&EPA
United States
Environmental Protection
Agency
Proceedings of the
Plight of Ecosystems in a
Changing Climate-. Impacts
on Services, Interactions, and
Responses Workshop
MAY 27 - 28, 2009
PLYMOUTH CHURCH
SEATTLE, WA
Office of Research and Development
National Center for Environmental Research
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Table of Contents
Tier I: Effects of Climate Change on Ecosystem Services Provided by Coral Reefs
and Tidal Wetlands
Effects of Sea Level Rise and Climate Variability on Ecosystem Services of
Tidal Marshes, South Atlantic Coast 1
Christopher B. Craft, Samantha B. Joy, Steven C. Pennings, Dick Park,
Jeffrey Ehman, Jonathan Clough
Climate-Linked Alteration of Ecosystem Services in Tidal Salt Marshes of Georgia and Louisiana 2
Mark W. Hester, Irving A. Mendelssohn, Samantha B. Joye, MerrylAlber
Linking Impacts of a Climate Change to Carbon and Phosphorus Dynamics
Along a Salinity Gradient in Tidal Marshes 3
Melanie A. Vile, Scott C. Neubauer, D J. Velinsky
Connectivity in Marine Seascapes: Predicting Ecological and Socioeconomic Costs
of Climate Change on Coral Reef Ecosystems 4
James N. Sanchirico, Kenneth Broad, Dan Brumbaugh, Alan Hastings,
FiorenzaMicheli, Peter J.Mumby
Effects of Climate Change on Ecosystem Services Provided by Hawaiian Coral Reefs 5
Paul L. Jokiel, Robert Buddemeir, Herman Cesar, Daphne Fuatin
Tier II: Nonlinear Responses to Global Change in Linked Aquatic and Terrestrial
Ecosystems
Hydrologic Forecasting for Characterization of Nonlinear Response of Freshwater Wetlands
to Climatic and Land Use Change in the Susquehanna River Basin 6
Denice Heller Wardrop, Robert P. Brooks, Kevin Dressier, Christopher Duffy,
William Easterling, Raymond Najjar, Richard Ready, James S. Shortle
Sustainable Coastal Habitat Restoration in the Pacific Northwest: Modeling and
Managing the Effects, Feedbacks, and Risks Associated With Climate Change 7
John Rybczyk, W. Greg Hood, Tarang Khangaonkar, Enrique Reyes, Zhaoqing Yang
Nonlinear Response of Pacific Northwest Estuaries to Changing Hydroclimatic
Conditions: Flood Frequency, Recovery Time, and Resilience 8
Anthony F. D'Andrea, Robert A. Wheatcroft
Nonlinear Response of Prairie Pothole Landscapes to Climate Change and Land Management 9
Carter Johnson, Richard Adams, Phil Fay, Glenn R. Guntenspergen,
Bruce V. Millett, Richard Voldseth
Innovative Management Options To Prevent Loss of Ecosystem Services Provided
by Chinook Salmon in California: Overcoming the Effects of Climate Change 10
Peter Moyle, Lisa Thompson, DavidPurkey, Andrew Engilis, Jr., Marisa Escobar,
Christopher Mosser, Melanie Allen Truan
Hydrologic Thresholds for Biodivestity in Semiarid Riparian
Ecosystems: Importance of Climate Change and Variability 11
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 Hi
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Thomas Meixner, Kate Baird, Mark A. Dixon, James F. Hogan, S. Joy Lite, Julie Stromberg
Nonlinear and Threshold Responses to Environmental Stressors in Land-River Networks
at Regional to Continental Scales 12
Jerry Melillo, Bruce Peterson, Charles Vorosmarty, Benjamin Felzer, David Kicklighter,
James McClelland, and Wilfred Wollheim
Tier III: Ecological Impacts From the Interactions of Climate Change, Land Use
Change, and Invasive Species
Integrated Bioclimatic-Dynamic Modeling of Climate Change Impacts on
Agricultural and Invasive Plant Distributions in the United States 13
Wei Gao, Xin-Zhong Liang
Global Change and the Cryptic Invasion by Transgenes of Native and Weedy Species 15
Cynthia L. Sagers, Peter K. Van de Water
A Multi-Scale Approach to the Forecast of Potential Distributions of Invasive Plant Species 16
John A. Silander, Daniel Civco, G. Wang, I. Ibanez, A. Gelfand, C. Reid
Predicting Relative Risk of Invasion by the Eurasian Saltcedar and New Zealand
Mud Snail in River Networks Under Different Scenarios of Climate Change
and Dam Operations in the Western United States 18
N. LeRoy Poff, Gregor T. Auble, Brian P. Bledsoe, Denis Dean, Jonathan Friedman,
David Lytle, David M. Merritt, David Purkey, David A. Raff, and Patrick B. Shafroth
Integrating Future Climate Change and Riparian Land Use To Forecast the Effects
of Stream Warming on Species Invasions and Their Impacts on Native Salmonids 19
Julian D. Olden, Timothy Beechie, Joshua J. Lawler, Christian E. Torgersen
Understanding the Role of Climate Change and Land Use Modifications
in Facilitating Pathogen Invasions and Declines of Ectotherms 20
Jason R. Rohr, Andrew Blaustein, Thomas R. Raffel
Beach Grass Invasions and Coastal Flood Protection: Forecasting
the Effects of Climate Change on Coastal Vulnerability 21
Eric Seabloom, Sally Hacker, Peter Ruggiero
Elevated Temperature and Land Use Flood Frequency Alteration Effects on Rates
of Invasive and Native Species Interaction in Freshwater Floodplain Wetlands 22
Curtis J. Richardson, Neal Flanagan, Song S. Qian
Ecological Impacts From the Interactions of Climate Change, Land Use Change,
and Invasive Species 23
Robert B. Whitlatch, Richard W. Osman
Appendices
Agenda
Participants List
Presentations
Meeting Summary
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Effects of Sea Level Rise and Climate Variability on Ecosystem Services
of Tidal Marshes, South Atlantic Coast
Christopher B. Craft1, Samantha B. Joye2, Steven C. Pennings3,
Dick Park4, Jeffrey Ehman5, and Jonathan Clough6
Indiana University, Bloomington, IN; 2 University of Georgia, Athens, GA;3
University of Houston, Houston, TX; 4Eco Modeling, Diamondhead, MS;
Pangea Information Technologies, Ltd., Chicago, IL; Warren Pinnacle Consulting, Warren, VT
The investigators employed field and laboratory measurements, geographic information systems (GIS),
and simulation modeling to investigate how tidal marsh area and delivery of ecosystem services will be
affected by accelerated sea level rise (SLR) along the South Atlantic (GA-SC) coast. Different habitats of tidal
marshes provide different quantities of ecosystem services. For example, aboveground biomass was 40 to 70
percent greater in tidal freshwater and brackish marshes than in salt marshes. Tidal freshwater and brackish
marshes also provided greater waste treatment per unit area than did salt marshes. These marshes sequestered
three times more N in soil and supported two to three times greater potential denitrification than salt marshes.
Model simulations using the IPCC mean (52 cm) and maximum (82 cm) estimates of SLR by 2100 for the
Georgia coast suggest that salt marshes will decline in area by 20 percent and 45 percent, respectively. Tidal
freshwater marshes will increase by 2 percent under the IPCC mean scenario but will decline by 39 percent
under the maximum scenario. Delivery of ecosystem services associated with productivity (macrophyte
biomass) and waste treatment (N accumulation in soil, potential denitrification) also will decline. These
findings suggest that tidal marshes at the lower and upper salinity ranges and their attendant delivery of
ecosystem services will be the most affected by accelerated SLR unless geomorphic conditions (i.e., gradual
increase in elevation) enable tidal freshwater marshes to migrate inland, or vertical accretion of salt marshes
increases to compensate for accelerated SLR.
The effects of climate variability were evaluated by analysis of climate (rainfall, temperature, salinity,
freshwater discharge) and selected ecosystem services data collected from 2000 to 2006 from permanent plots
of 10 marshes of the Georgia Coastal Ecosystems Long Term Ecological Research (LTER) study domain. The
data revealed that river discharge was the most strongly correlated with the measured ecological variables.
Discharge was positively correlated with Spartina alterniflora aboveground biomass and sediment deposition.
S. alterniflora on the marsh plain also was positively correlated with precipitation. Salinity was inversely
correlated with freshwater discharge. Increasing salinity was associated with reduced S. alterniflora
aboveground biomass and greater numbers of fiddler crabs. There was no association between temperature and
the measured ecological variables.
This work provides a basis to: (1) understand how ecosystem services vary among salt-, brackish-, and
tidal freshwater marshes; (2) determine how sea level rise will alter marsh area and delivery of ecosystem
services; and (3) elucidate how climate variability affects temporal patterns of macrophytes, epifauna, and
sediment deposition.
Reference:
Craft C, Clough J, Ehman J, Joye S, Park D, Pennings S, Guo H and Machmuller M. Forecasting the
effects of accelerated sea level rise on tidal marsh ecosystem services. Frontiers in Ecology and the
Environment 2009;7:73-78.
EPA Grant Number: 832220
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Climate-Linked Alteration of Ecosystem Services in Tidal
Salt Marshes of Georgia and Louisiana
Mark W. Hester1, Irving A Mendelssohn2, Samantha B. Joye, and Merry I Alber3
1 University of Louisiana, Lafayette, LA; 2Louisiana State University, Baton Rouge, LA;
University of Georgia, Athens, GA
The investigators' objective is to elucidate the effects of climate change on tidal marsh ecosystem services
in tidal salt marshes of Georgia and Louisiana. The goal of this research is to better understand how the
ecosystem services of eutrophication control, carbon sequestration, sustainable habitat, and faunal support are
influenced by climate change, specifically increased drought severity, in salt marshes with tidal amplitudes
ranging from mesotidal (Georgia) to microtidal (Louisiana).
This research project takes advantage of a unique and timely opportunity afforded by recent, multi-year,
severe drought events in the tidal salt marshes of both Louisiana and Georgia that resulted in large areas of
sudden salt marsh dieback. Within each state, six large study areas will be identified in which permanent plots
will be established in habitats that represent a range of salt marsh response to drought from relatively
unimpacted, reference (high vegetation cover) to severely impacted (complete dieback and loss of vegetation
cover). Additionally, Spartina alterniflora, the dominant salt marsh grass, will be artificially established at low
and high stem densities within areas of complete dieback (bare) marsh as a mechanism of controlling plant
density independently from the drought-induced dieback. Alteration to the ecosystem services mentioned
above will be evaluated at several scales over two growing seasons.
The proposed research will greatly increase the understanding of how climate change and severe drought
events impact crucial salt marsh ecosystem services. By conducting this research in a natural laboratory that
brackets a range of hydrogeomorphic conditions (deltaic plain/microtidal to coastal plain/mesotidal), the data
generated on the effects of climate change on tidal salt marsh ecosystem services will have widespread
applicability and value to coastal managers.
EPA Grant Number: R832221
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Linking Impacts of Climate Change to Carbon and Phosphorus
Dynamics Along a Salinity Gradient in Tidal Marshes
MelanieA. Vile , Scott C. Neubauer , andD.J. Velinsky
1Villanova University, Villanova, PA; 2 University of South Carolina at Columbia, Columbia, SC;
3Academy of Natural Sciences, Philadelphia, PA
Tidal freshwater marshes are often located in areas experiencing intense urbanization pressure, yet they
provide valuable services to coastal ecosystems by acting as water quality filters (removing nutrients and
sediments), sequestering carbon [C] and phosphorus [P], serving as nursery habitat for fishes, and buffering
storm and flood waters. A climate change stressor that is unique to tidal freshwater systems is the intrusion of
salt water into environments that have historically been dominated by freshwater flows. The investigators
especially are interested in how the increase in S042" concentration associated with salt water intrusion will
affect the biogeochemical interactions that govern the cycling of C and P in tidal freshwater marshes and how
it will affect the flux of elements between marshes, tidal waters, and the atmosphere.
The investigators will implement a novel, three-phase approach to determine changes in tidal marsh
metabolism (e.g., C02 and CH4 gas fluxes and S042" reduction), C and P sequestration (sediment deposition
and burial), sediment P speciation, and porewater chemistry at sites along a low-salinity transitional gradient in
the Delaware Estuary. Phase 1 consists of field observations (as a space-for-time substitute) to assess current
ecosystem services provided by tidal freshwater and low salinity marshes, and allow the investigators to
predict how these services may change as a result of salt water intrusion. Phases 2 and 3 provide a more
detailed look at specific biogeochemical processes that impact cycling of C, P, and S. In Phase 2, laboratory
experiments using marsh cores exposed to low salinity levels (< 5 psu) will be conducted to study the short-
term (weeks to months) impact of increased salinity on marsh sediment C and P biogeochemistry. Phase 3
involves large-scale manipulations in the field (reciprocal transplanting of cores between tidal freshwater,
oligohaline, and mesohaline marshes) to examine longer term (~l-2 yr) ecosystem-level responses of marshes
to elevated salinity.
This research will improve the assessment of how ecosystem services provided by tidal freshwater
marshes are likely to respond to predicted changes in climate-induced sea level rise and salinity. It is expected
that a small increase in salinity in tidal freshwater wetland sediments will increase rates of decomposition (but
decrease rates of C burial and emissions of the greenhouse gas CHO, and cause a release of sediment-bound P
from the soils. The results from this project can be used to improve existing climate change forecast models
and will allow appropriate management to moderate the impacts of future climate change in low salinity tidal
marshes.
EPA Grant Number: R832222
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Connectivity in Marine Seascapes: Predicting Ecological and Socioeconomic
Costs of Climate Change on Coral Reef Ecosystems
James N. Sanchirico , Kenneth Broaa, Dan Brumbaugh ,
Alan Hastings4, Fiorenza Micheli5, and Peter J. Mumby6
1 Resources for the Future, Washington, DC; 2 University of Miami, Miami, FL;
3American Museum of Natural History, New York, NY; 4 University of California, Davis, CA; 5Stanford
University, Palo Alto, CA; 6University of Exeter, Exeter, United Kingdom
This research project seeks to integrate theory and data from ecology, biology, and the social sciences to
address major questions about the potential consequences of climate change on coral reef ecosystems. The
researchers will establish a general framework starting at the habitat scale that is linked with population
biology and socioeconomic models. This structure will allow systematic exploration of several core questions,
including: (1) How do local impacts, including overfishing and mangrove deforestation, affect the
vulnerability of Caribbean coral reefs to climate change? (2) When do socioeconomic responses to changes in
the ecosystem triggered by climate change stressors exacerbate the vulnerability of coral-reef ecosystems to
future stressors? and (3) What are the critical ecological and/or socioeconomic uncertainties for predicting
climate change impacts on ecosystem services that will yield the greatest returns from investigation? In all
questions, ecosystem services will be measured through the effects on fisheries, biodiversity, and
social/cultural systems.
The investigators will develop an integrated ecological-socioeconomic model that will be representative of
Caribbean ecosystems and be formulated in discrete time and space. Data for estimating ecological and
socioeconomic response functions are already being collected by this team in an ongoing National Science
Foundation-funded biocomplexity project. This unique data set will allow the investigators to highlight and
measure the effects of local processes that are typically averaged out in more aggregate climate change models.
The model will include explicit spatial processes, such as larval and adult/juvenile dispersal and movements of
fishers, along with dynamic adjustment responses to predict the vulnerability of coral-mangrove ecosystems to
climate change stressors. Given the large uncertainties in both the nature of the relationships and measurement,
the researchers will perform a value of information analysis to learn about the impacts of reducing
uncertainties on various ecological and socioeconomic criteria.
Taking advantage of ongoing model development and data collection analysis of Caribbean coral-reef
ecosystems, the goals in this study are to develop a new understanding of changes in ecological services due to
climate stressors, provide a framework for evaluating different management scenarios on ecosystem services,
and highlight mechanisms where climate stressors can cascade through the ecological and socioeconomic
systems triggering responses that increase the vulnerability of the ecosystem.
EPA Grant Number: R832223
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Effects of Climate Change on Ecosystem Services
Provided by Hawaiian Coral Reefs
Paul L. Jokiel, Robert Buddemeir , Herman Cesar , and Daphne Fuatin
University of Hawaii at Honolulu, Honolulu, HI;2 University of Kansas, Lawrence, KS;
3Cesar Environmental Economics Consulting, Arnhem, The Netherlands
A robust, modular model of reef and ecosystem services responses to both the long-term mean and the
short-term extreme event components of climate change will be developed from the wealth of ecological and
physiological data available for the coral and reef communities of Hawaii. Its output will be the input for the
socioeconomic models, which will translate the climate change scenarios into a comprehensive picture of
possible futures of the ecosystem services and socioeconomic sectors, activities, and costs for the region. The
model (as well as the environmental data used and a comprehensive inventory of Hawaiian corals) will be
available for both online use and download from a Web site (www.kgs.ku.edu/Hexacoral). providing for
community involvement through hands-on testing and feedback.
This research project will integrate and extend existing models to develop a comprehensive, scenario-
based analysis of the range of possible effects of global climate change on ecosystem services provided by the
coral reefs of the Hawaiian archipelago, and on the economic valuation of predicted changes. It will build on
an extensive base of coral, reef, environmental, and economic data and analyses already assembled for the
region, using targeted surveys and experiments to characterize five diverse case study sites that will sample the
region. Cross-scale (reef to Global Circulation Model [GCM] cell dimensions) and cross-domain (biological,
environmental, economic) analyses will be carried out and integrated using domain-based typologies to
classify sites and services, and to scale and integrate the impacts on services and values. A Geographic
Information System (GIS) will be used extensively for visualization, analysis, integration, and communication
of results.
In addition to systematic identification and valuation of potential changes in ecosystem services, broken
down by service, environmental type, and socioeconomic sector, the project will emphasize the unique
suitability of Hawaii and its indigenous culture for advancing methods of valuing both unused resources (the
Northwest Hawaiian Islands) and the cultural and spiritual, as well as aesthetic, services provided by coral
reefs. In addition to elucidating the interactions among climate change stressors and their relative effects on
multiple ecosystem services, the project will develop and disseminate a suite of new and useful technical,
methodological, and conceptual tools that will be broadly applicable to other systems.
EPA Grant Number: R832224
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Hydrologic Forecasting for Characterization of Nonlinear Response
of Freshwater Wetlands to Climatic and Land Use
Change in the Susquehanna River Basin
Denice Heller Wardrop, Robert P. Brooks, Kevin Dressier, Christopher Duffy,
William Easterling, Raymond Najjar, Richard Ready, and James S. Shortle
Pennsylvania State University, University Park, PA
The objectives of this research project are to characterize nonlinear responses to global climate change in
linked aquatic and terrestrial ecosystems through: (1) selection of a linked terrestrial-aquatic ecosystem that
provides critical ecosystem services and ecological functions; (2) characterization of various global change
scenarios, incorporating both climate and land cover, and a method of assessing their effect on the identified
ecosystem through the primary forcing factor of hydrology (both alone and in conjunction with other human-
associated stressors); (3) identification of potential nonlinear ecological responses (sensu Scheffer et al., 2002)
in the selected ecosystem as a result of these changes; and (4) estimation of the resultant change in ecosystem
services on a watershed and Basin-wide scale in the Susquehanna River Basin (SRB).
The general approach to investigating the response of freshwater wetlands to climatic and land use change
is based on the tools and products of four previous U.S. Environmental Protection Agency Science To Achieve
Results (EPA STAR) grants, and involves the following series of activities:
1. Develop scenarios of climate and land cover change, operating on a scale of decades, relevant to the SRB.
2. Using these scenarios, in conjunction with a coupled surf ace-ground water model, develop a number of
predictive hydrologic scenarios for a collection of 11-digit HUC watersheds representing a range of
human-associated land uses in the SRB.
3. Characterize the relationships between hydrologic and landcover parameters and ecosystem characteristics
and services in wetlands of various types in the SRB, focusing on those with preliminary evidence of non-
linearity and/or thresholds.
4. Utilize the predicted hydrologic scenarios to forecast changes in ecosystem services across the entire SRB,
clearly identifying where and when non-linearities and/or thresholds in response occur, utilizing a series of
unique statistical tools to develop a probability surface.
The investigators will develop a unique analytical method for prediction of climate and land cover change
impacts, incorporating the forecasting of hydrologic conditions, which can be used to identify thresholds and
non-linearities in the functional performance of freshwater wetlands. Any set of hydrologic/land cover change
conditions can then be placed on the probability surface, allowing the statistical model to be used in a
predictive fashion. The method could be applied to a wide variety of aquatic ecosystems for which state
changes occur over either a spatial or temporal extent, or both.
EPA Grant Number: R833013
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Sustainable Coastal Habitat Restoration in the Pacific Northwest: Modeling and
Managing the Effects, Feedbacks, and Risks Associated With Climate Change
John Rybczyk , W. Greg Hooa, Tarang Khangaonkar , Enrique Reyes , and Zhaoqing Yang
1 Western Washington University, Bellingham, WA; 2Skagit System Cooperative, La Conner, WA;
3Battelle Memorial Institute, Pacific Northwest Division, Sequim, WA;
4East Carolina University, Greenville, NC
The overall objective of this research project is to develop a predictive landscape simulation model,
incorporating non-linear feedbacks, of the ecological and geomorphological consequences of climate-induced
sea level rise and river flow alteration in two of the most ecologically significant estuarine systems in Puget
Sound, Padilla Bay, and Skagit Bay. The investigators will use the model to guide the course of restoration and
management efforts, given climate change, as they relate to salmon habitat in Puget Sound.
The investigators will develop and link a spatially explicit hydrodynamic and sediment transport model of
Padilla Bay and Skagit Bay to a mechanistic wetland elevation dynamics and vegetation unit model and
models of tidal channel geomorphology and juvenile salmon abundance and distribution. The linked models
will be initialized, calibrated, and validated using extensive site-specific data sets that the investigators have
already developed and the data that they have collected. The model will be run under various sea level rise and
river flow scenarios.
Effective and sustainable habitat restoration needs to anticipate future environmental conditions to ensure
that restoration efforts will be robust and capable of surviving anticipated climate change. The investigators
will use this model to examine how recovery goals (e.g., hectares to be restored) should be adjusted depending
on how much marsh progradation or erosion occurs over the next century, and will characterize regions in the
estuary that would be high- or low-risk restoration sites depending on their likely vulnerability or resilience to
climate change. It is precisely this "vulnerability/resilience" response to climate change that is nonlinear. The
investigators anticipate immediately incorporating this model into planning and management processes used
by local tribes, local restoration planning organizations (e.g., the Skagit Watershed Council), and regional
restoration planning organizations (e.g., the Northwest Indian Fisheries Commission, Washington Shared
Strategy, and the Puget Sound Nearshore Restoration Program, among others).
EPA Grant Number: R833014
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Nonlinear Response of Pacific Northwest Estuaries to Changing Hydroclimatic
Conditions: Flood Frequency, Recovery Time, and Resilience
Anthony F. D 'Andrea and Robert A. Wheatcroft
College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, OR
Pacific Northwest (PNW) estuarine soft-sediment habitats are productive systems that play an important
role in the biodiversity and functioning of coastal ecosystems and provide economically important biotic
resources and diverse ecosystem services. Rainfall intensity is on the rise, and the sediment yield from PNW
basins has increased. Consequently, sediment input to estuaries has increased in magnitude and intensity, and
the input rate of fine-grained sediment from the surrounding drainage basin is likely to have important effects
on estuarine ecosystem services. It also may interact nonlinearly to impact the structure and function of
intertidal benthic communities and facilitate colonization by non-indigenous species (MS).
The investigators conducted a manipulative field study simulating different frequencies of flood
sedimentation events (zero, one, or two events in a single rain season) and tracked the initial mortality and
recovery of the benthic community from these events using a combination of high resolution benthic sampling
and univariate and multivariate analyses of benthic community metrics. Particular emphasis has been placed on
identifying changes in functional biodiversity, documenting recovery times and potential hysteresis effects of
having two sedimentation events in a rain season, tracking mortality and recovery of important functional
groups, and changes to the populations of NIS. Parallel sediment samples were collected and analyzed to track
changes in important sediment properties that have direct or indirect effects on survival or habitat suitability to
the benthic community.
This study will develop an empirical and theoretical framework for predicting the effects of flood
sedimentation events on tideflat macrobenthic communities in PNW estuaries and how these changes affect
ecologically and economically important biotic resources and ecosystem services. This research will be used to
quantify the resilience of intertidal benthic communities and identify important structural changes that may
indicate a threshold or catastrophic shift in the benthic ecosystem in response to sedimentation events. Because
neither sufficient data nor models currently exist to conduct risk analyses, these datasets will significantly
improve our ability to perform ecorisk assessments in PNW estuaries, which can be used by resource managers
to make better informed decisions regarding actions to minimize or eliminate the risks to these systems.
EPA Grant Number: R833015
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Nonlinear Response of Prairie Pothole Landscapes
to Climate Change and Land Management
Carter Johnson , Richard Adams , Phil Fay , Glenn R. Guntenspergen ,
Bruce V. Millett1, and Richard Voldseth5
1 South Dakota State University, Brookings, SD; 2Oregon State University, Corvallis, OR;
3Agricultural Research Service, Temple, TX; 4 University of Minnesota - Duluth, Superior, Wl;
5U.S. Forest Service, Grand Rapids, MN; U.S. Geological Survey, Reston, VA
This research project involves a multi-disciplinary, multi-institutional project that examines the possibility
that the response of prairie wetland ecosystems to climate change may be nonlinear or threshold in nature.
Wetland ecosystems of the Prairie Pothole Region (PPR) in the northern Great Plains are extremely
vulnerable to climate change. While aspects of this vulnerability have been examined in previous research,
strongly suspected threshold responses of these wetlands to environmental drivers remain largely unstudied.
The objective of this research project is to identify possible future climatic and land use conditions that could
sharply reduce biodiversity in wetlands across the PPR.
A multi-step, integrated research framework will examine nonlinear responses through the use of a tested
mathematical model (WETLANDSCAPE) that links upland and wetland processes at the landscape scale.
Simulations will focus on critical environmental thresholds that control key ecosystem processes upon which
most wetland biodiversity depends. Terrestrial (upland) conditions and their management will be incorporated
explicitly as they influence the environment of wetlands down slope. The potential to use land management to
mitigate for possible negative consequences of climate change on prairie wetland biodiversity will be
examined using a land use decision model that embeds economic variables. This will allow quantification of
the economic costs of land use alterations to achieve ecosystem goals.
A primary outcome of this research will be to inform the scientific and management community, and
ultimately the public, of the existence of critical thresholds in the hydrologic environment of prairie wetlands
which, if exceeded by future climate forcings, could produce major negative consequences for biodiversity.
The possibility that amphibian and waterfowl numbers will greatly diminish in North America because of
climate change in the PPR is of great concern among public and natural resource management agencies. This
research will provide new understanding of the complex relationships among climate, wetland environment,
and the habitat base for these and other elements of biodiversity. The research also will suggest the degree to
which human adjustments (beyond reductions in greenhouse gases) such as land use changes can lessen the
severity of impacts of climate change on natural ecosystems in the PPR. Finally, this study will provide
preliminary information on the economic feasibility of alternative land use options and indicate the magnitude
of required societal costs to achieve such outcomes.
EPA Grant Number: R833016
The Office of Research and Development's National Center for Environmental Research and EPA Region 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Innovative Management Options To Prevent Loss of Ecosystem Services
Provided by Chinook Salmon in California: Overcoming the Effects
of Climate Change
Peter Moyle1, Lisa Thompson1, David Purkey2, Andrew Engilis, Jr.1,
Marisa Escobar , Christopher Mosser , and Melanie Allen Truan ,
University of California—Davis, Davis, CA; Stockholm Environment Institute, Davis, CA
In this research project, investigators are using an integrated water resources management model
(WEAP21) to simulate potential changes in flow and temperature in the salmon spawning reaches of Butte
Creek, California, in response to climate change. The resulting data are being used to drive a fish population
model (SALMOD) that simulates response to changing environmental conditions, including threshold effects
on survival. Literature reviews, field surveys, and an expert panel are being used to develop a conceptual
model of the impacts of changes in the salmon marine-derived nutrient subsidy to terrestrial wildlife.
The basic objective of the research is to determine the flow and temperature thresholds that lead to long-
term losses or reductions in spring-run Chinook salmon in Butte Creek. Hypothesis 1: Climate induced
changes in flow and temperatures in Butte Creek will lead to critical reductions in the available habitat of
spring-run Chinook salmon. Hypothesis 2: The loss/reduction of Chinook salmon will reduce the diversity
and abundance of birds and mammals in the riparian corridor. The final objective is to evaluate management
options to ameliorate these impacts.
The approach to assessing non-linear and threshold responses to gradual climate change on spring-run
Chinook and the dependent terrestrial ecosystem services will be both analytical and expert-panel based. The
primary, linked analytical models are WEAP21—an integrated watershed hydrology, water and irrigation
management, and water quality model, and SALMOD—a population dynamics model that predicts the growth,
survival, and movement (habitat choice) of salmon in freshwater systems from spawning to the egg, juvenile,
and smolt life stages, based on water quantity and quality conditions. Model results, along with the knowledge
base of the study team, will provide information for expert panels in Years 2 and 3 of the project. These
experts will help assist in the evaluation of potential impacts of climate change and management policies to
address these impacts.
Expected results include greater insight into the sustainability of spring-run Chinook salmon and their role
in defining the terrestrial biodiversity of the riparian corridor. Bringing climate change to bear on the issues
will determine environmental thresholds that also will be decision-making thresholds. The investigators will
provide various stakeholder and management groups with a set of tools and new information to help
determine: (1) if salmon are in increased danger from climate change; (2) if there are strategies to save the fish
and fish-dependent wildlife species from climate change effects; and (3) when and how these strategies can be
implemented. The analytical process and expert panel opinion will lead to: (1) possible water management
strategies to counter climate change impacts on stream ecosystems and the services they provide; and (2) an
improved understanding of the potential tradeoffs between services provided by water diversion versus
services provided by water left in the stream.
Analytical tools developed will be made available to the research and water management communities.
Dr. Lisa Thompson (Co-Pi), who has an appointment in the University of California Cooperative Extension
(UCCE), will extend academic information about California inland fisheries to stakeholders such as private
landowners and government officials. David Purkey has worked with the U.S. EPA Office of Research and
Development to extend the WEAP21 modeling framework to incorporate climate change, and it was used in
the recent California Governors Report on Climate Change (http://www.climatechange.ca.gov/). Thus, the
results of this work will be relevant for water management decision makers far beyond the Butte Creek basin.
EPA Grant Number: R833017
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 10
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Hydrologic Thresholds for Biodiversity in Semiarid Riparian Ecosystems:
Importance of Climate Change and Variability
Thomas Meixner, KateBaird, Mark A. Dixon, James F. Hogan, S. Joy Lite, and Julie Stromberg,
University of Arizona, Tucson, AZ
Riparian ecosystems of the arid and semiarid Southwest are linear corridors of high productivity and
diversity. These ecosystems are sensitive to even small changes in the riparian water balance, with sharp
changes in vegetation as streams become intermittent and as groundwater declines below survivorship
thresholds. As a result, riparian vegetation has declined on many rivers due to water abstraction or has been
altered due to the hydrologic impacts of climate variability. Despite much disciplinary work on individual
rivers, a regionally comprehensive and integrated understanding of how aquatic-terrestrial ecotones respond to
hydrologic change, including those imposed by climate change, awaits development.
The investigators will determine region-wide sensitivity of riparian vegetation to climate change. Project
hypotheses include: (1) decadal scale climate change and variability alter riparian aquifer recharge through
mechanisms that depend on the magnitude, frequency, and seasonality of flooding, and exert the greatest
change in reaches that receive minimal groundwater inflow from the regional aquifer; (2) riparian vegetation
structure responds non-linearly as riparian aquifers are dewatered and as key hydrologic thresholds for
survivorship of plant species are exceeded; and (3) decadal scale climate variability and change alters riparian
ecosystem water budgets that in turn changes vegetation structure and function and the ecosystem services
provided to society.
For hypothesis 1, the investigators will: isotopically quantify riparian aquifer recharge along a regional
precipitation gradient. On one river, the San Pedro, a model that links storm flow and aquifer recharge,
calibrated with isotopic data, to estimate steam base flows and seasonal aquifer conditions will be developed.
For hypothesis 2, the investigators will: further evaluate established connections between vegetation
condition and hydrologic conditions of flood flows, groundwater depth, and stream flow permanence. For
hypothesis 3, the investigators will: develop five alternative scenarios of climate change and use a scenario
driven model to estimate the climate impacts on vegetation along the San Pedro River. For the other rivers,
climate scenarios and hydrologic and vegetation data will be used to develop a climate change sensitivity
matrix. Biodiversity and water quality ecosystem services of riparian systems will be quantified for each
scenario.
This research project will produce three useful products for resource managers in the Southwest. First, the
research will improve understanding of the linkages between climate (precipitation timing and amount),
hydrologic variability (stream flow and aquifer conditions), vegetation structure, and ecosystem services in
riparian ecosystems, and of the regional variability in these relationships. Second, a transferable coupled model
of hydrologic-vegetation processes in riparian ecosystems that will allow for modeling of non-linear responses
to hydrologic change resulting from climate change or other causes will be produced. Third, the climate
sensitivity matrix that is developed will be useful for projecting regional impacts of climate change and
anthropogenic impacts on riparian water budgets and ecosystem change.
EPA Grant Number: R833025
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 11
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Nonlinear and Threshold Responses to Environmental Stressors in
Land-River Networks at Regional to Continental Scales
Jerry Melillo , Bruce Peterson , Charles Vorosmarty , Benjamin Felzer
David Kicklighter1' James McClelland1, and Wilfred Wollheim2
1The Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA;
2 Complex Systems Research Center, University of New Hampshire, Durham, NH
Ecosystems of the United States are subject to a variety of human-caused stressors, including changes in:
climate, the chemistry of the atmosphere, the chemistry of precipitation, and land cover and land use. These
stressors can act singly or together to elicit nonlinear and threshold responses in freshwater ecosystems and
alter their capacity to deliver ecosystem services such as sufficient quantities of clean water. In this research
project, the investigators will explore how a set of environmental stressors acts to affect the physical, chemical,
and biological integrity of linked land-river networks using a coupled terrestrial-aquatic ecosystem model that
is process-based and is applied in a georeferenced context within drainage basins across the United States. The
research project will have two parts: building the linked land-river network model, and using the model in
both retrospective and prospective studies. Use of the model will be guided by two hypotheses: (1) nonlinear
and threshold responses in the coupled land-water systems are key to defining the observed variations in water
quality across the United States during the last 100 years, transforming and intensifying local and in some
cases regional-scale problems to fully continental-scale syndromes; and (2) future policy interventions can
slow and sometimes reverse these problems and syndromes, but the interventions will be complicated by the
reality of new stable states and the heritage of existing threshold responses requiring many years to reverse.
The research plan includes two workshops involving the science team, resource managers, and policy makers.
At the first workshop, the investigators will develop a set of "what if scenarios that include specific policy
interventions and use them in simulations. At the second workshop, the investigators will analyze how these
interventions affect nonlinear and threshold behaviors in the freshwater ecosystems within drainage basins, and
what the consequences will be for ecosystem services. This research will contribute significantly to the
development of a theoretical basis for effectively protecting and managing ecological systems that exhibit
nonlinear and threshold responses to environmental stressors. The successful development of research and
management tools, such as the ones we are proposing, will help scientists to predict ecological thresholds
before they are observed. These research tools also will help resource managers and policy makers select
among alternative courses of action as they work to maintain, and in some cases enhance, the services provided
to us by ecosystems.
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 12
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Integrated Bioclimatic-Dynamic Modeling of Climate Change Impacts on
Agricultural and Invasive Plant Distributions in the United States
Wei Gao andXin-Zhong Liang
Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO;
2 University of Illinois, Urbana, IL
Biological invasions of nonindigenous plants and pests are serious threats to U.S. natural and managed
ecosystems, causing more than $120 billion per year of major environmental damages and losses. In
agriculture alone, $27 billion per year is estimated for the crop production lost from alien invasive weeds and
herbicide application expense. It is well established that climate is the dominant determinant of the geographic
distribution of plant species, native or alien. This distribution is confined by the prevailing bioclimatic limits in
the regional resources of light, heat, water, and nutrients. Given the rapid growth in worldwide trade or
globalization, long-range transport of non-native plant species across national boundaries becomes
increasingly important, exacerbating U.S. invasive species problems. Although humans facilitate the initial
establishment, the invasion, spread, and subsequent distribution of nonindigenous species may be controlled
largely by local environmental factors. Recent climate change, such as general warming, earlier spring, longer
growing season, decreasing winter frost period, and altered hydrologic cycle has already caused unequivocal
shifts in the distributions and abundances of species, and even pushed certain native species to extinction.
The objective of this study is to quantify and understand the impacts and uncertainties of regional climate
changes from the present to 2050 on the U.S. agricultural and invasive plant species distributions, emphasizing
crop production, and to account for both adaptation of alternative crops and invasion of non-native species to
enable decision makers to design effective management and control strategies for a sustainable future
agroecosystem. The original contribution of this research will derive from the application of a state-of-the-art
bioclimatic-dynamic ensemble forecast system that integrates a species environmental matching or niche
modeling component (SEM) with a high-resolution dynamic regional climate-ecosystem predictive component
(CEP) over North America. Both components incorporate multiple alternative models representing the likely
range of climate sensitivity and ecological response under the conceivable anthropogenic emissions scenarios
to rigorously assess the resulting uncertainty to improve risk analysis. This study will account for both
adaptation of alternative crops and invasion of non-native species in response to projected climate changes.
Historical simulations of the observed climate and crop production first will be conducted using the CEP to
provide the best proxy of the actual soil and bioclimatic conditions fundamental to the plant survival and
reproduction. This module can generate a high-resolution (10-30 km in this study), physically consistent and
most complete list of climate variables.
The high-resolution CEP-integrated bioclimatic predictors, including total plant productivity as input, will
be used to establish the SEM functional relationships of species distributions with these environmental
envelopes. The optimized ensemble of multiple CEP and SEM component models driven by four combinations
of regional climate models/global climate models (RCM/GCMs) and emissions scenarios will be used to
represent the most plausible range and uncertainty of future projections of U.S. agricultural and invasive plant
species distributions in the 2050s. The coupled CEP will be used to study climate-crop interactions, focusing
on how they affect U.S. agricultural productivity at the present and in the future. The representative GCM
projected and RCM downscaled climate changes will be used in this study. The recent RCM incorporates the
most comprehensive surface boundary conditions and advanced physics schemes that improve surface-
atmosphere and convection-cloud-radiation interactions. More importantly, it has been coupled with
comprehensive crop growth models to realistically simulate U.S. crop yields. The coupled RCM-crop
modeling system will serve as the key CEP to predict the climate and crop production conditions needed for
the development and application of the ensemble SEM system. These conditions will be used as input to
develop a robust SEM to best capture the observed agricultural and invasive plant species distribution. Future
projections for the potential niche distributions of alternative crops adaptable to the likely range of climate
changes in the 2050s will subsequently be made using the CEP. These CEP simulations of the future soil and
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 13
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
bioclimatic conditions will be integrated by the SEM to project the geographic distribution and abundance of
U.S. agricultural weeds and invasive plant species in the 2050s.
Through the proposed application of this unique ensemble forecast system, the investigators will make
major contributions to the key goal of the U.S. Department of Agriculture Cooperative State Research,
Education, and Extension Service (USDA CSREES) to enhance protection and safety of the Nation's
agriculture and food supply. The advanced state of the system components will result in a more in-depth
understanding of complex interactions among regional climate and land use, focusing on agricultural crop
production and invasive plant species across a full range of spatial and temporal scales. The investigators
expect to model the risks associated with several high-profile, costly agricultural weeds in the United States.
By using a conceivable range of climate scenarios, we will evaluate, with a credible estimate of associated
uncertainties, how these weeds may change in future distribution across a wide suite of crop types and
environmental envelopes. This will lead to better targeting of harmful invasive species in response to climate
change. It is expected that that the results will greatly surpass the capability of existing studies for climate
change impacts on future U.S. agricultural productivity.
