No description available.

No description available.

No description available.

No description available.

Runoff generation refers to the physical processes by which water travels through the landscape, moving through the subsurface or over the ground surface, ultimately arriving at the stream channel. These physical processes vary in both space and time leading to difficulties in mechanistic modelling of storm response, contaminant transport and nutrient fluxes. Runoff generation has been extensively studied at the hillslope scale and in headwater catchments. However, only recently have empirical studies begun to collect similarly detailed datasets across multiple catchments with which to examine how these processes change with scale. This study examines runoff generation from a series of eight small nested forest catchments and focuses specifically on the controlling influences of antecedent moisture conditions and catchment topography. / End-member-mixing-analysis using stream water hydrochemistry from the eight catchments shows changing seasonal and storm-based source water contributions to the stream channel. Analysis identifies hydrochemical solutes with behaviour consistent with the assumptions of the mixing-model approach for all eight catchments. Results indicate that testing of solute selection is critical in the application of this method to multiple catchments. / Runoff generation observed for five storm events shows a strong nonlinear relationship between runoff and antecedent moisture conditions, supporting the hypothesis of varying 'states of wetness'. Detailed hillslope-scale investigation during the different 'states of wetness' indicates that while groundwater and soil moisture profiles show changing active-flow connectivity on a seasonal and storm-base dtime scale, there no significant change in spatial patterns of shallow soil moisture. These results suggest that a priori spatial patterns in shallow soil moisture in forested terrains may not be a good predictor of critical hydrologic connectivity that leads to the threshold change in runoff generation, as has been found in rangeland catchments. / Differences in storm response from the eight catchments are in part attributable to variation in topography and landscape organization. The multiple catchments have similar distributions of topographic index and yet differences in mean values of topographic index lead to significantly different estimates of mean residence time. Scaling of storm response is dominated by the behaviour of the three largest catchments. These three catchments distinguish themselves with larger MRT and larger valley bottom areas. It is these three catchments that, under dry antecedent moisture conditions, show significantly larger amounts of new water delivery to the stream channel, suggesting a significant change in dominant runoff mechanisms related to topography and landscape organization.

No description available.

A research study was undertaken to determine the predictive capability of the AGNPS model with respect to surface runoff, peak flow, and sediment yield produced by rainfall-runoff events on a 26 km$ sp2$ watershed in Quebec. Precipitation, stream discharge, surface runoff, and suspended sediment concentrations were monitored for rainfall-runoff events occurring from 1994-96, inclusive. Data describing stream patterns, topography, soil type, and land use were collected and input to the model. / Seven rainfall-runoff events were used for model calibration. Five storms were used to validate the model. Calibration curves were developed to correlate the antecedent precipitation index (API) to the SCS curve number. For model calibration, coefficients of performance of 0.05, 0.43, and 0.12 were obtained for surface runoff, sediment yield, and peak flow, respectively. For model validation, coefficients of performance (CP$ rm sp prime sb{A}$) of 0.02, and 0.01 were obtained for surface runoff, and sediment yield, respectively. Peak flow was generally overpredicted and yielded a CP$ rm sp prime sb{A}$ of 2.07. / A sensitivity analysis showed API and associated curve numbers to be the most sensitive input parameters. USLE factors were also sensitive. The surface condition parameter and Manning's n had negligible influence on model output. The hydrograph shape factor toggle parameter showed extreme sensitivity. / A simulation of best management practices on the basin estimated soil loss reductions of 15 to 25% for storms of varying magnitudes if a 4-year crop rotation implementing conservation tillage were to be adopted.

No description available.