RunCA: A cellular automata model for simulating surface runoff at different scales

- A cellular automata based surface runoff model is presented.
- The model is designed to capture both temporal and spatial variations in runoff.
- Modeling results agree well with measured runoff at different spatial scales.

SummaryThe Runoff Model Based on Cellular Automata (RunCA) has been developed to simulate surface runoff at different scales by integrating basic cellular automata (CA) rules with fundamental measureable hydraulic properties. In this model, a two-dimensional lattice composed of a series of rectangular cells was employed to cover the study area. Runoff production within each cell was simulated by determining the cell state (height) that consists of both cell elevation and water depth. The distribution of water flow among cells was determined by applying CA transition rules based on the minimization-of-difference algorithm and the calculated spatially varied flow velocities. RunCA was verified and validated by three steps. Good agreement with the analytical solution was achieved under simplified conditions in the first step. Then, results from runoff experiments on small laboratory plots (2 m × 1 m) showed that the model was able to well predict the hydrographs, with the mean Nash-Sutcliffe efficiency greater than 0.90. RunCA was also applied to a large scale site (Pine Glen Basin, USA) with data taken from literature. The predicted hydrograph agreed well with the measured results. Simulated flow maps in this basin also demonstrated the model capability in capturing both the spatial and temporal variations in the runoff process. Model sensitivity analysis results showed that the calculated total runoff and total infiltration were most sensitive to the input parameters representing the final steady infiltration rate at both scales. The Manning's roughness coefficient and the setting of cell size did not affect the results much at the small plot scale, but had large influences at the large basin scale.