Annual Report 2000 - 2001

Scientific Services

National Atlas & Thematic Mapping Organisation

The NATMO is a subordinate office under the administrative control of the Union Ministry of Science & Technology (Department of Science & Technology), Government of India. The Organisation has a total sanctional strength of 487 till date which includes a large number of qualified professional geographers and cartographers. It is a premier organisation in the country in the field of preparation of thematic maps.

Functions:

• Compilation of National Atlas of India in English and Hindi.
  
• Preparation of National Atlas Maps in regional languages.
  
• Preparation of thematic maps based on research studies on environmental and associated aspects and their impact on social and economic development.
  
• Any other work entrusted by the Central Government to NATMO.
  
• Installation of Automated Mapping System for increasing efficiency in Mapping.
  
• Geographical/Cartographical research and training.

Target & Achievements for the Year 2000-2001

Objectives

• Development of operational and regional scale numerical weather prediction(NWP) models for forecasting weather in medium range(3-10 days) time scale taking full advantage of existing and concurrent developments both in India and abroad in the field of atmospheric science
  
• Establish the state-of-the-art supercomputing infrastructure to develop suitable NWP models to issue medium range weather forecasts
  
• Set-up 127 agrometeorological advisory service(AAS) units, each unit representing one of the 127 agroclimatic zones spread all over India, to prepare/issue/disseminate and to receive user feedback as well
  
• Set-up satellite based V-SAT Network for enabling stable/fast dedicated communication with AAS units

Agriculture is heavily weather dependent world over and is more so in tropical monsoon country like India. Further, India is one of the few countries in the world severely affected by cyclonic storms, droughts and floods. Farming community needs to be advised in time by producing custom-tailored location specific weather forecasts sufficiently in advance to take suitable actions to increase the production and minimise the losses as well. Deterministic numerical methods, which are highly computing intensive applications, have proved to be viable tools for the production of weather forecasts in the medium range time scale(3-10 days). The sophisticated global scale numerical weather prediction models require very high performance super computers.

Computing requirements for the implementation of global scale forecasts models drastically increase with the reduction of the spatial scale of the models. With the advancement of computers, particularly in terms of memory and speed over the recent decades, this area of weather prediction also developed accordingly.

The global observing system is composed of a complex array of surface(land and ship) and space based devices(satellites and aircrafts). Several uncertainties do exists in the retrieval of various products from the fields measured by the satellites. At the moment, only cloud tracked winds(CMVs) from geostationary satellites(like INSAT), temperature/moisture profiles from polar orbital satellites of NOAA, USA are the most widely employed satellite derived products in NWP. Global communication satellites transmit these data sets by collecting from land based communication networks spread over various countries and transmit the global observational data in real-time to the global weather centers. This service is part of global telecommunication system(GTS) of World Meteorological Organisation(WMO. The quantum of data received through GTS at various centers is limited to the bandwidth of the communication line and is one of the major constraints involving the observational data reception in the Indian context. As compared to other centers, viz. in USA, Europe and Japan, only 50-60% of the data quantum is received in India. In the operational setting, the weather forecasts need to be produced in certain limited time frame. For improving the accuracy of the forecasts, multiple runs with different initial conditions and different models are also required.

The NWP system implemented on real-time at the NCMRWF consists basically a complete suite of job sequences made to run on main computing systems to produce tailor made weather forecasts as per the user requirements. Various components of the system are as follows:

• Scheme for generating Data Base of decoded meteorological observations
  
• Quality control and pre-processing of decoded meteorological observations
  
• Data Assimilation(run intermittently at 6-hrly interval daily)
  
• Forecast Model[global spectral model with triangular truncation of 80 Waves in the horizontal and 18 levels in the vertical(T80/L18)]
  
• Scheme for generating Database of Post-processed NWP products
  (graphical and digital output; verification and archiving)

One of the major activities of the Centre is the performance evaluation of the Global Data Assimilation-Forecast System(GDAFS) during successive summer monsoon seasons. This task is pursued thoroughly as the successful prediction of various facets of the summer monsoon is the most challenging task for any forecasting system which is highly sensitive to various factors viz. availability of good quality initial observational data; intraseasonal variability; variability of various regional/global scale surface boundary fields etc.

