Regolith 2004 In: Roach I.C. ed. 2004. Regolith 2004 . CRC LEME, pp. 54-56. 54 DEVELOPMENT OF A SEMI-DISTRIBUTED CATCHMENT HYDROLOGY MODEL FOR SIMULATION OF LAND-USE CHANGE STREAM-FLOW AND GROUNDWATER RECHARGE WITHIN THE LITTLE RIVER CATCHMENT, NSW P.W. Carlile 1 , B.F.W. Croke 2 , A.J. Jakeman 2 & B.G. Lees 1 1 CRC LEME, School of Resources Environment and Society, Australian National University, ACT, 0200 2 CRC LEME, Centre for Resource and Environmental Studies and Intergrated Catchment Assessment and Management/School of Resources Environment and Society, Australian National University, ACT, 0200 This paper outlines how PhD research was developed to investigate the ability of a rainfall-runoff, recharge- discharge model to accurately simulate the hydrology of a catchment at various scales. Catchment attributes are utilised to obtain values for conceptual model parameters. This way modelling can take place in an ungauged catchment without the need for stream gauge data to perform calibration. Two hydrological models were used as a starting point for model development. These were a lumped conceptual rainfall-runoff model (IHACRES) (Jakeman & Hornberger 1993) and a physics-based conceptual groundwater discharge model (Sloan 2000). The modelling aims to appropriately disaggregate the catchment in order to improve on previous catchment or sub-catchment hydrology models, so that hydrological modelling can be carried out at the management scale. SCALE OF CATCHMENT DISAGGREGATION AND MODELLING To disaggregate the catchment into meaningful hydrological units requires manipulation of a Digital Elevation Model (DEM) using a number of steps within a Geographic Information System (GIS). These steps include (in order): estimation of flow direction and flow accumulation surfaces; stream links based on a threshold flow accumulation; and, stream ordering. Once the stream orders were determined, outlet points are estimated based on the maximum flow accumulation of each stream order link. Using these outlet points sub- catchments are created for each stream order. Modelling can then take place based on 1 st , 2 nd , 3 rd or 4 th order sub-catchments. Within these sub-catchments it was seen as favourable to form hydrological response units (HRUs), in which all water passes directly to the stream. This was seen as important so that individual soil-vegetation combinations within these HRUs can be connected or unconnected on a hill-slope. In this way the level of model hydraulic connectivity may be further tested and/or model complexity may be reduced. This reduction in model complexity centers on whether the lateral movement of water from one unit to another on a hill slope is included in modeling. The units in the model have their own catchment moisture store (deficit) and supply recharge and lateral flow directly to the aquifer and the stream respectively. It is hoped that the sum of these management units effectively characterise the hydrologic response of each HRU within each sub- catchment. Each sub-catchment is required to be comprised of three units to ensure all the water from each unit passes directly to the stream. These are the left, right and headwaters of the stream. To form these units the locations of stream dangling nodes were used as outlet points in the formation of new sub-catchments. By joining these new sub-catchments with the streams and original sub-catchment coverage, new units were formed that represented the left, right and headwaters of each stream. Figure 1 illustrates the formation of these hydrologic response units in each sub-catchment and the soil-vegetation-management units within them.