Infiltration excess overland flow is the dominant mechanism for runoff generation in many dryland watersheds. Event-based rainfall-runoff models therefore partition precipitation into two components: loss and excess precipitation. The latter is then transformed into a runoff hydrograph. Numerous loss models have been developed over the past century ranging from simple empirical to sophisticated physically based methods. Complex models can lead to equifinality and associated uncertainty at larger spatial scales with varying soil and cover conditions. Simple models are therefore widely used in hydrologic practice. In the absence of measured data in many arid and semiarid regions, model parameters are often estimated based on laboratory or field infiltrometer tests. Given the documented importance of spatial scale on the runoff response in dryland catchments, it is not clear how models parameterized at the point or soil column scale will perform at the hillslope or catchment scale under real-world conditions. In this study, we compared the performance of seven simple loss models with three or less parameters: the Philip, Smith-Parlange, Horton, Kostiakov, curve number (CN), initial and constant (IC) and the linear and constant (LC) models. The latter is a modification of the IC model introduced in this study. We estimated parameters at the plot scale (2.8 m²) using rainfall simulation and then tested model performance at the hillslope (1.5–3.7 ha) and catchment scale (2.4–2.8 km²) based on measured rainfall-runoff data at two sites in New Mexico and Arizona, U.S. Results show that rainfall simulation can be used successfully to parameterize loss models at the hillslope scale. At the catchment scale, most models showed positive bias, suggesting that other losses (such as channel or transmission losses) play an important role in determining the catchment runoff response. Rainfall intensity and temporal distribution were found to be crucial for accurate runoff prediction. Models that are sensitive to rainfall intensity during the entire simulation (Philip, Smith-Parlange, Horton, Kostiakov, LC) therefore performed better than those with an initial abstraction term (CN, IC). During intermittent rain, the best results were achieved by methods expressing infiltration capacity as a function of cumulative infiltration (LC, Smith-Parlange).

This dataset contains comparisons of observed and simulated runoff for seven loss models (Philip, Smith-Parlange, Horton, Kostiakov, curve number, initial and constant, linear and constant) at three different spatial scales:

Plot Scale

19 runoff and infiltration time series were measured at seven test plots in central New Mexico. Tests were conducted using a portable rainfall simulator. A nozzle, suspended 3 m above ground, was used to spray water onto a 2.8 m² test plot at a constant rate of 194.1 or 58.4 mm/h. Below the nozzle, a sheet metal border was driven into the ground to a depth of approximately 5 cm. Any runoff originating inside the test plot was directed to a graduated container. Using a time lapse camera, runoff volume was recorded in one- or two-minute increments. For each time step, runoff rate was determined by dividing the incremental runoff volume by the plot size. Infiltration rate for each time step was calculated by subtracting the runoff rate from the rainfall rate. Each plot was tested under dry, intermediate and wet antecedent soil moisture conditions. Plot-scale data were used to optimize loss model parameters. 

Hillslope Scale

Optimized loss models were applied at the hillslope scale using data from the Walnut Gulch Experimental Watershed (https://doi.org/10.1029/2006WR005733). One example for each loss model is included in this dataset. The examples compare observed and simulated runoff from Walnut Gulch hillslope 102 resulting from the storm of August 4, 2002. Hillslope 102, with a contributing area of 1.5 ha, was modeled as a single basin. Precipitation data was obtained from a rain gage located on the hillslope. Excess precipitation was converted to runoff hydrographs by accounting for temporary storage effects using a linear reservoir model.

Catchment Scale

Optimized loss models were applied at the catchment scale using data from the Walnut Gulch Experimental Watershed (https://doi.org/10.1029/2006WR005733). One example for each loss model is included in this dataset. The examples compare observed and simulated runoff from Walnut Gulch catchment 4 resulting from the storm of July 25, 2008. The catchment was modeled using a 200 x 200 m grid. Precipitation for each grid cell was interpolated form adjacent rain gages. Excess precipitation was converted to runoff hydrographs by accounting for the translation time from each grid cell to the catchment outlet, and modeling temporary storage using a linear reservoir model.