A regional analysis of the effects of largest events on soil erosion

A large amount of geomorphic work is caused by a small number of extreme events that are mainly responsible for the time compression of geomorphic processes. The classic approach defines extreme events by their magnitude and they are quantified by certain deviation from a central value. Alternatively, we define extreme events as the largest events sorted by rank, whatever their absolute magnitude. In this case, events with equal rank from two different sites can be responsible for different magnitudes of geomorphic work, e.g., the amount of erosion. The new approach applied to soil erosion is that, whatever the magnitude of soil eroded, erosion is a time compressed process and the percentage contribution to total soil erosion by the few largest events (regardless of their magnitude) is negatively related to the total number of daily soil erosion events recorded. To verify and generalize this approach, we used the most extensive daily soil erosion dataset available (USLE database). In this paper, we present a geographical analysis of the effects of largest daily event on soil erosion by comparing 594 erosion plots located in agricultural fields under various climatic temperate conditions across the central-east of the USA. Plots differ in cover, soil types and length of records. Results indicate that: i) soil erosion in agricultural fields is a highly time compressed process and soil erosion mean values calculated over short time periods are biased, regardless of the magnitude of daily events recorded; ii) the relative effects of the n-largest daily events (whatever their magnitude) on total soil erosion depends on the length of records and, particularly, on the total number of events recorded; iii) the derived relationship of the required time length for records is generalized on a semi-continental scale; and iv) thus seems to be independent of climate conditions. This new approach can help us to define the minimum number of recorded soil events needed to avoid bias in soil erosion evaluation, in other words: the minimum period of field research on erosion should be evaluated not in years, but by the total number of daily erosive events. Thus, the number of daily erosive events is the key parameter for characterizing soil erosion processes at each measuring location.

► Soil erosion is a compressed process in time and independent of extreme events.
► It depends on the named “largest events”.
► The n-largest events contribution not differ between regions of USA.
► Time compression pattern is expressed by an empirical power law.
► Erosion research validation require a minimum number of daily events.