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 14
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Global Change and the Cryptic Invasion by Transgenes
of Native and Weedy Species
Cynthia L. Sagers and Peter K. Van de Water
Department of Biological Sciences, University of Arkansas, Fayetteville, AR
According to the U.S. Department of Agriculture (USDA): "The sustainability of agriculture, forest and
rangelands depends on understanding the factors that influence climate change, the mechanisms that may
enhance or mitigate this change, and its effects on food and fiber production and natural resources." (USDA:
http://www.csrees.usda.gov/). A global issue in agriculture is the increasing incidence of herbicide resistant
weeds. Weeds may become resistant to herbicides by mutation or by gene flow from sexually compatible crop
species genetically modified for herbicide resistance. The adventitious presence of transgenes in the
environment represents a potential threat to U.S. agriculture, and is an understudied aspect of global change.
The investigators have adopted commercially available canola genetically modified for herbicide resistance as
a model system. Canola is sexually compatible with a number of weeds in the United States; this project will
focus on field mustard (Brassica rapa L) and black mustard (Sinapis arvensis}. The investigators will travel to
sites in the midwestern United States to collect weeds and their seed progeny to: (1) evaluate the incidence of
gene flow from crop to weed, and (2) to assess population variability in the likelihood of hybridization. The
population measures, including flowering phenology and sexual compatibility, will be mapped and merged
with predictive models of climate change in the United States. The result will be an understanding of regional
variation in the likelihood of transgene flow, a predictive model of how these risks will change in the advent of
climate change, and a heightened awareness of the impact of global change on agriculture and food supply in
the United States.
The primary goal of this research project is to develop a predictive model of how populations of plant
agricultural pests may expand or contract in the face of climate change. The study system is genetically
modified canola (Brassica napus L. \BrassicaceaJ) and native and weedy plant species that are sexually
compatible with canola. To this end, the investigators will conduct plant surveys of the upper midwestern
United States where canola is currently an important crop system, greenhouse experiments to evaluate
population variability in compatibility, and GIS modeling efforts that incorporate these data with accepted
models of predicted climate change. This collaborative work will involve scientists from the University of
Arkansas, California State University (CSU), Fresno, and the U.S. Environmental Protection Agency's
National Health and Environmental Effects Laboratory, Western Ecology Division. It is anticipated that one
postdoctoral fellow and two graduate students will be recruited to the project. The results of the work will be
published in a series of peer-reviewed publications (at least three), one review article, four papers presented at
national or international meetings, and a symposium arranged by the collaborators to be held in the last year of
the project.
This research approach adopts methods from plant population biology and rapidly evolving geospatial
technologies. The investigators visit sites in the upper midwest (Montana, North Dakota, Minnesota, and
Wisconsin); midwest (Iowa, Illinois); and southeast (Arkansas, Oklahoma) to collect seeds of weed pests that
are sexually compatible with canola (primarily B. rapa and S. arvensis). These seeds will be used to address a
number of questions that include estimating the rate of gene flow from GM herbicide resistant canola, and
determining population variability in sexual compatibility with canola. Greenhouse studies will be completed
at the University of Arkansas, Fayetteville. Spatially explicit information regarding rates of gene flow, sexual
compatibility, and environmental data will be incorporated into a GIS layer, which in turn will be incorporated
into an emerging predictive model of climate change. The majority of the geospatial modeling will be
completed at CSU, Fresno. This work constitutes a novel approach to assessing the risks of transgene escape in
the face of climate change. This project is unique in melding traditional plant population biology with
emerging spatial technologies.
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 15
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
A Multi-Scale Approach to the Forecast of Potential
Distributions of Invasive Plant Species
John A. Silander, Daniel Civco, G. Wang, I. Ibanez, A. Gelfand, and C. Reid
University of Connecticut, Starrs, CT
Controlling and preventing the spread of invasive plant species are common goals among ecologists and
natural resource managers. Because these goals often are most successful when initiated early in the invasion
process, the ability to predict where invasives will spread is crucial. The objective of this research project is to
explain the distribution and abundance of invasive plants across the northeast United States as a function of
climate and land use, and then forecast their future spread across the region to mid-21st century. To achieve
reliable predictions on invasive species spread, the investigators propose a comprehensive approach that will
take into consideration the major variables that will shape plant invasions in the next few decades (i.e., climate
change, land use change, and the effects of elevated atmospheric C02).
The investigators will integrate experiments with predictive modeling to study plant invasions by focusing
on the factors affecting their establishment and spread at four spatio-temporal scales: (1) regional-level, in
which distributional ranges, based on the response to climate, will indicate the broad tolerance limits of each
species; (2) landscape-level, in which incorporating the structure and composition of the landscape will inform
predictions on the land use attributes that promote the spread and population growth of invasive species;
(3) local-level, in which local site attributes (e.g., habitats, microclimates, soils, biotic interactions, etc.) will
inform of establishment thresholds for these species; and (4) individual-level, in which changes in drought and
shade tolerance will be examined under elevated atmospheric C02. The focus is to identify where specific
species could establish and increase in abundance as successful invaders now and in the future.
An integral component of this project is to incorporate education and outreach for the public at large,
professionals, and scientists. The investigators will use as a model the outreach and networking tools that they
have implemented through the IPANE project (Invasive Plant Atlas of New England). The IPANE project has
developed extensive educational and outreach materials on invasive species through its Web site, IPANE.org.
Output from this project will be incorporated on model-prediction Web sites. The investigators plan to present
the results of this research at regional, national, and international meetings of relevant scientific societies (e.g.,
the Ecological Society of America) each year during the course of the project. It is anticipated that the results
of the research will be published in peer reviewed journals that focus on ecology, climate change, invasive
species, and related issues (e.g., Ecological Applications, Biological Invasions, Global Change Biology, etc.).
The investigators also will consider submitting articles, when appropriate, to high-profile general science
journals.
Using the IPANE data set (species presence/absence, canopy closure, habitat type, etc.) with climate and
land use and land cover (LULC), hierarchical Bayesian (HB) models will be constructed to predict potential
distribution of selected invasive species. This approach provides for the specification of uncertainty in model
components, as well as the predictions, and accepts prior knowledge and data from multiple sources. Regional
predictions of future climates, focusing on projected changes in temperature and soil moisture, then will be
incorporated. The climate models will be identified with co-Pi Wang after examining the temperature and soil
moisture changes projected by each of more than 20 global climate models (IPCC AR4). Climate projections
from the North American Regional Climate Change Assessment Program (NARCCAP) also will be examined
using forecasts for the middle of the 21st Century. The investigators will develop predictive LULC-change
models, using LULC-change data from co-Pi Civco. Co-Pi Gelfand will develop and implement the LULC
change models for the region. To evaluate the process of successful establishment of invasive species, in the
context of new climates, varying establishment factors, and new biotic environments, the investigators will
conduct a large-scale transplant study of invasive plant species across the region; this includes planting sites
from southern Connecticut to northern Vermont. Demographic variables will be estimated as functions of
environmental covariates using R and OpenBUGS software. Co-Pi Reid will implement a C02 enrichment
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 16
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
experiment with representative invasive and native species, grown under ambient and elevated (mid-21st
Century) C02, under an array of watering and light levels; this allows quantification of the potential
demographic advantage that projected elevated C02 levels may bring to some species.
The major objective of this project is to provide potential distribution maps and site information on
potential establishment and abundance of invasive plant species across the region now and in the future.
Predictions based on experimental data will reflect realistic plant responses to environmental conditions. This
model approach will provide measurements of the uncertainty in predictions, one of the advantages of using
statistical hierarchical Bayesian models. These models will be evaluated in part using Deviance Information
Criterion and cross validation analyses. Data documentation, data files, and model descriptions will be made
available through the IPANE Web Site. Periodic self-evaluation will be conducted by the project Pis.
Independent evaluation of the project will come from peer reviews of manuscripts submitted for publication.
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 17
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Predicting Relative Risk of Invasion by the Eurasian Saltcedar and New Zealand
Mud Snail in River Networks Under Different Scenarios of Climate Change
and Dam Operations in the Western United States
TV LeRoy Poff, Gregor T. Auble , Brian P. Bledsoe , Denis Dean , Jonathan Friedman ,
David Ly tie3, David M. Merritt4, David Pur key5, David A. Raff, and Patricks. Shafroth2
1 Color ado State University, Fort Collins, CO; 2U.S. Geological Survey, Fort Collins, CO;
3Oregon State University, Corvallis, OR; 4U.S. Forest Service, Washington, DC;
5Stockholm Environmental Institute, Davis, CA; 6U.S. Bureau of Reclamation, Denver, CO
Predicting the spread and establishment of invasive species in river ecosystems under climate change
requires developing models that mechanistically link species population success to climate-sensitive
environmental drivers. The goal of this research project is to build a general and mechanistic framework with
which to predict the future potential distribution of two invasive species expected to expand their ranges under
a warming climate in streams and rivers of the western United States. The investigators hypothesize that local
site invasibility will be regulated by climate-sensitive thresholds of hydrogeomorphic disturbance, which will
vary throughout river networks in response to reach-scale channel geomorphology, future precipitation
regimes, and operation of dams, which modify natural flow regimes.
In a geographic region predicted to support saltcedar snails in the near future, the investigators will
downscale projected scenarios of temperature and precipitation as inputs to the Water Evaluation and Planning
(WEAP) model framework, allowing generation of streamflow regimes at ca. 50 km2 sub-basins based on
precipitation and water management operations (including dams). An artificial neural network (ANN) model
will be used to spatially distribute the WEAP hydrologic predictions throughout river networks at the reach
scale (100s of meters). These reach-scale flow regime predictions, in conjunction with GIS-derived measures
of channel and valley bottom geomorphology, will allow application of the biological model to assess the most
likely locations in river networks for successful saltcedar and mud snail invasion given the flow-mediated
disturbance regimes of any of several future climate scenarios. Further, using the coupled WEAP-ANN model,
the investigators will explore how a range of water management operations might influence the likelihood of
invasive establishment in these climate contexts. Finally, innovative stochastic population models will be used
to evaluate the probability of long-term success of the invasive species across a range of habitat vulnerability.
This synthetic, multi-scale approach will generate a sequence of spatially explicit maps that will provide
science guidance to support strategic decision-making regarding the spatially distributed risk of, and possible
adaptation to, the spread of invasive species at local to regional scales in the western United States. The model
will be general enough that it can be applied to other riverine species and resources, including non-invasive
species.
Grant Number: R833833
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 18
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Integrating Future Climate Change and Riparian Land Use To Forecast the Effects
of Stream Warming on Species Invasions and Their Impacts on Native Salmonids
Julian D. Olden , Timothy Beechie , Joshua J. Lawler , and Christian E. Torgersen
University of Washington, Seattle, WA;2National Oceanic and Atmospheric Administration (NOAA),
Seattle, WA
This project develops and applies an analytical framework that quantifies how future climate change and
riparian land use influences the direct and indirect effects of invasive species on the survival of Pacific salmon
in the John Day River in Oregon. Climate change, increasing agricultural land use, and invasive species
threaten the functioning of freshwater ecosystems in the Pacific Northwest. Elevated stream temperature is one
of the most pervasive water quality issues in this region, and projected climate change and riparian vegetation
loss are predicted to exacerbate this problem. Rising temperatures have direct implications for coldwater native
salmon, but they also will alter the composition of aquatic biota by facilitating range expansion and altering the
impacts of warm water invasive species.
The investigators will integrate climate-change projections, geomorphic sensitivity, riparian land use,
stream thermodynamics, and ecological niche modeling to quantify the potential range expansion and
temperature-mediated impacts of invasive smallmouth bass (Microptems dolomieu) and northern pikeminnow
(Ptychocheilus oregonensis) in critical habitats that support endangered Chinook salmon (Oncorhynchus
tshawytscha). The proposed work will: (1) predict spatiotemporal patterns of riverine thermal regimes in
response to future climate change, geomorphic sensitivity, and riparian land-use; (2) forecast species-specific
responses to projected future thermal regimes; and (3) evaluate alternative scenarios of climate change to
identify critical opportunities for riparian habitat restoration and protection to mediate future climate-induced
warming of streams and species invasions.
This project provides both the science and decision-support tools required to forecast with certainty how
the interactive effects of climate change, land use change, and invasive species will affect native salmon in the
future. Model results provide spatially explicit predictions of the vulnerability of adult and juvenile Chinook
salmon to the direct effects of stream warming associated with climate and land use change, and the indirect,
temperature-mediated effects of smallmouth bass and northern pikeminnow range expansion. Model outputs
improve the scientific capabilities for guiding management strategies and policies aimed at minimizing the
future range expansion of invasive species through protection and restoration of riparian vegetation that creates
and maintains a coolwater habitat. More broadly, this project and the analytical framework it developed is
readily applicable to other species of concern and relevant in other river systems of the Pacific Northwest,
where the range expansion of warmwater fishes in response to climate change and riparian-habitat loss is
ongoing and of imminent threat to native fishes.
EPA Grant Number: R833834
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 19
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Understanding the Role of Climate Change and Land Use Modifications
in Facilitating Pathogen Invasions and Declines of Ectotherms
Jason R. Rohr , Andrew Blaustein , and Thomas R. Raff el
1 University of South Florida, Tampa, FL; 2 Oregon State University, Corvallis, OR
Invasive parasites of humans and wildlife are arising at an unprecedented rate and are debilitating our
ecosystems. For instance, pathogens have been implicated in many amphibian declines that are triggering state
changes and impairing ecosystem functions. Climate change and land use modifications might elicit disease
emergence, but few generalizations have materialized for how these factors facilitate parasite invasions. The
investigators recently documented immunosuppression in amphibians associated with agrochemical exposure
and temporal climatic variability, stimulating the agrochemical spread and climatic variability hypotheses.
These hypotheses predict that proximity to agriculture (a global land-use modification) and elevated temporal
variability in temperature (due to climate change), respectively, compromise host immunity and facilitate
parasite invasions. In preliminary work, both temperature increases and decreases caused suboptimal
immunity, but drastic seasonal drops in temperature caused the longest periods of suboptimal immunity,
stimulating the hypothesis that cold-tolerant parasites will benefit most from elevated climatic variability
driven by global climate change. The investigators propose to test these hypotheses on multiple parasites and
ectothermic taxa, but intentionally focus on the invasive Batrachochytrium dendrobatidis and amphibians
because this emerging chytrid fungus is cold-tolerant and implicated in many of the global amphibian declines.
The investigators will test these hypotheses by: (1) examining whether the timing of apparently disease-
induced amphibian extinctions in Central and South America are related to climatic variability, proximity to
agriculture, or alternative factors; (2) testing whether the distribution of extinct and threatened ectothermic
species worldwide is positively associated with the spatial pattern of climatic variability and agriculture across
the globe; and (3) conducting a series of manipulative experiments in which numerous ectothermic hosts and
cold- and warm-tolerant parasites will be exposed to constant and variable temperatures (across a temperature
range) and quantify subsequent host immunity and parasite infections.
This research project is expected to reveal general mechanisms by which climate change and specific land
use modifications facilitate parasite invasions. This will enhance risk assessment and management by allowing
decision makers to prioritize regions, localities, and species that are at risk for potentially debilitating parasite
invasions.
EPA Grant Number: R833835
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 20
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Beach Grass Invasions and Coastal Flood Protection: Forecasting
the Effects of Climate Change on Coastal Vulnerability
Eric Seabloom, Sally Hacker, and Peter Ruggiero
Oregon State University, Corvallis, OR
Increased storm severity and sea-level rise resulting from climate change have greatly elevated the risk of
catastrophic flooding and storm damage to coastal communities. These risks have been exacerbated by
alterations to coastal ecosystems and the introduction of exotic species. In the Pacific Northwest, coastal dunes
protect approximately one-half of the coastline, and our initial results suggest that climate change-induced sea
level rise could double the frequency with which waves overtop dunes. Intentional planting of exotic grasses
may have initially increased coastal protection from flooding by building tall foredunes parallel to the
shoreline. However, an unintentional second invasion appears to be decreasing foredune height by 50 percent,
thereby increasing risk exposure. In addition, many agencies are removing exotic beach grasses to restore
habitat for imperiled species listed in the Endangered Species Act. The effects of these conservation actions on
flooding risk are unknown. The objectives of this research are to determine: (1) the effects of climate change
on exotic beach grass invasion; (2) the effects of exotic beach grass invasion on coastal vulnerability; and (3) if
conservation management alters coastal vulnerability to flooding under a range of climate change, invasion,
and management scenarios.
The investigators will use published climate change scenarios, remotely sensed beach topography data
(LIDAR), and field experimentation to parameterize coastal process and vulnerability models. These
empirically parameterized models will be used to forecast the risk of flooding in coastal communities under a
range of climate change and invasion scenarios.
This research will yield an increased general understanding of interactions among the alteration of coastal
ecosystems, species invasions, climate change, and human risk in coastal environments. In addition, the
researchers will conduct a quantitative vulnerability assessment of a specific coastal community in
Washington. This case study will serve as a template for other applications of our models and data in coastal
dune systems worldwide.
EPA Grant Number: R833836
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 21
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Elevated Temperature and Land Use Flood Frequency Alteration Effects on Rates
of Invasive and Native Species Interaction in Freshwater Floodplain Wetlands
Curtis J. Richardson, Neal Flanagan, and Song S. Qian
Duke University, Nicholas School of the Environment and Earth Sciences, Durham, NC
The primary objective of this research project is to assess how predicted climate and land use driven
changes in hydrologic flux and temperature regimes of floodplain ecosystems affect plant communities in
terms of their vulnerability to the establishment and spread of invasive species and, in turn, ecosystem
functions and services. Future climate scenarios for the southeastern United States predict that surface water
temperatures will increase (in concert with air temperature) and that stream flows will likely decrease, with a
greater proportion of annual watershed hydrologic yield occurring during major storm events. Land use
changes (urban vs. forested, etc.) have been shown to raise water temperature and increase pulsed water
releases during storms. This research project focuses on the relationships between native species composition,
diversity, productivity, and invasibility of floodplain ecosystems affected by alterations of water temperature
and annual hydrographs driven by climate and land use changes. The investigators will use a combination of
varying scale experimental studies and one novel large-scale regional study to verify the experimental and
threshold modeling results.
There are four study levels: (1) A field-based warming experiment will allow the investigators to directly
evaluate and model treatment effects of temperature and hydrology on species invasions, community
composition, and ecosystem services of an experimental (restored) floodplain ecosystem. (2) There are 99
diversity plots on a floodplain that will be used to test how species richness affects species invasions. (3) There
are 102 permanent vegetation plots that will be distributed over three hydrogeomorphic zones in the floodplain
(stream bank, low terrace, and high terrace) to assess species invasions affected by pulsed waters. (4) Regional
studies on wetlands downstream of surface and bottom-releasing dams will be used to assess pulsed water and
temperature effects on invasive species as compared to control rivers. At each experimental level the
investigators will assess how feedbacks from invasive species alter ecosystems services such as flood control,
sediment retention, and maintenance of water quality. A unified Bayesian hierarchical model will be developed
as a decision support tool to predict temperature and hydrology thresholds for invasive species response to
alterations in floodplain ecosystems.
Experimental results will be used to estimate the effects of predicted temperature increases and increased
storm flow events on the ability of existing floodplain communities to resist invasive species. Proposed
Bayesian modeling methods can address nonlinear responses and provide a risk assessment probability
analysis to predict ecosystem threshold shifts.
EPA Grant Number: R833837
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 22
-------
The Plight of Ecosystems in a Changing Climate:
Impacts on Services, Interactions, and Responses Workshop
Ecological Impacts From the Interactions of Climate Change,
Land Use Change, and Invasive Species
Robert B. Whitlatch1 and Richard W. Osman2
University of Connecticut, Groton, CT; 2Smithsonian Environmental Research Center, Edgewater, MD
The five objectives of this research project are to: (1) work with environmental managers and
stakeholders to explore different scenarios for land use planning, development of coastal areas, habitat
restoration, or other management issues in the context of climate change and invasive species; (2) conduct
mesocosm experiments testing links between climate change and land use in altering the ability of invasive
species to affect native communities; (3) conduct field experiments to assess temporal and/or spatial scales of
potential efforts needed to effectively manage invasive species; (4) conduct field experiments examining the
survival of key predators of invasive species in areas of different land use; and (5) develop predictive models
to assess alternative management strategies. Focus will be placed on integrating management needs with
ecological predictions that allow managers to evaluate multiple stressors at different temporal and spatial
scales in different types of coastal systems.
Workshops with managers and stakeholders will discuss multi-stressor management needs and establish
the most useful management scenarios for coastal zone planning in a context of climate change and invasive
species and information dissemination methods. Mesocosm experiments will simulate predicted temperature
changes, and the population and community responses of native and recently introduced species will be
compared. Field experiments will determine the spatial and temporal scales for the effective management of
invasive species in the context of differences in coastal land use and climate change. An existing
population/community model will be modified to present easily understood scenarios to managers and
planners.
This study will directly examine climate change on shallow-water marine communities that are most likely
to suffer from the poleward spread of species as coastal waters warm. The adaptation of an existing model will
couple climate and land use changes to assess their combined effects on the susceptibility of habitats to species
invasion and subsequent ecosystem changes in a manner that can be used by managers and planners. Because
the invaders are easily recognized and their damage to native communities can be readily quantified, they can
be used by managers as highly visible indicators of stress, as well as to assess the success of various types of
implemented management plans.
EPA Grant Number: R833838
The Office of Research and Development's National Center for Environmental Research and EPA Region 10 23
-------
Appendices
-------
U.S. Environmental Protection Agency (EPA)
The Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and
Responses Workshop
May 27 - 28, 2009
Plymouth Church
1217 Sixth Avenue
Seattle, WA
AGENDA
Webinar Information
http://hawkeye.epa.gov/imtapp/app/sch mtg details.uix?mID=410300
Conference ID: 410300
Conference Key: 2546283
Call-in Information:
Toll-free dial-in number (US and Canada): (866) 299-3188
International dial-in number: (706) 758-1822
Conference code: 2023439850
Day One, Wednesday, May 27, 2009
8:00 a.m. - 8:30 a.m. Registration
8:30 a.m. - 9:00 a.m. Introductory Remarks
Roseanne Lorenzana, EPA, Region 10 Science Liaison
Brandon Jones, Project Officer, EPA, ORD, NCER
Tier I - Effects of Climate Change on Ecosystem Services Provided by Coral Reefs and
Tidal Wetlands
9:00 a.m. - 9:20 a.m. Effect of Sea Level Rise and Climate Variability on Ecosystem
Services of Tidal Marshes
Chris Craft, Indiana University, Bloomington
9:20 a.m. - 9:40 a.m. Climate-Linked Alteration of Ecosystem Services in Tidal Salt
Marshes of Georgia and Louisiana
Mark Hester, University of Louisiana at Lafayette
9:40 a.m. - 10:00 a.m. Linking Impacts of Climate Change to Carbon and Phosphorus
Dynamics Along a Salinity Gradient in Tidal Marshes
Melanie Vile, Villanova University
10:00 a.m. - 10:20 a.m. Break
* USDA Grantees
-------
Day One, Wednesday, May 27, 2009 (continued)
10:20 a.m. - 10:40 a.m. Connectivity in Marine Seascapes: Predicting Ecological and
Socioeconomic Costs of Climate Change on Coral Reef Ecosystems
Julie Kellner, Resources for the Future
10:40 a.m. - 11:00 a.m. Effects of Climate Change on Ecosystem Services Provided by
Hawaiian Coral Reefs
Paul Jokiel, University of Hawaii at Honolulu
11:00 a.m. - 12:00 p.m. Tier I Discussion
12:00 p.m. - 1:00 p.m. Lunch (on your own)
Tier II - Nonlinear Responses to Global Change in Linked Aquatic and Terrestrial Ecosystems
1:00 p.m. - 1:20 p.m. Hydrologic Forecasting for Characterization of Nonlinear Response
of Freshwater Wetlands to Climatic and Land Use Change in the
Susquehanna River Basin
Denice Wardrop, Pennsylvania State University
1:20 p.m. - 1:40 p.m. Sustainable Coastal Habitat Restoration in the Pacific Northwest:
Modeling and Managing the Effects, Feedbacks, and Risks
Associated With Climate Change
John Rybczyk, Western Washington University
1:40 p.m. - 2:00 p.m. Nonlinear Response of Pacific Northwest Estuaries to Changing
Hydroclimatic Conditions: Flood Frequency, Recovery Time,
and Resilience
Rob Wheatcroft, Oregon State University
2:00 p.m. - 2:20 p.m. Nonlinear Response of Prairie Pothole Landscapes to Climate
Change and Land Management
Carter Johnson, South Dakota State University
2:20 p.m. - 2:40 p.m. Innovative Management Options To Prevent Loss of Ecosystem
Services Provided by Chinook Salmon in California: Overcoming
the Effects of Climate Change
Lisa Thompson and David Purkey, University of California at Davis
2:40 p.m.-3:00 p.m. Break
3:00 p.m. - 3:20 p.m. Hydrologic Thresholds for Biodiversity in Semi-Arid Riparian
Ecosystems: Importance of Climate Change and Variability
Thomas Meixner, University of Arizona
3:20 p.m. - 3:40 p.m. Nonlinear and Threshold Response to Environmental Stresses in
Land-River Networks
Jerry Melilo, Woods Hole Oceanographic Institution
-------
Day One, Wednesday, May 27, 2009 (continued)
3:40 p.m. - 4:40 p.m. Tier II Discussion
4:40 p.m. Adjournment (continued discussion and dinner on your own)
Day Two, Thursday, May 28, 2009
7:30 a.m. - 8:00 a.m. Registration
Tier III - Ecological Impacts From the Interactions of Climate Change, Land Use Change, and
Invasive Species
8:00 a.m. - 8:20 a.m. Integrated Bioclimatic-Dynamic Modeling of Climate Change
Impacts on Agricultural and Invasive Plant Distributions in the
United States
* Wei Gao, Colorado State University
8:20 a.m. - 8:40 a.m. Global Change and the Cryptic Invasion by Transgenes of Native
and Weedy Species
* Cynthia Sagers, University of Arkansas
8:40 a.m. - 9:00 a.m. A Multi-Scale Approach to the Forecast of Potential Distributions of
Invasive Plant Species
*John Silander, University of Connecticut
9:00 a.m. - 9:20 a.m. Predicting Risk Invasion by Salt Cedar and Mud Snails
Leroy Poff, Colorado State University
9:20 a.m. - 9:40 a.m. Integrating Future Climate Change and Riparian Land Use To
Forecast the Effects of Stream Warming on Species Invasions and
Their Impacts on Native Salmonids
Julian Olden, University of Washington
9:40 a.m. - 10:00 a.m. Break
10:00 a.m. - 10:20 a.m. Climate Change: Pathogens and Decline of Ectotherms
Jason Rohr, University of South Florida
10:20 a.m. - 10:40 a.m. Beach Grass Invasions and Coastal Flood Protection: Forecasting
the Effects of Climate Change on Coastal Vulnerability
Eric Seabloom, Oregon State University
10:40 a.m. - 11:00 a.m. Elevated Temperature and Land Use Flood Frequency Alteration
Effects on Rates of Invasive and Native Species Interactions in
Freshwater Floodplain Wetlands
Curtis Richardson, Duke University
-------
Day Two, Thursday, May 28, 2009
11:00 a.m. - 11:20 a.m. Ecological Impacts From the Interactions of Climate Change, Land
Use Change, and Invasive Species
Robert Whitlatch, University of Connecticut
11:20 a.m. - 12:20 p.m. Tier III Discussion
12:20 p.m. - 12:30 p.m. Closing Remarks
12:30 p.m. Adjournment
-------
U.S. Environmental Protection Agency (EPA)
The Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions,
and Responses Workshop
May 27 - 28, 2009
Plymouth Church
1217 Sixth Avenue
Seattle, WA
PARTICIPANTS LIST
Balram Ambade
Pt. Ravishankar Shukla University, Raipur
Juliann Aukema
The Nature Conservancy
Marcia Bailey
U.S. Environmental Protection Agency
John Barich
U.S. Environmental Protection Agency
Judith Barry
Noblis
Thomas Baugh
U.S. Environmental Protection Agency
Chris Bellovary
U.S. Environmental Protection Agency
Jim Brennan
Washington Sea Grant
Liz Carr
Washington State Department of Health
Chris Castner
U.S. Environmental Protection Agency
Christopher Cora
U.S. Environmental Protection Agency
Christopher Craft
Indiana University
Anthony D'Andrea
University of the Virgin Islands
Ifeyinwa Davis
U.S. Environmental Protection Agency
Christine Davis
U.S. Environmental Protection Agency
Erin Donley
University of Washington
Robert Drake
U.S. Environmental Protection Agency
Daniel Duncan
U.S. Environmental Protection Agency
Kirsten Feifel
University of Washington
Neal Flanagan
Duke University
Nancy Cavallaro
U.S. Department of Agriculture
Roger Fuller
The Nature Conservancy
-------
Jill Gable
U.S. Environmental Protection Agency
Wei Gao
Colorado State University
Patty Click
National Wildlife Federation
Rachel Gregg
EcoAdapt
Glenn Guntenspergen
U.S. Geological Survey
Alan Hamlet
University of Washington
Lynn Helbrecht
Washington Biodiversity Council
Mark Hester
University of Louisiana at Lafayette
Harry Hill
U.S. Environmental Protection Agency
Greg Hood
Skagit River System Cooperative
Research
Bevin Horn
U.S. Environmental Protection Agency
Carter Johnson
South Dakota State University
Paul Jokiel
University of Hawaii at Honolulu
Brandon Jones
U.S. Environmental Protection Agency
Susan Julius
U.S. Environmental Protection Agency
Julie Kellner
University of California, Davis
Tarang Khangaonkar
Pacific Northwest National Laboratory
Tony Lafua
National Inland Water Ways Authority
Al LaTourette
U.S. Environmental Protection Agency
Peter Leinenbach
U.S. Environmental Protection Agency
Xin-Zhong Liang
University of Illinois at Urbana-Champaign
Shuyan Liu
University of Illinois at Urbana-Champaign
Roseanne Lorenzana
U.S. Environmental Protection Agency
Lisa Macchio
U.S. Environmental Protection Agency
Theogene Mbabaliye
U.S. Environmental Protection Agency
Thomas Meixner
University of Arizona
Jerry Melillo
Woods Hole Oceanographic Institution
Jamie Mooney
University of Washington
Robert Naiman
University of Washington
Tristan Nunez
University of Washington
Julian Olden
University of Washington
Rick Parkin
U.S. Environmental Protection Agency
-------
Erik Peterson
U.S. Environmental Protection Agency
Don Phillips
U.S. Environmental Protection Agency
LeRoy Poff
Colorado State university
Derek Poon
U.S. Environmental Protection Agency
David Purkey
University of California, Davis
Ramesh Reddy
University of Florida
Jonathan Reum
University of Washington
Curtis Richardson
Duke University
Danielle Rioux
University of Washington
Deborah Robinson
U.S. Environmental Protection Agency
Jason Rohr
University of South Florida
Keith Rose
U.S. Environmental Protection Agency
John Rybczyk
Western Washington University
Paul Rygiewicz
U.S. Environmental Protection Agency
Cynthia Sagers
University of Arkansas
Rhea Sanders
Oregon State University
Eric Seabloom
Oregon State University
Anne Sergeant
U.S. Environmental Protection Agency
John Silander
University of Connecticut
Elaine Somers
U.S. Environmental Protection Agency
Ginny Stern
Washington Deptartment of Health
Lisa Thompson
University of California, Davis
Joseph Trinh
University of Antioch Seattle
Ron Tressler
Seattle City Light
Melanie Vile
Villanova University
Garrett Voerman
Seattle City Light
Elizabeth Waddell
National Park Service
Denice Wardrop
Pennsylvania State University
Robert Warinner
Washington Department of Fish and Wildlife
Larry Wasserman
Swinomish Indian Tribal Community
Nathaniel Weston
Villanova University
Rob Wheatcroft
Oregon State University
-------
Robert Whitlatch
University of Connecticut
Sharon Wilson
U.S. Environmental Protection Agency
-------
Skagit Watershed, Puget Sound, WA
Largest river system and watershed in Puget Sound
Four native Indian Tribes
-30% of freshwater draining into Puget Sound
394 glaciers
Precipitation: 50" at mouth; > 140" ridge tops
Parts designated wild and scenic
All 5 Species Of Pacific Salmon (largest producer of wild salmon in
Puget Sound ; it produces 60% of wild Chinook salmon in the Sound and is the largest
*un of Chum in the lower 48 states)
Bald eagles, waterfowl, shorebirds, raptors
Intertidal/delta critical for salmon, ESA, agriculture
Most upland private timber, US Forest Service
Lowland highly developed for urban and agriculture
rrent EPA clim
EPA grants: @ $2,300,000
• Other grants, investments: $800,000+
• In-kind and matching
Total current investments
Slide Credit: Univ of Washington, Seattle City Light and others
-------
-------
Effects of Sea Level Rise and Climate Variability
on Ecosystem Services of Tidal Marshes,
South Atlantic Coast
Christopher Craft, Samantha Joye,
Steve Pennings, Richard Park, Jeff
Ehman and Jonathan Clough
Salt Marsh
Brackish Marsh
Tidal Freshwater Marsh
Tidal fresh-
water marsh
Regulation Functions
Shoreline Protection •*•
CO, & CH, Flux
Habitat Functions
• Macrophyte Diversity
Marsh Nekton -«•
Productivity Functions
-------
Climate Change = Saltwater Intrusion
Project Goal
Develop a conceptual model that
describes how tidal marsh ecosystem
services vary along the salinity
gradient and a simulation model of how
sea level rise and climate variability will
affect the delivery of ecosystem
services.
Hypotheses
Rising sea level (RSL) leads to inundation and loss of
tidal marshes, especially tidal freshwater marshes and
their ecosystem services.
Diking protects freshwater marshes against RSL. But,
when marshes are diked, ecosystem services
associated with connectivity are lost.
Greater inter-annual variability of climate leads to
greater frequency of drought and reduction in
ecosystem services in drought years. Greater
variability in rainfall leads to increased
delivery of ecosystem services in wet years.
How will accelerated sea level rise (SLR) affect
the area and spatial distribution of tidal marshes
AND their delivery of ecosystem services?
Wetland Habitat
Reduced salt and brackish
marsh habitat
Near complete loss of tidal
freshwater marsh
Increased submerged land
Ecosystem Services
• Reduced regulation functions
(shoreline protection, carbon
sequestration, N&Pretention,
denitrification, sediment
deposition, greater CH4 & CO2)
• Reduced habitat functions
(plant diversity, migratory songbir
habitat)
• Reduced production functions
(plant productivity, marsh nekton,
commercial shrimp yield)
SE US Coast Study Region
-------
Measurement of Ecosystem Services
Three estuaries:
Three marsh types:
Altamaha
Ogeechee
Satilla
Tidal fresh
Tidal brackish
Tidal salt
3 estuaries/3 marsh types/2 sites, n=18 sites
Nitrogen Accumulation in Soil
^
^ 12 -
E
& 10-
o
•8 8.
1
3 «-
o
C 4 .
O
H 2 -
Z
IB) 1
|B| ^H Oqeechee 1
82 + 05 7'5 * 1 ^" Altamaha |
2.9 ±0.8
i
_
n
I 1 Satilla
(A)
I 2.4 ± 0.4
II
TFr Swamp Tidal Fresh Brackish Salt
Potential Denitrification
250,
= _ 200
° E
1 •?
'SS, 150
II
— -? 100 •
.2 O
(A)
101.9 ±47.6
I
(B)
i i Ogeechee
I I Altamaha
l=] Satilla
49.1 ±23.1
f. ]
Tidal Fresh Brackish
(C)
21.0 ±12.3
lln
Salt
Potential Denitrification (g N/m2/yr)
Tidal fresh marsh
Brackish marsh
Salt marsh
6.6 + 2.7
4.6 +2.3
3000 -
~£ 2500 -
S
U> 2000 -
.2
m 1500 -
•o
C
B 1000 -
O)
0)
o
Si 500-
0-
Aboveground Biomass
(A) (A)
1400 ±200 1712 ±112
1 Ti
x
1
Tidal Fresh Brackish
i i Ogeechee
1 1 Altamaha
^m Satilla
(B)
996 ±152
T
T
!
JL
Salt
-------
SLAMM Version 5 (beta)
(Sea Level Affects Marshes Model)
SLAMM uses elevation, NWI,
tide range, historic sea level
rise and site-specific accretion
rate data to parameterize
the model.
A salinity algorithm is used
to simulate saltwater intrusion
into river-dominated estuaries
as sea level rises.
The simulation is run
using A1B SRES (mean, max)
scenario.
The Altamaha River Estuary (Georgia) as an Example
Altamaha River
(1999)
IfdUII* I |
Dov Oryll
inOn Dry Land |
MrvJwuud Swamp j
ElBuamM Qpvn WtHor
SLAMM simulation of the effects of accelerated
SLR on the southeast (Georgia) coast
SLAMM simulation of the effects of accelerated SLR (A1B
mean = 52 cm) on (wet)land cover types the Georgia coast
Initial Condition
(km*)
Dry land
Non-tidal swamp
Inland fresh marsh
Tidal fresh swamp
Tidal fresh marsh
Brackish marsh
Salt marsh
Transitional salt marsh
Tidal flat
Estuarine open water
Year 2100 Loss
(km*) (%)
-------
Landscape-scale (Georgia coast) N Accumulation
in Soil and Denitrification (A1B mean = 52 cm)
Landscape-scale (Georgia coast) N Accumulation
in Soil and Denitrification (A1B max = 82 cm)
Wetland
Change
(km*)
Denitrification
Wetland
Change
(km*)
N Denitrification
Tidal fresh +1
Brackish +41
Salt marsh -226
+ 8 +/
+307 +184
-542 -384
Tidal fresh -32
Brackish -4
Salt marsh -496
Cumulative -184
Cumulative -532
leads to loss of ecosystem services
-------
Lessons Learned
Lessons Learned (continued)
Different types of tidal marshes provide
different levels of ecosystem services.
Tidal fresh- and brackish-marshes have
greater aboveground biomass, N retention in
soil and denitrification than salt marshes.
Climate change (sea level rise - SLR) will
promote salt water intrusion and
submergence, leading to habitat conversion
and loss of tidal marshes, especially those at
either end of the salinity gradient.
Wetland loss may not be a great as predicted
because spatial models lack positive
feedback mechanisms that enable marshes
increase surface elevation.
-------
Lessons Learned (continued)
Challenges
Dikes, while protecting tidal marshes,
leads to loss of connectivity to estuarine
waters and the ecosystem services that
depend on connectivity.
Tidal marsh ecosystem services are more
strongly correlated with variation in
salinity, driven by river discharge, than by
variation in temperature and precipitation.
Difficulty in evaluating ecosystem
services of fauna/wildlife (fishes, birds).
Difficulty working with subcontractors
(esp. consultants).
Interaction with Clients
Outcomes
The Nature Conservancy
(Sea level rise (SLR) modeling of the Altamaha River
Bio-reserve and elsewhere in coastal Georgia)
US Fish & Wildlife Service
(SLR modeling of National Wildlife Refuges in
Georgia and South Carolina)
New Projects
NOAA National Estuarine Research Reserve System
DOE National Institute for Climate Change Research
GCELTER (Phase II)
Outcomes
Three publications in 2009 including...
Craft, C., J. Clough, J. Ehman, S. Joye, D. Park, S. Pennings, H.Guo and
M. Machmuller. 2009. Forecasting the effects of climate change on tidal
marsh ecosystem services. Frontiers in Ecology and the
Three "in review"
Three planned including.
Summary paper in Global Change Biology.
