home

Alabama

Land cover:

We have statewide leaf-off coverage of ETM+ imagery and are awaiting the acquisition of leaf-on scenes from the current MRLC contract.  We have completed unsupervised classifications of the leaf-off imagery.  During fall 2001, we acquired digital aerial videography of approximately 3,500 miles of flight line, completing this phase of our project.  Digital video will be interpreted and used to select training regions for further classification.  The year 2002 will be spent primarily on building our library of training regions.

Animal modeling:

We have developed a peer-reviewed list of vertebrates to model and will be initiating modeling and database-related activities during the coming year.

Land stewardship mapping:

Land stewardship mapping is approximately 85% complete.
We expect to complete this theme during the coming year and to build an associated database.

Alaska

Arkansas

California

Connecticut

Land cover:

Although the land cover map was completed in 1997, preliminary field assessment by cooperators indicates that there are significant classification errors, especially in the Cape Cod region.  Although new GPS-logged aerial videography was obtained for Massachusetts as part of an NSF-funded project with the Department of Computer Science at the University of Massachusetts, there are no plans to redo the land cover map at this time.

Animal modeling:

Predicted habitat maps are complete for birds, reptiles, and amphibians.  With completion of the expert review of mammal range maps, the predicted habitats maps for mammals will be completed.  Predicted habitat maps for all 273 vertebrates modeled in the Southern New England region will be finalized within the next six months.

Land stewardship mapping:

All conservation lands in the region are mapped and classified according to conservation status.  About 7% of the land area was classed in categories 1 & 2.  However, this database will be updated prior to the final analyses planned within the next 12 months.

Analysis: 

Once the mammal habitat maps are completed and the land stewardship maps are updated, the species richness analyses will be redone and updated.  The final gap analysis will be completed by December 2002.

Reporting and data distribution:

Once the databases and analyses are updated, all data layers will be made available on a new home page in cooperation with UMass Extension.  Until a new server is installed and the new home page developed, there will be only limited access to the data.  We also plan to distribute the final report via CD-ROM.
Availability of the data and final report is planned for December 2002.

Other accomplishments and innovations:

A draft of a manual detailing the use of GPS-logged aerial videography for land cover mapping has been completed and is under revision.

Delaware

Land cover:

Land cover mapping for the three-state area was completed during 2001.  An accuracy assessment was conducted with nearly 2,500 points randomly selected from the project video.  The final fuzzy accuracy assessment calculations will be completed early in 2002.

Animal modeling:

Models have been developed for all vertebrates, and distribution maps are currently being generated to be distributed for expert review beginning in January 2002. 

Land stewardship mapping:

The land stewardship mapping for the project was completed in 2001.

Analysis:

The tables for the land cover analyses have been created.  Early in 2002, the tables for the vertebrate layer will be completed.  The gap analysis for the project will be completed by spring 2002.

Reporting and data distribution:

The final report development is well under way with the methods sections completed for each chapter.  Results for land cover and stewardship were completed in 2001.  The complete report will be accomplished by July 2002.

Florida

Georgia

Land cover:

A statewide general land cover map has been completed and accuracy assessed. Accuracy assessment of the general land cover map was accomplished through digital video; overall accuracy was 85%.
From the general land cover, we have begun mapping vegetation alliances. A set of decision rules has been developed to make the transition from general land cover categories to alliances. We anticipate completion of alliance mapping by April 2002.

  Figure 1.  Natural sandhills habitats on the Georgia Coastal Plain.

Animal modeling:

Habitat association databases have been completed for all taxa.  All of the associations are currently out for review.  An interactive form developed in Microsoft Access allows reviewers to make changes and add comments in the database.  We have completed point occurrence databases for both herps and mammals.  Georgia's Department of Natural Resources Nongame Wildlife Division provided valuable assistance in compiling herp records, while the Game Management Division  helped with a number of mammal distributions.  The Nongame Division's Breeding Bird Atlas was our primary source for bird occurrence data.  General range maps for all taxa are currently out for review.  We are ready to begin writing final models for species distributions, pending modifications from reviewers.

Land stewardship mapping:

The stewardship layer was completed in 1999.  Subsequent land purchases by the State of Georgia have been incorporated into an updated version.  We will include other acquisitions that occur before completion of the project.

Other accomplishments and innovations:

The initial general land cover map will provide the base layer for the Georgia Land Use Trends (GLUT) project, which analyzes land cover change between 1974-1998.  We have completed a land cover map for Georgia from 1974 and begun analyses of changes.  Additional land cover maps from 5-year increments should be completed over the next year.

We have developed innovative methods for predicting distributions of stream salamanders and other aquatic herps using hydrologic models.  We expect to expand and improve upon them before project completion.

Aquatic GAP:

Georgia has initiated an Aquatic Gap Analysis Project in partnership with Alabama.  We will be working in the Alabama-Coosa-Tallapoosa and Apalachicola-Chattahoochee-Flint river basins.

Hawaii

Land cover:

The Hawaiian entry to the NVC has been expanded to include16 plant associations identified as invasive-dominated.  The description of these nonnative species assemblages will enable HI-GAP to accurately define variation within nonnative-dominated land cover along with the mapping of native-dominated plant associations.  This work was done in conjunction with the development of a crosswalk classification system, which ties the NVC to the more detailed avian fauna habitat mapping effort being produced through the Hawaii Forest Bird Interagency Database Project (HFBIDP) at the Pacific Islands Ecosystem Research Center.

A statewide cloud-free LandSat TM imagery set is nearly in place.  Images are still being sought for notorious cloud-cover areas, and multiple-date TM imagery is being sought for areas where only a single cloud-free coverage is in place.  An IKONOS consortium is being formed by the Hawaii Natural Heritage Program (HINHP), which will enable land managers in Hawaii to join together and share image purchases.  HI-GAP is administrated through HINHP and will soon have access to a large library of recently collected high-resolution spectral imagery.
While statewide coverage may be a long-term goal, this imagery should assist in the development of the statewide LandSat TM land cover.

HFBIDP is currently mapping upland forested areas in Hawaii in areas where native Hawaiian forest birds are known to occur.  Imagery and classification techniques are being transferred between the HFBIDP effort and the HI-GAP land cover mapping team.

Animal modeling:

  More than 3,000 database records from the Hawaii Stream Assessment conducted by the State of Hawaii's Division of Aquatic Resources have been geocoded to stream segments.
A distributable GIS data set has been produced and reviewed.  Refinements are ongoing.  The data set is the most comprehensive distribution of native aquatic biota currently available for Hawaii.

In addition to the traditional indigenous vertebrate mapping, HI-GAP is forming a group to undertake mapping of invasive vertebrates.
In Hawaii, invasive pig, goats, deer, sheep, cats, mongoose, and rats all pose significant threats to native biodiversity.  Statewide and islandwide distribution data for these critical threats is not currently available.  HI-GAP hopes to integrate existing data, fill gaps, produce spatial data sets, and disseminate data on the distribution of invasive vertebrates.  This important but controversial effort is going through a thorough internal scoping phase, as a strong response is expected both from land management agencies and local communities.

Land stewardship mapping:

  Stewardship mapping is underway.  Initial draft maps and data sets have been produced and distributed (Figure 1).  Review is ongoing and an updated version is expected in the spring.

Figure 1.  Stewardship layer for the island of Oahu.  Parcel data from the City and County of Honolulu were used to derive this map layer.
<%Case "ID"%>

Idaho

Land cover:

The Idaho land cover layer and final report chapter are complete.  The land cover classification recognizes 81 cover types and is mapped at a resolution of 0.09 ha with a 2 ha MMU.  The land cover data, metadata, and final report chapter can be downloaded at http://www.wildlife.uidaho.edu/idgap.htm or by contacting the Idaho Cooperative Fish and Wildlife Research Unit.

