Modelling catchment-scale water storage dynamics: Reconciling dynamic storage with tracer-inferred passive storage

Abstract

We investigate the storage-discharge relationships of two nested (3·6 and 30·4 km2) upland catchments using rainfall-runoff models: (1) a nonlinear discharge sensitivity function and (2) a tracer-constrained process-based conceptual model. Both approaches explicitly acknowledge that water storage is neither time nor space invariant and this provided insight into the respective scaling relationships. Both modelling approaches consistently demonstrated small seasonal storage fluctuations consistent with the wet, cool Scottish climate: the smaller catchment exhibited a greater average dynamic storage (ca 55 mm) compared to the larger scale (ca 40 mm). However, there are differences in simulated storage quantities and ranges inferred from both models, which can largely be explained by model uncertainties and model assumptions. In contrast to the dynamic (active) storage indicated by the rainfall-runoff models, input-output relationships of δ18O in both catchments were used to estimate the passive storage available for mixing and tracer damping in streams. This showed that catchment storage is an order of magnitude greater (ranging from 500 to 900 mm) than the dynamic storage estimated by both models, though again, storage estimates were greater for the smaller catchment. The passive storage inferred for mixing indicates that discussion of dynamic storage revealed by water balance considerations masks a much larger catchment storage which may in turn determine sensitivity to environmental change. © 2011 John Wiley & Sons, Ltd.