Grantee Research Project Results
2002 Progress Report: National Research Program on Design-Based/Model-Assisted Survey Methodology for Aquatic Resources
EPA Grant Number: R829096Center: Center for Air, Climate, and Energy Solutions
Center Director: Robinson, Allen
Title: National Research Program on Design-Based/Model-Assisted Survey Methodology for Aquatic Resources
Investigators: Stevens, Don L. , Gitelman, Alix I. , Urquhart, N. Scott , Herlihy, Alan T. , Muñoz-Hernández, Breda , Sifneos, Jean , Opsomer, Jean , Courbois, Jean-Yves , Conquest, Loveday , Murtaugh, Paul , Hughes, Robert , Smith, Ruben , Carroll, Steven , Lesser, Virginia
Current Investigators: Stevens, Don L. , Urquhart, N. Scott , Herlihy, Alan T. , Lesser, Virginia
Institution: Oregon State University , University of Washington , Iowa State University , Colorado State University
Current Institution: Oregon State University , Colorado State University
EPA Project Officer: Packard, Benjamin H
Project Period: October 15, 2001 through October 14, 2005 (Extended to October 13, 2006)
Project Period Covered by this Report: October 15, 2001 through October 14, 2002
Project Amount: $2,989,884
RFA: Research Program on Statistical Survey Design and Analysis for Aquatic Resources (2001) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Tribal Environmental Health Research , Watersheds , Water , Aquatic Ecosystems
Objective:
The main objective of this research project is to develop and implement design-based/model-assisted statistical methods for aquatic surveys.
Progress Summary:
The research effort for the National Research Program on Design-Based/Model-Assisted Survey Methodology for Aquatic Resources is organized into four projects: (1) integration and extramural outreach; (2) survey design methodology for aquatic resources; (3) parametric model-assisted survey methods for environmental surveys; and (4) nonparametric model-assisted survey estimation for aquatic resources. Research progress is reported by each project.
Integration and Extramural Outreach Project
The objectives of Project 1 are to: (1) identify and articulate issues in aquatic monitoring that require application of statistical expertise; (2) focus statistical expertise on the aquatic monitoring issues encountered by states and tribes; (3) ensure communication between the statistical methodology projects of the program; (4) provide expert survey design and analysis assistance to states and tribes; and (5) transfer this statistical expertise to states and tribes through a combination of supervised application of statistical tools and more structured distance learning techniques.
The major effort in this first year has been to collaborate with state agencies, particularly the three state programs identified in the Oregon State University (OSU) proposal as collaborators? the Oregon Department of Fish and Wildlife (ODFW), the Surface Waters Ambient Monitoring Program (SWAMP), and the San Francisco Estuary Institute (SFEI). We also have completed the ODFW report entitled, "Sampling Design and Statistical Analysis Methods for the Integrated Biological and Physical Monitoring of Oregon Streams." The San Francisco Estuary Regional Monitoring Program for Trace Substances (RMP) has been redesigned using a panel design similar to the one used by ODFW. The design divides the estuary into five segments, with a rotating panel design in each segment. Both of these designs use the Generalized Random Tessellation Stratified (GRTS) sample selection technique as described in Project 3. The SWAMP has suffered a number of setbacks since OSU's Science to Achieve Results (STAR) proposal was written. Most importantly, funding for SWAMP has been drastically cut because of California's economic difficulties. Consequently, progress towards a consistent statewide monitoring program has been slow. Work with both the State Water Resource Control Board and several regional boards is ongoing.
Another significant outreach effort is the West Coast Tidal Wetland Monitoring and Assessment Venture, which was formed by the U.S. Environmental Protection Agency (EPA) for tidal wetland monitoring and assessment on the West Coast. Participants include EPA Region 9, EPA Western Ecology Division, EPA Atlantic Ecology Division, Oregon Department of Environmental Quality (DEQ), SFEI, and the Southern California Coastal Water Research Project.
