A collection of research describing applications of GIS to assist in decision making and priority setting in watershed management.
Watershed management is a complex pursuit, encompassing hydrology, land-use, terrain, fish, groundwater, water quality, and many other factors. Regional problems and management goals vary based on differences in these characteristics, resources and training. A common theme is lack of time, money, and labor to address all management goals. Below is a list of research and resources describing GIS applications and tools to assist watershed managers in decision making and prioritization efforts in order to efficiently apply available resources to projects with the most potential for success. Riparian buffer zones are heavily emphasized as they represent a significant management practice to address many watershed issues, especially here in the Willamette Valley, Oregon. Wetlands constitute an important part of coastal watersheds and there are two resources in this area. Older articles were chosen based on their influence on later studies. Other topics include GIS-based decision support systems on the web, pollution assessment, fish habitat, and species selection for planting efforts.
________________________________________________________________
Meals, Donald W., E. Alan Cassell, David Hughell, Lynnette Wood, William E. Jokela, and Robert Parsons, 2008. Dynamic spatially explicit mass-balance modeling for targeted watershed phosphorous management: II. Model Application. Agriculture, Ecosystems and Environment (AGEE) 127:223-233. DOI: 10.1016/j.agee.2008.04.005.
Phosphorous (P) inputs, soil levels, and outputs in a watershed depend on a variety of factors and can be difficult and complex to assess and manage. Assessing effects of current or potential management practices is even more difficult. Meals et al. developed the dynamic interactive simulation of phosphorus loss areas (DISPLA) model to determine effects of three management practices and a combination of these practices over a long-term time scale (80 years) per 60 m cell. They tested the model in the Little Otter Creek (LOC) Watershed in Vermont. The watershed has mixed land use. A previously published paper (Meal et al. 2008) outlined the development of the model and the baseline (no management practices) results. This study looked at the effects of nutrient management, erosion control, change in land-use and all three combined on P soil tests and P outputs over an 80-year span.
DISPLA is based on a unit mass-balance model (PPBalModel) which calculated rates of entry and exit of P in a given pixel annually. The authors used ArcGIS in conjunction with the DISPLA model to compile, process and tabulate the data, and analyze spatial data needed to fulfill the model inputs. The model set prescribed changes in response to certain thresholds and triggers for each of the three management practices, starting in year ten of the model. The model provides for a minimum P input (needed to support crops) and maximum threshold at which each measure is triggered in a certain model year. In the combined analysis, the practices were staggered so that effects of changes could be distinguished. ArcGIS was also used to visually represent average pixel P exports and change per year of total P based on the different management practices at various years. The visualization allows for analysis of patches of high P soil levels and output areas to target for management. Analysis demonstrated that nutrient management had the highest impact on P exports and soil levels, erosion control had minimal impact and land-use change also has a significant impact, but not as much as nutrient management. As might be expected, the combination of all three had the highest impact of an average 74% decrease in P per pixel. The authors note that while farm management on 60 m grid basis is unrealistic, these results show the long-term trends and often multi-decadal lag between management implementation and a new equilibrium. Also, despite significant decreases in P as a result of nutrient management strategies, the soil P test and P export did not reach a balance during the duration of the model span. Even though this a simplified model, it is useful for demonstrating to landowners and managers the complexities involved in P management and a more appropriate temporal scale under which results should be considered.
Total Phosphorous Export
Maps of differences in P export from test watershed pixels between baseline and nutrient management scenarios simulated by DISPLA in year 20 (left panel) and in year 50 (right panel). Higher values represent lower TP export with nutrient management. Polygons outlined in red indicate zones of elevated initial soil test P. Areas of 1.0 RCAp are outlined in light blue.
________________________________________________________________
U.S. Environmental Protection Agency, BASINS (Better Assessment Science Integrating point & Non-point Sources). Accessed March 10, 2011, http://water.epa.gov/scitech/datait/models/basins/index.cfm.
