Groundwater recharge modelling using the Monte Carlo technique, Manawatu region, New Zealand

• Published in: Journal of hydrology (Amsterdam) 1999-11, Vol.224 (3/4), p.137-148
• Main Authors: Bekesi, G, McConchie, J
• Format: Article
• Language: English
• Subjects: Earth sciences; Earth, ocean, space; Engineering and environment geology. Geothermics; Exact sciences and technology; Hydrogeology; Hydrology. Hydrogeology; Monte Carlo method; New Zealand; Pollution, environment geology
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/S0022-1694(99)00128-6

The study area, the Manawatu region of New Zealand, covers approximately 4200 km2 and contains numerous aquifers in unconsolidated Quaternary sedimentary formations. Most of the 5000 bores are shallow, with over 75% being less than 50 m. Of concern, is that 20% of bores in the south tap groundwater with a nitrate-nitrogen concentration above the drinking water standard (11.3 mg/l) An effective regional vulnerability assessment was therefore developed to mitigate this problem. While four factors largely control aquifer vulnerability in the study area, this paper focuses on developing a regional rainfall recharge model. Percolating rainwater is the major transporting agent for contaminants through soil and unsaturated zone. Mean annual recharge was calculated at each rainfall station throughout the study area using the Monte Carlo technique to randomise soil moisture parameters and approximate the variability of soils. These point estimates of recharge were then modelled spatially to produce a 'recharge surface'. Good agreement between the modelled and actual groundwater levels was obtained. Likewise, the regional recharge surface is consistent with the physical processes and the expected pattern. Uncertainty of recharge estimation is expressed as the standard deviation of the mean annual recharge resulting from the Monte Carlo modelling. Estimates of both recharge and uncertainty are presented as readily assimilated colour-coded maps which can be used by resource planners and other laypersons. This methodology is considered appropriate for other non-arid areas with limited soil and groundwater data coverage.