2005 Progress Report: Combining Environmental Data Sets

EPA Grant Number: R829095C001
Subproject: this is subproject number 001 , established and managed by the Center Director under grant R829095
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: Space-Time Aquatic Resources Modeling and Analysis Program (STARMAP)
Center Director: Urquhart, N. Scott
Title: Combining Environmental Data Sets
Investigators: Hoeting, Jennifer A. , Breidt, F. Jay , Davis, Richard A. , Gitelman, Alix I. , Ritter, Kerry J. , Stevens, Don L.
Current Investigators: Hoeting, Jennifer A. , Breidt, F. Jay , Davis, Richard A. , Gitelman, Alix I. , Johnson, Devin S. , Ritter, Kerry J. , Stevens, Don L.
Institution: Colorado State University , Oregon State University , Southern California Coastal Water Research Project Authority
Current Institution: Colorado State University , Oregon State University , Southern California Coastal Water Research Project Authority , University of Alaska - Fairbanks
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 2001 through September 30, 2006
Project Period Covered by this Report: October 1, 2004 through September 30, 2005
RFA: Research Program on Statistical Survey Design and Analysis for Aquatic Resources (2001) RFA Text |  Recipients Lists
Research Category: Aquatic Ecosystems , Ecological Indicators/Assessment/Restoration , Water and Watersheds , Water , Ecosystems


The objective of this research project is to develop approaches for spatio-temporal design and modeling to further understanding of aquatic resources.

Progress Summary:

Geof Givens and Jennifer Hoeting finished their book, Computational Statistics (Wiley, 2005). The book is serving as a comprehensive text on modern and classical methods of statistical computing and computational statistics with detailed examples and problems drawn from diverse fields. The book includes examples utilizing Environmental Monitoring and Assessment Program (EMAP) data. It was the basis for two short courses offered by the authors in October 2004; a related course was offered as a part of this year’s Joint Statistical Meetings. Two further short courses are scheduled for the summer of 2006. The book has been adopted as a textbook at a number of universities, including Stanford University and the University of Wisconsin. There was a second printing of the first edition of the book in September 2005.

Andrew Merton, a doctoral student, has developed methods for selecting geostatistical models. Merton, Hoeting, and Davis have presented results of this research at the Joint Statistical Meetings, a National Science Foundation (NSF)-sponsored conference on statistics and the environment, the international biometrics conference, and the Fourth Annual Conference on Statistical Survey Design and Analysis for Aquatic Resources, Corvallis, Oregon, September 2005. Megan Dailey has begun a doctoral project related to modeling of aquatic data from Oregon and Washington EPA EMAP studies. Julia Smith is working on an M.S. degree project aimed at predicting grain (rock) size in wadeable streams using Geographic Information System (GIS) indicators developed for these data. These projects are joint work with another U.S. Environmental Protection Agency Science To Achieve Results grant project (principal investigators Poff and Bledsoe). The aim is for Dailey and Smith to produce papers to appear in the stream ecology and statistical literature. Kathi Georgitis of Oregon State University is working on her doctoral project on parameter estimation for geostatistical models under Gitelman and Hoeting. Urquhart has begun assembling spatially dense aquatic data from numerous sources, in streams, rivers, estuaries, and near coastal systems, and has helped several investigators use various aspects of these data. Josh French compiled some of these datasets and displayed a poster about his findings at the recent joint Designs and Models for Aquatic Resource Surveys/Space-Time Aquatic Resources Modeling and Analysis Program (STARMAP) meeting at Oregon State University. The focus of this project is to investigate the degree of spatial correlation present in aquatic data.

Specific Areas of Research

Develop New Models and Methodology for the Analysis of Compositional Data, Accounting for Correlation Over Space and Time. This includes the use of landscape variables developed under Project 3. Recent efforts have focused on fish and macroinvertebrate richness and is being done with Colorado State University macroinvertebrate ecologists. Johnson’s STARMAP-funded work at the University of Alaska at Fairbanks continued. Hoeting was invited to speak on work with Johnson at the International Conference on Bayesian Statistics and its Applications, Varanasi, India, in January 2005, and the Joint Statistical Meetings in August 2005. A paper on this work will appear in a refereed volume on Bayesian statistics and its applications, edited by S. K. Upadhyay, U. Singh, and D. K. Dey. A paper is in progress with a wildlife biologist at the National Marine Mammal Laboratory and will be submitted in Year 5 of the project.

Model Selection for Geostatistical Models. Hoeting, Merton, and Davis continue this work. A paper submitted to Ecological Applications will be published as part of a special issue on the interface of ecology and statistics. Merton’s doctoral work in this area focuses on the theory and application of the effect of sample size on covariance function estimation for geostatistical models. He cooperated closely with Erin Peterson (Project 3) applying techniques that he has developed to data she has assembled. A joint manuscript from this collaboration recently was accepted for publication.

Hierarchical Bayesian Models for Seasonal Radio Telemetry Habitat Data. Initial work on this project included the study of a model proposed by Ramsey and Usner that incorporates a persistence parameter into the analysis of habitat radio telemetry data. The work of Dailey and Gitelman has extended this model by allowing the persistence parameter to change with season. Along with the seasonal persistence parameter, they have specified a multinomial logit model incorporating a season component. They have developed models in a Bayesian framework applicable to aquatic resources. A manuscript has been accepted from this work; Dailey was the runnerup for best student presentation at the Western North American Region of the International Biometric Society Annual Meeting in Fairbanks, Alaska, in June 2005.

