- Title
- Multi-variable hydrological simulation in a semi-arid catchment towards effective water resources management
- Creator
- Binesh, A.; Yeo, I. Y.
- Relation
- 38th Hydrology and Water Resources Symposium (HWRS 2018). Hydrology and Water Resources Symposium (HWRS 2018): Water and Communities (Melbourne 3-6 December, 2018) p. 1015-1029
- Publisher
- Engineers Australia
- Resource Type
- conference paper
- Date
- 2018
- Description
- Water resources management require comprehensive understanding on the spatio-temporal hydrologic processes throughout landscapes to the catchment. Reliable estimates of hydrologic variables, such as soil moisture, evapotranspiration, and streamflow, over multiple spatial and temporal scales provide crucial information to assist the decision-making and water resources management. However, the hydrologic processes across water-limited regions is extremely difficult to understand and quantify, due to the extreme climate variability, complex soil characteristics and vegetation condition. This study attempts to improve understanding on the hydrological processes of a semi-arid catchment under the extreme climate variability. The aim is to provide more reliable prediction of soil moisture and streamflow at high spatial resolution using Soil and Water Assessment Tool (SWAT) to improve water balance estimation at the field and catchment scales. SWAT is a continuous semi-distributed, physically based hydrological model developed to predict hydrology and water quality processes. We proposed the multi-variable approach (i.e. using both on-site soil moisture at the field and streamflow from the catchment outlet) to calibrate the SWAT and compared the results against those obtained from the calibrated SWAT under the single-variable approach (i.e., using only streamflow observed from the catchment outlet as the observed variable). As the case study, SWAT is applied to the Merriwa sub-catchment, a semi-arid catchment located in the Upper Hunter Region of NSW. The study site has been extensively monitored with in-situ soil moisture network since 2002 and experienced high climatic variation over last two decades. The streamflow based sensitivity analysis highlighted the importance of those parameters controlling surface runoff and routing processes, while the sensitivity analysis under the multi-variable approach pointed to the importance of soil properties to make reliable field-scale soil moisture estimation. While both calibration set-ups led to the satisfactory streamflow prediction (with the NSE value>0.5), the SWAT calibrated only to the streamflow did not provide reasonable soil moisture estimation at the HRU scale. Furthermore, the two calibrations showed significantly different catchment water budget estimation over long run. The SWAT calibrated against streamflow showed significantly less soil moisture store (21%) while much higher estimation for evapotranspiration (61%), while the multi-variable calibration model showed higher soil moisture stored (34%) but less evapotranspiration (47%). This different partitioning of the precipitation between soil moisture and streamflow was persistent throughout the seasons and this, in turn, affected the streamflow generation processes. This difference in the model prediction clearly has important implication for water management for the region, highlighting the importance of considering multiple hydrologic variables to improve model parameterization.
- Subject
- SWAT; semi-arid region; hydrologic components; water management
- Identifier
- http://hdl.handle.net/1959.13/1403151
- Identifier
- uon:35114
- Identifier
- ISBN:9781925627183
- Language
- eng
- Reviewed
- Hits: 777
- Visitors: 754
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|