« Back

Vol. 9 No.3 June 2001   

Development of Mesoscale Weather Forecast Model

An operational fine resolution mesoscale model has been developed especially for the mountain weather forecast using MMS mesoscale model. MMS is the latest fifth-generation National Centre for Atmospheric Research (NCAR)/Penn State University (PSU) Model. This model has several advantages, which include: (i) multiple-nest capability (up to nine domains at a time ), (ii) non-hydrostatic dynamics, (iii) four-dimensional data assimilation capability, (iv) more physics options, and (v) portability to wide range of computing platform. The model source code is designed in such a way that it can use NCEP (National Centre for Environmental Prediction) and ECMWF (European Centre for Medium Range Weather Forecasting) global analysis data. Source code of the MMS has been downloaded from the public web site of NCAR/PSU .

The MM5  model system consists of six main modules, which include the graphics (GRAPH) module for post-processing the model output. The other modules are: TERRAIN, REGRID, RAWINDS, INTERP and MMS. Terrestrial and isobaric meteorological data are horizontally interpolated from latitude-longitude mesh to a variable high-resolution domain by using programs, TERRAIN and REGRID. This interpolation does not provide mesoscale details. For improvement of mesoscale forecast for the limited area, the interpolated data can be enhanced with observations from the standard network of surface and rawindsonde stations, which are locally available. This will be done by program RAWINS. The program INTERP performs vertical interpolation of analyzed data from pressure level to the sigma (terrain following) coordinate system before the data is input into the model.

Initially, TERRAIN module of the MM5 modeling system has been decoded. This module interpolates horizontally the latitude-Iongitude interval terrain elevation and land use categories into the chosen mesoscale domain. This module generates fine mesh terrain height and land use files, which will be used by other modules in the modeling system. A fine grid terrain height of N-W Himalayan region has been successfully generated from 30 arc second USGS (United State Geological Survey) digital topography data, which is available in public domain. The main part of the modeling system is the preprocessor of the model, which comprises REGRID, RAWINDS and INTERP modules. The output from files of this module gives the initialization of the model. The development of preprocessor for the MM5 mesoscale model has been taken up by DRDO in association with the Centre for Development of Advanced Computing (C-DAC), Pune.

Numerical Simulation of Circulation Pattern over Kashmir Valley

A numerical model was recently employed to study the thermally-generated mesoscale air  circulation pattern over Kashmir valley. The salient  features of the model are:

  • Dry, hydrostatic model with terrain  following coordinate system in vertical (z)  direction
  • Computation of planetary boundary layer  (PBL) height through prognostic equation
  • Parameterization of surface layer heat, momentum and moisture fluxes using  Monin-Obukhov similarity theory (PBL  parameterization)
  • Calculation of surface temperature through surface energy balance equation

The model was initialized with synoptic surface data and vertical profiles of pressure, Numerical Modeltemperature, humidity, wind speed and wind direction obtained through regular radio/rawin sonde observations over Srinagar and Manali. The model was run with  25 x 25 x 17 grids (i.e., 25 grid points in X (east-west) direction, 25 grid points in Y (north-south) direction and 17 grid points in Z (vertical) direction). The region of study was taken from 74 E to 76 E and 33 N to 35 N and from 76 E to 78 E and 31.5 N to 33.5 N, keeping the Srinagar and Manali locations respectively as the centre of the domain. The topography was derived from 30 arc second USGS digitized topography data.

Generally, in mountainous regions, rising motion along the slopes and sinking motion in the centre of valley are observed during day time due to differential heating. A similar pattern is very well simulated. This clearly shows the up-slope flow along the slopes of the valley during day time. Similarly, during night time, due to nocturnal cooling, sinking motion along the slopes and raising motion in the centre of the valley are observed, which is just reversal of day time wind pattern. The horizontal wind circulation due to up-slope flows (during day time) and due to down-slope flows (during night time) were well simulated. The results show that there is no change in the upper level winds. Overall, the model simulates the local wind circulation pattern over Kashmir valley.