Determination of input factors/weightings

Wetland Prioritization Study Main Page

 

Stakeholder/expert feedback

  • Arkansas Multi-Agency Wetland Planning Team
  • Caltrans Regional Advance Mitigation Planning
  • Maryland Watershed Resource Registry
  • NOAA Habitat Priority Planner Mississippi-Alabama Habitat Tool
  • TNC-ELI Duck-Pensaukee Watershed Approach Pilot
  • USACE Sunrise River Watershed-Based Mitigation Pilot

 

Analysis of field data

  • Idaho Department of Fish and Game
  • Montana Natural Heritage Program
  • Weller et al. (2007)

 


 

Stakeholder/expert feedback

 

Arkansas Multi-Agency Wetland Planning Team:1 After identifying a set of objectives, the team draws from the spatial datasets assembled for the wetland planning area (WPA) to develop quantitative factors that will represent each objective in the GIS model. For example, if the team identifies water quality as a prioritization objective, it may use a map of riparian corridors developed for the WPA to design an input layer that rates areas (30m2 pixels) higher as potential sites for wetland protection or restoration that are adjacent to a riparian corridor. The Multi-Agency Wetland Planning Team (MAWPT) employs the expertise of state wetland planners to determine whether priorities identified by the GIS tool meet WPA needs. If not, the team may reformulate WPA objectives or factors and rerun the prioritization analysis.

 

Caltrans Regional Advance Mitigation Planning (RAMP):2 The Caltrans' MARXAN greenprint analysis incorporates conservation targets defined by multiple stakeholders. In Thorne et al. (2009), habitat conservation targets defined by local stakeholders were inputted into the MARXAN greenprint analysis. For example, in its analysis of the Elkhorn Slough watershed, RAMP consulted with the Elkhorn Slough Foundation to identify habitat types and extents (i.e., weightings) that local stakeholders considered to be ecologically desirable for protection (e.g., 30% freshwater wetlands). RAMP then used these target habitat percentages as inputs in the MARXAN reserve selection algorithm. In another example, through an analysis of impacts to different habitat types, RAMP determines the required extent of habitat that would need to be mitigated based on typical mitigation ratios for each habitat type through an analysis of impacts to different habitat types. These habitat extents serve as inputs for the MARXAN greenprint analysis.

 

Maryland Watershed Resource Registry (WRR):3 WRR was developed by a multi-partner technical advisory committee that met regularly from 2008-2011 to select data inputs and set weightings for each GIS tool.

 

NOAA Habitat Priority Planner Mississippi-Alabama Habitat Tool:4 For NOAA's habitat priority planner tool, the developers facilitated the Coastal Habitats Coordination Team (CHCT) in identifying prioritization criteria for each of its ten priority habitat types (four of which were aquatic). The CHCT consisted of more than 60 state and local scientists, non-profit staff, environmental professionals (consultants), and local/state officials.

 

USACE Sunrise River Watershed-Based Mitigation Pilot:5 In a series of workshops, a stakeholder team collaborated to develop a framework for selecting mitigation sites that would best meet watershed needs. This stakeholder team consisted of representatives from the U.S. Environmental Protection Agency (EPA), Minnesota Pollution Control Agency (MNPCA), Minnesota Department of Natural Resources (MNDNR), Minnesota Board of Soil and Water (MNBSW), and local agencies responsible for implementing the Minnesota Wetland Conservation Act. This process, which was administered by the Corps, involved three phases:

1.

Identification of watershed vulnerability priorities: the stakeholder team identified subwatersheds within the Sunrise River watershed that it considered to be priorities for mitigation projects by drawing upon a baseline analysis that was prepared by the Corps.

2.

Identification of criteria used for site prioritization: the stakeholder team identified criteria that it considered to be most important for targeting wetland compensation mitigation efforts within the each subwatershed. These criteria included:

  • Hydrologic connection to tributaries
  • Land costs
  • Potential to reconnect riparian buffers
  • Potential beneficial effects on fisheriesf
  • Threats from urban growth
  • Adjacency to public lands
  • Opportunities to improve or protect areas of significant biodiversity
  • Distance from roads and population centers
  • Locations within the floodplain of a tributary
  • Opportunities to improve water quality impairments

An analysis of the input received during this process, which will identify overall stakeholder priorities for each subwatershed, remains under development.

3.

Following the workshops, stakeholders completed a web-based survey in which they ranked selected criteria against one another in a series of pairwise comparisons (see below). Survey results were used to assess the overall importance of each criterion to the group as a whole using the Analytic Hierarchy Process (AHP), a type of Multi-Criteria Decision Analysis (MCDA). These importance ratings were then used to determine the weightings to use for each criterion as part of the Spatial Decision Support System (SDSS) model. MCDA methods such as AHP provide a transparent, structured decisionmaking process for identifying stakeholder preferences based on complex, disparate, and conflicting preference data.6 The survey was completed online, rather than as a group, to minimize bias and avoid concerns related to group think.

