Projection of Climate Change Based on Multi-Site Statistical Downscaling over Gilan area, Iran

آب و خاک

Article 1, Volume 30, Issue 5 - Serial Number 49, November 2016, Pages 1686-1699   

Document Type: Research Article

DOI: 10.22067/jsw.v0i0.47008

Authors

Vesta Afzali^{* 1}; Masoud Reza Hessami Kermani²

¹Graduate University of Advanced Technology, Kerman

²Shahid Bahonar University of Kerman

Abstract

Introduction: The phenomenon of climate change and its consequences is a familiar topic which is associated with natural disasters such as, flooding, hurricane, drought that cause water crisis and irreparable damages. Studying this phenomenon is a serious warning regarding the earth’s weather change for a long period of time.
Materials and Methods: In order to understand and survey the impacts of climate change on water resources, Global Circulation Models, GCMs, are used; their main role is analyzing the current climate and projecting the future climate. Climate change scenarios developing from GCMs are the initial source of information to estimate plausible future climate. For transforming coarse resolution outputs of the GCMs into finer resolutions influenced by local variables, there is a need for reliable downscaling techniques in order to analyze climate changes in a region. The classical statistical methods run the model and generate the future climate just with considering the time variable. Multi-site daily rainfall and temperature time series are the primary inputs in most hydrological analyses such as rainfall-runoff modeling. Water resource management is directly influenced by the spatial and temporal variation of rainfall and temperature. Therefore, spatial-temporal modeling of daily rainfall or temperature including climate change effects is required for sustainable planning of water resources.
Results and Discussion: For the first time, in this study by ASD model (Automated regression-based Statistical Downscaling tool) developed by M. Hessami et al., multi-site downscaling of temperature and precipitation was done with CGCM3.1A2 outputs and two synoptic stations (Rasht and Bandar Anzali) simultaneously by considering the correlations of multiple sites. The model can process conditionally on the occurrence of precipitation or unconditionally for temperature. Hence, the modeling of daily precipitation involves two steps: one step, precipitation occurrence and the other step precipitation amounts and the modeling of daily temperature is performed in one step. The choice of predictor variables is one of the most influential steps in the development of statistical downscaling scheme because the decision largely determines the character of the downscaling results. It is essential to remember that predictors relevant to the local predict and should be adequately reproduced by the host climate model at the spatial scales used to condition the downscaled response. To test this approach over the current period and to compare the results with observed data, temperature and precipitation, from 2 stations, model is evaluated and calibrated by using NCEP (National Center for Environmental Prediction) reanalysis data before the use of GCMs as input variables. Then climate was predicted for three periods which each period consist of thirty years in the future, 2011-2040, 2041-2070 and 2071-2100. ASD reduces the problem of predictor selection and it is capable of performing all steps of statistical downscaling automatically. In order to select dominant predictors in multi-site modeling backward stepwise regression method was used; so that some predictors like 850 hPa geopotential, 850 hPa specific humidity, 1000 hPa specific humidity, screen air temperature (2m) were the most important variables for temperature models, and 850 hPa geopotential, 1000 hPa zonal velocity, 1000 hpa specific humidity, and screen air temperature (2m) played a main role in temperature modeling. For downscaling of precipitation, the amount of explained variance (R2) is 0.336 for NCEP data and it is 0.89, 0.922 and 0.855for maximum, mean and minimum temperature, respectively. The results underlined certain limitations to downscale the precipitation regime and its strength compared to downscale the temperature regime. To evaluate the performance of the multi-site statistical downscaling approach, several climatic and statistical indices were developed. For instance, based on daily total precipitation, two precipitation indices was used including percentage of wet days, maximum number of consecutive dry days. The results showed the increase of the average temperature and precipitation decreases for future mainly. In this case study, the decrease of 0.30 mm day-1 in the average rainfall on the second period of future, 2041-2070, and the reduction of 7.25% on the 3rd period, 2071-2100, in climate index of the percentage of wet days was predicted, compared to the based period. However, the results illustrated an increase in the mean of maximum, mean, and minimum temperature, 3.17, 2.5 and 2.8 °C, for the 3rd period of future from 2071 to 2100, respectively.
Conclusions: The aim of applying this new method is to demonstrate the importance of multi-site models and developing a suitable context to calculate the uncertainty of climate predictions. Further works are needed to evaluate in depth the fundamental assumption of multi- site statistical downscaling, i.e. the stability of the relationships between predictors and predict and in altered climate and test their plausibility and consistency.

