Gulf of Maine Watershed Information and Characterization System Database Development Documentation
Water Systems Analysis Group Technical Report 00-01
    Contents:
        Abstract
        Introduction
        Site Description
        Database Development
           Elevation
           STN-2min River Network
           Land Cover
           Soils
           Climate Data (Air Temperature and Precipitation)
           Stream Discharge
           Atmospheric Deposition
           Water Quality
        Web Interface
        Water Balance Modeling
        References
        List of Figures

ABSTRACT

     The goal of the Gulf of Maine Watershed Information and Characterization System (GM-WICS) project is to develop a regional, geographic information system (GIS) database and characterization system for drainage basins that empty into the Gulf of Maine, an important coastal zone of the Northeastern United States and Canada. Our objective is to provide an online reference source, for scientists, coastal and freshwater managers, citizen groups and students of digital watershed characteristic and water quality monitoring data for the upland basins emptying into the Gulf of Maine. This system includes an archive of relevant GIS-based data sets, historical and near real-time discharge and water chemistry data, and the results of water balance and water chemistry modeling. This report describes the development of data sets required for water balance and nutrient transport modeling for the Gulf of Maine Watershed region. The Water Balance Model (Vörösmarty et al.1998) operates on a monthly time step and requires the following gridded data inputs: mean monthly air temperature, total monthly precipitation, elevation, soil texture, and land cover. Discharge data for the gauged stream stations in the Gulf of Maine have been compiled for model comparison and validation. A gridded river network has been developed for the Gulf of Maine basins as the base layer for nutrient transport modeling.  In addition to the data sets prepared for use with the Water Balance Model, atmospheric deposition and water quality data sets were assembled and mounted on the web (http://www.gm-wics.unh.edu).

INTRODUCTION

    The management of inland landscapes and freshwater resources is a key determinant of coastal zone water quality, an aspect of coastal zone management that is still poorly understood (National Academy of Sciences 1994).  Within this context, riverborne loadings of biotically-active elements to the coastal zone are known to have increased several-fold since the beginning of the Industrial Era (Nixon et al. 1996). Increased freshwater nutrient inputs to the coastal zone with subsequent increases in primary production, producing eutrophic conditions in some areas, and resultant proliferation of nuisance and toxic algal blooms, have become a national and world-wide problem (Anderson 1989; Anderson 1994; Smayda 1990).
    There are numerous types of nutrient inputs to estuarine and coastal waters including sewage treatment plants, industrial discharges, ground water, atmospheric deposition, sediment regeneration, and non-point sources such as fertilizer and septic leakage. Some nutrient sources are relatively constant (i.e. sewage treatment plants) while others are either seasonally driven (spring snowmelt, fertilizer applications, and coastal water inputs) or event driven (rainfall).
    The major stresses impacting the Gulf of Maine ecosystem are nutrient over-enrichment (eutrophication), the introduction of toxic contaminants and toxic algal blooms. Natural inputs of nutrients to the Gulf of Maine are dominated by ocean inputs through the Northeast Channel, however coastal areas are primarily impacted by land-based nutrient loads. The geographical distribution of anthropogenic nutrient inputs to the nearshore ocean is generally related to population centers and are expected to increase as coastal development pressures increase (Pearch and Wallace 1995).
     This report describes the creation of geographically-referenced data sets for the biophysical properties of the Gulf of Maine upland watershed, use of these data layers in water balance modeling for the region, and the assessment of additional hydrologic data sets developed for water and nutrient transport modeling. These data sets make up the Gulf of Maine Watershed Information and Characterization System (GM-WICS) and this database has been mounted on the internet at http://www.gm-wics.unh.edu. These spatially distributed data layers include: climate (air temperature and precipitation), elevation, vegetation, soils, a gridded stream network, river discharge, atmospheric chemistry and water chemistry. Climate, elevation, vegetation and soils data have been used as the input data for the Water Balance Model (WBM) (Vörösmarty et al. 1998) to develop estimates of soil moisture variation, evapotranspiration, and runoff.

SITE DESCRIPTION – GULF OF MAINE WATERSHED

    The Gulf of Maine watershed covers 177,008 km2 in three states and three Canadian provinces, stretching from the north shore of Cape Cod, Massachusetts, to Cape Sable, Nova Scotia in Canada. It reaches 322 km offshore to underwater plateaus called the Georges and Brown banks. The US EPA has delineated 25 major estuarine drainage areas (EDA) and 11 minor coastal drainage areas (CDA) in the Gulf of Maine Watershed region (NOAA-EPA). The region includes 60 counties, 57 U.S. Geological Survey (USGS) Hydrologic Cataloging Units, and 453 subbasins (NOAA- EPA). Table 1 provides a summary of the land area in the watershed by state and province.
    The Gulf of Maine Basins provide freshwater inputs to a mosaic of habitats.  The Merrimack, Saco, Androscoggin, Kennebec, Penobscot, St. Croix, and St. John Rivers contribute on the average 946 million m3 of fresh water to the Gulf each year (Maine State Planning Office 1991). The Gulf's estuaries, such as the Annapolis River estuary in Nova Scotia and the Great Bay estuary in New Hampshire, are highly productive systems. There are over 200 species of fish and shellfish native to the Gulf of Maine, of which 40 to 50 species are of commercial value (Maine State Planning Office 1991). These estuaries are thought to be vital at some stage in life to 70 percent of the commercially valuable fish species of the Gulf (NOAA-EPA).

