• Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrological cycle. Nature, 419, 224232, doi:10.1038/nature01092.

    • Search Google Scholar
    • Export Citation
  • Ashfaq, M., Y. Shi, W.-W. Tung, R. J. Trapp, X. Gao, J. S. Pal, and N. S. Diffenbaugh, 2009: Suppression of South Asian summer monsoon precipitation in the 21st century. Geophys. Res. Lett., 36, L01704, doi:10.1029/2008GL036500.

    • Search Google Scholar
    • Export Citation
  • Ashfaq, M., C. B. Skinner, and N. S. Diffenbaugh, 2011: Influence of SST biases on future climate change projections. Climate Dyn., 36, 13031319, doi:10.1007/s00382-010-0875-2.

    • Search Google Scholar
    • Export Citation
  • Chaturvedi, R. K., J. Joshi, M. Jayaraman, G. Bala, and N. H. Ravindranath, 2012: Multi-model climate change projections for India under representative concentration pathways (RCPs): A preliminary analysis. Curr. Sci., 103, 791802.

    • Search Google Scholar
    • Export Citation
  • Cherchi, A., A. Alessandri, S. Masina, and A. Navarra, 2011: Effects of increased CO2 levels on monsoons. Climate Dyn., 37, 83101, doi:10.1007/s00382-010-0801-7.

    • Search Google Scholar
    • Export Citation
  • Dhar, O. N., and S. Nandargi, 2003: Hydrometeorological aspects of floods in India. Nat. Hazards, 28, 133, doi:10.1023/A:1021199714487.

    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., R. D. Koster, and Z. Guo, 2006: Do global models properly represent the feedback between land and atmosphere? J. Hydrometeor., 7, 11771198, doi:10.1175/JHM532.1.

    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., Y. Jin, B. Singh, and X. Yan, 2013: Trends in land–atmosphere interactions from CMIP5 simulations. J. Hydrometeor., 14, 829–849, doi:10.1175/JHM-D-12-0107.1.

    • Search Google Scholar
    • Export Citation
  • Dominguez, F., P. Kumar, X. Z. Liang, and M. Ting, 2006: Impact of atmospheric moisture storage on precipitation recycling. J. Climate, 19, 15131530, doi:10.1175/JCLI3691.1.

    • Search Google Scholar
    • Export Citation
  • Endo, H., and A. Kitoh, 2014: Thermodynamic and dynamic effects on regional monsoon rainfall changes in a warmer climate. Geophys. Res. Lett., 41, 17041710, doi:10.1002/2013GL059158.

    • Search Google Scholar
    • Export Citation
  • Goessling, H. F., and C. H. Reick, 2013: On the “well-mixed” assumption and numerical 2-D tracing of atmospheric moisture. Atmos. Chem. Phys., 13, 55675585, doi:10.5194/acp-13-5567-2013.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., V. Krishanmaurthy, and H. Annmalai, 1999: A broad-scale circulation index for the interannual variability of the Indian summer monsoon. Quart. J. Roy. Meteor. Soc., 125, 611633, doi:10.1002/qj.49712555412.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., and B. J. Soden, 2006: Robust responses of the hydrological cycle to global warming. J. Climate, 19, 56865699, doi:10.1175/JCLI3990.1.

    • Search Google Scholar
    • Export Citation
  • Joseph, S., A. K. Sahai, B. N. Goswami, P. Terray, S. Masson, and J. J. Luo, 2012: Possible role of warm SST bias in the simulation of boreal summer monsoon in SINTEX-F2 coupled model. Climate Dyn., 38, 15611576, doi:10.1007/s00382-011-1264-1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., S. Yukimoto, A. Noda, and T. Motoi, 1997: Simulated changes in the Asian summer monsoon at times of increased atmospheric CO2. J. Meteor. Soc. Japan, 75, 10191031.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., H. Endo, K. K. Kumar, I. F. A. Cavalcanti, P. Goswami, and T. Zhou, 2013: Monsoons in a changing world: A regional perspective in a global context. J. Geophys. Res. Atmos., 118, 30533065, doi:10.1002/jgrd.50258.

    • Search Google Scholar
    • Export Citation
  • Koster, R. D., and Coauthors, 2006: GLACE: The Global Land–Atmosphere Coupling Experiment. Part I: Overview. J. Hydrometeor., 7, 590610, doi:10.1175/JHM510.1.

    • Search Google Scholar
    • Export Citation
  • Kulkarni, A., 2012: Weakening of Indian summer monsoon rainfall in warming environment. Theor. Appl. Climatol., 109, 447459, doi:10.1007/s00704-012-0591-4.

    • Search Google Scholar
    • Export Citation
  • Lau, W. K. M., and K.-M. Kim, 2012: The 2010 Pakistan flood and Russian heatwave: Teleconnection of hydrometeorologic extremes. J. Hydrometeor., 13, 392403, doi:10.1175/JHM-D-11-016.1.

    • Search Google Scholar
    • Export Citation
  • Lee, J. Y., and B. Wang, 2014: Future change of global monsoon in the CMIP5. Climate Dyn., 42, 101119, doi:10.1007/s00382-012-1564-0.

    • Search Google Scholar
    • Export Citation
  • Li, C. F., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land sea thermal contrast. J. Climate, 9, 358375, doi:10.1175/1520-0442(1996)009<0358:TOAIVO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, Y., N. C. Jourdain, A. S. Taschetto, C. C. Ummenhofer, K. Ashok, and A. Sen Gupta, 2012: Evaluation of monsoon seasonality and the tropospheric biennial oscillation transitions in the CMIP models. Geophys. Res. Lett., 39, L20713, doi:10.1029/2012GL053322.

    • Search Google Scholar
    • Export Citation
  • May, W., 2002: Simulated changes of the Indian summer monsoon under enhanced greenhouse gas conditions in a global time-slice experiment. Geophys. Res. Lett., 29, 1118, doi:10.1029/2001GL013808.

    • Search Google Scholar
    • Export Citation
  • May, W., 2004: Potential future changes on the Indian summer monsoon due to greenhouse warming: Analysis of mechanisms in a global time-slice experiment. Climate Dyn., 22, 389414, doi:10.1007/s00382-003-0389-2.

    • Search Google Scholar
    • Export Citation
  • Meehl, G. A., and J. M. Arblaster, 2003: Mechanisms for projected future changes in South Asian monsoon precipitation. Climate Dyn., 21, 659675, doi:10.1007/s00382-003-0343-3.

    • Search Google Scholar
    • Export Citation
  • 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.

  • Moss, R. H., and Coauthors, 2010: The next generation of scenarios of climate change research and assessment. Nature, 463, 747756, doi:10.1038/nature08823.

    • Search Google Scholar
    • Export Citation
  • Sabade, S., A. Kulkarni, and R. Kripalani, 2011: Projected changes in South Asian summer monsoon by multi-model global warming experiments. Theor. Appl. Climatol., 103, 543565, doi:10.1007/s00704-010-0296-5.

    • Search Google Scholar
    • Export Citation
  • Seth, A., S. A. Rauscher, M. Biasutti, A. Giannini, S. J. Camargo, and M. Rojas, 2013: CMIP5 projected changes in the annual cycle of precipitation in monsoon regions. J. Climate, 26, 73287351, doi:10.1175/JCLI-D-12-00726.1.

    • Search Google Scholar
    • Export Citation
  • Singh D., and Coauthors, 2015: Severe precipitation in northern India in June 2013: Causes, historical context, and changes in probability [in “Explaining Extremes of 2013 from a Climate Perspective”]. Bull. Amer. Meteor. Soc.,95 (9), S58–S61.

