Tracey Frescino
AAron Poe
Chuck Werstak
Spring 1996
The realization that individual populations of organisms cannot survive without adequate space and resources has spawned many programs designed to identify and protect so called "critical habitat." Such protection for threatened and endangered species is now mandated by the Endangered Species act of 1973. In order for land managers to comply with this legislation it is helpful to have species' specific habitat relational models designed around the ecological requirements of each protected species.
The U.S Forest Service recognizes the following six Endangered/Threatened and Sensitive (ETS) species as occuring within Cache County.
 |
The purpose of our project is to make the public and local managers aware of the location of critical habitats for these ETS plants in Cache County. We created a model of ETS habitats based on available information for the individual ETS species' requirements. The model isolates potential areas in Cache County where each plant is likely to be found. The specific habitat information was integrated by overlaying various ecological coverages using Geographical Information Systems (GIS). We then were able to generate a map of locations where each species is predicted to exist.
There are many factors that contribute to the unique requirements of the survival of a plant. This study reflects the abiotic and biotic habitat characteristics specific to each ETS plant. The ETS habitat data was isolated from a compilation of references based on past research and site evaluations of the plants. We separated this information into different thematic layers of ecological features including: vegetative cover type, elevational distribution, aspect qualifications, geological characterics, and hydrologic features. Habitat requirements for three of the six ETS plants were unavailable or not unique enough to determine with high accuracy possible locations of these plants in Cache County. Our study focuses on the ETS plants that have sufficient data for an adequate depiction of critical habitats: Brownie ladyslipper, Cronquist daisy, and Maguire primrose. After creating thematic layers of aspects in Cache County specific to each of these plants, we intersected the layers, one at a time, using GIS applications.
Vegetation is one of the most widely accepted indicators for the locations of terrestrial plants (Austin 1991). The classification of vegetation is generally based on the distribution of the dominant species, most of which frequent general habitat types (Edwards, et al. 1995). Due to the sedentary lifestyle of plants, the satellite coverage aspect of GIS is an accurate reference of vegetative cover for the habitat characteristics of ETS plants. The source image we used for delineating vegetative cover types was the Final Utah Gap Vegetation Image, December 1994. Although ETS plants are frequently microsite dwellers, they have been found to consistantly occur within common vegetation types that we selected from this image. Using information from various sources, we were able to generate a map of the vegetation zones of each plant. Specific vegetative data were available for each ETS plant on the list.
Using RECODE from the GIS ANALYSIS tools, we were able to manipulate the attribute table of the Gap Vegetation Image in Imagine . Looking at each endangered plant individually, we created six new images of vegetation habitat. Each image has a unique attribute table that represents the specific habitat type of the associated plant. An example of the vegetative coverage of Brownie ladyslipper after recoding is as follows:
Brownie ladyslipper
(Cypripedium fasciculatum)
Elevational distribution was the next aspect we used to define each ETS plant's habitat. We used ARC, ARC/GRID and Imagine to process the altitudinal features and intersect this data with the repective species' vegetative characteristics. We began by creating a text document (look-up table) of the specific elevation ranges of each ETS plant. A separate grid was generated for each species using the RECLASS command in ARC/GRID. The RECLASS command's usage required the specific coverage being reclassed, the respective look-up table, and whether or not we wanted to include the rest of the data in the grid (data/nodata). The source image for this procedure was the DMA (3-Arc Second Digital Elevation Data; 90 meter resolution). In order to isolate the specific elevation range of each plant's habitat, we chose the nodata alternative in the reclass command. We then created an image from the grid using the GRIDIMAGE command in ARC. The gridimage command transformed the elevation grid to an image and saved it in the specified Imagine format. This allows us to use the image in conjunction with other images in order to further isolate the habitat for the specified plant. The MATRIX command in IMAGINE was used to intersect the coverages providing a more specific habitat representation. The only trick to using MATRIX was to make sure the image projections were the same. An example of a resulting image, Maguire primrose, having an elevation distribution of 1460 to 1830 meters is as follows:
Maguire primrose
(Primula maguirei)
Aspect was only pertinent to one ETS plant in Cache County, Cronquist daisy. Using the DMA as the source grid coverage, we generated an aspect grid utilizing the ASPECT command in ARC/GRID. The process for delineating the specific aspect range for Cronquist daisy was similar to the elevation distribution format previously described using the RECLASS command. We were able to isolate the specified aspect coverage for Cronquist daisy using the resulting aspect grid as the source.
Cronquist daisy was determined from our sources to characteristically be found on northern slopes. In a text editor we coded 315 to 45 degrees, encompassing northwest to northeast exposures. Using this look-up table data and specifying nodata in the RECLASS command statement, we RECLASSed the aspect image to represent only northern slopes:
Northern Slopes(315 to 45 degrees)
Again, we created an image in IMAGINE format from this grid using GRIDIMAGE in ARC and intersected this image with the vegetation and elevation habitat distributions of Cronquist daisy to further isolate the areas in Cache County that encompass the specific requirements. We used the MATRIX command in Imagine again to intersect this image with the elevation + vegetation image of Cronquist daisy.
Brownie ladyslipper was found to occur adjacent to streams. We used the vector coverage of the river network in Cache County(1:100,000 scale) to incorporate this characteristic towards Brownie's habitat requirements. When looking at the AAT(Arc Attribute Table) of the Cache County river vector coverage, we were able to distinguish the stream system from the canals, ditches, et al. by isolating the MINOR1 class. The generated image of the stream coverage in Cache County is as follows.
