Data analysis for Puttock, et al. (2021) and Graham, et al. (In Review).
Workflow for the processing and analysis of hydrological data from East Budleigh EA gauge to evaluate the impacts of Beaver on flow regimes...
Location of Budeligh brook Catchment Colaton Brook catchemnts, gauging station and Beaver dam sequence.
All events overlain including fitted General Additive Models (GAM). Used to estimate and compare changes in hydrological event geometry pre/post beaver.
Range of GLMs to investigate flow attenuation mechanism. M4/M5 are comparable and perform best; M5 model residuals indicate less deviance in residuals for larger events and therefore offering more reliable predictions for these key larger events. Attenuation increases up to a point - after which we cannot be confident in observing significant differences between factor levels (beaver presence/absence).
Predicted attenuation increases up to 94th percenile of event total rainfall. beyond the 97th percetile there is a large increase in our prediction uncertainties.