2004 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. , Johnson, Devin S. , Stevens, Don L. , Urquhart, N. Scott , Weisberg, Steven B.
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 , University of Alaska - Fairbanks
EPA Project Officer: Hiscock, Michael
Project Period: October 1, 2001 through September 30, 2006
Project Period Covered by this Report: October 1, 2003 through September 30, 2004
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

Objective:

The objective of this research project is to develop approaches for spatio-temporal design and modeling to further the understanding of aquatic resources. This is one of several projects conducted by the STARMAP Center. The progress of the other projects is reported in separate reports (see the Annual Reports for R829095 and R829095C002 through R829095C005).

Progress Summary:

With the addition of five new research areas and impressive output from the existing subprojects, the Combining Environmental Data Sets (R829095C001) group was quite productive in Year 3 of the project. Numerous A large papers were submitted to peer-reviewed journals and a number of talks were given, including several invited talks and one keynote address (Dr. Davis). These talks were presented to a number of audiences, including user communities. Highlights of the year include: completion of the Statistical Computing book and delivery of short courses (Dr. Hoeting); an important new interaction with the City of San Diego and submission of a related paper (Drs. Ritter, Leecaster, and Urquhart); submission of three papers by Gitelman (and co-authors) on three different subjects; submission and acceptance of a paper by Drs. Hoeting, Merton, and Davis; submission of a paper by Drs. Johnson and Hoeting; and establishment of a new subcontract with the University of Alaska–Fairbanks (UAF) to support Dr. Johnson’s continued work. Graduate students Kathi Georgitis (Oregon State University [OSU]), Megan Dailey (Colorado State University [CSU]), and Joshua French (CSU) have joined the Space-Time Aquatic Resources Modeling and Analysis Program (STARMAP) project. Andrew Merton’s Ph.D. research continues. Mr. Merton won an award for best student paper at an American Statistical Association (ASA) conference.

Specific Areas of Research

Book on Statistical Computing. Participants: J. Hoeting and G. Givens (CSU)

  • Goals: To complete a book on statistical computing.
  • Year 3 Outcomes: The book is essentially complete, with minor copy editing still remaining. A Web page has been created. Drs. Hoeting and Givens presented two short courses in Chicago, in October 2004, at a conference on Computational Environmetrics, which was co-sponsored by the Section on Statistics and the Environment (ASA) and the Center for Integrating Statistical and Environmental Science (CISES) at the University of Chicago.
  • Upcoming: Possible short course to be given at the Joint Statistical Meetings (JSM) in Minneapolis, Minnesota, in August 2005. The book is expected to appear in early 2005.

Models for Compositional Data. Participants: D.S. Johnson and J.A. Hoeting

  • Goals: To develop new models and methodology for the analysis of compositional data. These models will be used to develop a better understanding of species traits of benthic invertebrates and may be useful in the development of indicators of stream fitness.
  • 2004 Outcomes: Devin Johnson assumed a position in the Department of Mathematical Sciences at UAF in August 2003. He continues to work with STARMAP through a subcontract from CSU to UAF that started in summer 2004. He is participating in the development of a program in environmental statistics. Drs. Hoeting and Johnson submitted a manuscript to The Annals of Statistics in January 2004. An additional manuscript is being prepared for possible submission to The Journal of the American Statistical Association in 2005. Dr. Johnson gave invited talks at JSM in Toronto in August 2004, and at the Conference on Computational Environmetrics in October 2004 at the University of Chicago. He also gave a presentation at the STARMAP conference in Fort Collins, CO.
  • Upcoming: A paper will be submitted for publication. Dr. Hoeting will give invited talks at the International Society for Bayesian Analysis meeting in Varanasi, India, in January 2005, and at JSM in Minneapolis in August 2005.

