1. Introduction
Water is an essential resource for human societies, and climatic conditions limit the turnover rate of available renewable freshwater resources (RFWR). Although current global freshwater withdrawals are well below the upper limit for sustainable use, more than two billion people live in highly water-stressed areas (Oki and Kanae 2006). Water resource engineers consider surface water and groundwater as water resources exploitable by withdrawal and evapotranspiration as a loss of water from precipitated water. Precipitation minus evapotranspiration (PME) over the land can therefore be considered a measure of the maximum available RFWR (Nakaegawa et al. 2007). Satellite data and global climate models have been used to estimate PME (Swenson and Wahr 2006; Trenberth et al. 2007) and to investigate the potential predictability of PME seasonality (Nakaegawa et al. 2003). Because the global water demand is projected to increase even further because of expected increases of the global population and expansion of economic activity (Shiklomanov 2000), accurate projections of PME are essential to the informed assessment of future impacts.
Knowledge of the impacts of global warming on PME is essential for these projections because the water cycle is expected to change under a warmer climate regime (Meehl et al. 2007). Although research related to the impact of global warming and adaptation requires climate simulations based on a wide range of emission scenarios, generation of such climate scenarios with atmospheric–ocean general circulation models (AOGCMs) requires large computer resources. Some of the impact research being carried out for the Intergovernmental Panel on Climate Change Fifth Assessment Report will use so-called pattern scaling, which can produce climate scenarios under a wide range of emission scenarios with fewer computer resources than those required by AOGCMs (Mitchell et al. 1999; Schlesinger et al. 2000).
In pattern scaling, an AOGCM combined with a specified emission scenario is first used to estimate the relationship between the normalized spatial pattern of certain climate variables and global mean surface air temperatures (SATs). A simple climate model with a very wide range of emission scenarios is then used to estimate global mean SATs. Finally, changes in regional climates are estimated for a very wide range of emission scenarios by multiplying the global mean SAT changes by the scaling pattern of the specified climate variable. The basic assumption of pattern scaling is that the scaling pattern is the same for all emission scenarios (Shiogama et al. 2010a).
Both greenhouse gases and aerosols influence the water cycle. Aerosol particles absorb and scatter incoming solar radiation and thereby reduce the downward flux of short-wavelength radiation at the surface (Ramanathan et al. 2001; Ramanathan and Carmichael 2008). Aerosols also influence the microphysical properties of clouds by acting as cloud condensation nuclei, and modifications to the microphysical properties of clouds can lead to enhancement of their albedo and lifetime, the first and second indirect effects, respectively. These first and second indirect effects lead to reductions in downward fluxes of short-wavelength radiation at the surface (Lohmann and Feichter 2005). These reductions result in regional-scale decreases of evapotranspiration because of decreases in the energy available for evapotranspiration at the surface. A consequence of the decrease of evapotranspiration is a decrease in precipitation.
Previous research has indicated that differences in aerosol loadings between emission scenarios induce a scenario dependence (SD) of scaling patterns of precipitation and evapotranspiration on both a global (Shiogama et al. 2010a; Frieler et al. 2011) and a regional scale (Ishizaki et al. 2013). The SDs of precipitation and evapotranspiration scaling patterns may induce an SD in the scaling pattern of PME. In addition, it can be assumed that precipitation minus evapotranspiration is in balance with the convergence of vertically integrated atmospheric moisture on a long time scale. We thus investigated the SD of the scaling pattern of PME and its causes, with a focus on the SD of the changes of aerosol loadings and large-scale atmospheric circulation per 1 K increase in global mean SAT.
