Ethiopia is seen as one of the African countries most vulnerable to the impacts of climate change, with limited capacity to cope with short-term climatic shocks or adapt to longer-term trends (Conway et al., 2007). Due to a significant dependence on the agricultural sector for production, export revenues, and employment, Ethiopia is seriously threatened by climate change, which contributes to frequent flooding, drought, and rising average temperatures (Asaminew, 2013). 

Drought is a common occurrence that can last for several years. It is admittedly an environmental hazard that attracted the attention of many environmentalists, ecologists, meteorologists, hydrologists, and agricultural scientists. Drought is the most complex of all natural hazards making life unbearable to humans, animals, and plants. In Ethiopia, the most severe drought in recent history occurred in 2015-2016, affecting over 10 million people and resulting in widespread food insecurity, malnutrition, and displacement. Despite the government's efforts to mitigate the impacts of drought, it remains a significant challenge to Ethiopian development. The shortage of rainfall for two consecutive rainy seasons called Kiremt and Belg that normally feed 80-85% population of the country, has led to a devastating drought and greatly increased malnutrition rates across the country (Nobre, 2019). With this research, the author hopes to contribute to achieving SDG 13.1 and creating a more resilient and sustainable future in the face of drought. 

In Ethiopia, various studies have been carried out on drought risk assessment by using the Standardized Precipitation Index (SPI), the Crop Water Requirement Satisfaction Index (CWRSI), and the Livelihoods Resilience Index (LRI). These indexes use different indicators, such as rainfall, temperature, crop yield, and socioeconomic factors, to assess drought vulnerability and identify appropriate interventions. However, due to the direct relation of parameters with climate change, this study used NDVI (Normalized difference vegetation index) and LST (Land surface temperature) to provide an early warning system and information on climate change induced drought risk of Meyo district in Borena, Ethiopia.  


Optical and radar satellite data sources used 

To analyze drought severity, Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) datasets from the United States Geological Survey (USGS) with a spatial resolution of 30 m were acquired for the month of December of the years 2002, 2012, and 2022. Additionally, rainfall data for same period was downloaded from Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS). The CHIRPS dataset comprises gridded time-series data of rainfall with a geographic resolution of 0.05° (approximately equivalent to 5.3 kilometers), which is utilized for trend analysis and monitoring seasonal drought. The USGS and CHIRPS datasets thereby supports effective water resource planning and disaster preparedness. 

Dana analysis  

The flowchart outlining the methodology utilized for data processing and analysis is depicted in Figure 1. Image processing techniques were applied to compute the Normalized Difference Vegetation Index (NDVI) value of the study area in ArcGIS environment 10.8. 

 Figure 1. Methodological flowchart  


Normalized Difference Vegetation Index (NDVI)

The NDVI value acts as a measure of vegetation health, where higher values indicate robust vegetation, while lower values signal stress or poor conditions due to inadequate moisture in a specific region (Faridatul and Ahmed, 2020). It is considered as popular index to analyze agricultural drought and its value is always between -1 and +1 (Ganie and Nusrath, 2016). The basic concept underlying the NDVI is that green vegetation reflects Near-Infrared (NIR) while simultaneously absorbing a significant portion of red visible radiation (VIS) within the electromagnetic spectrum. Thus,  





Land Surface Temperature (LST) 

Land surface temperature (LST) is a satellite-observed variable that can exhibit rapid response to drought conditions as it coincide with soil moisture and vegetation temperature (Anderson et al., 2013). Vegetation temperature can provide valuable insights into root-zone soil moisture. This is because vegetation temperature is closely linked to water absorption by plants from the root-zone, as explained by the energy balance of the land surface (Choudhury and Idso, 1984; Wetzel and Woodward, 1987). Consequently, it serves as a means to detect drought by monitoring the impact of high temperatures and moisture scarcity on vegetation health as high LST values typically indicate severe aridity resulting from insufficient rainfall (Li et al., 2021; Bhuiyan et al., 2017).  

Weighted Overlay 

Considering the influence of each parameter on the severity of the drought, matrices for pairwise comparisons have been devised. After determining the relative ranking of each parameter, the weights were assigned to the parameters. . Finally, following the preparation of all thematic layers (including NDVI and rainfall), the layers were uniformly reclassified and subsequently merged using a weighted overlay approach (Fig. 1).  