  • Sperber, K. R., H. Annamalai, I.-S. Kang, A. Kitoh, A. Moise, A. Turner, B. Wang, and T. Zhou, 2013: The Asian summer monsoon: An intercomparison of CMIP5 vs. CMIP3 simulations of the late 20th century. Climate Dyn., 41, 27112744, doi:10.1007/s00382-012-1607-6.

    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Search Google Scholar
    • Export Citation
  • Turner, A. G., and H. Annamalai, 2012: Climate change and the South Asian summer monsoon. Nat. Climate Change, 2, 587–595, doi:10.1038/nclimate1495.

    • Search Google Scholar
    • Export Citation
  • Turner, A. G., P. M. Inness, and J. M. Slingo, 2007: The effect of doubled CO2 and model basic state biases on the monsoon-ENSO system. I: Mean response and interannual variability. Quart. J. Roy. Meteor. Soc., 133, 11431157, doi:10.1002/qj.82.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., A. Iwai, K. Kuwako, and M. E. Hori, 2006: Impact of anthropogenic forcing on the Asian summer monsoon as simulated by eight GCMs. Geophys. Res. Lett., 33, doi:10.1029/2005GL025336.

    • Search Google Scholar
    • Export Citation
  • Van der Ent, R. J., O. A. Tuinenburg, H.-R. Knoche, H. Kunstmann, and H. H. G. Savenije, 2013: Should we use a simple or complex model for moisture recycling and atmospheric moisture tracking? Hydrol. Earth Syst. Sci., 17, 48694884, doi:10.5194/hess-17-4869-2013.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and B. J. Soden, 2007: Global warming and the weakening of the tropical circulation. J. Climate, 20, 43164340, doi:10.1175/JCLI4258.1.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., B. J. Soden, A. T. Wittenberg, I. M. Held, A. Leetmaa, and M. J. Harrison, 2006: Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing. Nature, 441, 7376, doi:10.1038/nature04744.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., V. O. Magaña, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. J. Geophys. Res., 103, 14 45114 510, doi:10.1029/97JC02719.

    • Search Google Scholar
    • Export Citation
  • Yoon, J.-H., and T.-C. Chen, 2005: Water vapor budget of the Indian monsoon depression. Tellus, 57A, 770782, doi:10.1111/j.1600-0870.2005.00145.x.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Difference/comparison (mm day−1) of multimodel mean precipitation between RCP8.5 late (2070–2099), mid-, (2040–2069) and early-(2010–2039) twenty-first-century periods and the historical (1965–2005) period. (a) Difference in summer precipitation spatial pattern between late-twenty-first-century period and the historical period (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean). (b) Comparison of annual cycle of precipitation amount over continental South Asia between late-twenty-first century period and the historical period (“N/14” in the panel indicates N out of 14 models show the same sign of change as the multimodel mean). (c) As in (a), but for the midterm period. (d) As in (b), but for the midterm period. (e) As in (a), but for the early period. (f) As in (b), but for the early period.

  • View in gallery

    Comparison/difference (mm day−1) of multimodel mean between RCP8.5 late-twenty-first-century period and the historical period. (a) Comparison of MTG (K), U shear (m s−1), and V shear (m s−1). (b) Difference in summer atmospheric precipitable water, ET, and moisture convergence (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean).

  • View in gallery

    As in Fig. 2, but for the early period in the twenty-first century.

  • View in gallery

    As in Fig. 2, but for the midperiod in the twenty-first century.

  • View in gallery

    Climatological characteristics of South Asian precipitation sources over the historical period for NCEP–NCAR R1 data and 18 CMIP5 models’ mean. (a) Annual cycle of precipitation amount and its four moisture sources’ contributions (mm day−1). (b) Percentage contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; unitless). (c) Summer precipitation percentage contribution to each grid cell of the South Asian subcontinent from different sources (unitless).

  • View in gallery

    Climatological characteristics of South Asian precipitation sources over the historical period for NCEP–NCAR R1 data and 18 CMIP5 models’ mean. (a) Spatial pattern of South Asian precipitation amount (mm day−1). (b) Amount contribution to continental South Asian summer precipitation (mm day−1) from the evaporative source domain (normalized). (c) Summer precipitation amount contribution (mm day−1) to each grid cell of South Asian subcontinent from different sources.

  • View in gallery

    Annual cycle of South Asian precipitation amount and its four moisture sources’ contributions from all 18 CMIP5 models (mm day−1).

  • View in gallery

    Comparison/difference of multimodel mean between RCP8.5 late-twenty-first-century period and the historical period. (a) Comparison of annual cycle of precipitation amount (percentage) from its four moisture sources’ contributions (mm day−1) (unitless). (b) Difference in percentage contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; unitless). (c) Difference in summer precipitation percentage contribution to each grid cell of South Asian subcontinent from different sources (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean; unitless).

  • View in gallery

    Comparison/difference of mean of the CMIP5 14 models between the RCP8.5 late-twenty-first-century period and the historical period. (a) Difference in amount contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; mm day−1). (b) Difference in summer precipitation amount contribution to each grid cell of South Asian subcontinent from different sources (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean; mm day−1).

  • View in gallery

    As in Fig. 8, but for the early period in the twenty-first century.

  • View in gallery

    As in Fig. 9, but for the early period in the twenty-first century.

  • View in gallery

    As in Fig. 8, but for the midperiod in the twenty-first century.

  • View in gallery

    As in Fig. 9, but for the midperiod in the twenty-first century.

  • View in gallery

    (a) Future changes in MTG, U shear, and V shear for the three future periods (2010–39, 2040–69, 2070–99) using boxplot (minimum, 25th percentile, median, 75th percentile, maximum) to represent the intermodels’ variability; the line connecting the squares (i.e., multimodel mean) indicates significant difference at the 95% confidence level among the neighboring future periods. (b) As in (a), but for the precipitation sources’ contributions.

  • View in gallery

    The 850-mb wind difference between RCP8.5 late-twenty-first-century period and the historical period, and the 850-mb wind for the historical period during summer monsoon season (JJAS; m s−1).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 256 172 15
PDF Downloads 162 127 15

Dominating Controls for Wetter South Asian Summer Monsoon in the Twenty-First Century

View More View Less
  • 1 Computer Science and Mathematics Division, and Climate Change Science Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee
  • | 2 Atmospheric Sciences and Global Change Division, Pacific Northwest National Laboratory, Richland, Washington
  • | 3 Department of Atmospheric Sciences, The University of Arizona, Tucson, Arizona
© Get Permissions
Full access

Abstract

This paper analyzes a suite of global climate models from phase 5 of the Coupled Model Intercomparison Project (CMIP5) archives to understand the mechanisms behind a net increase in the South Asian summer monsoon precipitation in response to enhanced radiative forcing during the twenty-first century. An increase in radiative forcing fuels an increase in the atmospheric moisture content through warmer temperatures, which overwhelms the weakening of monsoon circulation and results in an increase of moisture convergence and therefore summer monsoon precipitation over South Asia. Moisture source analysis suggests that both regional (local recycling, the Arabian Sea, the Bay of Bengal) and remote (including the south Indian Ocean) sources contribute to the moisture supply for precipitation over South Asia during the summer season that is facilitated by the monsoon dynamics. For regional moisture sources, the effect of excessive atmospheric moisture is offset by weaker monsoon circulation and uncertainty in the response of the evapotranspiration over land, so anomalies in their contribution to the total moisture supply are either mixed or muted. In contrast, weakening of the monsoon dynamics has less influence on the moisture supply from remote sources that not only is a dominant moisture contributor in the historical period but is also the net driver of the positive summer monsoon precipitation response in the twenty-first century. The results also indicate that historic measures of the monsoon dynamics may not be well suited to predict the nonstationary moisture-driven South Asian summer monsoon precipitation response in the twenty-first century.