2. Model description
We analyzed the results of five models that were part of the fifth phase of the Coupled Model Intercomparison Project (CMIP5): Community Climate System Model, version 4 (CCSM4; Meehl et al. 2012); Commonwealth Scientific and Industrial Research Organisation Mark, version 3.6.0 (CSIRO Mk3.6.0; Rotstayn et al. 2012); L’Institut Pierre-Simon Laplace Coupled Model, version 5, coupled with NEMO, low resolution (IPSL-CM5A-LR; Hourdin et al. 2012); Model for Interdisciplinary Research on Climate, version 5 (MIROC5; Watanabe et al. 2010); and Max Planck Institute Earth System Model, low resolution (MPI-ESM-LR; Brovkin et al. 2013). Because a simulated hydrological cycle is sensitive to internal variability in the climate system, we selected five models that had at least three ensemble members each and analyzed the ensemble means for each model. All of the models can treat direct and indirect effects of carbonaceous aerosols (black carbon and organic matter) and sulfate aerosols. The emission scenario in each representative concentration pathway (RCP) specifies the anthropogenic aerosol emissions. We estimated the scaling patterns by using features of three different emission scenarios: RCP2.6 (van Vuuren et al. 2007), RCP4.5 (Wise et al. 2009), and RCP8.5 (Riahi et al. 2007), and twentieth-century simulations. Because a simulated hydrological cycle is sensitive to internal variability of the climate system, we did not use RCP6.0 (Hijioka et al. 2008), because only one ensemble member is available for that scenario in MIROC5.
3. Method of estimating scaling patterns
4. Results and discussion
a. Scaling patterns of PME in each RCP in MIROC5
The change of annual PME per 1 K change in global mean SAT in each RCP is projected to increase over high latitudes of the Northern Hemisphere, Indonesia, central Africa, and the Asian monsoon region and to decrease over the southwestern part of North America, Central America, and the southern part of Europe (EU; Fig. 1). These results are consistent with previous research on future changes in total runoff (Nohara et al. 2006) and PME (Christensen et al. 2007). Although the general features of the scaling patterns are similar for all of the RCPs, an F test revealed that the SD of the PME scaling pattern was statistically significant (p = 0.05) for 1) RCP8.5 and RCP2.6 over EU, the Congo River Basin (CRB), and East Asia (EA; Fig. 2a) and 2) RCP8.5 and RCP4.5 over the Maritime Continent (MC) and the CRB (Fig. 2b).
Scaling patterns (P, per 1 K global mean warming) of annual mean PME (mm day−1 K−1) in (a) RCP2.6, (b) RCP4.5, and (c) RCP8.5.
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
Scaling pattern differences in annual mean PME (mm day−1 K−1) between (a) RCP8.5 − RCP2.6 and (b) RCP8.5 − RCP4.5. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
b. The causes of the SD of PME for RCP8.5 and RCP2.6 in MIROC5
As technological advances lead to a reduction of emissions of anthropogenic aerosols in the future in both RCPs, the direct and indirect effects of aerosols decrease. However, the fractional decreases of anthropogenic aerosol emissions per 1 K change in global mean SAT are smaller in magnitude in RCP8.5 than in RCP2.6 over EU, EA, and CRB (Figs. 3a–c) because the projected global mean SAT is higher in RCP8.5 than in RCP2.6. The decrease of direct and indirect effects of aerosols per 1 K change in global mean SAT is thus smaller in RCP8.5 than in RCP2.6. As a result, the decrease of the net downward surface short-wavelength radiation flux per 1 K change in global mean SAT [(ΔS↓/S)/ΔT] is smaller in RCP8.5 than in RCP2.6 over EU and EA (Fig. 4a). In previous research (Ishizaki et al. 2012b), we attributed the cause of the scenario dependency of (ΔS↓/S)/ΔT to the scenario dependency of anthropogenic aerosol scaling patterns and confirmed that the impacts of land use change were not large. This SD of surface short-wavelength scaling patterns due to the SD of aerosol scaling patterns leads to smaller fractional increases of evapotranspiration per 1 K change in global mean SAT [(ΔE/E)/ΔT] in RCP8.5 than in RCP2.6 over the EU and EA (Fig. 5b).