Design and Analysis Methods for Geo-Referenced Binary Data. Participants: M. Leecaster (Idaho National Engineering and Environmental Laboratory), K.J. Ritter (Southern California Coastal Water Research Project [SCCWRP]), and S.B. Weisberg (SCCWRP)

  • Goals: To determine appropriate grid size for producing maps using Markov random field models and compare to geostatistical modeling approaches.
  • Year 3 Outcomes: This work has progressed from theory to application with a design implemented and executed by the City of San Diego in summer 2004. Ritter gave presentations at the Southern California Academy of Sciences Meeting, the Graybill Conference and the 2004 STARMAP/Designs and Models for Aquatic Resource Surveys meeting. A manuscript has been submitted to Ecological and Environmental Statistics.
  • Upcoming: The contract has been completed; there is the possibility for funding in 2005.

Geostatistical Models: Model Selection and Asymptotics. Participants: A. Merton (CSU doctoral student), J. Hoeting, and R. Davis

  • Goals: To produce new methodology and perhaps new theoretical results related to model selection for geostatistical models.
  • Year 3 Outcomes: A paper was accepted and will appear in Ecological Applications in 2005. Dr. Hoeting presented at the National Science Foundation (NSF) Conference on Statistics and Ecology in Jackson Hole, Wyoming, in December 2003. Dr. Davis gave the keynote address at the International Biometrics Conference in Cairns, Australia, in July 2004. Mr. Merton presented at the ASA Colorado-Wyoming Chapter Spring Meeting in Boulder, Colorado, in April 2004 (best student presentation award); the JSM in Toronto in August 2004; and the Monitoring Science and Technology Symposium in Denver, Colorado, in September 2004.
  • Upcoming: Mr. Merton has been invited to present at the 33rd Annual Meeting of the Statistical Society of Canada and at the Western North America Region of The International Biometric Society (WNAR) Annual Meeting, both in June 2005. Mr. Merton will have his Ph.D. preliminary exam on this work in late 2004. His goals continue to be to earn his doctoral degree and publish additional papers.

Hierarchical Models for Analyzing Radio Telemetry Habitat Association Data. Participants: M. Dailey, A. Gitelman, and F. Ramsey (OSU)

  • Goals: To develop hierarchical Bayesian models to model radio telemetry data on habitat selection by fish in Oregon streams.
  • Year 3 Outcomes: Drs. Gitelman and Dailey submitted a paper to Environmental and Ecological Statistics. They presented posters at the Graybill Conference in Fort Collins, Colorado; the STARMAP meeting in Fort Collins, Colorado; and The International Environmetrics Society (TIES) Conference in Portland, Maine. Dr. Dailey also gave an invited talk at JSM in Toronto.
  • Upcoming: The subproject has been completed, but several possible extensions to this work may be pursued.

Multiscale Analysis (new subproject). Participants: A. Gitelman and K. Georgitis

  • Goals: To develop methodology for multi-scale analysis
  • Year 3 Outcomes: A paper was submitted to Environmental and Ecological Statistics; a poster was presented at the TIES Meeting in Maine; and Ms. Georgitis gave a presentation at the STARMAP conference in Fort Collins, Colorado.
  • Upcoming: Dr. Gitelman will give invited talks at the WNAR Annual Meeting in Fairbanks, Alaska, in June 2005, and at the Ecological Society of America Annual Meeting in Montreal, Quebec, in August 2005.

Bayesian Spatial Models for Stream Data (new subproject). Participants: A. Gitelman, K. Georgitis, and J. Hoeting

  • Goals: To complete Bayesian spatial modeling of SO4 in streams using the U.S. Environmental Protection Agency Mid-Atlantic Highlands Assessment/Mid-Atlantic Integrated Assessment dataset. Note that this subproject is a joint OSU/CSU effort.
  • Year 3 Outcomes: The subproject started in October 2004. Exploratory data analysis has begun.
  • Upcoming: An invited paper will be submitted to Environmental and Ecological Statistics in spring 2005. Ms. Georgitis was invited to present at the WNAR Annual Meeting in Fairbanks, Alaska, in June 2005.

Bayesian Model Determination for Geostatistical Regression Models (new subproject). Participants: D. Johnson

  • Goals: To develop Reverse-Jump Markov Chain Monte Carlo methodology for model selection.
  • Year 3 Outcomes: Dr. Johnson co-taught a workshop on analyzing trend and abundance for resource monitoring data for agency personnel in Fairbanks, Alaska, in November 2004, and was invited to present at the Conference on Computational Environmetrics, in Chicago, Illinois, in October 2004. A manuscript is in preparation for invited submission to Environmetrics.
  • Upcoming: Submit the manuscript for publication.

