Blending field observations and AWRA outputs to estimate groundwater recharge in the Clarence-Moreton basin, eastern Australia

Abstract

Groundwater recharge is one of the hardest components of the water balance to estimate, largely because it cannot be directly measured. The Landscape component of the Australian Water Resources Assessment model (AWRA-L), has been shown to have the capacity to reproduce the temporal trends of groundwater recharge when compared to time series of field data at a limited number of locations. However, it cannot reproduce the heterogeneity in long term average recharge at a local scale. Many of our field techniques for estimating recharge can only produce a long term average recharge rate and provide no information on temporal variations. Blending the long term average recharge estimates from field observations with the continuous daily output of the AWRA-L model allow us to create a time series of recharge that honours the long term average of field observations whilst imposing the modelled temporal patterns. The Clarence-Morton Basin straddles the Qld-NSW border east of the Great Dividing Range and is one of the regions currently being studied as part of the Bioregional Assessments Programme. The majority of the basin has had limited investigations of its groundwater recharge processes with no knowledge of the long term average recharge rates and their inter-annual variabilities. Upon adoption of the chloride mass balance method for estimating recharge, 374 point estimates of recharge were obtained for this basin. These were subsequently used to create regression equations between the average annual recharge and the average annual rainfall for each of the surface geology types in the basin; this enabled those point estimates to be upscaled to the basin scale. The temporal pattern from AWRA-L was then imposed for each grid cell in the landscape to create a annual time series of recharge across the entire basin. The results of this investigation showed that the Walloon Coal Measures had the lowest annual average recharge rate (for a given rainfall) and the tertiary volcanics had the highest. The temporal trends in recharge from AWRA-L were similar between surface geology types with wet years having above average recharge and dry years having below average recharge. These results form an essential input into the numerical groundwater model used to investigate the impact of coal seam gas extraction upon water dependent assets in the Richmond River catchment, NSW.