-------
-------
Climate-Linked Alteration of
Ecosystem Services in Tidal Salt
Marshes of Georgia and Louisiana
Outline
MarkW. Hester
Coastal Plant Ecology Laboratory
University of Louisiana at Lafayette
In/ing A. Mendelssohn
Introduction
Project Goals*
Approach
Lessons Learned/Challenges*
Key Findings
Interactions with Clients*
Funding provided by US EPA STAR
Introduction
Drought-induced sudden salt marsh dieback of Spartina
a/tern/florat\da\ salt marshes
- Louisiana (2000-2001)
Georgia (2001)
Reported in sever
;oastal states since
Potential for drastic alteration of ecosystem services
- Driven by decrease in 5. alterniflora living stem density
- Directly linked to degree loss of ecosystem processes
Project Goals*
Elucidate the effects of climate change (increased drought
severity) on tidal salt marsh ecosystem services
- Eutrophication control
- Carbon sequestration
- Sustainable habitat
- Faunal support
Two hydrogeomorphic settings
- Louisiana (microtidal)
- Georgia (mesotidal)
Develop exploratory Structural Equation Model (SEM)
Measurable Ecosystem Processes
i Service latent variables (Ecosystei
-------
Experimental Approach
Manipulative field experiment of Spartina alterniflora plant density in
micro- and mesotidal salt marsh ecosystems (Main Plots)
Louisiana (Caminada-Moreau Headland)
Identify 6 dieback areas (Block
establishment of large researcf
- 24 plots per state
- Each plot 8.0-m x 7.5-m
- 4 vegetated conditions
stem density
stem density
Series of specific a-priori, process-driven hypothesis testing
(univariate and multivariate)
-.M (Structural Equation Modeling) to reveal relationships between
;m density, ecosystem processes, and ecosystem services
-------
Lessons Learned/Challenges^
Damage to UNO, resultant resignation
Prolonged drought in Georgia
- Affected achieving target plant densities (covariable)
nunication of rigors of large r
- During budgeting, setup, and a
Personnel changes at LSU and UGA; UNO to ULL
- Required continued effort in management and continuity
(and expected) component
Climate variability
Louisiana December 2008
-------
Louisiana December 2008
Spartina
alternijlora
Stem
Densities
• Gradient of stem
density treatments
established in
each state by 2007
Cover reflected
stem densities
Desired gradient
in Louisiana
ffl
Spartina
alternijlora
Cover
• Cover reflected
stem densities
• Desired gradient
in Louisiana
• Newdieback
occurred in 2008
in 3 Georgia
Reference plots
5 ,c
DLow Density DHigh Density DRi
-------
Additional
Experiment
Natural variation in
stem density
In (dynamic)
equilibrium with
resource supply
Overall non-
significant
differences in C
assimilation rates
Non-significant effect of stem
density on net CO2 assimilation
Additional
Experiment:
Higher density
Spartina more
efficient utilization
of leaf N for C
assimilation
Highly significant effect of
stem density on PNUE
Louisiana
Accretion & Net
Elevation Change
High Density plots had
equivalent accretion rates
to Reference plots
High Density plantings
increased surface elevation
Bare plots had lower
accretion rates and lost
elevation
Georeia
Accretion & Net
Elevation Change
Reference plots had
greatest accretion rates
High Density & Bare plots
had lower but equivalent
accretion rates
-------
Belowground
Productivity &
Decomposition
High Density and Reference Plots
had equivalent belowground
productivity rates
Litterbag decomposition
Reference and High Density
initially higher decomposition rates
No significant differences after
>400 days
Interstitial
Ammonium
• Much greater
ammonium in
Louisiana
• However, no
consistent
pattern with
stem density
ArTHTionium
Georgia Louisiana
I HIM)!.
Interstitial
Sulfides
Much greater in
Louisiana
Often below
detection in Georgia
Sulfides can inhibit
plant uptake of NH4
May be less tight
coupling of plant C
& N relations in
Louisiana
*
•
-------
Exploratory SEM of Spartina alterniflora
Stem Density on Ecosystem Processes & Services
Will be a 2-Group Model: LA and GA
Differences between Louisiana and Georgia in
Relative Strength of Relationships
Determining Photosynthetic N-Use Efficiency
Key Findings
Climate change (severe drought) can affect a suite of ecosyste
services
Density of Spartina alterniflora important driver of many
ecosystem services across hydrogeomorphic setting
Hydrogeomorphic setting important modulator of ecosystem
processes and services
A it—>cj patterns of carbon sequestration, eutrophication control
N-use, N cycling
nd faunal support
Net elevation change
7
-------
Interactions with Clients'
Louisiana
- LUMCON (Louisiana Universities Marine
Consortium)
- Burlington Land
• Permission to establish plots and infrastructure
- Louisiana Department of Natural Resources
Georgia
- Georgia Department of Natural Resources
- University of Georgia Marine Institute
Interactions with Clients*
Synergistic Activities Related to this Project
Interactions with Clients*
Another 7 Presentations Related to this Project
ublications Related to this Project
Alber, M., E. M. Swenson, S.C. Adamowicz, and LA. Mendelssohn.
2008. Salt marsh dieback: an overview of recent events in the US.
Istuarine Coastal and Shelf Science 80 (1): 1 -11.
"Georgia salt marshes healthy;
ow" The Darien News. Decembe
"Salt marsh s
November 20
ught sensitive." The Savannah Morning News.
"Drought, what drought?" The Savannah Morning News. Octobe
2007.
"Cause sought as marshes turn into barren flats" Boston Globe. J
2006.
Mendelssohn, I. A., M. Alber. E. Swensc '" " ' ' ~
dieback: a synthesis. I
4-8, 2007. Providence, R.I.
Alber, M. 2006. Salt marsh dieback in Georgia. Sudden wetland dieback meeting. May
structure and function.'
Estes Park, CO.
Salt marsh dieback in Georgia. September 2006. Univ. of
Kenemer, B., C. McFarlin, and M. Alber. 2006. Fiddler crabs dig it: A study of burrow
dynamics in a salt marsh. Poster presented at the Southeastern Estuarine Research
Society. October 2006. Savannah, GA
Outcomes*
)ata in process of final integration
Structural Equation Model
- Valuable managementtool
- Key differences in strength of relationships
Improved insights into climate variability
- Future management and planning
Tiarsh habitat
, -. jcosystem services
- State agencies (Louisiana & Georgia DNR)
- Federal agencies (EPA, DOE, NOAA)
Continued research opportunities
•"latory model of salt marsh ecosyst
F SEM ecosystem services apprc~
Acknowledgments
Merryl Alber (UGA)
Dale Bishop (UGA)
Joe Baustian (LSU)
Megan 1 & 2 (UGA)
Jodie Noel (ULL)
8
-------
Often higher in
Low and High
Density plots
Does not appear
due to previous
dieback
What is the
relationship
between stem
density, leaf N,
and PNUE?
Georgia
I I
Loui,i,n,
Ti
JL
T
in
_L
1
i
_L
I
_L
Spartina
alternijlora
Net CO2
Assimilation
• Trend towards
greater C
assimilation rates
in Low and High
Density plots
DLow Density DHigh Density DRf
Georgia
-------
Linking Impacts of Climate Change to
Carbon and Phosphorus Dynamics Along
a Salinity Gradient in Tidal Marshes
Nathaniel B. Weston1, Melanie A. Vile2,
Scott Neubauer3 & David Velinsky4
Department of Geography and the Environment, Villanova University
Department of Biology, Villanova University
3Baruch Marine Field Laboratory, University of South Carolina
3Patrick Center for Environmental Research, Academy of Natural Scieno
"Patrick Costello
'Amanda Foskett
Oliva Gibb
Anthony Geneva
Paul Kiry
Chris McLaughlin
Avni Malhotra
'Justin Meschter
Stephen Mowbray
*Michael Patson
*Tatjana Prsa
James Quinn
*DanielRusso
*Mariozza Santini
Kimberli Scott
Roger Thomas
*PaulWeibel
Environmental Protection Agency,
STAR Program
Laboratory for Molecular and
Systematic Ecology, Academy of
Natural Sciences
Villanova Department of Biology
en
T 50
ol-s°
i —
fi -100
Sea Level Rise
Global average sea level
l.Smmyr1
& accelerating
1850 1900 1950
Year
Intergovernmental Panel on Climate Change Fourth Assessment Report (2007)
Coastal Tidal Marshes
Marshes Must Accrete to Keep Pace With
Watershed
Inputs
Rising Sea-Levels
CO2
Primary Production
CO
A
Inorganic
Sediment
JM*
Export
It*
MSL
Organic Microbial Respiration
Matter ' '
CCX&CIL
Tidal Freshwater
Marsh
i i Freshwater
CZ1 Seawater
-------
Estuarine / Marsh Coastal Ecosystem
Changing
Precipitation /
Evaporation /
Evapotranspiration
Salinity Intrusion
Rising
Sea Level
Project Goals
Understand how salt water intrusion into
TFMs will impact C, N and P cycling
- Plant processes
- Microbial processes
Predict the response of TFMs and the
ecosystem services they provide to scenarios^
of future climate change
Delaware River Estuary
Salinity
Increase
Observed
What are the impacts of climate change on
TFMs?
Microbial Respiration Processes
Freshwater Marshes:
Methanogenes is
C6H1206 + 3 H20 ^ 3 CH4 + 3 HCO3- + 3 H+
Microbial Respiration Processes
Salt Marshes:
Sulfate Reduction
C6H12O6 + 3(SO^)^ 3 HS- + 6 HCO3- + 3 H+
Sulfate - Major constituent in seawater
-------
Delaware River Estuary
DDDDDDD
DDDDDDD
Long-Term Salinity Intrusion Experiment
Freshwater
CO2 & CH4
Gas Flux Rates
Sulfate Reduction
and
V Methane-genesis
Rates and other
Biogeochemical
Measurements
CO2 Flux
E »
•a
e «
Month
Sulfate Reduction
Sulfntc Reduction
[mmol m '2 d ')
5 o "3 o o c
-50 o
Vi
\
» » .
50 IOO 150 2OO
Days
Rates
o Freshwater
• Salinity Intnieion
O
250 30° 330 -loo
CH4 Flux
D Freshwater
• Satinirv* Intrusion
-*- ° o Increase
_
£ 100
5
— BO
Total C Gas Flux
D Freshwater
• Salinity Intrusion
-*-% Increase
Month
50% Higher C Flux over i Year
-------
'a
J,
fl 600 -
Q 580-
*£ 560 -
bfi
O 540 -
I
1
1 1
1
1
1
-50 o
Soil Organic Carbon
••I
j
r-1
o
j .*
'!•'
O Freshwater
--••-- Salt Water Intrusion
\
Days
400
Freshwater Marsh Undergoing Salinity Intrusion
Watershed
Input
Plant Response?
Inorganic
Sediment
Organic
Matter
Methanogens
Sulfate Reducers
CO, & CH
Microbial Response
Delaware River Transplant Experiment
BBBQBBQ
Delaware River Transplant Experiment
Field Site Monitoring
CO2 and CH4 flux
Plant Bioniass
Microbial Rate Measurements
Biogeochemistry
Microbial Community
Composition
,
J
Delaware River Transplant
Experiment - Conductivity
Jan Apr Jul Oct Jan Apr Jul Oct Jan
-------
Jan Apr Jul Oct Jan Apr Jul Oct Jan
n Apr Jul Oct Jan Apr Jul
Outcomes - Shift to Salt Marsh?
Watershed
Inputs
Response of Freshwater Marsh Plants to
Salinity Intrusion and Inundation
TFM Plant Biomass (g rrr2) =
- 10.8 [Conductivity (mS cm-1)]
+ 9.6 [Temperature (°C)]
-1.4 [Inundation (cm)]
-23.1
R2 = 0.37; p < o.ooi
Current Work - TFM 'Organs'
Elevation Relative
to Local
Marsh Platform
Wliatare the impactS""ofclimate""Criangeon
TFMs?
Watershed
Inputs I CO.
-------
Suspended Sediment in the Delaware River
ig6o's igyo's igSo's iggo's 2ooo's
Data from USGS Water Quality Monitoring Station
East / Gulf Coast Suspended Sediment Analysis
\
USGS Data
42 Rivers
East / Gulf Coast Suspended Sediment Analysis
oNo change (43%
_ • Increase (9%)
• Decrease (48%)
Suspended
Sediment
Watershed
Inputs
Outcomes - Loss of TFM?
Plant Response
dback
Inorganic
Sediment
Loss of Marsh
Organic
Matter
/TVIetnanogens
\Sulfate Reducer
Challenges
Controlling for marsh vertical elevation
critical in field experiments
Multiple TFM plant species (salt marshes
are easy!)
Understanding response of methanogens
Complex, interconnected processes (plant,
microbial, sedimentation, accretion)
Interaction with Clients
Integration with ongoing work in other groups
- Partnership for the Delaware Estuary
- University of Delaware, Rutgers, DEP, EPA
Communication with local stakeholders
— Delaware Estuary Environmental Summit
- Earth Day event
- Field site interactions
Communication at national meetings
- Society of Wetlands Scientists
- Estuarine Research Federation
- American Society of Limnology and Oceanography,
-------
^Patrick Costello
* Amanda Foskett
Oliva Gibb
Anthony Geneva
Paul Kiry
Chris McLaughlin
Avni Malhotra
* Justin Meschter
Stephen Mowbray
*Michael Patson
*Tatjana Prsa
James Quinn
* Daniel Russo
*Mariozza Santini
Kimberli Scott
Roger Thomas
*PaulWeibel
Environmental Protection Agency,
STAR Program
Laboratory for Molecular and
Systematic Ecology, Academy of
Natural Sciences
Villanova Department of Biology
7
-------
SR8
Connectivity in Marine Seascapes:
Predicting ecological and
socioeconomic costs of climate
change on coral reef ecosystems
Team
• James N. Sanchirico (Resources for the Future)
- Julie B. Kellner (post-doc)
• Kenneth Broad (University of Miami)
• Dan Brumbaugh (American Museum of Natural History)
• Alan Hastings (University of California, Davis)
• Fiorenza Micheli (Stanford University)
- Steven Y. Litvin (post-doc)
• Peter J. Mumby (Exeter University)
- Helen J. Edwards (post-doc)
Project Information:
• Research Category and Sorting Code: Effects Of Climate
Change On Ecosystem Services Provided By Coral Reefs
and Tidal Marches, 2004-STAR-J1
• Project Period: March 1, 2005 - February 28, 2008
Research Questions
How do environmental and anthropogenic impacts including
overfishing and mangrove deforestation affect the vulnerability
of Caribbean coral reefs to climate change?
When do socioeconomic responses to changes in the
ecosystem triggered by climate change stressors exacerbate
the vulnerability of coral-reef ecosystems to future stressors?
What are the critical ecological and/or socioeconomic
uncertainties for predicting climate change impacts on
ecosystem services that will yield the greatest returns from
investigation?
-------
Belize barrier reef
Threats to reefs
Coral bleaching
- Response of corals to elevated
temperatures or high levels of ultra
violet radiation
- Corals expel their symbiotic
- If exposure is weak, corals a
recover these algae
- Prolonged exposure can cause
mortality
Hurricanes
- Can damage, overturn and kill cor
- Movement of sediments and debr
causes scouring
- Increased nutrients can encourage
algal growth
-------
The importance of grazers
Macroalgae compete with
corals
* Reefs can switch from a
healthy (coral-dominated)
to an unhealthy (algal-
dominated) state
• Grazers influence
replenishment rate, growth &
fecundity of coral colonies
• Grazing underpins resilience of
coral reefs to disturbance
Managing the
resilience of coral
reefs in the face of
rising sea
temperature
Helen J Edwards
University of Exeter
Modelling the impacts of
disturbances on corals
Mortality caused by bleaching depends on
- Magnitude & duration of thermal stress (calculated
using degree heating months)
- Each coral's 'thermal history'
Mortality caused by hurricane depends on
- Strength of hurricane at reef location (Saffir-Simpson
category)
- Colony size
If a hurricane occurs, bleaching is prevented from
occurring that year
-------
Impacts of exploiting grazers
Parrotfish
exploited
Each line shows
average coral
cover(over 100
runs) for each
site
Managing the resilience of coral reefs in the
face of climate change
Peter J. Mumby
University of Exeter
Urchins & Parrotfish
I limited (max
reef per 6 mo)
Williams SPolunin (2001)
MEPS222: 187-196
Mumby (2006) Ecol.Apps. 16:
747-769
-------
What is resilience?
Resilience
Predictions of coral-reef grazer model (hysteresis)
NEGATIVE FEEDBACKS
(too little grazing intensity)
Reduced structural Reduced fish
complexity recruitment
POSITIVE FEEDBACKS
(surplus grazing intensity)
Increased structural |ncreasedSsn
complexity recruitment
GRAZING INTENSITY ^ 3
^ <^
Increased grazing
-------
Predictions of coral-reef grazer model
(hysteresis)
High grazing Low grazing
'redictions of coral-reef grazer model
(hysteresis)
High grazing
Low grazing
The location and nature of
the threshold and the shift is
a function of ecological,
climate, and socioeconomic
Resilience influenced by..
Shape (bifurcation points)
- Primary production (medium)
- Recruitment rate (weak)
- Coral growth rate (high)
Grazing Axis (X-axis)
- Disease of urchin Diadema antillarum
- Fishing of grazing fishes
- Seascape context (mangrove nurseries)
Coral axis (Y-axis)
- Bleaching
- Hurricanes
- Disease
Stable versus
unstable
unstabte eQulibria
jquilibr
80
!'J
-60
X
10
rt
ISya
nnRnuBDDnuDSBBSl
u
n
i
D U
°R:
n " " °
, yDpHH§yyB°.
80
70;
60
?
r so
•s 10
30
20
10
1
T
H
-SI
\
\ t1
-
\
X
X
X If left to fully
X_^ degrade
Parrotfish only^ ^ ^Parrotfish 8- urchins^
H^ — r — • — • — m^ "" "• ( •*" (
0.1 0.2 y& 0.4 0.5 0.6 0.7
Grazing (proportion of reef grazed in 6 months)
-------
05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Grazing (proportion of reef grazed in 6 months)
Uses of model for reef management
1) Managing grazers on reefs
2) Conservation of mangroves
3) Choose sites of naturally-lower
productivity
-------
Quantify resilien
0.05 0.1 0.15 0.2 0.25 0.3 0.35
Grazing (proportion of reef grazed in 6 months)
8
-------
Grazing 30%-33
of reef (10% rise)
Effect of mangroves in
re
Mangrove present
Mangrove absent -
.32 0.36 0.41
Mortality rate of pubescent and adult corals
-------
Ecosystem Services
Research questions
How does habitat (and
loss thereof) affect the
productivity of fisheries?
What does this imply for
the economic value of
habitat?
How do these values
impact coastal land use
decisions?
Species habitat use and coastal
land-use decisions
Obligate relationship
between species and
habitat results in less
clearing than if the
behavior is facultative,
everything else being
equal
Percent of mangroves cleared
Lessons learned (1)
Caribbean coral reefs appear to exhibit
alternate stable states
Threshold levels of coral cover, grazing,
nutrients etc
Restoring reef health becomes
disproportionately more difficult as health
declines
Act sooner rather later
Lessons learned (2)
Hysteresis plot
Resilience
ectory of reef between
nts & location of thresholds
ibability that reef does NOT
ed in shift towards stable
Derive by combining hysteresis plot with
stochastic simulation ; of disturbance
-------
Examples of activities of the group
• Chair of Remote Sensing Working Group of the World
Coral Reef Targeted Research Project (www.gefcoral.org).
- The ecological models developed under the EPA project have been
extended through collaborators on this World Bank project so that a
parameterization for coral disease was added.
• Intermittent Expert Hire of the U.S. EPA to help them design their
coral reef valuation work
• Bahamas National Trust board member
Examples of the work being included in policy
decisions
• The modeling we did on parrotfish exploitation was presented to
the Fisheries Administration, 170 fishermen, and stakeholders in
Belize (March 2005, 2008 and 2009). In partial response to this,
the Minister of Fisheries has just signed new legislation to ban
parrotfish harvesting in the country.
- A similar story is also true for Bonaire where they are currently drafting
regulations to ban fish traps.
Outreach (cont
• G3-6 teaching resources: Treasur
in the Sea
• College-level exercise: Marine
Reserves & Local Fisheries
Practitioners
• Newsletter: BBP in Brief
• Project meetings
• Presentations
Decision makers
• Project meetings & office visits
• BNT Council representation
Output - Outcomes
Integrative models useful and used for management
and education
Examples of peer-reviewed publication outlets
- Nature, Conservation Letters, Theoretical Ecology,
Science, Ocean and Coastal Management, Conservation
Biology, Coral Reefs, PNAS, Marine Biology, Ecological
Applications, and Journal of Ecology
Examples of presentations
- International Coral Reef Symposium, Ecological Society of
America, International Marine Conservation Congress,
American Fisheries Society, NOAA, WWF, TNC, etc.
11
-------
Uses&
Users
Visualiza
exploratii
experime
various
influenc
populati
and fish
sustain
Lab exercises
NZ, and The Bahamas
Stakeholder meetings in Tl
Additional details
)gram
AT&T)
Mor iloads via
http://bbp.amnh.org/website/curricula.html
Ongoing work
Multiple regression analysis to understand the
factors that explain key proxies for ability to
adapt to climate change
Development of a supplemental survey to better
understand how people have adapted to and
respond to hurricanes
- Use this information to better understand how
households in the Caribbean would likely adapt to
increased storm intensity
Mapping our fishermen income and effort levels
to the trophic model and reef resilience modeling
Discussion and ongoing work
Reef resilience exhibits hysteresis as a function
of grazing intensity
- Modeling endogenous thresholds
Model of grazers that takes into account
- Habitat dependencies (mangroves, sea grasses)
- Predator interactions (Grouper)
- Direct and indirect fishing pressure
Model of households that predicts changes in
fishing pressure via changes in labor market and
fishing returns
12
-------
Thank You!
We also want to acknowledge the NSF
biocomplexity team members.
(Bahamas Biocomplexity Project information can
be found at http://bbp.amnh.org/)
• ~ baugh,AMNH
ad, Univ. Miami, RSMAS
Steve Cantrell, Univ. Miami
Jackie Chisholm, College of The Bahai
Chris Cosner, Univ. Miami
Bob Cowen, Univ. Miami, RSMAS
Craig Dahlgren, Perry Inst. Marine Self
RobDeSalle,AMNH
Meg Domroese,AMNH
Christine Engels,AMNH
Nonong Gayanilo, Univ. Miami, RSMAJ
AlastairHarborne, Univ. Exeter
n'in Hastings, UC Davis
Julie Kellner, UC Davis
Phil Kramer, The Nature Conservancy
Steve Litvin, Stanford University
John McManus, Univ. Miami, RSMAS
Fiorenza Micheli, Stanford University
Jessica Minnis, College ofThe Baham
Peter Mumby, Univ. Exeter
Don Olson, Univ. Miami, RSMAS
Steve Palumbi, Stanford University
Claire Paris-Limouzy, Univ. Miami, RS
;-McManus, Univ. Miami, RSMA
Thank you.
Peer-Reviewed Publications
Published and in press publications related to the EPA Star grant:
imby, Alastair R. Harborne, Jodene Willia
ugh, Fiorenza Micheli, Katherine E. Holmes,
d Paul G. Blackwell. 2007. Trophic cascad
eserve. PNAS, 104:8362-8367.
, Carrie V. Kappel, Daniel
aig P. Dahlgren, Claire B.
cilitates coral recruitment i
Harborne, A. R., PJ. Mumby, F. Micheli, C. T Perry, C.P Dahlgren, D. Brumbannh
and P. Kramer. In press. The functional value of Caribbean reef habitats to ec°
processes. Advances in Marine Biology.
Mumby, P. J., F. Micheli, C. P. Dahgren, S. Y. Litvin, A. B. Gill, D. R. Brumbaugh, K.
Broad, J. N. Sanchirico, C. V. Kappel, A. R. Harborne, K. E. Holmes. 2006. Marine
Parks Need Sharks? - response. Science 312: 527.
Mumby, P.J., C. P. Dahlgren, A. R. Harborne, C. V. Kappel, F. Micheli, D. R.
Brumbaugh, K. E. Holmes, J. M. Mendes, K. Broad, J. N. Sanchirico, K. Buch, ^
" — ".W. Stoffle, A. B. Gill. 2006. Fishing, trophic cascades, and the process
-azmg on coral reefs. Science 311: 98-101.
imby Peter J., Hedley, John D., Zychaluk, Kamila, Harborne, AlastairR. &
I, Paul G. (2006) Revisiting the catastrophic die-off of the urc
on Caribbean coral reefs: Fresh insiahts on resilience from a si
END
13
-------
Models of coral
community structure,
environmental variation
and connectivity
Carrie Kappel
National Center for Ecological Analysis & Synthesis
Dan Brumbaugh, Craig Dahlgren,
AlastairHarborne, Katherine Holmes,
Fiorenza Micheli, Peter Mumby, Claire Paris
What environmental factors are important
to coral abundance and community
structure at the seascape scale?
What role does connectivity play?
Relating coral abundances and community
structure to environmental variation
Brooders (24 sc
66% of variance explained
Soawners (17 sc
46% of variance explained
Significant predictor variables of coral
abundance and community structure
Exposure is consistently a strong predictor of coral
abundance and community patterns.
Larval retention and to some degree larval subsidies
showed weak, but significant effects at species and
community scales.
The signal pf past disturbances from bleaching and
hurricanes is detectable at the community level and for
some species.
assemblage of spawning corals and to certain species
at the seascape scale.
relief not included in spawners' analysis
Relating coral abundances and
community structure to environmental
variation
Predictor Variables:
Water depth
Vertical relief
Wave exposure
Temperature history
Hurricane history
Grazing by parrotfish
- Human population density
- Tourism intensity
• Larval subsidies
• Larval retention
Study locations
We conducted
habitat
mapping,
fish and
benthic
surveys
across the
Bahamas
archipelago
14
-------
Connectivity among islands
estimated from simulations of larval
dispersal
"Typical" brooding coral:
• 1 day pre-competent period
• 42 day max competency
• Year round planulae release
(Apr-Dec here)
"Typical" spawning coral:
• 5 day pre-competent period
• 30 day max competency
• Spawning in Aug & Sept
15
-------
Effects of Climate Change on
Ecosystem Services Provided by
Hawaiian Coral Reefs
P. L. Joklel, PI
R. W. Buddemeier
P. van Beukering
W. Haider, Z. Hausfather, D. Fautin, K. Rodgers, S. Saving, Y. Liu, K.
Zimmerman, K. Shapiro, S. Garcia, A. Andersson, I. Kuffner, F. Cox, F.
MacKenzie
1. Project Goals
Integrate and extend existing models to develop a
comprehensive, scenario-based analysis of the range of
possible effects of global climate change on ecosystem
services provided by the coral reefs of the Hawaiian
archipelago, and on the economic valuation of predicted
changes.
Features and emphasis
Cross-scale (reef to GCM cell)
Cross-domain (biological, environmental, socio-economic)
Responses to long-term means and short-term events
Valuation of lightly used or unused resources
Aesthetic, cultural, and spiritual values
Development and dissemination of tools as well as results
Model available for both on-line use and
download from a website (www.kgs.ku.edu/Hexacoral), providing for
community involvement through hands-on testing and feedback.
2. Lessons learned/Challenges
Challenge of Model Building at 3 levels
•Climate Change Modeling
•Biological Response Modeling
•Ecosystem Services Modeling
Lessons Learned
Unexpected: Ocean acidification, corals, crustose coralline algae, calcification
(coral growth and mortality central to all our work)
Timing: Economic downturn at time of valuation survey.
3. Interaction with clients
3 Sept 2008 "Identifying Bleaching Thresholds" Paul Jokiel.
NOAA Climate Workshop, HIMB
3 Sept 2008 "Techniques for Bleaching Assessments"
PaulJokieland Ku'ulei Rodgers NOAA Climate Workshop, HIMB
4 Sept 2008 "Reef Restoration". NOAA Climate Workshop, HIMB
5 Sept 2008 "Indigenous Practices and Climate Change" by Paul L. Jokiel.
NOAA Climate Workshop, HIMB
12 Oct 2008 "Impact of Ocean Acidification on Hawaiian Coral Reefs"
The Nature Conservancy Workshop on Ocean Acidification,
St. Stephens Diocesan Center, Kane'ohe, Hawai'i.
April 5-7, 2009. Climate Change Symposium. Local and global panel
member and moderator. Exploratorium. San Francisco, CA.
"Impacts of Climate Change in the Hawaiian Islands"
and "Impacts of Climate Change on Coral Reefs in America" Ku'ulei Rodgers
March 2-6, 2009 Pacific Science Inter-Congress in Tahiti French Polynesia
Climate Change Symposium "Impact of ocean acidification on Hawaiian
coral reefs in the 21st century" Presenter and moderator Paul Jokiel
Also- Upcoming Bleaching response team (managers), LocalAction Strategy
Committee On Climate Change, Hawaii Conservation Conference
Training Graduate Students, Undergraduates, Interns, Docents
4. Outcomes
ENVIRONMENTAL PROTECTION AGENCY (15 April 2009)
[EPA-HQ-OW-2009-0224;FRL-8892-5]
Ocean Acidification and Marine pH Water Quality Criteria
AGENCY: Environmental Protection
Agency (EPA). ACTION: Notice of data availability (NODA)
http://www.epa.gov/fedrgstr/EPA-WATER/2009/April/Day-15/w8638.pdf
Four of our recent EPA funded papers were cited in the Federal Register
announcement (Jokiel et al. 2008, Kuffner et al. 2008, Andersson et al.2009,
Buddemeier et al. 2008). The NODA notes that "EPA has supported the
development of the Coral Mortality and Bleaching Output (COMBO) model
to project the effects of climate change on coral reefs by calculating impacts
from changing sea surface temperature and CO2 concentration, ...".
The notice also mentions the EPA biocriteria initiative. This NODA is an
important step related to possible future EPA action related to controlling
ocean acidification and climate change.
-------
Response of Hawaiian corals to increased temperature
Hawaii Institute of Marine Biology
Coconut Island, Kaneohe Bay, HI
•Experiments are conducted in continuous flow outdoor
mesocosms that simulate the reef environment.
•Treatments: acidified to produce carbonate saturation states
predicted for year 2100, plus controls (3x replication).
Feb-Mar2006
- Non-calcifying algae
Non-calcifying mixec
algal assemblages
(% cover)
= B fi SB
100
% 80
T3
2 60
|
QJ 40
3 20
O
+52%
i.. i
j-
r
a
CCA recruits -78%
C1 C2 C3 T1
a
T2
•
T3
>
- Conlrol Treatment
-^ -^^ -^
b
CCA Cover -92%
hi..
C1 C2 C3 T1 T2 T3
50
40 ^
>
30 S
I
20 s
10 B
Kuffner IB, Andereson AJ, Jokiel PL, Rodgers KB, Mackenzie FT (2008)
Decreased abundance of crustose coralline algae due to ocean acidification.
Nature Geoecience 1:114-117
No mortality Coral calcification rate reduced 15-20%
Skeletal density decreased, branches thinner
No evidence of acclimation
The calculated decrease in CaCO3, production, estimated using the scenarios
considered by the International Panel on Climate Change (IPCC), is
10% between 1880 and 1990, and 9-30% (mid estimate: 22%) from 1990 to
2100. (Galluso el al. 1999).
Recruitment
Rhodolith accretion
(g buoyant weightyr1)
Control
mean±1 s.e.
+ 0.6±0.3
Acidification
mean ±1 s.e.
- 0.9±0.3
Percent
Difference
-250
Two-sample
t-test
PO.0001
-------
Maro Reef - Crustose Coralline Reef Formation
Photo by Paul Jokiel
Mesocosm Wall Settlements
Recruitment
Crustose coralline algae
(% cover on walls)
Turf algae
(% cover on walls)
Vermetid tubes (no. per
m?)
Serpulorbis sp.
Oyster % cover
Dendrostrea
sandwichensis
Barnacles (no. per m2)
Ba/anussp.
Barnacle size (mm)
Ba/anussp.
Bare Substratum
(% cover on walls)
Control
mean±1 s.e.
25^4.0
16.6±4.0
78.4±35.1
5.7±1.9
8.3±3.5
5.1±0.6
53.2±2.1
Acidification
mean±1 s.e.
3.6±0.9
14.5±2.6
7.7±3.1
4.4±0.8
4.5±1.6
6.0±0.2
77.5±1.1
Percent
Difference
-
Q4
-13
-90
-23
-46
+18
C+4£i
Two-sample
t-test
_,
p = 0.03_P
p = 0.69
p = 0.18
p = 0.56
p = 0.37
p = 0.25
p=oqoog>~
Jokieletal.2008
•^Vv
.* * Settlements of reef coral
* "• Pocillopora damicornis
L * 'C^^M
Control Acidification Percent
Mean Mean Change
±1 s.e. ±1 s.e.
Settlements per m2 55±14 49±18 -1 1
Diam. (mm) 2.5±0.2 2.8±0.1 +12
no. of polyps per settlement 4.4±0.9 5.3±0.6 +20
Jokiel et al. 2008
Consistent with: Albright, Mason and Langdon (2008)
Two-sample
t-test
p = 0.81
p = 0.44
p = 0.47
Measured Variable Control
±s.e.
Acidified Percent Statistical
±s.e. Difference Significance
bundles g ~1 coral
eggs bundle-1
5.2±5.4 7.2±10.1
15.2±3.0 13.3±3.7
Jokieletal.2008 I
Consistent with Fine and Tchernov (2007)
p = 0.47
p = 0.06
Photo: Waikiki Aquarium
June 21-22,2006
Net Ecosystem Calcification
(NEC = CaCO3 production -
dissolution)
+4.5 mmol CaCO3 rr1
At present seawater pCO2
-0.1 mmol CaCO3 rr1
At twice present pCO2
CORALS WILL STILL
BE GROWING WHILE
REEFS ARE DISSOLVING!
Andersson, Andreas J., lisa B. Kuffner, Fred T. Mackenzie, Paul L. Jokiel, and Ku'ulei S.
Rodgers. Adrian Tan (2009) Net loss of CaCO3 from coral reef communities due to
human induced seawater acidification. Biogeosciences Discuss., 6,1-20, 2009.
Climate-response modeling (The COMBO Model)*
Buddemeier, R. W., PL. Jokiel, K.M. Zimmerman, D.R. Lane, J. M. Carey, G.C. Bohling,
J.A. Martinich. (2008) A modeling tool to evaluate regional coral reef responses to
changes in climate and ocean chemistry Limnol Oceanogr. Meth. 6:395-411
Excel-based model of coral cover incorporating steady-state temperature and CO2 effects
on coral growth and mortality, plus a probabilistic treatment of high temperature stress
(bleaching mortality).
Key features:
-User has control of all factors (sensitivities, probabilities, environmental inputs).
-Regionally appropriate default values are provided (versions for other areas have also
been developed).
-The effects of quasi-steady-state temperature, CO2 concentration, and temperature
variation are assessed independently and accumulated into net change in cover.
*Much of the initial development and testing was funded by the EPA Climate Change
Program (Dr. Jane Leggett) through a contract with Stratus Consulting. That support
has contributed to strong synergy and mutual advantage for the two projects.
-------
The response model
user interface offers
drop-down menus,
explanatory pop-ups:
andfill-in-the-blanks
value selections.
Calculations are
performed in linked, •
user accessible
worksheets, with
options for replacing
the built-in datasets.
As input values are
changed, output
plots and tables (%
original cover vs.
time) update
immediately.
Predicted Changes in Coral Cover, Hawaii
Cover, nearshore, 50th Percentile
Buddemeier, R. W., P.L. Jokiel, K.M. Zimmerman, D.R. Lane, J. M. Carey, G.C. Bohling,
J.A. Martinich. (2008) A modeling tool to evaluate regional coral reef responses to
changes in climate and ocean chemistry Limnol Oceanogr. Meth. 6:395-411
Initial COMBO temperature
predictions in the Hawaiian Islands
are older IPCC (AR3) data and are
limited to three 5° latitude x 15°
longitude 'boxes', or regions.
These locations were updated
to 1°x1° boxes centered on
Johnston Atoll, Oahu, French
Frigate Shoals, and Midway
#0 XSD OT> JOT JOW M» 2100
The data itself was updated to
reflect the latest model
ensembles for current IPCC
predictions (AR4).
Historical temperature analysis
(e.g. Satellite SST) was used
to "train" the variability in these
predictions.
From SST datasets - calculation
of probability density functions
(PDFs) of summertime
temperatures
Example of predicted temperatures
from monthly temperatures PDFs
Midway has a the largest deviation — much
larger than Johnston, e.g. the 'tails' are much
longer. Extreme thermal anomalies happen
most often here.
'bleaching threshold'
n Monthly Climatology Value* 1°C
Run this prediction many times
Results in a 50% probability of some coral bleaching occurring each year
(every other year) by -2030 at Midway, -10 years later at the other
locations
But coral mortality is often linked to heat exposure, the time
duration of extreme water temperatures, eg 'Degree Heating Weeks'
(DHW) or 'Degree Heating Months' (DHM)
Probabilities of 2 DHM, from literature, likely to result in
widespread coral mortality:
I I
20% (once every five years) at Midway by -2050
French Frigate and Oahu -2060
Johnston by -2065
-------
Temperature and predictions were derived from the World Climate
Research Programme (WCRP) Coupled Model Intercomparison Project
phase three (CMIP3) multi-model dataset, constrained and downscaled
with historical data.
Figure-10: Model results for fractional change in coral
cover: episodic mortality with no assumed
ability of corals to adapt; Qa sensitivity = 0.2.
Representations are the same as for Figure 6.
IOC
60
1
JO
G
D«gr*e Heating Months (C°munth)
Validation observations
Extremely unlikely that viable coral populations will persist In
the shallow waters of the Hawaiian archipelago In 2100;
precipitous declines will likely start In the northern region
sometime between 2030 and 2050 with steady decline over the
entire century throughout the region.
Hoeke et al. (manuscript)
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
year
SST only
MF-0.15. Sens !ia-0.2
- SST, Sens aa = 0.2
-MF-05. Sans iia- 0.2
SST, Sens Qa = 0.4
Figure 2. The effects of various mortality and growth scenarios on a coral community
with an initial 30% cover. The initial dose is 12.3 DHW, and subsequent thresholds are
progressively 2.6 DHW higher; other than this threshold increase, no adaptation is
assumed. The A1B climate scenario and the temperature datasets identified in the
text are used. A: projected effects on growth of gradual SST increase
(no bleaching events) and no 0 effect. B: SST effects, with moderate 0 sensitivity
(0.2). C: SST effects, with high 0 sensitivity (0.4). D: 15% mortality per event with
moderate 0 sensitivity (0.2). E: as in D, but with 50% mortality per event.
Modeling Regional Coral Reef Responses to Global Warming and
Changes in Ocean Chemistry: Caribbean Case Study
R.W. Buddemeier, D.R. Lane, J.A. Martinich (submitted to Climatic Change
Socioeconomic modeling -
Adapted a STELLA dynamic model previously created (Cesar et al. 2005)
to determine changes in reef-related Total Economic Value (TEV) over
time. Modifications include adding climate change factors to
anthropogenic degradation; TEV factors are tourism, diving/snorkeling,
amenity, biodiversity, coastal protection, fish catch, and cultural and
traditional values.
TEV, especially for the non-use values, are assessed through a series
of surveys that will determine "willingness-to-pay" based on stated
choice analysis.
GIS and geographic similarity analysis (typology) was used to
apportion vales and probable changes from the case study areas to
the archipelago as a whole.
TEV As adapted from Cesar & Van Beukering (2004)
Choose your preferred diving / snorkellng spot
Conservation Fee Choice Task
-------
Discrete Choice Experiments
an alternative to revealed preference analysis
avoid the problem of multi-colinearity
multi-attribute trade-off analysis
evaluation of non-existing alternatives
decision support systems
specify models on any appropriate scale
(somewhat bound by behavioral relevance to
respondent)
Web-based surveys supplemented by targeted
interviews are an efficient and versatile means of
collecting responses
Hedonic Pricing (estimated real estate damage)
Adding the coral reef variables results in a significant
improvement of the hedonic price model's fit. a relatively small, but statistically
significant improvement of the model's explanatory power, suggesting that
coral reef has a significant impact on house prices, both in terms of
presence and quality.
