Does the weighting of climate simulations result in a more reasonable quantification of hydrological impacts?

Abstract. With the increase in the number of available global climate models (GCMs), pragmatic questions come up when using them to quantify the climate change impacts on hydrology: Is it necessary to weight GCM outputs in the impact studies, and if so, how to weight them? Some weighting methods have been proposed based on the performances of GCM simulations with respect to reproducing the observed climate. However, the process from climate variables to hydrological responses is nonlinear, and thus the assigned weights based on their performances in climate simulations may not be translated to hydrological responses. Assigning weights to GCM outputs based on their ability to represent hydrological simulations is more straightforward. Accordingly, the present study assigns weights to GCM simulations based on their ability to reproduce hydrological characteristics and investigates their influence on the quantification of hydrological impacts. Specifically, eight weighting schemes are used to determine the weights of GCM simulations based on streamflow series simulated by a lumped hydrological model using raw or bias-corrected GCM outputs. The impacts of weighting GCM simulations are investigated in terms of reproducing the observed hydrological regimes for the reference period (1970–1999) and quantifying the uncertainty of hydrological changes for the future period (2070–2099). The results show that when using raw GCM outputs with no bias correction, streamflow-based weights better represent the mean hydrograph and reduce the bias of annual streamflow. However, when applying bias correction to GCM simulations before driving the hydrological model, the climate simulations become rather close to the observed climate, so that compared to equal weighting, the streamflow-based weights do not bring significant differences in the multi-model ensemble mean and uncertainty of hydrological impacts. Since bias correction has been an indispensable procedure in hydrological impact studies, the equal weighting method may still be a viable and conservative choice for the studies of hydrological climate change impacts.

[1]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[2]  G. Bastin,et al.  On the Accuracy of Areal Rainfall Estimation - a Case-study , 1987 .

[3]  S. Sorooshian,et al.  Effective and efficient global optimization for conceptual rainfall‐runoff models , 1992 .

[4]  D. Entekhabi,et al.  Observed Sacramento Basin streamflow response to precipitation and temperature changes and its relevance to climate impact studies , 1996 .

[5]  Alex J. Cannon,et al.  Recent Variations in Climate and Hydrology in Canada , 2000 .

[6]  K. Taylor.  Summarizing multiple aspects of model performance in a single diagram , 2001 .

[7]  F. Giorgi,et al.  Calculation of average, uncertainty range, and reliability of regional climate changes from AOGCM simulations via the reliability ensemble averaging (REA) method , 2002 .

[8]  C. Perrin,et al.  Improvement of a parsimonious model for streamflow simulation , 2003 .

[9]  M. Webb,et al.  Quantification of modelling uncertainties in a large ensemble of climate change simulations , 2004, Nature.

[10]  A. Raftery,et al.  Using Bayesian Model Averaging to Calibrate Forecast Ensembles , 2005 .

[11]  R. Wilby,et al.  A framework for assessing uncertainties in climate change impacts: Low‐flow scenarios for the River Thames, UK , 2006 .

[12]  C. Frei,et al.  Downscaling from GCM precipitation: a benchmark for dynamical and statistical downscaling methods , 2006 .

[13]  S. Sorooshian,et al.  Multi-model ensemble hydrologic prediction using Bayesian model averaging , 2007 .

[14]  A. Hense,et al.  Probabilistic climate change predictions applying Bayesian model averaging , 2007, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[15]  E. Maurer.  Uncertainty in hydrologic impacts of climate change in the Sierra Nevada, California, under two emissions scenarios , 2007 .

[16]  A. Pitman,et al.  Evaluation of the AR4 Climate Models’ Simulated Daily Maximum Temperature, Minimum Temperature, and Precipitation over Australia Using Probability Density Functions , 2007 .

[17]  Reto Knutti,et al.  The use of the multi-model ensemble in probabilistic climate projections , 2007, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[18]  Charles Doutriaux,et al.  Performance metrics for climate models , 2008 .

[19]  K. Taylor,et al.  Evaluating the present‐day simulation of clouds, precipitation, and radiation in climate models , 2008 .

[20]  T. Reichler,et al.  How Well Do Coupled Models Simulate Today's Climate? , 2008 .

[21]  R. Leconte,et al.  Uncertainty of the impact of climate change on the hydrology of a nordic watershed , 2008 .

[22]  D. Nychka,et al.  Spatial Analysis to Quantify Numerical Model Bias and Dependence , 2008 .

[23]  M. Hutchinson,et al.  Development and Testing of Canada-Wide Interpolated Spatial Models of Daily Minimum–Maximum Temperature and Precipitation for 1961–2003 , 2008 .

[24]  R. Toumi,et al.  Climate projections: Past performance no guarantee of future skill? , 2009 .