Case study of Meyo District Borena 

The Meyo district is found in Borena Zone Ethiopia, geographically located between 3°33' and 4°9' north latitude and 38°20' and 39°0' east longitude, covering an area of approximately 3027 km2. This district has a diverse topography, which varies between 682 and 1841 metreser above sea level (Fig. 2). 

Figure 2. Study area map 


Climatic variability and its impact on vegetation cover 

The results allow to compare accumulated December rainfall over the years (Fig. 3) and indicate that in December 2002, rainfall values ranged from 50.09 mm to 110.03 mm, while in December 2012, values were between 9.5 mm and 29 mm. By December 2022, the range was 7.18 mm to 18.88 mm, pointing to notable rainfall variability within the study area. The outcome further reveals deficient rainfall in the Northern and Eastern sections of the study area compared to other parts (Fig. 3). 

Figure 3.  Accumulated December rainfall over the years 


Utilizing ArcGIS 10.8, NDVI was computed for December in 2002, 2012, and 2022, employing Landsat 7 ETM+ and Landsat 8 OLI/TIRS, with temporal analysis evaluating the effects of drought-induced change. NDVI images depicted areas affected by drought in brown, indicating poor health, while green represented thriving, moist vegetation zones (Fig. 4). 

Figure 4. Change in NDVI index over the years 


In the analysis of the three years, 2002 yielded the highest NDVI value, and the lowest LST value (Fig. 4 and 5). The NDVI maps reveal that the NDVI values are declining over the years which can be attributed to changing climate. The peak value was 0.537572 in 2002, followed by 0.624331 in 2012, and then decreased to 0.554326 in 2022, which suggests decreasing plant health due to lower NDVI values. The lower NDVI value rose from -0.593583 in 2002 to -0.182837 in 2022, indicating high drought vulnerability due to reduced rainfall.  

Figure 5. LST result from 2002 to 2022 for the month December 


The study’s results reveal an inverse relation of LST and NDVIi.e., heightened surface temperatures corresponded with lowered NDVI values, indicating elevated vegetation stress due to diminished water availability for transpiration. In agreement with this result, different studies have reported the inverse relationship between NDVI and LST (Tran et al., 2017; Han et al., 2020; Shashikant et al., 2021). This correlation might be attributed to the presence of high moisture in vegetation during the wet season and can be changed with land cover types (Guha and Govil, 2020). 

During arid periods, rising leaf temperatures act as reliable markers of plant moisture stress, preceding the onset of drought. Correlation plots were constructed using December's NDVI and LST values for each year (2002, 2012, and 2022). Across the study period, the inverse correlation between LST and NDVI persisted (Fig. 4 and 5), signifying heightened vegetation stress due to decreased water available for transpiration. These findings are in line with the study performed by Cunha et al. (2015), which identifies this behavior as prevalent in drought environments. 

Drought risk map 

Through a comparative analysis of drought on temporal basis, it became evident that the drought is more severe during 2012-2022 than in 2002-2012 (Fig. 6). Moreover, considering precipitation indicators, the hydrological year 2002-2012 displayed more pronounced characteristics of extreme dry conditions compared to the hydrological year 2012-2022. 

The provided drought maps illustrate the severity of drought risk within the study area, depicting three classes of risk (Low, Moderate, and High). These maps clearly indicate a scenario of drought prevalence and its trend in the study area. In 2022, the extent of drought risk was higher than in 2002.2. Additionally, the results highlight variations in vegetation conditions during the study period, showcasing fluctuations in drought magnitude across both time and space (Fig. 4 and 6).  

Figure 6. Spatial and temporal distribution of drought 



Currently, Ethiopia is grappling with severe drought conditions, leading to widespread suffering among both the population and livestock. The adverse impact of this drought is taking a toll on the well-being and livelihoods of people and animals alike. Urgent measures are required to address this critical situation and provide essential relief to the affected communities. Thus, this study is aimed to identify and classify drought-prone areas within the study region, differentiating between low, moderate, and high drought-risk zones using space technology. The drought areas were mapped using weighted overlay analysis in GIS environment. The findings hold significance for decision-makers, policymakers, and stakeholders involved in devising effective drought management strategies not only in the study area but also in other drought-prone regions in Ethiopia and beyond, thereby contributing to the achievement of SDG 13.13. The high drought vulnerability zones identified can be prioritized for targeted interventions, including water resource management, agricultural practices, and livelihood support. This focused approach in vulnerable areas can enhance resilience and minimize the adverse impacts of drought on local communities. 