Animal modeling:

Second-generation wildlife habitat relationship models have been completed for 379 terrestrial vertebrates in Idaho.  The models are stored on Idaho's Web site as georeferenced TIFF images with a native resolution of 0.09 ha.  Accuracy assessments of the models were completed in 2001 and the results included in the metadata.  The models and metadata can be downloaded at the URL above or by contacting the Idaho Cooperative Fish and Wildlife Research Unit.

Land stewardship mapping:

The revised Idaho land stewardship database is also complete.  This data set represents a significant improvement over the original Idaho land stewardship layer by increasing spatial resolution to a 2 ha MMU and incorporating many of the smaller managed areas in Idaho.  This data set, its metadata, and final report chapter are available for download at the URL above or by contacting the Idaho Cooperative Fish and Wildlife Research Unit.

Analysis:

Analysis of the protection status of Idaho's land cover types and wildlife habitat distributions is complete.
We also completed a gap analysis of geomorphologic and climatic features in Idaho.

Reporting and data distribution:

The final report for ID-GAP is in draft format and should be completed by spring 2002.  (See report summary on page __.)  Final updates are being made to the metadata.  

Other accomplishments and innovations:

We have developed programs to create hypergrids (grids containing the distributions of all input grids in condensed, binary form) in order to facilitate the identification of areas that satisfy multiple selection criteria including species richness, species protection status, size of area, and contiguity of area.  

2001-2002 GAP-related publications include:

Wright, R.G., J.M. Scott, S. Mann, and M. Murray.  2001.  Identifying unprotected and potentially at-risk plant communities in the western USA.  Biological Conservation 98:97-106.

Scott, J.M., F.W. Davis, R.G. McGhie, R.G. Wright, C. Groves, and J. Estes.  2001.  Nature reserves: Do they capture the full range of America's biological diversity?  Ecological Applications 11:999-1007.

Karl, J.W., L.K. Svancara, P.J. Heglund, N.M. Wright, and J.M. Scott.  2002.  Species commonness and the accuracy of habitat-relationship models. Pages 573-580 in J.M. Scott, P.J. Heglund et al., editors.  Predicting species occurrences: Issues of accuracy and scale.  Island Press, Washington, DC.

Illinois

Land cover:

As part of the Interagency Land Cover Classification Project between the Illinois State Geological Survey, Illinois Department of Natural Resources, and the Illinois Department of Agriculture's National Agriculture Statistics Services, an updated statewide Level I land cover classification was completed in September 2001, using 1999 and 2000 TM imagery.  To speed up the classification process, we use these Level I classifications for the remaining scenes in the state and will classify them to Level II (community level) for GAP.   Current completed classified scenes can be found in Figure 1.  Editing and accuracy assessment are being started on TM scene 2233.

Animal modeling:

Draft vertebrate models have been completed and reviewed by scientists for all mammals, reptiles, and amphibians (Figure 2.)  The bird databases are 90% complete.  We will begin creating hexagon maps for all bird species in January 2002.  We are continuing work on our habitat associations database for mammals, amphibians, birds, and reptiles.

Land stewardship mapping:

We have developed a land stewardship map for Illinois, attributed general ownership categories, and assigned management status levels.  The GAP coding scheme for land units has been assigned to each property.
The database includes federal, state, and county properties.  In Illinois counties control the stewardship of about 500 properties.  This coverage is complete and is currently being reviewed and updated as needed.

Analysis:

We have completed some preliminary analysis using amphibian, bird, mammal, and reptile locational data to create species richness maps using the EMAP hexagons.  Preliminary species richness maps for birds were created using only locational data from Illinois Breeding Bird Atlas, Illinois Spring Bird count, the U.S. Department of Interior Bird Banding Lab, and the Illinois Natural Heritage Database.  Final species richness maps for birds will be created again using all available databases (including those listed above and Illinois Natural History Survey collections, Breeding Bird Survey, Audubon's Christmas Bird Count, Project Feeder Watch, and the Great Backyard Bird Count). yes; color: blue">  We have also conducted preliminary analysis of predicted distributions for species that occur in southern and northeastern Illinois.
We will do more analysis as our species and vegetation mapping progresses.

Reporting and data distribution:

  We have started writing some portions of the final report and will continue as our project progresses.

Other accomplishments and innovations:

• The IL-GAP Web page can be reached at www.inhs.uiuc.edu/cwe/gap/gapintro.html.
• Listed below are projects that are starting up, ongoing, or have been completed using the land cover database of Illinois as well as other data developed as part of IL-GAP.
• Applying spatial information technology to ecological risk assessment in Illinois. T. Weicherding, J. Levengood, S. Lavin.
• Distribution of Franklin's ground squirrel in Illinois.  J. Martin, E. Heske, and J. Hofmann.
• Population ecology and habitat use of eastern wild turkeys in Illinois.  P. Hubert, T. Van Deelen, P. Brown, and J. Garver.
• Fox and coyote ecology in central Illinois.
  T. Gosselink, T. Van Deelen, R. Warner, and P. Mankin.
• Deer in the urban ecosystem.  D. Etter, T. Van Deelen, and R. Warner.
• Identification and classification of critical wildlife habitat.  P. Brown, J. Aycrigg, and L. Suloway.
• Map Illinois.  J. Aycrigg, D. Stigberg, J. Westervelt, and M. Joselyn.
• Effects of traditional habitat and ecosystem management on the population ecology of northern bobwhite. J. Seigrist and J. Brawn.
• Urban white-tailed deer as biomonitors for zoonoses in Chicago, Illinois.  K. Hollis and R. Warner.

Indiana

Land cover:

Animal modeling:

Land stewardship mapping:

Analysis:

Reporting and data distribution:

Completed data are distributed through our regional server at the USGS Upper Midwest Environmental Sciences Center in LaCrosse, Wisconsin (contact Dan Fitzpatrick at (608) 781-6298 or Daniel_Fitzpatrick@usgs.gov).

Iowa

The Iowa Gap Analysis Project (IA-GAP) has finished its fourth and final year in 2001.  The IA-GAP home page is accessible at http://www.iowagap.iastate.edu/.

Land cover:

Land cover mapping is 100% complete.  Final maps will be published in the final report and on the IA-GAP Web site.  Data can also be viewed on the Iowa Geographic Image Map server at http://ortho.gis.iastate.edu/gaplandcover/gap_lc.html.

Animal modeling:

Models have been completed for about 98% of all species.  Iowa has been a cooperator in the Upper Midwest vertebrate modeling initiative along with North and South Dakota.  Final distribution maps will be published in the final report and on the IA-GAP Web site.

Land stewardship mapping:

Stewardship mapping and attribution is 100% complete.  Final maps will be published in the final report and on the IA-GAP Web site.  The IA-GAP stewardship image map server can be accessed at http://komodo.gis.iastate.edu/new_site/.

Analysis

Analysis is 100% complete.  Final maps will be published in the final report and on the IA-GAP Web site.

Reporting and data distribution:

Final maps will be published in the final report and on the IA-GAP Web site.

Aquatic GAP:

The Iowa Rivers Information System (IRIS) project started with Iowa DNR as partner.  Funding for a 3/4-time position was established through 6/02.  More information can be found at http://maps.gis.iastate.edu/iris/viewer.htm.  A proposal was accepted for Iowa Aquatic GAP in August 2001.  Staff was sent to an Aquatic GAP training session in Columbia, Missouri, in November 2001.  PIs are involved in regional coordination with KS, MO, and NE.

Goals for 2002 are to

(1) build the spatial database for Aquatic GAP and IRIS with the DNR's GIS Section using the National Hydrologic Dataset (NHD); (2) collect and add specific data attributes to the IRIS spatial structure; (3) regionalize data sets with KS, MO, and NE; and (4)integrate IRIS into a GIS environment where queries can be done for location or attribute information from a user's desktop.  This will involve writing programming code in an ArcView environment using Avenue, Visual Basic, or C++.

