• Adler, R. F., and Coauthors, 2003: The version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 11471167.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., and B. J. Soden, 2007: Large discrepancy between observed and simulated precipitation trends in the ascending and descending branches of the tropical circulation. Geophys. Res. Lett., 34, L18705, doi:10.1029/2007GL031460.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., and B. J. Soden, 2008: Atmospheric warming and the amplification of precipitation extremes. Science, 321, 14811484.

  • Allan, R. P., B. J. Soden, V. O. John, W. Ingram, and P. Good, 2010: Current changes in tropical precipitation. Environ. Res. Lett., 5, 025205, doi:10.1088/1748-9326/5/2/025205.

    • Search Google Scholar
    • Export Citation
  • Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 224232.

  • Bengtsson, L., K. I. Hodges, and E. Roeckner, 2006: Storm tracks and climate change. J. Climate, 19, 35183543.

  • Biasutti, M., and A. H. Sobel, 2009: Delayed Sahel rainfall and global seasonal cycle in a warmer climate. Geophys. Res. Lett., 36, L23707, doi:10.1029/2009GL041303.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and J. D. Neelin, 2004: Mechanisms of global warming impacts on regional tropical precipitation. J. Climate, 17, 26882701.

  • Chou, C., and M.-H. Lo, 2007: Asymmetric responses of tropical precipitation during ENSO. J. Climate, 20, 34113433.

  • Chou, C., and J.-Y. Tu, 2008: Hemispherical asymmetry of tropical precipitation in ECHAM5/MPI-OM during El Niño and under global warming. J. Climate, 21, 13091332.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and C.-A. Chen, 2010: Depth of convection and the weakening of tropical circulation in global warming. J. Climate, 23, 30193030.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, J.-Y. Tu, and C.-T. Chen, 2006: Regional tropical precipitation change mechanisms in ECHAM4/OPYC3 under global warming. J. Climate, 19, 42074223.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J.-Y. Tu, and P.-H. Tan, 2007: Asymmetry of tropical precipitation change under global warming. Geophys. Res. Lett., 34, L17708, doi:10.1029/2007GL030327.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, C.-A. Chen, and J.-Y. Tu, 2009: Evaluating the “rich-get-richer” mechanism in tropical precipitation change under global warming. J. Climate, 22, 19822005.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., C. M. Johanson, J. M. Wallace, and T. Reichler, 2006: Enhanced mid-latitude tropospheric warming in satellite measurements. Science, 312, 1179, doi:10.1126/science.1125566.

    • 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.

  • Kniveton, D. R., R. Layberry, C. J. R. Williams, and M. Peck, 2009: Trends in the start of the wet season over Africa. Int. J. Climatol., 29, 12161225.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., and S. Manabe, 1995: Time-mean response over the tropical Pacific to increased CO2 in a coupled ocean–atmosphere model. J. Climate, 8, 21812199.

    • Search Google Scholar
    • Export Citation
  • Li, H., T. Zhou, and C. Li, 2010: Decreasing trend in global land monsoon precipitation over the past 50 years simulated by a coupled climate model. Adv. Atmos. Sci., 27, 285292.

    • Search Google Scholar
    • Export Citation
  • Liu, S. C., C. Fu, C.-J. Shiu, J.-P. Chen, and F. Wu, 2009: Temperature dependence of global precipitation extremes. Geophys. Res. Lett., 36, L17702, doi:10.1029/2009GL040218.

    • Search Google Scholar
    • Export Citation
  • Lu, J., G. A. Vecchi, and T. Reichler, 2007: Expansion of the Hadley cell under global warming. Geophys. Res. Lett., 34, L06805, doi:10.1029/2006GL028443.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Neelin, J. D., M. Münnich, H. Su, J. E. Meyerson, and C. E. Holloway, 2006: Tropical drying trends in global warming models and observations. Proc. Natl. Acad. Sci. USA, 103, 61106115.

    • Search Google Scholar
    • Export Citation
  • O’ Gorman, P. A., and T. Schneider, 2009: The physical basis for increases in precipitation extremes in simulations of 21st-century climate change. Proc. Natl. Acad. Sci. USA, 106, 14 77314 777.

    • Search Google Scholar
    • Export Citation
  • Previdi, M., and B. G. Liepert, 2007: Annular modes and Hadley cell expansion under global warming. Geophys. Res. Lett., 34, L22701, doi:10.1029/2007GL031243.

    • Search Google Scholar
    • Export Citation
  • Richter, I., and S.-P. Xie, 2008: Muted precipitation increase in global warming simulations: A surface evaporation perspective. J. Geophys. Res., 113, D24118, doi:10.1029/2008JD010561.

    • Search Google Scholar
    • Export Citation
  • Seager, R., and G. A. Vecchi, 2010: Greenhouse warming and the 21th century hydroclimate of southwestern North America. Proc. Natl. Acad. Sci. USA, 107, 21 27721 282.

    • Search Google Scholar
    • Export Citation
  • Seager, R., N. Naik, and G. A. Vecchi, 2010: Thermodynamic and dynamic mechanisms for large-scale changes in the hydrological cycle in response to global warming. J. Climate, 23, 46514668.

    • Search Google Scholar
    • Export Citation
  • Seidel, D. J., Q. Fu, W. J. Randel, and T. J. Reichler, 2007: Widening of the tropical belt in a changing climate. Nat. Geosci., 1, 2124, doi:10.1038/ngeo.2007.38.

    • Search Google Scholar
    • Export Citation
  • Stephens, G. L., and T. D. Ellis, 2008: Controls of global-mean precipitation increases in global warming GCM experiments. J. Climate, 21, 61416155.

    • Search Google Scholar
    • Export Citation
  • Sun, Y., S. Solomon, A. Dai, and R. W. Portmann, 2007: How often will it rain? J. Climate, 20, 48014818.

  • Sun, Y., Y. Ding, and A. Dai, 2010: Changing links between South Asian summer monsoon circulation and tropospheric land-sea thermal contrasts under a warming scenario. Geophys. Res. Lett., 37, L02704, doi:10.1029/2009GL041662.

    • Search Google Scholar
    • Export Citation
  • Tan, P.-H., C. Chou, and J.-Y. Tu, 2008: Mechanisms of global warming impacts on robustness of tropical precipitation asymmetry. J. Climate, 21, 55855602.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B. Parsons, 2003: The changing character of precipitation. Bull. Amer. Meteor. Soc., 84, 12051217.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and Coauthors, 2007: Observations: Surface and atmospheric climate change. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 235–336.

    • 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.

  • Wang, B., and Q. Ding, 2006: Changes in global monsoon precipitation over the past 56 years. Geophys. Res. Lett., 33, L06711, doi:10/1029/2005GL025347.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., L. Ricciardulli, K. Hilburn, and C. Mears, 2007: How much more rain will global warming bring? Science, 317, 233235.

