Testing a Mechanistic Soil Erosion Model for Three Selected Soil Types from Iran

Hairsine and Rose (1991) developed a process-based soil erosion model which described the erosion transport of multiparticle sizes in sediment for rain-impacted f low in the absence of entrainment in overland f low. In order to test this model laboratory experiments were carried out in a detachment tray using simulated rainfall. Three contrasting soil types were subjected to simulated rainfall at diff erent rates (25-110 mm h- 1) in a 35×30 ×10 cm detachment tray. Rainfall was applied using a rainfall simulator with a single scanning nozzle located four meters above the soil surface that emitted drops with a mean diameter (volumetric D50) of 1.5 mm. Results showed that the Hairsine and Rose model can clearly describe the sensitivity of diff erent soils to erosion by introducing three terms of detachability, re-detachability and settling velocity, though the model is unable to describe aggregate breakdown which takes place in one of the soil at higher rainfall rates. The experimentally observed relationship between ponding water depth and soil detachability parameters did not agree with previously proposed theories. In addition, the results showed that the Hairsine and Rose model tends to over-predict values at the lower end of the scale, and under-predict values at the upper end, although the average sediment concentration predicted for the entire data set is not greatly diff erent f rom the average measured values