Southwest Regional Gap Analysis Program (SWReGAP) for Nevada, Utah, New Mexico, Colorado, and Arizona - Project Status
GAP Program Overview: The Gap Analysis Program is sponsored by the Biological Resources Discipline of the U.S. Geological Survey. A “gap” is the lack of representation or under-representation of an element of biodiversity (plant community or animal species) in an area intended for its long-term conservation management. Gap analysis is a process to keep common species common by identifying the gaps in our network of lands managed for biodiversity. The GAP methodology and its products are straightforward: 1) map the distributions of natural plant communities, 2) map predicted distributions of native terrestrial vertebrate species, 3) map the degree of management for biodiversity maintenance, and 4) analyze the representation of vegetation and animal species distributions in the conservation network to identify “gaps” in long-term security. This type of coarse-filter approach is used for land management and conservation planning purposes by land managers, scientists, and policy-makers at both regional and ecosystem levels.
The SWReGAP Project in Nevada
Field Data Collection Methodology: Data for land cover modeling has two critical characteristics: 1) it samples the range of environmental settings in a landscape, and 2) it contains enough data to robustly model each plant community of the classification. To classify the vegetation of Nevada thus requires thousands of training sites. Because of the large volume of data required, only enough time was spent at each site to collect the minimum amount of ecological detail to confidently classify it via the National Vegetation Classification System. Field crews selected training sites opportunistically based on homogeneity of plant species composition, landform, and spectral characteristics. Three essential types of data were collected at each training site: 1) characterization of the plant species composition and the ecological setting of the landscape, 2) a hand-digitized polygon delineating the training site's location, and 3) a digital photograph of the site for documentation and future reference. See Field Methodologies and Training Manual for Nevada Field Crews.(3929 KB)
Land Cover Mapping/Modeling: Once all field data for a region, or map zone (Figure 1), were collected, the hand-digitized polygons were intersected through contemporary Landsat 7 Thematic Mapper imagery, digital elevation models, digital climatic data, and ancillary GIS data layers to produce a large, multivariate dataset used to predict the geographic distribution of each plant community. The Nevada Landsat dataset utilized 63 triplicate ETM+ scenes (i.e. spring [leaf-on], summer [peak greenness], and fall [leaf-off]) from 1999-2001. Research has indicated that classification and regression trees (CART) tend to have greater predictive accuracy than traditional land cover mapping methodologies. Therefore, the SWReGAP modeling approach is centered on the use of classification and regression trees. CARTs recursively partition a dataset into increasingly “pure” subsets based on a multitude of predictor variables. The output of a classification tree is a set of decision rules that is implemented in an image processing software package.
Summary of 2002/2003 Vegetation Training Site Collection and Land Cover Mapping
From February 2002 - October 2003, approximately 17,500 field training sites were collected across the five mega map zones for Nevada (Figure 2) . Specifically, 1,000 sites were collected for the Sheldon Mapping Unit, 4,700 sites for the Lahontan Mapping Unit, 800 sites for the Reno/Tahoe Mapping Unit, 8,000 sites for the Great Basin Mapping Unit, and 3,000 sites for the Mohave Mapping Unit (Figure 3). The target level of classification within the National Vegetation Classification is the Ecological System. An Ecological System within the National Vegetation Classification nomenclature is described as groups of plant community types that tend to co-occur within landscapes with similar ecological processes, substrates, and/or environmental gradients (and spectral characteristics). Seventy-three natural and semi-natural ecological system land cover types were mapped for Nevada. The entire 5-state project area (535,175 sq. miles) was mapped to 125 land cover classes using 1.5 billion 30-meter Landsat pixels. An internal validation of the modeling effort estimated the predictive capability of our modeling approach at 61% accuracy. The minimum mapping unit for the dataset is 1.1 acre (0.45 ha). Land cover datasets and metadata can be downloaded from http://earth.gis.usu.edu/swgap/. Also see http://www.epa.gov/esd/land-sci/nv_geospatial/nv_geospatial_data_browser.htm for information as it specifically pertains to Nevada.
Animal Habitat Modeling
A second key task was the development of seamless animal-habitat models for all terrestrial vertebrate species for the Southwest region. Through the collaborative efforts of participating state, federal, and non-governmental organizations, 819 seamless habitat models were completed for the project area by September 2005.
The Southwest Regional Gap Analysis Project predicted habitat for terrestrial vertebrate species that reside, breed, or use habitat in the five-state region for a substantial portion of the their life history; approximately 520 species are estimated to occur within the state of Nevada.
These species habitat models are based on the concept of Wildlife Habitat Relationships (WHRs). We have defined WHRs as a statement describing resources and conditions present in areas where a species persists and reproduces or otherwise occurs. Relationships can be modeled to predict habitat composition, and if the relationships are represented in a cartographic plane they can predict the presence of habitat spatially. For each species, these relationships were identified by reviewing the available literature and then generating a spatial representation of habitat within the species known range.
An important factor for model implementation is to understand the objectives of the modeling effort and the assumptions of the models. The objectives of the habitat models are to: 1) provide maps that predict the distribution of terrestrial vertebrate species in the project area to support analysis of conservation status; and 2) develop a database of geographic range, wildlife habitat relationships, and predicted distribution of each vertebrate species for the long-term utility of GAP and its cooperators. The following assumptions apply to the GAP vertebrate habitat models:
- Species are assumed to occur within a polygon representing potential habitat but are not predicted to occur at any particular point within that polygon.