Other accomplishments and innovations:

Land cover accuracy assessment - Final report submitted to EPA Region VII.  The final report can be viewed on the IA-GAP home page (http://www.iowagap.iastate.edu/).

NatureMapping

In 1999, Iowa State University Extension Wildlife Programs began offering the Iowa NatureMapping Program to a wide-ranging audience.  NatureMapping, a citizen-based wildlife monitoring program, is an education and outreach component of IA-GAP.  Reliable, accurate, and up-to-date information about Iowa's wildlife collected by Iowans will give those considering decisions such as wildlife management and research, urban development, or conservation and preservation a valuable layer of data not otherwise available in traditional land use planning.
NatureMapping is a way to collect large data sets while reconnecting people to their local resources.  Final maps will be published in the final report and on the IA-GAP Web site.

Iowa Geographic Information Image Server

The server is serving orthophotos, topographic maps, and other Iowa grid data from http://ortho.gis.iastate.edu.  This service is heavily used by IA-GAP and our cooperators as well as many other Iowa users for a variety of applications.  The goal for 2002 is to update storage to provide data at higher resolution and serve more data, including integrating vector data and new Iowa color infrared photography coordinated by the Iowa DNR.

Kansas

Land cover:

The land cover layer and accuracy assessment are finished.  The land cover layer can be viewed or downloaded at http://mapster.kgs.ukans.edu/dasc/catalog/coredata.html.

Animal modeling:

The wildlife habitat relationship models and draft models are completed for all 364 vertebrates.  Experts have reviewed 80% of the vertebrate models, and 78% have been finalized.  We anticipate having all models reviewed and finalized by February 2002.  None of the vertebrate maps have been validated. 

Land stewardship mapping:

The stewardship boundary layer is 100% complete relative to land unit boundaries.  Data from surveys have been entered into attribute tables used to assign status codes to all land units.  Status codes have been assigned to units in the stewardship layer, and the majority of these codes have been reviewed by the appropriate state and federal agencies or nongovernmental organizations.  Review of remaining status codes should be completed by mid-February 2002.

Analysis:

We are currently working on methods for the gap analysis, which should be completed by the end of March 2002.

Reporting and data distribution:

We expect to complete reporting and data distribution within the next four months.

Aquatic GAP:

A meeting with local stakeholders was held in November 2001, at which representatives from various state, federal, and private agencies expressed a strong interest in an Aquatic GAP project for Kansas.  In addition, we received input from agency personnel on the potential uses of Aquatic GAP.
We have completed the initial processing of the National Hydrology Data set, which will be used as a base layer (i.e., by valley segments) to link biological and physical attribute data.  We are also in the process of constructing a relational database with fish and mussel data provided by the Kansas Department of Wildlife and Parks and Kansas Department of Health and Environment.

Other accomplishments and innovations:

We are currently conducting a pilot study to examine the feasibility of developing the KS-GAP vertebrate database as a discovery tool for elementary students in Kansas.  We have also received continuing funds from the Kansas Department of Wildlife and Parks to develop the KS-GAP vertebrate database into a decision support system for resource managers in Kansas.   <%Case "KY"%>

Kentucky

Land cover:

The final version of the statewide vegetation map is complete.  A first draft of the vegetation chapter has been written and is under review.

Animal modeling:

Modeling of the predicted distributions of terrestrial vertebrate species has been completed.  Land cover, digital elevation models, and National Wetland Inventory data were the basic input layers for our models.  These layers were manipulated to refine our predicted species' distributions.  Most of these refinements were a consequence of creating an edge buffer around map units or limiting a predicted distribution due to elevation constraints.  All layers were created in or converted to ARC/INFO Grid (raster) format, with a cell size of 30 meters.  All of our modeling was performed in the ARC/INFO Grid environment.  We used one model (i.e., AML) consisting of many different subprograms for all species in order to maintain quality control over the modeling process.

Land stewardship mapping:

The land stewardship layer (including metadata) is complete.
A first draft of the stewardship chapter has been written and is under review. 

Analysis:

Land stewardship and vegetation analyses are complete.  We are currently completing the gap analysis for the vertebrate species using the final stewardship layer.  We anticipate that the analysis will be complete in spring 2002.
The conclusions and management implications chapter of the final report is being written.

Reporting and data distribution:

The goal for 2002 is to have the entire report completed, reviewed, and accepted by National GAP by September 2002.

Other accomplishments and innovations:

We have worked with the Kentucky Department of Education and University of Kentucky Extension Service to produce middle school and high school teaching units using KY-GAP data.  Received a $6,000 grant from BRD to conduct teacher workshops in June 2002 to develop teaching modules.

Louisiana

Maine

Maryland

Land cover:

Land cover mapping for the three-state area was completed during 2001.  An accuracy assessment was conducted with nearly 2,500 points randomly selected from the project video.  The final fuzzy accuracy assessment calculations will be completed early in 2002.

Animal modeling:

Models have been developed for all vertebrates, and distribution maps are currently being generated to be distributed for expert review beginning in January 2002. 

Land stewardship mapping:

The land stewardship mapping for the project was completed in 2001.

Analysis:

The tables for the land cover analyses have been created.  Early in 2002, the tables for the vertebrate layer will be completed.  The gap analysis for the project will be completed by spring 2002.

Reporting and data distribution:

The final report development is well under way with the methods sections completed for each chapter.  Results for land cover and stewardship were completed in 2001.  The complete report will be accomplished by July 2002.

Massachusetts

Land cover:

Although the land cover map was completed in 1997, preliminary field assessment by cooperators indicates that there are significant classification errors, especially in the Cape Cod region.  Although new GPS-logged aerial videography was obtained for Massachusetts as part of an NSF-funded project with the Department of Computer Science at the University of Massachusetts, there are no plans to redo the land cover map at this time.

Animal modeling:

Predicted habitat maps are complete for birds, reptiles, and amphibians.  With completion of the expert review of mammal range maps, the predicted habitats maps for mammals will be completed.  Predicted habitat maps for all 273 vertebrates modeled in the Southern New England region will be finalized within the next six months.

Land stewardship mapping:

All conservation lands in the region are mapped and classified according to conservation status.  About 7% of the land area was classed in categories 1 & 2.  However, this database will be updated prior to the final analyses planned within the next 12 months.

Analysis:

Once the mammal habitat maps are completed and the land stewardship maps are updated, the species richness analyses will be redone and updated.  The final gap analysis will be completed by December 2002.

Reporting and data distribution:

Once the databases and analyses are updated, all data layers will be made available on a new home page in cooperation with UMass Extension.  Until a new server is installed and the new home page developed, there will be only limited access to the data.  We also plan to distribute the final report via CD-ROM.
Availability of the data and final report is planned for December 2002.

Other accomplishments and innovations:

A draft of a manual detailing the use of GPS-logged aerial videography for land cover mapping has been completed and is under revision.

Michigan

Land cover:

Land cover mapping follows the Upper Midwest GAP protocol at ftp://ftp.umesc.usgs.gov/pub/misc/umgap/98-g001.pdf.  The Upper Peninsula and Northern Lower Peninsula are complete, and draft versions are available from the USGS Upper Midwest Environmental Sciences Center (UMESC).  Mapping of the existing natural and seminatural land cover of southern Michigan continues.  In cooperation with the DNR's Integrated Forest Monitoring Assessment and Prescription project (utilizing new Landsat 7 imagery), we began field work to develop revised land cover information for the Upper Peninsula and the Northern Lower Peninsula.  A major effort to cross-walk the GAP classification to the NVCS is planned for 2002.