  • Xie, S.-P., C. Deser, G. A. Vecchi, J. Ma, H. Teng, and A. T. Wittenberg, 2010: Global warming pattern formation: Sea surface temperature and rainfall. J. Climate, 23, 966986.

    • Search Google Scholar
    • Export Citation
  • Yin, J. H., 2005: A consistent poleward shift of the storm tracks in simulations of 21st century climate. Geophys. Res. Lett., 32, L18701, doi:10.1029/2005GL023684.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Climatology (1981–2000) of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation from the 17 multimodel ensembles. The stippling denotes areas with the annual range of precipitation >1 mm day−1 and 100% of annual mean precipitation. The unit is mm day−1.

  • View in gallery

    As in Fig. 1, but for the GPCP precipitation in 1979–2008.

  • View in gallery

    Time series (1980–2100) for global averages of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation calculated from varied (solid) and fixed (dashed) maximum and minimum seasons. The blue curve in (c) is global averages of annual mean precipitation. The unit is mm day−1. The scale for the solid cures is on the left and for the dashed and blue curves is on the right.

  • View in gallery

    Linear trends of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation in 1981–2100. The unit is mm day−1 century−1. The trends that passed the 95% statistical confidence level of a Student’ s t test are stippled. In (a) and (b), the hatching denotes areas with pressure velocity ω at 500 hPa negative (i.e., ascents). In (c), the hatching denotes areas with distinct wet and dry seasons, the annual range of precipitation >1 mm day−1 and 100% of annual mean precipitation.

  • View in gallery

    Time series (1981–2100) of the moisture budget (see text) for maximum precipitation: (a) −〈ωpq〉, (b) −〈v · ∇q〉, (c) E, (d) residual, (e) , (f) , (g) vertically integrated water vapor, and (h) pressure velocity at 500 hPa. The unit is W m−2 in (a)–(f), kg kg−1 in (g), and Pa s−1 in (h).

  • View in gallery

    As in Fig. 5, but for minimum precipitation.

  • View in gallery

    Time series (1981–2100) of the moisture budget for the annual range of precipitation: (a) −〈ωpq〉, (b) −〈v · ∇q〉, (c) E, (d) residual, (e) , and (f) . All units are W m−2.

  • View in gallery

    As in Fig. 4, but for −〈ωpq〉. The unit is W m−2 century−1.

  • View in gallery

    As in Fig. 8, but for . The dashed lines denote at 500 hPa.

  • View in gallery

    As in Fig. 8, but for . The dashed lines denote ω′ = 0 at 500 hPa.

  • View in gallery

    As in Fig. 8, but for −〈v · ∇q〉.

  • View in gallery

    As in Fig. 8, but for evaporation.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1368 599 68
PDF Downloads 1103 510 41

Changes in the Annual Range of Precipitation under Global Warming

View More View Less
  • 1 Research Center for Environmental Changes, Academia Sinica, and Department of Atmospheric Sciences, National Taiwan University, Taipei, Taiwan
  • | 2 Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan
Full access

Abstract

The annual range of precipitation, which is the difference between maximum and minimum precipitation within a year, is examined in climate model simulations under global warming. For global averages, the annual range of precipitation tends to increase as the globe warms. On a regional basis, this enhancement is found over most areas of the world, except for the bands along 30°S and 30°N. The enhancement in the annual range of precipitation is mainly associated with larger upward trends of maximum precipitation and smaller upward trends or downward trends of minimum precipitation. Based on the moisture budget analysis, the dominant mechanism is vertical moisture advection, both on a global average and on a regional scale. The vertical moisture advection, moisture convergence induced by vertical motion, includes the thermodynamic component, which is associated with increased water vapor, and the dynamic component, which is associated with changes in circulation. Generally, the thermodynamic component enhances the annual range of precipitation, while the dynamic component tends to reduce it. Evaporation has a positive contribution to both maximum and minimum precipitation, but very little to the annual range of precipitation. Even though evaporation and horizontal moisture advection are small for a global average, they could be important on a regional basis.

Corresponding author address: Chia Chou, Research Center for Environmental Changes, Academia Sinica, P.O. Box 1-48, Taipei 11529, Taiwan. E-mail: chiachou@rcec.sinica.edu.tw

Abstract

The annual range of precipitation, which is the difference between maximum and minimum precipitation within a year, is examined in climate model simulations under global warming. For global averages, the annual range of precipitation tends to increase as the globe warms. On a regional basis, this enhancement is found over most areas of the world, except for the bands along 30°S and 30°N. The enhancement in the annual range of precipitation is mainly associated with larger upward trends of maximum precipitation and smaller upward trends or downward trends of minimum precipitation. Based on the moisture budget analysis, the dominant mechanism is vertical moisture advection, both on a global average and on a regional scale. The vertical moisture advection, moisture convergence induced by vertical motion, includes the thermodynamic component, which is associated with increased water vapor, and the dynamic component, which is associated with changes in circulation. Generally, the thermodynamic component enhances the annual range of precipitation, while the dynamic component tends to reduce it. Evaporation has a positive contribution to both maximum and minimum precipitation, but very little to the annual range of precipitation. Even though evaporation and horizontal moisture advection are small for a global average, they could be important on a regional basis.

Corresponding author address: Chia Chou, Research Center for Environmental Changes, Academia Sinica, P.O. Box 1-48, Taipei 11529, Taiwan. E-mail: chiachou@rcec.sinica.edu.tw

1. Introduction

As anthropogenic forcings, such as greenhouse gases, become pronounced, more and more climate changes are found in the world. Global warming is one of the most apparent climate changes. Besides the increase of surface temperature, other climate changes are also gradually becoming clear; the change in precipitation is one of them. Under global warming, both observations and model simulations show an increase in the global average of mean precipitation, but with a strong spatial variation (i.e., increases in some regions but decreases in others; Allen and Ingram 2002; Meehl et al. 2007; Trenberth et al. 2007). The change in mean precipitation is associated with changes not only in precipitation intensity but also in precipitation frequency. Previous studies show that both intensity and frequency increase for heavy rainfall under the impact of global warming (e.g., Trenberth et al. 2003; Sun et al. 2007; Allan and Soden 2008; Liu et al. 2009; O’ Gorman and Schneider 2009; Allan et al. 2010).

