Determine a suitable spatial extent for integrated ecosystem services assessment across the conterminous US with wavelet variance analysis

Authors: Liem Tran*, University of Tennessee at Knoxville
Topics: Spatial Analysis & Modeling, Quantitative Methods, Environmental Science
Keywords: spatial analysis, ecosystem services, wavelet
Session Type: Paper
Day: 4/13/2018
Start / End Time: 1:20 PM / 3:00 PM
Room: Proteus, Sheraton, 8th Floor
Presentation File: No File Uploaded


Metrics representing ecosystem services (ES) are often used in various integrated assessments to evaluate the overall health of the ecosystems at different scales. With the advance of spatial technologies (e.g., remote sensing and geographic information systems), multiple ES metrics are developed and available over large regions. For example, EPA’s EnviroAtlas program has a set of nearly two hundred ES metrics at HUC-12 level for the whole United States. While there have been a number of studies on integration methods, not much attention has been paid to the spatial extent of the study area (e.g., administrative units such as state or county, or natural units like ecoregions or watersheds) to make the integration of multiple ES metrics meaningful. In this presentation, we will discuss the use of wavelet variance analysis to detect a suitable spatial extent for integrated ES assessments at regional scale across the conterminous USA. First we perform scale-based decompositions of the variance for the ES metrics available in the EnviroAtlas’ database. Specifically, we use the maximal-overlap discrete wavelet transform (MODWT) estimator of the wavelet to quantify certain characteristics of both non- stationary and stationary ES metrics. We then compare those scale-based variances to detect the common denominator of variance across all ES metrics. We demonstrate that an integrated ES assessment at a right spatial extent would be more meaningful and valuable.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login