&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

-------