Research and Development

Extensive R&D work focussed towards the real-time operations is conducted for improvement of the representation of physical and dynamical processes in the forecast models and the quality/quantity of input observational data to the operational forecast model with an ultimate goal of improving the accuracy and skill of the medium range forecasts on continuous basis. Salient features of the R&D efforts are summarised in the following:

The scheme for mixed layer treatment of planetary boundary layer(PBL) with in the operational global spectral model has been changed from first order closure to non-local closure scheme. This update of the model enables us to generate the PBL heights as well. Exhaustive experimentation carried out with the updated version of the model demonstrated a clear improvement in the performance of the model(in seasonal as well as case studies) which facilitated us to replace the earlier operational forecast model by the updated model since November, 2000.

The operational global model was integrated with the incorporation of two different convection parameterization schemes namely, the Simplified Arakawa-Schubert scheme (SAS) and, the Relaxed Arakawa-Schubert Scheme (RAS). Results of the study suggest that the new convection schemes show a strong monsoon flow and better prediction of rainfall over the Indian subcontinent as compared to the existing Kuo convection scheme. Comparison between SAS and RAS indicated that the former scheme produces better forecast skill scores in terms of systematic errors and root mean square statistics. It is planned to implement the updated version of the global model with SAS soon.

Reanalysis of the operational analysis archives for the month of Feb., 1999 was carried out with the incorporation of the additional data collected during the INDOEX field phase observational program(largely from Sagarkanya and Ron Brown research vessels; data from land stations of Maldives and Diego Garcia; dropsonde data collected from aircrafts; METEOSAT-5 high resolution CMVs etc.) Detailed impact assessment of the augmentation of the INDOEX observations on the global data assimilation-forecast system(GDAFS) was carried out.

The shortwave radiation parameterisation scheme of the model has been changed from GFDL scheme to NASA/Goddard scheme. Though, the Clear sky part of both the schemes are alike, the treatment of cloudy sky part is different. While GFDL scheme assumes absorption and simple reflection by the clouds(leaving aside the scattering by clouds), the NASA/Goddard scheme takes care of both scattering and absorption process in the cloudy atmosphere(through a two-stream method involving delta-Eddington approximation). Further, direction dependent surface albedo is being computed (using diffuse surface albedo). Based on certain experiment and after considering different factors, it is decided to set Short-wave radiation call interval at 3Hrs and Long-wave radiation call interval at 12Hrs. Results of the experimentation with the NASA scheme suggests that improvements are seen in temperature and geopotential fields globally and also over the Indian region.

Parameterising the effects of multiple scattering appropriately in the computation of absorption characteristics is crucial particularly in the presence of clouds, aerosols and green house gases etc. In particular, significant scattering and absorption can occur in the same part of the atmosphere due to the presence of these constituents. A theory is developed based on certain physical principles in which the effects of multiple scattering are considered explicitly. Using this method, the optical depth and fluxes are computed and their impact on tropical forecasts is examined in detail. Further, the results are validated with the available observational findings. The theoretical model finding of this study might provide insight relevant to the investigations carried out under the umbrella of various internationally organized observational programmes(INDOEX, GAME and ISTEP).

The SAS convection scheme was also run in combination with a new boundary layer parameterization based on non-local closure (NLC) and NASA/Goddard radiation parameterization in the operational model. An ensemble of five days forecasts were made for the month of August, 1999 with the updated experimental version of the operational model. Results indicate that the new schemes when combined together produced better distribution of rainfall over the Western Ghat. The combined experiment also produced improvements in the temperature forecasts in particular after 72 hours.

Work is in progress for the utilization of cloud liquid water path (CLWP) derived from the MSMR (IRS-P4) satellite in the NCMRWF model. The CLWP derived from the model was used as a first guess for objectively analyzing the observations from the MSMR for the period 1-20 August, 1999. The analyzed CLWP from MSMR were compared with the climatological distribution of CLWP derived from SSMI as well as those obtained from the ECMWF model. Results showed that the CLWP derived from MSMR are often overestimated as compared to the SSMI and NWP model outputs. The MSMR values also showed many spurious values of the CLWP, which had to be filtered out during the analysis.

A mesoscale model (MM5) model was run for the case of a cyclone in the Bay of Bengal for 48 hours using the NCEP analyzed data of 8th November, 1995 at 30 km and 60 km resolutions. Work is in progress to run the MM5 using the local analysed fields.

A software was developed based on GrADS to draw the meteorological subdivisions of India and plot the weekly cumulative rainfall forecasts. The performance of the NCMRWF model in producing the weekly cumulative rainfall forecasts was analyzed for 5 years

Workstation version of meso-scale ETA model was also implemented during the year. Several case studies are being conducted with the 48Km/38-Levels version of the model.

