Location: Great Plains Agroclimate and Natural Resources Research
Project Number: 3070-11130-006-011-N
Project Type: Non-Funded Cooperative Agreement
Start Date: Apr 1, 2017
End Date: Mar 31, 2020
Objective:
The objective of this collaboration project is to thoroughly validate and evaluate the radionuclide fallout cesium-137 technique for estimating soil redistribution rate over a landscape. Four widely used cesium-137 redistribution models will be tested and evaluated against measured soil loss data in small watersheds selected from both the U.S. and China.
Approach:
USDA-ARS North Appalachian Experimental Watershed near Coshocton, Ohio measured surface runoff and soil loss from 1939 to 2016. Those data were the most appropriate data in the world for validating the Cs-soil redistribution models, which estimate soil loss since 1954 when nuclear bomb testing began. There are few small watersheds in the world that have continuous soil loss measurement since 1954. Three small watersheds (#109, #123, and #118) with drainage areas <2 acres under different tillage and cropping systems at Coshocton were sampled in a 10-m grid using a hydraulic probe. The probe i.d. is 5.2 cm and the sampling depth is 30cm. In addition, about 30 soil cores were taken on a reference site near watershed #102. Samples were dried and sieved to <2mm for Cs-137 activity measurement with a Gamma spectrometer by the Beijing Normal University. The measured daily precipitation, runoff, and sediment in runoff from the three watersheds will be compiled and analyzed. The measured soil loss will be compared with the soil losses estimated with four widely used Cs-redistribution models. The model errors will be analyzed and the best model identified.
Similarly, a few small watersheds or hillslope profiles will be selected from the Loess Plateau of northwest China and the Black soil region of northeast China. The selected sites will be sampled in a 10-m grid using soil probes. The Cs-137 activity will be measured, and will be converted to soil erosion rates using the best model selected from the U.S. watersheds.
