Researchers' Working Meeting on Climate Change Impacts
in the Mid-Atlantic Region
Penn State--June 8-9, 1998

EXISTING INFORMATION AVAILABLE FOR MARA

(Summary by Colin Polsky & Lubing Wang)

Panel:

Chair - Dave Abler, Penn State
John Paul, US EPA, Narragansett Lab
Lou Pitelka, University of Maryland Center for Environmental Science
Doug Burns, US Geological Survey
Linda Mortsch, Environment Canada/University of Waterloo
Adam Rose, Penn State
Eric Barron, Penn State & National Assessment Synthesis Team

John Paul discussed the information available from a series of projects related to the EPA-sponsored Mid-Atlantic Integrated Assessment (MAIA). EPA is increasing its work with states and the private sector to provide useful information about ecosystem health. They are working to develop improved environmental indicators for biology, habitat and land activities. These indicators include nutrients and toxic contaminants, watershed land-use, habitat quality, and plants and animals. Merging physical and chemical information with biological can enhance the potential for data to influence management decisions as well as public perceptions. The goal is that MAIA would help to answer the following questions: Are there problems with the ecological resources? What is the geographic distribution of the problem areas? What are the probable causes of the ecological problems? Have the ecological problems been changing with time? What does this information mean to people? What is being done to address the ecological problems?

MAIA includes the Mid-Atlantic Highlands Assessment, Gap Analysis Program, Chesapeake Bay Program, Delaware Estuary Program, Maryland Coastal Bays Program, National Water Quality Assessment Program, Forest Inventory and Analysis Program, Coastal Change Analysis Program, and others. Much of their data and more can be found on the Environmental Monitoring and Assessment Program (EMAP) web site (www.epa.gov/EMAP), including useful web links. Given this wealth of available data, the following measures can facilitate environmental assessments: First, establish strict guidelines for data requirements to reduce the temptation to integrate unreliable or unrelated data. Second, establish an external advisory group to ensure that the end product is scientifically credible and relevant to stakeholders.

Lou Pitelka summarized the Vegetation/Ecosystem Modeling and Analysis Project (VEMAP) and related data available for common use. The objective of VEMAP is to conduct a preliminary analysis of the effects of climate change on vegetation distribution and ecosystem processes in the continental United States. The approach links a climate model with a vegetation model and an ecosystem model. A range of results was provided by using three representatives of each type of model to create twenty-seven combinations of linked models. The GCMs used were OSU, UKMO and GFDL; the vegetation models were BIOME2, MAPSS and DOLY; the ecosystem models were CENTURY, BIOME-BGC, and TEM. The analysis was a sensitivity experiment, rather than a prediction.

To ensure that differences in results were due to differences in model structure, a common dataset was used in all models to describe initial conditions. The different linked models were run to 'equilibrium' to determine total carbon storage. Phase I of the project is now complete and available on WWW and a CD.

Phase II is targeting shortcomings found in Phase I such as the inability to produce transient runs and the projection of potential natural vegetation rather than actual vegetation (including agriculture). The data underlying Phase II analyses are characterized by physical consistency between daily and monthly figures and climate and topography. These data include the distribution of current vegetation; soil characteristics; historical climate (99 year record at 0.5° grid squares of minimum and maximum temperatures, precipitation, vapor pressure, solar radiation); and transient climate change scenarios from GCMs. Information on windspeed is planned.

Doug Burns discussed research ongoing in the Catskills (New York), where nitrogen deposition and its effects on the New York City water supply are being monitored. He introduced databases from USGS-related projects including Hydrological Benchmark Stations, the National Water Quality Assessment (NAWQA) and an upcoming integrative workshop. Additional information can be found at www.usgs.gov.

Linda Mortsch presented results and lessons learned from an investigation into the potential impacts of climate change and variability on the Great Lakes-St. Lawrence Basin region. Involving stakeholders has been a key ingredient to developing a solid, sustainable assessment and in developing solutions (adaptations) to the impacts of climate change. One question to keep in mind is "How can/will adaptation to climate change affect regional ‘sustainable development’ efforts?" Not every possible adaptation to climate change is necessarily desirable. The lay public has serious misperceptions on this topic, especially within the context of sustainable development. For communication purposes, she recommends framing the discussion on impacts in terms of "practice climates" or "what if" scenarios. In the assessment process, more than one climate change scenario should be used to give a range of vulnerabilities and to reflect the uncertainty in climate change knowledge. Analysts who present information on "net" effects of climate change may obscure some of the distributive effects that are important to stakeholders and policy makers. Proceedings from a 1997 Symposium on "Adapting to climate change and variability" is available at http://www.tor.ec.gc.ca/earg/glslb/sympMay97.htm.

Adam Rose discussed the IMPLAN System. Developed by the US Forest Service, IMPLAN is a widely used tool for regional economic impact analysis. Rose displayed the type of results IMPLAN can generate by describing the economic profile of the Mid-Atlantic Region. This region accounts for twenty percent of US Gross Domestic Product (GDP) and twenty-five percent of US employment; one-third of gross regional product is exported. Thus regional economic activity is heavily dependent upon external agents and decision-makers.

IMPLAN permits analysts to paint a detailed picture of a regional economy. Beyond generating descriptive statistics, one can also use IMPLAN and its US-wide, county-specific, 500 sector-strong underlying structure to create Input-Output (I-O) matrices, social accounting matrices and impact analyses through the development of I-O and Computable General Equilibrium (CGE) models.

Eric Barron, a member of the National Assessment Synthesis Team (NAST), reviewed the climate data that all regional research groups are to use in their impact assessments. Ideally, the climate data would consist of the following: (1) historical baseline climate data from 1850 to present, and (2) an ensemble of scenarios generated from GCM experiments, including some high-resolution scenarios, and some scenarios that are consistent with those planned for use in the upcoming third Inter-governmental Panel on Climate Change (IPCC) assessment.

Realistically, given the short time frame of the assessment, the NAST recognizes the need to rely on already-completed GCM runs, where the information is easily accessible and in a consistent format. A limited number of model "experiments" will be used, and the data will show changes that are significant compared to natural variability. The NAST hopes to obtain climate data used by the VEMAP project (gridded record; 1895-1993), and data from the National Climate Data Center's U.S. Historical Climate Network (HCN). NAST will also use data from the Canadian Climate Center (CCC) model, which has the capability to run simulations in both doubled-CO₂ and transient modes, both with and without IPCC aerosols. Further, regions will use runs from the Climate Model Inter-comparison Project (CMIP) to define ranges for climate variables, and apply downscaling methodologies when appropriate/available.

The differences between the ideal and realistic data constitute the limitations of regional assessments. The model scenarios are not "ensemble", so results will not be generated at different levels of atmospheric CO₂ concentration. There may not be full congruence with IPCC modeling parameters, so direct comparability with international results is unlikely. Interannual variability is not incorporated, so one of the principal variables to impact societies and agriculture is absent, at least for the moment.