2004 Progress Report: Local Inferences from Aquatic Studies

EPA Grant Number: R829095C002
Subproject: this is subproject number 002 , 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: Local Inferences from Aquatic Studies
Investigators: Breidt, F. Jay , Davis, Richard A. , Hoeting, Jennifer A.
Current Investigators: Breidt, F. Jay , Davis, Richard A. , Hoeting, Jennifer A. , Opsomer, Jean
Institution: Colorado State University
Current Institution: Colorado State University , Iowa State University
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 hierarchical spatio-temporal models for local inferences about 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, R829095C001, and R829095C003 through R829095C005).

Progress Summary:

Work continues in the use of nonparametric modeling for survey regression estimation, jointly supported by Designs and Models for Aquatic Resource Surveys (DAMARS) and Space-Time Aquatic Resources Modeling and Analysis Program (STARMAP). Numerous invited talks have been given, and several manuscripts have been submitted; three have been published and another accepted for publication. The problems of interest in this context combine landscape-level auxiliary data (such as those from GIS coverages) with field observations. The inferential problems range from model-assisted descriptive inferences for aquatic populations, to model-based small area estimates. Drs. Jay Breidt and Jean Opsomer, together with students and colleagues, extended earlier results on local polynomial survey regression estimation in a number of directions, some of which are described here. The M.S. project of Alicia Johnson on cumulative distribution function estimation was submitted for publication, and the M.S. project of Siobhan Everson-Stewart on two-dimensional kernel estimators was extended by Giovanna Ranalli, a post-doctoral research fellow, to the setting of penalized splines. Giovanna Ranalli and Jay Breidt also obtained promising preliminary results on the use of low rank radial basis functions for smoothing data from river networks.

Ji-Yeon Kim successfully defended her Ph.D. under the direction of Drs. Breidt and Opsomer, and submitted a joint paper on two-stage local polynomial regression estimation. New work conducted with Gerda Claeskens on nonparametric model-assisted estimation using penalized splines was submitted to Biometrika, and currently is being extended to the case of small area estimation. Ph.D. student Mark Delorey is extending penalized spline estimators further to the setting of two-stage sampling. Nan-Jung Hsu and Hsin-Cheng Huang are visiting research scientists collaborating with Dr. Breidt and other STARMAP researchers during the period from August to November 2004. Drs. Breidt and Hsu will continue work on semiparametric modeling for increment-averaged data, such as arise in soil and sediment core sampling. Drs. Hsu and Breidt also will work on Bayesian estimation for non-Gaussian, non-invertible, moving average models, with potential application to spatial data from river networks. Hsin-Cheng Huang will work on improving the power in trend detection of lake water quality using spatio-temporal models. Drs. Hsu, Breidt, and Huang also propose to develop a new variable selection method for multiple layers of GIS data using LASSO. Such a methodology would be immediately applicable to aquatic resource data, for which multiple layers of geospatial data are available as potential regressors.

Dr. Breidt and Ph.D. student Mark Delorey are working on the broad topic of small area estimation, investigating methods for incorporating spatial relationships. The small areas of interest, for now, consist of watersheds (HUC8’s) in the Mid-Atlantic Highlands Region, and the responses focus on characteristics of water quality. They currently are using Bayesian methods to construct a set of ensemble estimates for all watersheds, and plan to investigate the same problem using non-parametric/semi-parametric methods.

Dr. Breidt and colleague Jean Opsomer at Iowa State are collaborating on semi-parametric model-assisted regression estimation, as well as model-based estimation. These methods can be useful for small area estimation, and will be compared to fully parametric procedures. This project has made substantial progress in developing model-assisted estimates of distribution functions, primarily through the M.S. project of Alicia Johnson. (She completed her degree, spent a year in an environmental position with the Centers for Disease Control and Prevention, and now is beginning Ph.D. studies in Statistics at the University of Minnesota.) The estimator compares favorably to other parametric and nonparametric alternatives. Results have been presented at several statistical meetings, and a paper on this has been submitted to the Canadian Journal of Statistics.

As an extension of the nonparametric model-assisted estimation project, Jean Opsomer and Jay Breidt have begun research in applying the nonparametric regression methodology to the small area estimation context, and are applying it to the Northeastern Lakes survey. Preliminary results based on this work were presented at the STARMAP/DAMARS meeting at Colorado State University in September 2004. A simulation study was conducted by M.S. student Everson-Stewart. (She completed her degree, took a job with Amgen for 1 year, and now is continuing doctoral studies at the University of Washington.) Jean Opsomer also presented a talk on this topic at the 2004 Joint Statistical Meetings (JSM) as well as a number of seminars at universities in the United States and abroad.

The choice of the smoothing parameter has an important effect on nonparametric regression estimators, and this is also true for model-assisted estimators. Jean Opsomer and Curtis Miller, a graduate student at Iowa State University, have been developing a cross validation-based method for selecting the smoothing parameter, and presented preliminary results at the JSM in August, 2003.

Future Activities:

Jay Breidt, Jean Opsomer, Mark Delorey, and Giovanna Ranalli will continue to develop semiparametric methods for small area estimation, focusing on comparison of penalized splines methods in a spatial context. These semiparametric procedures will be compared to fully parametric methods. A paper on this work, possibly using acid neutralizing capacity data from The Mid-Atlantic Highlands Assessment, is planned.