Project Investigator Don Stevens has been working with the EPA Western Ecology Division to revise and unify the software used to select Environmental Monitoring and Assessment Program (EMAP)-type samples. Much of this software is based on algorithms developed for EMAP, and has been implemented as ad hoc prototypes. The intent of this effort is to rewrite the algorithms to minimize the reliance on a global positioning system (GPS), to make the code as portable as possible, and to develop a user interface to facilitate use. The goal is to make the sample selection into a tool that can be used by state and tribal agencies.
Survey Design Methodology for Aquatic Resources Project
The objectives of Project 2 are to: (1) extend existing EMAP survey design methodologies to incorporate auxiliary information in survey design; (2) develop design-based methods for analyzing rotating panel surveys; (3) extend methods for incorporating nonprobability-based data in design-based analyses; (4) develop estimation methods for species richness from probability sample data; and (5) make both existing and newly developed survey design and analysis tools accessible to the states and tribes.
The manuscript entitled, "Spatially-Balanced Sampling of Natural Resources in the Presence of Frame Imperfections," defines the mathematical and statistical theory that supports the latest evolution in EMAP's sampling methodology. We also provide a description of some of the spatial properties of sample points selected by the methodology, along with some supporting simulation studies that demonstrate the flexibility and robustness of the design. EMAP has used the sample selection method as described in this manuscript, and previous manuscripts have provided partial descriptions. This represents the first time that the full flexibility of the GRTS design, coupled with Reverse Hierarchical Ordering, has been described in print. This manuscript was submitted to the Journal of the American Statistical Association in August 2002.
The manuscript entitled, "Variance Estimation for Spatially Balanced Samples of Environmental Resources," derives a variance estimator for EMAP-type sampling designs, in particular, the GRTS design. The EMAP Surface Waters program has been using the estimator for about a year. The manuscript defines the statistical foundation for the estimator, in combination with supporting simulation evidence. The estimator developed in this paper is important to EMAP because it captures the increase in precision that results from using the spatially balanced GRTS design. The estimator previously used by EMAP was based on an approximation resulting from assuming independent sampling. Experience suggests that the previous estimator was biased high, usually by a factor of two or more. In contrast, the new estimator is approximately unbiased. The manuscript was accepted in August 2002, for publication in the journal Environmetrics.
We have completed the current investigations into cost models for ranked set sampling and are in the process of submitting an article to Environmetrics. We are preparing the publication of the spatial correlation/habitat association model. These methods were applied to the Oregon stream habitat area data and were found to give lower cost ratio estimates than ones that had been previously calculated (previous estimates used only balanced allocations).
Habitat association models have been useful in issues of stratification and sample allocation for estimating salmon redd abundance. We used a Markov habitat association model for spring Chinook salmon in Idaho's Middle Fork Salmon River. The Monte-Carlo Markov Chain (MCMC) technique provided estimates for the sampling distribution of statistics, which provided guidance for sampling decisions.
New research efforts were initiated on an approach to design-based trend assessment applying multi-phase regression to the coastal salmon data. The approach draws on the multi-panel structure of the sampling design that was developed and presented to ODFW personnel in early February. Secondly, we have extracted a data set from the EMAP MAHA/MAIA study. The data set includes selected response metrics for fish, benthos, and chemistry, along with landscape metrics. The data set will be used to develop small area estimation techniques. The approach models spatial covariance at two points on a stream as a function of the direct drainage area into the two points.
Parametric Model Assisted Survey Methods for Environmental Surveys Project
The objectives of Project 3 are to: (1) apply empirical orthogonal functions to data collected from probability sampling designs and to incorporate auxiliary information for improving estimation and prediction of survey estimates; (2) develop methods to correct for classification errors, which is a key problem associated with remote sensing; and (3) develop procedures to adjust for both item and unit nonresponse (missing data) expected in environmental survey data.
