Tropical forest backscatter anomaly evident in SeaWinds scatterometer morning overpass data during 2005 drought in Amazonia

Abstract

A severe drought occurred in southwestern Amazonia in the dry season (June–September) of 2005. We analyzed 10 years (7/99–10/09) of SeaWinds active microwave Ku-band backscatter data collected over the Amazon Basin, developing monthly means and anomalies from those means in an effort to detect landscape responses to this drought. We compared these to seasonal accumulating water deficit anomalies generated using Tropical Rainfall Monitoring Mission (TRMM) precipitation data (1999–2009) and 100 mm mo− 1 evapotranspiration demand as a water deficit threshold. There was significant interannual variability in dry-season monthly mean backscatter only for morning (c. 06:00 LST) overpass data, and little interannual variability in dry-season monthly mean backscatter for afternoon (c. 18:00 LST) overpass data. Strong negative anomalies in both morning-overpass backscatter and accumulating water deficit developed during July–October 2005, centered on the southwestern Amazon Basin, with a strong spatial correlation between morning-overpass backscatter anomalies and water deficit anomalies in September. This is the first reporting of tropical forest seasonal drought detection by active microwave scatterometry. Based on the differences between early-morning and late-afternoon backscatter variability, we hypothesize that as the drought persisted over several months, the forest canopy was increasingly unable to recover full leaf moisture content over night, resulting in anomalously low early-morning overpass backscatter.

Department

Earth Sciences, Earth Systems Research Center

Publication Date

3-15-2011

Journal Title

Remote Sensing of Environment

Publisher

Elsevier

Digital Object Identifier (DOI)

10.1016/j.rse.2010.11.017

Document Type

Article

Rights

Copyright © 2016 Elsevier B.V. or its licensors or contributors. ScienceDirect® is a registered trademark of Elsevier B.V.

Share

COinS