Daytime temperature extremes and percentile-based thresholds (p95–p99) in Sabah, Borneo (1981–2024): Climatic modulation, compound drivers, and emerging heat risks
Corrina Henry Mansa, Ricky Anak Kemarau, Ramzah Dambul, Zulfaqar Sa’adi, Wee Hin Boo, Zaini Sakawi, Oliver Valentine Eboy, Zaher Mundher Yaseen, Stanley Anak Suab, Wan Shafrina Wan Mohd Jaafar, Liew Juneng, Maggie Chel Gee Ooi, Noorashikin Md Noor, Khairul Nizam Abdul MauludThis study quantifies the magnitude, spatial organisation, and climatic modulation of daytime temperature extremes across Sabah, Borneo, in equatorial Southeast Asia during 1981–2024. Daytime maximum temperature (Tmax) extremes were assessed using ERA5-Land as the primary dataset and independently validated against CHIRTS-daily at the district scale (R 2 > 0.90). Ground-based evaluation using 1,544 valid monthly pairs from Kota Kinabalu, Kudat, Sandakan, and Tawau yielded r = 0.828, KGE = 0.691, RMSE = 0.968°C, MAE = 0.876°C, and a mean bias of −0.838°C. These results indicate that ERA5-Land represents regional temporal variability more reliably than absolute station-level temperature magnitude. Percentile-based thresholds (p95, p97, and p99) were used to characterise spatial amplification and evaluate modulation by large-scale climate modes. Mean Tmax increased by + 0.020–0.031°C yr -1 across Sabah, corresponding to total warming of +0.8–1.3°C since 1981. The p99 threshold increased by + 2.6°C, while extreme events that were previously decadal now occur almost annually. Pronounced spatial gradients were evident, with Tmax exceeding 37°C in the coastal and lowland-basin areas of Semporna, Lahad Datu, and Kinabatangan, compared with lower but consistently increasing temperatures in the high-elevation interior. Tmax extremes were significantly modulated by interacting ENSO, IOD, MJO, and monsoonal conditions, although their influences were neither independent nor temporally uniform. The strongest compound response occurred during concurrent El Niño, positive IOD, and Southwest Monsoon conditions, when Tmax anomalies exceeded +2.3°C relative to the climatological baseline. This district-level percentile framework provides a locally adaptive diagnostic for heat-risk assessment and demonstrates the increasing persistence of extreme heat across Sabah, with important implications for agricultural productivity, ecological resilience, and human health.