DATABASE DEVELOPMENT

     Table 2 lists the data gathered for the Gulf of Maine Watershed Information and Characterization System (GM-WICS). These data sets are available at http://www.gm-wics.unh.edu/html/download.html. The goal for time varying data was to have complete records of data from 1970 through the present. The soils and vegetation data were chosen because they cover the entire Gulf of Maine Watershed area.
    Climate, discharge and chemistry data sets were acquired as ASCII (text) files. These data sets were formatted using Unix commands, Perl scripts and short C programs. Point coverages were created in ArcInfo 8.0.2 (ESRI) or Global Hydrologic Archive and Analysis System- River GIS 2.1 (GHAAS-RGIS) (Water Systems Analysis Group, UNH). Gridded elevation, vegetation and soils data were acquired in ArcGrid (ESRI) or ArcInfo grid ascii format. These data sets were projected, formatted and resampled using ArcGrid 8.0.2 (ESRI) and GHAAS-RGIS 2.1 (Water Systems Analysis Group, UNH). ArcView 3.2 Spatial Analyst extension (ESRI) and Data Manager (Water Systems Analysis Group, UNH) were used to perform station data interpolations to create climate and nutrient concentration surfaces.   ArcView 3.2 (ESRI) is used to visualize spatial data sets and create map products.
    All gridded data sets have been standardized to 2 arc minute grid resolution, in geographic format, within the region of 40o – 49 o N, 62 o – 73 o W. The database resolution was chosen as the finest scale that would provide a reasonable representation of the Gulf of Maine Watershed rivers and also be within the memory restrictions of the Water Balance Model. Grid dimensions vary with latitude. Table 3 lists the dimensions of the 2 arc minute grids.

Elevation
    The GTOPO30 digital elevation model (DEM) was chosen as the base layer for this project (USGS EDC 1996). This DEM was downloaded from the USGS EDC website (http://edcdaac.usgs.gov/gtopo30/gtopo30.html). For this project the 30 arc second digital elevation model was resampled to 2 arc-minute resolution using GHAAS-RGIS (Water Systems Analysis Group, UNH).  Three elevation grids were created for the Gulf of Maine Watershed Area: minimum, mean and maximum. Table 4 lists statistics for these grids, minimum grid cell, maximum grid cell, and mean grid cell. The mean elevation grid was used to create the simulated river network in GHAAS-RGIS (Water Systems Analysis Group, UNH). Figure 1 illustrates the mean elevation grid for the Gulf of Maine Watershed Regional Area.

STN-2min River Network
    The basin and sub-basin attribute files and the Simulated Topological Network (STN) were generated from the GTOPO30 DEM (USGS EDC 1996), which represents topography at 30-second resolution. After tests of several different grid resolutions on portions of the basin, we discovered that a 2-arc minute resolution gridded networking system would be suitable for application at the regional scale. This network has more than 150,000 individual grid cells within the Gulf of Maine Watershed area. The final 2min resolution river network represents 825 individual basins, 400 of these have more than a single cell, the remaining 425 basins are single cells along the coast. There are 190 named basins in the network. Total watershed area at 2min resolution is 174,825 sq km. This data set represents a topological structure by which a hierarchy of nested basin and sub-basin characteristics was derived using existing software we had already developed (GHAAS-RGIS (Water Systems Analysis Group UNH)).  The grid-based structure was error-checked and edited using 1:1M digital line graph templates from the Digital Chart of the World (ESRI 1992), high resolution EPA River Reach files (EPA 1998) and the EDA/CDA basin boundaries from the NOAA Gulf of Maine Land-based Pollution Sources Inventory (NOAA-EPA).  This river network serves as the organizational structure upon which the land characterization is based. Summary statistics for these basins in the Gulf of Maine Watershed include stream order, stream length, and basin drainage area.
    Three levels of manual editing were completed using additional functions of the software. The initial editing compared the STN-2min network with the 1:1 million scale Digital Chart of the World (ESRI 1992). This comparison ensured that the major rivers were simulated correctly. Any major river, which deviated significantly from the DCW river location, was manually edited to create a better match. The second level of editing compared STN-simulated basin areas with the US EPA EDA/CDA basins designated for the Gulf of Maine Watershed in the Gulf of Maine Land-Based Pollution Sources Inventory (NOAA-EPA). Figure 2 shows the Gulf of Maine Watershed EDA/CDA basins. Basin boundaries were compared and small order river segments were redirected to fall within basin boundaries. The last level of editing was an integral part of geo-referencing the stream discharge gauging stations to the simulated river network. Discharge stations were linked to the stream network and the station’s upstream area was compared to the STN-simulated upstream area for the station. The discharge stations were georeferenced to the STN 2min river network through an iterative process. The gage location was adjusted to place the gage on the river network and minimize the difference between the measured catchment area and the STN estimated catchment area. These adjustments were made using the GHAAS-RGIS function STN Characteristics, and the STN Coordinates function (Water Systems Analysis Group UNH).  These procedures linked the station data to the STN-4km, generated characteristics about the stations, and then moved the actual stations to the closest reasonable STN cell center with most similar upstream area using a search algorithm.  Upstream reported areas of the river discharge stations were then compared to the estimated upstream area of the 2min STN river network. In some cases the river network was manually edited to improve the area comparison. The manual editing involved changing the direction of stream reaches grid cell by grid cell in GHAAS-RGIS (Water Systems Analysis Group UNH).These procedures were used to ensure that the gridded network would generate accurate representations of the actual river networks.  Figure 3 illustrates the manually-corrected STN-2min network. Darker stream segments represent higher order stream reaches. Stream order is based on the Strahler (1964) classification. The STN-2min network outside of the Gulf of Maine Watershed boundary has not been edited.

