University of Birjand Water Harvesting Research 2476-6976 9 1 2026 03 01 Multiannual Soil-Erosion Potential and Structural Connectivity in the Deli Watershed, Southwestern Iran (2005–2025) 140 154 4244 10.22077/jwhr.2026.11799.1209 EN Kohzad Heidary Department of Soil Conservation and Watershed Management Research, Khuzestan Agricultural and Natural Resource Research Center, Agricultural Research, Education and Extension Organization, Ahvaz, Iran. 0000-0001-5156-2511 Amin Zoratipour Department of Natural Engineering, Faculty of Agriculture, Khuzestan University of Agricultural Sciences and Natural Resources, Ahvaz, Iran. 0000-0002-4979-576X Journal Article 2026 07 30 Water erosion in semi-arid mountain watersheds depends on both sediment-source potential and the structural connection of source areas to the drainage network. A multiannual, sensitivity-aware screening workflow was developed for the 105.20 km² Deli watershed, southwestern Iran, for 2005–2025. Landsat 5/7 indices were transformed to an Operational Land Imager-equivalent scale using local near-coincident Landsat 7–8 pairs, annual cover-management factors were derived, rainfall erosivity was represented using a CHIRPS–modified Fournier proxy, soil erodibility was estimated from SoilGrids, and topographic controls were derived from a 30 m DEM. Annual RUSLE products were combined with structural connectivity, hotspot persistence and conditional support-practice scenarios. The uncertainty measure is the low-to-high structural sensitivity range across alternative R, K, C and P assumptions, not a probabilistic confidence interval. All erosion and delivery results are unitless comparative index values rather than physical soil-loss or sediment-yield rates. The long-term mean potential soil-loss index was 582.11 index units; annual means ranged from 138.33 in 2010 to 1040.20 in 2018, with no significant monotonic watershed-scale trend. The annual response closely followed the rainfall-erosivity proxy (Spearman ρ = 0.98, p < 0.001), whereas its association with the watershed-mean C factor was weak and non-significant (ρ = −0.07, p = 0.758). Hotspots recurring in at least half of the years occupied 10.42 km², and mean relative structural uncertainty was 37.40%. Moderate and strong conditional P scenarios, representing 10–45% and 55–72.5% local reductions across eligible cells, lowered the index within 4.85 km² of stable cropland by 36.87% and 68.38%, but reduced the watershed-wide mean by only 0.34% and 0.63% because the treated domain covered about 4.6% of the watershed. The maps provide first-tier evidence for prioritizing field verification, monitoring and sediment pre-treatment. CHIRPS intensity limitations, the approximately 250 m support of SoilGrids and the absence of sediment observations preclude calibrated rates or stand-alone siting decisions. https://jwhr.birjand.ac.ir/article_4244_7369f3c86bf3c0a354615432832d9e8f.pdf