Denotes Open Access content.

Corresponding author address: Dr. Rui Mei, Oak Ridge National Laboratory, P.O. Box 2008 MS-6301, Oak Ridge, TN 37831-6301. E-mail: meir@ornl.gov

Publisher’s Note: This article was revised on 10 April 2015 to include the open access designation that was missing when originally published.

Abstract

This paper analyzes a suite of global climate models from phase 5 of the Coupled Model Intercomparison Project (CMIP5) archives to understand the mechanisms behind a net increase in the South Asian summer monsoon precipitation in response to enhanced radiative forcing during the twenty-first century. An increase in radiative forcing fuels an increase in the atmospheric moisture content through warmer temperatures, which overwhelms the weakening of monsoon circulation and results in an increase of moisture convergence and therefore summer monsoon precipitation over South Asia. Moisture source analysis suggests that both regional (local recycling, the Arabian Sea, the Bay of Bengal) and remote (including the south Indian Ocean) sources contribute to the moisture supply for precipitation over South Asia during the summer season that is facilitated by the monsoon dynamics. For regional moisture sources, the effect of excessive atmospheric moisture is offset by weaker monsoon circulation and uncertainty in the response of the evapotranspiration over land, so anomalies in their contribution to the total moisture supply are either mixed or muted. In contrast, weakening of the monsoon dynamics has less influence on the moisture supply from remote sources that not only is a dominant moisture contributor in the historical period but is also the net driver of the positive summer monsoon precipitation response in the twenty-first century. The results also indicate that historic measures of the monsoon dynamics may not be well suited to predict the nonstationary moisture-driven South Asian summer monsoon precipitation response in the twenty-first century.

Denotes Open Access content.

Corresponding author address: Dr. Rui Mei, Oak Ridge National Laboratory, P.O. Box 2008 MS-6301, Oak Ridge, TN 37831-6301. E-mail: meir@ornl.gov

Publisher’s Note: This article was revised on 10 April 2015 to include the open access designation that was missing when originally published.

1. Introduction

The summer monsoon accounts for more than 75% of the annual precipitation over most of South Asia (Dhar and Nandargi 2003). The summer monsoon precipitation has a profound impact on water resources, agriculture production, and human lives throughout South Asia. The most severe flooding events in Pakistan in July–August 2010 (Lau and Kim 2012) and in northern India in June 2013 (Singh et al. 2015) are two recent examples. Given the dependence of a large population on monsoon precipitation and an expected increase in population, robust projections of the response of South Asian summer monsoon precipitation to increased radiative forcing is critical for long-term planning on adaptation and mitigation.

The response of the South Asian summer monsoon to an increase in radiative forcing has been primarily studied using numerical modeling approaches (e.g., Kitoh et al. 1997; May 2002, 2004; Meehl and Arblaster 2003; Ueda et al. 2006; Turner et al. 2007; Ashfaq et al. 2009; Sabade et al. 2011), as recently reviewed in Turner and Annamalai (2012). Monsoon circulation has been unanimously projected to weaken, consistent with the prevailing view of the weakening tropical Walker circulation (Vecchi et al. 2006; Vecchi and Soden 2007) as a result of the thermodynamic scaling argument (Allen and Ingram 2002; Held and Soden 2006) under global warming. With regard to monsoon precipitation change, although a regional modeling study (Ashfaq et al. 2009) finds a decrease in precipitation, attributed to the weakening large-scale monsoon circulation, suppression of the convective environment, and dominant intraseasonal oscillatory modes, most studies [including phase 3 of the Coupled Model Intercomparison Project (CMIP3)] using global climate models project an increase in summer monsoon precipitation. This simulated wetter response is mainly driven by an increase in atmospheric moisture with increasing temperature despite the weakening of the dynamics governing summer monsoon circulation. However, examination of individual models presents uncertainties in CMIP3 projection for the summer monsoon precipitation. For instance, while the multimodel mean response to an increase in radiative forcing under the A1B scenario is largely positive, less than two-thirds of the models simulate an increase in precipitation in their individual responses [Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4), section 11.4.3, and Figs. 11.9 and 10.3.5.2].

More recently, given the availability of a new generation of global climate simulations from CMIP5, several studies (Sperber et al. 2013; Li et al. 2012) show that the CMIP5 multimodel mean performs better than the CMIP3 counterpart through an evaluation with a comprehensive set of metrics. As for future climate change, consistent with the CMIP3 multimodel mean projection, CMIP5 studies (Seth et al. 2013; Kitoh et al. 2013; Lee and Wang 2014) project an increase in monsoon precipitation that is attributed to enhanced moisture convergence resulting from an increase in atmospheric moisture content. Furthermore, Chaturvedi et al. (2012) show that more models converge toward a wetter summer monsoon response in South Asia, going from low to high radiative forcing scenarios and from near-term to long-term twenty-first century projections. Similar conclusions have also been discussed in the IPCC Fifth Assessment Report (AR5, section 14.2.2.1 and Fig. 14.4).

While the simulation of the South Asian summer monsoon response to an increase in radiative forcing is admittedly more robust in the current generation of global climate models (GCMs), the two competing effects—the thermodynamic effect (through an increase in atmospheric moisture content) and dynamic effect (through a weakening of monsoon circulation) —have also been further researched and quantified to demonstrate that the thermodynamic effect overwhelms the dynamic effect, leading to an increase in precipitation over South Asia (Cherchi et al. 2011; Endo and Kitoh 2014). However, only a few studies have attempted to attribute the predominantly positive precipitation response to a moisture supply/source perspective, which suggests an increased moisture supply from the warmer Indian Ocean to be the potential cause. For example, May (2002) and Ueda et al. (2006) found a larger moisture convergence and precipitation over South Asia partly due to the enhanced evaporation over the south Indian Ocean. Similarly, Meehl and Arblaster (2003) conducted several sensitivity experiments to test the response of precipitation to anomalously warm Indian and Pacific Oceans (with SST anomalies derived from increased radiative forcing experiments) and found that the increase of precipitation in response to radiative forcing is due to the warmer Indian Ocean. However, none of these studies provides a clear explanation of various moisture sources that contribute to the summer monsoon precipitation in South Asia. Moreover, it is not clear how an increase in radiative forcing affects the moisture supply from the contributing sources. To this end, we address the need of a more rigorous understanding of various moisture sources by applying a Lagrangian moisture-tracking method, extended from the dynamic recycling model (DRM) developed by Dominguez et al. (2006). Further, we investigate the impacts of anthropogenic warming on South Asian summer monsoon precipitation using CMIP5 data, taking into account not only the anomalies in the monsoon dynamics and apparent moisture supply, which have been extensively examined in previous studies, but also the changes in the contribution of various moisture sources and their effects on precipitation response, which have not been thoroughly investigated so far.

