Date of this Version
Populations of North American beaver (Castor canadensis) have increased in recent decades throughout the agricultural Midwestern U.S., leading to an increase in the frequency of beaver dams in small streams. The impact of beaver dams on channel structure in this region is not known. Our field observations indicate that beaver dams are too dynamic and their affects on channel structure occur over longer time frames than is practical to study with field measurements. Modeling is therefore needed to determine if beaver dams will help stabilize and aggrade incised streams. The objective of this paper is to determine how a channel evolution model (CONCEPTS) might be used to predict the impact of beaver dams on channel structure.
The study area, Little Muddy Creek watershed in southeastern Nebraska, is predominantly in agricultural land use. The main reach of the third –order watershed was surveyed for beaver dams from 2003 to 2005. Dam locations were mapped, integrity of dam structure was noted, and water surface elevations were measured. Failure of dam structure was documented following runoff-producing storms. While some dams were repaired within weeks, others were abandoned and left to degrade, causing a significant and transient change in the water surface profile of the stream. Field observations showed that the spatial arrangement and hydraulic condition of beaver dams were temporally dynamic in both short and long term scales.
Laboratory tests were conducted to determine if beaver dams could be modeled as broad-crested weirs. Discharge-rating curves were developed for a simulated beaver dam and a conventional weir. The roughness caused by the sticks on the surface of the dam significantly altered the stage-discharge relationship, but did not violate the broad-crested weir model.
Accounting for the temporal dynamics of spatial and hydraulic characteristics of beaver dams represents the greatest challenge to modeling the impact of beaver dams on stream channel morphology. TheCONCEPTS model, however, enables manually inserting the appropriate temporal patterns of structural change into model simulations and, thus, allows prediction of dynamic, long-term effects.