Scaling pattern differences in (a) annual mean sulfate aerosols (10−6 kg m−2 K−1), (b) annual mean black carbon (10−7 kg m−2 K−1), and (c) annual mean organic carbon (10−6 kg m−2 K−1) between RCP8.5 − RCP2.6; and in (d) annual mean sulfate aerosols (10−6 kg m−2 K−1), (e) annual mean black carbon (10−7 kg m−2 K−1), and (f) annual mean organic carbon (10−6 kg m−2 K−1) between RCP8.5 − RCP4.5. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
Scaling pattern differences in annual mean net downward surface short-wavelength radiation (W mm−2 K−1) between (a) RCP8.5 − RCP2.6 and (b) RCP8.5 − RCP4.5. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
Scaling pattern differences in (a) annual mean precipitation and (b) annual mean evapotranspiration between RCP8.5 − RCP2.6; and in (c) annual mean precipitation and (d) annual mean evapotranspiration between RCP8.5 − RCP4.5. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
The smaller [(ΔE/E)/ΔT] value results in smaller fractional increases of precipitation per 1 K change in global mean SAT [(ΔP/P)/ΔT] in RCP8.5 than in RCP2.6 over EA and its precipitation source region (North Pacific Ocean; Fig. 5a). The SD of the evapotranspiration scaling pattern over some part of the precipitation source region causes the SD of the precipitation scaling pattern to have a greater effect than the SD of the evapotranspiration scaling pattern over the downstream landmass regions in EA (the Korean Peninsula and the region around the Yellow River). This is responsible for the SD of the scaling pattern of PME over that region.
The fractional decreases of black and organic carbon aerosol emissions per 1 K change in global mean SAT are smaller in magnitude in RCP8.5 than in RCP2.6 over the CRB. As a result, (ΔS↓/S)/ΔT is also smaller in magnitude in RCP8.5 than in RCP2.6. However, (ΔP/P)/ΔT is larger in RCP8.5 than in RCP2.6 (Fig. 5a). Thus, the SD of PME cannot be explained by the SD of black and organic carbon aerosols over CRB. Walker circulation is projected to weaken as global warming proceeds (Vecchi and Soden 2007). The sensitivity of African Walker circulation to global warming is significantly different (p < 0.05) between the RCPs over the CRB (Fig. 6a). As a result, the SD in African Walker circulation may induce a SD in the scaling pattern of PME over the CRB.
Scaling pattern differences of 500-hPa vertical velocities (hPa day−1 K−1) over central Africa between (a) RCP8.5 − RCP2.6 and (b) RCP8.5 − RCP4.5.
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
The northern part of the North Atlantic Ocean warmed more in RCP8.5 than in RCP2.6 because of the nonlinear response of the Atlantic meridional overturning circulation (AMOC) to global warming in MIROC5 (Ishizaki et al. 2012b). The nonlinear response of the AMOC increased evaporation more in RCP8.5 than in RCP2.6 over the northern part of the North Atlantic Ocean (Ishizaki et al. 2013, their Fig. 4c). As a result, there were cyclonic anomalies between RCP8.5 and RCP2.6 in the changes in water vapor transport per 1 K increase of global mean SAT, and the inflow of water vapor from the North Atlantic Ocean to Europe was greater in RCP8.5 than in RCP2.6 (Ishizaki et al. 2013, their Fig. 6). At equilibrium, precipitation minus evapotranspiration is balanced with the time-averaged and vertically integrated water convergence (Joshi et al. 2008). Thus, an SD of PME occurs over the EU.
In contrast, the SD of evapotranspiration scaling patterns was less in RCP8.5 than in RCP2.6 because of the SD of sulfate aerosol scaling patterns. The signs of the two effects are opposite, and these two effects cancel each other out. Thus, a scenario dependence of the precipitation scaling pattern does not occur over EU.
c. The causes of the SD of PME between RCP8.5 and RCP4.5 in MIROC5
The SD of the precipitation scaling pattern is significant (p < 0.05) over the MC and CRB for RCP8.5 and RCP4.5 (Fig. 5c). The SD of (ΔS↓/S)/ΔT is also significant for RCP8.5 and RCP4.5 (Fig. 4b) over the MC and the surrounding oceans because of the SD of organic and black carbon aerosol emissions (Figs. 3e,f). Consequently, the SD of the evapotranspiration scaling pattern over the MC and the surrounding oceans is significant (p < 0.05) for RCP8.5 and RCP4.5 as well (Fig. 5d). Because the MC is a region where water vapor converges from the surrounding oceans (Ishizaki et al. 2012c), the SD of the precipitation scaling pattern over the MC is significant (p < 0.05), and the magnitude of the SD of the precipitation scaling pattern is larger than the SD of the scaling pattern of evapotranspiration. The result is a SD of the scaling pattern of PME over the MC.
