Progress 07/15/08 to 07/14/09
Outputs
OUTPUTS: MODEL DEVELOPMENT--We developed a final model-building protocol after completing a review of the important, but often overlooked, assumptions and predictions from ecological (habitat selection and metapopulation) theory related as they relate to variation in species occurrence over time. We describe in a publication (see below) how these perspectives can be incorporated into models of species distribution with data on the detection/non-detection of species across environmental gradients. As a case study, we applied these models to evaluate model fit and predictive ability relative to more conventional modeling approaches using a relatively large-scale and long-term data set of bird populations inhabiting forests across Montana and northern Idaho. A University of Florida MS student, Chris Rota, has further developed these approaches in two articles (one in preparation and one in revision; see below). WEB DEVELOPMENT--We have a considerable amount of time during the past year working with GCS-Research to develop a web-based Decision Support Tool that can incorporate our landbird data and the bird distribution models. This is still unfinished, but the work in progress can be viewed here: http://development.gcs-holdings.net/UMAvian/GoogleMap.aspx COURSE DEVELOPMENT/MODIFICATION--RJF offered two courses in Landscape Ecology (WIS 4203C and WIS 6468C) at the University of Florida, which included lab and lecture material on species distribution models and decision support tools, and RLH modified and expanded the habitat relationships and modeling sections of one undergraduate course (BIOL 304, Ornithology) at the University of Montana. WORKSHOP COORDINATION--We are currently organizing a national workshop through the Avian Knowledge Alliance on the kinds of Decision Support Tools that will make the best use of our map-based products. The meeting is to be held in Reno, Nevada, 13-15 October 2009. PRENSENTATION--R. Fletcher (speaker) and C. Rota (MS student) presented three papers at the Wildlife Society 15th Annual Conference, 3 November 2008, Miami, FL, and Fletcher gave an invited seminar at the Michigan State University's Kellogg Biological Station. PARTICIPANTS: INDIVIDUALS--R. Hutto, PI, was involved with supervision of project and with design and implementation of web interface with partner business firm, GCS Research. R. Fletcher, PI, was involved with model building and supervision of graduate student who completed his M.S. degree using data from the project. C. Rota was the graduate student who finished his M.S. from the University of Florida in spring 2009. J. Young worked full time on data management and model building at the Avain Science Center. A. Noson worked part-time on extracting variables from GIS data layers for habitat relationships models. GCS Research is a partner organization contracted to build the Decision Support Tool web interface. Graduate student training was completed with the graduation of C Rota. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
IMPACT--Change in conditions: as outlined above, we built a new interactive web interface that will serve as a decision support tool for managers, we included new modeling material in courses, and we presented findings to peers at professional meetings. EVALUATION--none yet.
Publications
- Rota, C. T., R. J. Fletcher, Jr., R. Dorazio, and M. G. Betts. 2009. Occupancy estimation and the closure assumption. Journal of Applied Ecology, in press.
- Fletcher, R. J., Jr., J. S. Young, R. L. Hutto, A. Noson, and C. T. Rota. 2009. Insights from ecological theory on temporal dynamics and species distribution modeling. A. Drew, F. Huettmann, and Y. Wiersma, Editors, Predictive Modeling in Landscape Ecology. Springer, in press.