Animal modeling:

Wildlife Division research faculty at Michigan State University will continue working with the Michigan Natural Features Inventory and other Wildlife Division staff on a species distribution modeling project.  Most GAP program funding received from Upper Midwest GAP for fiscal year 2002 will be applied toward this work.

During fiscal year 2002, work will continue on the integration of existing species habitat databases into the GAP modeling process.
Further refinement of the databases will occur with additional literature review.  Initial distribution/occurrence models will be developed for most species.

Land stewardship mapping:

The stewardship map has been delivered to UMESC and is being reviewed.

Reporting and data distribution:

Draft land cover data and stewardship data are available from UMESC.  Contact Daniel Fitzpatrick at (608) 781-6298 or Daniel_Fitzpatrick@usgs.gov.

Minnesota

Land cover:

Land cover mapping follows the Upper Midwest GAP protocol (ftp://ftp.umesc.usgs.gov/pub/misc/umgap/98-g001.pdf).  The state DNR is finishing classification of the SW (mostly agricultural) quarter of the state.
The following classification units are completed, and a draft version is available from the USGS Upper Midwest Environmental Sciences Center (UMESC): North Shore, Border Lakes East, Border Lakes West, North Tamarack Lowlands, South Tamarack Lowlands, East Chippewa Plains, West Chippewa Plains, Nashwauk Uplands, Laurentian Uplands, St. Louis Moraines, Pine Moraines and Outwash Plains, Mille Lacs Uplands, Agassiz Lowlands East, Agassiz Lowlands West, Anoka Sand Plain, Blufflands, and Oak Savannah and Rochester Plateau.  A major effort to cross-walk the classification to the NVCS is planned for 2003.

Animal modeling:

The ongoing state DNR vertebrate mapping effort will expand in 2002.  Draft hexagon species range maps have been delivered to UMESC and are being reviewed. Species expert review teams are helping to develop habitat suitability.  The animal modeling coordinator for the Minnesota DNR is Jodie Provost, Department of Natural Resources, e-mail: Jodie.provost@dnr.state.mn.us).

Land stewardship mapping:

Stewardship mapping is completed, and a draft version is available from UMESC.  The Public Land Survey (PLS) was used as a base map.  The section corners are located, and the 40-acre tracts are generated from an algorithm.  Each 40-acre parcel is attributed for public landowner, manager, and stewardship status.  The coverage is clipped and served in 1:100k mapquad tiles.

Reporting and data distribution:

Draft land cover data and stewardship coverages are available from UMESC.  Additional land cover data are expected to become available in 2002.  Contact Daniel Fitzpatrick at (608) 781-6298.

Mississippi

Land cover:

The MS-GAP land cover map was completed in 1999 and has been used by numerous state, federal, and local resource agencies.  Over 100 individuals have requested these data for use in a wide variety of projects.  The data will soon be downloadable from the MS-GAP home page and from the Mississippi Automated Research Information System (MARIS) Web site.  MARIS is the GIS data clearinghouse for the state of Mississippi and can be found at http://www.maris.state.ms.us.

Animal modeling:

Modeling of animal ranges and distribution has generally been completed.
A few issues still persist on several herpetile species, but these issues will be resolved by late winter.  Distributions have been developed for 402 species including 58 mammals, 216 birds, and 128 reptiles and amphibians.  Reviews are complete on all models except the few herps.

Land stewardship mapping:

Land stewardship mapping has been completed thanks to the aid of the Mississippi Department of Wildlife, Fisheries, and Parks and the U.S. Forest Service.  As with most eastern states, Mississippi is comprised of mostly management status 4 lands.  Less than 1% of the state is in status 1, and 7% of the state is in status 2. 

  Analysis:

Analysis is mostly complete.  The analysis section of the report is the final step to completion of this project.  Once completed it will be sent to reviewers for comment.

  Reporting and data distribution:

The report is the current focus of MS-GAP.  We have completed the land cover and stewardship sections and have sent them out for external review.  The modeling is nearly complete.  Analysis is the final section to complete. 

Missouri

Near completion
Anticipated completion date: March 2002

Land cover:

Complete.  Phase I land cover was completed by the Missouri Resource Assessment Partnership.  Metadata has been compiled for the base land cover map as well as all derivative databases created from this base such as ecotones, interiors, etc.  Ancillary land cover databases (sink holes, wetlands, glades, etc.) compiled for this project were also documented.

Animal modeling:

Nearing completion.  Three hundred forty-eight vertebrates were modeled (66 mammals, 164 birds, 74 reptiles, and 44 amphibians).  Ninety-meter grids representing the species predicted distributions were created for final GAP reporting.  Metadata is being compiled for these predictive species maps.

Land stewardship mapping:

Complete.  The stewardship layer was created by the Missouri Resource Assessment Partnership.  Public lands comprise only 6.7% of Missouri with 4.7% under federal and 2% under state jurisdiction.  All areas greater than 16 hectares were analyzed for biodiversity components.  Metadata has been compiled for these stewardship maps.

Analysis:

Complete.  Upon review, another analysis was needed to meet national reporting guidelines.  This analysis has been completed and added to the final report and deliverables for Missouri.

Reporting and data distribution:

Nearing completion.  The third draft of the final report will be mailed in December 2001.  CDs are being assembled for distribution to the GAP Operations Office following the directory structures outlined for final reporting of data.
Data distribution from these analyses will be posted on the Missouri Spatial Data Information Service (MSDIS) at http://msdis.missouri.edu.  A link from this site will be created to the AMLs and programs written in support of this effort.

Montana

Nebraska

Land cover:

The land cover map has been completed.  We are finishing the accuracy assessment and preparing the metadata.
Plans for the next twelve months include publishing the land cover map as an outreach product.

Animal modeling:

We are in the midst of modeling vertebrate habitats using two approaches.  First, a recursive partitioning program (QUEST) permits the rapid production and revision of classification and regression tree models to relate georeferenced occurrence data with environmental (land cover, climate, soils, terrain) variables (see article by Henebry et al. on page  ). Second, for those species lacking sufficient occurrence data to permit effective application of the recursive partitioning algorithm, we have turned to the "literature gestalt" method.  Range maps are undergoing expert review.  Accuracy assessments are being conducted with voucher specimens located in other museums and occurrence data from the literature.  In 2002 we plan to complete habitat modeling and inversion to range maps, complete expert reviews and accuracy assessments, and prepare metadata.

Land stewardship mapping:

Land stewardship mapping has been completed.  We are now preparing the metadata.

Analysis:

We are in the process of analyzing aspects of the stewardship and land cover maps with the vertebrate occurrence data used in the modeling and some of the reptile and amphibian draft models.  Further analyses will be conducted as animal models are completed.

Reporting and data distribution:

Metadata assembly, data lineage, and methods documentation are ongoing.

Aquatic GAP:

We are collaborating with the Missouri Resource Assessment Partnership (MoRAP) to develop geospatial databases and implement Aquatic GAP in Nebraska.
In order to provide a regionally consistent Aquatic GAP product, this effort will also involve close collaboration with Aquatic GAP programs in Iowa, Kansas, and South Dakota.  Within Nebraska, Aquatic GAP will involve collaboration and cooperation with fisheries biologists, aquatic ecologists, and other scientists affiliated with the Nebraska Game and Parks Commission, the Nebraska Department of Natural Resources, the Nebraska Department of Environmental Quality, the University of Nebraska State Museum, The Nature Conservancy, the University of Nebraska School of Natural Resource Sciences, the USDA National Agroforestry Center, and many additional university faculty, state and federal agencies, and private organizations.

GAP research:

We are developing an approach to the regionalization of disparate land cover maps using image time series from coarse spatial but fine temporal resolution sensors, such as AVHRR and MODIS.  The objective is to provide an integrated regional land cover map of the Great Plains that provides improved discrimination among grassland types and elimination of discordance of land cover classes across state boundaries.

