Drought is a major challenge for sustainable agriculture and water security in the Ethiopian Highlands. However, there is still a lack of thorough assessments on its long-term patterns over time and space. This study aims to fill these existing gaps by using advanced geospatial techniques to analyse drought trends across a span of 24 years (from 2000 to 2023), using remote sensing data, particularly the Normalized Difference Vegetation Index (NDVI). Meanwhile, the Standardized Precipitation Evapotranspiration Index (SPEI) is calculated from observational data sourced from the CRU TS version 4.07 dataset, accessed from CSIC 2.10. By integrating satellite-based NDVI data from MODIS with seasonal SPEI data, this research employs Mann-Kendall trend tests to evaluate long-term trends. Moreover, a Random Forest regression model is applied to gain insights on how these factors impact crop yields. The results show that while NDVI did not display any significant long-term trends seasonally, peak vegetation vigor consistently occurred in the fall, hitting values of 0.58 in 2019. In contrast, the spring SPEI exhibited a significant upward trend (p < 0.01), suggesting that more moisture is becoming available. However, the summer and fall SPEI did not reveal any significant trends. The spatial analysis identified years like 2012 and 2015 affected by “severe” drought impacts, affecting about 287.5 hectares and over 50% of the area in certain years, faced by “mild” and “moderate” droughts. Additionally, correlation and regression analyses indicated a strong relationship between the drought indices and crop yields. The Pearson correlation values ranged from 0.96 to 0.97 for both indices, with R² values over 0.88 for SPEI and over 0.79 for NDVI, suggesting a strong predictive capability. Importantly, SPEI was found to be more effective than NDVI in explaining changes in crop yields. These findings highlight the critical need to combine vegetation and moisture indices for effective drought monitoring, agricultural planning, and policy making. This research contributes to developing targeted water management and resilience strategies to lessen the impacts of drought in the face of climate variability, offering valuable insights for policymakers and stakeholders in the Ethiopian Highlands. Graphical Abstract This study examines the trends and patterns of drought over time and space in the Ethiopian Highlands, a crucial area for agriculture and water resources. By employing geospatial technology, we are able to analyse seasonal and temporal shifts using NDVI and SPEI data. The map of the study area illustrates that the Ethiopian Highlands range in elevation from 1,000 to 4,620 m and experience distinct wet and dry seasons, which have a significant effect on vegetation vigor. Results show that NDVI values peak in the fall, coinciding with the harvest season. Data reveals that the fall of 2019 and 2023 were the wettest, whereas the springs of 2011 and 2015 saw the driest conditions due to insufficient rainfall and delays. The SPEI analysis highlights the variability in moisture availability from year to year and season to season, indicating that rainfall patterns are inconsistent. Records indicate that the summer of 2019 and spring 2023 were relatively wet, while the summers of 2002 and the spring of 2003 faced extreme drought conditions. Moreover, correlation and regression analyses show strong relationships between crop yields and both NDVI and SPEI. The correlation values suggest a solid connection between crops and these indices, with both NDVI and SPEI demonstrating a consistent relationship, despite minor variations. The R² values fall between 0.79 and 0.90, suggesting these indices are quite effective in predicting crop yields. Notably, SPEI shows stronger predictive power than NDVI, underscoring the importance of moisture balance for sustainable agriculture. This comprehensive assessment provides valuable insights for formulating effective water management strategies, and highlights the urgent need for adaptive agricultural practices in response to the ongoing climate challenges in the Ethiopian Highlands.

Mapping Drought Patterns Using Geospatial Techniques, in the Highlands of Ethiopia

期刊论文
发表日期：

2025年8月31日

查看全文

Yonas Tesfay Tela,
Aqil Tariq

摘要

Drought is a major challenge for sustainable agriculture and water security in the Ethiopian Highlands. However, there is still a lack of thorough assessments on its long-term patterns over time and space. This study aims to fill these existing gaps by using advanced geospatial techniques to analyse drought trends across a span of 24 years (from 2000 to 2023), using remote sensing data, particularly the Normalized Difference Vegetation Index (NDVI). Meanwhile, the Standardized Precipitation Evapotranspiration Index (SPEI) is calculated from observational data sourced from the CRU TS version 4.07 dataset, accessed from CSIC 2.10. By integrating satellite-based NDVI data from MODIS with seasonal SPEI data, this research employs Mann-Kendall trend tests to evaluate long-term trends. Moreover, a Random Forest regression model is applied to gain insights on how these factors impact crop yields. The results show that while NDVI did not display any significant long-term trends seasonally, peak vegetation vigor consistently occurred in the fall, hitting values of 0.58 in 2019. In contrast, the spring SPEI exhibited a significant upward trend (p < 0.01), suggesting that more moisture is becoming available. However, the summer and fall SPEI did not reveal any significant trends. The spatial analysis identified years like 2012 and 2015 affected by “severe” drought impacts, affecting about 287.5 hectares and over 50% of the area in certain years, faced by “mild” and “moderate” droughts. Additionally, correlation and regression analyses indicated a strong relationship between the drought indices and crop yields. The Pearson correlation values ranged from 0.96 to 0.97 for both indices, with R² values over 0.88 for SPEI and over 0.79 for NDVI, suggesting a strong predictive capability. Importantly, SPEI was found to be more effective than NDVI in explaining changes in crop yields. These findings highlight the critical need to combine vegetation and moisture indices for effective drought monitoring, agricultural planning, and policy making. This research contributes to developing targeted water management and resilience strategies to lessen the impacts of drought in the face of climate variability, offering valuable insights for policymakers and stakeholders in the Ethiopian Highlands.

Graphical Abstract

This study examines the trends and patterns of drought over time and space in the Ethiopian Highlands, a crucial area for agriculture and water resources. By employing geospatial technology, we are able to analyse seasonal and temporal shifts using NDVI and SPEI data. The map of the study area illustrates that the Ethiopian Highlands range in elevation from 1,000 to 4,620 m and experience distinct wet and dry seasons, which have a significant effect on vegetation vigor. Results show that NDVI values peak in the fall, coinciding with the harvest season. Data reveals that the fall of 2019 and 2023 were the wettest, whereas the springs of 2011 and 2015 saw the driest conditions due to insufficient rainfall and delays. The SPEI analysis highlights the variability in moisture availability from year to year and season to season, indicating that rainfall patterns are inconsistent. Records indicate that the summer of 2019 and spring 2023 were relatively wet, while the summers of 2002 and the spring of 2003 faced extreme drought conditions. Moreover, correlation and regression analyses show strong relationships between crop yields and both NDVI and SPEI. The correlation values suggest a solid connection between crops and these indices, with both NDVI and SPEI demonstrating a consistent relationship, despite minor variations. The R² values fall between 0.79 and 0.90, suggesting these indices are quite effective in predicting crop yields. Notably, SPEI shows stronger predictive power than NDVI, underscoring the importance of moisture balance for sustainable agriculture. This comprehensive assessment provides valuable insights for formulating effective water management strategies, and highlights the urgent need for adaptive agricultural practices in response to the ongoing climate challenges in the Ethiopian Highlands.