- Species are assumed to be present within a polygon, but no assumptions are made about the abundance of the species in the polygon.
- Species are assumed to be present in a polygon at least once in the last 10 years but need not be present every year in the last decade.
- Species are assumed to be present during some portion of their life history, not necessarily during the entire year.
All data for the SWReGAP vertebrate species are currently available on a ‘provisional’ status at http://fws-nmcfwru.nmsu.edu/swregap/HabitatModels/default.htm. All models were reviewed by taxon experts and various agency representatives relative to the wildlife habitat relationship models, range extent, and spatial depiction of predicted habitat for each species.
Land Stewardship Mapping
The third key task was the development of a seamless land stewardship map for the region. Through the collaborative efforts of participating state, federal, and non-governmental organizations a seamless product was completed by September 2005.
The two primary goals of SWReGAP are to provide an assessment of the management status for certain elements of biodiversity (vegetation communities and animal species) throughout the 5 Southwestern states, and to provide land stewards with information on the representation of these elements on their land so they can make informed decisions about their management practices regarding biodiversity.
To accomplish this, the mapped distributions of vegetation communities are compared to a map of land stewardship. In GAP, the land stewardship map combines attributes of ownership, management, and a measure of intent to maintain biodiversity. These comparisons do not consider viability, but provide a beginning to assess the likelihood of future habitat conversion—the most obvious cause of biodiversity decline.
We use the term “stewardship,” because legal ownership of a land area does not necessarily equate to the entity charged with managing the resource. Though we record the management and ownership entities of public lands and privately owned conservation lands, we also acknowledge that these attributes are complex and change rapidly. At the same time, it is necessary to distinguish between stewardship and biodiversity management status in that a single land steward, such as the Bureau of Land Management, may subdivide its land into units that may be managed for different purposes that affect biodiversity.
The four biodiversity management status categories can generally be defined as follows:
Status 1: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a natural state within which disturbance events (of natural type, frequency, intensity, and legacy) are allowed to proceed without interference or are mimicked through management.
Status 2: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a primarily natural state, but which may receive uses or management practices that degrade the quality of existing natural communities, including suppression of natural disturbance.
Status 3: An area having permanent protection from conversion of natural land cover for the majority of the area, but subject to extractive uses of either a broad, low-intensity type (e.g., logging) or localized intense type (e.g., mining). It also confers protection to federally listed endangered and threatened species throughout the area.
Status 4: There are no known public or private institutional mandates or legally recognized easements or deed restrictions held by the managing entity to prevent conversion of natural habitat types to anthropogenic habitat types. The area generally allows conversion to unnatural land cover throughout.
There are three primary pieces of information involved in developing the Stewardship coverage:
- Geographic boundaries of public land ownership (and voluntarily provided private conservation lands, e.g., Nature Conservancy Preserves);
- The manager/owner attributes of each mapped land unit; and
- The biodiversity management status category of each mapped unit. The importance of this map in the GAP process is not only for the spatial documentation of the existing network of conservation lands, it is also the base map from which future designs for the conservation network will come.
All data for the SWReGAP the stewardship maps are currently available on a ‘provisional’ status at http://fws-nmcfwru.nmsu.edu/swregap/Stewardship/Default.htm. Also see http://www.epa.gov/esd/land-sci/nv_geospatial/nv_geospatial_data_browser.htm for information as it specifically pertains to Nevada. All land ownership and land conservation status maps were reviewed by various agency representatives and iterated to a final product. Actual Gap analysis will include the process of combining the stewardship dataset with the land cover or habitat models to identify the representation of protection of these biodiversity elements.
The Nevada Geospatial Data Browser
The Landscape Ecology Branch of the U.S. Environmental Protection Agency (Las Vegas, NV) with the assistance and collaboration of the University of Idaho (Moscow, ID) and Lockheed-Martin Environmental Services (Las Vegas, NV) has developed the Nevada Geospatial Data Browser, a spatial data archive to centralize and distribute the geospatial data used to create the land cover, vertebrate habitat models, and land ownership/stewardship maps produced for the Nevada ecoregional component of the SWReGAP project. The purpose of the data browser is to provide a one-stop, easy-access product for the user community to assist in natural resource management and improve environmental decision-making. The Nevada Geospatial Data Browser utilizes data from a number of sources and has assembled 38 complete GIS datasets into 10 data categories (land cover maps, land cover training data, digital elevation model, soils and geology, climate data, ecoregional boundaries, political boundaries, hydrology, miscellaneous land cover [i.e., sand dunes and fire history], and miscellaneous vector data [i.e., roads, quad boundaries, Landsat scene boundaries, cities and towns]) for the entire state of Nevada. The data browser includes important metadata information relative to acquisition, location, processing level, projection, file size, and format. The Nevada Geospatial Data Browser is currently available on-line via the EPA Web site (http://www.epa.gov/nerlesd1/land-sci/nv_geospatial/nv_geospatial_data_browser.htm) and distributed in limited release on DVD.
Recommended citation: Kepner, W. G., T.D. Sajwaj, D. F. Bradford, and E. J. Evanson. 2005. Nevada Geospatial Data Browser. EPA/600/C-05/005. U.S. Environmental Protection Agency, Office of Research and Development, Las Vegas, NV ( http://www.epa.gov/nerlesd1/land-sci/gap.htm).