Other accomplishments and innovations:

• Presentation of our approach to habitat modeling for reptile and amphibian species at

  (a) the U.S. Landscape Ecology Symposium in Tempe, AZ, April 25-30, 2001,

  (b) the Nebraska GIS/LIS Association meeting in Lincoln, NE, May 8-11, 2001,

  (c) the National Gap Analysis Program meeting in Brookings, SD, June 17-20, 2001, and

  (d) the Ecological Society of America meeting in Madison, WI, August 5-10, 2001.

  (e) the Nebraska State EPSCoR conference, Lincoln, NE, April 4-5, 2002.

• Presentation of our approach to habitat modeling for avian species at

  a) the Mid-America GIS Consortium meeting, Kansas City, MO, April 15-19, 2002,

  b) the U.S. Landscape Ecology Symposium in Lincoln, NE, April 23-28, 2002.

  Collaborating with Oak Ridge National Laboratory on objective ecoregionalizations of the data layers used for modeling wildlife-habitat relationships.

• Providing land cover data to the National Park Service for the Revised General Management Plan for the Niobrara National Scenic River.
• Continuation of work with regional partners in Great Plains GAP consortium.
• Working with Kansas toward comparing approaches to modeling species habitat and geographic ranges.
• Cooperating with Rainwater Basin Joint Venture (US Bureau of Reclamation-$180,000; NE Game and Parks Commission) in land cover mapping using 1997-98 TM data.
• Cooperation with a $50,000 project from the Nebraska Game and Parks Commission to enhance and update land cover maps for the Niobrara River watershed.
• Cooperating with the Nebraska Research Initiative in Geospatial Decision Systems.
• Participation in the Great Plains Cooperative Ecosystem Study Unit (CESU).
• Participation in a $100,000 NSF Biocomplexity Incubation Activity project on the ecology and geology of the Sandhills of Nebraska.
• Participation in a $1.2M grant from the U.S. Environmental Protection Agency for a three-year project to develop methods for lake classification in Nebraska. 
• Participation in enhanced mapping of land cover for the Platte River watershed in Nebraska (including 1997-2000 change analysis) through Cooperative Hydrology Study (COHYST) consortium-$310,000.

New Hampshire

Land cover:  

Complete.  

Animal modeling:

Complete.

Land stewardship mapping:

Complete.

Analysis:

Nearly complete.

Reporting and data distribution:  

Digital coverages were submitted in mid-2001.  The final report will be distributed for peer review in early 2002.

New Jersey

Land cover:

Land cover mapping for the three-state area was completed during 2001.  An accuracy assessment was conducted with nearly 2,500 points randomly selected from the project video.  The final fuzzy accuracy assessment calculations will be completed early in 2002.

Animal modeling:

Models have been developed for all vertebrates, and distribution maps are currently being generated to be distributed for expert review beginning in January 2002. 

Land stewardship mapping:

The land stewardship mapping for the project was completed in 2001.

Analysis:

The tables for the land cover analyses have been created.  Early in 2002, the tables for the vertebrate layer will be completed.  The gap analysis for the project will be completed by spring 2002.

Reporting and data distribution:

The final report development is well under way with the methods sections completed for each chapter.  Results for land cover and stewardship were completed in 2001.  The complete report will be accomplished by July 2002.

New York

North Carolina

Land cover:  

The statewide land cover map will be complete in the winter of 2002.  In the spring an accuracy assessment based on ecological groups will assess the land cover using a stratified design with digital orthophoto quarter quads, aerial videography, and field visits being used to label the assessment sites.  A draft land cover has been mosaicked for the Roanoke-Tar-Neuse-Cape Fear Ecosystem for a collaborative project with the Fish and Wildlife Service.

Animal modeling:

Preliminary models were run for the species occurring in the Roanoke-Tar-Neuse-Cape Fear Ecosystem in the fall of 2001.  These models are being internally reviewed and revised and will be rerun when the statewide land cover becomes available in spring 2002.  Additional ancillary data required to improve the predicted distributions are being compiled in the winter of 2001/02.  The database of habitat relationships and ranges for each of the 414 species being modeled is complete.

Land stewardship mapping:

This layer is complete.  In developing this layer we have worked with a long list of cooperators including the NC Natural Heritage Program, the NC Center for Geographic Information and Analysis, NC Wildlife Resources Agency, NC Division of Coastal Management, Fish and Wildlife Service, and the NC Division of Parks and Recreation.  The final coverage includes 395 managed areas that have been assigned a protection status of 1, 2, or 3.  While we have been able to include many of the land trust lands with the help of our cooperators, we know land trust lands are underrepresented due to a lack of resources within the land trust offices to provide spatial representations of the lands they are protecting.  The methods and metadata have been created and are ready for internal review.   The final data layer and supporting documents will be available in spring 2002.

Analysis:

Draft analyses have been completed for the Roanoke-Tar-Neuse-Cape Fear Ecosystem, which includes large portions of the Coastal Plain, Piedmont, and Sandhills of North Carolina.  The analysis of the amount and distribution of statewide management lands has also been done.
In the spring of 2002, analysis of the protected status of the terrestrial vertebrates and land cover types will be completed.

Reporting and data distribution:

Data on the ranges of 414 terrestrial vertebrates and the land management status layer will be delivered via the NC-GAP Web site in the winter of 2002 (www.ncgap.ncsu.edu).  The final report, land cover, and predicted distributions will be delivered via the Web following external review in spring 2002. 

Other accomplishments and innovations:

We will complete the work on the GAP Ecosystem Data Tool in the spring of 2002.  This decision support tool was designed to help USFWS Refuges as well as Ecological Service offices address landscape conservation issues.  For a more thorough description of the tool see "Taking Refuge-GAP a step further: The GAP Ecosystem Data Tool in the Roanoke-Tar-Neuse-Cape Fear Ecosystem" 

North Dakota

Land cover:

The analysis of plant species composition and biomass data from 3,468 range surveys on state school lands in North Dakota (ND) was the primary activity in 2001.  Models predicting the relative abundance (% biomass) of common grass species in ND were created using Regression Tree Analysis.  Environmental predictor variables included mean annual precipitation and temperature; percent sand, silt, and clay in the soil surface layer; root zone depth; and available water capacity.  A plant community map was created from cluster analysis of the individual grass species' relative abundance maps.  The plant community map was intersected with a land cover map produced from supervised cluster analysis and maximum likelihood classification of multitemporal Thematic Mapper imagery.  The composite map provides information about the abundance of grass species and vegetation communities for the natural grassland class in the land cover map.
A hierarchical vegetation community classification was developed from the range survey data set using two-way indicator species analysis (TWINSPAN).  The TWINSPAN analysis included all plant species observed at the sample units, not just the common grass species used in the regression tree analysis.
Results from these analyses will be interpreted in conjunction with the National Vegetation Classification System to develop the map classes and legend for the natural grassland component of the ND-GAP vegetation and land cover map.  Primary activities in 2002 will be refinement of the regression tree and multitemporal Thematic Mapper imagery analyses, completion of a vegetation and land cover map for ND, and an accuracy assessment of the map.

Animal modeling:  

Range maps have been reviewed for the 289 breeding terrestrial vertebrate species included in ND-GAP.  The development of wildlife habitat relationship (WHR) models in a Microsoft Access database in conjunction with Iowa and South Dakota GAP projects was almost completed.  WHR models for all bird, amphibian, and reptile species have been reviewed. WHR models for all regional mammal species have been reviewed, and models for mammal species specific to North Dakota are in review.  Efforts were made to identify additional reviewers for WHR models.  A list of land cover types, ancillary layers, and ecotone layers necessary to run all WHR models was prepared.  An SAS program was created that converts WHR information in the Access database to Excel spreadsheets for inspection by model reviewers and develops a script for use in ARC/INFO to execute the WHR models and produce species distribution maps.  Efforts in 2002 will include the development of environmental data grids for modeling species distributions, the refinement of WHR models from reviewers' comments, and production of species distribution maps.

