TY - JOUR
T1 - Geospatial modelling of drought patterns in Oman
T2 - GIS-based and machine learning approach
AU - Mansour, Shawky
N1 - Publisher Copyright:
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.
PY - 2024/3/4
Y1 - 2024/3/4
N2 - Drought is one of the most devastating natural disasters, and its consequences affect various human and environmental aspects. In both arid and semi-arid regions, the impact of drought poses direct threats to livelihoods and socio-economic activities. For drought mitigation purposes, spatially accurate predictions of the areas to be affected are essential. By utilising an Artificial Neural Network (ANN) within a Geographic Information Systems (GIS) environment, this research aimed to project drought severity across Oman throughout the twenty-first century. Drought severity during the rainy season (DJF) was characterised using the Standardized Precipitation Evapotranspiration Index (SPEI) calculated for February at a three-month timescale. SPEI was computed based on the monthly data for a set of climatic variables (i.e. maximum and minimum air temperatures, total precipitation, wind speed, relative humidity) derived from the climate forecast system reanalysis (CFSR) dataset at a grid interval of 0.25° for the period between 1998 and 2012. The ANN model was forced with drought classes (i.e. mild, moderate, severe, extreme, and very extreme) employed as a dependent variable, while a wide spectrum of climatic (e.g., air temperature, precipitation, wind speed), topographical (e.g., elevation, aspect) and geographical (e.g., distance to coasts, vegetation cover) variables were used as independent variables. For consistency in projecting drought changes, the dependent and independent variables were re-gridded to a common grid interval (0.25 °C) using a spline interpolation algorithm. Our findings show that the ANN model provided a realistic simulation of drought occurrence incorporating the relevant climatic, topographical and geographic parameters across Oman. Regarding the projected spatial patterns of drought, the northern parts of the study area (e.g., North and South Al-Batinah governorates) are exposed to the severe and extreme intensification of drought, whilst predominately medium and low levels of droughts are expected to occur across the south and south-west areas of Oman. In a water-scarce region like Oman, the results of this study could have particular policy implications, specifically in terms of management of water resources, food production, agriculture, water supply, hydropower energy and biodiversity, amongst others. The projected changes in drought occurrence in Oman make it necessary to develop effective national initiatives to mitigate the impacts of drought and to build society's capacity for drought preparedness. Graphical abstract: (Figure presented.)
AB - Drought is one of the most devastating natural disasters, and its consequences affect various human and environmental aspects. In both arid and semi-arid regions, the impact of drought poses direct threats to livelihoods and socio-economic activities. For drought mitigation purposes, spatially accurate predictions of the areas to be affected are essential. By utilising an Artificial Neural Network (ANN) within a Geographic Information Systems (GIS) environment, this research aimed to project drought severity across Oman throughout the twenty-first century. Drought severity during the rainy season (DJF) was characterised using the Standardized Precipitation Evapotranspiration Index (SPEI) calculated for February at a three-month timescale. SPEI was computed based on the monthly data for a set of climatic variables (i.e. maximum and minimum air temperatures, total precipitation, wind speed, relative humidity) derived from the climate forecast system reanalysis (CFSR) dataset at a grid interval of 0.25° for the period between 1998 and 2012. The ANN model was forced with drought classes (i.e. mild, moderate, severe, extreme, and very extreme) employed as a dependent variable, while a wide spectrum of climatic (e.g., air temperature, precipitation, wind speed), topographical (e.g., elevation, aspect) and geographical (e.g., distance to coasts, vegetation cover) variables were used as independent variables. For consistency in projecting drought changes, the dependent and independent variables were re-gridded to a common grid interval (0.25 °C) using a spline interpolation algorithm. Our findings show that the ANN model provided a realistic simulation of drought occurrence incorporating the relevant climatic, topographical and geographic parameters across Oman. Regarding the projected spatial patterns of drought, the northern parts of the study area (e.g., North and South Al-Batinah governorates) are exposed to the severe and extreme intensification of drought, whilst predominately medium and low levels of droughts are expected to occur across the south and south-west areas of Oman. In a water-scarce region like Oman, the results of this study could have particular policy implications, specifically in terms of management of water resources, food production, agriculture, water supply, hydropower energy and biodiversity, amongst others. The projected changes in drought occurrence in Oman make it necessary to develop effective national initiatives to mitigate the impacts of drought and to build society's capacity for drought preparedness. Graphical abstract: (Figure presented.)
KW - Artificial neural network
KW - Drought risk patterns
KW - Forecasting
KW - GIS
KW - Oman
UR - http://www.scopus.com/inward/record.url?scp=85186553464&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85186553464&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/a46bc9b2-f13f-36a7-bd0c-2702d07d103f/
U2 - 10.1007/s40808-024-01958-9
DO - 10.1007/s40808-024-01958-9
M3 - Article
AN - SCOPUS:85186553464
SN - 2363-6203
JO - Modeling Earth Systems and Environment
JF - Modeling Earth Systems and Environment
ER -