Publications

Measho, S; Li, FD; Chen, G; Hirwa, H (2023). Characterizing Cropland Patterns Across North-East Africa Using Time Series Vegetation Indices. JOURNAL OF GEOPHYSICAL RESEARCH-BIOGEOSCIENCES, 128(3), e2022JG007075.

Abstract
The global community faces challenges of agroecosystem sustainability in the face of changing climate and intensified agriculture, particularly in Africa. In this study, annual and seasonal cropland Normalized Vegetation Difference Index (NDVI) trends and phonological variations, and associated key drivers in North-East Africa were generated using Moderate Resolution Imaging Spectral-radiometer NDVI of 8-day satellite images and climate data between 2001 and 2020, which were further analyzed by a modified dynamic threshold method. The results showed that the annual cropland NDVI increased at an average rate of 0.0106 per year. The seasonal cropland NDVI was positive for the most significant changes observed in the boreal autumn season across the region. The start of the season and end of the season cropland NDVI in the boreal autumn season increased at an average rate of 0.015 (13.4% significant, p < 0.05), and 0.0151 x (12.8% significant, p < 0.05), respectively. The findings indicated that the boreal autumn season was getting longer, whereas the boreal summer season was becoming shorter during the last two decades. The most rapid cropland NDVI changes and maximum farmland NDVI variability were observed in the sub-tropic warm arid and tropic warm semiarid agro-ecological zones. The land surface temperature was found to be the main driving factor for the cropland NDVI and associated agroecosystem changes. This study provides essential information on the cropland productivity and shifting of greening patterns, which will support climate change mitigation measures at national and regional scales, contributing to sustainable agriculture.Plain Language Summary Climate change and land use change impose various impacts on agricultural productivity and agroecosystem services in Africa. In this research, we analyzed annual and seasonal cropland changes using time series Normalized Vegetation Difference Index (NDVI) as monthly image composites from 2001 to 2020. We assessed key transition stages in the vegetation and crop phenology trends including annual cropland NDVI, the start of the growing season (SOS), and the end of the growing season (EOS). At an annual time-step, there were more greening patterns across large parts of North-East Africa with the highest greening trends in the croplands of Egypt followed by Uganda and South Sudan, while more browning patterns were in Djibouti, Eritrea, and Somalia. The most greening and significant SOS trends were shown in the boreal autumn season, while negative EOS trends were common in the boreal summer season. The boreal summer was getting shorter as the boreal autumn experienced longer periods. The sub-tropic warm arid and tropic warm semiarid agro-ecological zones showed the highest farmland productivity changes. The cropland NDVI changes was strongly associated with land surface temperature. The study provides important phonological information, which is helpful to understand the challenges of food security at a regional scale.

DOI:
10.1029/2022JG007075

ISSN:
2169-8961