Evaluating land-cover change effects on runoff and recharge in Kawela, Moloka'i, Hawai'i

Abstract

The Precipitation Runoff Modeling System (PRMS), a modular, physically based, distributed-parameter modeling system, was used to develop a watershed model for Kawela, Moloka‘i to evaluate the impact of changing watershed characteristics on surface-water runoff and groundwater recharge. Available spatial information was processed in a geographic information system (GIS) environment and assigned to the delineated hydrologic response units (HRUs) within the watershed. PRMS simulates different parts of the hydrological cycle based on a set of user-defined modules, and each component of the hydrological cycle is computed by empirical relations or process algorithms. For each HRU, an energy and a water balance is computed; the sum of all of the HRU’s water-budget components produces the watershed’s total hydrological response. The model was manually calibrated using a climatic adjustment coefficient and this calibration resulted in a reasonable match between simulated and observed hydrographs and flow volumes. To further minimize any differences between the simulated and observed streamflow values, the model was automatically calibrated using PEST (Parameter ESTimation) software. The simulated total runoff volume was within 8.7 percent over the entire simulation period (04/01/2006-03/31/2010). Simulation results for the four-year period indicate that 91 percent of the precipitation that falls on the watershed is partitioned into evapotranspiration (43 percent) and groundwater recharge (48 percent). A much smaller percentage of rainfall is partitioned into runoff (8 percent) that is measured at the outlet of the watershed. The calibrated model was used to assess different watershed restoration and degradation scenarios and evaluate the hydrological system’s sensitivity to changes in land cover. Compared to the current land cover, the tested land-cover change scenario of vegetation denudation resulted in a smaller component of fog-drip, which translated to a 4 percent decrease in precipitation and consequently only a 1 percent increase in the amount of precipitation partitioned into runoff. However, vegetation restoration decreases runoff by 16 percent, which, by inference, would lead to reduced sediment loading of the nearshore environment. The amount of precipitation partitioned into recharge changed by less than 5 percent in both scenarios. PRMS is a helpful management tool that can be used to evaluate changes in runoff and recharge under different land-cover change scenarios.