Land Cover
    The land cover grid for the Gulf of Maine Watershed Region was developed from the USGS North America Land Cover Characteristics (NALCC) Database (USGS EDC 1999). This data set was developed from 1 km AVHRR data from April 1992 – March 1993. The 1km2 resolution grid for North America was clipped to the Gulf of Maine Watershed Regional Area (40o – 49 o N, 62 o – 73 o W), and resampled in ArcGrid (ESRI) using a nearest neighbor resampling function to create the 2 arc minute grid. Two versions of the NALCC data set, International Geosphere-Biosphere Programme (IGBP) and U.S. Geological Survey Land Use/Land Cover System-Legend 2 (USGS-L2), were converted to the Water Balance Model (WBM) land cover categories (Table 5).  The original categories for the IGBP and USGS-L2 legends are listed in Appendix A.  These grids were compared to check the category conversions and were within 2% agreement.  The WBM land cover classification has hydrological importance, since WBM uses land cover to calculate rooting depth (Vörösmarty et al. 1996, 1998). Figure 4 illustrates the WBM land cover grid.

Soils
    The soil texture grid for the Gulf of Maine Watershed Region was developed from the Digital Soil Map of the World CD-ROM, Version 3.5 (FAO 1995). The FAO digital map contains soil unit attributes and slope and texture codes. The digital map was clipped to the Gulf of Maine Watershed Regional Area and converted to a 2 arc minute grid. The soil unit, texture and slope codes were used to create the WBM soil texture classifications (Vörösmarty et al. 1996, 1998). A slope grid was also created. The WBM soil texture classifications are listed in table 6. These classifications are based on ¾ majority for the texture, thus soil units which have a sum of greater than 75% coarse texture are considered to be coarse, soil units which have a sum of greater than 75% medium texture are considered to be medium, soil units which have a sum of greater than 75% fine texture are considered to be fine, etc. Figure 5 illustrates the WBM soil texture grid.

