Climate Dynamics

, Volume 47, Issue 3–4, pp 1007–1027

Deciphering the desiccation trend of the South Asian monsoon hydroclimate in a warming world

  • R. Krishnan
  • T. P. Sabin
  • R. Vellore
  • M. Mujumdar
  • J. Sanjay
  • B. N. Goswami
  • F. Hourdin
  • J.-L. Dufresne
  • P. Terray
Article

DOI: 10.1007/s00382-015-2886-5

Cite this article as:
Krishnan, R., Sabin, T.P., Vellore, R. et al. Clim Dyn (2016) 47: 1007. doi:10.1007/s00382-015-2886-5

Abstract

Rising propensity of precipitation extremes and concomitant decline of summer-monsoon rains are amongst the most distinctive hydroclimatic signals that have emerged over South Asia since 1950s. A clear understanding of the underlying causes driving these monsoon hydroclimatic signals has remained elusive. Using a state-of-the-art global climate model with high-resolution zooming over South Asia, we demonstrate that a juxtaposition of regional land-use changes, anthropogenic-aerosol forcing and the rapid warming signal of the equatorial Indian Ocean is crucial to produce the observed monsoon weakening in recent decades. Our findings also show that this monsoonal weakening significantly enhances occurrence of localized intense precipitation events, as compared to the global-warming response. A 21st century climate projection using the same high-resolution model indicates persistent decrease of monsoonal rains and prolongation of soil drying. Critical value-additions from this study include (1) realistic simulation of the mean and long-term historical trends in the Indian monsoon rainfall (2) robust attributions of changes in moderate and heavy precipitation events over Central India (3) a 21st century projection of drying trend of the South Asian monsoon. The present findings have profound bearing on the regional water-security, which is already under severe hydrological-stress.

Keywords

Recent trends in the South Asian Monsoon High-resolution model simulations Regional hydroclimatic response to climate change 

Supplementary material

382_2015_2886_MOESM1_ESM.eps (609 kb)
Auxiliary Figure A1Time-series of year-wise count of heavy rainfall events (intensity  100 mm day−1) over Central India (74.5°E–86.5°E, 16.5°N–26.5°N). The counts are for the June–September monsoon season from 1951–2005 based on IMD observations (black line), HIST1 (brown solid line), HIST2 (brown dashed line), HISTNAT1 (blue solid line) and HISTNAT2 (blue dashed line). The linear least-square trends and their statistical significance are presented in Table 4. (EPS 609 kb)
382_2015_2886_MOESM2_ESM.eps (3.1 mb)
Auxiliary Figure A2Difference maps of precipitation (mm day−1, shaded) and 850 hPa winds (ms−1, vectors) (a) RCP4.5 minus HISTNAT1 (b) RCP4.5 minus HIST1. The mean of RCP4.5 is for the period 2006-2060. For HIST1 and HISTNAT1, the means are for the period 1951-2005. Note the persistence of weak SAM circulation and rainfall anomalies in the RCP4.5 projection. (EPS 3197 kb)
382_2015_2886_MOESM3_ESM.eps (3.1 mb)
Auxiliary Figure A3Tropospheric temperature (TT) and circulation response to anthropogenic influence: Map showing the difference in JJAS mean of TT (°C) and tropospheric circulation (vectors: ms−1) between HIST1 and HISTNAT1 for the period (1951-2005). The temperature and wind fields are vertically averaged between 600 and 200 hPa. Note that the TT response over the near-equatorial areas is warmer as compared to that of the extra-tropics (poleward of 30°N). The cyclonic circulation anomaly over West-Central Asia is associated with cold air advection and subsidence over the Indian subcontinent. The anticyclonic circulation anomaly over the Indian region indicates weakening of the SAM circulation. (EPS 3202 kb)
382_2015_2886_MOESM4_ESM.eps (4.8 mb)
Auxiliary Figure A4Climatological mean monsoon rainfall and 850 hPa winds from observations/reanalysis, LMDZ4 high-resolution simulations, IPSL-CM5A models. a, GPCP and NCEP b, HIST1 c, HIST2 d, IPSL-CM5A-MR e, IPSL-CM5A-LR. The means are for the period 1951-2005, except for GPCP rainfall which is for the period 1979-2009. Notice the severe underestimation of monsoon winds and precipitation, particularly over the Western Ghats in the IPSL-CM5A models. (EPS 4950 kb)
382_2015_2886_MOESM5_ESM.eps (333 kb)
Auxiliary Figure A5Coupled variability of monsoon precipitation and low-level winds in observationsand simulations. The first empirical orthogonal function (EOF1) of JJAS precipitation over western Ghats and west-central peninsular India for the period 1941-2005 from (a) Observations (b) HIST1 (c) IPSL-CM5-LR (d, e, f) corresponding principal component (PC1) time-series (g, h, i) Pattern obtained by regressing the 850 hPa winds over the Arabian Sea upon the PC1 time-series of rainfall. Note the decreasing trend of PC1 time-series in observations and HIST1 high-resolution simulation, but not in the IPSL-CM5-LR model. Consistent with the decreasing trend of PC1, the regression pattern of westerly winds indicate weakening of the monsoon flow in NCEP reanalysis and HIST1. In contrast, note that the wind variations in the IPSL-CM5-LR are anti-correlated with the increasing trend of PC1 time-series as seen from the easterly anomaly. (EPS 333 kb)
382_2015_2886_MOESM6_ESM.eps (1.8 mb)
Auxiliary Figure A6Spatial map of projected future changes in the seasonal monsoon rainfall. Least-square linear trend of June–September monsoon rainfall from the RCP4.5 simulation expressed as mm day−1 (45 years)−1 (a) (2006 – 2050) (b) (2051 – 2095). Only values exceeding the 95 % confidence level are displayed. (EPS 1870 kb)

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • R. Krishnan
    • 1
  • T. P. Sabin
    • 1
  • R. Vellore
    • 1
  • M. Mujumdar
    • 1
  • J. Sanjay
    • 1
  • B. N. Goswami
    • 1
    • 2
  • F. Hourdin
    • 3
  • J.-L. Dufresne
    • 3
  • P. Terray
    • 4
    • 5
  1. 1.Centre for Climate Change Research (CCCR)Indian Institute of Tropical Meteorology (IITM)PuneIndia
  2. 2.Indian Institute of ScienceEducation and Research (IISER)PuneIndia
  3. 3.Laboratoire Meteorologie Dynamique, (LMD/IPSL), Centre National de la Recherche Scientifique (CNRS)Université Pierre et Marie Curie (UPMC)/ENS/Ecole PolytechniqueParisFrance
  4. 4.LOCEAN LaboratorySorbonne Universités (UPMC, Univ Paris 06)-CNRS-IRD-MNHNParisFrance
  5. 5.Indo-French Cell for Water SciencesIISc-NIO-IITM–IRD Joint International Laboratory, IITMPuneIndia

Personalised recommendations