The treatment of Land Surface Processes(LSP) in the regional spectral model(RSM) is made identical with that of the global operational model. Results of Monsoon-2000 with the updated version of RSM demonstrate the removal of lower level easterly bias in its forecasts found in its pre-updated version.

Necessary tools for the utilization of real-time analysed global sea surface temperatures(SSTs ) in the Global Data Assimilation-Forecast System(GDAFS) are implemented and its operational utilization will be started soon.

Implementation of the HYSPLIT-4 model for long range transport of aerosols, gases and also for computing 3-D air trajectories with the development of all necessary modules to utilise the operational analyses/forecasts.

Implemented all interfaces and tools of the NETCDF software for facilitating the handling of NETCDF format data files on the existing computing systems .

Implementation of the VIS5D graphics system on the existing computing systems of the Centre is successfully completed for augmenting the visualization tools.

Impact of anomalous SSTs over the Bay of Bengal on the Indian monsoon simulations is carried out involving the operational global spectral model. Idealized experiments are carried out by specifying anomalous values of SST(slowly varying with a 40-day periodicity and with an anomaly of G1.0°C from the observed values during 1997) over the Bay of Bengal. The impact of the above prescription was found to be global in nature and the intraseasonal variations of rainfall suggest that phase monsoon activity gets altered.

A dynamic quality control technique is developed to filter out the MSMR wind speed data having large residuals as compared to the background field with the data analysis system. The quality control primarily involves a 'gross check' and 'an inter-consistency check' with the background(first guess ) field. New assimilation modules are developed to assimilate the MSMR wind speed data after subjecting to the said quality control in the GDAFS. It is to be mentioned that the new analysis module is specially developed for assimilating wind speeds and augmented with the existing segment of GDAFS. Exhaustive assimilation-forecast experiments have been carried out involving the MSMR data at 150Km resolution to examine the impact assessment of MSMR data inclusion on GDAFS. These efforts lead to the generation of near surface wind analysis over the Indian seas for ocean sea state forecasting applications groups working at NIO, Goa, SAC, Ahmedabad, IIT, Delhi etc. Efforts are on to make real-time generation of surface wind analysis at our Centre and make it available to various groups mentioned above in very near future.

Efforts are also currently on to assess the quality and consistency of the MSMR fields of wind speed and total precipitable water content(TPWC) with the corresponding fields of SSM/I data for the period 1-20 July, 1999. As a part of this effort, detailed evaluation of the consistency, representativeness (bias) and collocation statistics of wind speed(TPWC) with regard to the analyses archives of GDAFS is carried out. Results of the study suggest that the SSM/I TPWC is found to be consistently moist than that of the analyses archives while that of the MSMR fields does not demonstrate such a clear cut bias. Though we assimilate TPWC values over the Indian seas generated from NOAA satellites(available at 500Km resolution) in our GDAFS, operational archives(analyses/forecasts) are found to be drier than the corresponding fields of JMA, Japan and ECMWF, UK. The real-time availability of SSM/I data and its positive bias with respect to our analyses provided an opportunity to examine its utilization and impact on our GDAFS. Results of the study with the incorporation of SSM/I TPWC demonstrated a great potential to improve the moisture content not only in the initial state of atmospheric circulation but in forecasts as well and the injected TPWC from SSM/I satellites is found to produce suitable adjustments in the model state leading to the positive impact in the medium range forecasts in general. The real-time utilization of the SSM/I surface wind and TPWC fields in the stand-by GDAFS started from November, 2000.

A regional analysis scheme based upon optimum interpolation technique has been developed. Plans are on to link this scheme with the regional spectral model/meso-scale models being run in the Centre. Numerical experimentation is going on to assess the impact of the regional assimilation system on the performance of the regional spectral model.

The wind observations of the National MST Radar Facility at Gadanki were assimilated in the GDAFS for the period 23-26 September, 1997. Though, the inclusion of the said data improves the quality of the forecasts, the vertical resolution of the forecast model is found to not enough to capture the small scale events observed in the MST radar data.

The real-time availability of the pseudo surface pressure observations(PAOBS) from Australia provided opportunity to incorporate them in GDAFS. After doing necessary software development for its inclusion viz. Decoding, pre-processing etc., pre-operational experimentation of the inclusion of said data in GDAFS is currently underway. It is planned to use these data on operational basis.