2. Data and methods

We analyze the coupled atmosphere–ocean general circulation model (AOGCM) outputs over the South Asian monsoon region from CMIP5 (Taylor et al. 2012) historical (1965–2005) and representative concentration pathway 8.5 (RCP8.5; 2010–99) experiments. RCP8.5 shows the highest level of radiative forcing (~8.5 W m−2) and global warming with greenhouse gas concentration exceeding 1370 ppm CO2e (carbon dioxide equivalent)(Moss et al. 2010) by the end of twenty-first century. We select 18 models for the historical period but 14 for the RCP8.5 period based on the availability of model output at the time of analysis, and use one ensemble run from each model as listed in Table 1. In addition, we analyze the NCEP–NCAR Reanalysis-I (NCEP R1) data (Kalnay et al. 1996) in comparison with the model output over the historical period. All model data were downloaded from the PCMDI data server (http://cmip-pcmdi.llnl.gov/) and analyzed at their native resolutions. Finally, we regrid the analyzed data to the NCEP R1 grid (2.5°) for comparison.

Table 1.

CMIP5 models used in this study. Forcing acronyms are as follows: GHG = greenhouse gases; Oz = ozone; SA = sulfate aerosols (direct and indirect effects); SD = sulfate aerosols (direct effects only); Sl = solar irradiance; Vl =volcanic aerosols; BC = black carbon; OC = organic carbon; LU = land use change; SS = sea salt; Ds = dust; MD = mineral dust; AA = inexplicit anthropogenic aerosols; Nat = inexplicit natural forcing; Ant = inexplicit anthropogenic forcing. Expansions of institutions and model names are available online at http://www.ametsoc.org/PubsAcronymList.

Table 1.

We first examine the South Asian summer monsoon [June–September (JJAS)] precipitation change in the three future time slices of RCP8.5, defined as early- (2010–39), mid- (2040–69), and late- (2070–99) twenty-first-century periods, with respect to the historical period (1965–2005). Further, we analyze changes in three dynamic indicators used to quantify the strength of the monsoon, together with three moisture terms (namely, atmospheric precipitable water, moisture convergence, and evaporation) used to quantify the apparent moisture supply for precipitation. The dynamic indicators include the meridional temperature gradient (MTG), defined as the difference of climatological mean temperature between the upper tropospheric layers (200–500 mb) at 30° and 5°N (Li and Yanai 1996); the local Walker circulation index (U shear), defined as the vertical easterly shear of zonal winds between 200 and 850 mb averaged over 0°–15°N, 50°–85°E (Webster et al. 1998); and the local Hadley circulation index (V shear), defined as the vertical meridional wind shear between 200 and 850 mb averaged over 5°–30°N, 70°–110°E (Goswami et al. 1999). These three indices have been widely used to indicate the strength of the monsoon circulation and to correlate well with the intraseasonal precipitation variability (MTG) (e.g., Kulkarni 2012) and seasonal mean precipitation (U shear and V shear) (e.g., Ashfaq et al. 2009). For instance, Kulkarni (2012) showed a weakening of Indian summer monsoon precipitation since the mid-1970s, which is accompanied by the weakening of MTG and therefore monsoon circulation.

Moreover, we use a Lagrangian moisture tracking method to determine the moisture sources that contribute to the moisture supply over the South Asian subcontinent during JJAS, and the effects of the increase in radiative forcing on their contributions to the changes in summer precipitation in the twenty-first century. The Lagrangian moisture tracking method is an extension of the DRM developed by Dominguez et al. (2006). The DRM is derived from the water vapor mass conservation equation with the assumption that the atmospheric column is well mixed, which dictates that the ratio of local recycled to neighboring advected precipitation for any grid cell is equal to the ratio of evaporated to advected water vapor of the atmospheric column. According to Dominguez et al. (2006), with the introduction of a Lagrangian coordinate system to follow the trajectory of the advected moisture, where represent the dimensions in space (west–east and south–north) and time, respectively, the local recycling ratio R accounting for the fraction of precipitation falling in a specific grid cell originating as evapotranspiration (ET) from the entire evaporative source region can be solved with this expression:
e1
where R, ε, and ω represent the recycling ratio, evaporation, and precipitable water, respectively; and the integration term is performed from the current time τ backward until time 0, when the back trajectory of the advected moisture travels beyond the predefined evaporative source region. The integration term in Eq. (1) is calculated with the numerical scheme as in Dominguez et al. (2006). To enable a general moisture tracking following the reasoning of the local recycling ratio, we can make an inference through extending the DRM that for any specific precipitation target grid cell, the contributed percentage during a small time interval over is from the moisture source cell j, where the back trajectory of moisture falls during that time interval . This leads to a general moisture tracking equation,
e2
where is the contributed percentage from moisture source cell j at time . With the above-mentioned equation, we can derive the precipitation contributed to any target grid cell from any evaporative moisture source grid cell.

One of the drawbacks of applying the extended DRM to NCEP R1 data is that the reanalysis data are not always water vapor mass conservative due to the data assimilation. A correction factor can be applied to either precipitation or evaporation as in Dominguez et al. (2006), which shows negligible effects on the results. However, water vapor mass conservation should not be a concern in the case of CMIP5 models due to the use of model physics. It is also important to note that during the moisture backtracking using the extended DRM, the corresponding trajectories follow an effective 2D wind field (Dominguez et al. 2006). This 2D approach has been recently under debate because the vertical wind shear of the horizontal winds can produce water transport patterns not accounted for in a vertically integrated moisture flux (Goessling and Reick 2013; Van der Ent et al. 2013). As such, the estimates of the extended DRM must be interpreted as first-order results. However, the extended DRM provides a computationally efficient tool for moisture tracking in this study.

The variables used in the analysis of precipitation sources include daily precipitation and evaporation, and 6-hourly three-dimensional horizontal winds and specific humidity. Note that 6-hourly three-dimensional horizontal winds and specific humidity are preprocessed into daily effective 2D wind and atmospheric precipitable water as done in Dominguez et al. (2006). We define four exclusive but complementary moisture sources for the South Asian monsoon precipitation region–land grid points between 5°–30°N and 70°–90°E (red box in Fig. 1a, land grids only): 1) local (i.e., the South Asian monsoon precipitation region itself); 2) the Arabian Sea—ocean grid points between 5°–30°N and 45°–80°E (green box in Fig. 1a, ocean grids only); 3) the Bay of Bengal—ocean grids points between 5°–30°N and 80°–100°E (blue box in Fig. 1a, ocean grids only); and 4) remote—the rest of the globe (inferred based on the first three sources because total = local + Arabian Sea + Bay of Bengal + remote). Note that when we conduct the moisture tracking analysis, we effectively consider only the moisture source domain spanning 30°S–45°N along the latitude and 0°–150°E along the longitude (Fig. 1a) (rather than the globe) for two reasons: 1) This domain can explain almost 90% of the total summer monsoon precipitation over South Asia, which also justifies the validity of the tracking method. 2) This domain covers the three regional moisture sources (local, the Arabian Sea, and the Bay of Bengal), and the domain portion that excludes those three sources reflects mostly the remote contribution because the whole domain explains almost 90% of the total summer monsoon precipitation over South Asia.

Fig. 1.
Fig. 1.

Difference/comparison (mm day−1) of multimodel mean precipitation between RCP8.5 late (2070–2099), mid-, (2040–2069) and early-(2010–2039) twenty-first-century periods and the historical (1965–2005) period. (a) Difference in summer precipitation spatial pattern between late-twenty-first-century period and the historical period (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean). (b) Comparison of annual cycle of precipitation amount over continental South Asia between late-twenty-first century period and the historical period (“N/14” in the panel indicates N out of 14 models show the same sign of change as the multimodel mean). (c) As in (a), but for the midterm period. (d) As in (b), but for the midterm period. (e) As in (a), but for the early period. (f) As in (b), but for the early period.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

In our results, the target precipitation grid cells are within the terrestrial South Asian monsoon region (i.e., local), and we present the aggregated contribution from the moisture cells within each of the four source regions as both spatially distributed maps and temporal time series. The percent contributions [i.e., recycled ratio in Eq. (2)] are calculated at the daily time scale and then averaged in time and aggregated in space to present the climatological analyses.

