Infrastructure Engineering - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemStochastic joint probability modelling of estuarine flood levels( 2004)The determination of the annual exceedence probability (AEP) of extreme water levels, such as the 1% AEP flood level, in complex estuarine systems is an important and yet highly challenging issue. Water level AEP is required for land and water resources planning, emergency management and flood insurance underwriting. Extreme water levels in estuaries are caused by the combined effects of environmental forcings (river floods, winds and coastal ocean levels (COLs)), estuarine hydrodynamics, floodplain topography and catchment conditions. A comprehensive flood study should therefore entail a detailed hydrological, hydraulic and terrain modelling of the entire system. Unfortunately, there is currently no standard procedure for undertaking such a study. The question asked in this thesis is: "Is it possible to estimate, in a scientifically rigorous but computationally efficient way, the AEP of extreme water levels in large and complex estuarine systems such that the spatial and temporal forcing characteristics ranging from catchment to synoptic scales are preserved?" This question is addressed by developing a generic modelling method for application to any estuaries, and testing it on the Gippsland Lakes in southeast Australia - a coastal lagoon system having water surface area of almost 400 km2 and contributing catchment area of over 20,000 km2. The new method is a stratified Monte-Carlo stochastic-deterministic hydro-climatic modelling-based joint probability (MBJP) method. Conceptually, two thousand years of stochastic event-based concurrent hourly forcing sequences (river flows, winds and COLs) that preserve the space-time cross-correlations are generated using a sequence of hydro-climatic models developed in this thesis. Monte-Carlo (MC) simulation of event-based water levels around the estuarine system is then carried out using a calibrated hydrodynamic model (HDM) driven by the generated stochastic forcing sequences. (From Abstract)
-
ItemSoil moisture and hydrology of the basalt plains of Western Victoria: a field and computer study of surface hydrology( 1979)This is a field and digital model study of surface hydrology. An attempt is made to relate field measurements to the assumptions of a physically-based deterministic rainfall-runoff model that has been proposed for widespread use in Australia. A major part of the field study was the measurement and interpretation of soil moisture contents obtained with a neutron moisture meter, and two methods were developed and used to provide calibrations appropriate to both locally variable soils and the large (55 km2) study area. Independent estimates of each of the major components of the water balance were made, utilizing data collected on the study by other authorities. Further measurements of field parameters were made to provide independent estimates of the appropriate parameter values for the digital model. A single year of data was used for development of the model and final selection of the parameter values. The prediction of the observed streamflow was used as the basis for this testing, and the hydrological validity of the assumptions upon which the model is based was then tested by a comparison of the internal prediction and the independent measurements of soil moisture contents. The model was found to behave reasonably well, but deficiencies in some of the algorithms were identified. A sensitivity analysis showed that the concept of an infiltration capacity is not important for runoff generation on this catchment, and that the spatial variability of the moisture holding capacity of the soil makes it virtually impossible to estimate this model parameter from field measurements alone. It was found that a simple approximation to this spatial variability provided a more realistic annual hydrograph, and reduced the accuracy required to determine the mean value of this moisture storage capacity from field measurements. The main conclusions are that deterministic models offer a useful approach to the understanding of surface hydrology, although further development is required before the parameter values for such a model can be determined solely from field measurements. The use of the field measurements and the model showed that the hydrology of this catchment is dominated by a nearly impermeable B horizon, that the concept of an infiltration capacity is irrelevant to the generation of storm runoff, and that the spatial variation of the moisture holding capacity of the soil is important. Particular topics for further study include the acceptance of spatial variability in the other algorithms of the model, and the hydrological validation of this and other models on other catchments.