The SD of the precipitation scaling pattern is also significant (p < 0.05) for RCP8.5 and RCP4.5 over the CRB (Fig. 6b) for the same reason that the SD of the precipitation scaling pattern is significant for RCP8.5 and RCP2.6. An SD of Walker circulation for RCP8.5 and RCP4.5 induces an SD of the PME scaling pattern.
d. Consistencies of the causes of the SD of PME among AOGCMs
Over the CRB, we found an SD of PME scaling patterns between RCP8.5 and RCP2.6 in multimodel (Fig. 7a) means as well as in MIROC5. The SD over the CRB was consistent among models (Fig. 8). We also found an SD of PME scaling patterns between RCP8.5 and RCP2.6 over the EU in multimodel ensemble means, although the SD was smaller than that of MIROC5 and insignificant over central parts of EU. A small SD of the PME scaling pattern between RCP8.5 and RCP2.6 was significant over EA (Fig. 7a) as well. Overall, the SDs of the PME scaling pattern due to the SDs of sulfate aerosol scaling patterns tended to be insignificant in multimodel ensemble means, because the SDs of precipitation and evapotranspiration scaling patterns tended to cancel each other out (Figs. 7b,c).
Scaling pattern differences of multimodel means in annual mean (a) PME (mm day−1 K−1), (b) precipitation (mm day−1 K−1), (c) evapotranspiration (mm day−1 K−1) between RCP8.5 − RCP2.6. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
Scaling pattern differences of annual mean PME (mm day−1 K−1) between RCP8.5 − RCP2.6 for (a) CCSM4, (b) CSIRO Mk3.6.0, (c) IPSL, (d) MPI. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
In addition to the CRB and some parts of EU, the SD of PME scaling patterns between RCP8.5 and RPC2.6 was significant over Central America, although a SD was not seen in MIROC5. Precipitation over most parts of Central America was projected to decrease (Christensen et al. 2007) because of a northward expansion and weakening of Hadley circulation under a global warming scenario (Meehl et al. 2007). However, the magnitude of the decrease of precipitation per 1 K change in global mean SAT over this region depends on RCPs (Fig. 7b). The SD of the changes in the northward expansion and weakening of Hadley circulation per 1 K change in global mean SAT resulted in an SD of the PME scaling pattern over most parts of Central America.
The significant SD of PME between RCP8.5 and RCP2.6 in multimodel ensemble means was apparent over the MC, although the magnitude of the SD was relatively weak in MIROC5. Precipitation was projected to increase in all of the RCPs over the MC (figures not shown). However, the magnitudes of the increases of precipitation per 1 K increase in global mean SAT differed significantly over the MC between RCP8.5 and RCP2.6 (Fig. 7b). The weakening of Walker circulation may be dependent on emission scenarios, and it may have led to the SD of the PME scaling pattern over the MC. An investigation of the causes of the SD of the weakening of Walker circulation per 1 K increase in global mean SAT is needed in future studies.
The SDs of the PME scaling patterns between RCP8.5 and RCP4.5 were also apparent in the multimodel ensemble means over the CRB and MC (Fig. 9a). The SD of the PME scaling patterns over the CRB seems to be due to the SD of African Walker circulation. The SD of the evapotranspiration scaling pattern over the MC and its surrounding regions, which seems to be due to the SD of the organic carbon aerosol scaling pattern, was apparent in multimodel ensemble means (Fig. 9c). This led to the SD of the precipitation scaling pattern over the MC (Fig. 9b). As a result, the SDs of PME scaling patterns between RCP8.5 and RCP4.5 over the MC were significant in the multimodel ensemble means as well, although the magnitudes and signs were model dependent (Fig. 10).