Climate Data (Air Temperature and Precipitation)
    The Canadian precipitation and air temperature data sets were compiled from several sources. Data for 1970-1993 were extracted from the Environment Canada: Canadian Climate Normals & Monthly Averages data set (Environment Canada 1993). These data were provided to the project by collaborators at the University of Delaware. Additional data for 1994-1999 were acquired during the summer of 2000. Monthly values were calculated from data on the Environment Canada Canadian Daily Climate Data Temperature and Precipitation CD-ROMs: Quebec 1998 and Atlantic Provinces 1999 (Environment Canada 1999, 2000). Additional data for Quebec 1999 were acquired as provisional data directly from the regional Environment Canada office. Monthly data are reported as total mm of precipitation and average degrees C temperature. For the 1994-1999 data, monthly values were calculated from daily records. The GM-WICS database includes 798 precipitation stations and 693 air temperature stations, which were active for some part of 1970-1999 and within the Gulf of Maine Watershed regional area (39 – 50 degrees latitude, -61 - -74 degrees longitude).  Stations outside of the Gulf of Maine Watershed were included in order to minimize interpolation edge effects when interpolation was applied to produce gridded fields.
    The United States precipitation, air temperature, and snow data were extracted from the National Climatic Data Center (NCDC) Cooperative Summary of the Day archive.  These data were acquired from the NCDC Cooperative Summary of the Day web site (http://nndc.noaa.gov/). Daily data values are reported for total inches of precipitation and degrees Fahrenheit temperature. These data were converted to millimeter of precipitation and degrees Celsius temperature. Monthly precipitation totals and average monthly temperatures were calculated from the daily data. There are 356 precipitation stations and 220 temperature stations within the Gulf of Maine Watershed regional area (39 – 50 degrees latitude, -61 - -74 degrees longitude) in the GM-WICS databank.
    The data from Canada and the United States were joined to create monthly time series files, station location files and interpolated climatology and time series data fields. The interpolations were processed using the Spheremap (NCGIA 1997) function in Data Manager (Water Systems Analysis Group, UNH). Spheremap is an algorithm that seeks to overcome variations in the distance between sampling locations that result from map projections (NCGIA 1997). Spheremap interpolates the data before projecting them into cartesian space (Willmott et al. 1984). These files were imported into RGIS and ArcInfo formats for display, visual analysis, and statistical analysis.
     Table 7 provides maximum, minimum, mean and standard deviation statistics for the precipitation climatology grids. Table 8 provides maximum, minimum, mean and standard deviation statistics for the air temperature climatology grids. Figure 6 illustrates the locations of the air temperature and precipitation stations used for the climate interpolations. Figures 7(a, b, c, d, e) illustrate the climatology interpolations for precipitation. Figures 8(a, b, c, d, e) illustrate the climatology interpolations for air temperature. Point data and the air temperature and precipitation interpolated time series can be viewed in the Interactive Data Explorer (http://www.gm-wics.unh.edu/explorer/).

Stream Discharge
      Canadian stream discharge data were extracted from the Environment Canada, Atmospheric Environment Service, Surface Water and Sediment Data, HYDAT CD-ROM version 4.92 (Environment Canada 1994) and the Environment Canada, Atmospheric Environment Program, Surface Water and Sediment Data, HYDAT CD-ROM version 98-1.05.8 (GIC 1999). The entire period of record was extracted for each station that was active between 1970 and 1999. There are 64 stations within the Gulf of Maine Watershed area for this time period. The data are reported as daily average stream (m3 s-1). Tab delimited ASCII (text) files were created for daily average discharge time series and monthly average discharge time series.
     The United States stream discharge data were extracted from the United States Geological Survey National Water Information System (NWIS-W) archive (http://waterdata.usgs.gov/nwis-w) and the Real-Time Water Data web pages. The provisional real-time data is downloaded on a daily basis and archived as hourly averages at UNH. The entire period of record was extracted for each station which was active between 1970 and the present. There are 132 stations within the Gulf of Maine Watershed area for this time period, real-time data is reported for 106 stations. The archived data were reported as daily average stream discharge (ft3 s-1). The real-time data is reported as 5 minute or 15 minute instantaneous measurements in the same units. The real-time, provisional data is archived at UNH as hourly averages (http://www.gm-wics.unh.edu/html/realtime.html). All US data were converted to m3 s-1. Tab delimited ASCII (text) files were created for daily average discharge time series and monthly average discharge time series. These files were merged with the Canadian data.  As explained earlier, the stations were geo-referenced to the STN using an iterative process in GHASS-RGIS (Water Systems Analysis Group UNH). This database includes the stations that drain areas equal to or larger than 100 km2. Basins smaller than 100 km2 are not adequately represented by the simulated river network.
     Figure 9 shows the stream discharge gage locations on the STN-2min simulated river network.  Many of the stations fall along large-flow, higher order segments of the river network.  Table 9 lists the 25 largest river basins in the Gulf of Maine Watershed STN 2min river network. Table 10 lists stream discharge station statistics, including the comparison between the measured upstream area and the estimated upstream area for each discharge station. Figure 10 shows a log-log graphical comparison between the gauging stations reported drainage area and the drainage area estimated by the STN 2min river network. There are only a few basins which are significantly different from the reported upstream area.  Figure 11 shows a comparison between cumulative watershed area and the STN-2min Gulf of Maine basins ranked by size (largest to smallest). The largest basin, the St. John River, covers 32% of the Gulf of Maine Watershed Area. The three largest basins, the St. John River, the Kennebec/Androscoggin River, and the Penobscot River, cover 58% of the watershed area. The 10 largest rivers cover 75% and the 50 largest rivers cover 90% of the watershed area.

Atmospheric Deposition
     The Canadian atmospheric deposition data sets were extracted from the National Atmospheric Chemistry (NAtChem) database, the Canadian Precipitation Monitoring Network (CAPMoN) database, and the New Brunswick Precipitation Monitoring Network (NBPMN) database (Environment Canada 1999). These data were received from the Environment Canada Air Quality Research Branch. These data are a combination of daily, weekly and monthly data. The concentration data are reported as mg L-1. These data have been formatted into tab delimited ASCII (text) files. There are 39 sampling sites within the Gulf of Maine Watershed regional area (39 – 50o N, -61 - -74o W).
     The United States atmospheric deposition data set was extracted from the National Atmospheric Deposition Program (NADP) archive. These data were acquired from the NADP website (http://nadp.sws.uiuc.edu/ default.html). Weekly concentration data were downloaded from this site. The data are reported in mg L-1 units. These data have been formatted as tab delimited ASCII (text) files. There are 17 sites within the Gulf of Maine Watershed regional area (39 – 50 o N, -61 - -74 o W).

