The Environmental Benefits Index Tool

The priority conservation areas were determined by using the Environmental Benefits Index tool.  The tool identified the highest potential for phosphorus reduction based on Soil Erosion Risk, Water Quality Risk, and Wildlife Habitat Quality.  For this project, Soil Erosion and Water Quality were weighted higher than Wildlife Quality, to identify the areas with the highest potential for conservation benefits to reduce phosphorus loading into the Kettle Watershed.

You may notice below that Pine County does not have all of the data to make this tool functional, so for areas that the EBI tool was unavailable, land use and proximity to surface water were the key components for prioritization.

Comprehensive EBI

This Environmental Benefits Index (EBI) is a composite score of multiple ecological benefits.  The score is based on a 0-300 scale, where a score of 300 is most valuable from a conservation perspective.  The EBI is the sum of the three independent layers: soil erosion risk, water quality risk, and a wildlife habitat quality layers. Each of those component layers contributes 0-100 points to the EBI.

This layer was created with the intention to rank CRP and other critical lands on multiple ecological benefits simultaneously. This approach is similar to the EBI used by the Farm Service Agency to rank farmers requests to enroll land in the Conservation Reserve Program. Our approach differs in that it offers flexibility in the weighting scheme, and allows users to explore both the spatial distribution of the data and the consequences of using alternative weighting systems. For example if, identifying lands of high soil erosion risk is important, the habitat quality and water quality risk maps can be downweighted (e.g. scaled from 0-50). This would produce a different map than when all attributes are weighted equally.


Water Quality Risk

Water Quality Risk

The risk score for Water Quality ranges from 0-100, with larger values indicating areas that are more likely to contribute overland runoff than smaller values.  This risk was defined by two data sources: Stream Power Index and Proximity to Water

Stream Power Index (SPI) measures the erosive power of overland flow as a function of local slope and upstream drainage area.  SPI was calculated statewide, but summarized by Terrain Zones, which represent physiographic regions of Minnesota with similar physiographic characteristics. The use of Terrain Zones removes bias from landscapes with extremely high relief.  Large SPI values (i.e. those in the 85thpercentile or higher) from each of the five terrain zones were used to create a critical area layer where overland erosion is likely to occur.  These critical SPI areas were summarized by SSURGO soil polygons: the proportion of SPI critical areas within each SSURGO polygon was used to assign a percentile rank to these polygons, the larger the proportion of critical SPI data, the larger risk score for that polygon.  This percentile rank represents 50 of the total 100 points for this risk layer.

The remainder of points was determined by calculating proximity from SSURGO polygons to the nearest DNR 24k surface water feature (Lake or Intermittent/perennial stream).  A percentile rank of these proximity values assigned to each SSURGO polygon represents the remaining 50 points, where the highest risk scores are given to the polygons closest to water features.


Soil Erosion Risk

Soil Erosion Risk

The potential for soil erosion is based on a number of factors, including climate, soil type, slope, and slope. We summarized this using factors from the Universal Soil Loss Equation. The Soil Erosion data layer represents a general risk score for potential soil erosion on a 0-100 point scale, 100 being the highest risk.  Larger values indicate soils that have a higher potential to erode if no conservation practices were in place and overland sheet or rill runoff was present.

A subset of the Universal Soil Loss Equation (USLE) was used to determine soil erosion risk values.  The USLE is a multiplicative equation using the formula A =R x K x LS x C x P where:

A = potential long term average annual soil loss in tons/acre/year

R = rainfall and runoff factor

K = soil erodibility factor

LS = slope length-gradient factor

C = crop/vegetation and management factor

P = support practice factor

The R (Rainfall), K (Soil Erodibility), and LS (Length/Slope) factors were used and calculated based on NRCS spatial and tabular SSURGO soils data, statewide county climate maps, as well as mathematical formulas based on standard USLE calculations.  SSURGO stands for Soil Survey Geographic Database.

The crop/vegetation and management factor and support practice factor were not used.  This is because there are no reliable statewide spatial data that represent these factors.  Although there exist statewide data depicting current cropping practices, there are no statewide data representing current tillage methods (e.g. fall plow, ridge tillage, no-till) or support practice (e.g. cross slope, contour farming, strip cropping) that are required for these calculations.  Furthermore these factors are temporal and will therefore shift over time.

Since only non-management factors were used, the resulting data layer should be viewed as a “worst-case” scenario, i.e. highest potential soil erosion of bare soil with no mitigating land use practices in place.  Although quantitative soils loss numbers (tons/acre/year) may be exaggerated under this model, the resulting data layer is used here in a qualitative, comparative capacity in order to compare the relative differences in soil loss risk between various parts of the landscape.


Wildlife Habitat Quality

Habitat Quality Mapping

The habitat mapping used in this plan was updated from the work done as part of Minnesota’s Statewide Conservation and Preservation Plan. The primary goal of habitat mapping was to collate the available information for Minnesota that can be used to prioritize important areas for conservation (protection, acquisition, restoration) by integrating both positive (resources) and negative (threats to resources) information on biodiversity, habitat quality, outdoor recreation (e.g., hunting and fishing), and water quality.  Positive components included features such as known occurrences of rare species, sites of biodiversity significance, or high levels of game species abundance, while negative components included the dominant drivers of environmental change as identified in Phase I of the plan. Negative influences on natural resources included such information as human development, land use, and road density.  By acquiring and objectively processing information related to these components, it was possible to rank areas in Minnesota according to their conservation priority.

The habitat analyses for the statewide plan are unique for several reasons. First, the analysis team comprised the major natural resource management agencies in the state, including several divisions of the DNR, the MPCA, BWSR, MN Dept of Agriculture, and others. This provided us with access to not only the most comprehensive and up-to-date statewide data sets, but also a wealth of expert knowledge, particularly as they relate to current issues facing the state. Second, the analyses were highly integrated: suites of habitat and stressor layers were combined using an additive modeling approach. This allowed us to generate composite maps of critical terrestrial and aquatic habitat that integrate across taxa and habitats, providing a ‘weight-of-evidence’ approach to the habitat rankings. Similarly, we were able to integrate data layers describing the fundamental drivers of change, using factors such as land use, population and road density, and other factors, to describe how environmental stressors, individually and cumulatively, are spatially distributed  across the state. Finally, the intersection of high-quality terrestrial and aquatic habitat with the composite environmental risk map identifies those regions of the state where critical habitats are most ‘at-risk’. To our knowledge, there have been few, if any, other statewide conservation assessments that have been able to conduct this kind of comprehensive assessment across the spectrum of natural resources.

High resolution data were used in this study; most of the data were derived or gridded to 30 m cells, the native resolution of the Landsat satellite imagery used for many of the statewide land cover classification and subsequent habitat analyses. These data were summarized, by township (terrestrial data) or lakeshed (watersheds surrounding lakes). The township summaries parallel the work of the state wildlife plan. Also, a key objective of the SCPP was to identify the general areas across the state with high conservation value, based on statewide data. For explicit land acquisition or planning purposes, it is necessary to conduct more specific analysis, using more detailed information that is available at local scales.

Twelve terrestrial data sets were identified and compiled from a variety of sources.  Each of these data sets were identified as important by the Land and Aquatic Habitat Conservation (LAHC) team and were, to the degree possible, available on a statewide basis.