Regarding the spatial variation of the change in mean precipitation, a tendency of wet regions getting wetter and dry regions getting drier has been found in both observations and model simulations (e.g., Allen and Ingram 2002; Neelin et al. 2006; Allan and Soden 2007; Wentz et al. 2007; Chou et al. 2009). One of the major mechanisms that induce this climate change is associated with the “rich get richer” mechanism, which is induced by increased water vapor due to global warming (Chou and Neelin 2004; Chou et al. 2009). In a warmer atmosphere, the tropospheric water vapor increases and roughly follows the Clausius–Clapeyron relationship with a constant relative humidity, at a rate of 7.5% K−1 (Allen and Ingram 2002; Held and Soden 2006; Stephens and Ellis 2008; Trenberth et al. 2003; Vecchi and Soden 2007; Wentz et al. 2007). The increased water vapor concentrates in the lower troposphere. Over climatologically wet regions dominated by ascending motion, the upward motion induces positive anomalies of moisture convergence, which then enhance precipitation. Over climatologically dry regions dominated by descending motion, on the other hand, the downward motion causes negative anomalies of moisture convergence, which create a unfavorable condition for processes that induce precipitation. This effect of increased water vapor transported by mean circulation is referred to as a thermodynamic component (Held and Soden 2006; Chou et al. 2009; Seager et al. 2010). The thermodynamic component is relatively robust among climate models. The change in precipitation modifies the release of latent heat, which alters the corresponding vertical motion. Precipitation is further changed via this so-called dynamic component, which is associated with changes in tropical circulation. The dynamic component is more complicated than the thermodynamic component and causes a disagreement in regional precipitation changes between climate models (Chou et al. 2009).

Based on the idea of the thermodynamic component, changes in precipitation could vary with seasons if the mean tropical circulation has a strong seasonal cycle, such as the Hadley circulation. Previous studies (Chou et al. 2007; Chou and Tu 2008; Tan et al. 2008) show a hemispherical asymmetry of tropical precipitation anomalies between two sides of the equator under global warming. Over the summer hemisphere, which is dominated by the ascending branch of the Hadley circulation, tropical precipitation (0°–30°) shows an upward trend. Over the winter hemisphere, which is dominated by the descending branch of the Hadley circulation, tropical precipitation has a slightly downward trend or is unchanged. Since summer is usually a rainy season and winter is a dry season, the hemispherical asymmetry of tropical precipitation implies that a wet season becomes wetter and a dry season becomes drier. In other words, the tendency of precipitation changes in the spatial variation (i.e., a wet region becoming wetter and a dry region becoming drier) is also found in the temporal variation. This asymmetry of tropical precipitation is also found in an interannual variation associated with ENSO, but due to a different mechanism (Chou and Lo 2007; Chou and Tu 2008).

The seasonal dependence of precipitation trends implies a possible climate change in the seasonal cycle of precipitation. Thus, we would like to examine in this study the climate change in the seasonal cycle, particularly in the annual range of precipitation, which is the difference between maximum and minimum precipitation within a year. In section 2, we first describe the data used in this study and define an annual range of precipitation, which is slightly different from the conventional definition. We then examine changes in the annual range of precipitation for global averages and spatial distribution in section 3. Mechanisms that induce these changes in the annual range of precipitation are discussed in section 4, followed by a discussion and the conclusions.

2. Data and analysis method

Data used in this study are from the World Climate Research Programme’ s (WCRP’ s) Coupled Model Intercomparison Project phase 3 (CMIP3) multimodel dataset (Table 1). The A1B scenario of anthropogenic emissions is used for the twenty-first-century global warming simulations since almost every climate model performed simulations with this scenario. Because of data availability, only one realization from each of the 17 models is analyzed. The results shown in this study are all multimodel ensembles from these 17 models. Observed monthly precipitation from the Global Precipitation Climatology Project (GPCP; Adler et al. 2003) is also used for comparison.

Table 1.

A list of the 17 coupled atmosphere–ocean climate model simulations in the A1B scenario from the CMIP3 archive.

Table 1.

Conventionally, an annual range of precipitation is defined as precipitation differences between fixed wet and dry seasons for the entire study domain and period (e.g., Wang and Ding 2006; Li et al. 2010), so both wet and dry seasons do not vary with time and space. This definition might not be able to give us the exact annual range of precipitation in a given year and at a given grid, since an annual range of precipitation should be the largest change of precipitation within one year (i.e., maximum minus minimum precipitation). Moreover, wet and dry seasons could shift under global warming, so the fixed wet and dry seasons might not be appropriate. Here we define the annual range of precipitation as the difference between the maximum and minimum precipitation in each year and at each grid. By using this definition for the annual range of precipitation, the seasons with maximum and minimum precipitation vary from year to year and are different between grids even in the same year. This definition can give us the exact range of how precipitation varies within one year and also avoids the effect of the shift of wet and dry seasons on the annual range of precipitation, which is another interesting issue (Biasutti and Sobel 2009; Kniveton et al. 2009). Here we focus on seasonal averages of precipitation, so the maximum and minimum precipitation amounts are a 3-month averages.

Figure 1 shows the spatial distributions of maximum and minimum precipitation, and the annual range of precipitation averaged over the period of 1981–2000 for a multimodel ensemble. To obtain the ensemble, we first calculate maximum and minimum precipitation for each model, and then average all 17 models. We note that the seasons for maximum and minimum precipitation are different between grids. The distribution of maximum precipitation is very similar to annual precipitation, but the maximum precipitation over oceans is much larger than that over land (Fig. 1a). This land–sea difference of the maximum precipitation is not clear in observations, particularly over South America (Fig. 2a). The minimum precipitation, on the other hand, shows much weaker amplitudes over the entire world, except for small areas within the ITCZ, the South Pacific convergence zone (SPCZ), and the Amazon, in which the minimum precipitation is over 4 mm day−1 (Fig. 1b). The observed minimum precipitation is slightly larger than the model ensemble over these regions (Fig. 2b). Most larger annual ranges of precipitation occur in the tropics, except for the equatorial region (Figs. 1c and 2c), which is different from the distribution of the maximum precipitation. We note that the annual range of precipitation is more meaningful over areas with distinct wet and dry seasons. Thus, we use a simple criterion to define these areas: the annual range of precipitation larger than 1 mm day−1 and 100% of annual mean precipitation (stippling in Figs. 1c and 2c). The land–sea difference in magnitude that is found in the maximum precipitation (i.e., larger over oceans and smaller over land) is much weaker in the annual range of precipitation. This implies that the simulated annual range of precipitation is slightly more similar to observations (Figs. 1c and 2c).

Fig. 1.
Fig. 1.

Climatology (1981–2000) of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation from the 17 multimodel ensembles. The stippling denotes areas with the annual range of precipitation >1 mm day−1 and 100% of annual mean precipitation. The unit is mm day−1.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

Fig. 2.
Fig. 2.