Water Quality
    There was no response from Environment Canada regarding the Canadian water quality database, Envirodat. At this time only the data codes are available on the web. A response from the Atmospheric Monitoring and Water Survey Directorate, indicated that data is archived in a central location through 1992 and has been archived in each region separately since 1992.
 The United States water quality data were extracted from the USGS CD-ROM: Data from Selected U.S. Geological Survey National Stream Water-Quality Monitoring Networks (WQN) – DDS-37(USGS 1996). The physical parameters, nutrient concentrations (mg L-1) and major ion concentrations (mg L-1) were formatted to create time series in tab delimited ASCII (text) format and point coverages of the station locations. These data have not been added to the web page. There are 22 WQN locations in the Gulf of Maine Watershed. Based on this limited number of water quality locations, we chose to focus our analysis of nutrient transport on four basins within the Gulf of Maine Watershed: the Ipswich and Parker Rivers, Great Bay rivers, the coastal rivers and streams at Wells NERR, and the Androscoggin /Kennebec basin. The analysis for the Ipswich and Parker Rivers was done in collaboration with Plum Island Ecosystem LTER researchers. We worked closely with T. Loder (CICEET project 11, 1997-1999) on analysis of data from the Lamprey, Oyster and Salmon Falls Rivers in the Great Bay drainage. In addition, we worked with T. Loder (CICEET project 11, 1997-1999) on sampling the Androscoggin/Kennebec river system. With the help of volunteers, we carried out a year-long sampling program at the mouth of the Kennebec and the Androscoggin and below the confluence of these rivers. We also worked with seven schools (Oakland, Carrabassett Valley, Gardiner, Dixfield, Oxford Hills, Lewiston and Auburn) in the Kennebec/Androscoggin drainage to develop weekly time series of river nutrient concentrations. Samples taken at the mouth of the Androscoggin/Kennebec have been used to estimate seasonal nutrient flux to the coast (Bredensteiner et al., in preparation). We estimate that 1621 metric tons of dissolved inorganic nitrogen (DIN), 127 metric tons of phosphate (PO4), and 19272 metric tons of silicate (SiO4) were delivered to the coastal zone by the Androscoggin/Kennebec Rivers in 1999 (Bredensteiner et al., in preparation). Of particular interest is the snowmelt contribution to this flux. In order to calculate the contribution of snowpack storage and subsequent melt to river nutrient flux, we completed five intensive river and three intensive snowpack sampling trips (River: February 1999, December 1999, January 2000, February 2000 and March 2000; Snow: February 1999, January 2000, February 2000). We have developed estimates of river nutrient flux and snowpack storage of inorganic nitrogen, phosphorus and silica from this data set. We estimate that the snowpack in the Androscoggin/Kennebec basins stored 760 metric tons of DIN, 11 metric tons of PO4, and 46 metric tons of SiO4 in 1999, and 1020 metric tons DIN, 13 metric tons of PO4, and 35 metric tons of SiO4 in 2000 (Bredensteiner et al. in preparation).

WEB INTERFACE

    The Gulf of Maine Watershed Information and Characterization System (GM-WICS) World Wide Web interface was designed and an initial homepage was put online during the summer of 1999. During the fall of 1999 climate and discharge station data was made available online. The Interactive Data Explorer was developed during the winter and put online in the spring of 2000. The Data Explorer provides an interactive tool for through which point data and gridded Gulf of Maine Watershed data layers can be examined. Real-Time stream discharge data is available for USGS stations within the Gulf of Maine Watershed. These data are downloaded from the USGS web sites daily and archived at UNH. Broad suites of specific data sets have been included on our WWWeb site. The site was expanded to contain two components: (i) the original site expanded to include focus basins and tiled datasets (http://www.gm-wics.sr.unh.edu/) and (ii) a new Interactive Data Explorer (http://www.gm-wics.unh.edu/explorer). These data repositories provide both station-based point data and gridded geospatial data. Table 1 lists the GM-WICS data sets gathered for this project.

