Fog Early Warning System for Egyptian Desert Highways A Five-Year METAR Climatology and Multi-Source Remote Sensing Validation
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Road fog constitutes one of the leading meteorological hazards for vehicular safety on Egyptian desert highways, with the Cairo–Alexandria Desert Road (210 km, ~90,000 vehicles/day) recording recurrent multi-vehicle collisions attributable to sudden zero-visibility conditions. This study presents the first five-year systematic fog climatology for HECA station (2019–2023), integrating half-hourly METAR observations (68,146 records), MSG SEVIRI satellite imagery, and SRTM terrain analysis to develop a prototype Egypt Fog Early Warning System (E-FEWS). Twenty-eight fog events were identified over the study period, overwhelmingly concentrated in the November–February winter season (75%). The Dew Point Depression (DPD = T_air − T_dew) emerges as a robust, operationally viable early warning signal: validated against the complete observational record (22,332 winter half-hourly observations, with no estimated values), DPD < 2°C preceded 26 of 28 confirmed fog events at least 1–3 hours before onset (Event-level Recall = 92.9%), with a mean lead time of 2.1 hours (range: 0.5–5.5 h) and a directly measured False Alarm Rate of 11.4% (morning windows). Validation against MSG SEVIRI Night Fog RGB imagery confirms a clear spectral signal for the most intense event (8 February 2024, VSI: Extreme), with blue channel values increasing from background ~170 to ~190 DN and a diagnostic green/blue ratio of ~0.9, compared to values of 3–30× for concurrent dust/sand events. A Visibility Severity Index (VSI) is introduced (Extreme / High / Moderate / Low), with 32% of events reaching Extreme status (visibility ≤ 0.12 statute miles). A 7-day Consecutive Fog Nights (CFN) sequence in January 2021 — the longest in the record — highlights the risk of driver complacency during persistent fog episodes. The study demonstrates clear phenological and spectral separation between fog (DPD < 5°C, wind < 10 kt, low G/B ratio) and dust/sand hazards (DPD > 8°C, wind > 10 kt, high G/B ratio), supporting a three-class (Fog/Dust/Mixed) operational classification framework for Egyptian highway warning systems. Full ground-truth validation across the entire winter record yields observation-level Recall = 89.2% and event-level Recall = 92.9%, comparable to or exceeding established fog warning systems in temperate climates, with the single-parameter DPD criterion identified as an effective screening layer that benefits from multi-parameter refinement (wind speed, time-of-day) for operational false-alarm reduction.