Decades-long erosion monitoring in runoff plots or small watersheds is costly and labor intensive. A cost-effective alternative for estimating erosion using fallout caesium-137 ( 137 Cs) has been explored since the 1970s. This 137 Cs technique has been widely used to estimate the net (loss and gain) soil redistribution for the period from 1954 to the sampling year with a one-time field visit. However, the 137 Cs technique has shortcomings in its central assumption of no 137 Cs loss or redistribution during transfer from rainwater to soil. To improve the technique, we elucidate the 137 Cs transfer and redistribution processes by quantifying 137 Cs uptake in suspended sediment and 137 Cs interception by plants during fallout. An improved 137 Cs mass balance model was developed and optimized using legacy soil loss data from runoff plots along with 137 Cs inventories measured during the fallout period of 1954–1976. Predicted soil loss proved highly sensitive to 137 Cs redistribution by runoff and eroded sediment and interception by plants during transfer. The two processes can be effectively simulated by two key parameters: 137 Cs uptake by sediment (ψ) and 137 Cs interception by plants. When ψ is >1, it simulates 137 Cs redistribution in runoff. When ψ is <1, it rectifies the uniform erosion assumption by accounting for rill incision from concentrated flows and can be estimated as the proportion of interrill erosion (i.e., 1 minus the rill erosion proportion). More experiments are needed to quantify ψ under different rainfall and topographic conditions to improve parameter estimation and therefore erosion prediction. The processes and estimates found in this study are broadly applicable to other radionuclides commonly used for erosion prediction.
Zhang et al. (Thu,) studied this question.