We present the results from our analysis with the multimodel mean. Changes in the early-, mid- and late-twenty-first-century periods are calculated as the mean difference from the historical period. We consider a projected signal as robust if two-thirds of the examined CMIP5 models simulate a response identical in sign to that of the multimodel mean.

3. Results and discussion

Figure 1b shows the annual cycle of South Asian precipitation climatology in the historical period (1965–2005) and the late-twenty-first-century future period (2070–99), area averaged over land grids only between 5°–30°N and 70°–90°E. All 14 GCMs show a positive response for regionally averaged future summer monsoon precipitation as the composite mean with a range of increase between 0.4 and 1.73 mm per day and a mean increase of ~1 mm per day by the end of the twenty-first century. The spatial response of precipitation, shown as a composite in Fig. 1a, is also robust, as two-thirds of the analyzed models agree on the sign of the change at most of the grid points over South Asia, while the magnitude of change exhibits differences from one model to another (not shown). In general, precipitation shows an increase over South Asia and the north Indian Ocean (including the Arabian Sea and the Bay of Bengal) and a decrease over the south Indian Ocean (mostly south of the equator). An analysis of the early- and mid-twenty-first-century periods reveals a gradual strengthening of the positive precipitation response with more and more models converging on the sign of precipitation change as radiative forcing increases during the course of the twenty-first century (Figs. 1c–f).

The fact that the examined CMIP5 models exhibit a robust precipitation response in the twenty-first-century future period motivates us to explore potential causes behind such a response. We begin with an examination of the monsoon dynamic indicators. First, the strength of all indicators (MTG, U shear, V shear) is underestimated with respect to the NCEP R1 in the multimodel mean (Fig. 2a) and most of the individual GCMs (not shown). While the robust nature of these biases in the simulation of summer monsoon dynamics may stem from common limitations/biases among models, such as coarse resolution and sea surface temperature errors (e.g., Ashfaq et al. 2011; Joseph et al. 2012), etc., further investigation is beyond the scope of this study. All of these dynamics indicators show a future weakening that is mostly robust in the twenty-first-century period (Figs. 2a, 3a, 4a) with 11 (12, 12), 14 (14, 14), and 10 (11, 8) out of 14 models demonstrating the same sign of change for MTG, U shear, and V shear in the late- (middle, early) twenty-first century period, respectively. The weakening of MTG is related to the projected enhanced upper-tropospheric warming in the tropics (Meehl et al. 2007; Ashfaq et al. 2009). The upper troposphere over the Indian Ocean warms faster than the Tibetan Plateau due to the increase in condensational heating over the ocean, thereby reducing the MTG and monsoon circulations indicated by U shear and V shear. Collectively, these changes suggest a weakening of the monsoon circulation in the twenty-first century in the examined CMIP5 GCMs, which is consistent with the IPCC AR4 and previous studies (e.g., Ueda et al. 2006). On the other hand, an inspection of the change in evaporation and moisture convergence—the only two apparent moisture supplies (Fig. 2b) for precipitation (Fig. 1a)—indicates that the simulated summer monsoon precipitation increase is mainly driven by an increase in moisture convergence based on the high degree of resemblance in their spatial anomalies. Given the weaker monsoon dynamics in the twenty-first century, such an increase in the moisture convergence over South Asia during the summer monsoon season is most likely caused by an almost uniform increase in the atmospheric precipitable water (Fig. 2b), which dominates over the weak monsoon dynamics. Likewise, analysis of the early- and mid-twenty-first-century future periods (Figs. 3, 4) also suggests a weakening of monsoon circulation compensated by an increase in the atmospheric moisture but at much weaker magnitude than that in the late-twenty-first-century future period. In particular, it is shown (see Fig. 14a) that the trend of weakening in MTG is significant across the three future periods, while that in U shear and V shear is at least significant across the first two future periods at the 95% confidence level.

Fig. 2.
Fig. 2.

Comparison/difference (mm day−1) of multimodel mean between RCP8.5 late-twenty-first-century period and the historical period. (a) Comparison of MTG (K), U shear (m s−1), and V shear (m s−1). (b) Difference in summer atmospheric precipitable water, ET, and moisture convergence (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 3.
Fig. 3.

As in Fig. 2, but for the early period in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 4.
Fig. 4.

As in Fig. 2, but for the midperiod in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

To understand the source of the excessive moisture over the South Asian monsoon region in the twenty-first century, we examine the characteristics of different moisture sources, defined in the methods section, that contribute to South Asian monsoon precipitation over the historical period and then analyze how their absolute/relative contribution changes over the course of the twenty-first century due to the gradual increase in radiative forcing. We note that all four moisture source regions, defined in our methods section, contribute to the total moisture supply during the summer monsoon season over South Asia (Fig. 5). The relative contribution of these moisture sources to the summer monsoon precipitation, as seen in the NCEP R1 data, is generally captured well in the multimodel mean of the examined CMIP5 GCMs. Among the four defined moisture sources, remote contributes the most, representing 46% (55%); followed by the local recycling, representing 30.4% (23%); the Arabian Sea, contributing 19.6% (15%); and the Bay of Bengal, contributing 4% (7%) in the analyzed GCMs (NCEP R1), respectively. Note that the finding that the Bay of Bengal is the smallest moisture contributor does not necessarily contradict the prevailing view that monsoon depression systems originating from the Bay of Bengal account for ~40–50% of summer monsoon precipitation (Yoon and Chen 2005) because the moisture in the low pressure systems may originate from other sources and pass over the Bay of Bengal as part of the monsoon circulation. Despite the biases in the simulation of the magnitude of the relative contributions from different moisture sources (Figs. 5, 6), the multimodel mean can still capture important features of the characteristics in the NCEP R1 from both temporal and spatial perspectives. Temporally, precipitation attributed to local recycling and the Bay of Bengal peaks at the later stage of the monsoon period and the postmonsoon period, while precipitation attributed to the remote area and the Arabian Sea peaks at an earlier stage (Fig. 5a). It suggests that during monsoon onset, moisture mainly comes from the remote and Arabian Sea sources to the subcontinent as precipitation. As the monsoon matures and retreats, continental local recycling through evapotranspiration enabled by increased soil moisture from the monsoon precipitation plays a more important role in supplying moisture for precipitation. Spatially, precipitation peaks over the southwest (i.e., western Ghats) and north-northeast (i.e., the foothills of the Himalayas) of the subcontinent (Fig. 6a), and its evaporative moisture source pattern (Figs. 5b, 6b) suggests that the moisture source signal density peaks locally and gradually weakens from the subcontinental land to the Arabian Sea, and from the west and the south Indian Ocean following the wind back trajectory. The evaporative source patterns (Figs. 5b, 6b) also suggest that the remote contribution mainly comes from the East African region, the Mediterranean region, and the south Indian Ocean; however, a number of grids in the south Indian Ocean contributing to the total moisture from the remote region are much more than the rest of the other two regions, meaning that the remote contribution predominantly comes from the south Indian Ocean. In terms of the continental “footprints” from the four moisture sources (Figs. 5c, 6c), the signature lies in north-northeast from the local and Bay of Bengal sources, in southwest from the Arabian Sea source, and in the north-northeast and southwest from the remote source. Although it is encouraging to see the above-mentioned consistencies in the characteristics of moisture sources between the multimodel mean and NCEP R1, individual model skill varies depending on whether the magnitude, relative contribution, or timing of the peak (Fig. 7) is of concern. For example, 12 out of 18 models can capture the relative contribution of each moisture source but only 7 out of 18 models can capture the increasing (decreasing) dominance of local (remote and Arabian sea) contribution from monsoon onset to retreat. We have also analyzed the NCEP Climate Forecast System Reanalysis (CFSR) and the ERA-Interim datasets (spanning the period 1979–2005) to make sure that the moisture sources’ characteristics based on NCEP R1 are robust. We note that the ordering of the relative contribution from each source and most spatial and temporal features are robust across all the reanalysis products (not shown).