WATER BALANCE MODELING

    The climate, and watershed characteristic data layers have been used as input layers for the Water Balance Model (WBM) (Vörösmarty 1998). WBM was run in the Data Assembler (Water Systems Analysis Group, UNH). This model runs using monthly input data and applying a quasi-daily time step. It estimates output variables such as: runoff, evapotranspiration, shallow groundwater, and soil moisture variations. The WBM runs summarized below used the precipitation and air temperature climatologies for 1970-1993. Additional synthesis of this data will be done based on comparisons of estimated stream discharge and measured stream discharge at the gauging stations throughout the watershed. Table 11 summaries the static soil parameters calculated in WBM.  These statistics include the minimum value, maximum value and mean value for the grid.
    Table 12 summarizes output variables from WBM for the Gulf of Maine Watershed climatology (1970-1993) run. These summaries include the minimum value, the maximum value and the mean value for the grid. The Water Balance Model simulates rainfall, snowfall, snowpack, soil recharge from snowmelt, potential evapotranspiration, actual evapotranspiration, soil moisture, change in soil moisture, depth to groundwater, and runoff. The precipitation input is separated by the model into rain and snow. Snowpack accumulates when monthly temperatures are below –1.0 oC. Snowmelt is a function of temperature and elevation. Potential evapotranspiration is calculated based on the Hamon (1963) function, and actual evapotranspiration is calculated as potential evapotranspiration minus change in soil moisture. Soil moisture, changes in soil moisture, depth to groundwater and runoff are calculated based on soil parameters, evapotranspiration and precipitation. Appendix B lists the WBM equations. Figures 12 (a, b, c, d) and 13 (a, b, c, d) illustrate estimated potential evapotranspiration (PET) and Runoff from the Gulf of Maine WBM climatology run. For comparison the runoff results from the UNH-Global Runoff Data Centre (GRDC) WBM analysis are included. The UNH-GRDC analysis was made at 30-minute resolution, and results presented here are for the Gulf of Maine Watershed regional area (39 – 50 o N, -61 - -74 o W). Table 13 summarizes the UNH-GRDC runoff results and Appendix C (a, b, c, d)  illustrates these monthly runoff grids.

REFERENCES

Anderson, D.M. 1989. Toxin variability in Alexandrium species. Toxic Marine Phytoplankton. 41-51.

Anderson, D.M. 1994. Red tides. Scientific American, 271: 62-68.

Bredensteiner, K., C. Vörösmarty, T. Loder, E. Penfold, and W. Wollheim. 2000. Regional Estimation of Snowpack Nutrient Storage. (In preparation for submission to Estuaries).

Environment Canada. 1993. Canadian Climate Normals 1961-1990 & Monthly Averages CD-ROM. Data provided by the University of Delaware, 1998. Cort Willmott and Mike Rawlins Center for Climate Research, Department of Geography, University of Delaware, Newark, DE 19716

Environment Canada. 1994. HYDAT CD-ROM Version 4.92. Surface Water and Sediment Data, Atmospheric Environment Service.

Environment Canada. 1999. Quebec 1998: Canadian Daily Climate Data - Temperature and Precipitation. CD-ROM.

Environment Canada. 1999. Quebec: Provisional Canadian Daily Climate Data – Temperature and Precipitation. Direct from Region via request.

Environment Canada. 2000. Atlantic Provinces: Canadian Daily Climate Data – Temperature and Precipitation. CD-ROM.

Environment Canada. 1999. Canadian National Atmospheric Chemistry Database. Atmospheric Environment Service, 4905 Dufferin Street, Downsview, Ontario, Canada M3H 5T4

Environmental Protection Agency. 1998. U.S. EPA Reach File 1 (RF1) for the Conterminous United States. U.S. Environmental Protection Agency, Office of Science and Technology Washington, D.C <http://www.epa.gov/ owowwtr1/monitoring/rf/rfindex.html>

Environmental Systems Research Institute, Inc (ESRI). 1982-2000. ArcInfo, ArcGrid, ArcTools 8.0.2. Redlands, CA.

Environmental Systems Research Institute, Inc (ESRI). 1992-1999. ArcView GIS 3.2. Redlands, CA.

Environmental Systems Research Institute, Inc (ESRI). 1992. Digital Chart of the World for use with ARC/INFO, Digital Data on CD-ROM.  Redlands, CA.

Food and Agriculture Organization (FAO). 1995. Digital Soil Map of the World CD-ROM. Version 3.5. Land and Water Development Division, Food and Agriculture Organization of the United Nations, Rome.

Greenland International Consulting Inc (GIC). 1999. HYDAT CD-ROM Version 98-1.05.8. Surface Water and Sediment Data, Atmospheric Environment Service. Data from Environment Canada Regional Offices.

Hamon, W.R. 1963. Computation of direct runoff amounts from storm rainfall. Int. Assoc. Sci. Hydrol. Publ. 63:52-62.

Maine State Planning Office. 1991. Gulf of Maine Watershed Map and Fact Sheet.

National Academy of Sciences.  1994.  Priorities for Coastal Ecosystem Science. Committee to Identify High-Priority Science to Meet National Coastal Needs.  National Academy Press, Washington, DC.  88 pp.

