racer: an innovative tool for guiding conservation in a climate-changed arctic

© L

ee

Na

rraw

ay/W

WF

-Cam

ada

Rapid Assessment of Circum-Arctic Ecosystem Resilience (RACER): An innovative tool for guiding conservation in a climate-changed arctic.

Canadian Society of Ecology & EvolutionKelowna, BC.

1

May 14th , 2013

James Snider*, Peter Ewins, Martin Sommerkorn*Advisor, Conservation Science & PracticeWWF-Canada

© M

onte

Hu

mm

el/

WW

F-C

anon

1. Conservation in a Changing World

2. Building “Resilience”

3. The RACER Methodology

4. Marine Case Study: Beaufort Sea

5. Terrestrial Case Study: Central Canadian Tundra

6. Current Applications

7. Lessons Learned

Outline

2

3

Conservation in a Rapidly Changing World

NASA Goddard Institute for Space Studies

4

Conservation in a Rapidly Changing World

5

Conservation in a Rapidly Changing World

Overland & Wang – Accepted, Geophysical Research Letters

Projections for the Future: Ice-free summers in the Arctic?

6

Projections in Arctic Vegetation Change:

Distribution of vegetation:Observed Predicted

Pearson et al., 2013. Nature Climate Change

7

How do we manage ecosystems under rapid change?

© Peter Ewins / WWF”

Building Resilience

Ability of a system to absorb disturbance and still retain its basic function and structure.

About maintaining/building

the capacity of systems

to adapt through change

(of the stability landscape)

Forward-looking, identifying

options for the future

(values, services)

Building Resilience

Requires an understanding of system functioning:

Drivers and their

trajectories (physical,

biological, social)

Ecosystem processes

and their response to

change

What role these and

feedbacks between them

play in building resilience

10

11

Rapid Assessment of Circum-Arctic Ecosystem Resilience (RACER)

Ecoregion-scale conservation planning

Landform heterogeneity

Primary productivity

RACER: (1)_Mapping Sources of Resilience: Areas of exceptional productivity and diversity confer resilience to the ecosystem

Literature data

© naturepl.com / Martha Holmes / WWF”

Asessing Persistence of Key Features:

wind surface temperature

soil moisture

nutrients ocean currents

sea ice

“DRIVERS”

An understanding of the functioning of a landscape / seascape that is not place-based, but process-based

14

RACER Marine Case Study: The Beaufort Sea

15

Drivers of an arctic marine shelf system:

Ingram et al. (2008)

Assessing Persistence for the Beaufort Sea:21st century sea ice projections for the Beaufort Coast and Continental Shelf ecoregion

Huard, 2010

Be

Be

Climate variables: Sea Surface Temperature (SST); Salinity; Sea-Ice thickness; Sea-Ice concentration (SIC); Precipitation (P); Surface Air Temperature (SAT).Persistence index: H – high; M – medium; L – Low

* relevant for the Mackenzie plume is the precipitation over the watershed of the Mackenzie River, i.e. outside the Beaufort coast and shelf ecoregion

Assessing Persistence of Key Features to projected change for the Beaufort Sea Coast and Continental Shelf Ecoregion

Key Feature Main drivers Current biological productivity & habitat heterogeneity

Main changes to GCM climate variables

Assessed persistence of Key Feature’s future above-average productivity / diversity

Barrow canyon & polynya Benthic topography Seasonal Ice Cover Water circulation/currents Sea Surface Temperature

High productivity and benthic habitat heterogeneity; warm saline Pacific water incursions.

SST Salinity SIC

H

Mackenzie canyon

Benthic topography Seasonal Ice Cover Water circulation/currents Sea Surface Temperature

High riverine plume nutrient inputs & heterogeneity, with upwelling driven by currents.

SST Salinity SIC

H

1. Mackenzie recurring shoreleads

Benthic topography Seasonal Ice Cover Water circulation/currents Sea Surface Temperature

Low absolute winter productivity, but open water regime allows light penetration/biotic activity.

SST Salinity SIC P

H-M

2. Kugmallit canyon Benthic topography Seasonal Ice Cover Water circulation/currents Sea Surface Temperature

High riverine plume nutrient inputs & heterogeneity, with upwelling driven by currents.

SST Salinity SIC

H

3. Mackenzie plume Salinity Nutrients Water circulation/currents Sea Surface Temperature

High sediment-laden nutrient inputs, but low habitat heterogeneity. Water circulation patterns influence nutrient availability.