|
Progress 07/15/07 to 07/14/08
Outputs
OUTPUTS: OUTPUTS Field work--During the 2008 field season, our goal was to obtain field data that could be used to validate bird habitat-relationships models that were built last year. Four observers conducted 951 point counts on 126 transects to complete the sample of 1900 points on 245 transects that we began in 2007. This sample was randomly selected from transects that were previously visited from between 1 and 8 previous years. On each transect, we visited 4 previously surveyed, permanently marked, on-road or on-trail points and 4 new off-road points that were located >300 m from the existing transect. Observers recorded 9246 bird detections distributed among 98 bird species. Our field work for this project is now complete. Model development--We built pilot models for four bird species (Dusky Flycatcher, Dark-eyed Junco, Hermit Thrush, and Ruby-crowned Kinglet) that represent extremes in the range of habitat types occupied and the variation in occupancy through time. We used existing data from the Landbird Monitoring Program (spanning 11 years and from over 2300 locations) to build species distribution models and for externally validating the performance of these models, based on predictions at new locations. For each species, we compared a number of different model types, including more conventional logistic regression approaches, metapopulation models, and models that explicitly incorporated predictions and assumptions from habitat selection theory (based on spatio-temporal variation in population size, as outlined in our proposal). For each species, models that incorporated predictions from theory fit the data better, based on model selection criteria, and also performed better, based on external validation. These results were presented at a professional meeting (see below) and are currently being written up for publication. Workshop--We attended the Northern Region Landbird Monitoring Program Annual Biologists' Workshop in Missoula, MT, on 4 December 2007 to conduct a workshop designed to obtain feedback on decision support tools that USFS Northern Region biologists feel they would use. DISSEMINATION Course development/modification--We incorporated new lecture material on habitat relationships modeling, and we conducted a pilot data collection project using Google Earth interface in Ornithology (Biology 304) during the spring semester, 2008. The data collection and presentation effort will be developed further as we refine our plans to build an effective decision support tool for the USDA project. Presentations in workshops/symposia-- "Novel use of a landbird database to inform management" presented by J. S. Young (speaker), R. F. Fletcher, and R. L. Hutto at the 4th International Partners in Flight Conference, McAllen, TX, 13-16 February 2008. Participant in workshop sponsored by Montana Fish, Wildlife and Parks to evaluate the plan to coordinate habitat conservation and restoration in Montana, Helena, MT, 1 November 2007. We discussed results from responses to a survey we conducted on desired decision support tools at the Northern Region Landbird Monitoring Program Annual Biologists' Workshop, Missoula, MT, 4 December 2007. PARTICIPANTS: INDIVIDUALS PI Hutto--provided project oversight and supervision on a continuing basis; participated in weekly model-building phone conferences. PI Fletcher--coordinated model-building effort; supervised graduate student project conducted out of the University of Florida. Jock Young--conducted all data management, field crew training and field crew supervision; provided all final model equations to GIS expert. Anna Noson--conducted all GIS data extraction and served as liason with USFS GIS personnel; provided all map services for our pilot effort to model the occurrence of bird species. Amy Cilimburg--supervised all Avian Science Center staff involved in the project; helped with field training. Chris Rota--Graduate student at University of Florida working on model output comparisons. TRAINING One M.S. graduate student is working on modeling approaches for part of his thesis. Forty undergrauate students in ornithology received training on modeling, conducted field work to map winter bird distributions in Missoula, and worked with a Google Earth data interface for data visualization as a pilot effort to inform our work on decision support tool design. TARGET AUDIENCES: TARGET AUDIENCES undergraduate students--through exposure to modeling in ornithology (included a cross section of social and economic groups) graduate student--through modeling effort EFFORTS taught modeling approaches in an undergraduate class at University of Montana and a graduate class at the University of Florida conducted a workshop for USFS biologists which involved the collection of survey data and a discussion of how to build a data visualization decision support tool. PROJECT MODIFICATIONS: The movement of one PI (Fletcher) from University of Montana to University of Florida has created challenges, but our weekly phone conferences seem to be working well.
Impacts
None to report for the current funding year; most will occur in the next year.
Publications
- Fletcher, R. J., L. Ries, J. Battin, and A. D. Chalfoun. 2007. The role of habitat area and edge in fragmented landscapes: definitively distinct or inevitably intertwined Canadian Journal of Zoology 85:1017-1030.
- Fletcher, R. J. 2007. Species interactions and population density mediate the use of social cues for habitat selection. Journal of Animal Ecology 76:598-606.
