Impact of initial and boundary conditions on regional winter climate over the Western Himalayas: A fixed domain size experiment P. Maharana, A.P. Dimri ⁎ School of Environmental Sciences, Jawaharlal Nehru University, New Delhi, India abstract article info Article history: Received 24 April 2013 Received in revised form 23 November 2013 Accepted 22 December 2013 Available online 31 December 2013 Keywords: Western Himalayas regional climate model Western Disturbances ensemble The Western Himalayas during winter receives precipitation due to the eastward moving low pressure synoptic weather systems, called Western Disturbances (WDs) in Indian meteorological parlance. The complex Himalayan topography, sparse observational data, less understanding of physical processes, etc. form many interesting re- search questions over this region. One of the important research goals is to study the change in the winter (Dec., Jan. and Feb. — DJF) climate over the Himalayas. In the presented study with modelling efforts having vary- ing initial and boundary conditions (ICBC) with same model physics option is attempted to provide a comment on important physical processes pertaining to precipitation and temperature fields. A 22 year (1980–2001) simulation with Regional Climate Model version 3 (RegCM3) forced with National Centre for Environmental Prediction/National Centre for Atmospheric Research (NCEP/NCAR) reanalysis 1 (NNRP1), NCEP/NCAR reanalysis 2 (NNRP2) and European Centre for Medium Range Weather Forecast 40 Year reanalysis (ERA40) as three differ- ent ICBC is carried out. The present study focuses on the winter climatology of the main meteorological param- eters viz., temperature, precipitation and snow depth and interannual variability of winter seasonal precipitation. The model shows overestimation of seasonal average precipitation and underestimation of seasonal average temperature fields over the Western Himalayas in all the three model simulations. The interannual variability of precipitation and temperature over this region is nicely captured by the model. The model simulation with NNRP2 as the ICBC shows more realistic results. In addition the ensemble mean of the three simulations has shown improved results and is closer to the abovementioned simulation. Precipitation bias explained in terms of the higher vertical integrated moisture flux and transport shows strong convergence zone over and along the southern rim of the Indian Himalayas. The energy balance over the Western Himalayas explains the cause of lower temperature in the model simulation and the cause of lesser convective precipitation and evaporation. © 2013 Elsevier B.V. All rights reserved. 1. Introduction The Himalayas are characterized by complex cascading ranges of mountains having west to east stretches starting from near Afghanistan in west to the northeast India in the east. The altitudinal variation and orientation of these mountain ranges make their role complex and define the climate of the region. Also, the precipitation characteristic is not uniform along the stretch; the Western Himalayas (WH) gets the precipitation during winter due to the WDs and the east- ern Himalayas (EH) gets the precipitation mainly during the Indian summer monsoon (ISM). The Himalayan region is termed as the third pole (Schild, 2008) and water tower of Asia (Xu et al., 2009). This is be- cause the Himalayas are the source of fresh water to many rivers of many countries surrounding it and hence influence lives of the millions of people in Indian subcontinent. So the study of the regional climate of the Himalayan region becomes important. Due to heterogeneous topography and variable landuse, it is very difficult to establish many weather stations over these regions and to maintain it subse- quently. Hence these regions are data sparse. The climate modelling/ downscaling could be one of the options to understand the meteorolog- ical processes which are taking place over this complex topographic ter- rain. Fowler et al. (2007) showed that the dynamical downscaling provides far better results than the statistical methods. Generally the global climate models (GCMs) are integrated at a very coarser resolu- tion in the order of 250 to 300 km horizontal resolution. This results in the loss of the climatic information over a small topographically com- plex region and subgrid processes. GCMs take longer time period for the simulation and huge computational resources. So regional climate models (RCMs) are widely used as they provide high resolution regional climate information from the GCMs or reanalysis (Giorgi, 2006) for the impact assessment of the climate and the detailed study of the atmo- spheric processes (Bhaskaran et al., 2012). The higher resolution infor- mation from the RCMs improves the understanding of the atmospheric processes associated with mountain orography, land cover and land–sea contrast (Rupa Kumar et al., 2006; Lucas-Picher et al., 2011). The problem of dynamic downscaling is that systematic Global and Planetary Change 114 (2014) 1–13 ⁎ Corresponding author at: School of Environmental Sciences, Jawaharlal Nehru University, New Delhi 10067, India. E-mail address: apdimri@hotmail.com (A.P. Dimri). 0921-8181/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gloplacha.2013.12.011 Contents lists available at ScienceDirect Global and Planetary Change journal homepage: www.elsevier.com/locate/gloplacha