Downscaled climate change projections with uncertainty assessment over India using a high resolution multi-model approach

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

  • Pankaj Kumar
  • Andrew Wiltshire
  • Camilla Mathison
  • Shakeel Asharaf
  • Bodo Ahrens
  • Philippe Lucas-Picher
  • Christensen, Jens Hesselbjerg
  • Andreas Gobiet
  • Fahad Saeed
  • Stefan Hagemann
  • Daniela Jacob

This study presents the possible regional climate change over South Asia with a focus over India as simulated by three very high resolution regional climate models (RCMs). One of the most striking results is a robust increase in monsoon precipitation by the end of the 21st century but regional differences in strength. First the ability of RCMs to simulate the monsoon climate is analyzed. For this purpose all three RCMs are forced with ECMWF reanalysis data for the period 1989-2008 at a horizontal resolution of ~. 25. km. The results are compared against independent observations. In order to simulate future climate the models are driven by lateral boundary conditions from two global climate models (GCMs: ECHAM5-MPIOM and HadCM3) using the SRES A1B scenario, except for one RCM, which only used data from one GCM. The results are presented for the full transient simulation period 1970-2099 and also for several time slices. The analysis concentrates on precipitation and temperature over land. All models show a clear signal of gradually wide-spread warming throughout the 21st century. The ensemble-mean warming over India is 1.5. °C at the end of 2050, whereas it is 3.9. °C at the end of century with respect to 1970-1999. The pattern of projected precipitation changes shows considerable spatial variability, with an increase in precipitation over the peninsular of India and coastal areas and, either no change or decrease further inland. From the analysis of a larger ensemble of global climate models using the A1B scenario a wide spread warming (~. 3.2. °C) and an overall increase (~. 8.5%) in mean monsoon precipitation by the end of the 21st century is very likely. The influence of the driving GCM on the projected precipitation change simulated with each RCM is as strong as the variability among the RCMs driven with one.

OriginalsprogEngelsk
TidsskriftScience of the Total Environment
Vol/bind468-469
Sider (fra-til)S18-S30
ISSN0048-9697
DOI
StatusUdgivet - 1 dec. 2013
Eksternt udgivetJa

ID: 186939888