Models for Compositional Data, Part 2 (new subproject). Participants: M. Dailey and J. Hoeting

  • Goals: To extend existing models for compositional data.
  • Upcoming: Dr. Dailey will begin research. Dr. Hoeting will give an invited talk at JSM in Minneapolis, Minnesota, in August 2005.

Evaluation of Degree of Spatial Correlation in Aquatic Systems (new subproject). Dr. Urquhart has sought aquatic data sets that support the exploration of possible spatial correlation from the Ohio River to estuaries to streams. Mr. French is evaluating spatial correlation across a number of responses and settings to try to establish the extent of spatial correlation across a number of aquatic systems; the work with the Ohio River data set is nearing completion. Expected output will include at least one professional presentation and possibly a journal article.

Chain Graph Models. Participants: A. Gitelma, A. Herlihy (OSU), and S. Jensen (OSU graduate student). (Note: this subproject was formerly listed under R829095C002, Local Inferences From Aquatic Studies.)

  • Goals: To develop spatial models using chain graphs and allied model selection tools.
  • Year 3 Outcomes: Drs. Gitelman and Herlihy submitted a paper to Environmental and Ecological Statistics. Dr. Gitelman gave an invited talk at the Graybill Conference and at the STARMAP meeting, both in Fort Collins, Colorado. Stephen Jensen gave a presentation at the STARMAP meeting in Fort Collins. Mr. Jensen completed an M.S. degree and now is employed at CleverSet, a small Bayes Network company in Corvallis, Oregon.
  • Upcoming: Dr. Gitelman and Mr. Jensen may prepare a paper for publication.

Future Activities:

Jennifer Hoeting and Geof Givens will present two short courses based on their book in progress in conjunction with the Computational Environmetrics Workshop in Chicago, Illinois, October 21, 2004:

  • Methods of Integration for Environmental Problems in Statistics: Quadrature, Monte Carlo Integration, and Markov Chain Monte Carlo Methods
  • Optimization Methods for Environmental Problems in Statistics: Numerical Maximum Likelihood, Combinatorial Optimization, and EM Algorithm

Jennifer Hoeting has been invited to present a talk at the International Workshop/Conference on Bayesian Statistics and its Applications, Banaras Hindu University, Varanasi, India, January 5-8, 2005.

Jennifer Hoeting has been invited to present a talk at JSM in Minneapolis, Minnesota, August 2005.

Jennifer Hoeting and Geof Givens may present a short course on material related to their upcoming book at JSM in Minneapolis, Minnesota, August 2005.

Andrew Merton will give invited talks at the 33rd Annual Meeting of the Statistical Society of Canada, Saskatoon, Canada and the WNAR Annual Meeting, Fairbanks, Alaska, both in June 2005.

Alix Gitelman will give invited talks at the WNAR Annual Meeting in Fairbanks, Alaska, in June 2005, and the Ecological Society of America Annual Meeting in Montreal, Quebec, in August 2005.

Kathi Georgitis will give an invited talk at the WNAR Annual Meeting in Fairbanks, Alaska, in June 2005.

Integration of STARMAP activities with our NSF Integrative Graduate Education and Research Traineeship Program for Interdisciplinary Mathematics, Ecology, and Statistics (PRIMES) continues. PRIMES visitors, such as A. Gelfand (Duke), enrich the STARMAP activities.-I.

Alix Gitelman, Kathi Georgitis, and Jennifer Hoeting will submit an invited manuscript to Environmental and Ecological Statistics in spring 2005.


Journal Articles on this Report : 1 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 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
    Exit
  • Abstract: ESA-Abstract
    Exit
  • 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:

    Relevant Web Sites: http://www.stat.colostate.edu/starmap Exit

    Progress and Final Reports:

    Original Abstract
  • 2002
  • 2003 Progress Report
  • 2005 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