Assessing land surface phenology of the savanna ecosystem in Southeast Asia using Moderate Resolution Imaging Spectroradiometer (MODIS) leaf area index from 2002 to 2020

The sustainability of the global savanna ecosystem is currently under threat from climate and anthropological change. Despite the immense threats, the existence of the savanna ecosystem is undervalued and understudied. This study examined the dynamics of the savanna ecosystem in the southern part of Southeast Asia (SEA) using MODIS leaf area index (LAI) data (MOD15A3H) with 500-m spatial resolution and 4-day data, from 2002 to 2020. The annual phenological metrics comprising the start of season (SOS), end of season (EOS), length of season (LOS), rate of greening, rate of browning, and the maximum peak values were derived from the daily interpolated data using the spline function. Additional oscillation and trend analysis using empirical ensemble decomposition methods (EEMD) was conducted to derive the nonlinear trend dynamics of the savanna ecosystem in East Nusa Tenggara (ENT). We found that the SOS at the Savanna ecosystem in SEA is 253.76 ± 2.1 days, the EOS is 161.12 ± 4.0 days, and the average LOS is 170.68 ± 6.5 days. The rainfall variabilities can explain around 35 of the variability in the LAI of the savanna ecosystem. Our EEMD analysis captured the decreasing LAI trend, showing a net change between 2002 and 2015 from 1.08 LAI units (scale of 10−2) to − 0.17 LAI units (scale of 10−2) from 2015 onwards. The result indicated a declining trend of LAI values of savanna ecosystem in ENT, thus requiring further monitoring to ensure the sustainability of this ecosystem. © 2021, Società Italiana di Fotogrammetria e Topografia (SIFET).