Jump to main content.


CREM Logo

Model Report

REVA DATA INTEGRATION METHODS

Last Revision Date: 08/25/2009 View as PDF
General Information Back to Top
Model Abbreviated Name:

Model Extended Name:

REVA DATA INTEGRATION METHODS
Model Overview/Abstract:
The core of the research effort in the Regional Vulnerability Assessment Program (ReVA) is a set of data integration methods ranging from simple overlays to complex multivariate statistics. These methods are described in the EPA publication titled, "Regional Vulnerability Assessment for the Mid-Atlantic Region: Evaluation of Integration Methods and Assessments Results," EPA/600/R-03/082, October 2003. In the near future, these methods will be implemented in a web-based tool accessible through the ReVA website
Keywords:
Model Technical Contact Information:
Betsy Smith
U.S. EPA
Office of Research and Development
National Exposure Research Laboratory
smith.betsy@epa.gov
(919) 541-0620
Model Homepage: http://www.epa.gov/reva/products.htm

User Information Back to Top
Technical Requirements
Computer Hardware
PC
Compatible Operating Systems
Any
Download Information
The report is available at: www.epa.gov/reva/products.htm
Using the Model
Basic Model Inputs
Spatial data for a variety of environmental metrics and indicators for the region under study.
Basic Model Outputs
Various data integration methods to assess the impact of multiple stressors on multiple resources/receptors.

Model Science Back to Top
Summary of Model Structure and Methods
Regional scale ecological processes bound processes at smaller scales.

The core of ReVA is the utilization of a variety of data integration techniques from the very simple to complex multivariate statistical methods. The data integration methods include the following:

1. best and worst quantile
2. simple sum
3. principal component analysis (PCA)
4. state space analysis
5. criticality analysis
6. analytical hierarchy process
7. clustering analysis
8. self-organizing maps
9. stressor-resource overlay
10. change analysis
11. stressor-resource matrix analysis

Mostly statistical in nature. They vary depending on the integration method used. These are described in more detail in the report.

Model Evaluation
Section 3 of the report describes the effect of various data issues on each integration method. The data problems looked at are:

1. discontinuity
2. skewness
3. imbalance
4. auto-correlation


Local Navigation


Jump to main content.