As in Fig. 1, but for the GPCP precipitation in 1979–2008.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

To understand changes in the annual range of precipitation, the vertically integrated moisture budget is used, which can be written as
e1
e2
where pressure velocity ω is assumed to be zero at the surface ps and at tropopause pT, and p and 〈〉 denote pressure and a mass integration through the entire troposphere, respectively. Specific humidity q is in energy units by absorbing the latent heat per unit mass L. For simplicity and consistency, the precipitation P is in energy units (W m−2) which, divided by 28 [=L/day (in seconds)], becomes millimeters per day. Here E is evaporation and v is horizontal velocity. The convergence of moisture flux −〈∇ · vq〉 on the right of (1) can be divided into two parts: the vertical moisture advection −〈ωpq〉 and the horizontal moisture advection −〈v · ∇q〉. The vertical moisture advection, the first term on the right of (2), is the part of the convergence of moisture flux induced by vertical motion or low-level convergence. The horizontal moisture advection, the second term on the right of (2), is the part of the convergence of moisture flux associated with horizontal velocity. The last term on the right of (2) δ is a residual term, which includes transient eddies and nonlinear effects.
Changes of the vertical moisture advection can be further divided into two terms:
e3
where () is climatology from 1981 to 2000 and ()′ is the change. The first term on the right is associated with changes in water vapor, which is mainly induced by temperature changes, so it is usually termed the thermodynamic component (Held and Soden 2006; Chou et al. 2009; Seager et al. 2010). The second term on the right is related to changes in vertical velocity, which is associated with tropical circulation, so it is termed the dynamic component (Held and Soden 2006; Chou et al. 2009; Seager et al. 2010). Here we neglect the nonlinear term −〈ω′∂pq′〉.

3. Changes in the annual range of precipitation

Figure 3 shows the time series for global averages of the maximum and minimum precipitation and the annual range of precipitation from 1980 to 2099 (solid curves), which are spatial averages of maximum and minimum precipitation at each grid point. The global average of precipitation in the maximum season (i.e., the season with maximum precipitation) shows a very clear upward trend, with a rate of 0.297 mm day−1 century−1. On the other hand, the average in the minimum season, the season with minimum precipitation, shows a very weak upward trend, with a rate of 0.016 mm day−1 century−1. Since the maximum precipitation increases faster than the minimum precipitation, the annual range of precipitation is widening, with a rate of around 0.282 mm day−1 century−1. This widening in the annual range of precipitation is much faster than the global average of annual mean precipitation shown in Fig. 3c. We note that if using monthly, instead of seasonal averages, the enhancement of the annual range of precipitation will be stronger, at a rate of around 0.427 mm day−1 century−1. All three trends have exceeded the 95% confidence level of a Student’ s t test. In other words, the seasonal cycle of precipitation tends to be stronger when the globe becomes warmer, which is mainly due to much wetter rainy seasons and relatively unchanged dry seasons. This result is similar to an enhancement in the hemispherical asymmetry of tropical precipitation (Chou et al. 2007; Tan et al. 2008), in which the precipitation enhances over the summer hemisphere and changes very little or slightly decreases over the winter hemisphere. Since the precipitation in the summer hemisphere might not always be an annual maximum at each grid point, the corresponding upward trends for the summer precipitation and the difference between summer and winter are smaller than the trends for the maximum precipitation and the annual range precipitation discussed here (Fig. 3). When using fixed wet and dry seasons in each year and at each grid, the corresponding upward trends are also weaker than those obtained from varied maximum and minimum seasons (dashed curves). We note that there is a jump around 2000, which might be due to a shift of the seasonal cycle and needs a further investigation.

Fig. 3.
Fig. 3.

Time series (1980–2100) for global averages of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation calculated from varied (solid) and fixed (dashed) maximum and minimum seasons. The blue curve in (c) is global averages of annual mean precipitation. The unit is mm day−1. The scale for the solid cures is on the left and for the dashed and blue curves is on the right.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

We next examine the spatial distribution of these long-term trends (1980–2099), beginning with the maximum precipitation (Fig. 4a). Most areas in the world show clear upward trends, with the strongest trends over the tropical Pacific and Indian Oceans. In the tropics, dominated by ascending motion, the distribution of the trends is similar to the climatology of the maximum precipitation shown in Fig. 1a, with a slightly eastward shift of the maximum trends, which could correspond to the enhanced equatorial warming response of SST that is induced by global warming (e.g., Meehl et al. 2007). Downward trends are found only over midlatitudes along 30° in the Southern and Northern Hemispheres, which are mainly dominated by descending motion. Overall, the sign of the trend is roughly consistent with the direction of the associated mean vertical motion (i.e., upward trends over ascending regions and downward trends over descending regions). Unlike the maximum precipitation, the trends of the minimum precipitation do not follow the direction of vertical motion very well (Fig. 4b). Most downward trends are found within 45°S–45°N, where they are dominated by descending motion. Most upward trends, on the other hand, are found over high latitudes, where they are also mostly dominated by descending motion. The strongest upward trends are over the equatorial central Pacific, where the trends of the maximum precipitation are also the largest. This might be due to the enhanced equatorial warming response of SST, similar to the possible cause for the maximum precipitation. Because of the strong cancellation between the downward trends within 45°S–45°N and the upward trends at high latitudes, the global average of the minimum precipitation is nearly unchanged (Fig. 3b). An examination of the annual range of precipitation reveals that it is enhanced almost everywhere in the world, except for narrow bands near 30°S and 30°N (Fig. 4c). The distribution of the trends is similar to that in the maximum precipitation, even over high latitudes where the trends of the minimum precipitation are upward. At high latitudes, the enhancement of the annual range of precipitation is associated more with smaller upward trends of the minimum precipitation than with the maximum precipitation. In the tropics (30°S–30°N), the enhancement of the annual range of precipitation is associated with the upward trends of the maximum precipitation and the downward trends of the minimum precipitation. The reduction of the annual range of precipitation near 30°S and 30°N is due to faster downward trends of the maximum precipitation than of the minimum precipitation. Moreover, stronger upward trends of the annual range of precipitation tend to occur over regions with distinct wet and dry seasons (i.e., the annual range of precipitation larger than 1 mm day−1 and 100% of the annual mean precipitation; Fig. 1c). This means that the annual range of precipitation that has already been large will become even larger under global warming. The spatial pattern of changes in the annual range of precipitation is similar to changes in annual mean precipitation, such as in Meehl et al. (2007) and Seager et al. (2010), since the annual range of precipitation is more associated with maximum than minimum precipitation and the spatial pattern of maximum precipitation is relatively similar to that of annual mean precipitation. However, over areas with less seasonality of precipitation, such as near the equatorial western Pacific and the Maritime Continent (Fig. 1), the enhancement in the annual range of precipitation is relatively smaller than that over other tropical convergence zones (Fig. 4).

Fig. 4.
Fig. 4.