​​Anderson, Martha C., Christopher Hain, Jason Otkin, Xiwu Zhan, Kingtse Mo, Mark Svoboda, Brian Wardlow, and Agustin Pimstein. "An intercomparison of drought indicators based on thermal remote sensing and NLDAS-2 simulations with US Drought Monitor classifications." Journal of Hydrometeorology 14, no. 4 (2013): 1035-1056. 

​Asaminew, T. G., A. Araya, G. Atkilt, and H. Solomon. "Modeling the potential impact of climate change on cotton (Gossypium hirsutum) production in Northeastern Semi-Arid Afar and Western Tigray Regions of Ethiopia." J Earth Sci Clim Change 8, no. 3 (2017): 1-7. 

​Bhuiyan, C., A. K. Saha, N. Bandyopadhyay, and F. N. Kogan. "Analyzing the impact of thermal stress on vegetation health and agricultural drought–a case study from Gujarat, India." GIScience & Remote Sensing 54, no. 5 (2017): 678-699. 

​Choudhury, Bhaskar J., and Sherwood B. Idso. "Simulating sunflower canopy temperatures to infer root-zone soil water potential." Agricultural and Forest Meteorology 31, no. 1 (1984): 69-78. 

​Conway, D., L. Schipper, M. Yesuf, Minale Kassie, A. Persechino, and Bereket Kebede. "Reducing vulnerability in Ethiopia: Addressing the implications of climate change: Integration of results from Phase I." (2007). 

​Cunha, A. P. M., Regina C. Alvalá, Carlos A. Nobre, and Magog A. Carvalho. "Monitoring vegetative drought dynamics in the Brazilian semiarid region." Agricultural and forest meteorology 214 (2015): 494-505. 

​Faridatul, Mst Ilme, and Bayes Ahmed. "Assessing agricultural vulnerability to drought in a heterogeneous environment: a remote sensing-based approach." Remote Sensing 12, no. 20 (2020): 3363. 

​Ganie, Mushtaq Ahmad, and Asima Nusrath. "Determining the vegetation indices (NDVI) from Landsat 8 satellite data." International Journal of Advanced Research 4, no. 8 (2016): 1459-1463. 

​Guha, Subhanil, and Himanshu Govil. "Land surface temperature and normalized difference vegetation index relationship: a seasonal study on a tropical city." SN Applied Sciences 2, no. 10 (2020): 1661. 

​Han, Yang, Ziying Li, Chang Huang, Yuyu Zhou, Shengwei Zong, Tianyi Hao, Haofang Niu, and Haiyan Yao. "Monitoring droughts in the Greater Changbai Mountains using multiple remote sensing-based drought indices." Remote Sensing 12, no. 3 (2020): 530. 

​Li, Bailing, Matthew Rodell, Benjamin F. Zaitchik, Rolf H. Reichle, Randal D. Koster, and Tonie M. van Dam. "Assimilation of GRACE terrestrial water storage into a land surface model: Evaluation and potential value for drought monitoring in western and central Europe." Journal of Hydrology 446 (2012): 103-115. 

​Nobre, Gabriela Guimarães, Frank Davenport, Konstantinos Bischiniotis, Ted Veldkamp, Brenden Jongman, Christopher C. Funk, Gregory Husak, Philip J. Ward, and Jeroen CJH Aerts. "Financing agricultural drought risk through ex-ante cash transfers." Science of the Total Environment 653 (2019): 523-535. 

​Shashikant, Veena, Abdul Rashid Mohamed Shariff, Aimrun Wayayok, Md Rowshon Kamal, Yang Ping Lee, and Wataru Takeuchi. "Utilizing TVDI and NDWI to classify severity of agricultural drought in Chuping, Malaysia." Agronomy 11, no. 6 (2021): 1243.  

​Tran, Hoa Thi, James B. Campbell, Tri Dinh Tran, and Ha Thanh Tran. "Monitoring drought vulnerability using multispectral indices observed from sequential remote sensing (Case Study: Tuy Phong, Binh Thuan, Vietnam)." GIScience & Remote Sensing 54, no. 2 (2017): 167-184. 

​Wetzel, Peter J., and Robert H. Woodward. "Soil moisture estimation using GOES-VISSR infrared data: A case study with a simple statistical method." Journal of Applied Meteorology and Climatology 26, no. 1 (1987): 107-117.​