Scaling pattern differences of multimodel means in annual mean (a) PME (mm day−1 K−1), (b) precipitation (mm day−1 K−1), and (c) evapotranspiration (mm day−1 K−1) between RCP8.5 −RCP4.5. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
Scaling pattern differences of annual mean PME (mm day−1 K−1) between RCP8.5 − RCP4.5 for (a) CCSM4, (b) CSIRO Mk3.6.0, (c) IPSL, (d) MPI. Dotted regions are statistically significant at the p = 0.05 level (F test).
Citation: Journal of Hydrometeorology 15, 1; 10.1175/JHM-D-12-0114.1
5. Summary
Because PME over land can be considered to be a measure of the maximum available RFWR, the projection of PME is very important for risk assessments of future water resources. In this study, we investigated the SD of the scaling pattern of PME among RCPs. Previous research has indicated that a SD of anthropogenic aerosol scaling patterns induces a SD of precipitation and evapotranspiration (Ishizaki et al. 2013). The SD of precipitation and evapotranspiration may induce a SD of PME. In addition, it can be assumed that precipitation minus evapotranspiration is in balance with the convergence of vertically integrated atmospheric moisture on a long time scale. We therefore investigated the SD of PME, with a focus on the SD of the changes of aerosol loadings and large-scale atmospheric circulation per 1 K increase in global mean SAT.
A significant SD of the PME scaling pattern was found over the CRB, over some parts of EU and EA, over Central America between RCP8.5 and RCP2.6, and over the CRB and MC between RCP8.5 and RCP4.5. Water stress is projected to increase in the future because of population increases and economic development in the CRB, some parts of EU, and the MC (Vörösmarty et al. 2000) and is already relatively severe under current climate conditions in EU and Central America (Oki and Kanae 2006). Serious errors may confound projections of the impact of climate change unless those projections take into consideration the SD of the PME scaling pattern.
The SD of the PME scaling pattern was induced mainly by the SD and nonlinear response of large-scale atmospheric circulation changes (Hadley circulation and Walker circulation) and oceanic changes (AMOC) per 1 K increase in global mean SAT. In contrast, when a SD of anthropogenic aerosol scaling patterns occurred, a significant (p < 0.05) SD in PME tended to be found because of suppression of increases in evapotranspiration, and as a result, there was a suppression of increases in precipitation. However, the SD of the scaling pattern of PME tended to be insignificant because the SDs of both precipitation and evapotranspiration tended to cancel each other.
Because the differences of the global mean SAT changes between RCP8.5 and RCP4.5 are smaller than those between RCP8.5 and RCP2.6, the SDs of anthropogenic aerosol scaling patterns and the sensitivity of Walker circulation, Hadley circulation, and AMOC to global warming tended to be less between RCP8.5 and RCP4.5 than between RCP8.5 and RCP2.6. Thus, when PME pattern scaling is applied to an emission scenario, it would be better to reduce the SD by using the RCP that projects global mean SAT changes similar to those projected by the emission scenario. We could not find a large impact of land use changes on PME scaling patterns. However, the SD of land use change may induce a SD of PME scaling patterns on a different time scale or a different spatial scale.
In this study, we did not consider the impacts of human activities, although such consideration is fundamentally important for risk assessments of water resources (Oki et al. 2003; Hanasaki et al. 2008a,b). In fact, Pokhrel et al. (2012) indicated that the incorporation of anthropogenic water regulation modules can significantly improve the simulation of low flow in most of the heavily regulated river basins such as the Colorado and Missouri. Further investigation of pattern scaling with a focus on human activities is also needed.
In addition, a simulated hydrological cycle is distinctly sensitive to the representation of topography (Giorgi and Marinucci 1996). Regional climate models can reproduce orographic precipitation and snowpack over complex mountainous regions more accurately than AOGCMs because of their more detailed representation of topography (Leung and Qian 2003; Rummukainen 2010; Ishizaki et al. 2012a). It would be better to use the results of future projections by RCMs over complex mountainous regions.
