|
 Statistics Department |
||||||
| Home
|
|
This Program, funded by The United States Environmental Protection Agency (USEPA), develops survey design and analysis methodology appropriate for large-scale aquatic monitoring programs. The program funds faculty research and several research assistantships. Some current research areas include: Small area estimation applied to surveys of lakes, streams, and estuaries. Estimates are often needed for geographical areas that have insufficient data to produce pure design-based estimates. This research area will examine ways of using ancillary information and spatial modeling to obtain estimates for small areas. Estimation of status and trend in condition using complex surveys over time. Rotating panels in conjunction with spatially balanced annual samples are being used in Oregon to monitor coho salmon populations and in California to monitor the San Francisco Estuary. Research in this area will examine the use of composite estimators and multi-phase regression to estimate both status and trends. Development of new survey design methods to assess ecological condition that incorporate remote sensing and other sources of auxiliary information. Often, auxiliary information relevant to survey objectives is available at the time a survey is designed. The research in this area will investigate ways of using such information to design more efficient surveys. Development of procedures to account for missing data in probability surveys due to denial of access by landowners and physically inaccessible locations. State-wide and multi-state environmental surveys can have severe missing data problems, often because much of the resource being surveyed is on private land. Even so, landscape-level, remotely-sensed data is often available. The research in this area is aimed at developing imputation procedures using such multi-tiered data. Developing procedures for combining probability survey data with non-probability survey data to estimate population characteristics.Historically, many aquatic monitoring programs were based on ad hoc or convenience-based site-selection procedures. The research focus in this area is to develop useful guidelines for determining the utility and practicality of combining probability and non-probability data and to investigate ways of extending existing techniques. Estimating total number of species within a region based on probability survey data. Species richness is frequently used as a metric to evaluate environmental condition. Methods to estimate species richness from a complex probability survey are not available. Research in this area will evaluate the use of Bayesian methods to develop design-based/model-assisted estimators. The research is being conducted in close cooperation with the Oregon Department of Fish and Wildlife, the California State Water Resources Control Board, and the San Francisco Estuary Institute, as well as several Regional offices of the USEPA. These agencies provide both data sets and a testing ground for the statistical methodology. |
|
|
| |
|
 Statistics Department |
Statistics Department |
 |