Geostatistic Modeling. Hoeting, Gitelman, and Georgitis continue to collaborate on geostatistical modeling through weekly conference calls. Hoeting visited Oregon State University in April and September 2005 to further this collaboration. Georgitis gave an invited talk on this work at the Western North America Region of the International Biometric Society in Anchorage, Alaska, in June 2005.

Chain Graph Models. A. Gitelman and A. Herlihy continued to develop spatial models using chain graphs and allied model selection tools. Gitelman and Herlihy had a paper accepted for publication in Environmental and Ecological Statistics. Gitelman gave invited talks at the regional Biometric Society meetings in Fairbanks, Alaska, and the Ecological Society Meeting in Montreal, Quebec, Canada.

Detecting Changes in Patterns of Variation in Long-Term Time Series. Davis and colleagues have expanded their work, implemented related computer software, and applied these tools to several environmental problems.

Future Activities:

Jennifer Hoeting, a STARMAP principal investigator, and Geof Givens will present a short course entitled “Statistical Computing: Techniques for Integration and Optimization” at the Alaska Chapter of the American Statistical Association in Juneau, Alaska, in July 2006, and at the Joint Statistical Meetings in Seattle, Washington, in August 2006.

Jennifer Hoeting has been invited to present a talk entitled, “Model Selection and Estimation for Geostatistical Models” at the Joint Statistical Meetings, in Seattle, Washington, in August 2006.

Jennifer Hoeting has been invited to present a talk at the Conference on Bayesian Methods in Wildlife Population Monitoring in Fort Collins, Colorado, in 2006.

Integration of STARMAP activities with Colorado State University’s NSF Integrative Graduate Education and Research Traineeship Program for Interdisciplinary Mathematics, Ecology, and Statistics (PRIMES) continues. PRIMES visitors enrich the STARMAP activities.

Anticipated Research Outputs

  • Tools to generate spatial weights matrices to be used in other statistical software (e.g., R, SPlus, WinBugs, etc.).
  • Research results concerning analysis methodology for use in analyzing aquatic monitoring data, published in both statistical and monitoring outlets.
  • Methodology and tools for predicting the probability that a point on a stream will be perennial or nonperennial. (This will be added to the sampling methodology described above.)
  • Methodology for executing the selection of spatial models.

Journal Articles on this Report : 7 Displayed | Download in RIS Format

Other subproject views: All 78 publications 10 publications in selected types All 8 journal articles
Other center views: All 302 publications 54 publications in selected types All 42 journal articles
Type Citation Sub Project Document Sources
Journal Article Dailey M, Gitelman AI, Ramsey FL, Starcevich S. Habitat selection models to account for seasonal persistence in radio telemetry data. Environmental and Ecological Statistics 2007;14(1):55-68. R829095 (2004)
R829095 (2005)
R829095 (Final)
R829095C001 (2005)
  • Abstract: Springer - Abstract
  • Journal Article Davis RA, Rodriguez-Yam G. Estimation for state-space models based on a likelihood approximation. Statistica Sinica 2005;15(2):381-406. R829095 (Final)
    R829095C001 (2005)
  • Full-text: Institute of Statistical Science Academia Sinica-Full Text PDF
  • Abstract: Statistica Sinica-Abstract
  • Journal Article Gitelman AI, Herlihy A. Isomorphic chain graphs for modeling spatial dependence in ecological data. Environmental and Ecological Statistics 2007;14(1):27-40. R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C001 (2005)
  • Abstract: SpringerLink Abstract
  • Journal Article Hoeting JA, Davis RA, Merton AA, Thompson SE. Model selection for geostatistical models. Ecological Applications 2006;16(1):87-98. R829095 (Final)
    R829095C001 (2004)
    R829095C001 (2005)
    R829095C004 (2005)
  • Abstract from PubMed
  • Full-text: Colorado State University-Full Text PDF
  • Abstract: ESA-Abstract
  • Journal Article Hoeting JA. Some perspectives on modeling species distributions. Bayesian Analysis 2006;1(1):93-98. (Comment on article by Gelfand et al.). R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C001 (2005)
  • Full-text: Carnegie Mellon University-Full Text PDF
  • Abstract: Carnegie Mellon University-Abstract
  • Journal Article Johnson DS, Hoeting JA. Bayesian multimodel inference for geostatistical regression models. PLoS ONE 2011;6(11):e25677. R829095C001 (2005)
  • Full-text: PLoS ONE-Full Text HTML & PDF Link
  • Journal Article Reese GC, Wilson KR, Hoeting JA, Flather CH. Factors affecting species distribution predictions: a simulation modeling experiment. Ecological Applications 2005;15(2):554-564. R829095 (Final)
    R829095C001 (2005)
  • Full-text: U.S. Forest Service-Full Text PDF
  • Abstract: Ecological Society of America-Abstract
  • Supplemental Keywords:

    latent processes, matern covariance function, model selection, remote sensing, sampling design,, RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Aquatic Ecosystem, Environmental Monitoring, EMAP, ecosystem monitoring, statistical survey design, spatial and temporal modeling, aquatic ecosystems, water quality, Environmental Monitoring and Assessment Program, modeling ecosystems

    Relevant Websites:

    http://www.stat.colostate.edu/starmap Exit

    Progress and Final Reports:

    Original Abstract
  • 2002
  • 2003 Progress Report
  • 2004 Progress Report
  • Final Report

  • Main Center Abstract and Reports:

    R829095    Space-Time Aquatic Resources Modeling and Analysis Program (STARMAP)

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R829095C001 Combining Environmental Data Sets
    R829095C002 Local Inferences from Aquatic Studies
    R829095C003 Development and Evaluation of Aquatic Indicators
    R829095C004 Extension of Expertise on Design and Analysis to States and Tribes
    R829095C005 Integration and Coordination for STARMAP