Ecological forecasting in the intertidal zone: from MODIS to mussels

1
Ecological forecasting in the intertidal zone
from MODIS to mussels

• Brian Helmuth, David Wethey, Venkat Lakshmi,
  Jerry Hilbish, Allison Smith, Lauren Szathmary,
  Christel Purvis
• University of South Carolina, Columbia

2
Intertidal zone is an interface between marine
and terrestrial ecosystems
3
Recent worldwide observations of intertidal
mortality linked to climate
Dead mussels
New Zealand
Bleached algae
Necrotic tissue
Photo Laura Petes
Oregon
Washington
4
Potential causes of mortality, range shifts, and
loss of biodiversity

• Direct physiological effects (acute and chronic)
• Changes in aerial body temperature
• Changes in water temperature
• Indirect effects
• Competition
• Predation
• Factors not related to climate (e.g.,
  anthropogenic influences)

5
Ecological niche modeling

• Fundamental vs. realized niche space
• Not all species range edges are set by climate
• Should not expect to see impacts of climate
  change everywhere
• Climate change may impact organisms in the middle
  of their ranges
• Organism performance changes spatially and
  temporally ? include physiological data in niche
  modeling

6
Theoretical Models of Organism Body Temperature
Climate and Remote Sensing Data
Physiological and Ecological Data
Determine Realized Niche Space
Make Hypotheses
Experimentally Test Hypotheses in the Field and
Laboratory
7
Goals

• How do we make predictions about geographic range
  boundaries?
• Poleward migrations
• Mosaic patterns
• Where and when do we look for the current and
  future effects of climate change on ecological
  patterns?
• Biodiversity
• Abundance
• Mortality
• How do we mitigate these effects?

8
Two organisms exposed to identical microclimates
can experience different body temperatures
Seastar at 12C
Mussel at 21C
See details on Szathmary et al. poster
9
Thermal mosaic over a large geographic range
Shady Cove Cattle Point Tatoosh Boiler
Bay Strawberry Hill Monterey Piedras
Blancas Cambria Lompoc Landing Jalama Alegria Boat
House Coal Oil Pt
(Helmuth et al. 2006 Ecol Monogr)
10
Topex-Poseidon R/S Data for Tidal Height
Complex patterns are likely to occur worldwide
due to tidal regimes
11
How do we measure animal temperature over large
scales in physiologically meaningful ways?
Avg Max Mussel Temp
Critical Physiological Temp
Avg Mussel Temp
Avg MODIS Temp
See Purvis poster for more details
12
Quantifying effects of climate over a cascade of
scales
http//uae2.gsfc.nasa.gov/terra_sat.gif
Weather station data
Remote sensing data
In situ animal temperature data
Wide spatial coverage Broad temporal
coverage Noninvasive
High spatial resolution More field
intensive Directly relevant to animals
13
Thermal engineering model of animal
temp. (Inside out/Outside in)
Tair
Qstored
m, cp
Tbody
Twater
Tground
14
NASA/NOAA data used as inputs to thermal
engineering models
Variable Name Data Source (Satellite Platform) Agency Time Period Spatial Resolution Temporal Resolution
Air Temperature Reanalysis NCEP/NARR 1948-present 1979-present 200 km (global) 32 km (NARR) 6 hour avg 3 hour avg
LST MODIS (Aqua, Terra) NASA Aqua Aug 2002-present Terra Feb 2000-present 1 km, 5 km Daily, 8-day average
SST MODIS (Aqua, Terra) NASA Aqua Aug 2002-present Terra Feb 2000-present 4.89 km, 9 km Daily
Solar Radiation Pinker (GOES) 1996-present 0.5 deg hourly
Solar Radiation Reanalysis NCEP/NARR 1948-present 1979-present 200 km (global) 32 km (NARR) 6 hour avg 3 hour avg
Wave Height, Wind Speed TOPEX/POSEIDON (Jason-1) NASA 1992-present 6 km, 0.5 deg, 1 deg 5-day, 10-day
Wind Speed Reanalysis NCEP/NARR 1948-present 1979-present 200 km (global) 32 km (NARR) 6 hour avg 3 hour avg
Relative Humidity Reanalysis NCEP/NARR 1948-present 1979-present 200 km (global) 32 km (NARR) 6 hour avg 3 hour avg
NCEP National Centers for Environmental
Prediction NARR North American Regional
Reanalysis incorporates NASA R/S Data and NOAA
ground-based data as part of reanalysis
15
Model Performance vs. Field Data
too cold too hot
Difference in Monthly Average Maximum
(Gilman et al., PNAS 2006)
16
Model Performance vs. Field Data
too cold too hot
Difference in Monthly Average Maximum
(Gilman et al., PNAS 2006)
17
Air Temperature vs. Field Data
too cold too hot
Difference in Monthly Average Maximum
(Gilman et al., PNAS 2006)
18
Hypothesis testing

• Produce short-range (8-day) forecasts and test
  using physiological measurements of stress (hsps,
  etc.)
• Generate hindcasts of body temperature back to
  1950s using historical data compare against
  biogeographic data
• Make long-range predictions using GCM models
  predict shifts in biodiversity/ ranges

19
Ecological forecasting
8-day forecasted mussel temperatures in upper
intertidal from July 23, 2006
http//tbone.biol.sc.edu/forecasting
20
Patterns of mortality match forecasting
predictions???
Photo Laura Petes
21
Biogeography of Barnacles in Europe
Physiological Information Semibalanus
balanoides have reproductive failure if SST gt
10C in winter
Remote Sensing Information 10C winter isotherm
moved north between 1984 and 1998
Prediction S. balanoides southern biogeographic
limit moved north to the winter 10C isotherm
22
Arrows indicate the southern limit of S.
balanoides based on field surveys from Southern
Portugal to Denmark in 1984 and 2005
February 1984
February 1998
Sea Surface Temperature (AVHRR 36km)
23
Remote Sensing

• MODIS, ASTER Land Surface Temperature
• MODIS Sea Surface Temperature
• Climate Measurements

• Skin vs. Modeled Temp
• R/Sing Climatic Inputs

• Skin vs. Body Temp

• GPS Positioning

• Thermal Engineering Models
• Wave Run-up Models
• Tide Cycles

• Predator/Prey Studies
• Population Genetics
• Physiological Tolerances

• In situ vs. Modeled Temp

In Situ Laboratory Research
Ecological Modeling
24
Acknowledgments

• Woody Turner and the NASA Ecological Forecasting
  Team
• NASA grant NNG04GE43G
• Nova Mieszkowska, Sierra Jones, Karly Jones,
  Sarah Gilman, Srinivas Chintala
• Bernardo Broitman, Carol Blanchette and
  Packard-PISCO (U.S. West Coast)
• Steve Hawkins, Alan Southward and MARCLIM (Marine
  Biological Association, Plymouth, U.K.)
• Cliff Cunningham and CORONA