1. Colorado River Basin streamflow projection under IPCC CMIP5 scenarios: from the global to basin scale using an integrated dynamic modeling approach Hsin-I Chang 1, Christopher Castro 1, 1 Department of Atmospheric Sciences University of Arizona Mar 28 th, 2014

2. Understanding uncertainties in future Colorado River streamflow (BAMS article, Jan 2014) Sources of climate projection uncertainty for CRB: 1.GCM and emission scenarios used 2.Spatial scale and topography dependency 3.How land surface hydrology represents precipitation and temperature change 4.Downscaling methodologies

3. Multi-model schematic: Regional climate simulations (25km resolution) Basin-scale simulations (1/8 degree resolution) Dynamical downscaling Bias correction Global Climate projections (1 to 2.5° degree resolution)

4. % Regional Climate Research using IPCC CMIP3 and CMIP5 climate projections CMIP3: NARCCAP (North American Regional Climate Change Assessment Program) – Time slice simulations: [1971- 2010], [2041-2070] – 50km in resolution CMIP5: CORDEX NA (Coordinated Regional climate Downscaling Experiment, North America) 25km resolution: Continuous simulations [100+ years) 12km resolution: time-slice simulations

5. Climate-Hydrology Projection Research (DOI) Objective: Characterize how the changing climate affects seasonal precipitation and streamflow projections in the Colorado River basins – Use the newest climate projections [IPCC CMIP5] that has good 20 th century climatology Research Question: How climate trends (mean and extremes) may change in the future, to anticipate worst-case scenarios in long-term water resource planning.

6. Complete CMIP5 model list ModelCenterAtmospheric Horizontal Resolution (lon. x lat.) Number of model levels Reference ACCESS1-0Commonwealth Scientific and Industrial Research Organization/Bureau of Meteorology, Australia 1.875 x 1.2538Bi et al. (2012) BCC-CSM1.1*Beijing Climate Center, China Meteorological Administration, China 2.8 x 2.826Xin et al. (2012) CanCM4Canadian Centre for Climate Modelling and Analysis, Canada 2.8 x 2.835Chylek et al. (2011) CanESM2*Canadian Center for Climate Modeling and Analysis, Canada 2.8 x 2.835Arora et al. (2011) CCSM4*National Center for Atmospheric Research, USA1.25 x 0.9426Gent et al. (2011) CESM1-CAM5-1- FV2 Community Earth System Model Contributors (NSF- DOE- NCAR) 1.4 x 1.426Gent et al. (2011) CNRM-CM5.1*National Centre for Meteorological Research, France1.4 x 1.431Voldoire et al. (2011) CSIRO-MK3.6*Commonwealth Scientific and Industrial Research Organization/Queensland Climate Change Centre of Excellence, AUS 1.8 x 1.818Rotstayn et al. (2010) EC-EARTHEC-EARTH consortium1.125 x 1.1262Hazeleger et al. (2010) FGOALS-S2.0LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences 2.8 x 1.626Bao et al. (2012) GFDL-CM3*NOAA Geophysical Fluid Dynamics Laboratory, USA2.5 x 2.048Donner et al. (2011) GFDL-ESM2G/M*NOAA Geophysical Fluid Dynamics Laboratory, USA2.5 x 2.048Donner et al. (2011) GISS-E2-H/R*NASA Goddard Institute for Space Studies, USA2.5 x 2.040Kim et al. (2012) HadCM3*Met Office Hadley Centre, UK3.75 x 2.519Collins et al. (2001) HADGEM2-CC (Chemistry coupled) Met Office Hadley Centre, UK1.875 x 1.2560Jones et al. (2011) HadGEM2-ES*Met Office Hadley Centre, UK1.875 x 1.2560Jones et al. (2011) INMCM4*Institute for Numerical Mathematics, Russia2 x 1.521Volodin et al. (2010) IPSL-CM5A-LR*Institut Pierre Simon Laplace, France3.75 x 1.8 39Dufresne et al. (2012) IPSL-CM5A-MRInstitut Pierre Simon Laplace, France2.5 x 1.2539Dufresne et al. (2012) MIROC4hAtmosphere and Ocean Research Institute (The University of Tokyo), National Institute for Environmental Studies, and Japan Agency for Marine- Earth Science and Technology, Japan 0.56 x 0.5656Sakamoto et al. (2012) MIROC5*Atmosphere and Ocean Research Institute (The University of Tokyo), National Institute for Environmental Studies, and Japan Agency for Marine-Earth Science and Technology, Japan 1.4 x 1.440Watanabe et al. (2010) MIROC-ESM*Japan Agency for Marine-Earth Science and Technology, Atmosphere and Ocean Research Institute (The University of Tokyo), and National Institute for Environmental Studies 2.8 x 2.880Watanabe et al. (2010) MIROC-ESM- CHEM Japan Agency for Marine-Earth Science and Technology, Atmosphere and Ocean Research Institute (The University of Tokyo), and National Institute for Environmental Studies 2.8 x 2.880Watanabe et al. (2010) MPI-ESM-LR*Max Planck Institute for Meteorology, Germany1.9 x 1.947Zanchettin et al. (2012) MRI-CGCM3*Meteorological Research Institute, Japan1.1 x 1.148Yukimoto et al. (2011) NorESM1-M*Norwegian Climate Center, Norway2.5 x 1.926Zhang et al. (2012)