Land stewardship mapping:

  Cooperators continued to provide significant in-kind resources with regard to public land stewardship maps.  Draft vectors of U.S. Fish and Wildlife Service (FWS) fee-title lands and easement refuges were received.  A draft vector of ND Game and Fish Department (NDGF) fee-title lands was received.
The FWS and NDGF expect to complete their vectors in 2002.  A vector of U.S. Forest Service Lands was received.  The management intent of some tracts may change, subject to the outcome of the final review and comment period for the Dakota Prairie Grasslands Management Plan.  Efforts to assemble the individual public land ownership vectors into a single public land stewardship vector for ND were begun.  Procedures for addressing edge-matching and polygon sliver problems, arising when vectors from different sources and scales are combined, were investigated.
Procedures for categorizing biodiversity management status were modified for application to public lands in ND.
The primary difficulty was reconciling the requirement for natural land cover with legislation and management objectives.  Many public lands in ND have only fragments of natural land cover but have a biodiversity conservation objective.  Strict adherence to the guidelines in the Mapping and Categorizing Land Stewardship Handbook would result in these lands being assigned a status 4, which is the lowest level.
In 2002, we will complete the acquisition of vectors for individual public landowners and assemble and attribute a single public land stewardship vector for ND.

Other accomplishments and innovations:

Strong, L., S. Magill, and D. Buhl .  2001.  Integration of GIS and remote sensing for mapping land cover of the Northern Great Plains.  11^th Annual National Gap Analysis Program Meeting, 12-20 June 2001, Brookings, SD.

Ohio

Land cover:  

Progress was made toward the goal of completing 30% of the land cover map by June 2002.  Accelerated mapping activities began with the hiring of a vegetation classification specialist in May 2001 and increased participation in the project by the Ohio Department of Natural Resources (DNR) beginning in the summer of 2001.  Methods and software for processing land cover images were successfully tested in a pilot project completed in September 2001.  Digital aerial photography is being used in place of aerial videography as part of the accuracy assessment of the land cover map.  Aerial photography was obtained from aircraft belonging to the Ohio DNR and Ohio Department of Transportation (ODOT).  Landsat 7 scenes are being acquired from the OhioView program for dates as close as possible to those of the aerial photography and ground-truthing activities.

The effort to obtain digital aerial images and ground-truthed data will span late summer and fall 2001 and spring to fall 2002.  In the fall of 2001 a total of 1,189 images were taken.  Completing the aerial photography is estimated to require 41 days of flight time in 2002 to obtain 28,946 digital aerial images including overlap and side lap of TM scenes for Ohio and adjacent states.  Ground-truthing was initiated in summer 2001 and will continue through the fall of 2002.  Individual photos note the following characteristics: size, shape, pattern, and texture of tree species, tone of image, shadow cast, habitat and location, and association.

With grant support from the Ohio Lake Erie Protection Fund, 20 wetlands in the Ohio Lake Erie Basin were field-assessed during the summer of 2001 to provide ground-truth information on wetland plant alliances.
Fieldwork and ground-truthing have shown that the aerial videography available from 1999 was not sufficiently detailed to map wetlands.  The aerial photography being obtained for the land cover map will provide a greatly improved source of data.  During the winter of 2001-02, aerial photos will be taken of wet woodlands.  Aerial photos from leaf-off periods will be used in the interpretation of the location and extent of  wooded wetlands that are often not discernable from dry woodlands during the leaf-on period.

Animal modeling:

The OH-GAP Project consists of terrestrial and aquatic components.

Terrestrial vertebrate species

Hexagon range maps were released for review in spring of 2001, and the expert review of these maps was completed in the fall of 2001 for all amphibians, birds, and mammals.  The expert review of reptiles is still under way; further comments on range information are expected from two herpetologists.  The reptile range maps will be 75% complete in June 2002 and fully completed in 2003.
The literature review of habitat affinity information for each terrestrial vertebrate species is being developed and is connected to an Access database.  The Vertebrate Modeling Database developed by Kansas GAP was used as a guide.  This work has been ongoing since 2000.  The literature review of terrestrial vertebrates is complete for 50% of the species.
The habitat affinity database and literature review are planned for completion in 2003.

Aquatic species:

The Ohio Aquatic GAP Project is closely following the Missouri Aquatic GAP protocol and methods.  This method uses the valley-segment classification of stream reaches statewide and relates the occurrence of fish species to valley-segment type(s).  The Ohio Aquatic GAP Project was about 50% completed in 2001.  Distribution maps of fish species were completed in July 2001 and released on the OH-GAP Web page for expert review.  These data include maps of 160 native and introduced fish species that reproduce in Ohio streams.  Corrections based on expert reviews will be finalized in January 2002.  The fish species data were compiled from published and unpublished records of the Ohio Environmental Protection Agency, ODOT, Ohio DNR, and U.S. Geological Survey.  According to Ohio DNR, the 4,525 data points in the Ohio fish database represent the fish species occurrence in about 50% of the 43,000 named streams in Ohio.  Fish distribution maps will be finalized and published on the OH-GAP Web page and on CD in spring 2002.

In 2002, fish species modeling will be undertaken using two approaches: simple overlays of species occurrence on valley-segment types and a statistical modeling approach.  Either the GARP model (Stockwell and Peterson 1999) or the WhyWhere model (Stockwell 2001) will be selected.  A journal article discussing the research findings for fish species of the Ohio Aquatic GAP Project is planned for 2002 and would be submitted for publication in early 2003.

Land stewardship mapping:

The land stewardship map was 25% complete by the end of 2001 and includes Ohio's state and federal lands.  In 2002, county and regional parks, natural areas, and privately owned preserves will be added, where available, for Ohio's 88 counties.  Land stewardship categories and status codes are being inventoried and developed as the data are received.  Completion of the land stewardship map is planned for May 2003.

Analysis:

The accuracy of simple overlays relating fish species occurrence to valley segment types will be compared to the accuracy of statistical models to determine the most effective and efficient method to analyze the aquatic data.
The statistical models are the GARP model and the WhyWhere model.  These are expert systems that test the fit of several types of statistical and nonstatistical models for spatial data analysis.  The GARP model was successfully pilot tested on a small Ohio watershed for the Ohio Aquatic GAP Project in 2000.  The WhyWhere model is relatively new and has not yet been published but is reported by the author to be superior to GARP.

Reporting and data distribution:  

A factsheet on the OH-GAP Project was published and released in October 2001 and included a description of the terrestrial, aquatic, and wetland components.  A CD of the Ohio Aquatic GAP data will be published by June 2002.
The data will describe Ohio valley-segment types and the Ohio fish database.

Other accomplishments and innovations:

New techniques for accuracy assessment of the land cover map were used by the OH-GAP Project in cooperation with the Ohio DNR, Division of Wildlife (DOW).  The DOW purchased a digital camera (Nikon D1X), GPS unit, laptop computer, and software to obtain digital aerial images of Ohio vegetation.  In the early fall of 2001, a total of 1,189 images were taken in areas represented by two different Landsat 7 TM scenes.  The images are of very high quality with a resolution of 3,008 pixels at a photo scale of 1:40,000 and offer a ground resolution better than 1 meter.  Each aerial photo covers 0.479 km².

In 2002, the OH-GAP Project plans to work with federal, state, and local stakeholders in Cuyahoga and Summit Counties, Ohio, to develop a Decision Support System to integrate watershed and transportation planning in the watershed of the Cuyahoga River.  The stakeholders represent the environmental and transportation agencies in northeastern Ohio.  The OH-GAP Project will provide the biodiversity data for this project.

