Comparison of SDSM and LARS-WG for simulation and downscaling of extreme precipitation events in a watershed

Abstract

Future climate projections of Global Climate Models (GCMs) under different emission scenarios are usually used for developing climate change mitigation and adaptation strategies. However, the existing GCMs have only limited ability to simulate the complex and local climate features, such as precipitation. Furthermore, the outputs provided by GCMs are too coarse to be useful in hydrologic impact assessment models, as these models require information at much finer scales. Therefore, downscaling of GCM outputs is usually employed to provide fine-resolution information required for impact models. Among the downscaling techniques based on statistical principles, multiple regression and weather generator are considered to be more popular, as they are computationally less demanding than the other downscaling techniques. In the present study, the performances of a multiple regression model (called SDSM) and a weather generator (called LARS-WG) are evaluated in terms of their ability to simulate the frequency of extreme precipitation events of current climate and downscaling of future extreme events. Areal average daily precipitation data of the Clutha watershed located in South Island, New Zealand, are used as baseline data in the analysis. Precipitation frequency analysis is performed by fitting the Generalized Extreme Value (GEV) distribution to the observed, the SDSM simulated/downscaled, and the LARS-WG simulated/ downscaled annual maximum (AM) series. The computations are performed for five return periods: 10-, 20-, 40-, 50-and 100-year. The present results illustrate that both models have similar and good ability to simulate the extreme precipitation events and, thus, can be adopted with confidence for climate change impact studies of this nature.

Modeling the Projected Changes of River Flow in Central Vietnam under Different Climate Change Scenarios

“…GCMs' output data can be used regionally, after disaggregation, to force hydrological models to simulate projected runoff and river discharge. An approach that is broadly used in many previous studies is to generate a baseline simulation based on observed meteorological data in the past and use the baseline meteorological data to force a hydrologic model to generate baseline hydrologic conditions [3,4,8,[11][12][13][14]. The evaluation of future changes in river discharge is based on the comparison of projected results with the baseline and their differences.…”

A coupled human–natural system to assess the operational value of weather and climate services for agriculture

“…In addition, the stochastic weather generator can also disaggregate a monthly forecast into daily values, which are needed to run the process-based model in the next step, without losing the generality of the statistical behaviour of the variables. The LARS-WG model (Semenov and Barrow, 1997) is selected for this task as it has been reported to outperform many other weather generators (Hashmi et al, 2011).…”

Climate Projections of Future Extreme Events in Malaysia

“…Despite numerous uncertainties in the different GCMs (Chu et al, 2010), these outputs provide hydrologists with priceless information. However, the coarse resolution of GCMs may lead to mismatch between the model's variables against observational variables for many climate change impact studies (Fowler et al, 2007;Hessami et al, 2008;Hashmi et al, 2009;Chu et al, 2010;Hashmi et al, 2010;Fatichi et al, 2011). The mismatch issues tend to produce inaccurate simulations of current regional climate for sub-grid scales (Chu et al, 2010;Guo et al, 2011).…”