Cache County Streams
We decided that a 90 meter distance on either side of the streams was an adequate width to measure adjancency. Within ARC, we used the BUFFER command to create buffered zones 180 meters wide following the paths of the Cache County streams. The next task involved transposing the buffer image into a grid coverage. Using the POLYGRID command in ARC, we were able to transform the buffered polygon coverage to a grid format having a 30 meter cell size. Again, using GRIDIMAGE in ARC, we created an image in IMAGINE format. This resulting image included a 90 meter buffered zone on either side of the Cache County vector coverage.
We were then able to MATRIX this image to the vegetation + elevation image of Brownie Ladyslipper that we had previously generated.
Due to the lack of detailed and consistent soils information for the three ETS plants, we investigated the possibility of using subsurface geology to further delineate the habitat for each plant. Information from a publication prepared by IHI Environmental concluded Maguire Primrose thrived in Laketown Dolomite and Cronquist Daisy in Laketown Dolomite and Garden City Limestone. We digitized these geologic units in Arc from the Geologic Map of the Logan Triangle, Utah map of 1946 by the USGS. Once the geologic coverages were created for both Maguire Primrose and Cronquist Daisy, digitizing corrections were made in Arc/Edit. The CLEAN command was then used to build topology and create labels for the polygons. Attribute tables for the coverages were created in Arc/Info. The attribute tables or lookup tables contain specific information for each coverage including: area, polygon id numbers, and polygon annotation. After the coverages were complete, it was necessary to convert them to grids using the same methods previously discussed. In order for these grids to be used in Imagine, they again, had to be made into images using the GRIDIMAGE command in ARC/GRID and saved in an Imagine format. The geologic coverages for Maguire Primrose and Cronquist Daisy could now be used as raster layers in Imagine. These new geologic layers were integrated using the MATRIX command in IMAGINE with the vegetation, elevation, and aspect layers previously generated which further delineated the predicted ETS plant habitats.
Example of the Digitized polygons:
Cronquist daisy
Erigeron cronquistii
The following images show our final predictions for the distributions of Cronquist daisy, Maguire primrose, and Brownie ladyslipper overlayed on a relief map of Cache County. By clicking on each image the succession of coverages we used to arrive at each final product can be seen.
Cronquist daisy(Erigeron cronquistii)
Maguire primrose(Primula maguirei)
Brownie ladyslipper(Cypripedium fasciculatum)
For the following plants we had only vegetation and elevation coverages to focus the habitat coverage.
Habitat:
Elevation: 2120 to 2880 m
Vegetation Type: mountain brush; spruce-fir
Geology: dolomitic/limestone substrates
Cache beardtongue(Penstemon compactus)
Habitat:
Habitat:
Unfortunately complete coverage data were not available for all six of our plants. For Maguire draba, Cache beardtongue, and Logan buckwheat we were only able to obtain vegetation and elevation information. We decided to intersect these coverages and were able to narrow down the areas where each plant should occur in Cache County. If additional specific information related to habitat characteristics for these three plants became available it could be easily incorporated into our predicted plant distribution based on this intersected vegetation/elevation coverage. For this reason we chose to include these 3 species in our project even though our coverage of their habitat is not very precise.
Soil requirements for all six plants were not specific enough to be used as key factors in further isolating their distributions. The only soil information available for these ETS plants was that they all occurred in "limestone based soils". This was not a surprise because most of the areas where these plants occur, according to our vegetation/elevation coverages, are characterized by these soils. To better determine the locations of these plants we had to turn to other coverage data.
Fortunately we found information in Bear River Endemics relating to specific geologic formations that are found in association with the Cronquist daisy and Maguire primrose. The geologic information from the cache geologic coverage (/auto/utah/phys.env/) could not be correlated to the limestone formations that were said to be in association with the Cronquist daisy and Maguire primrose. We had to use a map that displayed the "Ogc" (Garden City Limestone) and "So"(Laketown Dolomite) formations. This required that we use an older map called the Geologic Map of the Logan Triangle, Utah. Although this map was created in 1946 it appeared to be more specific than the on-line geologic coverage.
The stream coverage and aspect coverage could only be used for the habitat classification of Brownie ladyslipper and the Cronquist daisy, respectively. Brownie ladyslipper is found in association with riparian areas, so we were able to eliminate areas from our vegetation/elevation coverage that were not within 90 meters of a major stream. The aspect information predicted that the Cronquist daisy would only occur on north facing slopes within the area dictated by our vegetation/elevation coverage. These two coverages significantly narrowed down areas in Cache County where these two plants are likely to occur. Out of our predicted plant distributions these two are likely to be the most accurate.
An interesting application of our project occured to us while doing research on the habitat of these ETS plants. Their has been work done with transplanting and artificial cultivation of at least one of our species. Dr. Paul Wolfe of Utah State University has managed to successfully germinate seeds of Maguire primrose. A model similar to the one we have created could be useful in selecting preliminary sites for reintroduction programs designed to ensure the survival of these plants.
Endangered, Threatened, and Sensitive Species (ETS) plants require special management procedures and protection. GIS offers an excellent oppurtunity to assist in the managing of ETS plants. Projects similar to ours allow the identification and characterization of areas where these species are likely to exist. It is important to be aware of these areas so the habitat required by each species can be preserved and protected from human disturbance. Protecting ETS plants is of crucial importance in the struggle to maintain vegetative diversity.
ReferencesCome on over to the USU Department of Geography Homepage
A special thanks to ThadTilton and Bonnie Banner for their guidance, without whom this project could not have happend.
Buffered Streams