Literature cited:

Stockwell, D., and D. Peterson.  1999.  The GARP modeling system: Problems and solutions to automated spatial prediction.  Int. J. Geographical Information Science 13:143-158.

Stockwell, D.  2001. WhyWhere.  Accessed December 1, 2001, at URL http://biodi.sdsc.edu/ww_home.html.

Oklahoma

Land cover:

The land cover map of Oklahoma is complete.  The map contains 39 land cover types ranging from oak-pine forest in southeastern Oklahoma to eastern redcedar-oak woodland in central Oklahoma to grama-buffalograss prairie in the western Oklahoma panhandle.  The accuracy of the map was assessed.

Animal modeling:

Wildlife habitat relation models were run for 402 terrestrial vertebrate species, including 50 amphibians, 81 reptiles, 178 birds, and 93 mammals.
All maps of modeled species were reviewed by state experts and revised.  The accuracy of the maps was assessed.

Land stewardship mapping:

The land stewardship map is complete.  Over 94% of the land in Oklahoma is in private ownership.
Status 1 lands comprise only 0.2% and status 2 lands 1.7% of the area of Oklahoma.

Analysis:

Analyses indicate that because of the small percentage of status 1 and 2 lands in Oklahoma, few vegetation and animal species elements are actively managed for biodiversity conservation.

Reporting and data distribution:

The draft final report has been peer-reviewed.  The Oklahoma Cooperative Fish and Wildlife Research Unit and Oklahoma State University will be handling initial distribution of the OK-GAP final report and data.

Oregon

Pennsylvania

Rhode Island

Land cover:

Although the land cover map was completed in 1997, preliminary field assessment by cooperators indicates that there are significant classification errors, especially in the Cape Cod region.  Although new GPS-logged aerial videography was obtained for Massachusetts as part of an NSF-funded project with the Department of Computer Science at the University of Massachusetts, there are no plans to redo the land cover map at this time.

Animal modeling:

Predicted habitat maps are complete for birds, reptiles, and amphibians.  With completion of the expert review of mammal range maps, the predicted habitats maps for mammals will be completed.  Predicted habitat maps for all 273 vertebrates modeled in the Southern New England region will be finalized within the next six months.

Land stewardship mapping:

All conservation lands in the region are mapped and classified according to conservation status.  About 7% of the land area was classed in categories 1 & 2.  However, this database will be updated prior to the final analyses planned within the next 12 months.

Analysis:

Once the mammal habitat maps are completed and the land stewardship maps are updated, the species richness analyses will be redone and updated.  The final gap analysis will be completed by December 2002.

Reporting and data distribution:

Once the databases and analyses are updated, all data layers will be made available on a new home page in cooperation with UMass Extension.  Until a new server is installed and the new home page developed, there will be only limited access to the data.  We also plan to distribute the final report via CD-ROM.
Availability of the data and final report is planned for December 2002.

Other accomplishments and innovations:

A draft of a manual detailing the use of GPS-logged aerial videography for land cover mapping has been completed and is under revision.

South Carolina

Land cover:

Land cover mapping is complete.  The final detailed land cover was modeled from an initial generalized land cover and ancillary data.  The 27-class land cover used in the vertebrate modeling was derived from the final 54-class land cover.  The ancillary data sources included detailed county-level soil maps, National Wetlands Inventory coverages, and elevational data in the Blue Ridge Province.  Accuracy assessment was completed using a combination of aerial videography, ground assessment, and Digital Ortho Quarter Quads.  Accuracy of the land cover at classifications corresponding to Anderson Level I and II were 71.9% and 51.4%, respectively.

Animal modeling:

Animal modeling is complete for 455 vertebrate species.  Accuracy assessment was conducted using species lists from five sites throughout the state.  Overall species richness was highest in the deciduous forests comprising bottomland hardwoods in the coastal plain and deciduous forests of the mountains and upper piedmont.  Accuracy of the vertebrate predicted distributions was determined through comparison with sites possessing a list of species.  Accuracy ranged from 57% to 85.5%, depending on comparison data.

Land stewardship mapping:

Land stewardship mapping is completed.  Overall, various state, federal, and private landowners protect 6580 sq. km or 8.1% of South Carolina habitat in all GAP status categories. 

Analysis:

A gap analysis was conducted, indicating varying levels of protection for vertebrate species in South Carolina.  Protection for species of concern varies depending on knowledge of habitat requirements and the type of habitat in which the species occurs.  Species with well-known and specific habitat requirements tended to receive greater protection than many other species.

Reporting and data distribution:

The SC-GAP report is expected to be distributed within the year along with the standard data sets.  In addition, the South Carolina GIS Clearinghouse will begin making data available early in 2002.  The GIS Clearinghouse is maintained by SCDNR Natural Resources Information and Analysis Section and can be found at http://water.dnr.state.sc.us/gisdata/.  

Other accomplishments and innovations:

SC-GAP has been involved in a statewide inventory of ant species that will be completed by early 2003.  In conjunction with the ant survey, other invertebrates have been inventoried, and those data will be incorporated if appropriate

South Dakota

Land cover:

Completed in spring 2000.  No new work has been done.  Accuracy assessment of the Black Hills is planned.

Animal modeling:

Habitat models have been completed for all 362 species and are currently under expert review.  After reviews are completed, models will be finalized.

Land stewardship:

Completed in summer 2000.  No new work has been done.

Aquatic GAP:

Range maps and habitat models have been completed for 112 fish species.
Expert review is currently under way.

Analysis:

In addition to land cover and 11-digit hydrologic unit analyses completed in 2000, gap analysis was completed for simplified valley-segment codes (5 of 10 attributes-temperature, stream size, flow, gradient, and groundwater potential-were evaluated as concatenated valley-segment codes).  Most (154) valley-segment types are protected in less than 10% of their range, including 34 rare types with less than 10 reaches or less than 100 meters long.  Two codes (20101 and 20103) were represented in at least 10% of stewardship status 1 or 2 lands.  One code (24110) was represented in greater than 20% of status 1 and 2 lands, but less than 50% of its range.  This was a rare type only represented by five reaches.  Two codes (23110 and 24120) were protected in 100% of their ranges.  Both were rare types with less than 10 reaches represented.

Reporting and data distribution:

The final report is currently 50% complete; reporting of analysis of vertebrate and aquatic modeling is under way.  Metadata and final attributing per GAP standards is being completed for all coverages.

Southwest Regional GAP (SWReGAP)

Land cover

The RS/GIS Lab at Utah State University is acting as the regional land cover mapping lab for the five-state southwest region.  Coordination with the other four states is facilitated through a Web page that allows access to spatial data, procedural documents, and an Internet Map Server (http://www.gis.usu.edu/docs/projects/swgap).  During 2001 the RS/GIS Lab completed the following tasks for the five-state region: 

Coordinated with EROS Data Center (EDC) on imagery acquisition

 A tracking system was developed to monitor scene acquisition for approximately 80 ETM scenes for spring, summer, and fall dates (~240 scenes total).  At present, approximately 90% of all ETM scenes have been received from EDC and have been distributed among the five participating states.  The RS/GIS Lab at USU keeps copies of all ETM imagery, and a second copy is kept by the participating agency for each state.

Developed a methodology for image standardization

All imagery received from EDC has been terrain-corrected, but has not been radiometrically corrected.  USU has adopted a COST (Cosine Theta) atmospheric standardization procedure for image standardization.  To facilitate the use of this procedure for image standardization, USU developed a program that can be used to "read-in" information from the scene's header file and, with limited input from a technician, create an ERDAS Imagine model (.gmd) that can be used to perform the COST correction. 