Available for download from the EPA web site, BASINS 4.0 has been updated to now integrate an open-source GIS with national watershed data and various models for greater accessibility to watershed managers and decision makers at all government levels. BASINS allows for water quality assessment at certain stream sites or on a watershed scale based on point and non-point source data. This is another tool to assist watershed managers and decision makers in monitoring and targeting efforts towards areas of highest risk for water quality problems and Total Maximum Daily Loads (TMDLs) violations. Apart from access to an open source GIS and data downloads with automatic search for updated data layers and components, BASINS includes tools such as a project builder, manual and automatic watershed delineation, access to several hydrologic models and databases.
The web site provides download access, fact sheets, a user manual, access to a users’ listserver, FAQs, tutorials and downloadable lectures, metadata and links to more information about the included models, data layers and the open source GIS, MapWindow.
BASINS 4.0 GIS Interface
________________________________________________________________
Zhang, Yanli and Paul K. Barten, 2009. Watershed Forest Management Information System (WFMIS). Environmental Modelling & Software 24:569:575.
As with any watershed, managing a forested watershed requires considerations of many factors in order to adequately protect water quality. Zhang and Barten contend that many forested watershed assessment tools only consider single factors and therefore present the integrated Watershed Forest Management Information System (WFMIS) developed as an extension for ArcGIS 9.0 and higher. In WFMIS three sub-models focus on important elements of managing a forested watershed. The Watershed Management Priority Indices (WMPI) is intended to assist in prioritizing critical areas for restoration and conservation based on non-point source pollution. Using raster overlay, the multi-source GIS model generates a weighted total per pixel (Priority Index) of influence on water quality based on physical factors such as slope and permeability. This is compared with land cover classification of three broad categories, 1) forests and wetland (named conservation), 2) agriculture and open space (named restoration), and 3) residential, commercial and industrial (named stormwater management). The model results in a distribution of a priority index per land use classification, giving watershed managers a means to target locations based on goals (conservation, restoration or stormwater management) and priority ranking.
Results of the WMPI
Roads in a forested area contribute to sedimentation, but contribution varies based on type, slope, stream crossings, proximity to water bodies, and other factors. The Forest Road Evaluation System (FRES) evaluates these factors in relation to the road network and provides output maps enabling a manager to focus on reducing sediment load to reduce water treatment costs and improve aquatic habitat. Timber harvesting, as with roads, negatively impacts water quality in various ways. The Harvest Schedule Review System (HSRS) uses a devised disturbance threshold which can either be set as proportion of the area to be harvested or the proportion of biomass to be removed in the watershed. This disturbance threshold as well as a designated recovery time must be set by the user based on local conditions and knowledge. Once these are set, the model will delineate areas above and below the threshold for proposed harvesting, providing managers with information about harvest locations which may negatively impact water quality so that alternatives can be considered.
________________________________________________________________
Fengxia, GU anda LIU Wenbao, 2010.Applications of Remote Sensing and GIS to the Assessment of Riparian Zones for Environmental Restoration in Agricultural Watersheds. Geo-Spatial Information Science 13(4):263-268. DOI: 10.1007/s11806-010-0368-9.
This study in the Grand River watershed in Ontario, Canada used remote sensing data and ArcGIS 9.3 to prioritize areas by sub-basin for riparian restoration. This watershed has large agricultural area and is broken into 11 sub-basins. Eight potential buffer zones were examined in each sub-basin. The authors used Landsat TM data to define land use and established training points on the ground. Riparian buffer scenarios of eight different widths (ranging from 25 to 200 m) were created around a linear network hydrograph determined with DEM data. These buffer layers were converted to raster and the land cover maps resampled to finer resolution. These raster grids were overlaid in order to analyze area of land cover types in the various buffers by sub-basin. The sub-basins are of different sizes, so the results were reported in percent agricultural area per buffer zone for each of the buffer widths in each sub-basin. Eight of the eleven sub-basins had more than 50% agricultural area in the buffers, with four of these eight containing a large portion of the watershed’s headwaters. Fengxia and Wenbao report this analysis as unique by performing a sub-basin analysis of cropland area within buffers and by using varying buffer widths for the analysis. This is a relatively simple approach to give watershed managers a means of assessing high priority riparian buffer restoration areas in each sub-basin by focusing on lands that contain larger amount of relative agricultural area within the buffer area. Adding the consideration of potential effects on downstream waters, watershed managers can make decisions to target areas with the highest potential beneficial impact in an agricultural watershed. Additional considerations could include effects of soil types, wetness, slopes and vegetation types on riparian restoration decisions.
