University of Twente Student Theses

Login

Flood frequency and inundation estimations under climate variability in eastern Australia

Gensen, M.R.A. (2014) Flood frequency and inundation estimations under climate variability in eastern Australia.

[img] PDF
2MB
Abstract:In this research a flood risk assessment was conducted for the Upper Mary River catchment. The Upper Mary River catchment is a relatively small non-coastal catchment area being 985 km2 in size. The only usable stream gauge is the Bellbird Creek gauge as others were affected by a dam. In total 78 suitable yearly discharge peaks were derived and identified. It is expected that discharges in this catchment area are affected by various climate variability modes. This statement is tested for the El Niño Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO) and Interdecadal Pacific Oscillation (IPO). At first an unconditional (independent of the phases of the climate variability modes) flood risk evaluation was conducted using all 78 yearly peaks. The resultant flood frequency curve can be compared to the conditional flood frequency curves which were calculated later. For the unconditional flood frequency multiple probability distributions were tested in order to find the best fitting distribution. It became evident that the Log Pearson type 3 distribution best fitted the data, therefore being used throughout the research. The resultant 100 year flood event and 1% annual exceedance probability is 3459 m3/s. To calculate the conditional flood frequency curves the phases of the different states of the IPO, PDO and ENSO were identified. For the IPO three phases were distinguished since 1920, two positive phases from 1920 to 1944 and from 1978 to 1998 and one negative phase from 1946 to 1976. The PDO and ENSO states were identified using climate indices and dividing them into positive, neutral and negative years. Subsequently the discharge peaks were divided into the years corresponding to the identified climate states. Using these discharge series the conditional flood frequency curves were determined. It was expected that all three climate variability modes would show an effect, but for the IPO no signal was found. The PDO and the ENSO showed similar effects in which the negative PDO phase and the La Niña phase had a higher 100 year flood event, respectively 5008.9 and 5179.2 m3/s opposed to 3010.6 and 3439.5 m3/s for the positive states. It is unequivocal that these variabilities lead to uncertainty in flood risk analyses. This uncertainty was evaluated by calculating random 100 year floods in a Monte Carlo simulation using different data lengths considering only the ENSO variability. When using a data series consisting of only 30 years the 95% bandwidth was found to be between 2.0 * 103 and 8.8 * 103 m3/s. Applying a considerably longer data length of 200 years the remaining uncertainty remained large: a 5% chance of the true value of the 100 year flood event being under 2.7 * 103 m3/s or over 5.3*103 m3/s. The calculated 100 year flood events were applied in a built HEC-RAS model in order to estimate flood inundations. It was found that little variety in flooded area would occur under the different climate variability states. This is caused by the specific bathymetry of the researched area. In terms of flood inundation heights larger differences were found, with the La Niña 100 year flood event having 2.0 metres additional inundation in comparison to El Niño’s 100 year flood event. Besides climate variability, future climate change is a potential source of error in flood risk analyses. It was found that for the area of interest, Eastern Australia and Southeast Queensland in specific, little quantitate predictions of climate change in terms of extreme events have been made. Researches with qualitative predictions also vary in their expected future changes. Some expect a decreased flood risk, whereas others expect an increased flood risk. The few quantitate predictions were applied to the Upper Mary River catchment. However a significant change in flood risk could not be determined.
Item Type:Essay (Bachelor)
Faculty:ET: Engineering Technology
Subject:56 civil engineering
Programme:Civil Engineering BSc (56952)
Link to this item:https://purl.utwente.nl/essays/66393
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page