We hired two post-doctoral research fellows. Both post-doctoral associates were teaching until August 2002. Development for the design-based empirical orthogonal functions (EOF) was presented in a Ph.D. dissertation by Mu?oz-Hernandez in 1999. Our current research applies the EOF methods to data collected from a probability sampling design. These methods are being applied to the EMAP data collected from the lakes in the Northeastern United States. Classical methods for nonresponse assume that missing data are missing at random. The feasibility of using classical imputation methods to account for missing data in environmental studies is being investigated with the Oregon Department of Fish and Wildlife Coastal Coho Salmon study.
Nonparametric Model Assisted Survey Estimation for Aquatic Resources Project
The objectives of Project 4 are to: (1) develop nonparametric model-assisted estimators for data obtained in probability surveys of aquatic resources; and (2) apply these estimators in problems of distribution function estimation, small area estimation, longitudinal surveys, and nonparametric adjustment.
We have extended the scope of nonparametric model-assisted survey regression estimation methods in a number of directions. Semiparametric additive regression estimation extends the work of Breidt and Opsomer (Annals of Statistics, 2000), so that the approach can accommodate models with multiple auxiliary variables of both continuous and categorical types. Because many natural resource surveys come with auxiliary information in the form of categories or classifications with continuous variables, it is important for a practical regression estimation method to handle both types of variables. In the semiparametric additive model, some variables (including all categorical variables) enter the regression model as parametric terms, and other variables are incorporated as smooth terms. The latter terms are typically univariate and enter the model additively. During the last year, we worked on defining the estimator and deriving its design-based properties, including calibration, invariance, design consistency, and asymptotic distribution. In addition, we have shown that nonparametric model-assisted estimators are competitive with parametric model-assisted estimators that have been included in the literature when the parametric model specification is correct, but the nonparametric estimators dominate the parametric estimators when the parametric model specification is incorrect. The practical implication of this work is the introduction of simple but efficient distribution function estimators, which incorporate auxiliary information without the need for a careful model specification.
Future Activities:
We will continue research on a number of topics related to nonparametric model-assisted inference, such as: (1) semiparametric additive regression estimation, which can accommodate models with multiple auxiliary variables of both continuous and categorical types; (2) estimating distribution functions from survey data using nonparametric regression; and (3) nonparametric regression estimation for two-stage spatial sampling.
For Project 1, we will: (1) complete the SFEI Regional Monitoring Program Re-Design Report; (2) assist one or more California Regional Water Resource Control Boards on survey design; (3) develop training tools, and software handbooks and manuals; and (4) contact several tribal agencies.
For Project 2, we will continue work with the ODFW; in particular, we will address the imputation of missing data. The anticipated approach is to construct models that draw on both spatial and temporal covariance, as well as ancillary information, such as land use, land cover, ownership, and stream topography. We also will continue to work on trend estimation using the ODFW data set, while considering whether to incorporate some ancillary information (e.g., information on ocean conditions). We will continue to determine species richness and the amount of wetland area that has been lost to infer the number of species that have been lost. In addition, we will continue work on model-based design techniques.
For Project 3, we intend to examine the incorporation of auxiliary data, including GIS data, into the estimation methods for the design-based empirical orthogonal function. Initial work has begun, assuming that the nonresponse is not missing at random and is nonignorable. This requires some development of model-based and numerical methods that we are investigating. We also will investigate the use of kriging with multiple imputation as a method to account for item and unit nonresponse.