________________________________________________________________
Atkinson, Samuel F., Bruce A. Hunter, and April R. English, 2010. Prioritizing Riparian Corridors for Water Quality Protection in Urbanizing Watersheds. Journal of Water Resource and Protection (JWARP) 2:675-682. DOI: 10.4236/jwarp.2010.27078.
A Water Quality Corridor Management (WQCM) GIS model is proposed to prioritize highly functioning riparian areas for protection and preservation in urbanizing watersheds. Development of residential, commercial and industrial areas is known to degrade stream systems through increased impervious surfaces and stream-bank vegetation and tree clearing, thereby negatively impacting water quality and ecosystem equilibrium. Preserving already existing riparian areas is more economically efficient than trying to restore degraded areas and can serve as models for restoration work in other zones. The WQCM model weights (in parenthesis) the five categories of vegetation class (3), erosivity (potential for erosion) (2), slope (from 1% to 5%) (2), floodplain (defined as the ratio of FEMA designated floodplain to stream buffer area, meaning the larger area outside of the FEMA designated floodplain the more protection needed) (1), and corridor (the percentage of stream corridor within the subwatershed with a higher percentage requiring more protection) (1). In each category, the stream segments are given a ranking of 0 to 5. The model results in a score of 0 to 50. The model was applied to two watersheds with 133 and 90 sub-basins and validated with a subset of 40 sub-basins in the first watershed analyzed. The resulting stream segments and their scores were grouped into four categories, low, medium, high and highest; a higher score receiving higher priority for preservation and protection. This system represents a method for watershed managers and urban planners to prioritize for protection of the riparian areas providing the greatest services for water quality. The authors note that this data set could be used with an adjusted weighting and ranking system for prioritizing restoration efforts and are working on that model.
Prioritization for Protection of Functioning Riparian Areas
________________________________________________________________
Timm, Raymond K., Robert C. Wissmar, John W. Small, Thomas M. Leschine, Gino Lucchetti, 2004. A Screening Procedure for Prioritizing Riparian Management. Environmental Management 33(1):151-161. DOI: 10.1007/s00267-003-2980-z.
Timm et al. examine reaches in the Lower Cedar River Basin, near Seattle, WA. The spatially explicit linear additive model was designed in response to the Endangered Species Act (ESA) listing of salmonid species in the basin. A local government effort via the Tri-County Coalition for Puget Sound Counties identified salmonid-containing streams in the basin on which the model was applied to quantitatively prioritize stream reaches for protection and restoration. Using four defined habitat factors and four anthropogenic factors, the authors calculated “habitat potential” for eight different stream reaches to identify those that have highest potential habitat function for the ESA listed fish. Using raster grids of 5 m spatial resolution, a value was assigned to each grid for each of the eight factors and the eight grids were stacked and analyzed in ArcView 3.2 with the FRAGSTATS extension. The habitat factors analyzed were canopy, relic channels, wetlands, and slide-prone areas (considered a source of gravel important for spawning fish). A +1 value was assigned to a grid cell for the presence of these positive habitat factors or a 0 for absence. Conversely, a -1 value (or 0 for absence) was assigned to each grid in each grid layer for the anthropogenic factors of development (if more than ten percent of the area in the cell is impervious surface), zoning (based on defined classes considered encouraging of development), high real estate value, and channel constraints (e.g. roads, levees and revetments). The eight layers were summed for an unweighted index of habitat potential from -4 to +4. The authors then used a riparian area pre-defined as the Tri-County Inner Management Zone (IMZ) to perform an inverse distance weighting analysis to weight the habitat index. Additionally, for any weighted index over zero, the anthropogenic factors were filtered out to identify the difference between areas with only habitat factors present (i.e. 1+1+1+0= 3) compared to areas with a mix of positive habitat factors and negative anthropogenic factors present (i.e. 1+1+1+1-1=3) to further weight the more optimal habitat areas. Grids with like results were grouped as patches and mean patch size, number of patches and area fragmentation were also considered, resulting in a spatially explicit analysis of habitat zones in various reaches along the study area. Using this system provides fish and wildlife resource managers as well as city planners a quantitative system in which to prioritize riparian protection and restoration areas based on habitat function for endangered salmonid species. The authors note that the model can be applied to monitoring management efforts by updating the values assigned to the grids based on real or potential land-use changes and the model can be adapted to other habitat analysis depending on the target species.
