Hydrological modelling for conjunctive water use in the Murrumbidgee Catchment: groundwater recharge estimation

Abstract

Many regions of the world are under the stress of water scarcity. The stress is intensifying due to population and economic growth, and further exacerbated by global climate change. There is growing awareness of the need to manage surface water and groundwater in a more holistic way, that is conjunctively, which requires better understanding of the interactions between surface water and groundwater. Groundwater recharge is a vital interaction process and quantifying its amount in situations of interest needs considerable investigation. Recharge estimation, however, is fraught with uncertainty since the recharge process is full of complexity and plagued by heterogeneity and scale issues, and so benefits from use of multiple methods to help infer its range of possible values. Available estimation techniques include approaches that are physical, chemical and mathematical or a combination of them. Chemical and physical methods usually are expensive and their feasibility is limited to point or small scales. Numerical modelling is widely considered as an effective tool because of its potential to predict recharge rates in space and time. In this paper, we investigate a purely hydrological surface modelling approach, combined with extraction of recession-flow signals from their associated hydrographs, as one way to help appreciate the amount of recharge occurring in a catchment. Two hydrological models (SIMHYD and GR4J), one with and one without direct representation of the recharge process, are used to estimate groundwater recharge in unregulated sub-catchments in the Murrumbidgee catchment for the period 1976-2011. In addition, groundwater recharge is also estimated using the empirical and non-process-based RORA approach on the basis of daily streamflow observations and predictions. The results show that when the hydrological models are well calibrated, a notion of recharge can be estimated via the RORA approach applied to the model predictions, thereby circumventing the situation where the hydrological model itself has not included representation of the recharge process. For the hydrological model SIMHYD which has the recharge process represented, the recharge output directly from the model matches to a reasonable extent that derived from simulated streamflow using RORA. The results imply that surface hydrological modelling together with recession analysis using methods like RORA can be a useful adjunct to other methods to infer notional groundwater recharge to support decision-making on conjunctive water use. Show more