Fig. 5.
Fig. 5.

Climatological characteristics of South Asian precipitation sources over the historical period for NCEP–NCAR R1 data and 18 CMIP5 models’ mean. (a) Annual cycle of precipitation amount and its four moisture sources’ contributions (mm day−1). (b) Percentage contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; unitless). (c) Summer precipitation percentage contribution to each grid cell of the South Asian subcontinent from different sources (unitless).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 6.
Fig. 6.

Climatological characteristics of South Asian precipitation sources over the historical period for NCEP–NCAR R1 data and 18 CMIP5 models’ mean. (a) Spatial pattern of South Asian precipitation amount (mm day−1). (b) Amount contribution to continental South Asian summer precipitation (mm day−1) from the evaporative source domain (normalized). (c) Summer precipitation amount contribution (mm day−1) to each grid cell of South Asian subcontinent from different sources.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 7.
Fig. 7.

Annual cycle of South Asian precipitation amount and its four moisture sources’ contributions from all 18 CMIP5 models (mm day−1).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Despite the disagreement among the models in representing the magnitude and relative contribution of different moisture sources for the historical period (Fig. 7), there is substantial agreement in their simulated changes in the moisture contribution from different sources during the late-twenty-first-century period (Figs. 8, 9), which is consistent with their simulated precipitation response (Fig. 1). Overall, robust temporal and spatial changes include an increase in the absolute and percentage (relative contribution; i.e., percentage of the total amount) contributions from the remote and a decrease in the percentage contribution from all other relatively regional sources (local, the Bay of Bengal, and the Arabian Sea). Specifically, more than two-thirds of the 14 models simulate an increase in the absolute and percentage contributions from remote sources and a decrease in percentage contributions from all regional sources. However, there is no consensus among the models in the simulation of future changes in the absolute contribution from regional sources (Fig. 8a). Spatially, the percentage contribution from all regional contributions (remote contribution) presents a weakening (strengthening) signal throughout the subcontinent (Fig. 8c), while in terms of absolute amount, the contribution from remote sources presents a coherent increasing signal, whereas all regional contributions (local, the Bay of Bengal, and the Arabian Sea) demonstrate a mixed signal throughout the South Asian subcontinent (Fig. 9b). The decreases in percentage contributions from all regional sources can be considered as net results of an increase in total precipitation and insignificant/uncertain changes in the absolute contribution from regional sources. All the above-mentioned anomalies can also be interpreted by the evaporative moisture source change pattern (Figs. 8b, 9a). Similarly, an analysis over early- and mid-twenty-first-century future periods (Figs. 1013) depicts a similar conclusion but with weaker signals compared to the late-twenty-first-century period. It is worth mentioning that the future response of each individual model out of the 14 models is consistent with that of the multimodel mean shown in Figs. 813, which suggests that multimodel mean is representative of the multimodel behavior. Also note that the remote contribution in terms of both absolute amount and percentage present an increasing trend, while the local and Arabian Sea percentage contributions show a decreasing trend across the three future periods at the 95% confidence level (Fig. 14b).

Fig. 8.
Fig. 8.

Comparison/difference of multimodel mean between RCP8.5 late-twenty-first-century period and the historical period. (a) Comparison of annual cycle of precipitation amount (percentage) from its four moisture sources’ contributions (mm day−1) (unitless). (b) Difference in percentage contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; unitless). (c) Difference in summer precipitation percentage contribution to each grid cell of South Asian subcontinent from different sources (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean; unitless).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 9.
Fig. 9.

Comparison/difference of mean of the CMIP5 14 models between the RCP8.5 late-twenty-first-century period and the historical period. (a) Difference in amount contribution to continental South Asian summer precipitation from the evaporative source domain (normalized; mm day−1). (b) Difference in summer precipitation amount contribution to each grid cell of South Asian subcontinent from different sources (stippled area indicates more than two-thirds of the models show the same sign of change as the multimodel mean; mm day−1).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 10.
Fig. 10.

As in Fig. 8, but for the early period in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 11.
Fig. 11.

As in Fig. 9, but for the early period in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 12.
Fig. 12.

As in Fig. 8, but for the midperiod in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 13.
Fig. 13.

As in Fig. 9, but for the midperiod in the twenty-first century.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

Fig. 14.
Fig. 14.

(a) Future changes in MTG, U shear, and V shear for the three future periods (2010–39, 2040–69, 2070–99) using boxplot (minimum, 25th percentile, median, 75th percentile, maximum) to represent the intermodels’ variability; the line connecting the squares (i.e., multimodel mean) indicates significant difference at the 95% confidence level among the neighboring future periods. (b) As in (a), but for the precipitation sources’ contributions.

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

The fact that regional moisture sources show a mixed or weaker response to increases in atmospheric moisture content in the twenty-first century (Figs. 8c, 9b) suggests that the weakening of monsoon dynamics apparently has more influence on at least some of the regional moisture supplies (the Arabian Sea, the Bay of Bengal) that offsets the positive influence of excessive atmospheric moisture on the absolute contribution from these sources (Fig. 9b). For instance, anomalies in monsoon dynamics (Fig. 2a) lead to a weaker lower-level westerly jet over the Arabian Sea and a southwesterly flow over the Bay of Bengal (Fig. 15) that could offset any increase caused by the excessive atmospheric moisture over these regions by reduced moisture advection. However, interpretation of the mixed or weaker response of local recycling is not straightforward. There can be a number of potential reasons that could collectively lead to such a response. First, there can be inconsistencies in how land–atmosphere interactions respond to global warming due to the uncertainties in the simulation of land–atmosphere coupling strength among the models (e.g., Koster et al. 2006; Dirmeyer et al. 2006, 2013). Second, competitive effects of increased evaporative demands driven by warmer surface temperatures versus reduced evaporative demands driven by moister atmosphere could lead to inconsistencies in net evaporation response to increases in radiative forcing as reflected in the early- and mid-twenty-first-century periods (Figs. 3b, 4b). On the other hand, the remote moisture source that mainly consists of the south Indian Ocean (Fig. 5b) is least influenced by the weakening of the monsoon dynamics, as indicated by both the absolute and relative contributions from the remote sources showing a robust increase among all the analyzed models.

Fig. 15.
Fig. 15.

The 850-mb wind difference between RCP8.5 late-twenty-first-century period and the historical period, and the 850-mb wind for the historical period during summer monsoon season (JJAS; m s−1).