Application

Major activities include preparation, dissemination and verification of location specific agro-meteorological forecasts, development of appropriate statistical interpretation models specifically for different agro-climatic zones, establishment of agrometeorological advisory service units to prepare and issue agro-advisories for contingency crop practices for the farming community in India, to carryout studies on crop weather relationships, development of decision support system and to evaluate the impact of weather conditions on pests and diseases. The advisories issued for the AAS units contain advises for crop production strategies like ploughing, sowing, pest and disease management, harvesting, threshing and post harvest procedures to derive maximum benefit of the benevolent weather and to mitigate the impact of malevolent weather for enhanced agricultural production. Geographical Information System (GIS) was procured at NCMRWF to make use in generation of homogeneous surfaces of different agromet parameters using data at agroclimatic zone scale in the country.

Local weather Forecasting and Agromet Advisory Service

• Number of Agromet Advisory service (AAS) units established is 81 out of which 27 units are at State Agricultural Universities (SAUs), one at IARI, Delhi and 52 colocated with the ICAR Regional Research Centres.
  
• Agroclimatic zone specific 3 days medium range weather forecasts were prepared and disseminated to 81 units once a week and 46 units twice a week for formulation of agromet advisory bulletins. Number of units preparing these bulletins and disseminating to the progressive farmers have increased from 72 to 74 and remaining are expected to start the service in operational mode shortly. These agromet advisory bulletins from 69 units are disseminated to the farming community through print media, 45 units on AIR and 14 units on Door Darshan.
  
• Satellite based VSAT communication systems have been operational at 72 AAS units.
  
• Besides the serving the needs of AAS fully, NCMRWF is also catering the needs of several other governmental(IMD, IAF, Indian Navy and SASE, Manali etc.) and non-governmental organisations. Besides this, we have issued forecasts during the year for the following applications as well :

  Natural Disaster Management
  Defence Excercises
  Flood Forecasting
  Summer monsoon onset and its progression
  Weather forecasts for special occasions during national functions (Independence
  Day/Republic Day etc.)/festivals
  Enroute weather outlook for annual religious pilgrimages (J &K Tourism etc.)

• Special weekly weather outlook continued to be provided to the Crop Weather Watch Group of Ministry of Agriculture on every Monday along with active participation of Centre's representative.
  
• Perfect Prognostic Method (PPM) models for 5 more units during monsoon were developed for preparing location specific agrometeorological forecasts, thus increasing the total number of such units to 45 during monsoon. PPM models for all the stations were implemented.
  
• Dataset (years 1994-99) for development of PPM models for monsoon season based on T80 analysis has been prepared.
  
• Preparation of dataset using real time synop data received through GTS for bias removal from the location specific Direct Model Output has been operationalised.
  
• IXth Annual Review Meeting of AAS of NCMRWF was held at Orissa University of Agriculture and Technology, Bhubneshwar during 19-21 September 2000 to review the progress made by 81 AAS units in the country. Recommendations adopted include strengthening of observatory under AAS network, timely weather data return to NCMRWF, dissemination of bulletins through mass media, improve forecast verification methodology, increase grant-in-aid provided to the units, opening of few more AAS units etc.
  
• 20 more units are proposed to be sanctioned under AAS scheme during the current financial year

Agro-Technology Transfer

Crop models along with DSSAT software were installed at few more units. Familiarization on CERES Wheat, CERES Rice, CROPGRO Soybean and SPAW models and their database requirement were given to the agricultural scientists from AAS units. Employing medium range weather forecasts, these calibrated models can be used to schedule irrigation and fertilizer application, sowing and spraying etc. In this regards, CERES Rice model was tested against the field experiments in the climate of Andhra Pradesh.

Other Activities

Exhibition on activities of NCMRWF with the details of the operational forecast products and the AAS service is arranged during

• Swadeshi Vigyan Mela held at IIT Delhi during 2-6 February, 2000
  
• International Conference on Managing Natural Resources for Sustainable Agricultural Production in the 21st Century held at IARI, New Delhi during 14-18 February, 2000

Computer and Network

Scientific computing is vital for research in Atmospheric Sciences particularly so in the area of Weather Forecasting. NCMRWF has already added latest technology to its existing supercomputing facility. At present there are three low cost stand-by solutions. The new systems are : DIGITAL DEC ALPHA; CDAC PARAM 10000 and ORIGIN 200

NCMRWF has hosted a home page viz: www.ncmrwf.gov.in . The home page was designed and hosted with the help of in-house expertise. All the NWP products of the Centre are updated daily and are available in the home page.

Till the present financial year 70 Very Small Aperture Terminals (VSATs) have been installed/commissioned. During this year 10 VSATs were installed and 5 more are at different stages on installation viz: manufacturing, testing, inspection etc.