Smart models for aquifer management: TopNet modelling suite

SKU:
SR_2019-27-pdf
$0.00
(Inc. GST)
$0.00
(Ex. GST)
Write a Review

 Zammit C, Yang J, Griffiths J, Rajanayaka C. 2019 Smart models for aquifer management: TopNet modelling suite. Lower Hutt (NZ): GNS Science. 118 p. (GNS Science report; 2019/27). doi:10.21420/FBS9-G965

Abstract:
This report presents the application of the TopNet model suite as one of the ‘simple’ models considered in the Smart Models for Aquifer Management (SAM) research programme. Led by GNS Science, the SAM programme aimed to investigate the implications of model simplification for uncertainty and water resource management decision-making. Comparison of different models in terms of prediction uncertainty relevant to key decision variables, both pre (‘priori’) and post (‘posteriori’) calibration, has been carried out. Models investigated range from a simple conceptual model (e.g. V-Notch); through to a more complex semi-distributed model (the TopNet model suite); a complex distributed model (LUCI); and a highly complex model (MODFLOW). It should be noted that the aim of the programme was to compare the impact of model uncertainties on the decision-making process rather than an inter-comparison of the performance of the models per se. The TopNet model suite (i.e. TopNet-0 and TopNet-GW models) has been developed over the past 10 years by NIWA through MBIE funded programmes such as Waterscape (MBIE COX1006 2010–2016). The TopNet-0 hydrological model is routinely used for hydrological modelling applications in New Zealand. It is a spatially distributed, time-stepping model of water balance simulating water storage in the snowpack, plant canopy, rooting zone, shallow subsurface, lakes and rivers. The TopNet-GW hydrological model couples TopNet-0 to a conceptual groundwater module which replicates the surface water sub-catchment spatial layout as a groundwater sub-catchment network. In TopNet-GW, the connection between groundwater sub-catchments and surface water is based on observed or predicted locations of surface–groundwater interaction, rather than on related physically-based parameters. The model also currently only represents annually averaged (or static) conditions. In addition, such representation may be influenced by the impact of existing groundwater abstraction patterns rather than natural groundwater variability. As part of the SAM programme, the TopNet model suite was applied to three case study catchments (i) the Ruamāhanga Valley (Greater Wellington Region), (ii) Mid-Mataura (middle Mataura and Waimea catchments (Southland Region)) and (iii) a synthetic model developed to represent an idealised generic groundwater and surface water catchment with characteristics typical of many natural catchments in New Zealand. Water management decisions were represented by allocation scenarios for two of the case studies (i and ii). To identify the impacts on the water resource two hydrological metrics were evaluated: number of days below a river flow threshold, and number of consecutive days below the same flow threshold. These metrics were applied to both the natural flow time series and the flow series resulting from application of surface water or groundwater allocation scenarios. To assess the impact of model complexity and model status on key environmental decision variables (i.e. discharge in rivers for this report), a common process was developed across the modelling team for the different SAM hydrological/hydrogeological models considered. Prediction uncertainty was estimated for an uncalibrated model (i.e. a model parameterised using existing a-priori knowledge) and a calibrated model (i.e. a model parametrised following calibration based on an agreed objective function related to the hydrological metrics- refer to as posteriori). For this project a team of four modellers was used by NIWA to complete the TopNet model suite across the three case studies. As a result, model calibration methodology varied across the case studies, which reflects both the modeller experience and data availability. (auth)