[25]  B. Santer,et al.  Incorporating model quality information in climate change detection and attribution studies , 2009, Proceedings of the National Academy of Sciences.

[26]  J. Vaze,et al.  Estimating climate change impact on runoff across southeast Australia: Method, results, and implications of the modeling method , 2009 .

[27]  F. Chiew,et al.  Influence of Rainfall Scenario Construction Methods on Runoff Projections , 2009 .

[28]  B. D. Santera,et al.  Incorporating model quality information in climate change detection and attribution studies , 2009 .

[29]  S. Somot,et al.  Weighted frequency distributions express modelling uncertainties in the ENSEMBLES regional climate experiments. , 2010 .

[30]  F. Giorgi,et al.  Weight assignment in regional climate models , 2010 .

[31]  Reto Knutti,et al.  Risks of Model Weighting in Multimodel Climate Projections , 2010 .

[32]  F. Giorgi,et al.  Upgrades to the reliability ensemble averaging method for producing probabilistic climate-change projections , 2010 .

[33]  Reto Knutti,et al.  Challenges in Combining Projections from Multiple Climate Models , 2010 .

[34]  D. Jacob,et al.  Validation of temperature trends in the ENSEMBLES regional climate model runs driven by ERA40. , 2010 .

[35]  R. Leconte,et al.  Overall uncertainty study of the hydrological impacts of climate change for a Canadian watershed , 2011 .

[36]  Karl E. Taylor,et al.  An overview of CMIP5 and the experiment design , 2012 .

[37]  J. Seibert,et al.  Bias correction of regional climate model simulations for hydrological climate-change impact studies: Review and evaluation of different methods , 2012 .

[38]  F. Su,et al.  Evaluation of the Global Climate Models in the CMIP5 over the Tibetan Plateau , 2013 .

[39]  B. Stevens,et al.  Climate and carbon cycle changes from 1850 to 2100 in MPI‐ESM simulations for the Coupled Model Intercomparison Project phase 5 , 2013 .

[40]  F. Brissette,et al.  Performance and uncertainty evaluation of empirical downscaling methods in quantifying the climate change impacts on hydrology over two North American river basins , 2013 .

[41]  C. Perrin,et al.  ‘As simple as possible but not simpler’: What is useful in a temperature-based snow-accounting routine? Part 2 – Sensitivity analysis of the Cemaneige snow accounting routine on 380 catchments , 2014 .

[42]  Reto Knutti,et al.  A Representative Democracy to Reduce Interdependency in a Multimodel Ensemble , 2015 .

[43]  Amir AghaKouchak,et al.  A methodology for deriving ensemble response from multimodel simulations , 2015 .

[44]  Yan Zhao.  Investigation of uncertainties in assessing climate change impacts on the hydrology of a Canadian river watershed , 2015 .

[45]  F. Brissette,et al.  A comparative analysis of 9 multi-model averaging approaches in hydrological continuous streamflow simulation , 2015 .

[46]  E. Alfaro,et al.  Skill of CMIP5 climate models in reproducing 20th century basic climate features in Central America , 2015 .

[47]  F. Brissette,et al.  Assessing the limits of bias‐correcting climate model outputs for climate change impact studies , 2015 .

[48]  R. Knutti,et al.  Skill and independence weighting for multi-model assessments , 2016 .

[49]  F. Brissette,et al.  Transferability of optimally-selected climate models in the quantification of climate change impacts on hydrology , 2016, Climate Dynamics.

[50]  Q. Duan,et al.  A nonstationary bias‐correction technique to remove bias in GCM simulations , 2016 .

[51]  Ruth Lorenz,et al.  A climate model projection weighting scheme accounting for performance and interdependence , 2017 .

[52]  F. Brissette,et al.  Impacts of weighting climate models for hydro-meteorological climate change studies , 2017 .

[53]  Karsten Lehmann,et al.  Selecting a climate model subset to optimise key ensemble properties , 2017 .

[54]  Alex J. Cannon,et al.  Transferability of climate simulation uncertainty to hydrological impacts , 2018, Hydrology and Earth System Sciences.

[55]  Chong-yu Xu,et al.  Bias nonstationarity of global climate model outputs: The role of internal climate variability and climate model sensitivity , 2018, International Journal of Climatology.

[56]  Prospects and Caveats of Weighting Climate Models for Summer Maximum Temperature Projections Over North America , 2018 .

[57]  Robert Pincus,et al.  ESD Reviews: Model dependence in multi-model climate ensembles: weighting, sub-selection and out-of-sample testing , 2019, Earth System Dynamics.

[58]  S. Hostetler,et al.  The Dependence of Hydroclimate Projections in Snow‐Dominated Regions of the Western United States on the Choice of Statistically Downscaled Climate Data , 2019, Water Resources Research.