Environmental Evaluation Modeling System (EEMS)

The Environmental Evaluation Modeling System (EEMS) modeling system produces a high level metric across a landscape by combining different types of spatial data. Like EMDS (Ecosystem Management Decision Support), on which it is based, EEMS allows users to combine data of different types (e.g., road density percent, area of natural cover, and species presence/absence) in a tree-based logic model (Fig. 1) to provide answers to management questions such as Where is the landscape most ecologically intact? and What reporting units have the highest conservation value?

The strength of EEMS lies in its use of fuzzy logic. Fuzzy logic is based on the premise that there can be degrees of truth and degrees of truth can be quantified. For example, the weather might be considered “very nice,” “sort of nice,” “not nice but not bad,” “kind of bad,” or “really bad.” So the statement, “the weather is nice,” might be considered fully true, partially true, neither true nor false, somewhat false, or fully false. Fuzzy logic uses numerical values (called fuzzy values) associated with the level of trueness/falseness (Fig. 2) as well as a set of logic operators for combining fuzzy values.

An EEMS model operates from the bottom up. Each input data layer is translated into fuzzy values based on a simple user-defined conversion function. Moving up the tree, fuzzy logic operators combine fuzzy values, generating intermediate data layers before producing the final top-level layer. Each data layer is associated with one node in the EEMS logic tree. A data layer is composed of reporting units, or divisions of the larger landscape (think of counties within a state or cells in a map grid). Between layers, corresponding reporting units are related and affect values moving up the tree, but in a single layer all reporting units are independent and have no influence on each other. Each data layer can be displayed as a map layer using appropriate software (e.g., ArcMap, Data Basin, or a NetCDF viewer, depending on the data type).

EEMS is currently implemented for use in ArcGIS and for CSV (comma separated variable) and NetCDF data. The Arc version utilizes Arc ModelBuilder for building the tree-based model. A user only needs to install an Arc Toolbox -- no third party software is required. For use with CSV and NetCDF files, a user constructs the model using the EEMS scripting language. Download of EEMS python source files is required.

The EEMS Explorer in Data Basin

Data Basin allows users to explore files produced by EEMS using the EEMS explorer.

In the Current Terrestrial Landscape Intactness (1km) DRECP dataset (Fig 3.), the threshold values for converting input values into low linear development are 0 km per square km for Fully True (fuzzy value +1.0) and 2.5 km per square km for Fully False (fuzzy value -1.0). Values between 0 km and 2.5 km are assigned a value between Fully True and Fully False using linear interpolation. Data inputs for Low Linear Development (Ground Transportation Density, Utility Line Density and Pipeline Density) appear at the bottom of the diagram. The inputs are summed before being converted to fuzzy values.

CBI Scientists have used EEMS in a wide range of projects. For the Desert Renewable Energy Conservation Plan (DRECP), EEMS was used to model Terrestrial Landscape Intactness, Conservation Value, Climate Change Refugia, Climate Sensitivity, and others. For the Tehachapis and Southern Sierra, an EEMS model incorporated data on habitat presence, habitat linkage, and disturbance to find areas of high ecological value and to provide guidance for reserve design and wind energy siting. For the Bureau of Land Management Rapid Ecoregional Assessments (REA) of the Sonoran Desert and Colorado Plateau ecoregions, several EEMS models were developed and used to evaluate a variety of current and projected ecological conditions.

To listen to a recorded webinar about EEMS by Tim Sheehan, please click here. The webinar covers the fuzzy logic concepts EEMS uses, an overview to EEMS in ArcGIS, and a demonstration of the EEMS Model Explorer. To find datasets associated with an EEMS model, you can search for EEMS in the search box, located in the upper right corner of this site or browse datasets found within the gallery DRECP Implementation Models and Tools. Datasets will have an EEMS explorer icon in the capabilities section of the dataset description as well as an icon in the layer tab when opened in the map viewer (Fig. 4). To view the model and the dataset simultaneously in the map, click on the EEMS explorer icon in the layer tab.

A complete list of terms and definitions can be found by clicking on the “What is EEMS” link on the lower left hand side of the window (Fig. 5). To view the nodes in the EEMS model, you can search by scrolling in the model window or through the full model view. To display the data layer associated with an EEMS model node, click on the model node and the data layer will appear. Examining reporting unit values will enable user to see node values. The user can see what’s contributing to the reporting unit’s results by examining the input data at the bottom of the tree.