Citation: Journal of Climate 28, 8; 10.1175/JCLI-D-14-00355.1

The increasing influence of atmospheric moisture content and the decreasing influence of historically defined monsoon dynamics drivers during the course of the twenty-first century also suggest the emergence of a new state where monsoon precipitation variability is progressively less connected to the variability of the traditional monsoon dynamics. Previous studies (e.g., Webster et al. 1998; Goswami et al. 1999) mostly examine the precipitation variability over the historical period when the atmospheric moisture content is more or less stationary. In that context, precipitation anomalies in response to external forcing (e.g., ENSO) can be well dictated by anomalies in the monsoon dynamics. However, in the future period under an enhanced radiative forcing scenario (e.g., RCP8.5 in this case), atmospheric moisture presents a nonstationary response that shows an increase with warmer temperatures following the Clausius–Clapeyron relationship. Despite the weakening of monsoon circulation, it should be noted that the summer monsoon is still a dominating circulation system that leads to moisture convergence over South Asia. In such a warmer and moister world, the increase in atmospheric moisture content may overcome the weakening of the monsoon circulation, leading to an overall increase in moisture convergence and monsoon precipitation. Hence, monsoon dynamics indicators such as MTG, U shear, and V shear (Fig. 2a) could become less useful in predicting the precipitation response to increased radiative forcing in the twenty-first century.

4. Conclusions

We examine the response of the South Asian summer monsoon precipitation to anthropogenic global warming in the twenty-first century using 14 GCMs from the CMIP5 archives. We show that summer monsoon precipitation increases throughout the twenty-first century in the majority of the analyzed GCMs even when a robust weakening of the monsoon dynamics has been simulated, consistent with earlier studies (e.g., Ueda et al. 2006; May 2004). Weakening of the monsoon dynamics is manifested through a significant decrease in MTG and U shear, and a slight decrease in V shear during the summer months (Fig. 15). The key mechanism responsible for the precipitation increase is the enhanced moisture convergence over the South Asian monsoon regions, which is likely a direct reflection of the increase in atmospheric precipitable water due to atmospheric warming.

Precipitation source analysis reveals the relative contribution of four exclusive but complementary moisture sources with remote and local recycling representing the first and second largest sources, respectively; followed by the Arabian Sea; and leaving the Bay of Bengal as the smallest contributor. In addition, the precipitation source analysis unveils intrinsic characteristics of precipitation from different moisture sources, such as their spatial footprints over South Asia and temporal variability during the summer monsoon period. The CMIP5 ensemble mean captures the relative contributions of moisture from difference sources to precipitation over South Asia with some fidelity. More importantly, this analysis shows that only the remote contribution demonstrates a robust increase throughout the twenty-first century. Together with the understanding gained through the monsoon circulation and the apparent moisture supply analysis, it can be inferred that the summer monsoon precipitation increase is mostly sustained by the increase in moisture supply from remote sources instead of regional sources (local, the Arabian Sea, and the Bay of Bengal), facilitated through increases in atmospheric moisture content and moisture convergence over the South Asian monsoon region.

The primary discrepancies and commonalities between our study and previous studies (e.g., May 2002; Meehl and Arblaster 2003; Ueda et al. 2006) are in two aspects: 1) Our study uses a new generation of global climate model simulations from CMIP5, while previous studies use either an old version of a single model or a subset of CMIP3 models. Compared with the uncertainties in CMIP3 (IPCC AR4, section 11.4.3 and Fig. 11.9) mentioned before, the CMIP5 models examined in this study indicate more robustness in the projection of an increase in precipitation, which also confirms other CMIP5 studies (Sperber et al. 2013; Chaturvedi et al. 2012; Kitoh et al. 2013). 2) Previous studies explain the increase in precipitation with the increased moisture transported to South Asia from the warmer Indian Ocean through either the moisture convergence term (May 2002; Ueda et al. 2006), as we have also examined in this study, or through additional sensitivity experiments (Meehl and Arblaster 2003) to account for the impact from the moisture source of the Indian Ocean, while our study implements a moisture tracking method to directly and quantitatively analyze the source of precipitation. This method clearly reveals that the increase in summer monsoon precipitation in the future is ascribed to the increase from remote sources (including the south Indian Ocean as a major component) rather than regional sources (local, the Arabian Sea, and the Bay of Bengal). This moisture source perspective complements our previous understanding of the monsoon precipitation increase from its decomposition into thermodynamic and dynamic effects in other studies (Cherchi et al. 2011; Endo and Kitoh 2014). Our analysis also indicates that traditional measures of South Asian monsoon strength may be less applicable to evaluating precipitation changes in the warmer and moister climate of the twenty-first century.

Acknowledgments

Support for data storage and analysis is provided by the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract DE-AC05-00OR22725. We acknowledge the World Climate Research Programme’s Working Group on Coupled Modeling, responsible for CMIP, and we thank the climate modeling groups (listed in Table 1 of this paper) for producing and making available their CMIP5 model output. We also thank the U.S. Department of Energy’s Program for Climate Model Diagnosis and Intercomparison for providing coordinating support and leading development of software infrastructure in partnership with the Global Organization for Earth System Science Portals for CMIP. This work is supported by the Regional and Global Climate Modeling Program of the DOE’s Office of Science and Oak Ridge National Laboratory LDRD project 32112413. Pacific Northwest National Laboratory is operated for the DOE by Battelle Memorial Institute under Contract DE-AC05-76RL01830.

REFERENCES

  • Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrological cycle. Nature, 419, 224232, doi:10.1038/nature01092.

    • Search Google Scholar
    • Export Citation
  • Ashfaq, M., Y. Shi, W.-W. Tung, R. J. Trapp, X. Gao, J. S. Pal, and N. S. Diffenbaugh, 2009: Suppression of South Asian summer monsoon precipitation in the 21st century. Geophys. Res. Lett., 36, L01704, doi:10.1029/2008GL036500.

    • Search Google Scholar
    • Export Citation
  • Ashfaq, M., C. B. Skinner, and N. S. Diffenbaugh, 2011: Influence of SST biases on future climate change projections. Climate Dyn., 36, 13031319, doi:10.1007/s00382-010-0875-2.

    • Search Google Scholar
    • Export Citation
  • Chaturvedi, R. K., J. Joshi, M. Jayaraman, G. Bala, and N. H. Ravindranath, 2012: Multi-model climate change projections for India under representative concentration pathways (RCPs): A preliminary analysis. Curr. Sci., 103, 791802.

    • Search Google Scholar
    • Export Citation
  • Cherchi, A., A. Alessandri, S. Masina, and A. Navarra, 2011: Effects of increased CO2 levels on monsoons. Climate Dyn., 37, 83101, doi:10.1007/s00382-010-0801-7.

    • Search Google Scholar
    • Export Citation
  • Dhar, O. N., and S. Nandargi, 2003: Hydrometeorological aspects of floods in India. Nat. Hazards, 28, 133, doi:10.1023/A:1021199714487.

    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., R. D. Koster, and Z. Guo, 2006: Do global models properly represent the feedback between land and atmosphere? J. Hydrometeor., 7, 11771198, doi:10.1175/JHM532.1.

    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., Y. Jin, B. Singh, and X. Yan, 2013: Trends in land–atmosphere interactions from CMIP5 simulations. J. Hydrometeor., 14, 829–849, doi:10.1175/JHM-D-12-0107.1.

    • Search Google Scholar
    • Export Citation
  • Dominguez, F., P. Kumar, X. Z. Liang, and M. Ting, 2006: Impact of atmospheric moisture storage on precipitation recycling. J. Climate, 19, 15131530, doi:10.1175/JCLI3691.1.

    • Search Google Scholar
    • Export Citation
  • Endo, H., and A. Kitoh, 2014: Thermodynamic and dynamic effects on regional monsoon rainfall changes in a warmer climate. Geophys. Res. Lett., 41, 17041710, doi:10.1002/2013GL059158.