7. ModelRMSE (mm day -1 ) BCC-CSM1-11.92 CCSM41.53 CNRM-CM51.29 CSIRO-Mk31.09 CanESM20.44 GFDL-CM31.54 GFDL-ESM2M1.72 GISS-E2-R1.46 HadCM30.63 HadGEM2-ES0.75 INMCM41.11 IPSL-CM5A-LR0.99 MIROC-ESM1.32 MIROC51.58 MPI-ESM-LR1.09 MRI-CGCM32.08 NorESM1-M1.96 Annual mean RMSE for precipitation: 17 core CMIP5 models vs CMAP observed estimates for NAM region Sheffiled et al. 2013

8. Regional Climate Experimental Design Weather Research and Forecasting model (WRF) – Forcing from IPCC CMIP5(2) datasets – 25km resolution (CORDEX North America domain) Two 100+ yr continuous simulation (CMIP5, U.S. and Mexico) – 10km resolution (Southwest U.S.) 2x2 10-yr simulations (WRF-CMIP5) Higher resolution (~ 2km) runs will be considered for Colorado Headwaters domain

9. Preliminary Results (CMIP3): Regional Climate and Streamflow analysis

10. Dry Gets Drier and Wet Gets Wetter Hypothesis: : Increases in warm season precipitation and temperature extremes will be enhanced by natural variability. Dry Gets Drier and Wet Gets Wetter Trend in Global Monsoon Precipitation: Wang et al. 2012: “….. enhanced global summer monsoon not only amplifies the annual cycle of tropical climate but also promotes directly a ‘‘wet – gets – wetter’’ trend pattern and indirectly a ‘‘dry – gets – drier’’ trend pattern through coupling with deserts and trade winds.” Hsu et al. 2011: “results suggest that in the past 30 years with an increase in the global mean surface temperature, the global monsoon total precipitation is strengthened.

11. Interannual variability: Teleconnections at monsoon onset (late June, early July) The onset and variability of North American Monsoon System (NAMS) is partly controlled by warm season atmospheric teleconnections Teleconnections driven El Niño Southern Oscillation (ENSO) and Pacific Decadal Variability (PDV) Influence monsoon ridge positioning in early summer. Other drivers of natural variability: Atlantic Mutidecadal Oscillation (AMO), Indian monsoon, antecedent land surface conditions Castro et al. (2001)

12. Early warm season precipitation significantly related to global sea surface temperature anomalies (CMIP3) Climate control period 1950-2000 Climate change period 2000-2040 Average of dominant JJ EOFs with a significant relationship to global SST Regression of mode on global SSTA

13. Positive ENSO-PDV phase (El Nino) Dry SW monsoon Negative ENSO-PDV phase (La Nina) Wet SW monsoon Precipitation Extremes Anomaly(following ENSO signal) CPC: (1981-2010)-(1950-1980) Anti-phase relationship in precipitation variability between the Southwest U.S. and central U.S. is also found in both precipitation climatology and extreme anomaly trend Observed trends in precipitation anomaly is following the natural variability of ENSO signal. Wet – gets – wetter and dry – gets – drier

14. Precipitation Extremes Anomaly (Positive ENSO) WRF-MPI: (2001-2040)-(1950-2000) Obs: (1981-2010)-(1950-1980)

15. Projected Southwest drying trend is not as dire in AR5 Mean-Annual Precipitation Change, percent CMIP3, 1970-1999 to 2070-2099, 50%tile Mean-Annual Precipitation Change, percent CMIP5 - CMIP3, 1970-1999 to 2070-2099, 50%tile Mean-Annual Precipitation Change, percent CMIP5, 1970-1999 to 2070-2099, 50%tile IPCC CMIP3 vs CMIP5 projections for the Southwest