Changes in Vegetation Dynamics in Alaska: Implications for Arctic Herbivores

Eugénie Euskirchen

Wildlife Response to Environmental Arctic Change Workshop

Nov. 17 -18, 2008

Growing Season Change 1988 - 2000

Change in Gross Primary Productivity

1982 - 2003

Change in Snow Cover Duration

1970 - 2000

Bunn & Goetz, Earth Interactions, 2006

McDonald et al., Earth Interactions, 2004

Euskirchen et al., Global Ch. Biol., 2007

<-0.4-0.4 -0.3-0.2-0.05 0.1

>0.1

Days shorter

yr-1

Days longer yr-1

Pan-Arctic Detection of Recent Land Surface Changes

≤-3.0-2.0 -1.0 0.0 1.0 2.0

≥3.0

Days shorter

yr-1

Days longer yr-1

Studies have also documented changes in plant community composition at high-

latitudes

Increased shrubiness

Sturm et al., 2001; 2005

Changes in fire regimesIncreases in insect outbreaks

Treeline advance

Suarez, 1999; Lloyd & Fastie, 2003

•Changes in the length of the snow & growing seasons impact vegetation. •How well can we predict these changes in vegetation? (TEM-DVM)

•What is the magnitude of climate-related changes in the productivity of the major ecosystem types (including their dominant plant functional types, PFTs) in northern Alaska?

•How can we use predictions in vegetation changes to gain an understanding of how herbivores may be impacted by predicted changes in vegetation? (TEM-DVM linked to CARMODEL)

•What types of processes / dynamics and habitats do vegetation models need to consider?

Spruce forest

Tussock tundra

Willow-birch tundra

Spruce, Salix spp., other deciduous shrubs, evergreen shrubs not including spruce, sedges, forbs, lichens, and feathermoss

Betula spp., other deciduous shrubs, evergreen shrubs, sedges, forbs, lichens, feathermoss and Sphagnum moss

Salix spp., Betula spp., other deciduous shrubs evergreen shrubs, sedges, forbs, lichens, & feathermoss

TERRESTRIAL ECOSYSTEM MODEL WITH DYNAMIC VEGETATION

AND LEAF, WOOD, AND ROOT COMPONENTS (TEM-DVM)

Linked to other submodels including a soil thermal model (permafrost dynamics; Zhuang et al., 2001) and hydrology model (snow dynamics; Euskirchen et al., 2007)

RH

NETNMINCS NS NLOST

NINPUT

SOIL

LightNAV

LightNAV

Atmospheric Carbon Dioxide (CO2)

NAV

Euskirchen et al., in press, Ecol. Applications

• The study region in northern Alaska is classified as 77% sedge tundra, 13% shrub tundra, and 8% evergreen needleleaf forest, with each of these ecosystems comprised of 8 or 9 PFTs.

• The model is parameterized based on field data collected in the study region.

Ch

an

ge in

air

te

mp

era

ture

(°C

year-

1)

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

MAM

SON

Mean annual

JJA

DJF

A2 C

SIR

O

A2

Had

. C

M3

A2 P

CM

B1 C

SIR

O

B1

Had

. C

M3

B1 P

CM

B2

CS

IRO

B2

Had

. C

M3

B2

PC

M Mea

n

We used a wide range of future climate scenarios, based on the IPCC SRES scenarios and three global climate models (GCMs). Shown below is input air

temperature data.

230

155

115

75

35

-5

(Sed

ge

)

Forb

Salix

Gra

ss

Salix

Feath

er

Betu

la Sp

ruce

Sed

ge

Decid

.

Fea

ther

Forb

Decid

. Sed

ge

Dec

id.

Sed

ge

E.g

ree

nForb

Lic

hen

E.g

ree

n Feath

er

E.g

ree

nSp

hag

.G

rass

Betu

laLic

he

nLic

hen

(Fore

st)

(S

hru

b)

(Sh

rub

)

ForestShrub tundraSedge tundra

Ch

an

ge in

NP

P,

2003-2

100 (

g C

m-

2)

Change in NPP (2003 – 2100) across all ecosystem types and PFTs,

arranged in descending order.

