Crop residue in North Dakota: measured and simulated by the wind erosion prediction system.

摘要

Residue cover is very important for controlling soil erosion by water and wind. Thus, the wind erosion prediction system (WEPS) includes a model for the decomposition of crop residue. It simulates the fall rate of standing residue and the decomposition of standing and flat residue as a function of temperature and moisture. It also calculates residue cover from flat residue mass. Most of the data used to develop and parameterize this model have been collected in the southern USA. We compared WEPS-simulated residue cover with that measured in south-central North Dakota for 50 two-year cropping sequences from nine crops species that were grown using no-till management. Measured data included residue mass at the time of harvest and residue cover just after seeding the next spring. Simulated residue cover significantly (P<0.05) underestimated measured cover for 33 out of the 50 simulated cropping sequences and overestimated measured cover for five cropping sequences. Some of the differences may be explained by the fact that, for many WEPS crops, residue decomposition parameters are not based on measured field data, but on expert judgment. In addition, WEPS did not predict any stem fall for most of the crops during winter, which contradicts observations that storms flatten many residue stalks of crops such as sunflower. In addition to stem fall and residue decay by biological means, which are driven by temperature and moisture, the model needs to explicitly simulate stem fall by mechanical forces, such as wind- and snowstorms, which are important in northern climates. Furthermore, WEPS does not model the migration of unanchored residue caused by rain- or windstorms, although this does affect residue mass-to-cover ratios and susceptibility to erosion. This study will help improve the WEPS decomposition model and its parameterization, but more data on residue decay and stem fall are needed for different climates and crops to ensure the applicability of the model over a wide range of conditions.