weather information for sustainable agriculture in … information for sustainable agriculture in...

Weather Information for Sustainable Agriculture in India*

L.S. RATHORE1

Director General of Meteorology, India Meteorological Department, New Delhi - 110003

of diminishing returns is catching up even for theareas benefited by Green Revolution.

Therefore, the area and input intensiveagriculture is leading us into an era ofunsustainable agricultural growth as the crucialcomponents of sustainable agricultural productionand distribution system such as integration ofnatural biological cycles and controls; protectionand renewal of soil fertility and the naturalresource base; optimisation of the use of non-renewable resources and of on-farm & off- farminputs; ensuring dependable farm income;promoting opportunity in family farming and farmcommunities; reducing adverse impacts on healthand safety of the farming community; and otheraspects of integrated soil, crop, livestock, waterand pest management are being ignored. All thesecomponents of sustainable agriculture, in additionto other management and input interventions, alsorequire precise weather and climatic information.Weather and climatic information plays a majorrole in the entire crop cycle right from selectingthe most suitable crop/variety upto post harvestoperations and marketing; and if provided inadvance can be helpful in inspiring the farmer toorganize and activate their own resources in orderto reap the benefits by judicious application ofcostly inputs. Hence, the meteorologicalinformation may help the farmer make the mostefficient use of natural resources, with the aim ofimproving agricultural production; both inquantity and quality. It becomes more and moreimportant to supply meteorological informationblended with weather sensitive management

*Fourth Dr. B.P. Ghildyal-ISAP Memorial Lecturedelivered on December 13, 20131Email: [email protected]

Vol. 13, No. 2, pp. 89-105 (2013)Journal of Agricultural Physics

ISSN 0973-032Xhttp://www.agrophysics.in

Agriculture sector, world over, hasexperienced a phenomenal growth since middleof the last century. The growth of agriculture inIndia, driven by the Green Revolution, has madea tremendous contribution towards aggregatesupply of food grains, ensuring food security tothe growing population with the total food grainproduction increasing from about 50 m tonnes atthe time of independence to more than 250 mtonnes now. This growth, however, is fastapproaching the plateau; and sustainability of thisgrowth faces serious challenges. Some of theproblems such as the loss of soil productivityfrom excessive soil erosion and associated plantnutrient losses, surface and ground waterdepletion both in terms of quality and quantityand pollution from pesticides, fertilizers andsediments, impending shortages of non-renewableresources and low farm income from depressedcommodity prices and high production costs areassuming serious proportions. Besides, most ofthe increase in production and yield has beenrestricted to food grains that too in wheat andrice, primarily because crop varieties covered inthe green revolutions have high waterrequirements and are not suitable for use underrainfed conditions. They also require high inputresources for giving maximal yields. Thus thegreen revolution has remained confined toirrigated areas at the cost of other crops,particularly coarse cereals and pulses. But the law

Memorial Lecture

90 Journal of Agricultural Physics [Vol. 13

operations before the start of cropping season inorder to adapt the agricultural system to increasedweather variability. Subsequent to this, weatherforecast based Agro-meteorological advisoriesbecome vital to stabilize their yields throughmanagement of agro-climatic resources as well asother inputs such as irrigation, fertilizer andpesticides. Meteorological information also playsan important role in the following operations inagriculture:

• Higher and regulated crop yields• Lower production costs• Environmental protection• Better product quality• Land management: diversification, soil

conservation• Water management: issues related to farm

irrigation, urban water needs, waterconservation measures

• Combat unfavourable consequences ofweather including pest and diseasemanagement

• Breeds and varieties adapted to localagroclimatic conditions

• Better consideration of meteorological factorsthat affect animal behaviour and health

• Multiple cropping system for irrigated areasand tree based farming system for rainfed area

• Integrated pest management• Soil and water conservation• Watershed management• Agroforestry systems in dry lands/ sloppy

areas and erosion prone areas• Water technology

Hence, the meteorological services can helpthe farmers & the planners exploit the potentialof good weather and minimize the impact of badweather as there exist tremendous opportunitiesfor applications of meteorology in both day today and long term planning in agriculture andsuch applications could contribute substantiallyto promote sustainable agriculture and povertyalleviation; and the management ofagrometeorological data is the key to these

applications. Weather information and advisoriesare assuming greater significance under theincreased variability in the changing climate.