Linear trends of (a) maximum precipitation, (b) minimum precipitation, and (c) the annual range of precipitation in 1981–2100. The unit is mm day−1 century−1. The trends that passed the 95% statistical confidence level of a Student’ s t test are stippled. In (a) and (b), the hatching denotes areas with pressure velocity ω at 500 hPa negative (i.e., ascents). In (c), the hatching denotes areas with distinct wet and dry seasons, the annual range of precipitation >1 mm day−1 and 100% of annual mean precipitation.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

4. Moisture budget and mechanisms

a. Global moisture budget

To understand mechanisms that induce the changes associated with precipitation, we first examine global averages of the moisture budget in (2). Figure 5 shows the time series of the moisture budget associated with maximum precipitation. The vertical moisture advection −〈ωpq〉 shows a clear upward trend (Fig. 5a), with a rate of around 5.142 W m−2 century−1, which is equivalent to 0.184 mm day−1 century−1. The horizontal moisture advection −〈v · ∇q〉 tends to increase (Fig. 5b), but the trend is very weak. Evaporation E shows an upward trend at a rate of around 3.31 W m−2 century−1 or 0.118 mm day−1 century−1 (Fig. 5c). The residual δ tends to decrease (Fig. 5d) a little bit. Overall, −〈ωpq〉 contributes to ⅔ of the upward trend of the maximum precipitation, while evaporation contributes to the remaining ⅓ of the trend of the maximum precipitation. In other words, the most dominant effect for inducing the upward trend of maximum precipitation is the vertical moisture advection. We further examine changes in the thermodynamic and dynamic component (i.e., and ) in Figs. 5e,f. The thermodynamic component increases, while the dynamic component decreases. To understand the trends of the thermodynamic and dynamic components, atmospheric moisture and vertical velocity are also examined. Both the water vapor and the vertical velocity show upward trends (Figs. 5g,h), which indicates an increase in water vapor and a reduction in ascending motion as the globe warms up. In the thermodynamic component, has a positive contribution to the upward trend of −〈ωpq〉 (i.e., ), because the mean vertical motion is upward in the maximum season, the season with maximum precipitation, and the global warming-induced positive anomalies of water vapor concentrates in the lower troposphere. In the dynamic component, on the other hand, the reduced upward motion tends to reduce −〈ωpq〉 (i.e., ). Overall, the increased −〈ωpq〉 is mainly associated with changes in atmospheric water vapor, the thermodynamic component. The reduction of the upward motion is consistent with a weakening of tropical circulation in climate model simulations, which has been discussed by many studies (e.g., Knutson and Manabe 1995; Held and Soden 2006; Vecchi and Soden 2007; Chou and Chen 2010). We note that the dynamic component could also be associated with transient eddies, especially at mid- and high latitudes (e.g., Seager et al. 2010).

Fig. 5.
Fig. 5.

Time series (1981–2100) of the moisture budget (see text) for maximum precipitation: (a) −〈ωpq〉, (b) −〈v · ∇q〉, (c) E, (d) residual, (e) , (f) , (g) vertically integrated water vapor, and (h) pressure velocity at 500 hPa. The unit is W m−2 in (a)–(f), kg kg−1 in (g), and Pa s−1 in (h).

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

For precipitation in the minimum season (the season with minimum precipitation), the moisture budget is shown in Fig. 6. The vertical moisture advection −〈ωpq〉 shows a downward trend with a rate of around −2.035 W m−2 century−1, which is equivalent to 0.073 mm day−1 century−1 (Fig. 6a), which is the opposite and much weaker in amplitude compared to the upward trend in the maximum season (Fig. 5a). However, since minimum precipitation is smaller than maximum precipitation in amplitude, the rates of the changes in both seasons become closer in percentage, with 9% and 7% century−1 in the maximum and minimum seasons, respectively. Since −〈ωpq〉 is always a negative value in the minimum season, the downward trend indicates an enhancement of the magnitude of the vertical moisture advection. Examining changes in atmospheric water vapor and vertical velocity that can affect −〈ωpq〉 reveals that the water vapor increases (Fig. 6g), but the corresponding descending motion reduces (Fig. 6h), a weakening of tropical circulation. Thus, the enhancement in the magnitude of −〈ωpq〉, a downward trend (Fig. 6a), is mainly associated with an increase in atmospheric water vapor, the thermodynamic component (i.e., ; Fig. 6e). The dynamic component associated with changes in vertical motion actually reduces the increasing rate in the magnitude of negative −〈ωpq〉 (i.e., ; Fig. 6f). Similar to those found in the maximum season, both the horizontal advection −〈v · ∇q〉 and the residual term are relatively unchanged. Evaporation in the minimum season shows an upward trend at a rate of around 3.38 W m−2 century−1 or 0.121 mm day−1 century−1, similar to the trend of evaporation in the maximum season. This implies that the upward trends of evaporation in both maximum and minimum seasons could be induced by similar mechanisms, such as those discussed in Richter and Xie (2008). The increase in evaporation under global warming is associated with increases in relative humidity and stability at the surface and a reduction in surface wind speed. For the ensemble shown here, the amplitude of the upward trend of evaporation is stronger than that of the downward trend of −〈ωpq〉, so the trend of minimum precipitation is slightly upward (Fig. 3b). However, it might slightly vary among models, so the trend of minimum precipitation might be downward, instead of upward, in some models, similar to those changes in precipitation found in the winter hemisphere (Tan et al. 2008). Overall, the trend of the minimum precipitation (Fig. 3b) is very weak because of the cancellation between the increases in the magnitude of the negative −〈ωpq〉 trend and the positive evaporation trend.

Fig. 6.
Fig. 6.

As in Fig. 5, but for minimum precipitation.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

Changes in the moisture budget for the annual range of precipitation are different from those found in the budgets for the maximum and minimum precipitation discussed above. The difference of −〈ωpq〉 between the maximum and minimum seasons shows a clear upward trend (Fig. 7a). This upward trend is around 7.177 W m−2 century−1 (or 0.256 mm day−1 century−1). It is close to the trend of the annual range precipitation shown in Fig. 3c. The rest of the terms are almost unchanged (Figs. 7b–d). The trends of the horizontal moisture advection −〈v · ∇q〉 and the residual term are already small in both maximum and minimum seasons, so the differences in these terms are small. We note that the residual term, which includes transient eddies, might be important on a regional scale (Seager and Vecchi 2010). It could possibly modify the annual range of precipitation slightly (Seager et al. 2010), but it will not alter the outcome (i.e., the widening of the annual range of precipitation). Evaporation shows similar upward trends in the maximum and minimum seasons, so the difference in evaporation does not change too much, as a result of the strong cancellation between these two seasons. Thus, the change in −〈ωpq〉, which is mainly associated with the thermodynamic component (Fig. 7e), is the only major mechanism that enhances the annual precipitation range. Almost no contribution is from evaporation, which is different in the moisture budget for maximum and minimum precipitation. We note that the dynamic component, which is associated with changes in vertical motion and the corresponding circulation, also affects the annual range of precipitation, narrowing but not widening the range (Fig. 7f).

Fig. 7.
Fig. 7.

