tINfdt

dN

PitcherNutrient

Pool [N,P]

InquilineCommunity

ArthropodPrey

PlantGrowth


0

5000

10000

15000

20000

25000

30000

1 10 19 28 37 46

Year

Po

pu

lati

on

Siz

e

AR

OLS


• Natural History



• Modeling

• Collaboration

Aaron M. Ellison

Harvard Forest

Conclusions• Anthropogenic deposition of N is a major ecological

challenge


challenge• Carnivorous plants in ombrotrophic bogs are a model

system



system• Individual response

plants alter morphology and growth in response to N:P ratios





• Population response N and P environments affect population growth rate





• Population response N and P environments affect population growth rate

• Community response Further study of nutrient ↔ inquiline feedback loop

nicholas j. gotelli department of biology university of vermont burlington, vt 05405 u.s.a

Documents

mglhigh n

n peatlandslifespan

storageeffects of n

p1 low n low pn

low n habitatspoor competitors

p2 low n high pn

mglhigh p

p ratios