Articles | Volume 18, issue 4
Hydrol. Earth Syst. Sci., 18, 1273–1288, 2014
Hydrol. Earth Syst. Sci., 18, 1273–1288, 2014

Research article 03 Apr 2014

Research article | 03 Apr 2014

Modelling overbank flood recharge at a continental scale

R. Doble1, R. Crosbie1, L. Peeters1, K. Joehnk2, and C. Ticehurst2 R. Doble et al.
  • 1Water for a Healthy Country, National Research Flagship, CSIRO Land and Water, PMB 2 Glen Osmond, SA, 5064, Australia
  • 2Water for a Healthy Country, National Research Flagship, CSIRO Land and Water, PMB 1666 Canberra, ACT, 2601, Australia

Abstract. Accounting for groundwater recharge from overbank flooding is required to reduce uncertainty and error in river-loss terms and groundwater sustainable-yield calculations. However, continental- and global-scale models of surface water–groundwater interactions rarely include an explicit process to account for overbank flood recharge (OFR). This paper upscales previously derived analytical equations to a continental scale using national soil atlas data and satellite imagery of flood inundation, resulting in recharge maps for seven hydrologically distinct Australian catchments. Recharge for three of the catchments was validated against independent recharge estimates from bore hydrograph responses and one catchment was additionally validated against point-scale recharge modelling and catchment-scale change in groundwater storage. Flood recharge was predicted for four of the seven catchments modelled, but there was also unexplained recharge present from the satellite's flood inundation mapping data. At a catchment scale, recharge from overbank flooding was somewhat under-predicted using the analytical equations, but there was good confidence in the spatial patterns of flood recharge produced. Due to the scale of the input data, there were no significant relationships found when compared at a point scale. Satellite-derived flood inundation data and uncertainty in soil maps were the key limitations to the accuracy of the modelled recharge. Use of this method to model OFR was found to be appropriate at a catchment to continental scale, given appropriate data sources. The proportion of OFR was found to be at least 4% of total change in groundwater storage in one of the catchments for the period modelled, and at least 15% of the riparian recharge. Accounting for OFR is an important, but often overlooked, requirement for closing water balances in both the surface water and groundwater domains.