Investigations of the impacts of aerosol sensitivity experiments that remove aerosol emissions from some regions or remove some kinds of aerosols (e.g., Shindell et al. 2012; Shiogama et al. 2010b) would be ideal, although these kinds of sensitivity studies require huge computer resources. In the future, similar sensitivity experiments should be conducted. In addition, modified approaches based on a statistical analysis of anthropogenic aerosols, as suggested by Frieler et al. (2012), would be required. Averaging limited numbers of ensemble members in each model may not remove the influences of internal variability in the climate system for some AOGCMs (Tebaldi et al. 2011). Some of the dependencies may be caused by internal variability. Furthermore, we used only five GCMs from the CMIP5. Our findings may be model dependent. Thus, dependencies of the scaling pattern of PME on the other AOGCMs must be investigated in future work.
Acknowledgments
This work was supported by the Global Environment Research Fund (S-10) of the Ministry of the Environment, Japan. We acknowledge the World Climate Research Programme’s Working Group on Coupled Modeling, which is responsible for CMIP, and we thank the climate modeling groups for producing and making available their model output. Comments by two anonymous reviewers and the editor are highly appreciated.
APPENDIX
Statistical Test for the Differences of the Slopes of Regression Lines through the Origin by an F Test
REFERENCES
Brovkin, V., and Coauthors, 2013: Effect of anthropogenic land-use and land-cover changes on climate and land carbon storage in CMIP5 projections for the twenty-first century. J. Climate,26, 6859–6881, doi:10.1175/JCLI-D-12-00623.1.
Christensen, J. H., and Coauthors, 2007: Regional climate projections. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 847–940.
Frieler, K., Meinshausen M. , von Deimling T. S. , Andrews T. , and Forster P. , 2011: Changes in global-mean precipitation in response to warming, greenhouse gas forcing and black carbon. Geophys. Res. Lett., 38, L04702, doi:10.1029/2010GL045953.
Frieler, K., Meinshausen M. , Golly A. , Mengel M. , van der Merwe K. , Donner S. , and Hoegh-Guldberg O. , 2012: Limiting global warming to 2°C is unlikely to save most coral reefs. Nat. Climate Change, 3, 165–170, doi:10.1038/nclimate1674.
Giorgi, F., and Marinucci M. R. , 1996: An investigation of the sensitivity of simulated precipitation to model resolution and its implications for climate studies. Mon. Wea. Rev., 124, 148–166, doi:10.1175/1520-0493(1996)124<0148:AIOTSO>2.0.CO;2.
Hanasaki, N., Kanae S. , Oki T. , Masuda K. , Motoya K. , Shirakawa N. , Shen Y. , and Tanaka K. , 2008a: An integrated model for the assessment of global water resources—Part 1: Model description and input meteorological forcing. Hydrol. Earth Syst. Sci., 12, 1007–1025, doi:10.5194/hess-12-1007-2008.
Hanasaki, N., Kanae S. , Oki T. , Masuda K. , Motoya K. , Shirakawa N. , Shen Y. , and Tanaka K. , 2008b: An integrated model for the assessment of global water resources—Part 2: Applications and assessments. Hydrol. Earth Syst. Sci., 12, 1027–1037, doi:10.5194/hess-12-1027-2008.
Hijioka, Y., Matsuoka Y. , Nishimoto H. , Masui M. , and Kainuma M. , 2008: Global GHG emissions scenarios under GHG concentration stabilization targets. J. Global Environ. Eng., 13, 97–108.
Hourdin, F., and Coauthors, 2012: LMDZ5B: The atmospheric component of the IPSL climate model with revisited parameterizations for clouds and convection. Climate Dyn., 40, 2193–2222, doi:10.1007/s00382-012-1343-y.
Ishizaki, Y., Nakaegawa T. , and Takayabu I. , 2012a: Validation of precipitation over Japan during 1985–2004 simulated by three regional climate models and two multi-model ensemble means. Climate Dyn., 39, 185–206, doi:10.1007/s00382-012-1304-5.