Euskirchen et al., in press, Ecol. Applications

Salix

0100

200

300

400

Betu

la

Decid

.

Everg

r.

Sed

ges

Forb

s

Gra

sses

Lic

hen

s

Feath

er.

01

23

45

6

Betu

la

Decid

.

Everg

r.

Sed

ges

Forb

s

Gra

sses

Lic

hen

s

Feath

er.

Salix

Change in Vegetation Carbon Change in Vegetation Nitrogen

Shrub Tundra 2003 - 2100V

eg

eta

tion

C c

han

ge (

g C

m-2)

Veg

eta

tion

N c

han

ge (

g N

m-2)

Increases in NPP resulted in increases in biomass, and consequently, decreases in

summer albedo.

0.1

0.12

0.14

0.16

0.18

0.2

2003 2023 2043 2063 2083

Su

mm

er

Alb

ed

o

Forest

Shrub tundra

Regional

Sedgetundra

The albedo of the shrub tundra is approaching that of the forest at the end of the simulation. Can the shrub tundra ecosystem still be a considered shrub tundra ecosystem?

How can TEM-DVM be used to relate vegetation changes to herbivores?

•Arctic calving caribou provide an example of a species that is likely to be affected by climate change, but with long annual migrations and a circumpolar distribution the effects on populations will not be uniform.

• An explicit energetics model (CARMODEL) is available that allows assessment of projected climate induced changes in potential forage intake on fat and protein dynamics.

CARMODEL (Russell et al., 2005)

En

erg

etic

s M

od

el

Energy Submodel

Daily forage intake

Hourly digestion

Rumen contentsNext hour

Cow weight

Growth Submodel

Daily Growth of Cow & Calf

• Maintenance• Activity• Gestation• Lactation

Daily Energy Allocation

Next day

Forage

• Biomass of a PFT• Digestibility• Cell wall content• Nitrogen content• Diet

Snow depth

Activity budgets

Reproductive strategy

• Conception date• Gestation period• Wean date

Future climate:• Air temperature • Precipitation • Solar radiation

Daily

Daily

Daily

Yearly

• Compare and contrast the implications of climate induced changes in forage biomass and quality on fat and protein dynamics in two groups of caribou with characteristically different diets at calving: a) caribou that calve in the wetter portions of the coastal plain (Central Arctic, Teshekpuk), and b) caribou that calve in uplands and drier portions of the coastal plain (Western Arctic, Porcupine).

• Analyses based on a range of climate scenarios will allow managers and subsistence users to more effectively understand the range of possible future conditions and to plan adaptation strategies.

1 kmX 1 km

Integration of the model presented here (TEM-DVM) with other land cover maps / habitat classification schemes:

• Finer scale representation of the ecosystem types within a landscape or region is possible.

• It is also possible to incorporate other ecosystem/habitat types than the three presented earlier.

• Some parameterization data for PFTs in a given habitat type is necessary.

• One recent model development is a dynamic soil layer module (Yi et al., manuscript in prep.) that incorporates different types of soil layers (e.g., live moss, humic, fibric) and drainage classes (e.g., moderately well-drained, poorly drained).

Future Work & Challenges:

• Incorporating shifts from one ecosystem type to another

• Incorporating changes in vegetation (at the ecosystem & PFT level) under changes in the fire regime and insect outbreaks

• Understanding how the phenology of the various PFTs responds to changes in growing season length in terms of both budburst in the spring and leaf senescence in the fall

• Incorporating water competition among the PFTs

• Increased levels of herbivory may lead to increases in productivity of ecosystems under climate change. How then does this increased herbivory feed back on the ecosystem? Field experiments suggest that the increased herbivory will ultimately modulate this increased productivity (e.g., Sjögersten et al., 2008).