________________________________________________________________
Hyatt, Timothy L., Tyson Z. Waldo, Timothy J. Beechie, 2004. A Watershed Scale Assessment of Riparian Forests, with Implications for Restoration. Restoration Ecology 12(2):175-183.
Much riparian restoration work in the Pacific Northwest (PNW) focuses on the presence of trees for shade, organic inputs and generation of large woody debris (LWD). LWD is considered essential for pool forming which has many functions including providing habitat for endangered salmonid species in the region. This study focused on stream segments in the Nooksack River Basin in Northwest Washington State which are known to have anadromous or resident salmonid species. Hyatt et al used remote sensing data and field observations to create a multiple regression model for predicting stream width, and then designated stream segments and associated riparian zones to those stream segments based on aerial photography assessments. Using GIS and stereo aerial photography, conditions of riparian stand polygons were mapped based on diameter class and then ranked by potential for LWD contributions. Rankings were established by comparing the stand’s diameter class to a minimum pool-forming diameter of LWD derived from an equation incorporating stream width. Pass indicates the stand’s diameter class is greater than the needed minimum diameter, fail if less than, and threshold for potential LWD contribution if the diameter class was in the same range as the minimum pool-forming diameter. Threshold ranking indicates a need for field verification and assessment. These results were mapped and further analyzed against other characteristics such as shade class and forest type. Random field verification indicated 69% agreement with model results. The authors discuss limitations of the model and factors such as landowner willingness that are not included in the assessment. The results of the model provide guidance in targeting areas for preservation (pass) and restoration (fail) and further assessment (threshold) in terms of potential contributions of LWD, especially as it relates to providing essential habitat for salmonids.
Riparian buffer analysis showing areas of pass, fail and threshold for potential LWD contributions
________________________________________________________________
Vennix, Sharon and William Northcott, 2004. Prioritizing Vegetative Buffer Strip Placement in an Agricultural Watershed. Journal of Spatial Hydrology 4(1):1-19.
The Agricultural Non-point Source Pollution (AGNPS) model was developed by USDA’s Agricultural Research Service to assist in identifying areas of higher pollution during a single storm event (10-year, 24-hour) in agricultural watersheds. He (2001) integrated AGNPS with an ArcView interface, creating the ArcView Non-point Source Model (AVNPS) which Vennix and Northcott used in the East Bad Creek (EBC) watershed in Michigan. A major pollutant identified in the watershed was sediment, so the objective was to identify and prioritize the most effective buffer strips for reducing sediment load. The authors defined seven stream segments in agricultural areas and used the AVNPS to define 30 m grid cells and the necessary 22 inputs for the AGNPS model using soil type and land-use cover maps and a DEM. They also made some assumptions for inputs such as worst-case scenario ranking for farming conservation practices and no impacts from pesticides and nutrients since these were not the focus of the study. The model ran baseline results (no buffer strips) and then a buffer strip scenario with 30 m buffer strips around the stream segments, resulting in a 17% decrease in sediment load at the watershed outlet with the buffers. Analysis on the stream segment, field and cell scales further contributed to identifying the most effective areas for buffer restoration to reduce sediment load entering the stream. At the cell scale, cells reducing sediment by more than 0.5 ton were shown, providing explanation for the efficiency in certain fields and segments over others. This analysis will aid watershed managers in targeting efforts on the most efficient buffer segments, reducing landowners targeted, field labor, time and cost.