Time series (1981–2100) of the moisture budget for the annual range of precipitation: (a) −〈ωpq〉, (b) −〈v · ∇q〉, (c) E, (d) residual, (e) , and (f) . All units are W m−2.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

b. Spatial distribution

We next examine the spatial distributions of the trends in the water vapor budget. Figure 8 shows the trends of −〈ωpq〉. In the maximum season (Fig. 8a), the spatial pattern of the −〈ωpq〉 trends is very similar to the distribution of the maximum precipitation trends (Fig. 4a). Upward trends are found in the tropics and high latitudes, while downward trends are over midlatitudes around 30°S and 30°N. The area with negative trends is slightly larger than that with the maximum precipitation trends. In the minimum season (Fig. 8b), on the other hand, downward (negative) trends of −〈ωpq〉 are found over most areas. Upward trends are only found over the equatorial central Pacific and around 60°S. This spatial pattern of −〈ωpq〉 is different from the distribution of the minimum precipitation trends (Fig. 4b), especially over high latitudes, where the trends of the minimum precipitation are upward, but the trends of −〈ωpq〉 are downward. At high latitudes, precipitation is not mainly associated with convection, so −〈ωpq〉 is less dominant. The contribution from high latitudes is one of the reasons that causes the difference of trends between the globally averaged minimum precipitation and −〈ωpq〉 in the minimum season (Figs. 3b and 6a). The trends in the lower latitudes can also contribute to this difference since the amplitude of the negative −〈ωpq〉 trends is stronger and the amplitude of the positive −〈ωpq〉 trends is weaker than the corresponding minimum precipitation trends shown in Fig. 4b. The differences of −〈ωpq〉 between the maximum and minimum seasons (Fig. 8c) show a very similar spatial pattern to the trends of the annual range of precipitation (Fig. 4c). This implies that −〈ωpq〉 dominates the changes in the annual range of precipitation not only on a global average, but also on a regional scale.

Fig. 8.
Fig. 8.

As in Fig. 4, but for −〈ωpq〉. The unit is W m−2 century−1.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

Changes in −〈ωpq〉 are mainly associated with the thermodynamic and dynamic components, which are related to changes in water vapor and vertical motion, respectively. We first examine the thermodynamic component (Fig. 9). Because tropospheric water vapor increases everywhere in the world in the maximum and minimum seasons (q′ > 0) and the increased water vapor concentrates in the lower troposphere, the sign of the thermodynamic component mainly depends on the corresponding mean vertical motion . In the maximum and minimum seasons, upward trends are found over areas with mean ascending motion , while downward trends are found over areas with mean descending motion . The trends for the differences of are upward almost everywhere in the world (Fig. 9c), with stronger magnitudes in the tropics. We note that the spatial distribution of the trends of the thermodynamic component (Fig. 9) is very close to the distributions in climatology for the maximum and minimum precipitation and the annual range of precipitation shown in Fig. 1.

Fig. 9.
Fig. 9.

As in Fig. 8, but for . The dashed lines denote at 500 hPa.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

We next examine the dynamic component, . The sign of is determined by the direction of the changes in vertical motion ω′ because mean tropospheric water vapor is positive everywhere and concentrates in the lower troposphere. In the maximum season, when dominated by mean ascending motion , the trend of the dynamic component shows a strong spatial variation, which is different from the pattern of (Figs. 9a and 10a). Upward trends are found at high latitudes and the equatorial Pacific, and are stronger at the equatorial Pacific. Most downward trends are found within 45°S–45°N. Compared to the magnitude of the upward trends, these downward trends are generally stronger. At high latitudes, has a similar effect as , which is positive. In the tropics, on the other hand, they are opposite. Most areas show negative values of the trend and positive values of the trend. Based on the direction of the −〈ωpq〉 trend (Fig. 8a), the thermodynamic component dominates in most tropical regions, while the dynamic component tends to reduce the amplitudes of the −〈ωpq〉 trends, which are positive over most tropical regions. However, in some regions, such as the equatorial Pacific and the central South America–Caribbean Sea, is clearly more pronounced than . These changes associated with changes in vertical velocity could be induced by either changes in local convection or a shift in large-scale circulation. In the minimum season (Fig. 10b), most areas are dominated by upward trends of the dynamic component, especially in the tropics, which is opposite to the trends of the thermodynamic component (Fig. 9b). Thus, the dynamic component tends to reduce the magnitudes of the −〈ωpq〉 trends in the minimum season, which is downward over most areas (Fig. 8b). The differences of the dynamic component between the maximum and minimum seasons are negative in most areas within 45°S–45°N, except the equatorial eastern Pacific (Fig. 10c). Weak upward trends are also found at high latitudes of the Northern Hemisphere. Similar to those trends of the dynamic component found in the maximum and minimum seasons (Figs. 10a,b), these trends are also opposite to the trends of the thermodynamic component (Fig. 9c) in most areas, except at high latitudes of the Northern Hemisphere.

Fig. 10.
Fig. 10.

As in Fig. 8, but for . The dashed lines denote ω′ = 0 at 500 hPa.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

Overall, thermodynamic component dominates the trends of −〈ωpq〉 in most regions, especially in the tropics. The dynamic component usually just slightly modifies the amplitudes of the trends, but it can also be important in some regions, such as in the equatorial Pacific. The mechanisms associated with −〈ωpq〉, which are discussed here, are similar to those mechanisms found for changes in mean precipitation (Seager et al. 2010). We note that the weaker impact of the dynamic component might also be due to the inconsistency of changes in vertical motion among climate models (O’ Gorman and Schneider 2009; Allan et al. 2010) since the results shown here are 17-model ensembles.

Figure 11 shows the distributions for the trends of −〈v · q〉. The trends in the maximum and minimum seasons and the trends for the differences between these two seasons show clear spatial variations. The horizontal moisture advection could be associated with the upped-ante mechanism (Chou and Neelin 2004; Chou et al. 2009), which reduces precipitation over convective margins. It could also be associated with a shift of an atmospheric circulation and transient eddies (e.g., Seager et al. 2010). The trends over most regions are relatively weak compared to the trends of −〈ωpq〉 (Fig. 8), but they can be strong in some regions, such as in the East Asian region and its neighboring oceans, and the Bay of Bengal. Over the East Asian region, the downward trend for the difference of −〈v · q〉 is associated with downward trends in the maximum season and upward trends in the minimum season. Over the Bay of Bengal, the trend for the difference of −〈v · q〉 range is associated with upward trends in the maximum season and downward trends in the minimum season.

Fig. 11.
Fig. 11.

As in Fig. 8, but for −〈v · ∇q〉.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

The trends of evaporation show increasing trends for most areas in both maximum and minimum seasons, with the strongest amplitudes over East Asia and the North Pacific (Figs. 12a,b). At high latitudes, the increase in evaporation (Fig. 12b) dominates the increase in precipitation in the minimum season (Fig. 4b), not the contribution of −〈ωq〉, which is negative (Fig. 8b). However, how the extra moisture transfers into precipitation should be examined in the future. Because of the cancellation, the trends for the differences of evaporation between the maximum and minimum seasons are relatively small for most areas, the tropics in particular, and their spatial pattern is very different from the distributions in the maximum and minimum seasons. The trends for the differences of evaporation show relatively strong amplitudes around 30°–60°N, such as strong downward trends over East Asia, the North Pacific, the North Atlantic, and Europe. Thus, evaporation could be important regionally, especially for its contribution to the annual range of precipitation. We note that the dipole pattern of the trends over the North Pacific is similar to that in the horizontal moisture advection (Fig. 11c). This could be due to changes in the Asian summer monsoon circulation, which should be further examined.