Usefulness of Weather Forecasts for Agri-culture

• Climate-based strategic agronomic planningFor optimal productivity at a given location,crops and cropping practices must be suchthat their cardinal phased weatherrequirements match the temporal march of therelevant weather elements, and at the sametime endemic periods of pests, diseases andhazardous weather are avoided. In suchstrategic planning of crops and croppingpatterns, short –period climatic data, bothroutine and processed (such as initial andconditional probabilities), have a vital role toplay.

• Occurrences of erratic and adverse weatherAgronomic strategies have to be devised tocope with the effects of erratic and adverseweather on agricultural production. Forexample, delay in the start of crop season canbe countered by using short duration varietiesof crops and thicker sowings and the effectsof frosts can be prevented by resorting toirrigation or lighting trash fires.Medium range weather forecasts with avalidity period that enable farmers to organizeand carry out appropriate cultural operationsto cope with, are clearly useful. The followingkind of forecast is mainly useful inoperational agrometeorology.Short and Medium Range Forecast foragriculture is useful for the followingactivities:

• Preparatory activities, including landpreparation and preparation of plant material

• Planting or seeding/sowing; Management ofcrops, fruit trees and vines

• Application of fertilizer, irrigation; thinning,topping, weeding; pest and disease control;Management of grazing systems

• Harvesting, on-farm post-harvest processing

2013] Weather Information for Sustainable Agriculture in India 91

• Transport of produce; Livestock productionactivities (for dairy enterprises, beef systems,lamb and other livestock systems)

Weather and Climate Observation System

India Meteorological Department (IMD) hasprogressively expanded its infrastructure formeteorological observations on real time basis andto provide current and past observations foroperational and research use. Surfaceobservatories are located one in each district soas to meet the requirements of agricultural,transport and other sectors. IMD’s present Upperair observational network comprises of radiosondeand pilot balloon, RS/RW observatories, DopplerWeather Radar etc. spread all over the country.Also a Doppler Weather Radar (DWR) providesvelocity and spectrum width data (in addition toreflectivity). This capability permits the forecasterto view weather events like thunderstorms,hailstorm, heavy rainfall etc. in greater detail. TheDoppler radar shows promise in depicting windfield patterns associated with cyclones, which canbe used in assessment of wind damage potentialand determination of the associated storm surgeaffecting the meteorological, hydrological andaviation products which are very useful inestimating the storm’s centre, fixing its positionand predicting its future path. In NumericalWeather Prediction models the Doppler data indigital form can directly be assimilated, whichresults in better rainfall estimation, betterprediction of the storm track and point of landfallfor tropical cyclones.

IMD is observing weather with satellite andradar. At present IMD is receiving and processingmeteorological data from two Indian satellitesKalpana-1 and INSAT-3A. Kalpana-1 andINSAT-3A both have three channel Very HighResolution Radiometer (VHRR) for imaging theEarth in Visible (0.55-0.75 µm), Infra-Red (10.5-12.5 µm) and Water vapour (5.7-7.1µm) channelshaving resolution of 2×2 kms in visible and 8×8kms in WV and IR channels. In addition theINSAT-3A has a three channel Charge CoupledDevice payload for imaging the earth in Visible(0.62-0.69µm), Near Infra-Red (0.77-0.86µm) andShort Wave Infra Red (1.55-1.77µm) bands of

Spectrum. All the received data from the satelliteis processed and are archived in National SatelliteData Centre (NSDC), New Delhi.