Roy Brouwer, Sebastiaan Hess, Pieter van Beukering, Yi Liu,
and Sonia S. Garcia (manuscript)
A Hedonic Price Model of Coral Reef Quality in Hawaii
Projected impacts of climate change
Global temperature change (relative to pre-indusirial)
1*C 2*C 3'C 4°C 5'C
Falling crop yields in many anas.
particularly developing regions
Possible rising yields in
some high latitude regtons
Ecosystems
Extreme
Weather Rising intensity of storms, forest fins, droughts, flooding andttm
Events
Risk of Abrupt and
Major Irreversible
Changes
Increasing nsk of dangerous fwdfracM and
abrupt, large-scale shifts in the climate system
Business as usual (BAU) emissions vs. paths for
stabilizing CO2 concentration to limit ATaverage
20 |
Path for 50% chance
demanding than path
otai cost estimated at
.1% of global economy
>er year.
cf avoiding AT3Vfl >2°C (gold} is much more
or 60% chance of avoiding >3DC (green).
Solomon et al. 2008.
Irreversible climate
change due to carbon
dioxide emissions.
PNAS 106:11704-1709.
1800 2DOD 220D 24GO 2600 2BOO 3DOC
Proportions of
HCO3- and COf
adjust in
response to
added CO2, with
less CO3Z-
photoxynthexix
/'Alkal
Saturation state,
Q-arag, reflects
ease of
calcification, and
is controlled
largely by [CO3Z-]
Carbonate Chemistry
1xCO2 2xCO2
_J£ 280
C02oq < > H2C03 C™*.7C 8
(co, t H,O) yx ~d
— ,s ~\s
nityX itot*-^. HC03- + H' 1635
*| j>
C°Z* / ..,..,.,. CO ^ + H-
1915
CaC03 / 2300
calcification
8.17
£2 = [Caz-][CO>]/K«, Ql3
560
16
1867
2061 TC02
2300 T1IK
7.93 pH
2.8^ Q-arag
-------
Changes in aragonite saturation predicted to occur as atmospheric CO2
concentrations (ppm) Increase.
Hoegh-Guldberg et al. (2007). Science 318:1737-1742
-------
Wolfgang Haider
School of Resource and Environmental Management
WTP for mitigation of climate change effects to Hawaiian coral reefs:
A contingent choice study
Research Goal
To estimate the willingness to pay (WTP) for mitigating the effects of
climate change on coral reefs in Hawaii
Challenges
• To separate use values from non-use values
• To control for key components of the reef ecosystem
• To design a payment vehicle that is applicable from the present,
but leads to uncertain outcomes in the future
Method: Contingent Choice Survey
Environmental Valuation
Discrete Choice Experiment
Coral Cover
Coral Health
Fish#
Species
Diversity
Water Clarity
Mitigation Fee
Turtle
Relief
0%-9%
Poor
Few
Low
Diversity
Low
$10 $20
10 &4 9%
Moderate
Moderate
Moderate
Diversity
Moderate
$40 $60
No Turtle
Low
50%-89%
Good
High
High
Diversity
High
$80 $100
90%-100%
Very Good
Very High
Very High
Diversity
Very High
$150 $200
Turtle
Medium
High
Ideal for visualization
Discrete Choice Experiment -Visualization
Fish Numbers
Water Clarity
-------
Discrete Choice Experiment -Visualization of Turbidity
The Survey Instrument
Wtlewn* to our study on Hiwi.lan coral rwfa.
ih*c«oul •canomc ana eeologictf u«vM Th*MrMHl>
cMtangM «t« v* twc ««*> *ohM
Survey- Introductory Questions
Survey- Introduction of Coral Reefs
Scenic Beauty Estimation
(12 per respondent)
-------
—.1 0*e*i ,uu wrf mil u*» o4 n»T4*«uMBnt and HM HI MkN tu OWOM mi »r aMuiwt »• MM*
Choose your pretemd divinn / inorMkw ipot
13. ChoOM a future sconino For tlM nurinr protecEed ar
Eacn Of IhB WICMnng tug« (ril Uio* y«u DTI* CUtnM «n<) Iwa pOWbtn (u[uf» f ml condilicn* wrth gnO wChOul rtldSKion fo>
ffi* P4pafi»n*j™ic*u**B Mann* NttionAl MWiunwot iPMNMl und*> COMJriKXU ot eWnMfl eh«%D« (ap(XO»m«r«ly 30 yMTt n
Tfie tan below each image Oee» ni can-on] iMnilo twno ol tlnsu: icAnamu may uem unl^aly lho> may occur urxtar
special orcurTHtoncas Rcmombef . tho PUNM n • marine protadad area, and n NOT avUlabh tar nxmlion or may aOtm
f£*£^ ;;"—• Jl SUSJ, IS0"*' I MV.lund*»Hrlw4«iir'^lntl
CwafHMhh "~*r-i«r C "..—-•
MMwafR* M4). MantwdMc ifi»
MlSf»clK UMOwnm I (*fi'4»0" Miyli^i
«K«t^u» v^»nO«rtjy IMUN-
L^^'X- ""~ "*" "|^t*.»j=
-------
Results - Part Worth Utility
Water Clarity
• Linear Coded
• Main: (z= 7.5854)
•Hw:(z= 4.1891)
Coral Cover
• Linear Coded
• Main: (z= 5.2785)
• Hw: (z= 3.5037)
Mitigation Cost
• Linear Coded
• Main: (z= -16.588)
• Hw: (z= -11.2983)
js ll5
5 '
3
£ L5
5
l
1 '"
:- 05
I ~15
Water Oarity
-^_ — — — ^™-
m« 5-,,,. B-2,,,. 3-3,,,,
.•M,,l,nd • •H».«.d,nt
Coral Cover
-^™ — ^—-
o»-» ,*« =»« ™.,™
| • •M.nl.nd • •H».«,,,d,nt |
Mitigation Cost
•-•-=
1
10$ 2D$ 40$ 6D$ SD$ 1CO$ 150$ t2EO$
[^amland . *Hw. Resident
Results - Part Worth Utility
Fish Numbers
• Linear Coded
• Hw: (z= 3.1383)
Species Diversity
• Effects Coded
• Main: Moderate Div. (z= -2.6745)
Coral Health
• Effects Coded
• Main: Moderate Health (z= -3.1115)
• Main: Very Good Health (z= 2.6669)
• Hw: Moderate Health (z = -2.2689)
js ll5
1 -°5'
1
- 05
1 -°5'
I
js L5
= 05 •
*
Fish Numbers
Low (1-10) Moderate High (21-33) Very High
(11-33) (31-40)
Species Diversity
Low dru. tModerate High dv. Very High
| ••Mainland •Hw.Reidmt |
Coral Health
—
P.., ,„„*„,, B..d »,,B..d
• -M.nl.nd .H».«,,d,nt
Results - Decline Index
Current Scenario
Without Mitigation With Mitigation
Decline Index = 7
15
5 o
1-°:?
!-=
Decline Index
• J .. —
| .'MLlmd .•H».«,,d,B |
.
(Improvement Index = 4)
Explaining Heterogeneity
Which of the following statements best reflects your opinion about climate change?
• There will be climate change, but the implications will only be noticeable later
• Climate change is a fact and the first indications are evident already
• Evidence about climate change is still too uncertain; it is too early to know what will happen
• I do not believe in climate change
• Other (please specify...)
Climate Change Believers
Climate Change Skeptics
Decision Support Tool - Scenarios
Attribute
Water Clarity
Coral Health
Coral Cover
Fish Number
Species Diversity
Turtle
Mitigation Cost
Market Share Hawaii
Main/and
NetWTP Hawaii
Main/and
Best -case scenario
Without With
Current Mitigation Mitigation
Very Very
Good Good Good
25 fish 35 fish 35 fish
High Very High Very High
No
0$ 0$
72 8% 27 2%
• ($36.25)
V ($32.69)
Worst case scenario
Without With
95% 5 5
Very
Good Poor Poor
35 fish 5 fish 5 fish
0$ 200$
• 97 1% 2 9%
• ($129.04)
/ V ($68.42)
Maximum Difference
Without With
Current Mitigation Mitigation
Very Very
Good Poor Good
35 fish 5 fish 35 fish
Very High Low Very High
n| ^^^B n|
4 8% 95 2%
3 0% 97 0%
$109.82
/ $125.64
No gain, no pain guj|t
(residual for branding)
-------
Mike Taylor
Ben Beardmore
(REM, SFU)
Pieter van Beukering
Roy Brouwer
(CesarConsulting, NL)
The rest of the Team
Explaining Heterogeneity
S*\
Mainland
Water Clarity
Coral Cover
FishB
Turtle
Mitigation Fee
Climate Belief
Intercep
/ \ 3 Class
(lass 1 |n=669] Class 2 |n=234] Class 3 =81]
0.3391 (0.0573) • 0.2538(0.0571) • -0.4 0.3148)
0.1416(0.0365)''
t™,S'.'
0.957 (0.2474) •
0.0846(0.0663)
-0.2558(0.2233)
0.0302 (0.1424)
0.70BB/0.2197) •
0.252B/0.125S) •
-0.252B (0.1253) •
0.1424 (0.0237) • -0.02 (0.0331)
0.045 (0.1733) -0.22 (0.5416)
-0.1768 (0.111) 0.17 (0.2904)
-0.051(0.1626) 0.3 .49]
0.031(0.0459,1 -0.0 .137J
-0.0621(0.1554] 0.29 492J
-0.0807(0.1057) -0.03 3267J
0.2282(0.1177] -0.09 3323J
-0.0853(0.1616] -0.16 4I94J
-0.1158(0.1321] 0.43 3967J
0.1158(0.1321] -0.43 3967J
-0.2348 (0.0573) • -0.7369 (0.0622) • -0.96 0 305) •
-0.7883 (0.1247) • -0.2983 (0.1434) • 1.08 194) •
2.1391 (0.2112) • 0.5307 (0.2419) • -2.669 0.3523) •
tattnbutSthat were rffedsW!*/
[) Standard Error
Class 1=
Pro-Mitigation Climate Believers
Anti-Mitigation Climate Believers
Class 3=
Anti-Mitigation Climate Skeptics
-------
Hydrologic Forecasting for Characterization of
Non-Linear Responses of Freshwater Wetlands to
Climatic and Land Use Change in the
Susquehanna River Basin, USA
Denice Wardrop1; Christopher Duffy3; Kevin Dressier2; Raymond
Najjar4;Richard Ready5;Kristen Hychka1;
Susan Yetter1; and Mary M. Easterling1
EPA Project Officer Brandon Jones
CO;, Climate, and Ecological Processes |
Air Temperahyfe — [ £ Precipitation Regime
Characterizing non-linear responses through:
Selection of a linked terrestrial-aquatic ecosystem that provides
critical ecosystem services and ecological functions,
Characterization of various global change scenarios, incorporating
both climate and land cover, and a method of assessing their effect
on the identified ecosystem through the primary forcing factor of
hydrology (both alone and in conjunction with other human-
associated stressors),
Identification of potential nonlinear ecological responses (sensu
Scheffer et al., 2002) in the selected ecosystem as a result of these
changes, and
Estimation of the resultant change in ecosystem services on a
watershed and Basin-wide scale.
-------
Scaling Issues
What is our assessment unit?
How do we stratify the study area for the purpose of
sampling, modeling, and subsequent "scaling up"?
At what scale do we express final results?
How do we resolve differences in scale (both extent and
resolution) of different disciplinary components of the
project?
Assessment unit that:
Integrates freshwater wetlands with
important contextual landscape
Spatial and temporal scale that matches
ecosystem services
Scale capable of being modeled
Representative of the range of conditions in
theSRB
-------
Climate Scenarios
Or, the weather's fine, wish you were
here
Climate models
World Climate Research Programme's (WCRP's)
Coupled Model Intercomparison Project Phase 3
(CMIP3) multi-model dataset
Daily & monthly averages of 2-m temperature and
precipitation
21 models from 12 countries
Some models: multiple realizations
Horizontal resolution: -1.5° to 4.5°
20th century: observed forcing
21st century: A2 scenario
-------
Model Performance
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Climate Considerations
Models differ dramatically in their ability to
predict the climate of SRB
Model mean is superior to any individual model
What are the relevant metrics for model
evaluation
Raw model output is not as bad as thought (precip
is so local, would think model would be awful;
e.g., # of extreme wet days)
Hydro model runs
Daily output from model 1960-1990 (baseline)
Same thing, but 2035-2065
• Effect of climate change
What's the impact of the change in mean
climate versus change in variability
• Repeat first run , modify by change in mean annual
cycle (#l-#2)
I
Hydrology model
Or, water flowing underground.
-------
-------
-------
Hydrological Modeling
Considerations
Scale-appropriate and ecologically-relevant
hydrologic scenarios
Ecologically-relevant and powerful metrics are
difficult to identify
Spatially heterogeneous response to a
homogeneous forcing function
Absolute values of predictions are difficult to
utilize in a meaningful way
Ecological Non-linearities
Ecological Response
Or, it's all so complex
Could changes in the hydrologic regime
result in:
The loss of wetland area?
The loss of function through physical changes
and the loss of functional process zones?
-------
-------
Floodplain, Moderate Disturbance
-------
10
-------
Interactions with Clients
Lessons Learned
Pennsylvania Climate Change Impacts
Assessment (PA Climate Change Advisory
Panel)
Chesapeake Bay Climate Impacts Assessment
(CBPSTAC)
Integrated Riparian Assessment Unit for
Pennsylvania DEP
Climate change impacts for wetlands, Mid-
Atlantic Wetland Workgroup
Scale, scale, scale
Choosing climate models for ecological
applications
The good and bad news of, "which hydrology
metric would you like?"
Use data visualization tools whenever possible
Legend
I \<
11
-------
Sustainable Coastal Habitat Restoration
In the Pacific Northwest
Greg Hood, Tarang Kangaonkar,
John Rybczyk, Zhaoqing Yang
Modeling and
— Managing —
the Effects, Feedbacks
andR isks Associated
with Climate Change
Skagit River Delia 10,000 years ago
'-SHf 'ViiiWii:
Mudflats .:'
'** *iSJ
**!*sx M**
courtesy of Padiil/t Bay InEeipretiva Center
Skagit Historical and Current Deltas
B. Collins2001 (Uni
V. G. Hood, unpublished (Skagit System Cooper
1900 1060 2000 2€60 2100
Rahmstorf S. 2007. A semi-empirical approach to projecting
future sea-level rise. Science 315: 368-370.
-------
Approach: Link sea level rise
predictions to LIDAR data and to
known elevation distributions of
vegetation in the tidal marshes of
the Skagit delta.
Esluarine ernergenl
Esluartne shrub
Riverine tidal shrub
Riverine lid.il roresle
1) Cutoff from historical sources of sediment
2) No opportunity for upslope migration
3) An increasing rate of sea level rise
-------
Are the eelgrass beds in Padilla Bay at risk?
Are they accreting at rate that keeps pace sea level rise?
SET Site 1 (scouring corrected)
Padilla Bay
(2002 - 2008)
• Geologic Uplift.
0.02 cm/year
• Eustatio Sea Level Rise 0.34 Cm/year
• SET Elevation Change 0.25 ± 0.13cm/yi
Elevation Deficit = 0.57 cm/year
-------
These types of analyses ignore
climate change induced
changes in salinity, tidal
regime, river flow, and
sedimentation, for example,
and they imply linearity.
Project Goals
• Develop a predictive simulation model, incorporating non-linear
elevation feedbacks, of the ecological and geo-morphological
consequences of sea level rise and river flow alterations in Padilla and
SkagitBays.
• Use the model to guide the course of restoration efforts, given climate
change, in the Skagit River estuary.
• Specifically:
- Link a spatially explicit hydrodynamic/sediment model to a
mechanistic wetland elevation dynamics and vegetation model.
- Model will be initialized and calibrated using extensive, site specific
data collected as part of this r.
-------
Spatial Extension
Extended the REM to model a 3D
surface instead of a single point
Model input/output is in the form
of map grids covering the spatial
area of interest (AOI)
Grid cell size: 50 m x 50 m
Elevation/vegetation change for
each cell is modeled individually
AOI can be divided into
independently calibrated regions
2002 2102
0.56 meter increase in 100 years
- .„., ~ ,.
2002 2102
1.27 meter increase in 100 years (Rahmstorf 2007)
-IPCC-Low
-IPCC-M 3
-IPCC-High
-IPCC-High-WoMn
-IPCC-High+lc*
-R«hrrwtc«T-Md
- Rahmstoff-High
Next Step: Integration with
FVCOM
-------
Finite Volume Coastal Ocean Model (FVCOM)
Model Setup- Bathymetry and Grid
I Elements: 43810
Nodes: 25070
>• Layers: 10
Simulated surface salinity and velocity during flood and ebb tides.
Model Calibration- Salinity
SK2
SKI Sk-gitRiv,
"I*
I
— SK2 "v
|PS1
Stillaguamish Riv<
PS2
Interaction With Clients
« — - -
Skagit Climate Science Consortium:
SRSC, NOAA, USGS, UW, WWU, PNL,
Seattle City Light and others.
S^*^ Runoi
/\
iff, Rainfall, Rivet' Flow
& Groundwater
Coastal and Estuarine
Circulation - Habit!
Restoration
/Climate
{Change
-------
Sensitivity to Sea Level Rise and Climate
Change
What is the Effect of climate
change and sea level rise on
nearshore habitat?
- Estuarine rearing habitat for
juvenile salmon
• Availability of brackish
environment
• Stability of marshes and
mudflats
- Effective and sustainable
habitat restoration
• Need to anticipate future climate change
and sea level nse conditions
Recommendations
Reconsider our habitat restoration goals for salmon
recovery.
We need to run faster just to stay in place. We may
have seriously underestimated the amount of tidal
habitat restoration necessary to recover Chinook
salmon, because we have not accounted for the
restoration (dike and levee setbacks) that will likely be
necessary to compensate for sea level rise. The
uncertainty involved with climate change also argues
for ecologically conservative estimates of future fish
needs.
Model Boundary Conditions
Tide Elevation- NOAA Xtide
River Inflow- USGS gage
Wind- Pain field
.tide
•nyer
' wind
Model Cal
"
SKI Sk.git River
"
1
SK2
Stillaguamish River
PS2
- SN,jjj»SN2
.!i!Ll'
Yang and Khangaonkar (2008). Ecological Mode
(nrpvipwl
ibi
1
I,:
8
I? 0 •
ng -o/
•ation-
05 6/9/05
•A.
6/1
Velocity
2/05
JOvUM
^AV^
!
i^wv*^-,
o*rH
Dat
pn,
u
l^AJJWv
6.2,»5 604/01
/W
AsJVJi--
Data
/05 ,0/,4/05 0/10/05 ,0/22/05 0/26/0
|V
/06
^
SN1
WWW
0/16/06
I
1
—
^
.:..=
^
0/20/05 10/24/06 10/28/Of
Date
7
-------
Nonlinear response of Pacific Northwest
estuaries to changing hydroclimatic
conditions: flood frequency, recovery
time, and resilience
Anthony F. D'Andrea '2, Robert A. Wheatcroft2, Rhea Sanders2
Floods are increasing...as is sediment delivered
Wilson River, Tillamook Count/, OR
100 year rainfall patterns
Tillamook, Oregon
• climate models predict increase in
total ppt especially the frequency
of extreme (high rain) events
• river flow and flooding to PNW
increasing and amplified by
seasonal rainfall patterns
• combination of watershed (timber,
roads) and estuarine changes
(diking, channelization) have
decreased buffering capacity for
water and has led to increased
sediment flux for aiven
>reciDitation amount
With potentially important (but unknown) effects..
• rapid sedimentation during floods can lead
to abrupt changes in benthic intertidal
communities
> last 30 years in Pacific estuaries
- deposition up to 12cm thick
- reduction in benthic abundance/diversity
- alteration of tidef lot habitat
- rapid growth of NIS populations
> However most studies anecdotal or focused
on only one or several species
> Need: Community-level studies of flood
sedimentation impacts on estuarine benthic
communities
APPROACH: manipulative field study simulating the
effects of the frequency of floods (none, one, two) on
PNW benthic intertidal communities
Kev research Questions:
1. What are the rates and timescales of recovery (i.e.,
resilience)?
2. What is the impact of flood sedimentation on the functional
composition of the benthic community?
3. Does the within year frequency of floods alter the
response, composition, or recovery times of the community?
4. Do flood events increase the susceptibility of community to
colonization by non-indigenous species?
Project Goals
Why Netarts Bay?
1) Design and implement a manipulative field study to determine the
ecological effects of flood sedimentation on intertidal benthic
macroinvertebrate communities
2) Use a combination of high resolution benthic sampling and multivariate I
;; analyses of benthic community metrics to track the initial t ' '"
recovery, and resilience of the benthic community. •
to the benthic community
direct or indirect effects on survival or habitat suitability of sediments
4) Synthesize the datasets from this study to develop an empirical and
theoretical framework for predicting the effects of flood sedimentation
events on tidef lot macrobenthic communities in PNW estuaries and how
these changes impact ecologically and economically important biotic
resources and ecosystem services.
6th largest estuary in OR
large intertidal area (65%)
1. small watershed with no river
so no previous flood events
2. conservation estuary: historic
loss <1% (1900-1990)
3. relatively pristine - no port,
industry, shoreline alterations
4. marine dominated minimizing
physical/chemical variability
5. small size = accessibility
-------
Challenges and Lessons Learned II
Multiple uses complicate field work
• trade-offs necessary in long-term field studies
• Netarts Bay has a number of uses by stakeholders
including recreational clamming and oyster
aquaculture
• limited potential field site locations but did initiate
interaction with local users of system
Repeated samDlina-minimal disturbance
Main
Field
Experiment
accessibility
and minimal
-------
Part 1. Flood Layer and Physical Properties
Temporal chanae in field plots
• photodocumentation
• Benthic tripod
-ADV, CT sensor, OBS
Sediment Physical Properties
• total organic carbon
• sediment phytopigments
• porosity
• grain size
Sediment Geochemistry
• O2 microprofiles
• O2 core incubations
• Benthic photosynthesis
and O2 production rates
TOC patterns in flood layers
0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5
* I I
control plots consistently low TOC (<0.4%)
7-fold increase in TOC in flood plots relative to controls
some initial compaction of layer
persistent feature through at least first 150 days
O2 patterns in control and flood plots - Light
O2 concentration (nM)
0 200 400 600 0 200 400 600 0 200 400 600
t=12d Dark
Light
Part 2. Benthic Community Measurements
Chances in Community Structure and Diversity
• species composition
• abundances
• depth and seasonal patterns
• recovery times
Functional Chances
• feeding types, mobility
• deep vs. shallow dwelling
• native, non-indigenous, cryptogenic
Univariate and Multivariate Analyses
• identify key community metrics
• track community changes in control and flood plots
• indicator species for sediment-stressed communities
-------
Community Composition at Study Site
mean = 101.1 x Id3 nr2 (± 23.6 x Id3 )
1
^^~
1=1 Tonoids
M Conophiid amphipod:
1=1 Gammarid amphip.d!
1 1 Oimaceanx
1=1 Bivalves
^•1 Polychnetei
11 10 1C
Density (xlO3 nr2)
• 43 taxonomically
verified species to
date (-10% NIS)
' high abundance,
deposit-feeding
1 numerically
dominated by
tana ids
1 biomass dominated
by corophiid
amphipods in
summer and fall
• two components: a
surface- and deep-
dwelling community
time „ . _. ,
point Species Richness
1 Significant reduction in
species richness for at least
first 72 days post flood
sedimentation
' approximately 50% decrease
in number of species in flood
plots relative to controls
1 changes in functional
diversity have not yet been
assessed (planned for 2009-
2010)
time Infaunal Abundances
point
• Impacts observed by 2d
• Mean abundances in flood
plots consistently lower than
controls since day 2
• Effect of disturbance
measurable and significant >
2.5 months after initial flood
deposit with a >50%
depression of abundances
• Driven in large part by
changes in density of the
tanaid L. dubia
6 8 10 12 14
Number of species
Total infaunal abundance (xlO3 m~2)
-------
Outcomes
Flood sedimentation alters benthic intertidal habitat...
• The deposited flood layer persisted for >1 year with little
physical or biological mixing
• properties of flood sediments were distinct from ambient
intertidal sediments (TOC, grain size, O2, benthic 1°)
• remaining benthos in flood plots may be food limited as indicated
by combination of high TOC, deep oxygen penetration, and slow
recovery of benthic microalgae
Outcomes
Interaction with Clients
amphipods immediately left plots in response to disturbance
significant decrease in abundance and species richness -
combination of organism behavior and smothering stress
depressed abundances last for first 70+ days
flood layer not readily recolonized, even by mobile species
species traits (e.g. behavior)
may be important in determining
community response and
resilience to rapid sedimentation
disturbance events
Local Area Residents. Users, and Stakeholders
• communicated by direct discussions and press releases
• includes residents, recreational users, commercial oyster growers
• locally well-received by Netarts Bay residents and stakeholders
Oregon Resource Agencies
• Oregon Department of Fisheries and wildlife
• Oregon Department of State Lands
• Oregon Department of Agriculture
• Oregon Department of Land Conservation
Future Interactions...
• Project is ongoing so much of interaction both with resource
managers and residents will be done in future (6oal 4)
• empirical and theoretical framework for assessing risk to estuarine
benthic resources by river flood sedimentation events
-------
-------
Prairie Pothole Landscapes,
Climate Change, and Land
Management
W. Carter Johnson
Bruce V. Millett
Richard Voldseth
Brett Wern**'
Mirela Tulbu
David Naug
Overarching Goal
Complete and test new simulation model
(WETLANDSCAPE) to examine non-
linear or threshold effects caused by
climate change and land management on
complexes of glaciated prairie wetlands
-------
Wetland Vegetation Cover Cycle
V- '
Dry Marsh
-closed" or "choked" phas
(< = 25% open wat«r)
Regenerating Marsh
'•hemi-tn.Trsh" pha*e
(26% • 74% open wat«r)
Drought or I
Drawdown
,*
Lake Marsh
-opi-n W*t*f** ph ttsn
( > = 75% open water)
Muskrat Damage
simain«t High water Degenerating Marsh
-^ - -h^mi.mArsh" ph**«
{26% - 74% open water)
Temperature +4 C
ipitatjon +10%
Crop Types and Wetland Drought*
-------
Conceptual map of the modeling process to
determine cost-effective mitigation of climate
impacts on waterfowl productivity
Challenges
Expected, but surmountable challenges
in fine-tuning and calibrating a new
simulation model
Interactions with Clientele
Professional
-U. S. Forest Service: new project
proposed to adapt our wetland models to
forested wetlands of the northeastern
U.S.
-U. S. Fish and Wildlife Service:
collaboration with Wildlife Refuge System
on wetland monitoring and climate
change detection
—Research findings reprinted in two
new textbooks (Wetlands by Mitsch and
Gosselink; Biology of Freshwater
Wetlands by van der Valk)
Interactions with Clientele
Public
• Associated Press article carried
in 60 U. S. newspapers including NY
Times, LA Times, USA Today,
Washington Post
• Frequent radio interviews: 10 commercial
stations plus public and Earth Watch
Radio-Madison, Wl
• Frequent television interviews and press
conferences: Sioux Falls and
Minneapolis
Outcomes
Wildlife conservation community (federal,
state, private) using our research findings
to develop long-range plans to mitigate
for climate change effects on waterfowl.
Participation in national workshop to write
white paper on waterfowl and climate
change policy at Ducks Unlimited
Headquarters.
-------
IK PAVIS
Innovative Management Options to Prevent
Loss of
Ecosystem Services Provided by Chinook
Salmon in California:
Overcoming the Effects of
Climate Change
Dr. Peter Moyle, Dr. Lisa Thompson, Dr. David Purkey, Mr. Andrew Engi
Dr. Marisa Escobar, Mr. Christopher Mosser, Dr. MelanieTruan
Project Goals / Objectives
Long Term Goals
- Investigate how climate change and land use practices
change temperature and flow regimes within California
watersheds
- Determine if these changes will lead to a reduction in salmon
habitat and thus a reduction in salmon abundance
- Determine how a reduction in salmon abundance will affect
local biodiversity through food web interactions
Year 1 Goals
- Develop watershed model
- Parameterize baseline salmon population dynamics model
- Develop site specific food web conceptual model
BUTTE CREEK AND
FEATHER RIVER
BUTTE CREEK
SUBWATER SHEDS
Area:
Butte Creek: 382 km2
WB Feather River: 130km2
Average Precipitation:
900-1700 mm/year
Input Data, Models and Outputs
daily: P, Tair, RH, Wind
Legend
watershed_elevationband_inte2
2Esaoth Soils, Land Use
51201
Legend
Soi l_uni on_all_Cli p3_Di ssol v1
I |
SoilDeptJ
-------
Unimpaired
Hydrology
WEAP modeling period: 1986-2003
• Input Climate data - Daymet : 1980-2003
• Calibration Unimpaired flows data - ResSim: 1985-2005
Operations
Infrastructure
- Diversions
- Reservoirs
- Powerhouses
Operations
- Flow
Requirements
- Operation
Rules
Calibration Point Statistics with Operations - Butte Creek
Note:
Model period = 1986-2003
USGS Gage = Butte Creek near Chico 11390000
1 m3/s = 35 cfs
35 -
30 -
J?25 -
™E,20 -
I15"
"- 10 -
5 -
0 -
Butte Creek at Chico Gage 1986-2003
USGS Gage
WEAP Model
ONDJFMAMJJAS
Month
Spawning reach: Aprox. 17 km
Divided into:
- 5 reaches (A-E)
- 40subreaches(A1-A5, B1-
C12, D1-D8, E1-E7)
-------
Potential Management Options
Philbrook and Round Valley Reservoir stratification
data for summer 2004, 2005, 2006
Lessons Learned / Challenges WEAR
Input Climate Dataset:
-Daymet: 1980-2003
- Mauer dataset: to 2005
Temperature calibration:
- Short period for calibration: summer 2001,
2002, 2003
- Some sites with less data
Reservoir Temperature Routine:
- Reservoir stratification data: 2004, 2005, 2006
- Link with Matlab existent 1D routine
SALMOD
SA
Returning Adults >•
Outmigrating^
Juveniles
LMOD Structure
Holding /Spawning
Adults
1
Eggs and Alevin
Temperature
Habitat
Flow
1
Fry
1
0+ Parr
1
1+ Parr
,,
Fecundity
Growth
Mortality
Movement
SALMOD Data Sources
Government Reports
- California Department of Fish and Game - Butte Creek Chinook Life
History Investigations (1995 - present)
- EPA-Watertemperatureeffects
- USFWS-Survival, Flow - Habitat Relationships
Peer Reviewed Publications
- Crisp 1981
- German and Quinn 1991
- Clarke and Shellbourne 1985
Books
- Pacific Salmon Life Histories (Groot and Margolis)
- Behavior and Ecology of Pacific Salmon and Trout (Quinn)
SALMOD Relationships
Egg Mortality vs. Temperature
Fry Growth Rate vs. Temperature
|0.4-
I0.2
°-°4
0.02
0 10 20 30 40 50 0 10 20 30 40 50
T'mp"*Ur"°C| T.mp.ratur.f'C]
Fecundity vs. Weight
0 20,000 40,000 60,000
Weight (g)
-------
Adult Summer Survival and Spawning
SflLMOD OUTPUT
Tim&Steos 545to700/CompUnts. tKJ 13
Bulls Cr&ak Spn ri'>Run r.hin.if.k Sdnwn T«ST
M« at Hi sn IM i)« wo (to t» in ui (» uo t» «o wo ™
Tttin Si.|n
Input returning
adults in spring
Baseline model
captures shift
from holding to
spawning and
mortality
Calibrate with
summer
mortality and
spawner
survey data
Juvenile Outmigration
,-!•.••:< a.ii-.n
Tims Steps MS to WO (Comp Urals I to 15
Bute Creek Spring.Sun Chmwjk Salmon test
Integrates
watershed
effects on all
life-stages
Baseline model
generates
reasonable
juvenile
abundance
Calibrate with
CDFG
outmigrant trap
data?
Lessons Learned / Challenges - SALMOD
• Program limitations
- Weekly time step
• Summer maximum temperatures
- Temperature & Flow habitat unit restrictions
• Program calibration
- Adult sampling method changes
- Juvenile outmigrant estimation
Food Web Conceptual Model
Role of spring-run Chinook salmon in delivering
marine-derived nutrients (MDN) to the Butte Creek
ecosystem
Develop an integrated conceptual model
- Fate of salmon-derived nutrients
- Nutrient flowpaths
- Aquatic-terrestrial trophic linkages
Expert panel
- Evaluate aquatic and terrestrial community structure and
function under different climate and management scenarios
Terrestrial Consumers of MDN
-------
Use of Stable Isotopes in Tracking MDN
MDN (C, N and S) in adult salmon tissues are
isotopically distinct from corresponding nutrients
from freshwater and riparian ecosystems
Isotopes accumulate at successively higher
trophic levels due to food web dynamics and
trophic relationships
Compare isotope ratios in areas with and without
salmon
- Surrogate for loss of salmon due to climate change
MDN above and below barriers to migration
Clltn R \,-,-k i. K
" ^.." "..
T
Invertebrates
Lessons Learned / Challenges
MDN-Food Webs
Lack of data on components of the terrestrial
ecosystem
Broaden spatial and temporal extent
• Deploy motion-sensor cameras across a broader
spatial and temporal scale
Need greater resolution of isotopic relationships
• Preliminary results lack clear resolution between
marine and freshwater derived nutrients in
producers and consumers
• Need better baseline samples (salmon tissue,
aquatic and terrestrial invertebrates)
Interactions with Clients
Presentations
- Spring-run Chinook Salmon Symposium (July 08)
- California Department of Fish and Game (May 09)
- Centerville Historical and Recreation Association (Sept 09)
Other
- Baltic Nest Institute, Stockholm Resilience Centre, Stockholm
University (Thorsten Blenckner)
- California Department of Fish and Game
- California Sportfishing Protection Alliance (Chris Shutes)
- Friends of Butte Creek (Allen Harthorn)
- National Center for Atmospheric Research
- Pacific Gas & Electric Co.
Outcomes
Tasks on track
Efficient and effective multidisciplinary
research program
Stakeholders and other parties
interested in research outcomes
- Resource managers
-Watershed groups
- Implications for restoration of spring-run
Chinook in San Joaquin River
-------
Hydrologic Thresholds for
Biodiversity in Arid and Semiarid
Riparian Ecosystems: Importance of
Climate Change and Variability
Tom Meixner and James Hogan,
University of Arizona
Julie Stromberg,
Arizona State University
Project Goals - Hypotheses
• 1) Decadal scale climate change and variability
alter riparian aquifer recharge through
mechanisms that depend on the magnitude,
frequency and seasonality of flooding, and exert
the greatest change in reaches that receive
minimal groundwater inflow from the regional
aquifer.
• 2) Riparian vegetation structure responds non-
linear/y as riparian aquifers are dewatered and as
key hydro/ogic thresholds for survivorship of plant
species are exceeded.
• 3) Decadal scale climate variability and change
alters riparian ecosystem water budgets that in
turn change vegetation structure and function and
the ecosystem services provided to society.
Project Overview
thine surface hydrafogtc mcdcdng
gttuttdwNtr modeling «n It
in. '.•. .1. ' • • 'i^i ••••' . • i -
Task 3E - Sensitivity
Matrix
.Combine vacated •
and hyrtiologlt i
osystem Services Impacts I
Statement of the problem: Mountain
* Mountain Front
Block Recharge
f
MOIST A
DRY
Recharge
\ Ephemeral Channel
Recharge
Riparian Water Sources
Recharge during
monsoon runoff
80
-6
Isotopes of water - natural tracer of source
Riparian wells span range between end members
Baseflow skewed toward monsoon runoff
Quantify % using simple mixing model
Uncertainty associated with runoff end member 5
Regional Climate Gradier
Montly Precipitation in Sonora,
Arizona and Utah
-------
Study #1-1
Study Area
Lower Bill Williams'is in
most arid part of the basin
Predictable floods
- Planned floods
- Upper basin USGS
gauges
& lake levels as
indicators
Study#1-2
Study site: Upper San Pedro Basin, AZ
The Upper San Pedro basin is about 4500 km2 with mean
annual precipitation of 41 cm. Historically, July through
September are the wettest months. Q
Study #1-2
10
Study #1-2
-£• Percent Summer Hoodwater In Riparian Gioundwater
5" 1M 1 1 1 1 1 r
*
1
_
1 «
3
1
7B%
6J16
T
X1 > ^. 1 -.:s. ,H^
4S%
15% 4 • !4% J, td '
10% P
T . $ I T
t 4s ?
*• Model Segment
11
Study #1-3
Mountain System Recharge (MSR)
Mountain Block
Recharge
Mountain Front
Recharge
Mountains are source region for 50% of rivers on globe
Significant component of recharge in many semi-arid basins
12
-------
Study#1-3
Statement of the problem:
Most ground-water models apply
temporally and spatially static
recharge rates across a
ground-water basin.
1
Empirical relationships
Model calibration
Limitations:
Q Complexity of recharge processes
a Lack of observational data
Q Lack of spatial analysis tools
| Groundwater Response
—I
13
Empirical Models:
Temporal Discretization
Seasonal Isotopic Ratios
70% MSR winter and 30% summer ~
(Wahi et al (2008))
Normalized Seasonal Wetness Ind
tSum
JZT
jf fvnxxtiaa
Seasonal MSR Ratios
.* (Hydrologic data)
P .and ET values (NARR)
f"^ 1 ft/Wff *
*y
rtwr
31
»11»»»lliM
'XL™ 14
Comparison between hydrologically &
isotopically scaled MSR
t>
Summer
y = 1.15x * 0.20
R1 = 0.81
Incorporating ET values enhanced
V1SR predictions especially for
summer season.
Upper San Pedro basin
Winter
Hypothesis 2
> Threshold #1: Flow permanence and decline of
hvdric (obligate wetland) herbaceous plants
> Thresholds #2: Groundwater depth and decline of
shallow-rooted and deep-rooted obligate riparian
herbaceous plants.
> Threshold #3: Groundwater depth and decline of
shallow-rooted and deep-rooted pioneer trees.
16
Study #2-1: Threshold #1: Flow permanence and
spatial patterns decline of hydric herbaceous plants
Problem statement: The regional uniformity of the
response of riparian vegetation to declines in stream
flow permanence is unknown.
Methods:
Surface flow monitored
monthly for presence/
absence at ephemeral
to perennial sites at
multiple rivers;
Vegetation sampled
along active channel.
Lov^flow channel zone, pre-monsoon season
Results:
Wetland
perennial
herbaceous
plants show
consistent
pattern of
sharp decline
in abundance
as stream flow
becomes non-
perennial
Conclusion: Abundance of a key stream community
type (riverine marshland) will decline with increasing aridity
-------
Study #2-2, Variance through time
Problem statement: Temporal and spatial response
of streamside vegetation to fluctuations in stream
flow poorly known.
Methods: Multi-
year field
monitoring of
vegetation (and
soil seed banks)
at ephemeral,
intermittent, and
perennial sites
through wet-dry
period.
II !
Stromberg JC, AF Hazelton, MS White, JM White, RA Fischer. 2009 (expected).
Ephemera! wetlands along a spatially intermittent river: Temporal patterns of vegetation
development. 19
Results: In
years with wet
winters, flood
runoff sustains
flows at
ephemeral
sites,
allowing for
development of
"ephemeral
wetlands"
ni
1
;
i
Je "spider" charts show numbers of hydric, mesic, and
ric plant species present only at a perennial site, only
an ephemeral site, or at both hydrologic site types,
ring different years.
20
Soil seed banks provide
resilience, allow distinct
plant communities to
develop in years with
varying flow conditions.
Diversity of seed banks
influenced by proximity to
perennial reach.
Conclusion: Spatial
distribution of wet and dry
reaches influences
vegetation response to
stream flow changes.
Hassayampa River Soil Seed Bank
Density and diversity of
wetland species in soil seed
banks of ephemeral reach
decline with distance
downstream from a perennial
reach
21
I'm C -
/
Citizen Wet Dry Mapping
Annual volunteer effort to Map Wet and Dry
reaches of San Pedro
Simple but cril
Citizen Scienc
Pal
V «* «
S *— •
a 6 . mt
° *•"•
» "• a »<^
* —
i ^ —
cal data
e needed
ominas Charleston To
— TZH
in "ir
~:
"""
mbstone
.