NCDC 1999. Cooperative Summary of the Day, National Climatic Data Center, Asheville, NC. NOAA.

NCGIA (National Center for Geographic Information and Analysis). 1997. Spherekit: The Spatial Interpolation Toolkit. University of California, Santa Barbara, CA  http://www.ncgia.ucsb.edu/pubs/spherekit/

Nixon, S.W., J.W. Ammerman, L.P. Atkinson, V.M. Berounsky, G. Billen, W.C. Boicourt, W.R. Boynton, T.M. Church, D.M. Ditoro, R. Elmgren, J.H. Garber, A.E. Giblin, R.A. Jahnke, N.J.P. Owens, M.E.Q. Pilson, and S.P. Seitzinger. 1996. The fate of nitrogen and phosphorus at the land-sea margin of the North Atlantic Ocean. Biogeochemistry 35: 141-180.

National Oceanic and Atmospheric Administration and Environmental Protection Agency (NOAA-EPA). 1996. Gulf of Maine Land Based Pollution Sources Inventory. http://www-orca.nos.noaa.gov/projects/gomaine/

Pearce, J. and G. Wallace. 1995. The Health of the Gulf of Maine Ecosystem: Cumulative Impacts on Multiple Stressors, Dartmouth College  Hanover, New Hampshire, Regional Association for Research on the Gulf of Maine.

Pierce, L.T. 1958. Estimating seasonal and short-term fluctuations in evapotranspiration from meadow crops. Bulletin of American Meteorological Society 39:73-78.

Smayda, T.J. 1990. Novel and nuisance phytoplankton blooms in the sea: Evidence for a global epidemic. Toxic Marine Phytoplankton. pp. 29-40.

Strahler, A. N. 1964. Quantitative geomorphology of drainage basins and channel networks Handbook  of Applied Hydrology. Chow, V.T. Editor, McGraw-Hill Book Company, Inc.

United States Geologic Survey (USGS) Earth Resources Observation Systems (EROS) Data Center (EDC). 1996. Global 30 Arc Second Elevation Data Set (GTOPO30). Sioux Falls, South Dakota http://edcdaac.usgs.gov/gtopo30/gtopo30.html

United States Geologic Survey (USGS). 1996. Data from Selected U.S. Geological Survey National Stream Water-Quality Monitoring Networks (WQN) – DDS-37 CD-ROM. Reston, VA.

United States Geologic Survey (USGS) Earth Resources Observation Systems (EROS) Data Center (EDC). 1999. North American Land Cover Characteristics Database (NALCC), Sioux Falls, South Dakota http://edcdaac.usgs.gov/glcc/na_int.html

United States Geologic Survey (USGS). 1998. NWIS-W Historical Stream Discharge Archive. http://waterdata.usgs.gov/nwis-w/US/ Reston, VA

United States Geologic Survey (USGS). 2000.  Real-Time Water Data Pages, http://water.usgs.gov/realtime.html State offices (Maine, Massachusetts, New Hampshire)

Vörösmarty, C.J., B. Moore, M.P. Gildea, B. Peterson, J. Melillo, D. Kicklighter, J. Raich, E. Rastetter, and P. Steudler. 1989. A continental-scale model of water balance and fluvial transport: application to South America. Global Biogeochemical Cycles 3:241-365.

Vörösmarty, C.J., C.J. Wilmott, B.J. Choudhury, A. L. Schloss, T.K. Stearns, S.M. Robeson, and T.J. Dorman. 1996. Analyzing the discharge regime of a large tropical river through remote sensing, ground-based climatic data, and modeling. Water Resources Research 32:3137-50.

Vörösmarty, C.J., C.A. Federer, and A. Schloss. 1998. Potential evaporation functions compared on U.S. watersheds: Implications for global-scale water balance and terrestrial ecosystem modeling. Journal of Hydrology 207:147-69.

Vörösmarty, C.J. and B. Moore. 1991. Modeling basin-scale hydrology in support of physical climate and global biogeochemical studies: an example using the Zambezi River. Stud. Geophys 12:271-311.

Water Systems Analysis Group, University of New Hampshire (UNH). 1996. Data Assembler 1.3. Developed at the Water Systems Analysis Group, Complex Systems Research Center, University of New Hampshire, Durham, NH.

Water Systems Analysis Group, University of New Hampshire (UNH). 1996. Data Set Manager 2.0. Developed at the Water Systems Analysis Group, Complex Systems Research Center, University of New Hampshire, Durham, NH.

Water Systems Analysis Group, University of New Hampshire (UNH). 1994-2000. Global Hydrologic Archive and Analysis System-RiverGIS (GHAAS-RGIS) Version 2.1. Developed at the Water Systems Analysis Group, Complex Systems Research Center, University of New Hampshire, Durham, NH.

Willmott, C.J., C.M. Rowe, and W.D. Philpot. 1984. Spheremap. Center for Climatic Research, Department of Geography, University of Delaware, Newark DE.
 
