Quantifying changes in savanna rangeland grass phenology and biomass due to an El Niño event

Abstract

The effects of El Niño events on rangeland grass phenology and biomass are often presumed and, therefore, seldom quantified. Here, the effects on the grass start of growing season (SOS), end of growing season (EOS), length of growing season (LGS), and aboveground biomass (AGB) are quantified. The SOS, EOS, LGS, and peak phenology dates in a normal and an El Niño season were identified using trends in 16-day Enhanced Vegetation Index (EVI) values derived from the MODIS (MODerate Resolution Imaging Spectrometer) MOD13Q1 product. During the El Niño rainy season, grass AGB samples were collected from dispersed sampling sites soon after the peak phenology stage. Their Soil Adjusted Vegetation Index (SAVI) values, computed from 10 m resolution Sentinel-2 MSI images, were correlated with the AGB. A statistically significant (p < 0.05) linear regression model for predicting grass AGB from SAVI values was developed. Using the model, the non-El Niño season's peak phenology AGB values at the respective sites were predicted from Sentinel-2 MSI image SAVI values. The El Niño event caused erratic seasonal rainfall, a two-month delay to the SOS, a one-month late EOS, and a one month shorter LGS. On average, grass AGB was 59% lower due to the event.