* *
»*
0 10 20 30 40 50 60 70 SO
Qs Distinct Front Surt of Survey • Mexico km
^^^^^^^^^M^^^^^^^^^^M ^T^
Cooperative effort of Upper San Pedro Partnership and UA-Project NEMO
22
Study #2-3 Thresholds #2: Groundwater depth and decline
of woody riparian plants
Problem statement: Regional uniformity of riparian
vegetation response to declines in water table is unknown
Methods: Monitoring wells
monitored at multiple sites,
multiple rivers; woody
vegetation sampled for
abundance and composition.
Preliminary
results:
Woody species,
grouped by
strategy type,
show similar
trends among
rivers.
Hypothesis 3: Decadal scale climate variability
and change can alter riparian ecosystem water
budgets that in turn change vegetation
structure and function as well as the services
provided to society by these ecosystems.
Floods, groundwater, recruitment and patch
dynamics of hvdromesic pioneer trees and shrubs.
24
-------
Study #3-1 Scenarios
Wet Dry Scenarios
Study #3-2: Modeling
Problem statement: Recruitment response of riparian
tree species to interactions between depth to water
table and flood patterns not yet quantified.
Methods: Modeling approach being used to estimate
potential seedling densities of riparian tree
species (Mark Dixon, Univ. South Dakota).
Results: Modeled
densities vary among
San Pedro
River sites with
different stream
hydrology and among
years with different
flow conditions.
26
Study #3-3: Historic legacies
Problem statement: Legacies of past extreme flood
events may be shaping current vegetation trajectories
and response to climate change.
Climate extremes + land use extremes i=> Historic entrenchment
of San Pedro River
"It was probably during the 1896 flood that a
channel almost 244 m wide and 6 m deep
developed..." (Hereford and Betancourt 2009).
Methods: Aerial photographs of the Upper San Pedro River
from 1935, 1955, 1978 and 2003 analyzed to assess temporal
and spatial trends in vegetation cover type abundance.
Results: As a ™°
legacy of past BOO
extreme jjj 600
disturbance, | 400
pioneer woody
vegetation has
Flood
n
II.
been expanding ° ^
over past 1/2 ^i*1*1
century.
Status in 1955
Populus/Saljx
Shrub. /wood.
Grassland
Bare ground
Farm + urban
Sum
„
J
In
= 1978
i= 2003
\»"4 \»"4 o^4 yi^ o»ie ^
" &* ^tJ^t1'4®*"^ ^
w
Status in 2003
Populus
Salix
15*
10%
m
56%
0%
100%
Shrub./
wood.
3%
46%
22%
29%
0%
100%
Grass-
land
7%
4%
4IS
48%
0%
100%
Bare
ground
9%
23%
im
50%
0%
100%
Farm
+urban
0%
0%
0%
0%
0%
100%
Most
Populus/Salix
points mapped
ground (as
mapped in 1955)
28
Pioneer trees have
sequentially established
during years with
suitable flood conditions.
As forest density
increased, the sediments
stabilized and flood
intensity decreased.
Recruitment now shifting
to a 'fringe replacement
. mode' with narrow bands
of young trees stabilizing
the channel.
29
As the pioneer forests expanded in the post-entrenchment
floodplain, water availability shaped species composition.
Conclusion:
Riparian forest
patterns are a
product of
interactions
between recent
climatic cycles and
land and water use
and past extreme
events that set in
motion trajectories
of change.
ShrublandAvoodi.w! ("wax, D/:J
show rr
in drier reaches
'V '
, I , i , I I , i II II .i II ,i
in 2007/2008, based on data from TNC
30
-------
Future and ongoing work:
Greenhouse studies of plant rooting depth and
response to water table decline underway.
Classification of riparian plants into strategy groups
based on response to drought and flooding
underway.
28
31
Lessons Learned-Challenges
• One PI left science another moved institutions
- Maintain flexibility
- Fortunate to have redundancy of skill with a graduate
student on team
• Shifted to simpler surface water model rather than
HEC-RAS
- Our questions do not need more complex model
- Simplest model should be preferred
32
Interactions with clients/stakeholders
Stromberg J, M Tluczek, AF Hazelton. Long-term
riparian forest change on the Upper San Pedro River.
Upper San Pedro Partnership Technical Committee
Meeting. May 20, 2009. Sierra Vista, AZ.
MeixnerT, S. Simpson Flood Water Sources in Planet
Valley Aquifer and the Bill Williams. Bill Williams
River Steering Committee. September 16, 2008,
Phoenix, AZ.
Continuing cooperation with Bill Williams Steering
Committee, Upper San Pedro Partnership and TNC
Hassayampa Preserve
33
Outcomes
• Flood recharge a critical process in all three
systems
• More so in two farther north systems
- Likely a geologic geographic control
- rather than seasonality
• Vegetation response has significant lag times
• Perennial reach presence has significant
downstream influence on vegetation
34
Next Steps
• Build seasonal groundwater model of San Pedro
• Developing scenarios
• Greenhouse studies of plant rooting depth and
response to water table decline underway.
• Classification of riparian plants into strategy
groups based on response to drought and flooding
underway.
35
Acknowledgemen
Co-l's
Andrea Hazelton, Meg White, Melanie Tluczek, Scott Simpson, Hoori Ajami
Environmental Protection Agency (GAD-R833025, fP-916987)
Project NEMO, Upper San Pedro Partnership
City of Scottsdale & the Staffords
Chris Eastoe (UA-Geosciences)
Andrew Hautzinger (USFWS)
Stan Culling (USFWS)
36
-------
ntegrated Biochmatic-Dynamic
Modeling of Climate Change Impacts
on Agricultural and Invasive Plant
Distributions in the United States
Wei Gao
USDAUV-B Monitoring a
Colorado State University
Xin-Zhong Liang, Shu
University of Illinois
Thomas Stohlgren
I Research Program
in Liu'
Biological invasions of nonindigenous species are
serious threats to the U.S. natural and managed
ecosystems ($ 120B/yr damage, $27B/yr crop loss)
Rapid growth in trade worldwide or globalization
exacerbates U.S. invasive species problems
Climate is the dominant determinant of the geographic
distribution of plant species, native or alien
Climate change has already caused unequivocal shifts in
distributions and abundances of species, and even
pushed certain native species to extinction
Non-native Species Established in Species-Rich Counties
Non-native plant species/county
Native Plant Species Density (#fkm2)
Invasive
Species
The Rich
Get Richer
Stohlgren, T.J., D. Barnett, C. Flather, J. Kartesz, and
B. Peterjohn. 2005. Plant species invasions along the
latitudinal gradient in the United States. Ecology 86:
2298-2309.
Stohlgren, T.J., D. Barnett, C. Flather, P. Fuller, B.
Peterjohn, J. Kartesz, and L.L. Master. 2005. Species
fishes in the United States. Biological Invasions 8:
427-457.
To better understand how global climate changes
affect the U.S. agricultural and invasive plant species
distributions focusing on crop production
To account for both adaptation of alternative crops
and invasion of non-native species to enable decision
makers to design effective management and control
strategies for a sustainable future agroecosystem
Proposed Research
To develop a robust SEM (species environmental
matching) to best capture the observed agricultural and
invasive plant species distribution using the conditions
from CEP (climate-ecosystem predictive).
To make projections for the potential niche
distributions of alternative crops adaptable to the likely
range of climate changes in the future using CEP.
To project the geographic distribution and abundance
of U.S. agricultural weeds and invasive plant species by
integrating newly-developed SEM and future soil and
bioclimatic conditions simulated by CEP.
-------
Northeast U.S. Assessment
PDF lor Precipitation
PDF for T
HCN
HCM
PCM
HCN
PCM
PCM
(b)
20 40 60
Precipitation Rate (mm/day)
•ID U 10 30 80 40 fiD
Much More Than That...
PGR JJA 1flB6_1BB5 P
110W 100W 90W 80W 110W 100W 90W 80W
1.0 15 2.0 2.5 3.0 3.5 4,0 4.5 5.0 55 6.0 6.5 7.0 7.5 8.0
Optimized Physics-Ensemble Prediction
KF Climate Mean (mm/day) QR
110W 100W BOW
now ioow BOW eow
110W 100W BOW
now ioow BOW eow
0.5 1.5 2.5 3.5 4.5 5.5
Propagation of GCM Present Climate Biases
into Future Change Projections: Temperature
PCIACB5 TA 1990s
POR-PCM TA 1990s
PGR-PCM TA 2090s A1R
Bias
Summer 2m Temperature (°C)
Champaign
Future projection
Present
n
1
2050
.
. li
2100
Jl
n
III 111 111 111 111 111 111 111 111 111
Bl B2 A2 A1F1 Bl B2 A2 A1F1
-------
CWRF
Climate-Ecosystem Interaction
Mosr Ecosystem -lodel
&?•" US DA
2003*
U.S. Cotton Yield in 1979-2005
(a) Observed yield 19TO-2005 (kg/ha)
Dynamic crop growth model
can be integrated with
satellite remote sensing to
predict annual yields, thus
help regulate market
supply-demand, make
strategic assessment of
optimal operation practices,
project food trend due to
climate change...
predictive'
Tom Stohlgren
National Institute of Invasive Species Science (www.NUSS.org)
Method: relate observed species distributions to
environmental envelopes
Assumption: the fitted observational relationships
to be an adequate representation of the realized
niche of the species under a stable equilibrium or
quasi-equilibrium constant
For this study
• Model: MaxEnt
• Presence point data: cheatgrass
• Environmental layers: 10
Cheatgrass was brought in 1898 from Eurasia into Washington
state by researchers looking for new grasses to make hay
It is widely distributed throughout the U.S. except for Florida
Its seeds can germinate after years of dormancy
Wind can carry its seeds into areas that have been cleared
No insects are yet available to control its spread
Hand pulling cheatgrass is very hard work
-------
Precipitation
Annual mean
1982-2007
\
Summer mean
June-August
1982-2007
Grow season mean
April-September
1982-2007
•'
T r
. ••'
,
Temperature
"
Tmaxsummer
Crop suitability
(968 crops whose soil and climate requirements were identified from
a literature search, Bowen and Hollinger, 2004)
Temperature
Precipitation
Growing season lengl
Minimum winter
temperature
0 Unsuitable ^^.
1 Slightlysuitable
2 Moderately suitable
3 Suitable
4 Highly suitable
Follow the law of the minimun
if the suitability score for either
unsuitable for the crop
EPA STAR
2009-2012
Xin-Zhong Liang
R. Srinivasan
Pushpa Tuppad
Jeff Arnold
Donald Wuebbles
Award in process
August2009
FOCUS
Nutrients
Pathogens
Bacteria
Sediments
Tfn n r""••
USGS SO. 03OTOOO
-------
uture Perspective
Expand the modeling system to predict most
major crops
Incorporate multi subgrids of land use/land
* Develop capability to simulate air pollution
impacts on crop
7S-- '
* Develop capability to study agriculture water
quality problems
* Develop capability to study the agroecosystem
-------
Global change and the
cryptic invasion by
transgenes of native and
weedy species
Cynthia Sagers
Biological Sciences
University of Arkansas
Pete Van De Water
Geosciences
Cal State U. Fresno
Steven Travers
Biology
North Dakota State University
transgene - a gene from one species that has
been introduced into the genome of another
species through biotechnology
Herbert Baker (1971) Human influence on plant
evolution. Bioscience21:108
Evidence for crop-to-weed gene flow
Rank
1
4
5
6
8
9
10
Crop
Wheat
Rice
Mane
Soybean
Barley
Cotton
Sorghum
Millet
Brfn;
Rapeseed
Scientific name
T lurgidum
O glabemma
_*-,.-- ->ys mays
Glycinemax
vulgare
Gossypium
hirsutum, G
barbadense
Sorghum
Eleusme
:-.•!-• :'.--lusvulgaris
Brassica napus,
B. rapa
Kile-hectares
228,131
149,555
143,633
67,500
65,310
51,290
45,249
38,077
28,671
24,044
Evidence for
hybridization
+?
m
+
Ellstrandetal. 1999
Evidence for crop-to-weed gene flow
Rank
1
3
4
5
6
7
8
9
10
Crop
Wheat
Rice
Maize
£• •!..-.-..
Barley
Cotton
Sorghum
Millet
Beans
Rapeseed
Scientific name
T turgidum
Oryza saliva,
O glabemma
Zea mays mays
Glycme max
Hordeum
Gossypium
hirsutum, G
Sorghum
bicolor
Eleusme
FtiaseoiutiviJiq&'K
Brassica napus,
B. rapa
Kilohectares
228,131
149,555
143,633
67,500
65,310
51,290
45,249
38,077
28,671
24,044
Evidence for
hybridization
m
+
Ellstrandetal. 1999
-------
Making canola:
Srass/ca rapa v Brassica oleracea
(AA, 2N = 20) ^ (CC, 2N = 18)
Srass/ca napus
(AACC, 2n =38)
Canola is sexually
compatible with least 44
brassicaceous species
Canola Harvests^ Acres
Making hybrids:
Srass/ca rapa
(AA, 2N = 20)
Srass/ca napus
(AACC, 2N =38)
F, hybrid
2N = 20-29
What factors promote gene flow?
• Coexistence
• Sexual compatibility
• Hybrid vigor
• Selective benefit
-------
Herbivory, heterosis and gene
flow in engineered populations
of Brassica and Brassica-
hybrids
us EPA, NRC
US EPANHEERLWED, Corvallis, Oregon
Hybrids:
B. rapa X canola
B. rapa X GM canola
Conclusions
Risk of transgene flow is a function of:
1) genetic background
2) competition
3) level of selection
-------
How will climate and land
use changes influence the
adventitious presence of
transgenes?
' 'ZONE
ZONE
ZONE
• ZONE
• ZONE
: ZONE
ZONE
ZONE
LJ ZONE
• ZONE
ZONE
Aacvr do
Objectives:
•Characterize variability among weedy populations in
traits related to outcrossing
•Incorporate these parameters into existing climate
change/land use change models to assess changing
risk of transgene flow
What factors promote gene flow?
• Coexistence
• Sexual compatibility
• Hybrid vigor
• Selective benefit
B. rapa L.
Sinapis arvensis L.
-------
Weed surveys
Objectives:
1) map local distributions
2) monitor transgene flow
3) model risk
Sentinel plant study
Objectives:
1) measure transgene flow
2) assess geographic variation
In gene flow rate
Long distance gene flow in Agrostis stolonifera
(Watrudetal.2002).
Greenhouse study
Objectives:
1) evaluate genetic variability of functional
traits among B. rapa populations
2) measure pollinator preference in controlled
environment
Modeling
Objectives:
1) develop phenological maps
for sexually compatible
relatives
2) create a probabilistic
model of changing risks of
transgene flow
"...seed crops hybridize with their ancestral races to produce
weedy derivatives wherever wild and cultivated kinds
are sympatric." DeWet and Harlan (1975)
August, Reuters: US rice farmers sue Bayer CropScience over GM rice
Rice farmers in Arkansas, Missouri, Mississippi, Louisiana, Texas and California have
sued Bayer CropScience, alleging its genetically modified rice has contaminated the crop,
attorneys for the farmers said on Monday. The lawsuit was filed on Monday in the U.S.
District Court for the Eastern District of Arkansas in Little Rock, law firm Cohen, Milstein,
Hausfeld & Toll said in a statement. The farmers alleged that the unit of Germany's Bayer
AG failed to prevent its genetically modified rice, which has not been
approved for human consumption, from entering the food chain. As a result, they said,
Japan and the European Union have placed strict limits on U.S. rice imports and U.S. rice
prices have dropped dramatically. A Bayer representative could not be immediately
reached for comment.
-------
Acknowledgements
Collaborators
Robert Bacon (UA, Fayetteville)
Paola Barriga (UA, Fayetteville)
Nonnie Bautista (OSU)
Connie Burdick (US EPA)
John Fowler (OSU)
Christine Hauther (U. Memphis)
E. Henry Lee (US EPA)
Jason Londo(US EPA)
Tom Millican (UA, Fayetteville)
Sharon Morgan (UA, Fayetteville)
Chris Pires(Mizzou)
C. Neal Steward (U. Tennessee)
Lidia Watrud (US EPA)
Agency support
Dynamac Corp.
National Research Council
USDANRI
US EPA
Escape from cultivation and the influence of crop
plants on the evolution of native populations
Allele frequencies in sink
populations are a function of
initial frequencies, migration
rate, and selection.
Question: what factors promote interspecific gene flow?
hybrid vigor
selective benefit of transgene
sexual compatibility
Question: what factors promote interspecific gene flow?
hybrid vigor
selective benefit of transgene
sexual compatibility
Predict:
Heterosis (F1 > parentals)
Performance GM > Performance non-GM
Transgenic seeds in non-transgenic plants
Risk = (probability of an accident) x (losses per accident)
Outcrossing:
Brass/ca rapa y Brass/ca o/eracea
(AA, 2N = 20) (CC, 2N =
Brass/ca napus
(AACC, 2N =38)
-------
2006 Seed mass data - by mating type
Canola X canola
6. rapa X canola
6. rapa X 6. rapa
B. rapa X GM-canola
GM-canola X GM-canola
Seed mass (g)
2007 - Seed mass - herbivory by mating type
interaction
Canola X canola
6. rapa X canola
6. rapa X 6. rapa
B. rapa X GM-canola
GM-canola X GM-canola
Seed mass (g)
Question: what factors promote interspecific gene flow?
hybrid vigor
selective benefit of transgene
gene flow
Found:
\ Heterosis (F1 > parentals)
(relative to the weedy parent, in nearly every case)
V Performance GM > Performance non-GM
V Transgenic seeds in non-transgenic plants
Question: what factors promote interspecific gene flow?
hybrid vigor
selective benefit of transgene
sexual compatibility
Predict:
Heterosis (F1 > parentals)
Performance GM > Performance non-GM
Transgenic seeds in non-transgenic plants
Allele frequencies in sink populations are a function of:
•initial allele frequencies
•migration rate
•selection
What factors promote gene flow?
• Hybrid vigor
• Sexual compatibility
• Benefit of transgene
• Ecological factors
-Population size and density
- Community structure
- Physical environment
7
-------
How will climate and land
use changes influence the
adventitious presence of
transgenes?
ZONE
ZONE:
ZONE
ZONE
ZONC
ZONE:
ZONE
ZONE
ZONE
ZONE
ZQNE:
SELDW
-60 TO
-40 TO
-30 TO -8
0 -1
Modelling effort
"Breeders have found that, with
rare exceptions, the crops do
not successfully cross-breed
with other plants in the
environment, especially plants
in crop-growing regions."
Martina McGloughlin, Director of
Biotechnology at the University
of California at Davis,
Washington Post, 2000
Hybridization occurs in nature.
Whittenmore and
Schaal 1991
Hybrids can be stable.
I
J-
Reiseberget al. 2003
I MM
uixt
Reduced herbivory and increased seed
set in sunflower (Helianthus annuus)
associated with the presence of Bt
transgene.
Snow etal. 2003
8
-------
At the end of the day...
Population biologists are in an ideal position to
address pressing questions of the effects of global
change on natural/not-so natural populations.
These projects operate at the juncture of policy, economics
and biology.
These issues are an invitation for cooperation and
collaboration across disciplines to provide recommendations
and contributions to basic and applied sciences,
to regulatory agencies, and to producers and developers
in agricultural industries.
Island models of gene flow: equilibrium predictions
Island models of gene flow: equilibrium predictions
favorable alleles:
@ P2*=P,
neutral or unfavorable alleles:
P2*>m
Project design
Brassica napus x Brassicarapa X GM-B. napus
F, hybrid GM-F, hybrid
Split plot design:
Two factors: mating type, herbivory
2006- low herbivory
2007- high herbivory
Response variables:
biomass (g)
seed mass (g)
seed counts
Global change and the cryptic
invasion by transgenes of native
and weedy species
USDACSREESNRI
-------
Delzie Demaree
Malvern, Arkansas
1926
Victor Muehlenbach, St. Louis
Missouri 1969
Three general assumptions in
risk assessment studies
• we understand the problem
• we know what to measure
• we can use available data to predict
the behaviors of novel traits in nature
Gene flow, selection and perverse effects:
Influences of modern
agriculture on the
evolution of
native species
C.L. Sagers
Biological Sciences
University of Arkansas
2006 Biomass - by mating type
CanolaXcanola
6. rapa Xcanola
6. rapa X 6. rapa
B.rapa XGM-canola
GM-canola X GM-canola
Biomass (g)
2006 Biomass - by mating type
CanolaXcanola
6. rapa Xcanola
6. rapa X 6. rapa
B. rapa X GM-canola
GM-canola X GM-canola
40 60
Biomass (g)
2006 Biomass - by mating type
CanolaXcanola
6. rapa Xcanola
6. rapa X 6. rapa
B.rapa XGM-canola
GM-canola X GM-canola
40 6C
Biomass (g)
10
-------
2006 biomass - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa
B. rapa X GM-canola
GM-canola X GM-canola
40 60
Biomass (g)
2007 Biomass - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa I
6. rapa X GM-canola
GM-canola X GM-canola
0 20 40 60
Biomass (g)
2006 Biomass - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa
B. rapa X GM-canola
GM-canola X GM-canola
Biomass (g)
2007 Biomass - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa
6. rapa X GM-canola
GM-canola X GM-canola
0 20 40 60
Biomass (g)
2006 Seed mass data - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa
6. rapa X GM-canola
GM-canola X GM-canola
Seed mass (g)
2006 Seed mass data - by mating type
Canola Xcanola
6. rapa Xcanola
6. rapa X 6. rapa
6. rapa X GM-canola
GM-canola X GM-canola
Seed mass (g)
11
-------
2007 - Seed mass - herbivory by mating type
interaction
CanolaXcanola
6. rapa Xcanola •
6. rapa X 6. rapa
6. rapa X GM-canola
GM-canola X GM-canola
Seed mass (g)
with herbivores
1 without herbivores
2007 - Seed mass - herbivory by mating type
interaction
CanolaXcanola
6. rapa Xcanola
6. rapa X 6. rapa
B.rapa XGM-canola
GM-canola X GM-canola
r
A
1 1 1 with herbivores
^| without herbivores
5 10 15 20 25
Seed mass (g)
2007 - Seed mass - herbivory by mating type
interaction
Canola Xcanola
B.rapa Xcanola
6. rapa X 6. rapa
6. rapa X GM-canola
GM-canola X GM-canola
| Q with herbivores
i | without herbivores
^^_
0 5 10 15 20 25
Seed mass (g)
2007 Seed mass - herbivory by mating type interaction
6. rapa Xcanola
6. rapa X 6. rapa
B. rapa X GM-canola
with herbivores
without herbivores
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Seed mass (g)
2007 Seed mass - herbivory by mating type interaction
6. rapa Xcanola
6. rapa X 6. rapa
6. rapa XGM-canola
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Seed mass (g)
Question: what factors promote interspecific gene flow?
hybrid vigor
selective benefit of transgene
sexual compatibility
12
-------
13
-------
A Multi-scale Approach to the
Forecast of Potential Distributions of
Invasive Plant Species
John Silander
University of Connecticut
Alien Invasive Species in
New England
111 invasive plant species identified in New England:
the vast majority (66%) are native to East Asia or
Of these, the most pervasive are woody invasives that
are native to East Asia.
The majority (61%) of invasion sites are dominated by
18% of all invasive species that are fleshy-fruited and
bird dispersed.
Data from:
www.IPANE.or
A primary objective has been to predict the where
invasive species will potentially spread in the regional
landscape
Euonymusatatus B&rberis itiunbergit Celasttus orbtculaws
Elaeagnusumbellata Alliariapetiolata RossmutiHIara
Our approach to modeling potential
distribution is to use spatially explicit
Hierarchical Bayesian models:
Response variable:
Presence/Absence data
Celastrus orbiculatus
• native of East Asia
• woody Liana
• edge habitats
-------
Incorporate native (Japanese)
Presence/Absence data (releves)
From the Japanese
releve plot
dataset:
presence/absence
data from about
20,000 plots
(PRDB database
from N. Tanaka et
al. 2005)
[Also used Nagano
Flora database]
Celastrus orbiculatus
Climate data layers
New England & Japan
•Max Temperature of Warmest Month
•Min Temperature of Coldest Month
•Annual Precipitation
•Precipitation Seasonality (Coefficient of
Variation)
•Precipitation of Warmest Quarter
The collapsed LULC neighborhood
around field survey sites n km gnd display]
vr loped
U.-.I.III-HI
Foiesls
.. ..
Crops
.A*
Local field survey
site characteristics
Habitat type (collapsed)
- Edge
- Deciduous forest
- Evergreen forest
- Open
- Marine influence
- Closed canopy wetland
- Open canopy wetland
Canopy closure (%)
An ordinal variable
Data from:
Potential distribution ~ /(Climate, Habitat, Canopy, LULC)
Celastrus orbiculatus
Models examined:
1. New England climate only
2. Japanese climate only
3. NE + J climates
4. NE+J climates + local
habitats + LULC
Prediction maps for New England:
probability of species occurrence and uncertainty
Celastrus orbiculatus
OMM
3I-OJ
| j?-08
| ji-Ot
-.
Heroanum specimens
Best model fits include: climate variables from New England &
Japan, LULC, and local site characteristics; predictions
validated by comparisons with independent herbarium records
-------
Re
tt
OJ
~(D
ro
'o
(D
-a
'E
en
ro
^
r n,*ir-
1.26
on
J J5
[00
044
i ,;•
Cred*."L< int^'j
f-t 30 .07?)
(0 10 o ;Sj
OT10.085)
Land-use
history is critical
to predicting the
distribution of
invasive species
in new England
But these are simply static models;
how do invasive plant species spread
across the landscape over time?
Celastrus spread (herbarium records)
-------
Bird dispersal mutualism?
European Starling
(Sturnus vulgaris)
nvasive fruits
that attract
birds
Invasive bird introduced
from Europe to New York
1896
Mutual spread across the region?
Estimate decadal starling
spread across New England
1920-2009, based on
Christmas Bird Count data
New York
1896
Joint spread of starlings & Celastrus?
Celastrus progression Starling progression
-
Feeding choice behavior
Starlings prefer
invasive fruits
e.'i r-uropean Sailing; I ] invasive*
(b)Amencan Robins I I invasives
s* X
Starling movement of invasive fruits
i;
Ingested seed
pass-through
times
Local seed dispersal
Movement of ingested
seeds during foraging
Cumulative
distribution of
flight movements
Long distance dispersal
data on over 24,000 banded starling recaptures
Frequency distribution
starling recaptures > 10
kms over 1 day
) 20 50 100 200 300 500 1000 3000 M
kilometers between recaptures
Upto200+km/day
frequency distribution
starling recaptures >10 kms
over 1 year
20 50 100 200 300 500 1000 3000 K
kilometers between recaptures
-------
Develop a Cellular Automaton model of
the dispersal and growth of Celastrus across
the New England region
• Grid of cells = LULC across the region
(5x5 min cells ~8x8kms) -6500 cells.
• Set of population asymptotic growth (A)
rules for Celastrus based on LULC specific
demographic responses.
• Local dispersal kernel for Celastrus linking
cells on grid (based on starling data)
• Long distance dispersal rule
LULC CA grid
Collapse LULC categories by majority rule
e.g. 25 5x5 min grid cells
Model grid
-6500 cells
Dispersal kernel (local)
LULC grid (5x5 minute ~ 8x 8km)
Spread Prediction Evaluations
Using independent data: herbarium
specimens (& field data)
99% spread
envelopes
Model display and evaluation
1940
o = herbarium record accounted
for by model
x = herbarium record not
accounted for by model
CA Model predictions for the spread of oriental bittersweet.
summary of 200 replicate runs:
- proportion of model runs predicting presence (0 - 1).
- if 80+% model runs predict presence, then if herbarium
record is present at that time, scored as correct prediction "o"; if
<80%, scored as incorrect prediction "x"
- example here is 50% correct.
Seeding the model
First sites in herbarium record:
New Haven, CT & Falmouth, MA;
a few years later, Durham, NH
These were used as seed points fo
the model runs
LULC
developed
agriculture
Deciduous
forest
Conifer/mix
forest
water
Starling
use*
30%
37%
18%
3%
12%
Proportion of
landscape
4%
7%
27%
51%
11%
Growth
rate (A)
1.5
1.5
1.3
0.95
0
* Dispersal kernel is adjusted by starling
behavioral use of the landscape.
Exponential local dispersal kernel with
mean = .5 (rate =2)
1 long distance dispersal per generation
moving drawn from a uniform
distribution on [3,20]
-------
Light >
Soil
moisture
Celastrus performance
(growth, survival,
reproduction) across
transplant sites
Predictions by 2009 with local or
long distance dispersal only
Local dispersal only long distance dispersal only both
Predicted spread 1920-2009
Herbarium record validations
1040 • I960 19HO
[IPANE field
data validations]
Future spread
and years
into the
future...
The
landscape
does not fill
Model parameter sensitivity
Increase long
distance
dispersal events
increase or decrease growth rates
A+20% /^| A-10% /"~^\ A-2'
Model simplifications?
Binary landscapes
Binary Landscape
Slightly poorer spread & performance than the
more fully specified landscape heterogeneity
-------
Uniform landscapes?
n/w~v
;;:.nfi
The landscape fills overtime
Random landscapes?
Celastrus fails to spread in the landscape
Joint spread of starlings & Celastrus?
Joint sequence
3
Starlings
Celastrus-30-40 yr lag
Summary
The most pervasive invasive plant species in New
England tend to be woody and with bird dispersed fruits.
HB models provide accurate, static predictions of the
potential distributions of species using climate, land-use,
and local site traits as explanatory variables. Native
range data together with invaded range data are critical
to accurate predictions.
CA models, calibrated from invasive plant demographic
data (Celastrus) and starling movement, yield predictions
that agree with the observed spread of invasives over
space and time.
Regional land-use patterns are critical to the patterns of
spread of both invasive plants and starlings.
-------
Predicting relative risk of invasion by saltcedar
and mud snails in river networks under different
scenarios of climate change and dam
operations in the western United States
Climate change likely to enhance spread of
invasives in river ecosystems ... but how?
Eurasian saltcedar (Tamarix)
New Zealand mud snail
(Potamopygus antipodarum)
LeRoy Poff
Brian Bledsoe
Denis Dean
Jonathan Friedman
Greg Auble
Pat Shafroth
David Purkey
David Merritt David Raff
• Alter ecosystem structure and function
• Contribute to native species declines
• Economic damage
Workina Hvoothesis
Within thermally suitable envelope ...local invasion
success will be dictated by habitat suitability and dynamics
(hydrologic, geomorphic) and biotic factors, which can be
modeled at the ecologically relevant scales.
Account for human responses to climate change, which will
contribute to risk of invasion.
Scaling the problem:
GCM -> Hydrologic Mode
;ubcatchments)-» Ecological Respons
Framework:
Hydrogeomorphic Template
Species population success is a function of magnitude,
frequency, timing, and duration of flow events that limit
establishment success or cause mortality.
Effectiveness of flow regime varies with geomorphic
settings (e.g., canyon vs. alluvial river reaches).
Plan: Combine flow regime and geomorphic setting (=
natural disturbance regime) to explain current
distribution of salt cedar and mudsnail and to project
future likelihood (risk) of invasion.
Generalize to disturbance-sensitive species that vary in
flow-sensitive species traits.
-------
Develop empirical "mechanistic" ecological response models
to explain the current distribution and dominance
(probability of occurrence) of two invasive species across
the interior West at the stream segment scale.
Hypotheses:
The current distribution and abundance ofsaltcedar and
NZMS can be explained statistically in terms of site-scale
hydrogeomorphic setting and dynamics.
Probability of species occurrence or dominance at a site will
reflect a hydrogeomorphic threshold.
Study Region
Use GCM output to identify "thermal envelopes" where
minimum winter temperatures will warm to > -30° C and
thus promote salt cedar range expansion. Overlay with
areas on edge of current salt cedar range to identify study
Region of conservation and
management concern
Goal 2
For a geographic region in the western US, use downscaled
projections of regional climate change to predict future
streamflow regimes, and incorporate the effect of water
management on those future flow regimes.
pothesis: The WEAR modeling
platform can be used to generate
sub-basin scale, weekly flow
regimes at a spatial grain of ca.
100s of km2 and can be used to
infer the effects of dam operations
on natural flow regimes for
subbasins in the region. [More
below]
WEAR - Water Evaluation and Planning Program
(http://weap21.org)
Rainfall-Runoff Model based on spatially distributed land
use/land cover types and climatic inputs to catchment;
operational rules of water management infrastructure are
incorporated to generate hydrographs throughout network.
Green River Basin, WY
Selection of "pour points"
•USGS stream flow gauges
•major management points
(dams, major ditch diversions)
•Upstream and downstream of
junctions
Catchment Objects
-landcover, soils, etc.
-Climate data (ppt,
temp, wind)
-------
Goal 3
Disaggregate (as necessary) the subbasin-scale flow
regime output from the WEAR model and construct a
reach scale flow regimes for the drainage network in
the entire region. Create geomorphic basemap (OEMs
at reach scale) and overlay hydrology (natural plus
management).
Hypothesis: An artificial neural net (ANN) model can be
constructed to predict streamflow at the river reach scale
based on subbasin-scale hydrologic output from the
WEAR model, on GIS landscape variables, on projected
climate data, and on river network structure.
Goal 4
Use the "mechanistic" ecological response model to examine
the "risk of invasion" for river reaches throughout the
region for different combinations of climate change
scenarios and modes of dam operations.
(empirical models vs. process-based ecological models)
Hypothesis: A reach-scale geomorphic base map can be
combined with projected reach-scale streamflow
regimes to project relative risk (probability of occurrence or
dominance) for the two species under various realizations
of future runoff and streamflow.
GoalS
Model long-term invasion success for the two species
under interannual flow regimes across a range of
hydrogeomorphic settings.
Hypothesis: Stochastic population dynamics models can
estimate year-to-year population sizes based on reach
geomorphology and long-term (projected) flow regime
(including dam operations) and thus assess long-term
viability of non-natives.
-------
bxpectea uutcomes
(what we're doing, what we're not, and what we
might)
1. More mechanistic (dynamic) and appropriately scaled
basis for projecting invasion risk.
. Risk map - decision support system given high
uncertainties in multiple, linked models. (Not precise
point predictions)
3. Framework for thinking about the spatial distribution of
threats and how to contemplate proactive
management. (Not make precise predictions)
• General application: Network scale model that enables
questions of managing for ecological resilience or
conservation planning in metapopulation context?
4. Future inclusion of social processes to examine cost-
benefits of spatially-distributed water mangement?
Interaction with Clients
Discussions with BuRec (upper
Colorado)
Discussions with The Nature
Conservancy (threats assessment of
CRB)
Planned discussions with Wyoming and
Colorado state agencies
Challenges / Lessons (being)
learned
1. Projecting ecological response models for
salt cedar and NZMS that can be applied to
future (novel?) environmental conditions?
2. Scaling climate and hydrologic models to
match ecological response/measurement
scale.
• Is weekly hydrograph good enough?
3. Representing "risk" in a robust way that
allows for linked multi-model uncertainties.
• Quantitative models ... Qualitative interpretation
IV. Outcomes (to date)
Developing a WEAR model for the upper
Green River and Yampa River basins that
can eventually be used to address a number
of water management issues in the Green
River and Yampa basins.
Generating interest among NGOs, states, and
feds.
-------
Integrating future climate change and
riparian land-use to forecast the effects
of stream warming on species invasions
and their impacts on native salmonids
Julian D. Olden
School of Aquatic & Fishery Sciences
W UNIVERSITY OF
WASHINGTON
Research Team
Joshua J.
Lawler
College of
Forest
Resources,
University of
Washington
Christian E.
Torgersen
Forest and
Rangeland
Ecosystem
Science
Center, USGS
Timothy J.
Beechie
Northwest
Fisheries
Science
Center, NOAA
Fisheries
David
Lawrence
School of
Fisheries,
University of
Washington
Aaron
Ruesch
College of
Forest
Resources,
University of
Washington
Challenge Synopsis
The prospect of dramatic climate change over the next century
underscores the need for innovative science and new decision-
support tools for efficiently managing freshwater ecosystems
Climate-induced changes in the geomorphic and physical
processes that drive stream ecosystems in the PNW are
imminent, including
- warmer temperatures (2.3-2.9°C)
- lower accumulation of winter snowpack (-44%)
- earlier onset of spring flows (4-6 weeks)
- lower summer baseflows (-10-35%)
Cumulative effects and complex interactions among multiple
agents of environmental change may limit the success of
current and future river management efforts
Rivers in hot water
• Climate changes will have direct implications for stream
temperatures, which are only exacerbated by the removal or
alteration of riparian habitat by logging and grazing that reduces
shading and modifies channel morphology
• Elevated stream temperature is one of the most pervasive water
quality issues threatening freshwater ecosystems in the PNW
1990 - damaged by livestock
overgrazing
2003 - after livestock were
removed and vegetation recovered
Management efforts are further complicated by the fact that
Pacific salmon (Oncorhynchus spp.) now share the riverine
landscape with a number of non-native fish species
Significant shifts in species ranges and the outcome of biological
interactions are highly possible
,
if.
CAT CRA LMB SUB WAL YEP
:
LMB SVB CAT WAL
ierson iaj(2l»9)
Project Goals
What are the predicted effects of regional climate change and
local riparian management on riverine thermal regimes?
How will Chinook salmon, smallmouth bass and northern
pikeminnow respond to projected temperature changes?
What are critical areas for
riparian restoration and
protection to mitigate the
negative ecological impacts
of climate-induced stream
warming in the future?
-------
Ecological Setting
Land use and resource
extraction vary longitudinally
Unregulated
One of the few remaining
wild spring Chinook salmon
runs in the Columbia River
Basin
Active region of upstream
invasion by smallmouth bass
and northern pikeminnow
Research Elements
1. Develop climate-change projections of
temperature and precipitation
The Columbia River Basin is
predicted to show consistent
average increases in air
temperature, higher winter-spring
Q, lower summer Q, and earlier
timing of spring peak events
We are downscaling simulating
future climate data from a suite of
GCMs under three green-house
gas emissions scenarios (Bl, A1B,
A2) for decadal time periods
(2020-2100)
Projections for the John Day Basin
5
Average Precipitation Change
2. Characterize channel
geomorphology and riparian land cover
Thermal sensitivity of stream reaches
to climate warming varies with
geomorphic setting and degree of
channel incision
Stream reaches will be classified
according to drainage characteristics,
lithology, and field measured and
modeled channel incision
LandSatTM imagery will be used to
quantify riparian land cover
Non-incised
3. Quantify multi-scaled thermal regimes
Thermal regimes will be quantified
using a network of digital temperature
loggers at point locations
Forward looking infrared (FLIR) thermal
imagery will be used to map spatially
continuous longitudinal patterns of
stream temperature
-------
4. Develop a spatially-explicit stream
temperature model
• Stream temperature is dependent on both heat load and stream
discharge
Heat Source (v. 7) allows for the
simulation of water temperature at the
reach scale using high resolution
spatially continuous data, coupled with
deterministic modeling of hydrologic
and landscape processes
It includes important processes:
- mass transfers from tributaries
- groundwater inflows
- landscape thermal radiation
- adiabatic cooling
- robust radiation modeling
- etc ...
.oyd and Kasper (2003;
5. Forecast thermal regimes under
scenarios of climate change and land
use management
• Future spatiotemporal patterns in stream temperature will be
predicted according to scenarios of projected climate change and
riparian land-use
1 1 '
Future climate
Future vegetation
Restored vegetation
Restored tributaries
No PODS
Ecological targets
Scenarios of projected temperature and hydrology
Scenarios of projected land development
Complete restoration to estimated potential vegetation
John Day Fish Habitat Enhancement Program
Conservation and acquisition priorities (TNC, TFT)
Tributaries flow and temperature set to estimated potential
No points of diversion
Scenarios targeting specific ecological outcomes
What it might look like ...
6 & 7. Model ecological responses to
future thermal regimes
• Fish species responses to climate change and riparian management
will be estimated according to psychological preferences and
tolerances
• A number of additional key temperature benchmarks will be explored.