 

 

Back to top

 


 

Analysis of field data

 

Idaho Department of Fish and Game:7 To identify input factors/weightings for its model, IDFG first compiled a list of as many spatial layers as possible that could potentially serve as indicators of wetland condition based on a review of existing models and existing spatial data for Idaho (e.g., percent urban coverage, population density, etc.). IDFG then applied the Analytical Tools Interface for Landscape Assessments (ATtILA) tool8 (an ArcView 3.x extension) to these layers to calculate landscape metrics for 20,158 total wetlands.2

Using statistical analysis, IDFG then correlated each of these landscape datasets with four different field-based data sources for wetland sites throughout the study areas to evaluate how well each metric correlated with on-the-ground wetland conditions. These field data included:

  • Streams, rivers, and lakes obtained as part of the Idaho Department of Environmental Quality (IDEQ)'s Beneficial Use Reconnaissance Program (BURP) dataset.
  • Riparian and aquatic habitat maintained by the PACFISH/INFISH Biological Opinion Effectiveness Monitoring Program (PIBO).
  • Wetland sites and plant communities, including ecological indicators, maintained by the IDFG Idaho Conservation Data Center databases.
  • An IDFG-developed rapid wetland assessment applied to ensure adequate representation of a variety of wetland environments across the landscape.

Those metrics that passed a five-part screening process, which included criteria such as ecological relevance, range of values, and significance of correlation with field conditions, were considered most predictive of wetland condition. Additionally, metrics found to be negatively correlated with wetland condition (e.g., elevation) were used to calculate an "index of environmental vulnerability" for each wetland.2

 

Montana Natural Heritage Program (MTNHP):9,10 MTNHP began development of the Landscape Integrity Model (MTLIM) with an attempt to find landscape-level predictors of wetland condition. The original data set came from thousands of Montana Department of Environmental Quality Rapid Assessment Method (MTDEQ-RAM) sites throughout the state. Many of these assessments were part of a statewide project to evaluate amphibian habitat. MTNHP built a GIS layer of landscape level stressors based on the literature and expert judgment, then used a Classification and Regression Tree (CART) analysis to select those with the greatest predictive value. In central eastern Montana, where there is a high density of seldom used roads, but heavy livestock use, no landscape predictors were particularly reliable. In western Montana, road density showed some predictive value for all wetland conditions.2,3

 

Weller et al. (2007):11 This approach developed landscape assessment models that predicted wetland condition for flat and riverine wetlands in the Nanticoke watershed. Researchers first applied EPA's Environmental Monitoring and Assessment Progam sample design to obtain five different field-based Functional Capacity Index (FCI) scores (based on HGM variables) for riverine wetlands, with four of these five FCI scores obtained for flat wetlands. Researchers then used regression analysis to evaluate relationships between each of these nine FCI scores and 27 landscape indicators to identify nine sets of landscape indicators that best predicted FCI scores. Through this analysis the researchers generated nine equations (four for flat wetlands and five for riverine) that predicted scores for each of the nine FCI models. The underlying methods applied by these researchers can be reapplied to prioritize wetland restoration or conservation for wetland types of any watershed for which a random sample of Rapid Assessment Method (RAM) scores can be obtained.

 

Back to top

 

Wetland Prioritization Study Main Page

 


 

1The Multi-Agency Wetland Planning Team. The Standard GIS Methodology for Wetland Analysis. Accessible from: www.mawpt.org/pdfs/Standard_Methodology_of_Analysis.pdf.
2 Thorne JH, Huber PR, Girvetz EH, Quinn J, and McCoy MC. 2009. Integration of regional mitigation assessment and conservation planning. Ecology and Society 14(1): 47
3 Interviews on 8/3/2011 with Ellen Bryson, USACE Baltimore District, and on 8/11/2011 with Ralph Spagnolo, USEPA Region III.
4 Feedback provided on 4/26/2012 by Roberta Swann, Director of Mobile Bay National Estuary Program.
5 Webinar "Watershed Based Identification and Evaluation of Compensatory Mitigation Site." Presented by Timothy Smith and Thomas Mings, U.S. Army Corps of Engineers, St. Paul District.
6Smith T, Burks-Copes KA. 2010. Development of a GIS-Based Spatial Decision Support System to Target Potential Compensatory Mitigation Sites in Minnesota. National Wetlands Newsletter 32(6) 14-15.
7 Idaho Department of Fish and Game. 2010. Development of a landscape-scale wetland condition assessment tool for Idaho.
8 ATtILA was developed by the EPA Landscape Ecology Branch and is available for download at: http://www.epa.gov/nerlesd1/land-sci/attila/index.htm
9
Feedback received on 5/15/2012 from Linda Vance, Senior Ecologist/Spatial Analysis Lab Director, Montana Natural Heritage Program.
10 Vance LK. 2009. Assessing Wetland Condition with GIS: A Landscape Integrity Model for Montana. A Report to The Montana Department of Environmental Quality and The Environmental Protection Agency. Montana Natural Heritage Program, Helena, MT. 23 pp. plus appendices.
11 Weller DE, Snyder MN, Whigham DF, Jacobs AD, and Jordan TE. 2007. Landscape indicators of wetland condition in the Nanticoke River watershed, Maryland and Delaware, USA. Wetlands 27(3) 498-514.