 


List of Figures

1 Gulf of Maine Watershed Area 2-arc minute Mean Elevation Grid. This grid was created from the USGS EDC GTOPO30 30-arc second digital elevation model. GTOPO30 DEM was resampled using a nearest neighbor resampling algorithm in ArcGrid (ESRI).

2 Gulf of Maine Watershed Basins. Basins created from the edited Simulated Topology Network (STN) River Network. The subbasins river networks were edited to match the EDA/CDA Basins used in the Gulf of Maine Land Based Pollution Sources Inventory, a joint project between the National Oceanic and Atmospheric Administration (NOAA) and the Environmental Protection Agency (EPA).

3 Gulf of Maine Watershed 2-arc minute Simulated Topological Network (STN) River Network. This network was created using a branching search algorithm in GHAAS-RGIS from the Gulf of Maine Watershed Area 2-arc minute mean elevation grid.

4 Gulf of Maine Watershed 2-arc minute Land Cover Grid (Water Balance Model classification). This grid was created from the USGS EDC 1-km AVHRR North American Land Cover Classification Database. The grid was resampled and reclassified using ArcGrid (ESRI).

5 Gulf of Maine Watershed Area 2-arc minute Soil Texture Grid (Water Balance Model categories). This grid was created from the FAO Digital Soil Map of the World (original scale 1:5,000,000). The polygon coverage was converted to grid format in ArcInfo (ESRI) and the grid was reclassified in ArcGrid (ESRI).

6 Gulf of Maine Watershed Region Air Temperature and Precipitation Stations. These stations were compiled from the US National Climatic Data Center Archives and the Environment Canada Archives. These are the stations that were active between 1970 and the present. There are 913 air temperature stations and 1154 precipitation stations.

7a Gulf of Maine Watershed Winter (December, January, March) Precipitation Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean of total monthly precipitation for 1970-1993.

7b Gulf of Maine Watershed Spring (March, April, May) Precipitation Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean of total monthly precipitation for 1970-1993.

7c Gulf of Maine Watershed Summer (June, July, August) Precipitation Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean of total monthly precipitation for 1970-1993.

7d Gulf of Maine Watershed Fall (September, October, November) Precipitation Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean of total monthly precipitation for 1970-1993.

7e Gulf of Maine Watershed Total Annual Precipitation Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean of total annual precipitation for 1970-1993.

8a Gulf of Maine Watershed Winter (December, January, March) Air Temperature Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean monthly air temperature for 1970-1993.

8b Gulf of Maine Watershed Spring (March, April, May) Air Temperature Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean monthly air temperature for 1970-1993.

8c Gulf of Maine Watershed Summer (June, July, August) Air Temperature Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean monthly air temperature for 1970-1993.

8d Gulf of Maine Watershed Fall (September, October, November) Air Temperature Climatology Grids. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean monthly air temperature for 1970-1993.

8e Gulf of Maine Watershed Mean Annual Air Temperature Grid. These grids were created using the spheremap interpolation algorithm in Data Manager (DM) (WSAG UNH). The grids represent the mean annual air temperature for 1970-1993.

9 Gulf of Maine Watershed Stream Dischage Gage Locations. These stations were compiled from the U.S. Geologic Survey NWIS-W Historical Archive, the U.S. Geologic Survey Real-time, Provisional Data Web Sites, and the Environment Canada HY-DAT Archives.

10 Stream Discharge Gaging Stations Upstream Area Comparison between Reported Drainage Areas and Simulated Drainage Areas. This graph highlights the close fit in drainage area between the simulated basins and the reported basin area.

11 Gulf of Maine Watershed Cumulative Land Area Graph. This graph highlights the size of the large river basins within the Gulf of Maine Watershed. The St. John basin covers 32% of the GOM watershed area and the 10 largest rivers cover 75% of the GOM watershed area.

12a Gulf of Maine Watershed Water Balance Model Results – Winter (December, January, February) Potential Evapotranspiration Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

12b Gulf of Maine Watershed Water Balance Model Results – Spring (March, April, MAy) Potential Evapotranspiration Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

12c Gulf of Maine Watershed Water Balance Model Results – Summer (June, July, August) Potential Evapotranspiration Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

12d Gulf of Maine Watershed Water Balance Model Results – Fall (September, October, November) Potential Evapotranspiration Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

13a Gulf of Maine Watershed Water Balance Model Results – Winter (December, January, February) Runoff Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

13b Gulf of Maine Watershed Water Balance Model Results – Spring (March, April, May) Runoff Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

13c Gulf of Maine Watershed Water Balance Model Results – Summer (June, July, August) Runoff Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.

13d Gulf of Maine Watershed Water Balance Model  Results – Fall (September, October, November) Runoff Grids. These output grids are results from a WBM simulation which used the air temperature and precipitation climatology grids as input layers.