Recommended citation: Sorí, R., Stojanovic, M., Pérez-Alarcón, A., Ernst, J., Samy-Mohamed, M., Mekhiel,A., Nieto, R., Gimeno, L. (2025). Response of soil water content to dry and wet conditions across Northern Africa. International Geographical Union Thematic Conference 2025. Cairo, Egypt. 12-19 April 2025.

Abstact

The northern part of the African (NA) continent encompasses diverse climate types, ranging from warm to cold desert climates, as well as arid and Mediterranean climates. However, common hazards in these regions include insufficient precipitation, intensive water consumption, and consequently, water scarcity. NA is also one of the most water-stressed regions in Africa and globally. Thus, water availability in the soil becomes crucial for various socioeconomic activities. Hence, this study aims to investigate the response of soil water content (swvl) to individual changes in precipitation and precipitation minus atmospheric evaporative demand in NA.

To conduct the study, datasets of soil water content from the European Space Agency (ESA), precipitation and evapotranspiration obtained calculated from ERA5 Land datasets were utilised, along with the volume of water in soil layers (swvl): layer 1 (0–7 cm), layer 2 (7–28 cm), and layer 3 (28–100 cm). According to ERA5-Land data, swvl characterise the soil texture, depth, and groundwater level. The Standardised Precipitation Index (SPI) and the Standardised Precipitation minus Evapotranspiration Index (SPEI) were calculated for temporal scales ranging from 1 to 12 months to assess dry conditions. These indices were used to determine the optimal temporal relationship between dry conditions and swvl across the three soil layers. In addition, a trend analysis of swvl and its attribution based on SPI and SPEI evolution was performed using multiple regression.

The results show that soil water content data from ESA and ERA5-Land are positively and strongly correlated in the first soil layer. For deeper layers, the maximum correlation occurs in the northwest region. SPI and SPEI values are also mostly positive and highly correlated with soil water content, particularly at shorter temporal scales, and especially with SPI in the first soil layer. This analysis identifies the shorter temporal scales of the SPI and SPEI as optimal for soil water content variations. Further analysis revealed a significant decrease in soil water content, particularly in the first and second soil layers in the northwestern region. The attribution analysis showed that, in the first layer, trends are dominated by the SPI evolution, as expected for a water-limited region. Our findings contribute to understanding the historical evolution and natural drivers of swvl variation in a region highly vulnerable to water availability.