SST Salinity SIC SAT P*

H-M

4. Cape Bathurst slope Benthic topography Water circulation/currents Sea Surface Temperature Nutrients

Habitat heterogeneity high, with resultant diversity of benthic fauna and current-induced nutrient availability.

SIC SST

H-M

Cape Bathurst-Amundsen Gulf polynya

Benthic topography Seasonal Ice Cover Water circulation/currents Sea Surface Temperature

Low absolute winter productivity, but open water regime allows light penetration/biotic activity.

SAT SST Salinity SIC

M

Continental shelfbreak and slope

Benthic topography Water circulation/currents

Low productivity currently in deep water, but very extensive high seabed habitat heterogeneity.

SIC Salinity

H

19

Integrating RACER Results into Spatial Planning:

20

Terrestrial Pilot Study Unit: Central Canadian Tundra

Raynolds et al., 2008

21

Mapping Areas of Exceptional Primary Productivity:

Normalized Difference Vegetation Index (NDVI)

10-year median of maximum summer NDVI

Significance calculated by highest percentiles for each bioclimatic subzone.

22

Validation of remote sensing for primary productivity:

1. Overlay of percentiles with known areas of biological importance:

• Key Bird Habitat Areas

• Protected Areas

• Caribou Calving Grounds

Remote Sensing Validation Part 2A:

CAVMVegetationClasses

Remote Sensing Validation Part 2B: Landsat

Northern Land Cover Classification (Olthof et al., 2009)

n = number of pixels% = percentage of total pixels

Fisher-Freeman-Halton Test, p-value = 0.000999alternative hypothesis: two.sided(Monte Carlo Estimation, s = 1000)

Statistical comparison of Landsat Landcover classification for significantly high NDVI pixels(relative to all other vegetated pixels within Bioclimatic Subzone #5)

Non-Sig Sig Total

Landsat Landcover Class (n) (%) (n) (%) (n) (%)

tussock graminoid tundra 62,538,979 29% 33,594 2% 62572573 29%

wet sedge 10,667,262 5% 28,686 2% 10695948 5%

moist to dry non-tussock graminoid / dwarf shrub tundra

19,066,292 9% 4,729 0% 19071021 9%

dry graminoid prostrate dwarf shrub tundra 6,259 0% 0 0% 6259 0%

low shrub (< 40cm; > 25% cover) 42,896,927 20% 168,375 12% 43065302 20%

tall shrub (> 40cm; > 25% cover) 14,980,136 7% 1,088,136 79% 16068272 7%

prostrate dwarf shrub 42,927,724 20% 3,403 0% 42931127 20%

sparsely vegetated bedrock 5,660,647 3% 736 0% 5661383 3%

sparsely vegetated till-colluvium 1,912,855 1% 1,348 0% 1914203 1%

bare soil with cryptogam crust - frost boils 3,509,303 2% 182 0% 3509485 2%

wetlands 9,383,754 4% 52,452 4% 9436206 4%

26

Landform Heterogeneity: “Topographic Position Index” (TPI)

Areas of Exception Landform Heterogeneity

Validation of Landform Heterogeneity Analysis

Comparison with beta-diversity of Landsat vegetation classes

Drivers of terrestrial arctic ecosystems

CAVM, 2003

31

Drivers of Arctic Vegetation: Surface Temperature

Raynolds et al., 2008

Drivers of Arctic Vegetation: Precipitation/Snow

Step 2: Vulnerability/Persistence Analyses Global Climate Modeling

Assessing Persistence for the Central Canada Ecoregion:Climate projections for precipitation & temperature

35

Current Applications: RACER in Antarctica

36

Assessing Suitability of Terrestrial Remote Sensing

37

Lessons Learned

1) Reframing the conservation target• Focus on nature’s potential to provide species• The “new” normal - may mean tough conservation decisions

(e.g. assisted migration, triage, accepting invasive species)

2) Get comfortable with uncertainty• Mis-matched rate of development & scientific rigor

3) From ecological to social-ecological • We manage people not ecological processes• Emphasize benefits of well functioning natural systems on humans

4) Tools inform decision making*Ideally imbedded with formal institutional structures.. (e.g spatial planning initiatives)

38

Thank you!

More on RACER? Panda.org/arctic/racer

jsnider@wwfcanada

@snider_james