- Title
- Testing the statistics of dynamically downscaled rainfall data for the east coast of NSW
- Creator
- Parana Manage, Nadeeka; Lockart, Natalie; Willgoose, Garry; Kuczera, George; Kiem, Anthony; Chowdhury, A. F. M. Kamal
- Relation
- 36th Hydrology and Water Resources Symposium: The Art and Science of Water (HWRS2015). Proceedings of the 36th Hydrology and Water Resources Symposium (Hobart, Tas. 7-10 December, 2015) p. 1051-1058
- Publisher
- Engineers Australia
- Resource Type
- conference paper
- Date
- 2015
- Description
- This study performs statistical analysis of downscaled rainfall data, concentrating on the attributes important for hydrology predictions used in reservoir reliability simulations, specifically long-term persistence and variability of the rainfall. The dataset used was produced by the NARCliM (NSW/ACT Regional Climate Modelling) project which provides high resolution (10km) dynamically downscaled climate data for south-eastern Australia. NARCliM has produced twelve ensembles of simulations for current and future climates as well as three simulations driven by the NCEP/NCAR reanalysis for the 60 year period 1950-2009. In this study, statistical testing results are presented for the three reanalysis datasets as well as the Australian Water Availability Project gridded (0.05 0.05 ) daily rainfall dataset. The validation has been performed for the Hunter region, which has a range of climate types (e.g. coastal, inland, rainshadow, East Coast Lows. The spatial variability of statistics of the NARCliM and AWAP gridded rainfall were calculated and plotted. An autocorrelation analysis was performed for daily, fortnightly, monthly and annual time resolutions. The autocorrelation analysis showed that all three reanalysis datasets overestimate the magnitude of the autocorrelations shown in AWAP; however the timing of the seasonal cycle is similar to that of AWAP. In order to understand the rainfall distribution of each dataset, cumulative probability distribution plots were also generated at the daily time resolution. These plots showed that NARCliM reanalyses overestimate the magnitude of rainfall at all probabilities.
- Subject
- NSW; rainfall data; climatology; hydrology predictions; statistical analysis
- Identifier
- http://hdl.handle.net/1959.13/1330842
- Identifier
- uon:26490
- Identifier
- ISBN:9781922107497
- Language
- eng
- Reviewed
- Hits: 1602
- Visitors: 1730
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|