Journal Articles: 16 Displayed | Download in RIS Format
Other center views: | All 142 publications | 19 publications in selected types | All 16 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Andrews B, Davis RA, Breidt FJ. Maximum likelihood estimation for all-pass time series models. Journal of Multivariate Analysis 2006;97(7):1638-1659. |
R829096 (2003) R829095 (Final) R829095C002 (2003) R829095C002 (2004) |
Exit Exit |
|
Breidt FJ, Hsu N-J. Best mean square prediction for moving averages. Statistica Sinica 2005;15(2):427-446. |
R829096 (2003) R829096 (2005) R829095 (Final) R829095C002 (2003) R829095C002 (2004) R829095C002 (2005) |
Exit Exit |
|
Breidt FJ, Hsu N-J, Coar W. A diagnostic test for autocorrelation in increment-averaged data with application to soil sampling. Environmental and Ecological Statistics 2008;15(1):15-25. |
R829096 (2005) |
Exit |
|
Buchanan RA, Conquest LL, Courbois J-Y. A cost analysis of ranked set sampling to estimate a population mean. Environmetrics 2005;16(3):235-256. |
R829096 (2002) R829096 (2003) R829096 (2004) R829096 (2005) |
Exit |
|
Cooper C. Sampling and variance estimation on continuous domains. Environmetrics 2006;17(6):539-553. |
R829096 (2005) |
Exit |
|
Courbois JP, Urquhart NS. Comparison of survey estimates of the finite population variance. Journal of Agricultural, Biological, and Environmental Statistics 2004;9(2):236-251. |
R829096 (2003) R829096 (2004) R829096 (2005) R829095 (2004) R829095 (2005) R829095 (Final) R829095C003 (2003) R829095C003 (2004) |
Exit |
|
Da Silva DN, Opsomer JD. Properties of the weighting cell estimator under a nonparametric response mechanism. Survey Methodology 2004;30(1):45-55. |
R829096 (2004) R829096 (2005) R829095 (2003) R829095 (2004) R829095 (2005) R829095 (Final) R829095C002 (2004) R829095C002 (2005) |
Exit Exit |
|
Montanari GE, Ranalli MG. Nonparametric model calibration estimation in survey sampling. Journal of the American Statistical Association 2005;100(472):1429-1442. |
R829096 (2004) R829096 (2005) R829095 (Final) R829095C002 (2004) R829095C002 (2005) |
Exit |
|
Munoz B, Lesser VM. Adjustment procedures to account for non-ignorable missing data in environmental surveys. Environmetrics 2006;17(6):653-662. |
R829096 (2005) |
Exit |
|
Munoz B, Lesser VM, Ramsey FL. Design-based empirical orthogonal function model for environmental monitoring data analysis. Environmetrics 2008;19(8):805-817. |
R829096 (2003) R829096 (2004) R829096 (2005) |
Exit |
|
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. |
R829096 (2004) R829096 (2005) R829095 (2004) R829095 (2005) R829095 (Final) R829095C002 (2004) |
Exit Exit |
|
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. |
R829096 (2003) R829096 (2004) R829096 (2005) R829095 (2004) R829095 (2005) R829095 (Final) R829095C002 (2004) R829095C002 (2005) |
Exit Exit |
|
Opsomer JD, Claeskens G, Ranalli MG, Kauermann G, Breidt FJ. Non-parametric small area estimation using penalized spline regression. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 2008;70(1):265-286. |
R829096 (2005) R829095C002 (2005) |
Exit Exit |
|
Stevens Jr. DL, Olsen AR. Variance estimation for spatially balanced samples of environmental resources. Environmetrics 2003;14(6):593-610. |
R829096 (2003) R829096 (2005) |
Exit |
|
Stevens Jr. DL, Olsen AR. Spatially-balanced sampling of natural resources. Journal of the American Statistical Association 2004;99(465):262-278 |
R829096 (2002) R829096 (2004) |
not available |
|
Thomas DL, Johnson D, Griffith B. A Bayesian random effects discrete-choice model for resource selection: population-level selection inference. Journal of Wildlife Management 2006;70(2):404-412. |
R829096 (2005) R829095 (Final) |
Exit |
Supplemental Keywords:
public policy, decisionmaking, community-based, monitoring, risk assessment, watersheds, streams, rivers, estuaries., RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Aquatic Ecosystem, Environmental Monitoring, EMAP, estuarine research, risk assessment, ecosystem monitoring, statistical survey design, spatial and temporal modeling, aquatic ecosystems, Environmental Monitoring and Assessment ProgramProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.