Spatial patterns of vegetation composition across a gradient of human resource use intensity in semi-arid African savannas

Authors: Kyle Woodward*, University of North Carolina - Wilmington, Narcisa Pricope, University of North Carolina - Wilmington, Andrea Gaughan, University of Louisville, Forrest Stevens, University of Louisville, Nicholas Kolarik, University of Louisville, Michael Drake, University of Colorado - Boulder
Topics: Remote Sensing, Human-Environment Geography, Africa
Keywords: remote sensing, land degradation, resource use
Session Type: Poster
Day: 4/4/2019
Start / End Time: 9:55 AM / 11:35 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded

Land degradation is a global threat to dryland environments and the human populations that depend on them. In Africa, increasing climatic variability and human land use create complex ecological responses that diminish an ecosystem’s ability to provide natural resources. Remote sensing technologies are leveraged to disentangle human and environmental drivers of land degradation using field-based biophysical, socioeconomic, and land use data. Natural resource use is an important element of land use practices; the degree of a community’s natural resource dependence often indicates their degree of economic and environmental vulnerability. However, resource use is rarely included in remotely sensed assessments of landscape degradation because it is difficult to quantify resource use in situ, and resource use information is not typically known for spatially explicit study areas. We present an experimental multi-scale, interdisciplinary approach to detect impactful resource use activities along a gradient of known resource use intensity in semi-arid African savannas. We will use Unmanned Aerial Systems (UAS) imagery, high- and medium-resolution satellite imagery, field-collected resource area (RA) spatial data, and household-level socioeconomic data collected in three community-based organizations in Botswana, Namibia, and Zambia during 2016-2018 dry seasons. We will use UAS metrics to train an object-based vegetation classification on PlanetScope imagery, then scale up to a Landsat vegetation fractional cover product. In a GIS, we will use a digital transect approach to compute geospatial statistics of vegetation composition across each RA’s boundary to determine any significant difference between vegetation communities inside and outside each RA.

Abstract Information

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

To access contact information login