Non-stationary drought patterns in hyper-arid regions: Spatiotemporal and multi-timescale drought analysis.

Area of Science:

• Climatology
• Hydrology
• Environmental Science

Background:

• Droughts are severe natural disasters, especially in arid regions like Oman.
• Stationary drought assessment may be inadequate due to climate change.
• Previous studies often used point-based methods, missing spatial drought variability.

Purpose of the Study:

• To evaluate stationary and non-stationary drought trends in Oman using high-resolution data.
• To compare the performance of different drought models at various timescales.
• To analyze spatial and temporal drought dynamics for improved monitoring.

Main Methods:

• Utilized ERA5-Land data (1950-2024) for Standardized Precipitation Evapotranspiration Index (SPEI) at 3-, 6-, and 12-month scales.
• Applied pixel-wise stationary and non-stationary log-logistic SPEI parameter models.
• Selected optimal models using the Akaike Information Criterion (AIC) for detailed drought analysis.

Main Results:

• Non-stationary models outperformed stationary models, especially for long-term droughts (SPEI-12).
• Short-term droughts (SPEI-3) were best described by non-stationary location shifts; SPEI-12 required dynamic mean and variability changes.
• Northern and central Oman showed increasing wetness trends, while eastern coastal areas trended towards dryness.

Conclusions:

• Pixel-wise non-stationary modeling provides a superior approach for drought assessment in arid regions.
• Findings highlight the need for dynamic drought monitoring under climate change.
• Results offer crucial insights for developing climate-resilient water policies and early warning systems in Oman.