Remote Sensing-Derived Drivers of Vegetation Persistence in East Africa

Keywords: remote sensing, vegetation, land cover, persistence, East Africa
Session Type: Virtual Poster Abstract
Day: Friday
Session Start / End Time: 2/25/2022 03:40 PM (Eastern Time (US & Canada)) - 2/25/2022 05:00 PM (Eastern Time (US & Canada))
Room: Virtual 39

Abstract

Interest in human impact on the natural environment has increased dramatically over the last quarter century. Over the same period, repeat digital, synoptic measures of the Earth‘s surface have stimulated many research questions designed to improve understanding of human-environment interactions. Satellite remote sensing is ideal for this task, providing consistent, repeatable measurements across a suite of spatial scales, and is well-suited to capture processes of land surface change. Detection and characterization of such changes is often the first step in understanding the mechanisms and identifying their drivers.

Directional Persistence (D) is a measure of the cumulative direction of change over the time series relative to the first observation. Cumulative frequency can be used to determine critical values of D, corresponding to the upper and lower desired significance level. The smallest value of D which equals, or is less than the generally used levels of significance can be extracted from the simulated frequencies of directional persistence generated under the conditions of the null hypothesis: values of the indices, such as NDVI, are independently and normally distributed. The greater the absolute value of the preceding NDVI, the less likely the series is to continue in that direction (increasing/decreasing). These approaches for examining vegetation change are conducted in a spatially explicit manner and will further understandings of the impact of climate variability, and policy and institutional change on the landscape by supporting the identification of sites within the study area experiencing critical changes in vegetation now, and back across the preceding 20 years.