Director's Corner

First of all, thanks to everyone for your understanding and patience during this very uncertain budgetary process we are going through. We hope that fiscal year '97 and the transition of the National Biological Service to the U.S. Geological Survey will be much easier.

As more states complete their projects, we will have increasing opportunities to test predictions of vertebrate occurrences at various thematic and spatial scales. We will also be able to examine representation of species and vegetation types in special management areas across their full range of geographical and ecological occurrence. In addition, our maps provide a sampling framework for more detailed mapping of biological structure and function and thus permit us to gather unbiased estimates of these features. Such efforts should contribute greatly to transboundary planning.

The accuracy assessment of GAP products is an important area in which we can, and have, made important contributions. It is important that we report the results of our work on accuracy assessment in the refereed literature (see Edwards et al. In press and 1996). A small workshop involving researchers from several GAP projects as well as from the U.S. Forest Service reviewed current methods used for accuracy assessment and identified those most appropriate for use with regional mapping efforts like GAP. The workshop was convened by Western States coordinator Patrick Crist in Denver this spring. He will report on the results of this meeting at our Key Largo meeting. In addition, we will continue to budget for further work on accuracy assessment issues. A request for proposals to address issues of accuracy, scale habitat relationships, and other topics that test the assumptions and products of GAP will be sent out October 1, 1996.

During the last five years, GAP investigators have developed new techniques for mapping land cover. It's time to review all that has been done. Jim Merchant will be documenting land cover mapping methods used by GAP investigators, comparing the different methods to identify strengths and weaknesses, costs, etc. and will host a workshop in March 1997. We hope that his findings will be useful to new GAP investigators as well as to our second-generation land cover mapping efforts. As one example of the lessons we have learned, recent land cover mapping efforts in New England and the Midwest have identified the importance of multiple dates for satellite coverage as an aid in obtaining more accurate and thematically more detailed vegetation maps. To meet the need for multiple dates, GAP is joining with its other Multi-Resolution Land Cover Consortium partners to purchase triple-date coverage of the coterminous United States. Scenes will be selected from available dates for summer 1995 to summer 1997. We believe that this second joint purchase of satellite scenes will greatly help coordinate interagency mapping efforts to develop a fully integrated, second-generation land cover map for the country. This map will be pixel-based with a MMU of 2 ha, a 50-fold higher spatial resolution than our current standard.

Ross Kiester and others used results of Idaho GAP to examine use of different algorithms to prioritize the selection of locations for conservation action and research and found complementarity rather than species richness to be the more defensible approach (Kiester et al. In press). Blair Csuti, working with research groups in England and Australia and the Biodiversity Consortium in Oregon, came to a similar conclusion based on a collaborative analysis of the Oregon data set (Csuti et al. In press).

Several GAP project investigators are testing the assumptions of the vertebrate models. Bill Krohn and his group are using information from the various accuracy assessments conducted to date to identify those species that we have difficulty reliably predicting, looking for commonality in life history, behavior, and demographics. It is hoped that this information can be used to develop more reliable vertebrate models. Of particular interest are any shared behavioral and/or life history or demographic characteristics that difficult-to-predict species may have.

Craig Allen, Wiley Kitchens, and the rest of the folks with Florida GAP have put together what promises to be a very interesting and stimulating program for the 1996 Annual GAP Meeting as well as some great field trips after the meeting. I look forward to seeing you in Key Largo!

Literature Cited

Csuti, B., P.H. Williams, R.L. Pressey, S. Polasky, J.D. Camm, M. Kershaw, A.R. Kiester, B. Downs, R. Hamilton, M. Huso, and K. Sahr. In press. A comparison of reserve selection algorithms using data on terrestrial vertebrates in Oregon. Biological Conservation.

Edwards, T.C., Jr., E.T. Deshler, D. Foster, and G.G. Moisen. 1996. Adequacy of wildlife habitat relation models for estimating spatial distributions of terrestrial vertebrates. Conservation Biology 10:263-270.

Edwards, T.C., Jr., G.G. Moisen, and D.R. Cutler. In press. Assessing map accuracy in an ecoregion-scale cover-map. Ecological Applications.

Kiester, A.R., J.M. Scott, B. Csuti, R.F. Noss, and B. Butterfield. In press. Conservation prioritization using GAP data. Conservation Biology.

J. Michael Scott, Director
National Gap Analysis Program
Moscow, Idaho