Ishizaki, Y., and Coauthors, 2012b: Temperature scaling pattern dependence on representative concentration pathway emission scenarios. Climatic Change, 112, 535–546, doi:10.1007/s10584-012-0430-8.
Ishizaki, Y., Yoshimura K. , Kanae S. , Kimoto M. , Kurita N. , and Oki T. , 2012c: Interannual variability of H2 18O in precipitation over the Asian monsoon region. J. Geophys. Res., 117, D16308, doi:10.1029/2011JD015890.
Ishizaki, Y., and Coauthors, 2013: Dependence of precipitation scaling patterns on emission scenarios for representative concentration pathways. J. Climate, 26, 8868–8879, doi:10.1175/JCLI-D-12-00540.1.
Joshi, M. M., Gregry J. M. , Webb M. J. , Sexton D. M. H. , and Johns T. C. , 2008: Mechanisms for the land/sea warming contrast exhibited by simulations of climate change. Climate Dyn., 30, 455–465, doi:10.1007/s00382-007-0306-1.
Leung, L. R., and Qian Y. , 2003: The sensitivity of precipitation and snowpack simulations to model resolution via nesting in regions of complex terrain. J. Hydrometeor., 4, 1025–1043, doi:10.1175/1525-7541(2003)004<1025:TSOPAS>2.0.CO;2.
Lohmann, U., and Feichter J. , 2005: Global indirect aerosol effects: A review. Atmos. Chem. Phys., 5, 715–737, doi:10.5194/acp-5-715-2005.
Meehl, G. A., and Coauthors, 2007: Global climate projections. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 747–845.
Meehl, G. A., and Coauthors, 2012: Climate system response to external forcings and climate change projections in CCSM4. J. Climate, 25, 3661–3683, doi:10.1175/JCLI-D-11-00240.1.
Mitchell, J. F. B., Johns T. C. , Eagles M. , Ingram W. J. , and Davis R. A. , 1999: Towards the construction of climate change scenarios. Climatic Change, 41, 547–581, doi:10.1023/A:1005466909820.
Nakaegawa, T., Sugi M. , and Matsumaru K. , 2003: A long-term numerical study of the potential predictability of seasonal mean fields of water resource variables using MRI/JMA-AGCM. J. Meteor. Soc. Japan, 81, 1041–1056, doi:10.2151/jmsj.81.1041.
Nakaegawa, T., Kusunoki S. , Sugi M. , Kitoh A. , Kobayashi C. , and Takano K. , 2007: A study of dynamical seasonal prediction of potential water resources based on an atmospheric GCM experiment with prescribed sea-surface temperature. Hydrol. Sci. J., 52, 152–165, doi:10.1623/hysj.52.1.152.
Nohara, D., Kitoh A. , Hosaka M. , and Oki T. , 2006: Impact of climate change on river discharge projected by multimodel ensemble. J. Hydrometeor.,7, 1076–1089, doi:10.1175/JHM531.1.
Oki, T., and Kanae S. , 2006: Global hydrological cycles and world water resources. Science, 313, 1068–1072, doi:10.1126/science.1128845.
Oki, T., Agata Y. , Kanae S. , Saruhashi T. , and Musiake K. , 2003: Global water resources assessment under climatic change in 2050 using TRIP. IAHS Publ.,280, 124–133.
Pokhrel, Y., Hanasaki N. , Koirala S. , Cho J. , Kim H. , Yeh P. J.-F. , Kanae S. , and Oki T. , 2012: Incorporating anthropogenic water regulation modules into a land surface model. J. Hydrometeor., 13, 255–269, doi:10.1175/JHM-D-11-013.1.
Ramanathan, V., and Carmichael G. , 2008: Global and regional climate changes due to black carbon. Nat. Geosci., 1, 221–227, doi:10.1038/ngeo156.