Keywords

ASD model; General Circulation Models; Predictand; Predictor

References

1- Gaitan C.F., Hsieh W.W. and Cannon A.J. 2014. Comparison of statistically downscaled precipitation in terms of future climate indices and daily variability for southern Ontario and Quebec, Canada. Climate Dynamics, 43:3201-3217.

2- Gangopadhyay S., Clark M., Rajagopalan B. 2005. Statistical downscalingusing K-nearest neighbors. Water Resources Research, 41:W0202:1-23.

3- HarphamC. and Wilby R.L. 2005. Multi-site downscaling of heavy daily precipitation occurrence and amounts. Journal of Hydrology, 312:235-255.

4- Hessami M., Gachon P., Quard, T.B.M.J. and St-Hailaire A. 2008. Automated regression-based Statistical Downscaling tool. Environmental Modeling and Software, 23(6):813-834.

5- IPCC. 2007. IPCC Fourth Assessment Report, Cambridge University Press.

6- Jeong D.I., St-Hilaire A., Ouarda T.B.M.J. and P. Gachon. 2013. Projection of future daily precipitation series and extreme events by using a multi-site statistical downscaling model over the great Montreal area, Quebec, Canada. Hydrology Research, 44.1:147-168.

7- Khalili M., Nguyen V. and Gachon P. 2013. A statistical approach to multi-site multivariate downscaling of daily extreme temperature series. International Journal of Climatology, 33:15-32.

8- Kim J.W., Change J.T., Baker N.L., Wilks D.S. and Gates W.L.1984. The statistical problem of climate inversion: determination of the relationship between local and large-scale climate. Monthly Weather Review, 112:2069-77.

9- Mehrotra R., Sharma A., Nagesh Kumar D.and Reshmidevi T.V. 2013. Assessing future rainfall projections using multiple GCMs and a multi-site stochastic downscaling model. Journal of Hydrology, 488:84-100.

10- Murphy J. 1999. An evaluation of statistical and dynamical techniques for downscaling local climate. Journal of Climate, 12:2256-2284.

11- Richardson C. 1981. Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resources Research, 17:182-190.

12- Liu W., Fu G., Liu C. and Charles S.P. 2012.A comparison of three multi-site statistical downscaling models for daily rainfall in the North China Plain. Theoretical and Applied Climatology, 113(3):585-600.

13- Wigley T.M.L., Jones P.D., Briffa K.R. and Smith G. 1990. Obtaining sub-grid-scale information from coarseresolution general circulation model output. Journal of Geophysical Research, 95:1943-1953.

14- Wilby R.L., Hay L.E. and LeavesleyG.H. 1999. A comparison of downscaled and raw GCM output: implications for climate change scenarios in the San Juan River basin, Colorado. Journal of Hydrology, 225:67-91.

15- Wilby R.L., Dawson C.W. and Barrow E.M. 2002.SDSM A decision support tool for the assessment of regional climate change impacts. Environment Model Software, 17(2):145-57.

16- World Meteorological Organization (WMO). 1988. General Meteorological Standards and Recommended Practices, Technical Regulations Basic documents,WMO: 49.

17- Lu Y. and Qin X.S. 2014. Multisite rainfall downscaling and disaggregation in tropical urban area. Journal of Hydrology, 509: 55-65.