Highest Efficiency Riparian Cells
________________________________________________________________
Xiang, Wei-Ning, 1996. GIS-based riparian buffer analysis: injecting geographic information into landscape planning. Landscape and Urban Planning 34:1-10.
In this study, GIS was used to assist a county to determine desirable riparian buffer width based on pollution detention time. Discrepancy with existing regulated buffers and an estimated cost of acquiring the discrepancy area were calculated using a GIS. The desired buffer width is calculated by comparing a buffer area to a conceived reference buffer using a reference buffer ratio, Manning roughness coefficient ratio, saturated hydraulic conductivity (as an indicator of permeability), slope, and soil moisture storage capacity. Soil, slope, hydrology and land-use data were overlaid so that each land area has values for all variables so that the buffer delineation model fgenerates the desirable buffer width. This value ranged from 7.9 m to 176 m depending on site characteristics. This vector data was then converted to raster grid for comparison with existing, regulated buffer zones to create a discrepancy grid layer. Xiang calculated that 37% of the desirable buffer area was outside of the already regulated buffer areas. Overlaying tax parcel data with the discrepancy layer and applying the average unit land value in dollars/square km based on the most recent appraised land values, he estimated approximately $30-52 million for acquiring the remaining desirable buffer area. Xiang reports the county’s interest in acquiring the land, but existing financial limitations. This analysis helps the county locate areas of the largest discrepancy and areas with highest cost efficiency for improving buffer areas. These results are useful to county planners for future acquisitions and buffer determinations.
________________________________________________________________
Carver, A.D., S.D. Danskin, J.J. Zaczek, J.C. Mangun, and K.W.J. Williard, 2004. A GIS Methodology for Generating Riparian Tree Planting Recommendations. Northern Journal of Applied Forestry (NJAF) 21(2):100-106.
This GIS-based decision support model is intended to assist watershed managers and landowners in improving success of riparian area tree-planting efforts. In the Cypress Creek Quadrangle area in Southern Illinois, the authors selected eight bottomland tree species to ultimately determine the tree species of highest suitability per 10 m pixel. This model incorporated digital soil surveys with spatial data to increase detail and automation. Using six characteristics adopted from Baker and Broadfoot’s (1977) valuation guide based on data available from the geo-referenced databases used, a suitability map containing standardized values associated with polygons was created and then converted into a raster grid. The six characteristics assessed were slope position, average soil pH, average depth to the water table, average depth of topsoil, frequency and duration of floods, and silt, sand and/or clay soil texture. All species were shown to be more suitable to the riparian areas (versus upland areas) as predicted. The results are considered consistent with current silvicultural practices, supporting the model output. These results support watershed managers’ and landowners’ efforts to plant in buffer zones, increasing potential for more productive growth. Higher productivity increases benefits from riparian areas, and decreases costs from unsuccessful planting efforts, as well as increasing timber benefits for the landowner. The expense of extensive site visits can be reduced with this automated tool. The study did not assess such factors as landowner support of planting efforts, landowner timber preferences, and other soil qualities not available through the spatial databases used.
Planting Recommendations by Species in the 50 m Buffer Area
________________________________________________________________
Russell, Gordon D., Charles P. Hawkins, Michael P. O’Neill, 1997. The Role of GIS in Selecting Sites for Riparian Restoration Based on Hydrology and Land Use. Society for Ecological Restoration 5(4S):56-68.