The payload of INSAT-3D adds a newdimension to weather monitoring through itsatmospheric sounding system, which providesvertical profiles of temperature, humidity andintegrated ozone from surface to top of theatmosphere, and is expected for the next sevenyears making a difference for the weatherforecasting and disaster warning systems for thecountry. This has two kinds of payloads ‘Imager’and ‘Sounder’. The six channel imager consistsof broad visible (VIS), short-wave infrared(SWIR), Middle infrared (MIR), water vapour(WV) and two split-thermal infrared (TIR) bands.The optical bands (VIS & SWIR) have spatialresolution of 1 km. The MIR and TIR bands have4 km spatial resolution. The WV band has 8 kmspatial resolution. The 19 channel sounder willbe able to probe the vertical structure of theatmosphere at regular time interval in a day. TheImager data will be able to provide satellitemeteorological (satmet) products such as rainfall,land surface temperature (LST), land surfacealbedo, incident solar radiation, cloudiness, uppertroposphere humidity (UTH), outgoing longwaveradiation (OLR). The combination of all theseproducts from 3D, improved weather forecasts,vegetation index product from INSAT 3A CCDand in situ data are valuable ingredients togenerate real-time value-added information forenhanced operational agro-met advisory servicesin the country. The long-term datasets fromINSAT suite will be able to aid in digital agro-climatic characterization.

IMD is maintaining a network of agro-meteorological observatories in farmenvironment. Beside this network, there areevapotranspiration (ET) stations, pan evaporationstations, soil moisture (SM) observatories, dewfall and radiation observations. Agro-meteorological Automatic Weather Stations(Agro-AWS) have been installed at 127Agrometeorological Field Units (AMFUs). Thedata are being received in AWS Laboratory, Puneregularly. IMD has drawn plans to upgrade andstrengthen its observation network in the coming


years. Nationwide network of Automatic WeatherStations and Doppler Weather Radar network willalso increase.

IMD has long time series of various climatedata in its archive. National Climate Center(NCC) at Pune generates many climate dataproducts for smaller spatial and temporal scalesfor the user community. These data productsinclude (i) Daily gridded (1o × 1o) rainfall andtemperature data, (ii) Daily gridded (0.5o × 0.5o)rainfall data, (iii) District wise normal for varioussurface parameters, (iv) Marine climatesummaries for Indian Ocean region. IMDoperationally prepares gridded rainfall data on0.5o × 0.5o spatial resolution by processingspatially continuous satellite derived rainfall(TRMM) and weather observation from existingweather network (IMD Gauge). These data helpin monitoring strength of monsoon, studyingoscillation of different periodicity, extended rangeprediction and to fine tune the dynamical models.The Centre will very soon bring out daily rainfalldata on 0.250× 0.250 spatial scale. Similarly dailygridded temperature data would also be broughtout on smaller spatial scale.

Weather Forecasting System

India experiences large spectra of weatherevents having spatial scale of less than 1 km tomore than 1000 km and temporal range of lessthan an hour to more than a week. Different partsof the country experience different kinds ofweather conditions such as Winter season (Jan-Feb) is characterised by Western Disturbances,Cold Wave, Fog; Pre-Monsoon (Mar-May) byCyclonic Disturbances, Heat Wave, ThunderStorms, Squalls, Hail Storm, Tornado; Monsoon(Jun-Sep) by Southwest Monsoon Circulation,Monsoon Disturbances; and Post-Monsoon (Oct-Dec) by Northeast Monsoon, CyclonicDisturbances. Meteorological Department in thecountry was one of first scientific departmentswhich started scientific research on weathersystems as early as in 18th century. Indianmonsoon is a field of active research all over theworld as it is a unique weather event which coversseveral oceans and continents but depicts its

characteristic phase over Indian sub-continent.The monsoon rains are crucial for agriculturalproduction in the country as most parts of thecountry depend on this rain for day-to-dayagricultural operations.

Prediction of weather systems in differentspatial and temporal scale over the Indian regiontherefore assumes considerable importance.Several organisations are engaged in developingtechniques for prediction of these systems. In therecent past, IMD has made enormousimprovement in the accuracy and lead time offorecasts for various usage including AAS.

