Multi‐site rainfall simulation at tropical regions: a comparison of three types of generators

Abstract

ABSTRACT:Rainfall modelling is an essential component of different hydrological studies. However, rainfall modelling in tropical regions, especially urban areas, remains inadequate. To determine the applicability of different types of rainfall modelling approaches, simulations by two Markov models (Matlab-based weather generator (MulGETS) and multi-site rainfall simulator (MRS)) and a Neyman-Scott based Poisson cluster model (RainSim) were compared with a proposed modified k-nearest neighbour (KNN) model for 30 years rainfall in Singapore. The MRS model was determined to be suitable for single-site applications in tropical regions. However, for multi-site conditions, RainSim was adjudged the most suitable given its accuracy in preserving observed spatial information, despite its performance lagging in few statistical indicators. The KNN model was found to perform satisfactorily during the wet seasons, and was the only model that could repeat the extreme precipitation values closely. Although typical studies evaluate the performance of models based on a set of criteria, there exists a lacuna in understanding the quality of simulations by these models. Therefore, uncertainty analysis based on two different criteria was implemented to understand the performance of the stochastic processes within. Although the MulGETS model exhibited the lowest differences between Prediction Intervals, RainSim's Prediction Intervals were found to subsume observed data more often. The proposed study would be useful for users examining rainfall models for differing objectives.

Flood risk analysis under consideration of river system dynamics and spatial temporal scales

“…Vallam & Qin, 2016). Literature suggests w values of 14 days in Europe (Sharif and Burn, 2006) or 20 days in tropical climate (Vallam & Qin, 2016). Preliminary testing of different window sizes, ranging from 10 to 20, suggests little influence of w to reproduce extreme discharge.…”

“…The KNN-CAD v4 is a block-bootstrap non-parametric generator with perturbation based on the K-nearest neighbour (KNN) algorithm (King et al 2015). KNN types of weather generators have been well documented and applied in many research studies, proving to produce adequately rainfall and temperature (Sharma & Lall, 1999;Buishand and Brandsma, 2001;Bannayan & Hoogenboom, 2008;Ly & Qin, 2014;Vallam & Qin, 2016) and extreme flows (Leander & Buishand, 2007;Wit & Buishand, 2007).…”

“…The parameters of KNN-CADv4 are typically selected on a trial and error basis (Vallam & Qin., 2016). The KNN-CADv4 requires a set of three main parameters.…”

“…The temporal window w, which controls the neighbours sample size k, is usually optimized on a trial and error basis with objective to minimize seasonality disruption (e.g. Vallam & Qin, 2016). Literature suggests w values of 14 days in Europe (Sharif and Burn, 2006) or 20 days in tropical climate (Vallam & Qin, 2016).…”

“…Besides, few direct comparisons of multi-site weather generator performance have been done. In their study,Vallam & Qin (2016) demonstrated that performances may vary with the specific characteristics of rainfall to be preserved (i.e. extreme values and spatial correlation).…”