An early application of GIS to riparian buffer prioritization, Russell et al use a soil wetness index and land cover classification to determine wetland preservation and restoration priorities in San Luis Rey River watershed in southern California. At the time of the study, wetland mitigation focused restoration efforts “on site, in kind,” meaning restoring or creating sites as similar to and near lost wetlands as possible. This approach may restrict targeted areas and increase fragmentation, also neglecting important local concerns of flood attenuation and endangered species habitat. Russell et al used USGS DEM and Landsat image raster grids in ArcInfo with which they classified pixels into seven land cover categories. Land classification proved difficult; accuracy assessment with an error matrix created using aerial videography showed frequent misclassifications of the vegetated covers. Manual interpretation was used to improve accuracy. Using the wetness index overlaid with the land cover types, preservation and restoration priorities were established. Restoration areas were further refined based on patch size and proximity to existing wetland riparian areas. This is a simple tool that presented an early use of GIS to prioritize wetland riparian restoration and preservation on a watershed scale on which many others based their work in similar areas.
Prioritization scheme based on wetness index and land cover classification
________________________________________________________________
Kauffman-Axelrod, Jennifer L. and Steven J. Steinberg, 2010. Development and Application of an Automated GIS Based Evaluation to Prioritize Wetland Restoration Opportunities. Wetlands 30:437-448. DOI: 10. 1007/s13157-010-0061-7.
Estuaries and wetlands are important areas of Oregon’s coastal watersheds. Over 2,000 potential wetland restoration areas had recently been defined in Oregon, with over 530 in the Coos watershed. The Spatial Wetland Assessment for Management and Planning (SWAMP) exists as extension to ArcView 3.x, but the authors find many disadvantages with this tool. Kauffman-Axelrod and Steinberg used ArcGIS 9.x’s ModelBuilder to calculate standardized, weighted prioritization rankings for the 530 potential restoration sites and catchments in the coastal watershed to identify the high priority wetland areas to target for field feasibility studies. ModelBuilder was selected as a tool that is transparent, adaptable, repeatable and transferable. Nine characteristics were evaluated and weighted; a separate model was built for eight of the nine and a PYTHON script was created for calculating adjacent wetland area. Standardized, weighted rankings were calculated, a higher score indicating less overall disturbance and better landscape characteristics, representing better potential for successful restoration. This method provides an objective and quantifiable prioritization method that will save money and time determining where to target resources for assessment and restoration. The model is intended to be adaptable to different coastal watersheds in Oregon and can be applied to other Pacific coastal watersheds with the appropriate data layers and weighting adjustments as needed.
Model Results Show Priority Wetlands in the Coos Watershed
________________________________________________________________
U.S. Environmental Protection Agency, Automated Geospatial Watershed Assessment Tool. Accessed March 10, 2011, http://www.epa.gov/nerlesd1/land-sci/agwa/.
The U.S. Environmental Protection Agency (EPA) in conjunction with the U.S. Department of Agriculture’s (USDA) Agricultural Research Service (ARS) and the University of Arizona developed the Automated Geospatial Watershed Assessment Tool (AGWA). AGWA is a GIS interface which combines two hydrologic models – the Soil and Water Assessment Tool (SWAT) and the KINematic Runoff and EROSion (KINEROS2). AGWA is available in version 1.5 for ArcView or 2.0 for ArcGIS. Outputs include peak and volume runoff, sediment yield, and nitrogen (N) and phosphorous (P) from SWAT. GIS data layers such as DEM, soil maps, precipitation, and land cover are used to run the model and display results. Current conditions as well as potential changes such as proposed or considered land use changes, climate/precipitation changes, pre- and post-fire conditions, buffer zones and addition of retention structures can be analyzed for watershed managers to assess and predict impacts of change on erosion, sediment yield and water quality. This can assist managers and decision makers in focusing monitoring or restoration activities and make the efficient use of available funds for the most positive impact or protective measures.