- Fletcher, R. J., Jr., and R. L. Hutto. 2008. Partitioning the multi-scale effects of human activity on the occurrence of riparian forest birds. Landscape Ecology 23:727-739.
|
Progress 07/15/06 to 07/14/07
Outputs
Field work--During the 2007 field season, 4 observers conducted 961 point counts on 119 transects that were randomly selected from a pool of 240 transects that were previously visited in at least 8 different years; the other half of the transect pool is scheduled to be surveyed in 2008. On each transect, we visited 4 previously surveyed permanently marked, on-road or on-trail points and 4 new points that were located >300 m from the existing transect. Observers recorded 11537 bird detections distributed among 80 bird species. Model development--We built pilot models for four bird species (Dusky Flycatcher, Dark-eyed Junco, Hermit Thrush, and Ruby-crowned Kinglet) that represent extremes in the range of habitat types occupied and the variation in occupancy through time. We used existing data from the Landbird Monitoring Program (spanning 11 years and from over 2300 locations) to build species distribution models and for externally validating the performance of these
models, based on predictions at new locations. For each species, we compared a number of different model types, including more conventional logistic regression approaches, metapopulation models, and models that explicitly incorporated predictions and assumptions from habitat selection theory (based on spatio-temporal variation in population size, as outlined in our proposal). For each species, models that incorporated predictions from theory fit the data better, based on model selection criteria, and also performed better, based on external validation. These results were presented at a professional meeting (see below) and are currently being written up for publication. Course development/modification--We offered a team taught graduate seminar (BIOL 594, Habitat Relationships Modeling) during the fall semester, 2006, and we modified and expanded the habitat relationships and modeling sections of one undergraduate course (BIOL 304, Ornithology) during the spring semester, 2007.
Presentations in workshops/symposia-- Workshop coordinators--Cilimburg, Young, and Hutto presented results and coordinated discussion related to modeling efforts associated with the Northern Region Landbird Monitoring Program at the Annual USFS Biologists Workshop, Missoula, MT; 7 December 2006. Presentation--Fletcher (speaker) and Hutto presented a talk entitled "Partitioning human impacts on riparian bird distribution along the Madison and Missouri Rivers, Montana" at the Wildlife Society Montana Chapter meetings, Bozeman, MT; 6-9 February 2007. Invited symposium presentation--Fletcher (speaker), Young, Noson, and Hutto presented a talk entitled "Incorporating ecological theory into models of habitat quality: the importance of variation in space and time" at the US-International Association of Landscape Ecologists meeting in Tucson, AZ, 9-13 April 2007. Workshop presentation--Hutto and Cilimburg provided instruction to K-12 teachers on use of the landbird monitoring data to model
bird distribution using a GIS-based system at the "Montana K-12 teacher GTEC workshop," Missoula, MT, 18 June 2007.
Impacts
Expected impact--Land managers are faced with the challenge of balancing resource extraction and environmental protection. The U.S. Forest Service and other public land management agencies must assess whether their management actions meet the requirements of several legal mandates requiring that they operate in a fashion that is both sustainable and does not compromise the ecological integrity of the systems within which they work. These requirements affect agricultural productivity in at least two ways: (1) legal challenges may delay timber harvests and other land-management activities, and (2) lack of knowledge about the ecological consequences of alternative management scenarios hinders rapid movement toward solutions that produce the greatest ratio of agricultural benefit to ecological cost. If managers could perceive potential ecological costs of alternative land-use scenarios in advance, they could avoid legal quagmires that might otherwise result, and the
sustainable level of wood fiber production and other kinds of resource extraction should improve. One approach for understanding ecological costs associated with proposed land-use activities is to build models to predict the occurrence of species (effects of management) across a range of vegetation and land-use categories. We are using a long-term database on bird populations, available from the USFS Northern Region Landbird Monitoring Program, to develop and validate habitat-based models for a large suite of species, and to provide this information to managers in ways that can inform land-use decisions.
Publications
- Papers published-- Fletcher, R. J., Jr. 2007. Species interactions and population density mediate the use of social cues for habitat selection. Journal of Animal Ecology 76:598-606.
- Fletcher, R. J., Jr., L. Ries, J. Battin, and A. D. Chalfoun. 2007. The role of habitat area and edge in fragmented landscapes: definitively distinct or inevitably intertwined? Canadian Journal of Zoology: in press. Invited article.
|
|