Fig. 12.
Fig. 12.

As in Fig. 8, but for evaporation.

Citation: Journal of Climate 25, 1; 10.1175/JCLI-D-11-00097.1

5. Discussion and conclusions

Global warming can modify precipitation characteristics, such as mean (amount), frequency and intensity. In this study, we focused on an annual range of precipitation, the difference between maximum and minimum precipitation, which is associated with a seasonal cycle of precipitation. For global averages, the enhancement in the annual range of precipitation is apparent, which is due to a faster increase of precipitation in the maximum season than in the minimum season (Fig. 3). Normalized by global averages of changes in surface temperature, the maximum precipitation increases at a rate of 3.64% °C−1, the minimum precipitation increases at a rate of 0.21% °C−1, and the annual range of precipitation widens at a rate of 4.54% °C−1. If considering monthly averages (i.e., the maximum and minimum months, instead of seasonal averages) the changing rates are also similar, at a rate of 4.06%, −0.03%, and 4.63% °C−1 for the maximum and minimum precipitation and the annual range of precipitation, respectively. To obtain the rate of the 17-model ensemble, we did not directly use the time series shown in Fig. 3. We calculated the rates for each model, and then averaged each of 17 models. Thus, the rate in the annual range of precipitation is slightly different from the rates of the maximum–minus–minimum precipitation. Compared to the increased rate of global averaged annual mean precipitation, which is around 2.97% °C−1, the maximum precipitation and the annual range of precipitation increase only slightly faster, but slower for the minimum precipitation. Unlike the result shown in Fig. 3c, in which the annual range of precipitation increases much faster than the global mean precipitation, this is due to larger amplitudes of the annual range of precipitation than of the global mean precipitation.

On a regional basis, the widening of the annual range of precipitation occurs almost everywhere in the world, except for two zonal bands around 30°N and 30°S (Fig. 4c), but the causes are different. At high latitudes, the widening of the annual range of precipitation is due to the faster increase of precipitation in the maximum season than in the minimum season, similar to the global averages. In the tropics, on the other hand, the widening over most areas is associated with an increase in precipitation in the maximum season and a decrease in precipitation in the minimum season. The reduced annual range of precipitation around 30°N and 30°S, where it is dominated by mean descending motion, is mainly associated with a faster reduction of precipitation in the maximum season than in the minimum season. Overall, the widening of the annual range of precipitation is more apparent over areas with distinct wet and dry seasons (i.e., changes in the seasonal cycle of precipitation).

To understand mechanisms that cause the widening of the annual range of precipitation, we examined a vertically integrated water vapor budget, which is equation (2). On both a global and regional basis, the dominant mechanism for inducing the enhancement of the annual range of precipitation is associated with the change in vertical moisture advection, −〈ωpq〉. Other effects, such as horizontal moisture advection and evaporation, are relatively weak because of different reasons. Horizontal moisture advection is generally small in the maximum and minimum seasons. Evaporation, on the other hand, is small because of the cancellation between the maximum and minimum seasons. The change in vertical moisture advection is mainly determined by changes in moisture and vertical motion, thermodynamics, and dynamic components, respectively. In general, the thermodynamic component is dominant over most areas and the dynamic component tends to oppose the effect of the thermodynamic component. That is why the increased rate of the annual range of precipitation for the global average (4.54% °C−1) is smaller than what the Clausius–Clapeyron thermal scaling (~7.5% °C−1) indicates.

In the thermodynamic component, since changes in the increases of water vapor are all positive and concentrate in the lower troposphere, the mean vertical motion becomes the determining factor for the sign of the thermodynamic component. Thus, the spatial patterns of the trends of the thermodynamic components are very close to the climatology of the corresponding precipitation. In the dynamic component, on the other hand, the determining factor is the change in vertical motion, which is much more complicated. In the tropics, the amplitudes of both ascending and descending motions generally reduce over most areas (i.e., a weakening of tropical circulation), so the dynamic component tends to reduce the annual range of precipitation. The poleward expansion of the Hadley circulation (Fu et al. 2006; Lu et al. 2007; Previdi and Liepert 2007; Seidel et al. 2007) and the poleward shift of transient eddies (Yin 2005; Bengtsson et al. 2006; Seager et al. 2010) could also be important. Tropical circulation may also shift regionally because of various effects, such as changes in the spatial distribution of SST, which can be associated with the feedback of ocean dynamics (e.g., Chou et al. 2006, 2009; Xie et al. 2010), and changes in land–sea contrast (e.g., Sun et al. 2010). At high latitudes, changes in vertical motion are consistent with mean vertical motion for most areas in the maximum season (i.e., an increase in the amplitude of vertical motion), so the dynamic component further increases the annual range of precipitation, similar to the thermodynamic component.

From a global point of view, the fact that the thermodynamic component is more important than the dynamic component implies that the atmospheric circulation might vary very little in its spatial distribution. It could also imply that changes in the atmospheric circulation are very inconsistent among climate models (O’ Gorman and Schneider 2009; Allan et al. 2010), while the thermodynamic component is more consistent. The only consistent change associated with the dynamic component or the atmospheric circulation is the weakening in its strength. Here we only examined changes in the annual range of precipitation from climate model simulations, so the results should be compared to observations, which is an ongoing study.

Acknowledgments

We thank Prof. R. Seager and two anonymous reviewers for their valuable suggestions for greatly improving the manuscript. We would also like to thank Ms. Sharon Lee for graphics. We acknowledge the international modeling groups for providing their data for analysis: the Program for Climate Model Diagnosis and Intercomparison (PCMDI) for collecting and archiving the model data, the JSC/CLIVAR Working Group on Coupled Modelling (WGCM) and their Coupled Model Intercomparison Project (CMIP) and Climate Simulation Panel for organizing the model data analysis activity, and the IPCC WG1 TSU for technical support. The IPCC Data Archive at Lawrence Livermore National Laboratory is supported by the Office of Science, U.S. Department of Energy. This work was supported by the National Science Council Grants NSC99-2111-M-001-003-MY3 and NSC98-2625-M-492-011.

REFERENCES

  • Adler, R. F., and Coauthors, 2003: The version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 11471167.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., and B. J. Soden, 2007: Large discrepancy between observed and simulated precipitation trends in the ascending and descending branches of the tropical circulation. Geophys. Res. Lett., 34, L18705, doi:10.1029/2007GL031460.

    • Search Google Scholar
    • Export Citation
  • Allan, R. P., and B. J. Soden, 2008: Atmospheric warming and the amplification of precipitation extremes. Science, 321, 14811484.

  • Allan, R. P., B. J. Soden, V. O. John, W. Ingram, and P. Good, 2010: Current changes in tropical precipitation. Environ. Res. Lett., 5, 025205, doi:10.1088/1748-9326/5/2/025205.