Developed a DEM-derived landform model

One of the most promising avenues by which a higher level of classification accuracy and community definition will be achieved, is to improve the modeling of biophysical parameters that predict potential vegetation.  To this end, USU has developed a DEM-derived landform model that can be generated using an ArcInfo AML.  The model provides 10 basic landform classes and can be refined further by incorporating life zones using a vegetation index or elevation grid.

Developed image processing and classification procedures-

Land cover classification methods have traditionally been based on somewhat subjective methods of analysis.  USU has been developing a land cover classification methodology that incorporates Classification and Regression Tree (CART) tools.
As part of developing this protocol, USU hosted a regional workshop (Jan. 3-4, 2002) to train participating state land cover analysts using this approach

Land cover activities for 2001 have focused on (1) initiating and maintaining a cooperative network of agencies, organizations, and individuals; (2) final selection, acquisition, and selected preprocessing of satellite imagery; (3) collection of vegetation data at analytical training sites; (4) collection and organization of ancillary data; (5) review of the National Vegetation Classification System (NVCS); and (6) hiring of land cover staff.

Imagery selection, acquisition, and preprocessing

SWReGAP is using Landsat 7 Thematic Mapper imagery from 1999-2001.  Final selection of satellite imagery was completed during 2001, and the list was submitted to EDC through the regional land cover lab at USU.  This was facilitated through the Regional Coordinator and in coordination with adjacent states for concurrence on imagery.  Imagery began arriving to states in summer 2001, and delivery is expected to be complete in January 2002.  The New Mexico lab is assisting with regional land cover mapping by preprocessing of selected imagery for New Mexico and Arizona.

Collection of vegetation data

A field data collection workshop was held in May 24-25, 2001, for states participating in the SWReGAP project.  This workshop was sponsored by the regional land cover mapping lab and was hosted by Nevada project staff.  The purpose was to standardize field collection protocols regionwide.

Training sites continue to be obtained through both fieldwork and cooperative interaction in order to get an initial representation of all National Vegetation Classification System (NVCS) alliances in the project area.  In the southwest states, existing training sites have been obtained from a variety of sources such as the Colorado Vegetation Classification Project (CDOW Basinwide Inventory) led by the Colorado Division of Wildlife and the Bureau of Land Management (BLM); military installations in New Mexico (Fort Bliss Military Reservation and White Sands Missile Range); and New Mexico Natural Heritage Program.

Current field training site data continue to be collected by project personnel throughout the mapping zones assigned to states.
Some examples of field data collection efforts in New Mexico and Colorado include the following.  The New Mexico project has visited 1,904 sites among the 17 mapping zones assigned to the project.  Within these mapping zones, data collection represents 87 of the approximately 155 alliances.  New Mexico is using these data, available satellite imagery, and known alliances to target specific alliances within each mapping zone for additional sampling.  In Colorado, the Southern Piedmont (298 field sites), High Plains (215 field sites), Northern Piedmont (106 field sites), and North Mountain Parks (44 field sites) mapping zones have been the focus of field collection efforts.  In spring 2002 additional field crews will be deployed for regionwide work where more sampling is needed to improve the accuracy of vegetation type predictions.

Collection of ancillary data

Ancillary data have been obtained from a variety of sources across the five-state region.  Example sources include the BLM, Earth Data Analysis Center at the University of New Mexico, U.S. Geological Survey, and military installations such as White Sands Missile Range.  These ancillary data include GIS coverages (e.g., urban and agriculture coverages) and remotely sensed data (i.e., digital orthophoto quads, NALC triplicates, etc.). 

Classification system

The NVCS is being employed to develop classification consistency across the five-state area.  Keith Schulz with NatureServe works with the southwest states to ensure SWReGAP land cover types are regionally consistent and linked directly to NVCS alliances.  During the course of 2001 Keith has provided periodic alliance updates for the region and by state.  In addition he has participated in various land cover mapping workshops and coordinated with field crews such as Colorado to address their plant identification questions.

The southwest states, in turn, work with NatureServe to refine their alliance-level classification as needed and continue to review the NVCS as training site data are obtained.  Data to potentially describe new alliances are sent to Keith Schulz by individual states.  For example, the NVCS is being expanded for Arizona by adding alliances already described in other states but not attributed to Arizona and describing new alliances to Arizona and the NVCS.  The Arizona project is providing the field data and initial analysis necessary to get these provisional alliances and "new" alliances into the NVCS.

NatureServe has developed an Access database to manage the state land cover type information by mapping zone as southwest states work together to develop cover types that can be consistently applied across the southwestern region.  This database will also be used to manage information on new alliances that are developed over the course of this project.  

Animal habitat modeling:

The New Mexico Cooperative Fish and Wildlife Research Unit (NMCFWRU) in Las Cruces is acting as the regional animal habitat modeling lab for the five-state southwest region.  During 2001 the New Mexico lab has focused on identifying the list of animal taxa for habitat modeling and orchestrating review of this list among the five projects. 

The New Mexico lab submitted, for regional review, a packet on taxa inclusion and the preliminary assignment of species to each project in May 2001.  There were 866 species-level taxa included in the review package, as well as a written Decision Rule for Taxa Inclusion, Exclusion, and Modeling Allocation.  The objective of this packet was to provide specific, consistent criteria to select the candidate taxa across the five-state region and build consensus on allocation of species for modeling.  Responses from states indicated interest in shifting allocation of modeling responsibility for 57 taxa.  At end of year, reconciliation of taxa inclusion and project responsibilities was still ongoing. 

Efforts also were under way to complete consistent taxa naming and coding by cross-referencing Integrated Taxonomic Information System (ITIS) and NatureServe (former TNC) conventions.  Lee O'Brien, Colorado Project Coordinator, has provided taxonomic/animal modeling expertise in facilitating cross-walking between SWReGAP and reference material from ITIS.   

In preparation for animal habitat modeling, the New Mexico project has identified the 8-digit HUC (watershed codes) as the coverage to be used in range delineation.  Once the taxa list has been reviewed with regional consensus, the methodology and coverage will be sent to respective states for this range delineation.

Land stewardship mapping:

Land stewardship mapping is expected to begin in the spring of 2003.  States are completing various preliminary tasks in preparation.  Arizona is focusing on obtaining current information on tribal lands.  New Mexico has reviewed the existing Public Land Survey System layer for New Mexico.  Both Colorado and New Mexico propose to bring personnel on board during summer and fall 2002 to assist further with land stewardship mapping activities.  Currently the region is considering collaboration with the BLM on obtaining regionwide stewardship information.

Regional cooperation:

Regional cooperation continues to be critical to the proper functioning of SWReGAP.  Individual states contributed to the regional project during 2001 by participating in (1) a field data collection workshop in May, hosted by NV-GAP with planning by UT-GAP; (2) regional breakout sessions held at the National Gap Analysis Meeting to coordinate animal habitat modeling and land cover mapping activities; (3) preparation of a cooperative poster for the region; (4) assistance to the SWReGAP coordinator in presenting SWReGAP to federal agencies; and (5) regional Web site and listserv. 

Field data collection workshop

The objectives of the field data collection workshop were to arrive at a consensus on the field data collection protocols that would act as the standard among the five participating states.  This was accomplished through both in-field meetings and group discussion.  Issues that were discussed included minimum data requirements, field sampling methods, and sample data stratification. 

Regional poster and presentations

Through a coordinated effort with the Regional Coordinator and other projects, the New Mexico project produced an information poster to be displayed and used throughout the region.  The poster is entitled "Southwest Regional Gap Analysis Project: Enhancing Your Conservation Options Beyond Political Boundaries" and is available for use by all southwest states through the SWReGAP Web site.  The Colorado project assisted the SWReGAP coordinator in scheduling and making presentations to the Region 2 Forest Service office and the Region 8 Environmental Protection Agency office in Denver."