Plant Populations and

Grassland DynamicsDavid Hartnett

Harmony Dalgleish

Jackie Ott

Maipelo N’Guessan

Gail Wilson

Understanding grassland plant demography is important to:

• Provide a mechanistic understanding of patterns and dynamics at the community and ecosystem levels

• Link population and ecosystem level processes

• Predict responses to environmental change – many global change phenomena are demographic processes (e.g. species invasions, cover change)

In perennial grasslands, vegetative reproduction and and belowground bud banks are important regulators of vegetation dynamics and productivity

Preburn 2May 16May 29May 12Jun 26Jun 11Jul 24Jul 15Aug

Per

cent

cha

nge

-20

0

20

40

60

80

100

Live

ste

ms

/ m2

600

800

1000

1200

1400

1600

1800

2000

SPB20B

a.

Fire

26A

pr

*

* * * * ** *

*

Fig. 13. Overall dynamics of live stems on the annually (spring) burned water-shed (black symbols) and the planned 20 yr burned watershed (open symbols). Error bars represent +/- 1SE of the mean. Asterisks indicate a significant difference between watersheds with p < 0 05 a Average live stem density for

b.

Benson & Hartnett. 2004. Plant Ecology 6:1-15

4Apr 19Apr 2May 16May29May12Jun 26Jun 11Jul 24Jul 15Aug

See

dlin

g em

erge

nce

(num

ber /

m2 )

0

1

2

3SBP20B

Fig. 2. The sample date on which each seedling was first detected, of the seedlings present at the last sampling date. Black bars represent seedlingsemerging in the annually (spring) burned prairie. White bars represent seed-lings emerging in the planned 20 year burn watershed.

Fire

26A

pr

Recruitment from Seed = 0.6 %

Recruitment from the bud bank (belowground meristems) = 99.4%

Plant recruitment in undisturbed sites

Rogers & Hartnett. 2001. Amer. J. Botany 88: 1634-1642

B.

Sampling Period

J96 J96 A96 S96 O96 M97 J97 J97 A97 S970

1

2

3

4

5

6

7

8

9

10

11

12 Large mounds

SDRAIN

CLONAL

SDBANKREGROW

80 %From buds

20 % From seed

Stems 0.01m-2

Plant recruitment in disturbed microsites

Long-term Average Annual Precipitation (mm/year)

200 300 400 500 600 700 800 900 1000

Ave

rage

Bud

Ban

k(b

elow

grou

nd m

eris

tem

s/m

2 )

0200400600800

100012001400160018002000

SEVSGSNVPKANKNZROC

P = 0.02r2 = 0.81

aa

b

b

b

c

Dalgleish & Hartnett 2006

Site

Meristem Limitation Index

(total buds/total stems)Rockefeller Prairie 3.72Konza Prairie 1.35Niobrara Valley Preserve 0.45Kanopolis State Park 0.50Short Grass Steppe 0.33Sevilleta 0.09

Dalgleish & Hartnett 2006

ANPP, Community Structure

Active budpool(meristems/m2)

Dormantbud pool(buds/m2)

Mortality

Mature rhizomepopulation

Bud outgrowth

Ramet population

(No./m2)

seedrain

seedbank

0.42

0.22

2,000

1,550

0.4x10-3

0.27

15K

4K

0.78

0.36

2.14

720

Stage-based matrix models: a tool to predict grassland population responses to environmental change

Consequences of bud bank populations and meristem limitation -

1. Grassland responses to environmental change

- tracking changes in resource availability

- resistance to invasibility

- demographic buffering capacity in rare species

- change in growth form dominance (grass-woody)

2. Genetic/evolutionary consequences

AN

PP

(g/m

2 )

Year

Precipitation (m

m)0

500

1000

1500

0

100

200

300

400

500

600

700

800

19951990198519801975

annually burned uplandannually burned lowland

2000

Future (LTER VI) Questions:

1. Partitioning variation in ANPP in response to environmental change

Variation in tiller size (physiology and RGR)

Variation in tiller number (bud bank dynamics)

Variation in species composition

Variation in Net Primary Productivity

Bud FloweringTiller

VegTiller

P1 P2 P3

G2G1

VfVv

A =

P3G200P2G1

VfVvP1

Response to increased resources (Nitrogen) in Sporobolusheterolepis

Elasticities

Control + NP1 0.10 ± 0.003 0.09 ± 0.003P2 0.19 ± 0.005 0.19 ± 0.007P3 0.01 ± 0.001 0.02 ± 0.003G1 0.35 ± 0.003 0.33 ± 0.004G2 0.01 ± 0.002 0.02 ± 0.003Vv 0.33 ± 0.005 0.31 ± 0.007Vf 0.01 ± 0.002 0.02 ± 0.003

S. heterolepis - 2004

Tiller Size

2004 2005 2005

Ave

rage

Tille

r Siz

e (g

)

0.0

0.1

0.2

0.3

0.4Control+ N

S. heterolepis K. macrantha

P = 0.08P = 0.28

P = 0.19

Number of clips1 2 3 4 5 6 7 8

RG

R (g

g-1

d-1)

0.00

0.02

0.04

0.06

0.0820062005

a

a

b b

cc

c c

N’Guessan (2007)

Schyzachrium scoparium

Num

ber o

f bel

owgr

ound

bud

s/Ti

ller

N’Guessan (2007)Number of Clips

0 1 2 3 4 5 6 7 8 90

2

4

6

8

10

abca ab

cabc ab ab

abc abc bc

20062005

Reserve (dormant season) bud bank in Schyzachrium scoparium

Future Questions:

2. Patterns of inter-specific and inter-site variation in bud bank dynamics

Kalahari-KonzaComparative Study

Watershed R20A

Hartnett, Ott, & Dalgleish (in prep)

• Does bud bank size in African grass species influence their resilience to drought and grazing?

•Are seed production and bud production inversely related (trade-offs)?

• How does changing resource availability influence relative allocation to seed vs. vegetative reproduction?

• To what extent do seed banks and reserve bud banks buffer effects of environmental change on plant population dynamics?

Future Questions:

3. Long-term study of factors driving population dynamics of dominant, sub-dominant, and rare species

• Role of intrinsic and extrinsic factors on population dynamics of grass guilds (C4, C3, caespitose, rhizomatous grasses)

• Relative influence of weather, fire, and grazing on population dynamics

(c.f. Jonas and Joern 2007 synthesis on grasshopper dynamics)

Meristem population density (No. m-2)

Commun

ity

stabil

ity

Invasibility

Low

HighInvasibilitythreshold

Low High

Future Questions:

4. Influence of belowground demography (bud bank densities) on invasibility of grasslands

Meristem (bud) Density (buds/m2)

500 1000 1500 2000 2500

Exo

tic P

lant

Spe

cies

Ric

hnes

s

0

2

4

6

Data from: Benson & Hartnett 2004, Dalgleish & Hartnett (unpublished)

1998

2004-2006

Future Questions:

5. De-coupling effects of fire suppression and woody plant invasion.

1. Split plot study on unburned watershed

• K20A, 20C, N20A, N20B

• Mechanical removal of woody vegetation on one portion of unit

2. Comparison of population dynamics and community trajectories in replicate“wooded” and “un-wooded” sites within unburned watersheds

Unburned Prairie (K20 non-woody)

Native (burned) Prairie

Unburned Prairie (K20 shrub island)

Restoration 1: Herbicide-Treated Shrub Islands

Restoration 2: Burned Shrub Islands

0.95

0.08

0.540.62

Herbaceous community similarity (Morisita’s Index) between tallgrass prairie sites.

Andrade (2007)