Rainfall and foliar dynamics in tropical southern Africa: Potential impacts of global climatic change on Savanna vegetation

Douglas O. Fuller, Stephen D. Prince

Research output: Contribution to journalArticle

60 Scopus citations

Abstract

Foliar dynamics in tropical southern Africa are examined using meteorological satellite observations (NOAA-AVHRR) collected from 1981- 1990, processed as monthly Normalized Difference Vegetation Index (NDVI) images, and resampled to 7.6 km resolution. Time series of NDVI and raingauge data are presented and analyzed using a variety of statistics. The analysis of time series from individual locations revealed positive correlations between NDVI and rainfall at semiarid locations where rainfall tended to be highly variable; whereas the relationships between these variables was insignificant in more mesic sites where the climate tended to be more predictable. In addition, there appeared to be an annual rainfall threshold of approximately 600 mm beyond which relationships between rainfall and NDVI were insignificant at the monthly time scale. Relationships between rainfall and NDVI were stronger at annual time scale, which suggests that factors other than contemporaneous rainfall account for photosynthetic activity in any given growing season. Using a rainfall surface and NDVI imagery, a large area of 'early' greening behavior is identified, which corresponded approximately to the distribution of mesic, plateau woodlands. These so-called, 'miombo' woodlands may be especially vulnerable if the arrival of spring rainfall were to undergo a positive shift in phase.

Original languageEnglish (US)
Pages (from-to)69-96
Number of pages28
JournalClimatic Change
Volume33
Issue number1
DOIs
StatePublished - Jan 1 1996
Externally publishedYes

ASJC Scopus subject areas

  • Global and Planetary Change
  • Atmospheric Science

Fingerprint Dive into the research topics of 'Rainfall and foliar dynamics in tropical southern Africa: Potential impacts of global climatic change on Savanna vegetation'. Together they form a unique fingerprint.

  • Cite this