Chinook sahnol
g/Alevin Juvenile
Mi^rafco
Holding
Spawning
Northern
pikeminnow
Low
Low
Low
Pref
Upp
Upp
Upp
r lethal limit
r tolerance limit
r growth limit
rred temperature 6.
r growth limit
r tolerance limit
r lethal limit
.7
.0
.5
10.0
2.8
W
8.9
0.8
4.5
10.0
12.0-13.3
15.6
19.1
25.1
0.8
3.3
7.2-14.5
21.0
22.0
10.0
12.3
20.2
25.0-26.0
27.0
29.5
36.9
8.1
10.1
16.1
<;• j .2.1.8
24.4
26.0
38.0
Statistical models linking species occurrence, abundance and
spawning activities (SMB) to projected changes in thermal and
hydrologic regimes will be developed
We will explore bioenergetic models and age-structured models that
account for environmental change, fish population dynamics, and
harvest rates
Peterson and Kwak (1999)
-------
Field surveys
• Continuous stream segments over 100 km of the Middle and North
Forks will be systematically surveyed to map and obtain counts of
fish and spawning nests in June, August and October
Prioritizing riparian management
in a changing climate
• Our findings will help guide management strategies and policy
aimed at minimizing the future range expansion of invasive
species through protection (i.e., conservation easements) and
restoration (i.e., riparian fencing) of riparian vegetation that
creates and maintains coolwater habitat.
Results from this project will make it possible to rank stream
segments in terms of their ability to:
- mediate the effects of climate change on stream temperatures
- create suitable thermal habitat that favors native species over invasive species
- establish thermal barriers to prevent upstream invasion
Management portfolios (based on different ecological endpoints)
will be distributed to local and regional agencies and NGOs
Outcomes
Products from our research project will be integrated into a Graphical
User Interface providing the user with animated maps and timelines of
stream temperature change, salmon habitat availability, and bass and
pikeminnow spread for a given climate change or land use scenario, or
the option to export data for quantitative analysis
Outcomes
Products from our research project will be integrated into a Graphical
User Interface providing the user with animated maps and timelines of
stream temperature change, salmon habitat availability, and bass and
pikeminnow spread for a given climate change or land use scenario, or
the option to export data for quantitative analysis
Challenges
Social
• Continuous land access
Scientific
• Incorporating climate-induced vegetation change into stream
temperature modeling
Management
• Preparing managers for the possibility of implementing
unconventional strategies
Interactions with clients
WASTED
DOLLAR
STUPID
-------
Interactions with clients
Th<-Naturc I
Conservancy
Freshwater Trusf
Thank you
-------
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
Eric Seabloom
Sally Hacker
Peter Ruggiero
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
1. System Overview
2. Project Goals
3. Outcomes
4. Lessons learned/Challenges
5. Interaction with clients
History of dune grass invasions on the Pacific coast
Prior to 1900, beaches and dunes were sparsely vegetated,
little grass, shifting sand
Coastal dune grass invasions along the Pacific coast
European beach grass, Ammophila
arenaria (L.)
American beach grass, Ammophila
breviligulata Fern.
Colonized coastal dunes and beaches
for nearly 110 years
40% of west coast shoreline consists of
sandy beach or dune
History of dune grass invasions on Pacific coast
Pre-1900: Native beach grass, Elymus mollis
~1900: Widely introduced European beach grass, Ammophila arenaria
By 1950s: A. arenaria present along entire west coast, Canada to Mexico
1935: Introduced American beach grass, Ammophila breviligulata, to Columbia River
Ammophila breviligulata
Dune grass formed foredunes
-------
Consequences of Foredune
• Increases coastal protection from waves, wind, and possible tsunamis
• Increases land stabilization for development behind the foredune
Foredunes:
Coastal Protection
Cape Hatteras, before and after Hurricane Ivan, 2003
Unintended consequences
Redistribution of sand
• overall open dune habitat decreases;
"sand starved"
• dynamic nature of shifting sand gone
Decline in some species of native plants
and animals
• Six federally listed endangered plants
• federally listed threatened Western
snowy plover
Increase invasion of other species
• Scotch broom, gorse, weeds, etc.
Before
Climate change and coastal Dunes
•Climate change is accelerating sea level rise and increasing storm
intensities (IPCC 2007, Websteretal. 2005)
•Over the past 5 years, annual losses due to hurricanes averaged
$35.8 billion, a 3-fold increase over the early 1990's (MSB 2006)
•Coastal dunes comprise 40% of the Oregon and Washington coasts
(Cooper 1958, Komar 1997).
I
"?• - ; s
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
1. System Overview
2. Project Goals
3. Outcomes
4. Lessons learned/Challenges
5. Interaction with clients
2
-------
Project Goals
1 . Determine effects of climate change on beach grass invasion
2. Determine effects of beach grass invasion on the ability of
dunes to mediate risk of climate change (G -^ E)
3. Determine effects of exotic grass management on the ability
of dunes to mediate risk of climate change (H -> E)
Project Goals
1 . Determine effects of climate change on beach grass invasion
Simulation models to estimate a range of likely sediment budgets under
expected climate change regimes (A -> D)
Field experiments to determine the outcome of invasions under predicted
sediment budgets (D -> G)
I
^* Flo^
Project Goals
2. Determine effects of beach grass invasion on the ability of
dunes to mediate risk of climate change (G -> E)
Field surveys and lidar to determine effects of species invasion on dune
morphology (G -> F)
Simulations modeling to determine the effects of dune morphology on risk
under various climate change scenarios (F -> E)
Project Goals
3. Determine effects of exotic grass management on the ability of
dunes to mediate risk of climate change (H -> E)
•Field surveys and lidar to determine effects of conservation management on
species invasion (H->G; Fig. 1) and dune morphology (H->F)
<*^ ^
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
1. System Overview
2. Project Goals
3. Outcomes
4. Lessons learned/Challenges
5. Interaction with clients
Preliminary Project Outcomes
Dune Bag Experiments at Hatfield Marine Science Center, Newport, OR
Phoebe Zarnetske PhD research
I
Species interactions x sand supply
3 sand addition levels: none, low, high
3
-------
Preliminary Project Outcomes
Sand deposition can alter competitive
interaction among native and exotic
dune grasses
i Med High
Sand Deposition
Preliminary Project Outcomes
Dunes dominated by the secondary invader (A.
breviligulata) are 40% lower than those dominated by
current invader (A. arenaria)
•
\
Preliminary Project Outcomes
A. Breviligulata continues to spread and increase in
dominance - a continued decrease in dune height
Change in dominance over 20 years
!i
Washington
Oregon
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
1. System Overview
2. Project Goals
3. Outcomes
4. Lessons learned/Challenges
= 5. Interaction with clients
Project Challenges
£300
ran
nmn
!,._,
Lidarwork conducted by Jeremy Mull
(MS thesis) with Peter Ruggiero
• Surveyed Washington on 9/18/2002.
• Surveyed Oregon on 9/20/2002 and
9/21/2002.
• 532 nm green laser used.
• Pulsed at 5,000 Hz.
Project Challenges
= shoreline determined from LIDAR survey
-------
Beach grass invasions and coastal flood
protection: forecasting the effects of climate
change on coastal vulnerability
1. System Overview
2. Project Goals
3. Outcomes
4. Lessons learned/Challenges
= 5. Interaction with clients
Interactions with clients
Quantitative Assessment of Flooding
Risk on Long Beach Peninsula, WA
In cooperation with City of Long Beach, Coastal
Communities of Southwest Washington,
Washington Department of Ecology, Oregon
Department of Geology and Minerals
2011 Meeting with Land Managers and
Researchers
Conduit to provided information to individual land
managers, scientists, and policy makers who are making
the critical decisions about invasive species
management.
Based on highly successful 2008 PNW Dunes Workshop
we conducted using funds from Oregon Seagrant
About 40 participants from 15 Federal, State, Local
Agencies and NGO's
Coastal Vulnerability
Assessment
Voigt, Ruggiero, Kaminsky. 2000
Current conclusions
1. It is likely that changing sediment loads resulting from climate
change will alter the composition of the dunegrass community
2. A rapidly spreading invasive dune grass is likely lowering
dune heights and reducing their ability to protect coastal
communities
3. Exotic grass management will require careful balancing to
preserve endangered species and coastal protection function
Thanks:
Collaborators:
Graduate Students:
Phoebe Zarnetske
Jeremy Mull
Field & Lab Assistants:
Vince & Autumn Adams
Lindsay Fitzgerald
my He
Funding:
EPA
NOAA
Oregon Sea Grant Program
HU!
ilbrahin
Thatcher Jone
Travis Lewis
Micah Rogers
John Schaefer
Chris Soto
Dave Worth
JayZametske
-------
Elevated Temperature and Land Use
Flood Frequency Alteration Effects on
Rates of Invasive and Native Species
Interactions in Freshwater Floodplain
Wetlands
Curtis J. Richardson, Neal Flanagan,
Mengchi & Song Qian
Duke University Wetland Center
Nicholas School of the Environment,
Duke University
Global climate change and freshwater
ecosystem studies & models suggest two
key findings:
1. water temperatures will increase (2 to 4° C)
(IPCC2007)
2. the frequency and intensity of high flow
stream events will increase
what are the implications of warmer
water and altered hydroperiod on the
establishment, abundance, and
distribution of invasive species in river
floodplain ecosystems?
Likely Future Scenario
Southeastern stream ecosystems will
experience
1. lower baseflows with more extended
drought periods punctuated by
2. more frequent and more intense storm
events.
Likely Future Scenario
Southeastern freshwater
wetlands;
1. will be inundated for less of each
year than currently, and
2. will undergo a greater number of
rapid wetting and drying cycles as
a result of extreme events.
Project Goals & Study Questions
Quantify effects of elevated wetland water temperature
and pulsed water on
- rates of species invasion
- patterns of sediment and nutrient retention services?
Assess how species-richness, diversity, productivity &
invasibility change under varying water temp regimes?
Determine have interactions between hydrology and
temperature affected the current community
composition/invasibility of SE floodplain ecosystems at
the regional scale?
-------
Experimental Levels
Experimental Level 1
- role of plant diversity on invasive species
- pulsed water effects on wetland species
- elevated temperature and pulsed water in
controlled (experimental) wetlands
Experimental Level 2
- regional floodplain hydrology and temperature
shifts in naturally occurring wetlands
Duke University Sandy Creek Watershed
d-17B7 hrt
O Unl«.n»i«y - ;ut n*
r 1 WaHcMifl - 24 tia
99 Research Diversity Plots: The role of hydroperiod shifts & water
pulses on diversity & wetland functions-denitrification, P cyclin
Hydroperiod and pulsed water experiments
-------
Experimental Levels
Experimental Level 1
- role of plant diversity on invasive s
- elevated temperature and pulsed water in
controlled (experimental) wetlands
Experimental Level 2
- regional floodplain hydrology and temperature
shifts in naturally occurring wetlands
-------
Experimental Level 2
(Regional Scale)
We identified nine flood plains sites
located on rivers throughout the North
Carolina and southern Virginia.
Wetlands studies downstream of:
- 3 surface (warm water)
— 3 bottom-releasing dams (cool water)
- 3 undammed reference watersheds
Siting Criteria
Temperature regime
Located within the Piedmont Ecoregion
Headwaters in mountains
High degree of hydrologic connectivity,
- Frequently flooded
Similar nutrient regimes
Reference sites have no upstream dams
- small ponds only
-------
Site Layout
Temperature Off Temperature
Near
Treatment Average Near Shore
Soil Temperature
Is-
3.
Total Phosphorus
3 5 10 15
f
09/23/08 10/02/08
--o-- Reference
-•- Warm
-••••- Cold
I
4 I
-------
Smith Mountain Lake - Cold
----- Temperature Probe Datum (off Shi
6
-------
Lake Gaston - Warm
Johns River - Reference
Temperatuit; Probe Datum fc" Shor
Average Average
Duration Return Period Depth
Frequency (hrs) (hrs) max/mean
Preliminary Year one
Results
-------
Shannon-Wiener Diversity Index
n Near Shore
D Off Shore
Warm Sles
Referenci
Cold Sles
Pielou's Evenness Index
D Near Shore
D Off Shore
Species Richness
D Near Shore
D Off Shore
Warm Sles
Referenci
Cold Sles
No. Invasive Species
ONearShi
DOffShoi
~
8
-------
% Invasive Species
D Near Shore
D Off Shore
Lesson Learned/Challenges
What is meant by invasive species definition varies greatly
Plant community respond to both temperature and pulsed water
events can be detected
Threshold responses to disturbance may vary by season
lifficult to separate out pulsed water effects from temperature
effects at regional scale
Mesocosm scale studies will allow for more temperature and wa
control to help in effect studies, but are difficult to set up and
maintain & scaling is an issue
Separating out environmental disturbance from climate change
effects is difficult and will require new approaches to threshold
nalysis to augment Bayesian threshold analysis.
Interaction with Clients: A Broad
Interest
International interest-collaborations with Peking University, Potsdam
Institute for Climate Impact Research, Finnish Environmental Institute,
University of Liverpool, University of Utrecht, and Eurolimpacs Climate
Change Program.
Presentations- Ecological Society of America, Society of Wetland Scientists,
numerous academic institutions and presentations to state government
officials and review panels.
Information requests & Coop with government agencies (USGS, USAGE,
South Carolina Sea Grant, North Carolina Wildlife Commission, and
Creation of Duke Wetland web site with research project findings & reports,
Popular Wetland Wire (www.env.duke.edu/wetland)
9
-------
-------
Mean daily minimum temp (Jan-March) °C
Timing of recruitment in relation to inter-annual variations in seawater
temperatures: non-native vs resident species
INVADERS
Diplosoma Botrylloides Ascidiella
March water temperature (°C)
Correlation between resident species richness and the
fraction of exotic species in areas of different
coastal land use
u '
U) Q)
• = Primarily rural
• = Primarily residential
= Primarily industrial
3 4 5 6 7 8 9 10 11 12 13 14 15 16
Number of Resident Species
-------
Goal #1 (cont)
Goal #2: Conduct mesocosm experiments examining the interactions of climate
change (temperature increase) and land-use (nutrients) and the interactions
between them in altering the ability of invasive species to Influence native
Aquakicfs host Molly I
yours truly
New London Day
Featured segment on AquaKids
Episode 18
- aired in Connecticut 24 Jan 2009
— aired nationally - week of 19 Jan
12- 3 m diameter tanks
3 habitat types: eelgrass, rocky, unvegetated
Treatments: temp (1-2° C above ambient), nutrients (0.5-1.Ox above
Ambient), temperature x nutrient increases
Native and non-native species added - response variables: growth, mortality,
competitive interactions, predation
YeaM: Conducted small-scale pilot
study (4 tanks) to establish experimental/
monitoring protocols - full experiments
will be conducted in Year 2 and Year 3
Goal #3. Conduct field experiments to assess temporal and spatial
scales of potential efforts needed to manage invasive species -
to be conducted in Years 2 and 3
Given the 'openness' of marine systems (aka larval transport),
Attempts to control invasive species most likely will occur at
Local scales
Removal experiments at different spatial scales
Seasonal experiments
Effects of variations in landuse
Goal #4 - Survivorship of predators and
their effectiveness in controlling
invasive species in different land use
conditions
Effects of macro-predators (seastars,
fish, crabs) feeding on juvenile and
adult ascidian life stages (7 day expt.)
Solitary forms
10-20 25-35 10-20 25-35
Colony Size (min. linear dimension in mm)
Life Stage
Goal #4. Conduct field experiments to examine
survival of key predators on invasive species and
how it varies with land-use.
Effects of predation micro predators (snails, small crabs):
Percent mortality of ascidian post-recruits (7 Day experiments
using different colony sizes or different ages of native and non-
native ascidians
Colonial forms
Solitary forms
h.
Colony Size (mm)
-------
xampie or
Larval sources
Source Area: A seagrass bed in a channel
Habitat + Hydrodynamics -> Larval
Distribution
urce: Seagrass bed in a Channel Region
-------
Lessons Learned
Very important to include input by managers/stakeholders in the early stages
of the project
Managers often are dealing with the most current 'brush fire' and often are in a
rapid response mode (a reality lesson, but also a challenge to the scientist)
Critical importance of long-term environmental data bases and associated
population/community data
Challenges:
The most recent sea squirt alien,
Didemnum vexillum, in eastern L
Island Sound (-25 m depth) - pot
new stable point
Concerns about "new" stressors -
coastal acidification, power infra-
structure disturbances - and how they
interact with climate change and land
use patterns
-------
Some fundamental goals and objectives of SC2 are to:
•Foster collaborative, interdisciplinary research to understand
and quantify the diverse impacts of climate change on the Skagit basin
•Serve as an objective and non-politically affiliated sourc
scientific information, data, and services to support long-term planning
and climate change adaptation by stakeholders in the basin
•Identify new scientific information, data, or services that are
needed to address climate change impacts in the basin, and generate
research funding to address these needs
•Establish and maintain long-term relationships between
scientists and stakeholders in the basin in the interest of generating trust,
fostering effective collaboration and sharing of information.
•Develop and maintain a web-based "clearing house" for
scientific products and services addressing climate change impacts
and adaptation in the basin.
•City of Anacortes, WA
•CSES Climate Impacts Group (UW)
•Lawrence Livermore National Labs
•Montlake Fisheries Science Center (NOAA)
•North Cascades National Park
•Pacific Northwest National Labs
•Seattle City Light
•Skagit River System Cooperative
•Swinomish Tribe
•University of Washington
•USGS
•Western Washington University
-------
Changes in Glaciers
2006
Recession of Whitechuck Glacier
(Sauk Headwaters)
Photos courtesy of Dr. Mauri Pelto, Nichols College
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
U.S. Environmental Protection Agency
Office of Research and Development
The Plight of Ecosystems in a Changing Climate: Impacts on Services,
Interactions, and Responses Workshop
Plymouth Church
1217 Sixth Avenue
Seattle, WA
May 27-28, 2009
MEETING SUMMARY
INTRODUCTION AND OVERVIEW
The U.S. Environmental Protection Agency (EPA) Office of Research and Development's (ORD) "The
Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses"
Workshop was held on May 27-28, 2009, in Seattle, Washington. The workshop brought together
researchers from academia, private industry, regulatory agencies, and government to discuss ongoing and
potential research on climate change and its effects on the environment, including ecosystem services.
The meeting also served as a stimulus for increased collaborations among the various researchers and
agencies. Approximately 88 individuals attended, and there were 49 people who called into the con-
ference over the 2-day period.
DAY1: MAY 27, 2009
Introductory Remarks
Brandon Jones, EPA, ORD, National Center for Environmental Research, and Roseanne
Lorenzana, EPA, Region 10
Dr. Brandon Jones thanked the participants for attending and Dr. Roseanne Lorenzana of Region 10 for
her help in organizing the meeting. He noted that EPA's new Administrator has placed a focus on
ecosystems and introduced Dr. Lorenzana.
Dr. Lorenzana welcomed the participants to Seattle and explained some of the logistics of the meeting.
She noted that Region 10 is comprised of Washington State, Oregon, Idaho, and Alaska and is particu-
larly interested in climate change and its effects. One regional project involves the Skagit Watershed, one
of the largest watersheds in Washington State. It is very important to the Puget Sound as it provides the
system with 30 percent of its freshwater. The three EPA climate-related grants in the area total
approximately $2.3 million, with an additional $800,000 provided by other grants or investments. She
noted that collaborative research is important and invited Dr. Alan Hamlet of the University of
Washington to say a few words about the Skagit Climate Science Consortium.
Dr. Hamlet acknowledged the instrumental efforts of Mr. Larry Wasserman, Swinomish Indian Tribal
Community, in creating the partnership of the Skagit Climate Science Consortium. The group is an
extension of the regional-scale planning occurring in the area with a focus on the Skagit River Basin,
which is located in the North Cascades region of Washington State and has important influence in the
area as it provides the largest freshwater drainage into the Puget Sound. The lower part of the basin has
the largest human use, with extensive farmlands and a growing number of towns. The upper basin is fairly
pristine, with the exception of several large hydroelectric projects. The basin provides a unique oppor-
The Office of Research and Development's National Center for Environmental Research
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
tunity to explore the science of climate change in the context of ecosystem services and human
development because of its still-functioning ecosystem and significant human use; it is a good area in
which to explore how to maintain a balance between human use and functioning ecosystems.
Research has identified several climate change impact pathways in the region. The fundamental goals of
the Skagit Climate Science Consortium are to understand the diverse impacts on climate change and assist
long-term planning and adaptation in the basin. The specific goals are to: (1) foster collaborative and
interdisciplinary research to understand and quantify the diverse effects of climate change on the basin;
(2) serve as an objective, nonpolitically affiliated source of scientific information; (3) identify new
science and specific information, data, and/or services that are needed and fund them; (4) establish and
maintain long-term relationships between stakeholders and scientific organizations and practitioners; and
(5) develop and maintain a Web-based clearinghouse of scientific information that can be accessed by
stakeholders and provide the inputs that are needed for long-term planning and adaptation. The
Consortium has many affiliations and partnerships, and more are expected to be added over time.
Currently, the group is actively planning to fund three projects. The first project will foster understanding
of hydrologic extremes, the second will explore changes in glaciers and the sediment regime, and the third
will examine changes in ecosystem functions from the headwaters to the basin.
TIER I: EFFECTS OF CLIMATE CHANGE ON ECOSYSTEM SERVICES PROVIDED BY CORAL
REEFS AND TIDAL WETLANDS
Effect of Sea Level Rise and Climate Variability on Ecosystem Services of Tidal Marshes
Chris Craft, Indiana University Bloomington
Dr. Chris Craft explained that there are large areas of tidal marshes on the East Coast from North Carolina
to Florida; closer to the ocean, these marshes are salt marshes. The salinity of these marshes fluctuates
based on tidal inundations. As the salinity decreases in the marshes further inland, species diversity
increases. The marsh scale developed by William E. Odum describes regulation, habitat, and productivity
functions. Among other things, salt marshes provide shoreline protection.
The main effect of climate change is rising sea levels, which cause erosion and saltwater intrusion.
Therefore, the project goal is to develop a conceptual model that describes how tidal marsh ecosystem
services vary along the salinity gradient and a simulation model of how sea level rise and climate
variability will affect the delivery of ecosystem services. The project is based on three explicit
hypotheses: (1) Rising sea level leads to inundation and loss of tidal marshes, especially tidal freshwater
marshes and their ecosystem services. (2) Diking protects freshwater marshes against rising sea levels, but
when marshes are diked, ecosystem services associated with connectivity are lost. (3) Greater interannual
variability of climate leads to greater frequency of drought and reduction in ecosystem services in drought
years; greater variability in rainfall leads to increased delivery of ecosystem services in wet years.
The researchers are examining how accelerated sea level rise will affect the area and spatial distribution
of tidal marshes and their delivery of ecosystem services. The research is based on wetland habitats,
particularly reduced salt and brackish marsh habitat and the near complete loss of tidal freshwater marsh,
and ecosystem services, particularly reduced regulation functions (e.g., nitrogen and phosphorus
retention, denitrification) and reduced production functions (e.g., plant productivity). The study region
runs along the East Coast from the border between North Carolina and South Carolina to the St. Mary's
River, which forms the border between Georgia and Florida. The focus includes the measurement of
ecosystem services in three marsh types (tidal freshwater, tidal brackish, and tidal salt) near three river
estuaries (Ogeechee, Altamaha, and Satilla) in coastal Georgia.
An example of ecosystem services measurement is nitrogen accumulation in soil; the researchers
calculated rates of nitrogen accumulation in each marsh type and in each area, and the results indicated
The Office of Research and Development's National Center for Environmental Research 2
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
that nitrogen accumulation increases with a decrease in salinity and that potential denitrification increases
in freshwater systems. The researchers scaled up to annual rates to compare the types of data and found
that ecosystem services vary across the marsh gradient. Freshwater systems provide a higher level of
ecosystem services with respect to nitrogen cycling. Research also indicated that brackish marshes have
increased aboveground biomass and species diversity compared to salt marshes.
The Sea Level Affects Marshes Model version 5 (SLAMM 5) is used to model how rising sea levels
affect wetland area and habitat conversation. SLAMM 5 uses elevation, the National Wetlands Inventory,
tide range, historic sea level rise, and site-specific accretion rate data to parameterize the model. The
researchers developed a salinity algorithm that is used to simulate saltwater intrusion into river-dominated
estuaries as sea level rises. SLAMM 5 runs in 25-year increments to predict future scenarios; by the year
2100, the model predicts an increase in open water and decreases in salt marshes, tidal freshwater
marshes, and tidal swamp in the Altamaha River estuary. SLAMM 5 also simulated the effects of
accelerated sea level rise along 200 miles of the Georgia coast. The model predicted 20 and 24 percent
losses of salt marsh and tidal fresh swamp, respectively, and minor gains in tidal freshwater and brackish
marshes. There is a predicted cumulative loss of 12 percent of the wetland habitat, mostly salt marsh, but
only a 4 percent loss in ecosystem services because freshwater marshes have increased amounts of eco-
system services per area. The researchers also examined the effect of diking, and as expected, the loss of
connectivity in an area results in a loss of ecosystem services. Diking causes losses of connectivity,
sediment deposition, water quality improvement functions, nitrogen retention, phosphorus storage, and
denitrification; however, there is an increase in waterfowl habitat. Other research indicated that Spartina
alterniflora aboveground biomass is positively correlated with freshwater, river discharge, and
precipitation, and crab hole density is more strongly correlated with salinity. Sediment deposition is
positively correlated with river discharge.
The researchers have identified several lessons learned from the project. (1) Different types of tidal
marshes provide different levels of ecosystem services. (2) Tidal freshwater and brackish marshes have
greater aboveground biomass, nitrogen retention in soil, and denitrification than salt marshes. (3) Climate
change (i.e., sea level rise) will promote salt water intrusion and submergence, leading to habitat
conversion and loss of tidal marshes, especially those at either end of the salinity gradient. (4) Wetland
loss may not be as great as predicted because spatial models lack positive feedback mechanisms that
enable marshes to increase surface elevation. (5) Although diking protects tidal marshes, it leads to loss of
connectivity to estuarine waters and the ecosystem services that depend on connectivity. (6) Tidal marsh
ecosystem services are more strongly correlated with variation in salinity, driven by river discharge, than
by variation in temperature and precipitation. The researchers faced challenges in evaluating ecosystem
services of fauna and wildlife and working with subcontractors. The researchers interacted with The
Nature Conservancy and the U.S. Fish and Wildlife Service (FWS). Outcomes include SLAMMView, an
interactive Web-based tool to visualize sea level rise that can be found at http://www.slammview.org, and
several additional projects and publications.
Climate-Linked Alteration of Ecosystem Services in Tidal Salt Marshes of Georgia and Louisiana
Mark Hester, University of Louisiana at Lafayette
Dr. Mark Hester explained that drought-induced, sudden dieback of S. alterniflora tidal salt marshes had
been observed in Louisiana in 2000, Georgia in 2001, and several other coastal states since; therefore,
there is the potential for drastic alteration of ecosystem services driven by a decrease in S. alterniflora
living stem density, which will be directly linked to degree of loss of ecosystem processes. The project
goals are to: (1) elucidate the effects of climate change (increased drought severity) on tidal salt marsh
ecosystem services (e.g., eutrophication control, carbon sequestration, sustainable habitat, faunal support)
in two hydrogeomorphic settings (microtidal in Louisiana and mesotidal in Georgia); and (2) develop an
exploratory structural equation model to explore causal relationships and ecosystem service latent
variables.
The Office of Research and Development's National Center for Environmental Research 3
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
The researchers are examining several ecological processes (nutrient cycling and transformation, primary
productivity, decomposition, erosion, sedimentation, infaunal abundance, composition, and diversity) and
their effects on ecosystem services, such as eutrophication control, carbon sequestration, maintenance of a
sustainable and healthy habitat, refugia, and habitat support for fauna. The experimental approach will
include a manipulative field experiment of S. alterniflora plant density in micro- and mesotidal salt marsh
ecosystems in Georgia and Louisiana to identify six dieback areas within each state for establishment of
large research plots with four vegetative conditions. A series of specific, a priori, process-driven
hypothesis testing will be completed, and a structural equation model will be used to reveal relationships
between stem density, ecosystem processes, and ecosystem services. Challenges the researchers have
faced include Hurricane Katrina's landfall following setup of the experimental sites, a prolonged drought
in Georgia, and personnel changes. The lesson learned is that improved communication and adaptive
management are necessary components of the project.
Results indicated that, in terms of aboveground primary productivity and carbon assimilation, S.
alterniflora cover reflected stem densities, and although the desired gradient is present in Louisiana, new
dieback occurred in 2008 in three Georgia reference plots. An additional experiment indicated that higher
S. alterniflora density provides more efficient utilization of leaf nitrogen for carbon assimilation. The
researchers examined sediment accretion and net marsh surface elevation change and found that, in
Louisiana, high-density plots had equivalent accretion rates to reference plots, high-density plantings
increased surface elevation, and bare plots had lower accretion rates and lost elevation; in Georgia,
reference plots had the greatest accretion rates, high-density and bare plots had lower but equivalent
accretion rates, and the reference plots are losing elevation over time. When the researchers examined
belowground productivity and decomposition, the high-density and reference plots had equivalent below-
ground productivity rates. In terms of biogeochemistry, cyanobacteria mats in Louisiana plots were most
abundant in low-density and bare plots, which has implications for a potential shift from a detrital to
grazing food web. There was much greater interstitial ammonium in Louisiana, but there was no
consistent pattern with stem density. Interstitial sulfides were much greater in Louisiana and often below
detection in Georgia; because sulfides can inhibit plant uptake of ammonium, there may be less tight
coupling of plant carbon and nitrogen relations in Louisiana.
In terms of secondary productivity, the presence or absence of certain meiofauna species showed
interesting differences and loss of some services in Louisiana. In Georgia, meiofauna increased in
vegetated areas, whereas in Louisiana, meiofauna increased in bare areas. Nematodes were larger in
vegetated treatments in Georgia, but there was no difference in nematode size in Louisiana or copepod
size in either state. The exploratory structural equation model will examine S. alterniflora stem density as
the main driver on ecosystem processes and services that can be measured. It will be a two-group model,
and Louisiana and Georgia are expected to be different. Key findings so far are that climate change
(severe drought) can affect a suite of ecosystem services, S. alterniflora density is an important driver of
many ecosystem services across the hydrogeomorphic setting, and the hydrogeomorphic setting is an
important modulator of ecosystem processes and services. The researchers have interacted with clients
throughout Georgia and Louisiana, resulting in several presentations and synergistic activities related to
the project. Currently, the data are in the process of final integration. The structural equation model is a
valuable management tool that identifies key differences in the strength of relationships between S.
alterniflora density and hydrogeomorphic setting. The work has resulted in improved insights into
climate variability that will help federal, state, and local agencies with future management and planning.
Linking Impacts of Climate Change to Carbon and Phosphorus Dynamics Along a Salinity Gradient in
Tidal Marshes
Nathaniel Weston, Villanova University
Dr. Nathaniel Weston noted that sea level rise is of great concern, especially as its rate is increasing and
accelerating; coastal tidal marshes are affected significantly by this rise. Marshes must accrete to keep
The Office of Research and Development's National Center for Environmental Research 4
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
pace with rising sea levels and do so via watershed inputs and carbon dioxide primary production. Rising
sea levels and river evaporation will increase the amount of salt in the system. The goals of this project
are to: (1) understand how salt water intrusion into tidal freshwater marshes will impact carbon, nitrogen,
and phosphorus cycling; and (2) predict the response of tidal freshwater marshes and the ecosystem
services they provide to scenarios of future climate change. The study site is the Delaware River estuary,
which is comprised of freshwater between Philadelphia, Pennsylvania, and Wilmington, Delaware, and
brackish below Wilmington. An increase in salinity has been observed over time in the brackish area of
the estuary. The researchers will study watershed inputs, inorganic sediments, and microbial processes to
determine the impact of climate changes on tidal freshwater marshes. Microbial respiration is carried out
via methanogenesis in freshwater marshes and via sulfate reduction in salt marshes. The researchers will
determine the importance of salt water intrusion in microbial processes via a long-term salinity intrusion
experiment that measures sulfate reduction and methanogenesis rates as well as other biogeochemical
measurements.
Results indicated that the increase in carbon dioxide flux following a saltwater intrusion event was
statistically significant for 8 months following the event. Additionally, there was an increase in the
amount of organic matter being mineralized. Sulfate reduction rates did not change, and methane flux
increased significantly for 4 months following the event. Total carbon gas flux was significantly higher
for 6 months following the saltwater intrusion event, with a 50 percent higher carbon gas flux during the
course of the year following the event. This is linked to a decrease in soil organic matter, which becomes
apparent 3 months after the event. The results indicate that relatively little is known about how microbial
communities respond to changing environmental drivers (e.g., climate change); microbial response is
important to composition. Therefore, the researchers initiated a field transplant experiment at four sites to
understand how microbial communities respond to climate change. The field site monitoring measured
carbon dioxide and methane flux, plant biomass, microbial rates, biogeochemistry, and microbial
community composition. During the first year, the first two sites had a good seasonal signal of plant
biomass, but the plants died at the third site during mid-summer as the salt levels increased. The plants at
the fourth site died almost immediately. During the second year, species common to brackish or salt
marshes grew, indicating a shift from freshwater to brackish or salt marsh. The response of freshwater
marsh plants to salinity intrusion and inundation indicate that there is a significant negative relationship of
plant biomass to productivity and inundation. The current transplant experiment researchers are con-
ducting controls for elevation.
Inorganic sediment is a major input for watersheds. The researchers are examining monitoring data from
the U.S. Geological Survey (USGS) from the 1970s to the present and found that there has been a steady
and serious decline in suspended sediment in the Delaware River. USGS data on 42 additional East and
Gulf Coast rivers indicates a significant decrease in suspended sediment in 48 percent of the rivers.
Results indicate that the decline in plant production, which leads to decreased deposition of organic
matter—combined with increases in microbial response and carbon dioxide and methane flux and a
decrease in watershed inputs—leads to a loss of freshwater tidal marshes.
The challenges the researchers faced in carrying out the projects included controlling for marsh vertical
elevation critical in field experiments, the increased diversity of plant species in tidal freshwater marshes,
understanding the response of methanogens, and the number of complex and interconnected processes.
The researchers integrated ongoing work with other groups (e.g., Partnership for the Delaware Estuary,
University of Delaware, Rutgers University); communicated with local stakeholders; and presented their
work at several national meetings.
The Office of Research and Development's National Center for Environmental Research
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
Connectivity in Marine Seascapes: Predicting Ecological and Socioeconomic Costs of Climate Change
on Coral Reef Ecosystems
Julie Kellner, Resources for the Future
Dr. Julie Kellner explained that the deteriorating health of the world's coral reefs threatens global
biodiversity, ecosystem function, and the livelihoods of millions of people living in tropical coastal re-
gions. The researchers have initiated a collection of team projects that focus on coral reef resilience,
particularly in response to bleaching, hurricanes, changes in trophic and habitat relationships, fishing, and
coastal development and management. The goals of the project are to: (1) integrate theory and data from
ecology, biology, and the social sciences to address major questions about the potential consequences of
climate change on coral reef ecosystems; (2) predict how fishing pressure, tourism development, and local
economies will be affected by climate change stressors; and (3) provide guidance for future management.
The study sites include a variety of different systems, including marine reserves and unprotected areas, in
or near the Bahamas archipelago, Belize, and Bonaire.
Trophic relationships in the Caribbean are very complex interactions that involve predators and their
recovery and relationships to lower species. Threats to reefs include coral bleaching and hurricanes. Coral
bleaching is the response of corals to elevated temperatures or high levels of ultraviolet radiation in which
they expel their symbiotic algae. Corals can recover these algae following weak exposure, but prolonged
exposure can cause mortality. Hurricanes can damage, overturn, and kill corals, and the movement of
sediments and debris causes scouring. Increased nutrients, as a result of hurricanes, can encourage algal
growth. Macroalgae compete with corals, and reefs can switch from a healthy, coral-dominated state to an
unhealthy, algal-dominated state. Grazers are important because they can influence the replenishment
rate, growth, and fecundity of coral colonies. Grazing underpins the resilience of coral reefs to distur-
bance. Many different models, especially those simulating bleaching and hurricanes, have been used to
determine the impacts of disturbances on coral reefs. Modeling indicates that mortality caused by
bleaching depends on the magnitude and duration of thermal stress and each coral's thermal history,
whereas mortality caused by hurricanes depends on strength of the hurricane (based on the Saffir-
Simpson Hurricane Wind Scale) at the reef location and colony size.
When parrotfish (grazers) were not exploited, the models predicted the health of Belize's coral reefs and
indicated that bleaching or bleaching combined with hurricanes will significantly decrease the amount of
coral cover, leaving almost no coral cover by the year 2100, whereas hurricane-only scenarios increased
the amount of coral cover. When parrotfish were exploited, all three scenarios (hurricanes only, bleaching
only, and hurricanes and bleaching combined) reduced coral cover significantly by 2030, with no coral
cover remaining by 2090. In terms of coral cover and the role of grazers, the researchers asked whether
the loss of grazers reduces resilience and how the systems can be recovered for future resilience against
disturbance. There is the potential for two different stable states in the system—low coral/high algae or
high coral/low algae. The adaptive capacity of the system may change if left degraded. The model
prediction was that increased grazing equals increased stability. The negative feedback loops in the
system include increased macroalgal cover and reduced structural complexity, fish recruitment, coral
cover, grazing intensity, and coral recruitment; the positive feedback loops are the opposite. Grazing
intensity increases as herbivore biomass and coral cover increase. The modeling allows managers to meet
the challenge of keeping reefs highly resilient by showing how resilient the reefs will be based on their
initial coral cover; this illustrates to managers where their efforts should be focused.
No-take marine reserves decrease the fishing pressure, which is important for resilience and results in
faster coral recovery within the marine reserve. The presence or absence of mangroves affects the density
and biodiversity of species, and one ongoing project examines how important mangroves are to the
community structure of important herbivores. The presence of mangroves increased the grazing intensity
of two types of parrotfish. Absence of mangroves decreased coral cover and increased algae; modeling
indicated that a current coral reef with 10 percent coral cover will have 12 percent coral cover in 50 years
The Office of Research and Development's National Center for Environmental Research 6
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
in the absence of mangroves. Presence of mangroves, however, increased resiliency, particularly follow-
ing hurricanes, and decreased coral mortality; modeling indicated that a current coral reef with 10 percent
coral cover will have 60 percent coral cover after 50 years in the presence of mangroves.
With regard to habitat conservation and ecosystem services, the researchers asked the following
questions: (1) How does habitat (and loss thereof) affect the productivity of fisheries? (2) What does this
imply for the economic value of habitat? (3) How do these values impact coastal land-use decisions?
Bringing ecology into habitat valuation will enable the development of a model that allows for obligate
and/or facultative associations, explores multiple habitat types, and links recent findings in ecology to the
tools of economic valuation.
The researchers learned a variety of lessons during their research. Caribbean coral reefs appear to exhibit
alternate stable states. There are threshold levels of coral cover, grazing, nutrients, and so forth. Restoring
reef health becomes disproportionately more difficult as health declines. There is a need to act sooner
rather than later. Coral resilience is linked with the probability that the reef does not become entrained in
a shift toward a stable algal state. The researchers have conducted outreach to educators, practitioners,
decision-makers, and the public via a Web site, booklets, videos, posters, teaching resources, newsletters,
meetings, and presentations. The researchers have found that integrative models are useful, particularly
for management and education.
Effects of Climate Change on Ecosystem Services Provided by Hawaiian Coral Reefs
Paul Jokiel, University of Hawaii at Manoa, Hawaii Institute of Marine Biology
Dr. Paul Jokiel explained that the goal of the project, which finished recently, was to integrate and extend
existing models to develop a comprehensive, scenario-based analysis of the range of possible effects of
global climate change on ecosystem services provided by the coral reefs of the Hawaiian archipelago and
on the economic valuation of predicted changes. The developed model is available for online use and can
be downloaded at http://www.kgs.ku.edu/Hexacoral; this availability provides for community involve-
ment through hands-on testing and feedback. The challenges the researchers faced included building a
model at three levels (climate change, biological response, and ecosystem services) and occurrence of the
economic downturn at the time of the project's valuation survey. Many lessons were learned, but an
unexpected lesson was that coral growth and mortality were central to all of the work. The researchers
interacted with clients through presentations at various workshops and meetings. Additionally, four
published papers from the project were cited in the Federal Register announcement of EPA's Ocean
Acidification and Marine pH Water Quality Criteria Notice of Data Availability.
