Application of Multi-site Weather Generators for Investigating Wet and Dry Spell Lengths under Climate Change: A Case St
- PDF / 686,722 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 30 Downloads / 169 Views
Application of Multi-site Weather Generators for Investigating Wet and Dry Spell Lengths under Climate Change: A Case Study in Southern Taiwan Hung-Wei Tseng & Tao-Chang Yang & Chen-Min Kuo & Pao-Shan Yu
Received: 19 February 2012 / Accepted: 4 September 2012 / Published online: 15 September 2012 # Springer Science+Business Media B.V. 2012
Abstract The study compared the performances of three weather generators (WGs), including a parametric model and two non-parametric models, in producing synthetic daily rainfall time series for multiple sites. The observed daily rainfalls of six raingauges during 1979~2008 in the catchment of Tseng-Wen Reservoir in Southern Taiwan were used as the data set. The generated results reveal that the k-nearest neighbor WG with a fixed window (i.e., a non-parametric model) is the best for daily rainfall generation at each site and performs well in preserving spatial correlation of rainfall among sites. The best WG was further applied to assess the impact of climate change on rainfall temporal characteristics (i.e., annual number of wet day, annual maximum number of continuous wet days and annual maximum number of continuous dry days) by using the downscaling results of 24 GCMs under the A1B emission scenario during 2020~2039. It is found that the rainfall temporal characteristics will change in the future which may make Southern Taiwan tend to face a longer period with no rain. Keywords Climate change . Weather generator . Rainfall characteristics 1 Introduction In Taiwan, the major sources of rainfall are Mei-Yu front (stratiform precipitation) and typhoon events. The mean annual rainfall is approximately 2,500 mm. Though the rainfall amount seems abundant, the rainfall temporal distribution is extremely uneven. For example, around 90 % of annual rainfall occurs during the wet season (from May to October) in Southern Taiwan, which makes a big challenge on water supply and allocation. Due to the rainfall temporal distribution during a year plays a critical role in water resources management, this study aims at finding a suitable weather generator (WG) for the catchment of Tseng-Wen Reservoir in Southern Taiwan to assess the impact of climate change on rainfall H.-W. Tseng : T.-C. Yang : C.-M. Kuo : P.-S. Yu (*) Department of Hydraulic and Ocean Engineering, National Cheng Kung University, No. 1, University Rd., Tainan 701, Taiwan e-mail: [email protected]
4312
H.-W. Tseng et al.
temporal characteristics (i.e., annual number of wet day, annual maximum number of continuous wet days and annual maximum number of continuous dry days). Different kinds of WGs have been developed to reproduce the rainfall characteristics of observed data. Generally speaking, WGs can be roughly categorized into two major categories (i.e., parametric and non-parametric WGs). The parametric WGs, such as Weather Generator (WGEN) (Richardson 1981; Richardson and Wright 1984) and CLImate GENerator (CLIGEN) (Min et al. 2011), adopt the first order Markov chain and probability distributions to generate several
Data Loading...