The EPA web site includes general information and links to more information about AGWA, SWAT and KINEROS, as well as a variety of resources such as fact sheets, a user manual, quality assurance plan, and updates under Basic Information. AGWA is available for download from the EPA or the University of Arizona AGWA web sites.
Visualization of AGWA Inputs
________________________________________________________________
Miller, Ryan C., D. Phillip Guertin, and Philip Heilman, 2004. Information Technology in Watershed Management Decision Making. Journal of the American Water Resources Association (JAWRA) 40(2):347-357.
Watershed management is increasingly moving towards a bottom-up instead of a top-down decision making process, involving stakeholders early and throughout the process. Education and access to technology-based decision making tools such as hydrologic models and GIS will support and improve these efforts. An internet-based decision making support system (DSS) can facilitate the use of these tools by watershed managers and foster the informed participation of stakeholders in the watershed management process. Miller et al. describe the problems facing use of information technology by watershed managers and stakeholders, specifically interoperability, security, access to internet, and bandwidth. They present a prototype internet-based spatial DSS that addresses interoperability and security and does not require downloads and user maintenance of the system or data. Access and bandwidth are not addressed by the prototype system.
Targeting rangeland watershed management, the prototype system brings the Kinematic Runoff and Erosion Model (KINEROS) component of the Automate Geospatial Watershed Assessment (AGWA) application to a web-based application with ArcGIS 8.3 by ESRI. Users can manipulate the model and change inputs and management practices in the model to simulate effects of land-use change, precipitation and test “what if” scenarios on runoff, sediment yield and cost. This allows for increasingly informed decisions regarding best management practices (BMPs), land use, vegetation cover, water sources, and other factors. The system can be used in a single session or form-based authentication is available for saving data and simulations to return to later. Availability of an internet-based DSS increases transfer of technological advancements in watershed management to decision makers and stakeholders for an improved decision making process.
________________________________________________________________
Sugumaran, Ramanathan, James C. Meyer, and Jim Davis, 2004. A Wed-based environmental decision support system (WEDSS) for environmental planning and watershed management. Journal of Geographical Systems 6:307-322. DOI: 10.1007/s10109-04-0137-0.
In order to address negative impacts of developing and increasing urbanization, Sugumaran et al. designed a Web-based decision support system (WEDSS) for the purposes of prioritizing sensitive watersheds among the 23 watersheds within Boone County, Missouri. The city of Colombia in Boone County experienced rapid population growth of 20.5% between 1990 and 2000, spurring concerns of effects on surface and ground water quality and storm water management. The WEDSS combines a client-side graphical user interface (GUI) on a web browser and server-side internet information server (IIS) and GIS server, in this case, Microsoft IIS and ArcView Internet Map Server (AvIMS), respectively. A database management system was created with 13 characteristics of environmental sensitivity which would be calculated from both spatial and non-spatial data and then weighted in the model. The model management system created a Multi-Criteria Evaluation (MCE) with weighted linear combination, selected based on its simplicity for a broad user base via the internet.
A GIS layer was created for each of the 13 criterion. Each watershed was assigned a value based on average data or presence/absence of certain elements (e.g. endangered species or overall stream health), a standardized rating value was calculated and summed with the weighted scale creating an overall Environmental Sustainability Index (ESI) for the watersheds. These results can be displayed in a map based on user requests on the client-side with a web browser. The users can zoom and pan, measure lengths and distances, include cartographic elements and change the weighting values if desired. Other features include transparency for comparison with overlay. The WEDSS provides users with access to watershed environmental sensitivity information on which to base prioritization decisions on a county-wide scale. The system can be adapted to other geographic regions or to the sub-basin scale using GIS to delineate the basins. Future work is planned to update the system to the ESRI Internet Map Server.
WEDSS in Action
Contact: Kristen Larson
larsonkr@onid.orst.edu
References are in the style of the Journal of the American Water Resources Association (JAWRA) as per the Instructions for Authors.