    • Search Google Scholar
    • Export Citation
  • Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 224232.

  • Bengtsson, L., K. I. Hodges, and E. Roeckner, 2006: Storm tracks and climate change. J. Climate, 19, 35183543.

  • Biasutti, M., and A. H. Sobel, 2009: Delayed Sahel rainfall and global seasonal cycle in a warmer climate. Geophys. Res. Lett., 36, L23707, doi:10.1029/2009GL041303.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and J. D. Neelin, 2004: Mechanisms of global warming impacts on regional tropical precipitation. J. Climate, 17, 26882701.

  • Chou, C., and M.-H. Lo, 2007: Asymmetric responses of tropical precipitation during ENSO. J. Climate, 20, 34113433.

  • Chou, C., and J.-Y. Tu, 2008: Hemispherical asymmetry of tropical precipitation in ECHAM5/MPI-OM during El Niño and under global warming. J. Climate, 21, 13091332.

    • Search Google Scholar
    • Export Citation
  • Chou, C., and C.-A. Chen, 2010: Depth of convection and the weakening of tropical circulation in global warming. J. Climate, 23, 30193030.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, J.-Y. Tu, and C.-T. Chen, 2006: Regional tropical precipitation change mechanisms in ECHAM4/OPYC3 under global warming. J. Climate, 19, 42074223.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J.-Y. Tu, and P.-H. Tan, 2007: Asymmetry of tropical precipitation change under global warming. Geophys. Res. Lett., 34, L17708, doi:10.1029/2007GL030327.

    • Search Google Scholar
    • Export Citation
  • Chou, C., J. D. Neelin, C.-A. Chen, and J.-Y. Tu, 2009: Evaluating the “rich-get-richer” mechanism in tropical precipitation change under global warming. J. Climate, 22, 19822005.

    • Search Google Scholar
    • Export Citation
  • Fu, Q., C. M. Johanson, J. M. Wallace, and T. Reichler, 2006: Enhanced mid-latitude tropospheric warming in satellite measurements. Science, 312, 1179, doi:10.1126/science.1125566.

    • 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.

  • Kniveton, D. R., R. Layberry, C. J. R. Williams, and M. Peck, 2009: Trends in the start of the wet season over Africa. Int. J. Climatol., 29, 12161225.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., and S. Manabe, 1995: Time-mean response over the tropical Pacific to increased CO2 in a coupled ocean–atmosphere model. J. Climate, 8, 21812199.

    • Search Google Scholar
    • Export Citation
  • Li, H., T. Zhou, and C. Li, 2010: Decreasing trend in global land monsoon precipitation over the past 50 years simulated by a coupled climate model. Adv. Atmos. Sci., 27, 285292.

    • Search Google Scholar
    • Export Citation
  • Liu, S. C., C. Fu, C.-J. Shiu, J.-P. Chen, and F. Wu, 2009: Temperature dependence of global precipitation extremes. Geophys. Res. Lett., 36, L17702, doi:10.1029/2009GL040218.

    • Search Google Scholar
    • Export Citation
  • Lu, J., G. A. Vecchi, and T. Reichler, 2007: Expansion of the Hadley cell under global warming. Geophys. Res. Lett., 34, L06805, doi:10.1029/2006GL028443.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Neelin, J. D., M. Münnich, H. Su, J. E. Meyerson, and C. E. Holloway, 2006: Tropical drying trends in global warming models and observations. Proc. Natl. Acad. Sci. USA, 103, 61106115.

    • Search Google Scholar
    • Export Citation
  • O’ Gorman, P. A., and T. Schneider, 2009: The physical basis for increases in precipitation extremes in simulations of 21st-century climate change. Proc. Natl. Acad. Sci. USA, 106, 14 77314 777.

    • Search Google Scholar
    • Export Citation
  • Previdi, M., and B. G. Liepert, 2007: Annular modes and Hadley cell expansion under global warming. Geophys. Res. Lett., 34, L22701, doi:10.1029/2007GL031243.

    • Search Google Scholar
    • Export Citation
  • Richter, I., and S.-P. Xie, 2008: Muted precipitation increase in global warming simulations: A surface evaporation perspective. J. Geophys. Res., 113, D24118, doi:10.1029/2008JD010561.

    • Search Google Scholar
    • Export Citation
  • Seager, R., and G. A. Vecchi, 2010: Greenhouse warming and the 21th century hydroclimate of southwestern North America. Proc. Natl. Acad. Sci. USA, 107, 21 27721 282.

    • Search Google Scholar
    • Export Citation
  • Seager, R., N. Naik, and G. A. Vecchi, 2010: Thermodynamic and dynamic mechanisms for large-scale changes in the hydrological cycle in response to global warming. J. Climate, 23, 46514668.

    • Search Google Scholar
    • Export Citation
  • Seidel, D. J., Q. Fu, W. J. Randel, and T. J. Reichler, 2007: Widening of the tropical belt in a changing climate. Nat. Geosci., 1, 2124, doi:10.1038/ngeo.2007.38.

    • Search Google Scholar
    • Export Citation
  • Stephens, G. L., and T. D. Ellis, 2008: Controls of global-mean precipitation increases in global warming GCM experiments. J. Climate, 21, 61416155.

    • Search Google Scholar
    • Export Citation
  • Sun, Y., S. Solomon, A. Dai, and R. W. Portmann, 2007: How often will it rain? J. Climate, 20, 48014818.

  • Sun, Y., Y. Ding, and A. Dai, 2010: Changing links between South Asian summer monsoon circulation and tropospheric land-sea thermal contrasts under a warming scenario. Geophys. Res. Lett., 37, L02704, doi:10.1029/2009GL041662.

    • Search Google Scholar
    • Export Citation
  • Tan, P.-H., C. Chou, and J.-Y. Tu, 2008: Mechanisms of global warming impacts on robustness of tropical precipitation asymmetry. J. Climate, 21, 55855602.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B. Parsons, 2003: The changing character of precipitation. Bull. Amer. Meteor. Soc., 84, 12051217.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and Coauthors, 2007: Observations: Surface and atmospheric climate change. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 235–336.

    • 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.

  • Wang, B., and Q. Ding, 2006: Changes in global monsoon precipitation over the past 56 years. Geophys. Res. Lett., 33, L06711, doi:10/1029/2005GL025347.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., L. Ricciardulli, K. Hilburn, and C. Mears, 2007: How much more rain will global warming bring? Science, 317, 233235.

  • Xie, S.-P., C. Deser, G. A. Vecchi, J. Ma, H. Teng, and A. T. Wittenberg, 2010: Global warming pattern formation: Sea surface temperature and rainfall. J. Climate, 23, 966986.

    • Search Google Scholar
    • Export Citation
  • Yin, J. H., 2005: A consistent poleward shift of the storm tracks in simulations of 21st century climate. Geophys. Res. Lett., 32, L18701, doi:10.1029/2005GL023684.

    • Search Google Scholar
    • Export Citation
Save