    • Search Google Scholar
    • Export Citation
  • Goessling, H. F., and C. H. Reick, 2013: On the “well-mixed” assumption and numerical 2-D tracing of atmospheric moisture. Atmos. Chem. Phys., 13, 55675585, doi:10.5194/acp-13-5567-2013.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., V. Krishanmaurthy, and H. Annmalai, 1999: A broad-scale circulation index for the interannual variability of the Indian summer monsoon. Quart. J. Roy. Meteor. Soc., 125, 611633, doi:10.1002/qj.49712555412.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., and B. J. Soden, 2006: Robust responses of the hydrological cycle to global warming. J. Climate, 19, 56865699, doi:10.1175/JCLI3990.1.

    • Search Google Scholar
    • Export Citation
  • Joseph, S., A. K. Sahai, B. N. Goswami, P. Terray, S. Masson, and J. J. Luo, 2012: Possible role of warm SST bias in the simulation of boreal summer monsoon in SINTEX-F2 coupled model. Climate Dyn., 38, 15611576, doi:10.1007/s00382-011-1264-1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., S. Yukimoto, A. Noda, and T. Motoi, 1997: Simulated changes in the Asian summer monsoon at times of increased atmospheric CO2. J. Meteor. Soc. Japan, 75, 10191031.

    • Search Google Scholar
    • Export Citation
  • Kitoh, A., H. Endo, K. K. Kumar, I. F. A. Cavalcanti, P. Goswami, and T. Zhou, 2013: Monsoons in a changing world: A regional perspective in a global context. J. Geophys. Res. Atmos., 118, 30533065, doi:10.1002/jgrd.50258.

    • Search Google Scholar
    • Export Citation
  • Koster, R. D., and Coauthors, 2006: GLACE: The Global Land–Atmosphere Coupling Experiment. Part I: Overview. J. Hydrometeor., 7, 590610, doi:10.1175/JHM510.1.

    • Search Google Scholar
    • Export Citation
  • Kulkarni, A., 2012: Weakening of Indian summer monsoon rainfall in warming environment. Theor. Appl. Climatol., 109, 447459, doi:10.1007/s00704-012-0591-4.

    • Search Google Scholar
    • Export Citation
  • Lau, W. K. M., and K.-M. Kim, 2012: The 2010 Pakistan flood and Russian heatwave: Teleconnection of hydrometeorologic extremes. J. Hydrometeor., 13, 392403, doi:10.1175/JHM-D-11-016.1.

    • Search Google Scholar
    • Export Citation
  • Lee, J. Y., and B. Wang, 2014: Future change of global monsoon in the CMIP5. Climate Dyn., 42, 101119, doi:10.1007/s00382-012-1564-0.

    • Search Google Scholar
    • Export Citation
  • Li, C. F., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land sea thermal contrast. J. Climate, 9, 358375, doi:10.1175/1520-0442(1996)009<0358:TOAIVO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, Y., N. C. Jourdain, A. S. Taschetto, C. C. Ummenhofer, K. Ashok, and A. Sen Gupta, 2012: Evaluation of monsoon seasonality and the tropospheric biennial oscillation transitions in the CMIP models. Geophys. Res. Lett., 39, L20713, doi:10.1029/2012GL053322.

    • Search Google Scholar
    • Export Citation
  • May, W., 2002: Simulated changes of the Indian summer monsoon under enhanced greenhouse gas conditions in a global time-slice experiment. Geophys. Res. Lett., 29, 1118, doi:10.1029/2001GL013808.

    • Search Google Scholar
    • Export Citation
  • May, W., 2004: Potential future changes on the Indian summer monsoon due to greenhouse warming: Analysis of mechanisms in a global time-slice experiment. Climate Dyn., 22, 389414, doi:10.1007/s00382-003-0389-2.

    • Search Google Scholar
    • Export Citation
  • Meehl, G. A., and J. M. Arblaster, 2003: Mechanisms for projected future changes in South Asian monsoon precipitation. Climate Dyn., 21, 659675, doi:10.1007/s00382-003-0343-3.

    • Search Google Scholar
    • Export Citation
  • 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.

  • Moss, R. H., and Coauthors, 2010: The next generation of scenarios of climate change research and assessment. Nature, 463, 747756, doi:10.1038/nature08823.

    • Search Google Scholar
    • Export Citation
  • Sabade, S., A. Kulkarni, and R. Kripalani, 2011: Projected changes in South Asian summer monsoon by multi-model global warming experiments. Theor. Appl. Climatol., 103, 543565, doi:10.1007/s00704-010-0296-5.

    • Search Google Scholar
    • Export Citation
  • Seth, A., S. A. Rauscher, M. Biasutti, A. Giannini, S. J. Camargo, and M. Rojas, 2013: CMIP5 projected changes in the annual cycle of precipitation in monsoon regions. J. Climate, 26, 73287351, doi:10.1175/JCLI-D-12-00726.1.

    • Search Google Scholar
    • Export Citation
  • Singh D., and Coauthors, 2015: Severe precipitation in northern India in June 2013: Causes, historical context, and changes in probability [in “Explaining Extremes of 2013 from a Climate Perspective”]. Bull. Amer. Meteor. Soc.,95 (9), S58–S61.

  • Sperber, K. R., H. Annamalai, I.-S. Kang, A. Kitoh, A. Moise, A. Turner, B. Wang, and T. Zhou, 2013: The Asian summer monsoon: An intercomparison of CMIP5 vs. CMIP3 simulations of the late 20th century. Climate Dyn., 41, 27112744, doi:10.1007/s00382-012-1607-6.

    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Search Google Scholar
    • Export Citation
  • Turner, A. G., and H. Annamalai, 2012: Climate change and the South Asian summer monsoon. Nat. Climate Change, 2, 587–595, doi:10.1038/nclimate1495.

    • Search Google Scholar
    • Export Citation
  • Turner, A. G., P. M. Inness, and J. M. Slingo, 2007: The effect of doubled CO2 and model basic state biases on the monsoon-ENSO system. I: Mean response and interannual variability. Quart. J. Roy. Meteor. Soc., 133, 11431157, doi:10.1002/qj.82.

    • Search Google Scholar
    • Export Citation
  • Ueda, H., A. Iwai, K. Kuwako, and M. E. Hori, 2006: Impact of anthropogenic forcing on the Asian summer monsoon as simulated by eight GCMs. Geophys. Res. Lett., 33, doi:10.1029/2005GL025336.

    • Search Google Scholar
    • Export Citation
  • Van der Ent, R. J., O. A. Tuinenburg, H.-R. Knoche, H. Kunstmann, and H. H. G. Savenije, 2013: Should we use a simple or complex model for moisture recycling and atmospheric moisture tracking? Hydrol. Earth Syst. Sci., 17, 48694884, doi:10.5194/hess-17-4869-2013.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and B. J. Soden, 2007: Global warming and the weakening of the tropical circulation. J. Climate, 20, 43164340, doi:10.1175/JCLI4258.1.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., B. J. Soden, A. T. Wittenberg, I. M. Held, A. Leetmaa, and M. J. Harrison, 2006: Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing. Nature, 441, 7376, doi:10.1038/nature04744.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., V. O. Magaña, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. J. Geophys. Res., 103, 14 45114 510, doi:10.1029/97JC02719.

    • Search Google Scholar
    • Export Citation
  • Yoon, J.-H., and T.-C. Chen, 2005: Water vapor budget of the Indian monsoon depression. Tellus, 57A, 770782, doi:10.1111/j.1600-0870.2005.00145.x.

    • Search Google Scholar
    • Export Citation
Save