There have been a number of background studies since the 1970s regarding the response of Hawaii's
coral reefs to temperature increases. The current mesocosm study experiments are conducted in conti-
nuous flow outdoor mesocosms that simulate the reef environment, and the experimental treatments
include acidification to produce the carbonate saturation states predicted for the year 2100. Following
acidification, noncalcifying algae increased by 52 percent, and crustose coralline algae recruits and cover
decreased by 78 and 92 percent, respectively. Results of the mesocosm coral growth experiment showed
that no mortality occurred in the acidified or control treatments, but coral calcification was reduced by
15-20 percent in the acidified treatment. Corals grown in the acidified treatment produced a more delicate
skeleton, including thinner branches and a decrease in skeletal density, and there was no evidence of
acclimation. Rhodoliths, which are accretions of crustose coralline algae, showed a 250 percent decrease
in calcification following acidification compared to the control group. The mesocosm wall settlements
experiment indicated that crustose coralline algae significantly decreased and bare substratum signifi-
cantly increased following acidification. Results confirmed previous studies that showed that ocean
acidification is affecting calcification but not the organic components of settlements. Additionally, there
was no change in reproduction rates, consistent with prior studies. A flow-through experiment illustrated
The Office of Research and Development's National Center for Environmental Research
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
the net ecosystem calcification and indicated that the system as a whole is decalcifying despite coral
growth; therefore, corals will continue to grow even as the reefs are dissolving.
The main focus of the research was to develop the Coral Mortality and Bleaching Output (COMBO)
Model. Within the model, the user has control of all factors (sensitivities, probabilities, environmental
inputs) via a user-friendly interface, and regionally appropriate default values are provided. Finally, the
effects of quasisteady-state temperature, carbon dioxide concentration, and temperature variation are
assessed independently in the model and accumulated into net change in cover. Calculations are
performed in linked, user-accessible worksheets with options for replacing the built-in datasets, and
output plots and tables are updated immediately as input values are changed. The sample output of the
COMBO Model indicates that Hawaiian reef coral cover could largely disappear by the end of the century
if fossil fuels use continues to expand. Additionally, the COMBO Model was updated to reflect the latest
model ensembles for the current Intergovernmental Panel on Climate Change (IPCC) predictions and is
more sophisticated and complex as a result. The resolution was increased and the noise was extracted to
allow inclusion of bleaching thresholds and how these will affect the vulnerability of susceptible areas.
The model was run many times with the same conclusion: It is extremely unlikely that viable coral
populations will persist in the shallow waters of the Hawaiian archipelago in 2100, and precipitous
declines likely will start in the northern region sometime between 2030 and 2050, with a steady decline
during the entire century throughout the region. The model also indicates that bleaching events are
important. Although Dr. Wolfgang Haider will present briefly on the socioeconomic modeling aspect of
the project, Dr. Jokiel noted that hedonic price modeling suggests that coral reef presence and quality
have a significant impact on house prices.
Tier I Discussion
Dr. Denice Wardrop asked EPA staff how the Agency will utilize the results of the studies, especially
considering the management implications for long-term planning and policies. Dr. Jones responded that
one of the goals of the workshop was for EPA to receive input from the researchers about possible next
steps. With the new Administrator's emphasis on climate change and its impact on ecosystems, the
Agency would like to take basic science information and put it in a format that can inform the chain of
command. The workshop discussions should include next steps, including input regarding the next
Request for Applications (RFA). Dr. Lorenzana added that it is not too late for researchers to partner with
the regions, which can be used to reach state and local partners. Each region has a Regional Science
Liaison, who can be used as a resource. She noted that Mr. Thomas Baugh, the Region 4 Regional
Science Liaison, was present at the workshop. Mr. Baugh explained that Region 4 is comprised of eight
states (Kentucky, Tennessee, North Carolina, South Carolina, Georgia, Florida, Mississippi, and Ala-
bama) and has the highest population of any of the regions. Dr. Lorenzana noted that Dr. Jones could help
any of the grantees contact their Regional Science Liaison.
Mr. Baugh asked Dr. Craft to comment on the quality and quantity of the historical sea level rise data that
are input into SLAMM 5. Dr. Craft responded that National Oceanic and Atmospheric Administration
(NOAA) maintains long-term monitoring stations at various coastal locations, and the researchers used
the NOAA rate from Ft. McAllister, Georgia, which is 3.1 mm/year. Generally, NOAA data are available
for 50-100 years within 50-100 miles of almost any coastal site. The model's weakness actually is the
elevation data. The researchers used the National Elevation Dataset, which is rather coarse. The
researchers are working to obtain Light Detection and Ranging (LIDAR) data. In terms of accretion rate
data, it should be site-specific for input into SLAMM 5. Many of the data required for input are publicly
available. He noted that it would be helpful if Region 4 could help acquire better LIDAR and elevation
data.
Dr. Wardrop asked Drs. Craft and Jokiel about their experiences resulting from the Web availability of
their models. Dr. Jokiel stated that people were actively using the Web site and models, which would
The Office of Research and Development's National Center for Environmental Research 8
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
evolve, especially as the subject matter (coral reefs) will be extinct by the end of the century if changes
are not made. Dr. Craft explained that the SLAMMView Web Site is a method by which to visualize the
modeling concept. The model itself is available at several Web sites (e.g., FWS). There have been so
many requests for technical support that the staff are becoming overwhelmed and may need to start
seeking financial compensation for technical support services in the future.
Dr. Robert Whitlatch asked the marsh researchers whether there was a manner by which to examine the
ecosystem services provided by ground marshes that could provide some insight into loss of minerals and
services. Dr. Hester responded that some current approaches could be applied to this problem. In terms of
eutrophication and primary productivity of the sites, there may be seasonal variations in primary
productivity that can affect nutrient cycling; the processes that can be measured at the site can be related
to the services. Dr. Whitlatch asked whether the current models are easily adaptable to different systems.
Dr. Hester responded that if the goal was to examine relationships between drivers of ecological processes
and ecosystem services, then a structural equation model approach could be used; conceptually, this
would be fairly easy to accomplish. Dr. Craft added that although SLAMM 5 does not incorporate this, it
can be adapted with bathymetry data. Dr. Hester added that erosion could be incorporated into a structural
equation model depending on the strength of the relationships. When using a structural equation model, it
is possible to use the best available scientific data to create an a priori model and then further develop the
model into the confirmatory model using hypothesis testing.
Dr. David Purkey asked Dr. Weston for his thoughts on the hypothesis that, in terms of the USGS
sediment data, the level of water management and reservoir infrastructure development may be related to
the decline in sediment. Dr. Weston agreed that this was probable. Damming and other efforts to decrease
suspended sediments—because they are not helpful to rivers and streams—have been successful and do
not consider the fact that marshes depend on receiving sediment from rivers and streams. The manage-
ment implications are challenging because it is not desirable to deposit large amounts of sediment into
rivers, although this would greatly help marshes and their management. Dr. Craft added that measuring
accretion can be challenging, and marshes increase their accretion rate with organic matter in addition to
sediment. In microtidal systems, organic matter appears to be more important than sediment amount.
Dr. Thomas Meixner noted that the tidal marsh studies deal with the influence of freshwater input into the
marsh systems. As changes occur, are there management practices that can be put in place? Dr. Weston
stated that there was no easy answer to this question. Climate change will affect precipitation patterns,
which in turn will affect salinity levels. Warmer temperatures also increase evaporation and therefore
salinity. Work can be focused on decreasing water withdrawals in the summer months so that there are
less low river flows.
Mr. Baugh asked Dr. Kellner to elaborate on the connection between coral reefs and human health that
she mentioned. Dr. Kellner explained that she was speaking in general terms; most models focus on the
socioeconomic benefits (e.g., fishing, tourism). There are some biodiversity issues in terms of what coral
reefs can provide for medical products.
Dr. Jones noted that ORD is interested in data translation and transfer issues. It is necessary to ensure that
the right people receive the information in a timely matter, particularly in relation to global climate
change (e.g., the expected loss of coral reefs within the next century). It is necessary to reach Congress
regarding the impact of valuation on constituents. Part of the focus needs to be on how to best translate
and transfer the information to the right people. ORD will develop a series of one-pagers regarding the
grants with different language to reach different audiences. Dr. Purkey commented that it is challenging
to translate basic science into real-life management decision-making. He suggested that the next climate-
change-related Science To Achieve Results (STAR) RFA demonstrate a preference for proposals that are
motivated by real-life management decisions that incorporate climate change so that the researchers will
be motivated by this, making the audience more receptive to the research information. Ms. Lisa Macchio
The Office of Research and Development's National Center for Environmental Research 9
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
of Region 10 noted that basic research is very beneficial, but regional programs want to know its
applications. It is important for researchers to explain how their results can be applied and inform
decision-making (e.g., how models are useful in writing total maximum daily load reports). It would be
beneficial to the Agency if researchers explained how to use the research in the grant applications and
final grant reports.
Dr. Ramesh Reddy of the University of Florida asked whether there is a movement within EPA to place
an economic value on ecosystem services. Dr. Jones responded that there had been an Economics and
Decision Sciences Research Program within ORD, but it did not receive a good deal of support, so ORD
is examining methods and partnerships to revive it; there have been many discussions within the Agency
about how to accomplish this. NOAA has an advantage in this respect because of its role with fisheries,
recreational areas, and so forth. EPA partners with other agencies that address valuation of ecosystem
services. Dr. Lorenzana added that partnering with regional offices can help; Region 10 employs an
economist who is working on ecosystem services and valuation.
Mr. Jerry Kuhn of Region 10 emphasized how critical it is for researchers to highlight the applications of
their research for EPA staff because this information is essential for writing regulations; researchers need
to use their science to advise the Agency. There is an ecosystem services valuation report available for the
Puget Sound; it is the most comprehensive work on ecosystem services to date. Mr. Kuhn asked Dr. Jones
to add the report to the workshop Web site.
Willingness to Pay for Mitigation of Climate Change Effects on Hawaiian Coral Reefs:
A Contingent Choice Study
Wolfgang Haider, Simon Fraser University
Dr. Haider explained that the goal of the project was to estimate people's willingness to pay for mitigating
the effects of climate change on coral reefs in Hawaii. The challenges were to separate use values from
nonuse values, control for key components of the reef ecosystem, and design a payment vehicle that is
applicable from the present but leads to uncertain outcomes in the future. The method chosen was a
contingent choice survey. Environmental valuation includes bequest, option, and existence values as
components. The attributes that were chosen for the discrete choice experiment were selected after much
discussion and include coral cover, coral health, fish numbers, species diversity, water clarity, mitigation
fees, presence or absence of turtles, and levels of relief (low, medium, or high). Survey respondents were
provided images so that they could visualize their responses. The survey instrument was a fairly complex
Web-based survey that included 1,000 mainland residents and 500 Hawaii residents. Pictures were
combined with textual information to decrease variable interpretations of the questions. Dr. Haider
showed examples of the survey questions and explained that the conservation fee referred to in the survey
was the cost to visit a coral reef; this determined the nonuse value. Results indicated that water clarity,
coral cover, mitigation cost, fish numbers, species diversity, and coral health were significant. The decline
index also had a significant effect; people's willingness to pay increases with the forecast of a negative
future. Additionally, respondents had a significant belief in climate change; mainland residents had an
increased willingness to pay compared to Hawaii residents, whereas climate change believers had an
increased willingness to pay compared to climate change skeptics.
The Office of Research and Development's National Center for Environmental Research 10
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
TIER II: NONLINEAR RESPONSES TO GLOBAL CHANGE IN LINKED AQUATIC AND TERRESTRIAL
ECOSYSTEMS
Hydrologic Forecasting for Characterization of Nonlinear Response of Freshwater Wetlands to
Climatic and Land Use Change in the Susquehanna River Basin
Denice Wardrop, Pennsylvania State University
Dr. Wardrop explained that the Susquehanna River Basin is a 26,000-square-mile watershed, which
provides 51 percent of the freshwater to the Chesapeake Bay. Wetlands in the area comprise approxi-
mately 2-4 percent of the landscape, and 73 percent of the wetland area is associated with headwater
streams. The area is predominantly forested, and each individual wetland is less than 10 acres. The
objective of the project is to characterize nonlinear responses through: (1) selection of a linked terrestrial-
aquatic ecosystem that provides critical ecosystem services and ecological functions; (2) characterization
of various global change scenarios, incorporating both climate and land cover, and a method of assessing
their effect on the identified ecosystem through the primary forcing factor of hydrology (both alone and in
conjunction with other human-associated stressors); (3) identification of potential nonlinear ecological
responses in the selected ecosystem as a result of these changes; and (4) estimation of the resultant change
in ecosystem services on watershed and basin-wide scales. Scenarios of climate and land cover change
generated by various departments are input into predictive hydrologic scenarios. These scenarios,
combined with hydrology and ecological function models and functional loss estimates of plants and
macroinvertebrates, are used to obtain valuation of change in ecosystem services and identification of
nonlinearities.
The researchers were challenged with scaling issues, needing to determine: (1) the assessment unit;
(2) how to stratify the study area for the purpose of sampling, modeling, and subsequent upscaling; (3) the
scale in which to express the final results; and (4) how to resolve extent and resolution differences in scale
of the various components of the project. The assessment unit needed to: (1) integrate freshwater wetlands
with important contextual landscape, (2) be of a spatial and temporal scale that matched ecosystem
services, (3) be of a scale capable of being modeled, and (4) be representative of the range of conditions
in the Susquehanna River Basin. The various interdisciplinary researchers solved this by discussing the
spatiotemporal scale used in each discipline. The biological data were at a wetland scale, whereas the
other disciplines were working at a reach scale or greater; the researchers realized that the reach scale
could work for all data. The researchers downloaded data from the IPCC's fourth report, which includes
21 models from 12 countries; some models have multiple realizations with different horizontal resolu-
tions. To chose the most ecologically relevant model, the model output and observational data were
placed on a one degree grid within the Susquehanna River Basin; the models were a tolerable fit,
predicting wetter springs and drier summers compared to actual precipitation. An overall squared error
was computed for each variable and model and normalized over all models to compute the overall
performance index by averaging over all of the metrics. Six of the 10 selected metrics deal with mean,
and four with variability and extreme events. In terms of model performance, the mean generally was the
best fit except in extreme scenarios. The researchers identified several lessons learned. Models differ
dramatically in their ability to predict the climate of the Susquehanna River Basin. The model mean is
superior to any individual model and specific to region. The raw model output was not as bad as expected.
The three scenarios prepared for the hydrology model were: (1) daily output from 1960-1990 to establish
a baseline, (2) daily output from 2035-2065 to show the effect of climate change, and (3) impact of the
change in mean climate versus change in variability. The multiscale Penn State Integrated Hydrologic
Model (PIHM) incorporates climate and land-use effects. The PIHM finite volume approach uses a
triangular irregular network (TIN) to allow nested grids and is calibrated across the ecoregions of the
Susquehanna River Basin. The model can decompose nearly 100 square miles of watershed into nearly
1,000 TINs and predict changes in terms of depth to water table, stream flow, and left- and right-bank
baseflow. In considering hydrological modeling, the researchers learned that: (1) scale-appropriate and
The Office of Research and Development's National Center for Environmental Research 11
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
ecologically relevant hydrologic scenarios can be predicted, (2) ecologically relevant and powerful
metrics are difficult to identify, (3) there is a spatially heterogeneous response to a homogeneous forcing
function, and (4) absolute values of predictions are difficult to utilize in a meaningful way. The research-
ers also asked whether changes in hydrologic regime could result in loss of wetland area and/or loss of
function through physical changes and the loss of functional process zones. A reach classification was
devised that encompasses the entire riparian area. The researchers also are examining land-use change
and disturbance in addition to hydrology. The assumption is that hydrological complexity leads to ecolog-
ical complexity.
As more work is completed, the researchers are utilizing feedback loops to revisit the climate change
scenarios to assess whether the right climate model was chosen. There has been a good deal of interaction
with clients, including assessments for various groups near the Chesapeake Bay and Mid-Atlantic region.
The major lesson learned is the importance of scale. The researchers will continue to perform hydrologic
modeling, characterize and validate physical and functional reach characterizations across physiographic
provinces, and define the distributions of hydrologic parameters for each reach type and disturbance cell
and extrapolate the results basin-wide.
Sustainable Coastal Habitat Restoration in the Pacific Northwest: Modeling and Managing the
Effects, Feedbacks, and Risks Associated With Climate Change
John Rybczyk, Western Washington University
Dr. John Rybczyk stated that the project's study site is the Skagit River System, with a focus on the
estuarine delta area. Historically, the Padilla Bay was part of the Skagit River Delta, but estuaries
currently exist only on the fringes because of human intrusion. The Padilla and Skagit Bays now are
isolated from each other with no dynamic exchange. Historically, the Skagit Bay was comprised of all
estuary habitats of some type, but because of dikes and levies, these estuaries exist only on the fringe.
How can the area be restored, especially in light of recent sea level changes? The experimental approach
was to link sea level rise predictions to LIDAR data and known elevation distributions of vegetation in
the tidal marshes of the Skagit River Delta. Sea level rise scenarios were run to determine how the
vegetative communities have changed and how and where to plan for restoration.
Approximately 3,000 hectares of intertidal eelgrass beds, an important habitat for the Pacific Northwest,
were examined in the Padilla Bay, which has no compensation mechanisms in the face of sea level rise as
it is cut off from its historical source of freshwater and sediment. The researchers investigated whether
eelgrass beds in Padilla Bay were at risk and whether they were accreting at a rate that keeps pace with
sea level rise; it appears that the areas are erosional rather than accretional and are losing elevation at a
rate of approximately 0.57 cm/year. These types of analyses must be taken with caution when making
long-term predictions, however, because they ignore climate change-induced alterations in salinity, tidal
regime, river flow, and sedimentation. The analyses imply linearity, but because of system feedbacks,
response to sea level rise is nonlinear. Decomposition, primary production, and sediments change in
relation to elevation, and these nonlinear dynamics also must be considered.
The goals of the project are to: (1) develop a predictive simulation model, incorporating nonlinear
elevation feedbacks, of the ecological and geomorphological consequences of sea level rise and river flow
alterations in Padilla and Skagit Bays; (2) use the model to guide the course of restoration efforts, given
climate change, in the Skagit River estuary; (3) link a spatially explicit hydrodynamic/sediment model to
a mechanistic wetland elevation dynamics and vegetation model; and (4) initialize and calibrate the model
using extensive site-specific data collected for the project. Currently, the researchers are cataloging soil
salinity, vegetative communities, and elevation as well as developing an extensive data network. The unit
(elevation/vegetation) model builds a sediment cohort that grows in response to above- and belowground
productivity and mineral matter inputs and then compacts and decomposes; what remains contributes to
wetland elevation, which in turn is affected by sea level rise. The relative elevation model was extended
The Office of Research and Development's National Center for Environmental Research 12
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
to model a three-dimensional surface instead of a single point, which was used to determine how eelgrass
beds will shift with various sea level rise scenarios. The next step will be to integrate the unit model with
the hydrodynamic model, which incorporates tides, winds, and river flows and simulates salinity, flow
rate and direction, and sediment. The researchers have interacted with a number of clients, including the
Skagit Climate Science Consortium, which includes a diverse group of stakeholders. The focus is to
integrate the results of several modeling efforts within the Consortium into one overarching model.
Nonlinear Response of Pacific Northwest Estuaries to Changing Hydroclimatic Conditions: Flood
Frequency, Recovery Time, and Resilience
Anthony D'Andrea, Oregon State University
Dr. Anthony D'Andrea explained that floods and delivered sediment are increasing. This project focuses
on flooding events, an important focus for the Pacific Northwest, which receives a great deal of precipi-
tation. Climate models predict an increase in total precipitation, particularly the frequency of extreme,
high-rain events. River flow and flooding in the Pacific Northwest are increasing and further are
amplified by seasonal rainfall patterns. Additionally, a combination of watershed and estuarine changes
have decreased buffering capacity and led to increased sediment flux with potentially important but
unknown effects; rapid sedimentation during floods can lead to abrupt changes in benthic intertidal
communities. During the last 30 years, documented changes have occurred in Pacific estuaries, including
reduction in benthic abundance and diversity, alteration of tideflat habitat, and rapid growth of
nonindigenous species.
Because most studies are anecdotal or focused on limited numbers of species, the overarching goal of this
project is to complete community-level studies of flood sedimentation impacts on estuarine benthic
communities. The approach is a manipulative field study simulating the effects of the frequency of floods
on Pacific Northwest benthic intertidal communities. Researchers are focusing on four interconnected key
research questions dealing with recovery, impacts of flood sedimentation, within-year frequency of
floods, and impact of floods on community susceptibility to nonindigenous species. The specific project
goals are to: (1) design and implement a manipulative field study to determine the ecological effects of
flood sedimentation on intertidal benthic macroinvertebrate communities; (2) use a combination of high-
resolution benthic sampling and multivariate analyses of benthic community metrics to track the initial
mortality, recovery, and resilience of the benthic community; (3) collect and analyze sediment samples
parallel to the benthic community samples to track changes in important sediment properties that have
direct or indirect effects on survival or habitat suitability of sediments to the benthic invertebrate commu-
nity; and (4) synthesize the datasets from the study to develop an empirical and theoretical framework for
predicting the effects of flood sedimentation events on tideflat macrobenthic communities in Pacific
Northwest estuaries and how these changes impact ecologically and economically important biotic
resources and ecosystem services.
The study site is Netarts Bay, chosen because of its large tidal area, relatively pristine state, and
accessibility, among other reasons. The study plots are of uniform intertidal height and divided into three
experiment groups: control, single flood, and multiple floods. Challenges and lessons learned are that
ironic weather cannot be predicted, and the multiple uses of the bay complicate fieldwork. Dr. D'Andrea
described the procedure for creating the flooding events and the sampling approach. The researchers
observed flood layer and physical properties, including temporal change and sediment physical properties
and geochemistry. In terms of temporal change, the key observation was that there is persistence as well
as potential long-term impacts despite high current speeds at the site. Researchers observed a seven-fold
increase of total organic carbon in the flood plots relative to control, which is a persistent feature through
at least the first 150 days following the flooding event. The researchers tracked changes in oxygen
patterns—which can be an indicator of stress to benthos—to assess impacts on the benthic microalgal
community that can actively oxygenate surface sediments and provide food resources for benthos. Under
dark conditions, there was no significant difference between the experimental groups; however, under
The Office of Research and Development's National Center for Environmental Research 13
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
dark conditions, there was little or no benthic photosynthesis in flood groups, which could add additional
food limitation stress for surviving benthos.
The researchers measured the benthic community by observing changes in community structure and
diversity and functional changes. Uni- and multivariate analyses were used to identify key community
metrics and track community changes. The study community was comprised of two smaller communities,
surface-dwelling and deep dwelling; the deep-dwelling community is the less mobile. There was a large
amount of behavioral response to flooding; mobile species immediately vacated the flood plots, and there
was a significant reduction in species richness for at least the first 72 days following the flood event.
Infaunal abundances were consistently lower in flood plots compared to controls, and the effect of the
disturbance was measurable and significant more than 2.5 months following the flood event; this
appeared to be driven by changes in density of Leptochelia dubia, which in normal conditions comprises
approximately 60 percent of the surface population.
The researchers concluded that flood sedimentation alters benthic intertidal habitat. The deposited flood
layer persisted for more than 1 year with little physical or biological mixing, and the properties of flood
sediments were distinct from ambient intertidal sediments. The remaining benthos in flood plots may be
food limited, as indicated by a combination of high total organic carbon, deep oxygen penetration, and
slow recovery of benthic microalgae. There were significant decreases in abundance and species richness,
and depressed abundances lasted at least for the first 70 days. Additionally, the flood layer was not readily
recolonized, even by mobile species. Finally, species traits (e.g., behavior) may be important in
determining community response and resilience to rapid sedimentation disturbance events.
Nonlinear Response of Prairie Pothole Landscapes to Climate Change and Land Management
Carter Johnson, South Dakota State University
Dr. Carter Johnson stated that the overarching goal of his research project is to complete and test a new
simulation model (WETLANDSCAPE) that will examine nonlinear or threshold effects caused by climate
change and land management on complexes of glaciated prairie wetlands. The prairie pothole region in
which the study is taking place is comprised of 1 million square kilometers and has a high amount of
biodiversity; approximately two-thirds of all ducks in North America are produced in this region. There is
a north-to-south temperature gradient as well as an east-to-west precipitation gradient across the area; the
temperature gradient is the strongest. The area is comprised of three types of wetlands: temporary,
seasonal, and semipermanent. WETLANDSCAPE simulations show the differences in the water regimes
in the three wetland types. Productive prairie wetlands must cycle through four well-known vegetative
cover cycles: dry marsh, regenerating marsh, degenerating marsh, and lake marsh. Climate can be
evaluated by assessing how well the wetlands progress through the cycle; this is the basis of the Cover
Cycle Index (CCI).
WETLANDSCAPE can predict the CCI of geographic regions under different scenarios. The best
predicted climate is in the area with the fewest wetlands, and the areas that are predicted to have the best
productivity and waterfowl nesting currently are forested. Dr. Johnson showed sample graphs of CCI
plotted against warming trends. Most weather stations in the prairie pothole region have reported
productive conditions in the 20th century; only three sites have been enhanced by increased warming.
Most sites currently have optimal conditions, so any increase in temperature will decrease productivity.
Hydroperiod frequency is defined as the number of days per year there is standing water; frog populations
need at least 100 days of standing water to reach "boom" population levels. Historically, 22 years per
every 100 have a hydroperiod frequency of at least 100 days. A 2°C rise in temperature allows for only 7
"boom years" out of 100, and a 4°C rise reduces this number to 1 year. Waterfowl depend on boom years
and will not frequent wetlands that only have 1-5 boom years out of 100. Area crop types differ in their
rates of evaporation, transportation, and runoff. In areas with unutilized grass, a 41-year simulation
predicts that wetlands will dry up in approximately one-half of the years. Small grain crops cause dry
The Office of Research and Development's National Center for Environmental Research 14
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
years one-third of the time, row crops one-quarter of the time, and grazed grass one-fifth of the time. This
suggests that there may be mitigation options via farm management; simulations suggested that mitigation
can help in the 2°C warming scenarios.
Dr. Johnson displayed a conceptual map of the modeling process to determine cost-effective mitigation of
climate impacts on waterfowl productivity. Four models provide input into one another; the climate
model provides weather scenarios to WETLANDSCAPE, which provides wetland/watershed character-
istics to the mallard model, which provides waterfowl response to the economic cost-effectiveness model,
which provides the ultimate values. The challenges that the researchers faced were the expected-but-
surmountable challenges in fine-tuning and calibrating a new simulation model. Additionally, the
researchers interacted with the U.S. Forest Service, FWS, and the public. The wildlife conservation
community (federal, state, and private) is using the research findings to develop long-range plans to
mitigate for climate change effects on waterfowl, and the researchers participated in a national workshop
at Ducks Unlimited headquarters to write a white paper on waterfowl and climate change policy.
Innovative Management Options To Prevent Loss of Ecosystem Services Provided by Chinook Salmon
in California: Overcoming the Effects of Climate Change
David Purkey and Lisa Thompson, University of California, Davis
Dr. Purkey explained that most of the Chinook salmon habitat in California has been dammed, and the
last place the species thrives is in the Butte Creek Watershed; however, increasing temperatures threaten
this habitat. The long-term goals of the project are to: (1) investigate how climate change and land-use
practices change temperature and flow regimes within California watersheds, (2) determine whether these
changes will lead to a reduction in salmon habitat and a resulting reduction in salmon abundance, and
(3) determine how a reduction in salmon abundance will affect local biodiversity through food web
interactions. The goals during the first year of the project, which just finished, were to develop a water-
shed model, parameterize a baseline salmon population dynamics model, and develop a site-specific food
web conceptual model.
The Butte Creek Watershed has several subwatersheds. Two reservoirs located in adjacent watersheds
operate seasonally. Salmon visit a series of deep pools in this area in which the water temperature can be
managed. The analytical approach to the problem is to combine climate data with the Water Evaluation
And Planning System (WEAP) and Salmonid Population Model (SALMOD) models to examine tradeoffs
between freshwater services and the salmon population. The models are built with elevation bands and
information on soils and land use and cover. The WEAP software was calibrated with historical data from
1983-2003 and a model-to-model comparison. Following calibration of the hydrology, operations were
input, including infrastructure (diversions, reservoirs, powerhouses), flow requirements, and operation
rules. The overall watershed hydrologic response can be compared; the model is well representative of the
overall volumes, and other statistical indicators are within acceptable ranges. The temperature model
domain was divided into 40 subreaches characterized by a series of riffles, runs, and pools; the propor-
tional length of each of these geomorphic types was combined in the modeling assumption. There is
reasonably good calibration of the model, although a small amount of divergence occurs below the
powerhouse. Modeling indicates that potential management options include using Philbrook Lake to
modify the temperature in Butte Creek when necessary. Temperature calibration was one challenge that
the researchers faced.
Dr. Lisa Thompson explained that SALMOD is a computer model developed by the USGS that simulates
the dynamics of freshwater salmonid populations. SALMOD structure includes holding/spawning adults,
eggs and alevin, fry, 0+ parr, and 1+ parr; inputs into the model include temperature, habitat, flow,
fecundity, growth, mortality, and movement. Data sources for the model include government reports,
peer-reviewed publications, and books. SALMOD can graph relationships, including egg mortality versus
temperature, fry growth rate versus temperature, and fecundity versus weight. Preliminary output of the
The Office of Research and Development's National Center for Environmental Research 15
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
model matches what is expected to be observed within the system. Juvenile outmigration is the gold
standard to ensure that the watershed is productive; the baseline model generates reasonable juvenile
abundance, and calibration is planned with California Department of Fish and Game outmigrant trap data.
Challenges with SALMOD included program limitations and calibration. The researchers are examining
marine-derived nutrients above and below the migration barriers to determine how much benefit the
salmon currently bring to the ecosystem and will use this information as a surrogate for what would be
lost if the salmon were lost.
The researchers have interacted with clients via presentations and have worked with several international,
national, state, and local organizations. Currently, the tasks are on track, and an efficient and effective
multidisciplinary research program has been established. Stakeholders and other parties, such as resource
managers and watershed groups, are interested in the research outcomes.
Hydrologic Thresholds for Biodiversity in Semi-Arid Riparian Ecosystems: Importance of Climate
Change and Variability
Thomas Meixner, University of Arizona
Dr. Meixner explained that the three hypotheses of the research project are that: (1) decadal scale climate
change and variability alter riparian aquifer recharge through mechanisms that depend on the magnitude,
frequency, and seasonality of flooding, and exert the greatest change in reaches that receive minimal
groundwater inflow from the regional aquifer; (2) riparian vegetation structure responds nonlinearly as
riparian aquifers are dewatered and key hydrologic thresholds for survivorship of plant species are
exceeded; and (3) decadal scale climate variability and change alter riparian ecosystem water budgets that
in turn change vegetation structure and function and the ecosystem services provided to society. Riparian
ecosystems are classified as wet, moist, or dry, and biodiversity decreases as the system moves from wet
to dry. Hydrology systems generally are understood by mountain-block, mountain-front, and basin-floor
recharge; additionally, basin groundwater mixes with flood recharge. Riparian well water composition
falls between that of basin groundwater and monsoon runoff; storms and floods propagated in the system
are critical sources for approximately one-half of the water in the riparian system. The experimental
design uses a climatic gradient to understand how differences in hydrology affect vegetation and how
climate change will affect the winter storms that are a critical source of water in the system.
Three projects are planned to study the first hypothesis. The study area for the first project is a 14,000-
square-kilometer watershed in Arizona. Results indicate that downstream wells have the least evaporation
signals, and upstream wells have the most. Additionally, there is a multidecadal storage of flood waters
within the basin. The study site for the second project is the Upper San Pedro Basin in Arizona. A very
simple flood model based on vegetation was used that captures the storage of floodwater and its re-
release. The third project attempts to understand mountain recharge systems. The idea is that climate is a
driver, and empirical relationships have been developed. Dr. Meixner described the temporal discreti-
zation of the empirical model; the model works well to estimate seasonal recharge.
The second hypothesis is being investigated via a series of three projects. The first project monitored
surface flow monthly at ephemeral to perennial sites at multiple rivers, and vegetation was sampled along
the active channel. Results indicated that wetland perennial herbaceous plants show a consistent pattern of
sharp decline in abundance as stream flow becomes nonperennial; the researchers concluded that the
abundance of a key stream community type (i.e., riverine marshland) will decline with increasing aridity.
The second project examines variance through time via multiyear field monitoring of vegetation and soil
seed banks at ephemeral, intermittent, and perennial sites through the wet/dry period. Results indicate that
in years with wet winters, flood runoff sustains flows at ephemeral sites, allowing for the development of
ephemeral wetlands. Additionally, soil seed banks provide resilience and allow distinct plant communities
to develop in years with varying flow conditions, and the diversity of seed banks is influenced by proxi-
mity to perennial reach. The conclusion is that spatial distribution of wet and dry reaches influences
The Office of Research and Development's National Center for Environmental Research 16
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
vegetation response to stream flow changes. Citizen wet/dry mapping, an annual volunteer effort to map
the wet and dry reaches of San Pedro, provides critical data. Another threshold is groundwater depth and
decline of woody riparian plants, which the third project investigates via monitoring at multiple wells and
rivers and sampling woody vegetation for abundance and composition. Preliminary results indicate that
woody species, grouped by strategy type, show similar trends among rivers.
Three projects are being undertaken to study the third hypothesis. The first project investigates wet/dry
scenarios, and the second project involves a modeling approach that estimates the potential seedling
densities of riparian tree species. Preliminary results for the second project indicate that modeled densities
vary among San Pedro River sites with different stream hydrology and among years with different flow
conditions. The third project explores historic legacies by analyzing aerial photographs of the Upper San
Pedro River from 1935, 1955, 1978, and 2003 to assess temporal and spatial trends in vegetation cover-
type abundance. Results indicate that as a result of past extreme disturbance, pioneer woody vegetation
has been expanding since 1953. Recruitment events are relatively rare, but when they occur there is a
large return of trees. Riparian forest patterns are a product of interactions between recent climatic cycles
and land and water use as well as past extreme events that set in motion trajectories of change.
Currently, greenhouse studies of plant rooting depth and response to water table decline are underway.
The researchers also are classifying riparian plants into strategy groups based on response to drought and
flooding. In the course of the study, the researchers learned that flexibility is critical, and a simpler model
is preferable to a more complex model for these studies. Several presentations were made to stakeholders
and clients. The next steps are to build a seasonal groundwater model of San Pedro, develop scenarios,
and continue the greenhouse studies and classification of riparian plants.
Tier II Discussion
Dr. Jokiel showed diagrams, figures, and graphs that highlighted: (1) the projected impacts of climate
change on food, water, ecosystems, and extreme weather events and the risk of abrupt and major
irreversible changes; (2) projections based on continuing "business as usual" for emissions versus paths
for stabilizing carbon dioxide emissions to limit temperature changes; (3) irreversible climate change
resulting from carbon dioxide emissions; and (4) carbonate chemistry of coral reefs. It is important to
make the general public and decision-makers aware of all of the possible outcomes resulting from climate
change. He noted that the carbon dioxide that already has been produced cannot be retrieved.
Mr. Burney Hill asked Dr. Jokiel whether there are any species or refuges that may continue to exist in
spite of climate change. Dr. Jokiel responded that the ocean as a whole is becoming acidified, so this
would be unlikely. Caribbean reefs will disappear first, followed by Pacific reefs and the Indonesia coral
triangle; this is not a hopeless situation if it is reversed now. Because fossil fuel cannot be burned indefi-
nitely in any case, it behooves mankind to stop now rather than later and use solar, wind, and other types
of power. When people realize the potential future conditions, they will be motivated to change their
habits; alternative energy resources are vast.
Dr. Curtis Richardson of Duke University asked Dr. Johnson whether there were any efforts to determine
whether to put resources in the eastern United States versus the western United States. Dr. Johnson
responded that there were few hold-outs for maintaining the level of restoration in the West. He noted that
restoration only resulted in recovery of 1 percent of wetlands in the previous few decades. FWS is close
to making policy decisions in this area, but the decision to choose East versus West is politicized.
Dr. Richardson noted that waterfowl populations were not the only issue in the West; water retention and
loss is a significant issue, as are landscape problems. Dr. Thompson asked whether efforts could move
north to Canada. Dr. Johnson responded that northern Canada appeared to be affected most, but there is
hope in the aspen parklands; there may be some potential if the current drainage in the area ceases and
restoration begins.
Th e Office of Research and Development's National Center for Environmental Research 17
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
Dr. Purkey asked Dr. Wardrop, in terms of scaling up from headwater systems, what the plans were to
introduce the water management dynamic into the analysis. Dr. Wardrop responded that most of the
management already is in the headwaters, with large management projects in place, as the headwaters are
the major producers of ecosystem services in the system. Dr. Purkey asked Dr. Wardrop whether the
model allows for pumping scenarios, and Dr. Wardrop replied that it does. A participant asked whether
soil mechanisms are included in the model. Dr. Wardrop answered that soils are a complicated geology,
but they are included.
Dr. Wardrop noted that condition assessments in her watershed indicate that more than one-half of the
wetlands are in fair or degraded condition and asked whether other researchers were making the argument
that the resources are in an impaired condition and any actions that improve the resource condition can
help in response to climate change. Dr. Meixner responded that in the San Pedro Basin the argument has
been made that current usage will drain the basin eventually, and climate change will worsen this
condition. Resource managers, however, do not know how to respond to this argument, and people are
nonresponsive to using to less water. Dr. Purkey reported that the same response was seen to the news
that the salmon habitat in California is very vulnerable.
Dr. Derek Poon of Region 10 stated that the question is what strategies can be undertaken if resources and
watersheds are degraded before climate change effects are seen. This is a very difficult question, not
answered in many places; the answer will not come from a piecemeal approach.
Ms. Susan Julius, EPA, noted that EPA Oregon produced a report on climate adaptations, including a
series of management activities that could be undertaken for various ecosystems. There must be a
paradigm shift that includes short- and long-term strategies. Solutions will be different for different areas
and even for different seasons within the same area.
Dr. Richardson noted that the European climate change program is approximately 5 years ahead of the
United States' and also struggles with how to separate climate change effects from disturbance. There
may be synergies that can be exploited. He noted that projects need to be brought to the user groups to
encourage forward progress. As some government policies and recommendations have been disastrous in
the past, governments often are cautious about making recommendations; sound science is necessary to
ensure the proper actions and recommendations.
Dr. Jason Rohr asked whether uncertainty had been considered and sensitivity analyses had been
performed in the mathematically based risk assessment models. Dr. Purkey noted that his group still is in
the model-building phase, but these will be included. Climate uncertainties are relatively easy to capture;
it is necessary to identify areas in which there is acceptable uncertainty and find stability. Dr. Wardrop
agreed and added that the concern is that no uncertainty analysis will mitigate a bad scientific decision. It
is necessary when using external tools to determine whether they are ecologically relevant to the current
work. It is important, as Albert Einstein noted, to define the question to be solved; sometimes researchers
attempt to answer too many questions. Dr. Rybczyk added that uncertainty and sensitivity analyses are
relatively easy to perform in ecogeomorphic models; many of the problems are in the initiation,
calibration, and validation, especially when scale is increased.
Dr. Haider noted that, as the only social scientist present at the workshop, he had a unique perspective. He
stated that people are responsible for adaptation; the general public and policy-makers need to be
convinced of the consequences of climate change. There are many organizations in various fields that
communicate this idea to the public; the information that the researchers at the workshop generate can
feed into this decision process, but it needs to be translated. EPA's RFAs must include a social science
component to interpret, translate, and determine how to proceed once the basic science information is
generated.