Breidt and Opsomer are finalizing a manuscript on semiparametric additive regression estimation, which can accommodate models with multiple auxiliary variables of both continuous and categorical types, and can be a useful estimation procedure for regions with moderate sample sizes. Jay Breidt, Jean Opsomer, and Siobhan Everson-Stewart plan to complete and submit a paper on nonparametric regression estimation for two-stage spatial sampling. Applications to small area estimation will be considered.


Journal Articles on this Report : 11 Displayed | Download in RIS Format

Other subproject views: All 83 publications 20 publications in selected types All 18 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 Andrews B, Davis RA, Breidt FJ. Maximum likelihood estimation for all-pass time series models. Journal of Multivariate Analysis 2006;97(7):1638-1659. R829095 (Final)
R829095C002 (2003)
R829095C002 (2004)
R829096 (2003)
  • Full-text: ScienceDirect-Full Text PDF
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Journal Article Andrews B, Davis RA, Breidt FJ. Rank-based estimation for all-pass time series models. Annals of Statistics 2007;35(2):844-869. R829095 (2005)
    R829095 (Final)
    R829095C002 (2003)
    R829095C002 (2004)
  • Full-text: Project Euclid-Full Text PDF
    Exit
  • Abstract: Project Euclid-Abstract
    Exit
  • Other: Cornell University-Full Text PDF
    Exit
  • Journal Article Breidt FJ, Hsu N-J. Best mean square prediction for moving averages. Statistica Sinica 2005;15(2):427-446. R829095 (Final)
    R829095C002 (2003)
    R829095C002 (2004)
    R829095C002 (2005)
    R829096 (2003)
    R829096 (2005)
  • Full-text: StatisticaSinica-Full Text PDF
    Exit
  • Abstract: StatisticaSinica-Abstract
    Exit
  • Journal Article Breidt FJ, Opsomer JD, Johnson AA, Ranalli MG. Semiparametric model-assisted estimation for natural resource surveys. Survey Methodology 2007;33(1):35-44. R829095 (Final)
    R829095C002 (2003)
    R829095C002 (2004)
  • Full-text: Statistics Canada-Full Text PDF
    Exit
  • Abstract: Statistics Canada-TOC
    Exit
  • Journal Article Da Silva DN, Opsomer JD. Properties of the weighting cell estimator under a nonparametric response mechanism. Survey Methodology 2004;30(1):45-55. R829095 (2003)
    R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
    R829096 (2004)
    R829096 (2005)
  • Full-text: Statistics Canada-Full Text PDF
    Exit
  • Abstract: Statistics Canada-Abstract
    Exit
  • Journal Article Francisco-Fernandez M, Opsomer JD. Smoothing parameter selection methods for nonparametric regression with spatially correlated errors. Canadian Journal of Statistics 2005;33(2):279-295. R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
  • Full-text: Universidade da Coruna-Preprint PDF
    Exit
  • Abstract: Wiley-Abstract
    Exit
  • Journal Article Francisco-Fernandez M, Jurado-Exposito M, Opsomer JD, Lopez-Granados F. A nonparametric analysis of the spatial distribution of Convolvulus arvensis in wheat-sunflower rotations. Environmetrics 2006;17(8):849-860. R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
  • Full-text: Universidade da Coruna-Prepublication PDF
    Exit
  • Abstract: Wiley Interscience Abstract
    Exit
  • Journal Article Montanari GE, Ranalli MG. Nonparametric model calibration estimation in survey sampling. Journal of the American Statistical Association 2005;100(472):1429-1442. R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
    R829096 (2004)
    R829096 (2005)
  • Abstract: American Statistical Association-Abstract
    Exit
  • Journal Article Opsomer JD, Botts C, Kim JY. Small area estimation in a watershed erosion assessment survey. Journal of Agricultural, Biological, and Environmental Statistics 2003;8(2):139-152. R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C002 (2004)
    R829096 (2004)
    R829096 (2005)
  • Full-text: Research Gate-Abstract and Full Text HMTL
    Exit
  • Abstract: SpringerLink-Abstract
    Exit
  • Journal Article Opsomer JD, Miller CP. Selecting the amount of smoothing in nonparametric regression estimation for complex surveys. Journal of Nonparametric Statistics 2005;17(5):593-611. R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
  • Abstract: Taylor & Francis-Abstract
    Exit
  • Journal Article Opsomer JD, Breidt FJ, Moisen GG, Kauermann G. Model-assisted estimation of forest resources with generalized additive models. Journal of the American Statistical Association 2007;102(478):400-409. R829095 (2004)
    R829095 (2005)
    R829095 (Final)
    R829095C002 (2004)
    R829095C002 (2005)
    R829096 (2003)
    R829096 (2004)
    R829096 (2005)
  • Full-text: US Forest Service-Full Text PDF
    Exit
  • Abstract: Journal of the American Statistical Association-Abstract
    Exit
  • Supplemental Keywords:

    path analysis, kernel regression, thin plate splines, 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, STARMAP

    Relevant Websites:

    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