An improved export coefficient model to determine total phosphorous loss spatial and temporal hotspots

Non-point source (NPS) phosphorus pollution remains a key driver of eutrophication in complex, data-limited watersheds. This research introduces an improved export coefficient model (ECM) called CAIBI-ECM, which incorporates the Contributing Area Index (CAI), Buffer Retention Index (BI), and monthly rainfall fluctuations to enhance the estimation of total phosphorus (TP) loss. The model was applied to the Shirin–Darreh watershed in northeastern Iran, characterized by heterogeneous land use and a eutrophic reservoir. Using GIS-based terrain and rainfall data, CaiBI-ECM simulated monthly TP exports across the watershed. Validation against field observations showed that the enhanced model reduced relative prediction errors by over 29.13% compared to a traditional ECM, with annual accuracy improving from 89.25 to 9.00%. Spatial diagnostics revealed that just 15–20% of the watershed contributes about 70% of the TP load. Temporal analysis highlighted April–June as the dominant export period, accounting for over 40% of the total annual loss due to runoff from snowmelt and rainfall. Hotspots were concentrated in eastern dry farming zones where low BI values (< 0.1) Limited retention. Reservoir TP concentrations reached 101.34 µg/L, far exceeding mesotrophic thresholds, and requiring a 72.2% reduction in loading. Scenario testing showed that only spatially and seasonally targeted best management practices (BMPs) could achieve meaningful mitigation. The CAIBI-ECM offers a scalable, low-data approach for identifying critical source areas, supporting adaptive watershed management under Total Maximum Daily Load (TMDL) frameworks.