Ramanathan, V., Crutzen P. J. , Kiehl J. T. , and Rosenfeld D. , 2001: Aerosols, climate, and the hydrological cycle. Science, 294, 2119–2124, doi:10.1126/science.1064034.
Riahi, K., Gruebler A. , and Nakicenovic N. , 2007: Scenarios of long-term socio-economic and environmental development under climate stabilization. Technol. Forecasting Soc. Change, 74, 887–935, doi:10.1016/j.techfore.2006.05.026.
Rotstayn, L. D., Jeffrey S. J. , Collier M. A. , Dravitzki S. M. , Hirst A. C. , Syktus J. I. , and Wong K. K. , 2012: Aerosol-induced changes in summer rainfall and circulation in the Australasian region: A study using single-forcing climate simulations. Atmos. Chem. Phys. Discuss., 12, 5107–5188, doi:10.5194/acpd-12-5107-2012.
Rummukainen, M., 2010: State-of-the-art with regional climate models. Wiley Interdiscip. Rev.: Climate Change, 1, 82–96, doi:10.1002/wcc.8.
Schlesinger, M. E., and Coauthors, 2000: Geographical distributions of temperature change for scenarios of greenhouse gas and sulfur dioxide emissions. Technol. Forecasting Soc. Change, 65, 167–193, doi:10.1016/S0040-1625(99)00114-6.
Shiklomanov, I. A., 2000: Appraisal and assessment of world water resources. Water Int., 25, 11–32, doi:10.1080/02508060008686794.
Shindell, D. T., Voulgarakis A. , Faluvegi G. , and Milly G. , 2012: Precipitation response to regional radiative forcing. Atmos. Chem. Phys., 12, 6969–6982, doi:10.5194/acp-12-6969-2012.
Shiogama, H., and Coauthors, 2010a: Emission scenario dependencies in climate change assessments of the hydrological cycle. Climatic Change, 99, 321–329, doi:10.1007/s10584-009-9765-1.
Shiogama, H., Emori S. , Takahashi K. , Nagashima T. , Ogura T. , Nozawa T. , and Takemura T. , 2010b: Emission scenario dependence of precipitation on global warming in the MIROC3.2 model. J. Climate,23, 2404–2417, doi:10.1175/2009JCLI3428.1.
Swenson, S., and Wahr J. , 2006: Estimating large-scale precipitation minus evapotranspiration from GRACE satellite gravity measurements. J. Hydrometeor.,7, 252–270, doi:10.1175/JHM478.1.
Tebaldi, C., Arblaster J. M. , and Knutti R. , 2011: Mapping model agreement on future climate projections. Geophys. Res. Lett., 38, L23701, doi:10.1029/2011GL049863.
Trenberth, K. E., Smith L. , Qian T. , Dai A. , and Fasullo J. , 2007: Estimates of the global water budget and its annual cycle using observational and model data. J. Hydrometeor., 8, 758–769, doi:10.1175/JHM600.1.
van Vuuren, D., den Elzen M. , Lucas P. , Eickhout B. , Strengers B. , van Ruijven B. , Wonink S. , and van Houdt R. , 2007: Stabilizing greenhouse gas concentrations at low levels: An assessment of reduction strategies and costs. Climatic Change, 81, 119–159, doi:10.1007/s10584-006-9172-9.
Vecchi, G. A., and Soden B. J. , 2007: Global warming and the weakening of the tropical circulation. J. Climate, 20, 4316–4340, doi:10.1175/JCLI4258.1.
Vörösmarty, C. J., Green P. , Salisbury J. , and Lammers R. B. , 2000: Global water resources: Vulnerability from climate change and population growth. Science, 289, 284–288, doi:10.1126/science.289.5477.284.
Watanabe, M., and Coauthors, 2010: Improved climate simulation by MIROC5: Mean states, variability and climate sensitivity. J. Climate, 23, 6312–6335, doi:10.1175/2010JCLI3679.1.
Wise, M. A., and Coauthors, 2009: Implications of limiting CO2 concentrations for land use and energy. Science, 324, 1183–1186, doi:10.1126/science.1168475.
