| Multi-Sensor Screening of Environmental Stress and Potential Pollution Anomalies in South Pars Coastal Waters |
| کد مقاله : 1201-IWWA |
| نویسندگان |
|
سید مصطفی داورپناه شهری *1، زهرا صداقی2، صبا علی نژاد2 1عضو سازمان نظام مهندسی سانکا خراسان رضوی 2دانشجوی دکترای دانشگاه تبریز |
| چکیده مقاله |
| Coastal ecosystems adjacent to large petrochemical complexes face environmental pressures that are difficult to monitor through conventional field sampling alone. This study presents a multi-temporal, multi-sensor remote sensing and GIS framework to screen for environmental stress and potential pollution anomalies in the coastal waters of the South Pars / Asaluyeh Special Economic Energy Zone, Iran, without in-situ measurements. Six periods spanning January–August 2026 were analyzed using optical, thermal, and radar indicators derived via Google Earth Engine. Multivariate anomaly detection, spatial hot-spot analysis, and physical zonal validation of SAR-derived oil-slick candidates were cross-checked against wind-speed confounding, while land-cover classification agreement was quantified across three independent methods. Results reveal a statistically robust, spatially clustered pattern of anomalies corroborated by independent temporal persistence and a genuine, large-effect-size reduction in SAR backscatter within candidate polygons. A significant declining trend in permanent-water-class area was also identified. Conversely, coastal vegetation health correlated positively, not negatively, with water-quality indices, suggesting a natural ecological signature rather than pollution-driven stress. These findings demonstrate that a cross-validated, multi-sensor framework can distinguish genuine anomalies from natural variability, serving as a low-cost screening tool for data-scarce industrial coastal zones. |
| کلیدواژه ها |
| Remote sensing; coastal water quality; SAR oil-slick detection; Getis-Ord Gi*; anomaly detection; Random Forest classification. |
| وضعیت: پذیرفته شده برای ارسال فایل های ارائه پوستر |