The Office of Research and Development's National Center for Environmental Research 18
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
Dr. Jones explained that EPA's National Center for Environmental Economics has a Web site that can be
searched for information, reports and other publications, and workshops regarding valuations of climate
change, ecological benefits, and so forth. EPA realizes that many different disciplines need to be a part of
these types of decisions. Scientific knowledge takes time to acquire, but some decisions need to be made
quickly. The role of scientists is to provide data and tools to the decision-makers, and scientists can
suggest workshops that are necessary to move forward.
Dr. Purkey described a grant with Google to use Google Earth technology to understand climate change
as well as collaborative work with the University of Kent that uses the same approach that Dr. Haider
suggested. Dr. Wardrop added that it is necessary to match the scale of the social and physical sciences to
the decision-making scale. Dr. Meixner stated that it is critical to consider that decision-makers often do
not make rational decisions based on science but rather decisions based on instinct. Dr. Poon added that
EPA does not have land-use authority and cautioned not to lose sight of the fact that local government
changes are critical.
Dr. Jones thanked the participants for a productive discussion and recessed the meeting at 4:41 p.m.
DAY 2: MAY 28, 2009
Before the presentations began, Dr. Jones asked participants to provide input to the Agency regarding
how to best use their research. Dr. Wardrop noted that it would be helpful for EPA to include in its RFAs
more direction and guidance regarding how to ensure that the research can be translated and applied. Dr.
Jones noted that he would take this suggestion back to his office and explained that the new management
structure has a new focus on outreach and communication.
Integrated Bioclimatic-Dynamic Modeling of Climate Change Impacts on Agricultural and Invasive
Plant Distributions in the United States
Wei Gao, Colorado State University, and Xin-Zhong Liang, University of Illinois at
Urbana-Champaign
Dr. Wei Gao explained that biological invasions of nonindigenous species are serious threats to natural
and managed ecosystems, causing approximately $120 billion in damage annually. Additionally, the rapid
growth in trade worldwide and globalization exacerbates the United States' invasive species problems.
Climate is the dominant determinant of the geographic distribution of native or invasive plant species, and
climate change already has caused unequivocal shifts in distributions and abundances of species and
pushed certain native species to extinction. The overall objectives of this project are to: (1) better under-
stand how global climate changes affect U.S. agricultural and invasive plant species distributions with a
focus on crop production; and (2) account for adaptation of alternative crops and invasion of normative
species to enable decision-makers to design effective management and control strategies for a sustainable
future agro-ecosystem. The proposed research will: (1) develop a robust species environmental matching
model to best capture the observed agricultural and invasive plant species distribution using the
conditions from a climate-ecosystem predictive model; (2) make projections for the potential niche
distributions of alternative crops adaptable to the likely range of climate changes in the future using the
climate-ecosystem predictive model; and (3) project the geographic distribution and abundance of U.S.
agricultural weeds and invasive plant species by integrating a newly developed species environmental
matching model and future soil and bioclimatic conditions simulated by the climate-ecosystem predictive
model.
Dr. Gao introduced Dr. Xin-Zhong Liang, who explained that most general circulation models (GCMs)
for climate prediction are based on IPCC models, but the researchers on this project used a model
computational domain design to regionally downscale global model projections. The results of an assess-
ment of the northeastern United States showed that the researchers' model predictions were closer to
The Office of Research and Development's National Center for Environmental Research 19
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
actual precipitation and temperature than IPCC's model; the IPCC model was even less accurate when
predicting mean summer precipitation because of the biases present. The regional model can reduce
biases, and combining models into a mosaic ecosystem model further reduces biases. A dynamic crop
growth model can be integrated with satellite remote sensing to predict annual yields and help regulate
market supply-demand, make strategic assessment of optimal operation practices, and project food trends
as a result of climate change. The species environmental matching method relates observed species
distributions to environmental envelopes with the assumption that the fitted observational relationships
are an adequate representation of the realized niche of the species under a stable equilibrium or quasi-
equilibrium constant. Cheatgrass was used in the study because it is not native to Washington State.
Cheatgrass entered the United States through Washington State from Eurasia during the 19th century and
now is widely distributed throughout the mainland United States, with the exception of Florida; no insects
are available to control its spread. The presence of cheatgrass throughout several states was examined,
and the modeling results predict that increasing temperatures will significantly increase the amount of
cheatgrass.
Because it is important to determine how crop productivity will be affected by climate change, the
researchers will examine the possible distribution of crops in the future. Additionally, another STAR
grant, which will begin in August 2009, was secured to study water quality. Water quality and agricultural
impacts on the United States following climate change will be studied. The next steps in the current grant
are to expand the modeling system to predict most major crops, incorporate multisubgrids of land use and
land cover, develop the capabilities to simulate air pollution impacts on crops, study agriculture water
quality problems, and study the agro-ecosystem carbon cycle.
Global Change and the Cryptic Invasion by Transgenes of Native and Weedy Species
Cynthia Sagers, University of Arkansas
Dr. Cynthia Sagers noted that the formal definition of atransgene is neutral (i.e., it does not imply "good"
or "bad"); it simply is defined as a gene from one species that has been introduced into the genome of
another species through biotechnology. Because of increasing global population, increases in the quality
and quantity of food are necessary, and biotechnology can introduce beneficial traits into existing crop
systems to help with this goal. Agricultural systems have a marked influence on native species, and there
is evidence for crop-to-weed gene flow and hybridization with native species. Evolution of crop and weed
systems ensures sexual compatibility between native species and crops in some part of their range. The
inevitable transgene flow from crops to weeds and natives is a serious issue because it can introduce
herbicide resistance and result in aggressive weeds. Factors that support gene flow are coexistence, sexual
compatibility, hybrid vigor, and selective benefit. The manner by which native and weedy species will
respond to climate and land-use change and whether the likelihood of transgene escape shifts with these
changes are important issues.
Canola has remarkable genetics and was the result of crossing a weed with cabbage in Canada. It is a
robust crop that is becoming increasingly important as alternative food sources are explored. Because
canola is sexually compatible with least 44 brassicaceous species, it is inevitable that it will expand into
the wild. The number of areas in the United States in which canola is cultivated has grown exponentially
since 1992 because of its increased use as a biodiesel and nontrans-fat cooking oil, among other reasons.
Additionally, it spontaneously hybridizes with congener Brassica rapa.
This project began in Oregon with an EPA National Health and Environmental Research Laboratory
project that studied insect resistance in canola. The project utilizes green fluorescent protein to determine
whether the plants carry the transgene. There are three levels of competition between the parentals and
hybrids. The conclusion of the research is that risk of transgene flow is a function of genetic background,
competition, and level of selection.
The Office of Research and Development's National Center for Environmental Research 20
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
A new project in North Dakota has begun that utilizes greenhouse and field work to determine how
climate and land-use changes will influence the adventitious presence of transgenes. The objectives of
this project are to: (1) characterize variability among weedy populations in traits related to outcrossing;
and (2) incorporate these parameters into existing climate and land-use change models to assess the
changing risk of transgene flow. Three different sexually compatible weeds— B. rapa L., B. nigra L., and
Sinapis arvensis L.—that are predominant throughout North America will be examined. The project will
begin with weed studies, the objectives of which are to map local distributions, monitor transgene flow,
and model risk. The objectives of the sentinel plant study are to measure transgene flow and assess
geographic variation in gene flow rate. The greenhouse study will evaluate genetic variability of
functional traits among B. rapa populations and measure pollinator preference in a controlled environ-
ment. The objectives of the modeling are to develop phenological maps for sexually compatible relatives
and create a probabilistic model of the changing risks of transgene flow. Transgene bans create problems,
and the policy requires attention, especially given that three different federal agencies regulate transgenic
species.
A Multiscale Approach to the Forecast of Potential Distributions of Invasive Plant Species
John Silander, University of Connecticut
Dr. John Silander explained that the New England area is an invaded landscape that is dominated by
woody bird-dispersed species. There have been 111 invasive plant species identified in New England, the
vast majority from East Asia or Eurasia. Of these, the most pervasive are woody invasives that are native
to East Asia. The majority of invasion sites are dominated by 18 percent of all fleshy-fruited, bird-
dispersed invasive species.
A primary objective of the project has been to predict the areas in which invasive species potentially will
spread in the regional landscape. The project's approach to modeling potential distribution is to use
spatially explicit hierarchical Bayesian models based on a prior U.S. Department of Agriculture (USDA)
project that incorporate many different explanatory variables (e.g., climate, site, land use). The response
variable is presence/absence data regarding Celastrus orbiculatus, which is a woody liana native to East
Asia that is found in edge habitats; native presence/absence data from Japan and New England and Japan
climate data layers also will be incorporated. The characteristics of the local field survey sites include
habitat type and canopy closure. The potential distribution of C. orbiculatus is predicted as a function of
climate, habitat, canopy, and land use/land change using four models: (1) New England climate only,
(2) Japanese climate only, (3) New England and Japanese climates, and (4) New England and Japanese
climates with local habitats and land use/land change. The researchers examined which models had the
best fit and verified the results by comparisons with independent herbarium records. The best model fits
include climate variables from New England and Japan, land use/land change, and local site character-
istics. Factors across species that influence invasive species richness at a site include positive (edges and
open canopies, road density, deciduous forests, and warmer summers) and negative (conifer forests and
active agriculture) factors.
Results indicated that constantly forested landscapes (i.e., land-use characteristics) discourage the occur-
rence of invasive species. The researchers examined 70 years of digital land-use change from aerial
photographs, satellite images, and groundtruthing and determined that land-use history is critical to
predicting the distribution of invasive species in New England. Because these are static models, however,
the goal was to develop a model that would determine the manner by which invasive plant species spread
across the landscape over time. The species tend to be distributed by birds, so the question was whether
birds, particularly the invasive European starling introduced in New York from Europe in the 19th
century, are assisting with the spread of invasive plant species. Is there mutual spread across the region?
Results indicate that there appears to be parallel, joint spread, with the birds arriving in advance of C.
orbiculatus. When the feeding choice behavior of the starlings was examined, it was determined that they
The Office of Research and Development's National Center for Environmental Research 21
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
prefer invasive fruits. The birds travel long distances, up to 200 km per day, and can disperse seeds over a
wide area. The modeling approach to determine how birds respond to landscape characteristics was to
develop a cellular automaton model of the dispersal and growth of Celastrus across the New England
region using grids of five landscape types. The model was evaluated and seeded, and results indicated that
the model was more accurate when local and long-distance dispersals were used. The model provides
historical, present, and future predictions of spread. Future predictions indicate that the landscape does
not fill because conifer forests have a blocking feature against the spread of these invasive species.
Simplifying the model using binary landscapes shows a slightly poorer spread and performance than the
more fully specified landscape heterogeneity. Using uniform landscape scenarios, the landscape fills over
time; however, in random landscape scenarios, the landscape does not fill with Celastrus. Additionally,
there appears to be a 30-40 year lag between the spread of the invasive starlings and Celastrus.
The researchers concluded that the most pervasive invasive plant species in New England tend to be
woody and with bird-dispersed fruits. Hierarchical Bayesian models provide accurate, static predictions of
the potential distributions of species using climate, land use, and local site traits as explanatory variables.
Native range data combined with invaded range data are critical to accurate predictions. Models
calibrated from invasive plant (Celastrus) demographic data and starling movement yield predictions that
agree with the observed spread of invasives over space and time. Finally, regional land-use patterns are
critical to the patterns of spread of both invasive plants and starlings.
Predicting Risk Invasion by Salt Cedar and Mud Snails
Leroy Poff, Colorado State University
Dr. Leroy Poff explained that climate change is likely to enhance the spread of invasive species in river
ecosystems, which is particularly important because these species can alter ecosystem structure and
function, contribute to native species declines, and cause economic damage. The goal is to understand
how climate change influences the spread of these species, as local factors will determine the success of
invasion. The working hypothesis is that within the thermally suitable envelope, local invasion success
will be dictated by habitat suitability and dynamics (e.g., hydrologic, geomorphic) and biotic factors,
which can be modeled at the ecologically relevant scales to establish the likelihood of success. Human
responses to climate change must be accounted for because they will contribute to the risk of invasion.
The challenge of the project is scaling the problem, and the project framework uses a hydrogeomorphic
template with the idea that species population success is a function of magnitude, frequency, timing, and
duration of flow events that limit establishment success or cause mortality. The key point is that effective-
ness of flow regime varies with geomorphic settings. The research plan is to combine flow regime and
geomorphic setting (i.e., natural disturbance regime) to explain current distribution of nonnative salt cedar
and New Zealand mud snail and project the future likelihood of invasion. The specific goals of the project
are to: (1) develop an ecological response model to explain the current distribution and dominance of two
invasive species across the interior West in terms of climatically driven, local-scale environmental
drivers; (2) use downscaled projections of regional climate change to describe possible future streamflow
regimes and incorporate the effect of water management on those future flow regimes in a geographic
region of the western United States; (3) disaggregate the subbasin-scale flow regime output from the
WEAP model and construct reach-scale flow regimes for the drainage network in the entire region;
(4) use the ecological response model to examine the risk of invasion for river reaches throughout the
region for different combinations of climate change scenarios and modes of dam operations in a geomor-
phic context; and (5) model long-term invasion success for the two study species under interannual flow
regimes, representing a range of hydrogeomorphic settings.
One hypothesis under Goal 1 is that the current distribution and abundance of salt cedar and New Zealand
mud snail can be explained statistically in terms of site-scale hydrogeomorphic setting and dynamics. The
second hypothesis is that the probability of species occurrence or dominance at a site will reflect a hydro-
The Office of Research and Development's National Center for Environmental Research 22
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
geomorphic threshold. To identify the study region, researchers used GCMs to identify areas in which
temperature changes will promote salt cedar range expansion and then overlaid these areas with areas on
the edge of the current salt cedar range. The region of conservation and management concern chosen is an
area surrounding the Green, Yampa, and White Rivers in Wyoming, Utah, and Colorado. The Goal 2
hypothesis is that the WEAP modeling platform can be used to generate subbasin scale and weekly flow
regimes and infer the effects of dam operations on natural flow regimes for subbasins in the region. The
researchers, based on lessons learned, probably will not pursue Goal 3 but will use WEAP instead. To
accomplish Goal 4, an empirical ecological response model will be applied to develop risk-based
predictions of invasion risk throughout the region. Under the model, the geomorphic base map combined
with the flow regime base map results in an invasion risk map. The hypothesis regarding Goal 5 is that
stochastic population dynamics models can estimate year-to-year population sizes based on reach
geomorphology and long-term (projected) flow regime and therefore assess the long-term viability of
nonnative species.
There are several expected outcomes of the project, including a more mechanistic (dynamic) and appro-
priately scaled basis for projecting invasion risk, a risk map, a framework for thinking about the spatial
distribution of threats and how to contemplate proactive management, and possibly the future inclusion of
social processes to examine cost-benefits of spatially distributed water management. The challenges have
been projecting ecological response models for salt cedar and New Zealand mud snails that can be applied
to future environmental conditions, scaling climate and hydrologic models to match ecological response
and measurement scales, and representing risk in a robust manner that allows for linked multimodel
uncertainties. The researchers have interacted with the U.S. Bureau of Reclamation and The Nature Con-
servancy and have planned discussions with Wyoming and Colorado state agencies. Currently, the
researchers are developing a WEAP model for the upper Green and Yampa River Basins that eventually
can be used to address a number of water management issues in these basins and have generated interest
from nongovernmental organizations and state and federal agencies.
Integrating Future Climate Change and Riparian Land Use To Forecast the Effects of Stream
Warming on Species Invasions and Their Impacts on Native Salmonids
Julian Olden, University of Washington
Dr. Julian Olden explained that the project began in September 2008. The prospect of dramatic climate
change during the next century underscores the need for innovative science and new decision-support
tools for efficiently managing freshwater ecosystems. Climate-induced changes in the geomorphic and
physical processes that drive stream ecosystems in the Pacific Northwest are imminent. Cumulative
effects and complex interactions among multiple agents of environmental change may limit the success of
current and future river management efforts. Climate changes will have direct implications for stream
temperatures, which are exacerbated by the removal or alteration of riparian habitat by logging and gra-
zing that reduces shading and modifies channel morphology. Elevated stream temperature is one of the
most pervasive water quality issues threatening freshwater ecosystems in the Pacific Northwest. Manage-
ment efforts are further complicated by the fact that Pacific salmon now share the riverine landscape with
a number of nonnative fish species, and significant shifts in species ranges and the outcome of biological
interactions are highly possible. The goals of the project are to: (1) determine the predicted effects of
regional climate change and local riparian management on riverine thermal regime; (2) investigate how
Chinook salmon, smallmouth bass, and northern pikeminnow will respond to projected temperature
changes; and (3) ascertain the critical areas for riparian restoration and protection to mitigate the negative
ecological impacts of climate-induced stream warming in the future.
The study site is unregulated, and land use and resource extraction within the site vary longitudinally. It
contains one of the few remaining wild spring Chinook salmon runs in the Columbia River Basin as well
as an active region of upstream invasion by smallmouth bass and northern pikeminnow. The researchers
are utilizing a combination of a GCM, land cover, geomorphology, and stream thermal regimes to deve-
The Office of Research and Development's National Center for Environmental Research 23
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
lop a stream temperature model, which will be used to run future riparian vegetation scenarios to
determine future stream thermal regimes. A field study will be performed to validate the model. The ulti-
mate goal is to determine how Chinook salmon are affected.
To develop climate change projections of temperature and precipitation, the researchers are downscaling
simulated future climate data from a suite of GCMs under three greenhouse gas emissions scenarios for
decadal time periods from 2020-2100. Channel reach morphologies will be classified to describe the
thermal sensitivity of stream reaches to changes in temperature and riparian vegetation cover, and reach
classifications will be based on a hierarchical scheme that accounts for differences in valley fill, degree of
channel incision, and channel pattern. Thermal regimes will be quantified using a network of digital
temperature loggers at point locations, and thermal imagery will be used to map spatially continuous
longitudinal patterns of stream temperature. The Heat Source Model Version 7 allows for the simulation
of water temperature at the reach scale using high-resolution spatially continuous data coupled with
deterministic modeling of hydrologic and landscape processes, allowing the development of a spatially
explicit stream temperature model. To forecast thermal regimes under scenarios of climate change and
land-use management, future spatiotemporal patterns in stream temperature will be predicted according to
scenarios of projected climate change and riparian land use. To model ecological responses to future
thermal regimes, fish species responses to climate change and riparian management will be estimated
according to thermal preferences and tolerances, and a number of additional key temperature benchmarks
will be explored. Field surveys will be completed to verify the models.
The project findings will help guide management strategies and policy aimed at minimizing the future
range expansion of invasive species through protection and restoration of riparian vegetation that creates
and maintains cool-water habitats. Results from this project will make it possible to rank stream segments
in terms of their ability to mediate the effects of climate change on stream temperatures, create suitable
thermal habitat that favors native species over invasive species, and establish thermal barriers to prevent
upstream invasion. Management portfolios based on different ecological endpoints will be distributed to
local and regional agencies and nongovernmental organizations. Products from the project will be integra-
ted into a graphical user interface providing the user with animated maps and timelines of stream
temperature change, salmon habitat availability, and bass and pikeminnow spread for a given climate
change or land-use scenario or the option to export data for quantitative analysis. The challenges include
the issue of continuous land access, incorporating climate-induced vegetation change into stream temp-
erature modeling, and preparing managers for the possibility of implementing unconventional strategies.
The researchers have interacted with several local, state, and federal agencies and nongovernmental
organizations.
Elevated Temperature and Land Use Flood Frequency Alteration Effects on Rates of Invasive and
Native Species Interactions in Freshwater Floodplain Wetlands
Curtis Richardson, Duke University
Dr. Curtis Richardson noted that global climate change and freshwater ecosystem studies and models
suggest two key findings: Water temperatures will increase approximately 2-4°C, and the frequency and
intensity of high flow stream events will increase. The question is what the implications will be of warmer
water and altered hydroperiod on the establishment, abundance, and distribution of invasive species in
river floodplain ecosystems. A likely future scenario is that stream ecosystems in the southeastern United
States will experience lower baseflows with more extended drought periods punctuated by more frequent
and intense storm events. Another likely future scenario is that freshwater wetlands in the southeastern
United States will be inundated less each year than currently and undergo a greater number of rapid wet-
ting and drying cycles as a result of extreme events.
The goals of the project are to: (1) quantify the effects of elevated wetland water temperature and pulsed
water on rates of species invasion patterns of sediment and nutrient retention services; (2) assess how
The Office of Research and Development's National Center for Environmental Research 24
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
species richness, diversity, productivity, and invasibility change under varying water temperature
regimes; and (3) determine how interactions between hydrology and temperature affected the current
community composition/invasibility of floodplain ecosystems in the southeastern United States at the
regional scale. There are two experimental levels to the project. The first is site-specific and involves the
role of plant diversity on invasive species, pulsed water effects on wetland species, and elevated
temperature and pulsed water in controlled wetlands. The second experimental level involves regional
floodplain hydrology and temperature shifts in naturally occurring wetlands. The study site is in the Cape
Fear River Basin and includes 10,000 acres of forest and wetland on the Duke University campus and 24
hectares of wetland on the edge of the main Duke University campus. The three phases of the project are:
(1) stream reconfiguration, (2) dam and impoundment, and (3) treatment wetland. A fourth phase, stream
and floodplain restoration, is pending. There are 99 study plots, which were used in a previous study
regarding the effect of diversity on ecosystem functions. The current study will examine the role of
hydroperiod shifts and water pulses on diversity and wetland functions. Previous results indicated that the
plots with the highest diversity prevented invasive species. Pulses are related to diversity and invasive
species, with areas of high and low marshes.
Results that show the fluctuation of plant species in high and low marsh conditions indicate that high
marshes decrease species diversity, and low marshes increase species diversity; invasive species appear to
be intolerant of flooding. To accomplish the third phase, a Weir system is used to divert stormwater and
raise water temperature. To accomplish the regional experimental level, nine flood plains sites located on
rivers throughout North Carolina and southern Virginia were identified; some have a 5-6°C temperature
gradient downstream of dams. Criteria for choosing the site include the presence of mountain headwaters,
a high degree of hydrologic connectivity, and similar nutrient regimes; the reference sites must have no
upstream dams. The researchers asked whether water temperature relates to species variety; there is no
pattern of difference in water quality.
Results indicate that there is a large number of nonnative, invasive species. The most frequently found
species at each type of site included at least one nonnative, invasive species, with the exception of near-
shore reference sites, in which they were absent. Diversity indices indicate that diversity increases in
warm conditions when species richness, number of invasive species, and percent invasive species were
examined. The projected outcomes of the project are to: (1) provide data that will quantify climate change
effects of temperature and pulsed water on invasive species in wetlands and provide information on
community structure shifts, (2) explicitly link hydrographic variation and elevated temperature with
ecological functions, (3) identify specific hydrologic and biogeochemical characteristics of floodplains
that enhance or inhibit establishment of invasive species, (4) identify feedbacks between invasive species
and ecosystems processes that alter the invasibility of floodplain ecosystems, (5) identify potential
management strategies for controlling invasive species, and (6) validate a new quantitative modeling
approach to evaluate shifts in linear or nonlinear thresholds of invasive species. The researchers learned
that invasive species definitions vary greatly, threshold responses to disturbance may vary by season, and
it is difficult to separate out pulsed water effects from temperature effects at the regional scale.
Additionally, mesocosm scale studies will allow for more temperature and water control to help in effect
studies but are difficult to set up and maintain; scaling is an issue. Separating out environmental
disturbance from climate change effects is difficult and will require new approaches to threshold analysis
to augment Bayesian threshold analysis. The researchers have interacted with a number of international
institutions and federal agencies and have presented at several national scientific conferences.
Climate Change: Pathogens and Decline of Ectotherms
Jason Rohr, University of South Florida
Dr. Rohr explained amphibians are highly threatened, and there has been global enigmatic amphibian
decline. The project focuses on amphibian diseases because many enigmatic extinctions are thought to be
a result of infectious disease, often chytridiomycosis caused by the fungus Batrachochytrium dendro-
The Office of Research and Development's National Center for Environmental Research 25
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
batidis (Bd). The fungus has been implicated in hundreds of amphibian extinctions during the last four
decades and is thought to be the most deadly invasive species on the planet behind humans. There is some
evidence that 5J-related declines are linked to climate change. Additionally, amphibian declines are
parallel to reptilian declines.
The researchers have an interest in climate variations because it is hypothesized that the ectotherm
immune system is temperature dependent. If immunity lags behind temperature change, then increased
temperature variability associated with climate change could make ectotherms more susceptible to patho-
gens. This hypothesis was tested via a seasonal field survey of newt immune parameters, and results
showed that seasons with dramatic temperature changes coincided with suboptimal immunity. The
greatest level of suboptimal immunity was at the monthly time scale, which led to the question of whether
variability in temperature at the monthly scale explains widespread amphibian extinctions putatively
caused by disease and, if so, how this predictor compares with other hypothesized predictors. The goal of
the project is to use a weight-of-evidence approach to evaluate the level of support for the hypothesis that
temperature variability facilitates parasite invasions in ectothermic hosts and subsequent host declines.
The project focused on the genus Atelopus because 71 of 113 Atelopus species are presumed extinct,
theoretically as a result of chytridiomycosis; most of these extinctions have occurred since 1980.
Spatiotemporal data on the extinctions are available for the last year species were observed and the year
they were thought to decline. There are four contrasting hypotheses regarding enigmatic/Bd-related
declines. A spatiotemporal hypothesis is that declines are caused by the introduction and spread of Bd,
independent of climate. Three of the hypotheses are based on climate—the chytrid-thermal-optimum
hypothesis, the mean-climate hypothesis, and the climate variability hypothesis.
The objective of the study is to evaluate the level of support for the spatiotemporal-spread and climate-
based hypotheses using published Atelopus extinction (i.e., last year observed) data. The first question the
researchers addressed was whether there is a spatial structure to the timing of the Atelopus extinctions.
Mantel's Test and Bayesian model averaging were used to identify parsimonious locations of Bd intro-
duction and subsequent spread. An evidentiary path spreading through the environment was found that
supports the spatiotemporal theory, and the extinctions through time followed classic disease dynamics
and were consistent with a spatially spread epidemic. The researchers then investigated the climate-based
hypotheses, the ultimate hypothesis of which is that the El Nino-Southern Oscillation (El Nino) drives
amphibian declines. The researchers asked whether it is necessary to control for the density-dependent
spatiotemporal spread when testing climatic hypotheses and detemined that it is. The years of amphibian
decline and last years observed match well with El Nino, suggesting a strong correlation. The researchers
then investigated what features of El Nino years are associated with amphibian declines by examining
regional predictors (e.g., cloud cover, temperature-dependent Bd growth, precipitation, temperature) with
and without a 1-year lag. When the univariate predictors were examined, none were significant without a
1-year lag; with a 1-year lag, however, several univariate predictors became significant, including mean
absolute value of monthly differences in temperature, Bd growth (negative predictor), frost frequency,
precipitation, temperaturemax, and wet day frequency. Results of best subset model selection indicate that
mean absolute value of monthly differences in temperature and diurnal temperature range are significant
and consistent with climate-based hypotheses.
The researchers investigated whether monthly temperature variability explained Atelopus extinctions by
examining data through time and found that there was a significant correlation between monthly
variations in temperature and extinction. Amphibian extinctions often have occurred in warmer years, at
higher elevations, and during cooler seasons; therefore, the researchers asked whether monthly variability
in temperature also is greater at these times and locations. Results confirmed that these times and
locations have greater monthly variability in temperature. Because the belief is that El Nino increases
temperature range and month-to-month variability, the researchers ran an experimental test with Cuban
tree frogs and Bd. The researchers were curious about the finding of negative relationships between
temperature-dependent Bd growth in culture and amphibian extinctions. Frogs die more frequently at cold
The Office of Research and Development's National Center for Environmental Research 26
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
temperatures, although Bd growth increases in culture at warm temperatures; therefore, it is necessary to
examine the interaction between the host and pathogen to understand the extinctions. The researchers
found that temperature variability increases Bd loads on frogs.
The temporal pattern of extinctions is consistent with a density-dependent spreading epidemic, and there
is a strong El Nino signature after controlling for density-dependent spread. The pattern of extinctions
appears more congruent with the climate-variability hypothesis than with the chytrid-thermal-optimum or
mean-climate hypotheses; experiments support the climate-variability hypothesis. The researchers plan to
quantify the impact of diurnal variability on Bd spread and virulence, test the climatic variability hypothe-
sis on other ectothermic hosts and pathogens, and test whether temperature variability can explain global
ectothermic declines.
Beach Grass Invasions and Coastal Flood Protection: Forecasting the Effects of Climate Change on
Coastal Vulnerability
Eric Seabloom, Oregon State University
Dr. Eric Seabloom explained that dunes in the Pacific Northwest are unique, understudied systems, and
the physical environment is strongly shaped and influenced by plants; therefore, dune grass is an
important species. Prior to 1900, beaches and dunes were sparsely vegetated, with little grass and shifting
sand; since then, there has been a history of dune grass invasions on the Pacific Coast. European beach
grass was introduced in 1900 and dominated along the West Coast from Canada to Mexico by the 1950s;
American beach grass was introduced from the East Coast in the 1930s. Dunes now have a gradient of
landscapes: ocean, foredune, deflation plane, beach grass hummocks, and transverse ridges. The foredune,
intentionally created to enhance protection, increases coastal protection from waves, wind, and possible
tsunamis and increases land stabilization for development behind the foredune. Unintended consequences
of foredunes include redistribution of sand, a decline in some species of native plants and animals, and
increased invasion of other species. A balance between protecting the coast and decreasing extinctions is
needed as climate change accelerates sea level rise and increases storm intensities.
Climate change affects sea level rise and wave environment, which in turn affect risk of flooding, sedi-
ment supply, dune morphology, and species invasion, the latter two of which are targets for conservation
management. There is a complex set of interactions between biological conditions and mandated manage-
ment. The objectives of the project are to determine the effects of: (1) climate change on beach grass
invasion, (2) beach grass invasion on the ability of dunes to mediate risk of climate change, and (3) exotic
grass management on the ability of dunes to mediate risk of climate change. Simulation models to
estimate a range of likely sediment budgets under expected climate change regimes and field experiments
to determine the outcome of invasions under predicted sediment budgets will be used to accomplish
Objective 1. To accomplish Objective 2, field surveys and LIDAR will be used to determine the effects of
species invasion on dune morphology, and simulations modeling will be used to determine the effects of
dune morphology on risk under various climate change scenarios. To accomplish the last objective, field
surveys and LIDAR will be used to determine the effects of conservation management on species
invasion and dune morphology.
In examining how sand supply rate affects species interaction, the researchers found that sand deposition
can alter competitive interaction among native and exotic dune grasses. This potential change in sand
supply has the potential to change species populations and distribution, which has strong implications for
dune size. Dunes dominated by the secondary invader (American beach grass) are 40 percent lower than
those dominated by the current invader (European beach grass), which has obvious implications for
coastal protection. During the previous 20 years, there has been a change in dominance; as American
beach grass increases, European beach grass decreases. There has been no change in native beach grass.
The American beach grass has moved from Washington State to Oregon in the past 20 years, and there is
The Office of Research and Development's National Center for Environmental Research 27
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
the potential for it to spread along the entire West Coast. Continued domination of American beach grass
will result in a decrease in dune height along the coast.
A challenge of the project will be to determine the shoreline from LIDAR surveys and create a risk map
for the entire Washington State/Oregon coastline. The researchers have interacted with local and state
clients from Washington State and Oregon and will participate in a 2011 meeting with land managers and
researchers. Thus far, the researchers have concluded that it is probably that changing sediment loads
resulting from climate change will alter the composition of the dune grass community, and a rapidly
spreading invasive dune grass likely is lowering dune heights and reducing their ability to protect coastal
communities. Finally, exotic grass management will require careful balancing to preserve endangered
species and a coastal protection function.
Ecological Impacts From the Interactions of Climate Change, Land Use Change, and Invasive Species
Robert Whitlatch, University of Connecticut
Dr. Whitlatch stated that the project's system focuses on shallow, subtidal, sessile invertebrate commu-
nities, which involve multiple taxa, diverse life histories, and economically important species. Following
20 years of study of these communities, researchers have determined that in New England there are four
different community states that are dominated by different factors. There are a variety of forcing functions
that occur on a variety of spatiotemporal scales that contribute to the movement of one community state
to another. During the past 30 years, nonnative ascidians have become a dominant component of southern
New England's fouling community, defined as a group of organisms that grow on hard substrata in
marine environments. Nonnative marine species must be considered because of their detrimental effects
on native species, biodiversity, habitats, and ecosystem services. They can impact commercially or
economically important species and man-made structures. Marine biofouling results in world-wide da-
mages of approximately $50 billion annually and regularly contributes a majority of the total production
costs of marine aquaculture operations.
Long Island Sound temperatures during the past 20 years have shown an increasing trend of large
interannual variation and increased temperature. Rising winter temperatures increase the recruitment
abundance of recent invaders and decrease the recruitment abundance of resident species. When the
timing of recruitment of nonnative and resident species in relation to interannual variations in seawater
temperatures was examined in Long Island Sound, it was determined that invaders respond to increased
water temperatures by recruiting earlier, but native species do not. In coastal Connecticut waters,
however, there tends to be an inverse relationship between the occurrence of invasive species and resident
species. Native biodiversity is important because habitats with higher diversity of resident species appear
less vulnerable to invasion. Coastal Connecticut has varied land use, with different areas being primarily
industrial, residential, or rural. Organisms respond to variations in land use; primarily industrial areas of
the coastline have dominant numbers of invasive species, whereas rural areas have dominant numbers of
native species.
The goals of the project are to: (1) work with environmental managers and other stakeholders on different
management scenarios for land-use planning in the context of climate change and invasive species,
(2) conduct mesocosm experiments examining the interactions of climate change and land use and the
interactions between them in altering the ability of invasive species to influence native communities,
(3) conduct field experiments to assess temporal and spatial scales of potential efforts needed to manage
invasive species, (4) conduct field experiments to examine the survival of key predators on invasive
species and how it varies with land use, and (5) develop predictive models to assess potential alternative
management strategies to evaluate multiple stressors at different spatial and temporal scales in different
types of coastal systems. Clients are embedded in the project goals.
The Office of Research and Development's National Center for Environmental Research 28
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
To accomplish Goal 1, the researchers established a project management advisory board, conducted a
workshop with various managers and stakeholders, met with local planning and conservation commis-
sions, and conducted various types of outreach. To accomplish the second goal, a small-scale pilot study
was conducted to establish experimental and monitoring protocols; the full experiments will be conducted
during the next 2 years, as will Goal 3. Under Goal 4, the effects of macropredators feeding on juvenile
and adult ascidian life stages were examined; nonnative species are not eaten by predators. Also under
Goal 4, the role of macropredators was examined, and results indicated that snails influence the mortality
of nonnative species and can be an effective measure for invasive species control. To accomplish Goal 5,
a spatially explicit, individual-based model driven by a hydrodynamic model is being developed, but it is
a complicated process. Another complication is dealing with a two-phase life history (larval and benthic).
The researchers run a hydrodynamic model and examine larval distribution over time; how shoreline
modifications affect larval distribution is examined. How invasive species might benefit from climate
change also will be explored; it is important to observe the effects of distribution and climate change
because they play a very important role in the success of invasive over native species.
The lessons learned have been that it is very important to include input by managers and stakeholders in
the early stages of the project; managers often are dealing with the most current crisis and are in a rapid
response mode. Additionally, long-term environmental databases and associated population and commu-
nity data are critically important. New stressors (e.g., coastal acidification) and power infrastructure
disturbances, and the manner by which they interact with climate change and land-use patterns, are
challenges.
Tier III Discussion
Dr. Lorenzana, in response to concerns regarding how researchers might interact with decision-makers,
stated that the primary function of EPA Regional Science Liaisons is to be a conduit between researchers
(inside and outside of ORD) and decision-makers at the state and local levels. Ms. Nancy Cavallaro of
USDA added that researchers also can interact with USDA extension agents who work with states.
Ms. Cavallaro commented to Dr. Seabloom that a stabilized shoreline destabilizes the coast farther down
and asked how to prioritize which area of the coast is of greatest concern. Dr. Seabloom responded that
any location on the coast that has sand is stabilized; destabilization and changes occur in areas that have
jetties. In these areas, huge sections are accreting, and others are eroding. Part of the project modeling
work examines erosional areas.
Ms. Cavallaro asked the three USDA grantholders about their client interactions. Dr. Sagers responded
that stakeholders in her project are farmers, developers of transgenic organisms, and everyone who
consumes any type of crop. Because many transgenic crops are herbicide resistant, herbicide development
has decreased. As herbicide resistance increases, the food supply is decreased because relatively few
herbicides are in development. North Dakota has been a receptive environment for her research. Her
expected outcome is to provide recommendations to agencies and producers. Dr. Silander noted that there
are a variety of stakeholders for his project because the data are valuable to make predictions. The
hundreds of citizens, from high school through retirement age, who help collect the citizen data are
stakeholders. The goal is to predict where to look in the landscape to best mitigate climate change effects.
Other stakeholders include landscape managers in national and state parks and forests and local conser-
vation organizations. Dr. Gao replied that policy-makers and farmers are the stakeholders for his project,
which will provide predictions to its stakeholders. The researchers work closely with the major cotton
farmers and provide them with results, with a focus on crop production.
Ms. Cavallaro noted that at the recent American Geophysical Union meeting that took place in Toronto
there was a great deal of discussion about climate change and modeling. In the last decade, the eastern
United States has cooled on average, whereas the western United States has warmed at an average higher
The Office of Research and Development's National Center for Environmental Research 29
-------
U.S. EPA Plight of Ecosystems in a Changing Climate: Impacts on Services, Interactions, and Responses Workshop
than that of the global warming average. Because of the lag in how climate is changing, variability is
important. Dr. Silander agreed and noted that the major challenge is dealing with the limitations of
climate modeling. There is a range of 16 different deterministic GCMs available, and changes are
modeled on a large, grid-cell basis. The thought is that temperature predictions are more reasonable on a
means basis, and precipitation has even greater uncertainty. Stochastic models incorporate uncertainty, so
the hope is to obtain regional climate models that are not merely scaled down. It would be beneficial for
researchers to communicate and collaborate to achieve this. The National Center for Atmospheric
Research is trying to develop stochastic versions of its models. Dr. Liang added that he had not presented
information about how his project's regional model had been developed and noted that it improved
hybridization and decreased uncertainty; the laboratory has published many papers regarding this, and he
is willing to share the results. Dr. Purkey noted that the State of California is using six models and two
emissions scenarios with 12 future climate projections for its Capital Improvement Plans involving
hydrology. It is important to articulate that different types of analyses are needed to help decision-makers
understand what the predictions mean and how to make a decision.
Dr. Wardrop noted that it is challenging to educate people regarding what changes are ecologically
related; it is necessary for funding agencies to convey to climate scientists what they would like to know.
Dr. Poff commented that he is interested in the issue of regionally downscaling models because that is
where the real uncertainty is; from a thermal perspective, this is not so difficult to accomplish, but from a
precipitation standpoint it is very challenging. It is difficult for models to capture atmospheric processes
at the regional scale that are important in driving regional precipitation. He asked Dr. Gao whether his
laboratory's modeling approach produces ecologically relevant hydrographs that can be used. Dr. Gao
agreed with Dr. Poff s assessment regarding temperature and precipitation. His publications generally
show how model improvements were made. Because of the differences between regions, the data must be
used together. Dr. Poff asked whether historical precipitation was captured. Dr. Liang responded that the
hydrologic model did capture it very well.
Ms. Cavallaro noted that predicting competition of invasive species is complicated, especially as
competition depends on the rate of climate change. Competition becomes very difficult to predict if the
rate of climate change cannot be predicted.
Closing Remarks
Dr. Jones thanked the presenters, organizers, and USDA for their efforts. He noted that he has many
suggestions and comments to share at EPA headquarters. He will contact the presenters regarding
permission to publish the presentations on the workshop Web site; proceedings of the workshop will be
available in the future. Dr. Lorenzana thanked the attendees for their participation and adjourned the
meeting at 12:22 p.m.
The Office of Research and Development's National Center for Environmental Research 30
------- |