IMD has made Global Forecast System (GFST574/L64, based on NCEP) operational inDecember 2009, implemented on IBM based HighPower Computing Systems (HPCS). Thisincorporates Global Statistical Interpolation (GSI)scheme as the global data assimilation for theforecast up to 7 days. In horizontal, it resolves574 waves (> 22 km in the tropics) in spectraltriangular truncation. The model has 64 verticallevels (hybrid; sigma and pressure). Currently, itruns twice in a day (00 UTC and 12 UTC). Inaddition to this, the meso-scale forecast systemWRF (ARW) with 3DVAR data assimilation isbeing operated daily twice, at 27, 9 and 3 kmhorizontal resolutions for the forecast up to 3 daysusing initial and boundary conditions from theIMD GFS-574/L64 (horizontal resolution over thetropics ~ 22 km). At ten other regional centres ofIMD, very high resolution mesoscale models(WRF at 3 km resolution) are made operational.NWP based objective forecast products areprepared to support cyclone warning service.Doppler weather and mesoscale dynamical modelbased nowcast system was made operational forthe National Capital of Delhi. A full range ofNWP products at different spatial and temporalscale are routinely made available on the IMDweb site www.imd.gov.in.

Next generation model will target moreaccurate ways to initialize forecasts by improvingdata assimilation techniques, more accurateparameterizations, and improved estimates ofuncertainty in the modelling system and inforecasts. Coupling of the atmosphere to the


required for the Integrated Agro-advisory Serviceof India. Five NWP models considered for thisdevelopment work are: (i) IMD GFS T574, (ii)ECMWF T799, (iii) JMA T899, (iv) UKMO and(v) NCEP GFS. Model outputs of the constituentmodels are interpolated at the uniform gridresolution of 0.250×0.250lat/long. Further, theweight for each model at each grid is determinedobjectively by computing the correlation co-efficient between the predicted rainfall andobserved rainfall. High resolution gridded rain-gauge data produced operationally at NationalCentre of IMD Pune are used for developmentand validation of the forecasts. The ensembleforecasts (day 1 to day 5 forecasts) are generatedat the 0.250×0.250resolution. The ensembleforecast fields are then used to generate districtlevel forecasts by taking average value of all gridpoints falling in a particular district. The productscomprise of quantitative forecasts for 7 weatherparameters viz., rainfall, maximum temperature,minimum temperatures, wind speed, winddirection, relative humidity and cloudiness. Theseproducts are disseminated to RegionalMeteorological Centres and MeteorologicalCentres of IMD located in different states. Theseoffices undertake value addition to these productsusing local conditions and synoptic interpretationof model out put and communicate to 130AgroMet Field Units (AMFUs) on every Tuesdayand Thursday.

Long Range Forecast

IMD has been issuing long-range forecasts(LRF) based on statistical methods for the south-west monsoon rainfall over India (ISMR) for morethan 100 years. The forecast for the south-westmonsoon rainfall is issued in two stages. The firststage forecast for the seasonal (June toSeptember) rainfall over the country as a wholeis issued in April and the update of the Aprilforecast is issued in June. Along with the updateforecast, forecast for seasonal rainfall over fourbroad geographical regions of India and Julyrainfall over country as a whole are also issued.

For issuing the forecast for the seasonalrainfall over the country, a new statisticalforecasting system based on the ensemble

underlying surface – land and ocean – is importantand also critical for forecast lead times extendinginto fortnightly and monthly timescales. Toadequately resolve the transient systemsresponsible for rainfall during monsoon and theirmovement, it is important to use models with highresolutions both in the horizontal and vertical. By2020, plan is to run global models of about 10km resolution with about 100 levels in vertical(high-resolution GFS such as T1148-L64 (16 km)to T1500-L128 (13 km)), and regional modelswith about one km resolution. The physicalparameterizations need to be improved/optimizedfor higher resolution models. As the model gridsize approaches the horizontal scales of moistdeep convection, special attention will be givento the parameterization of moist processes. Cloudresolving models are required to be embedded inthe global and meso-scale models as clouds arethe major cause of uncertainty in present daymodels. All current and future global NWPsystem will have to address the uncertaintyinherent in the prediction over all temporal andspatial scales. The challenge is to increase thenumber of ensemble members which will furtherimprove its value for all applications.

Recent studies suggest that an interactiveupper-ocean in a high resolution coupled generalcirculation model significantly enhance theprediction skill of weather in medium range.Experiments at ECMWF and other centers havestarted showing the benefits of coupling inimproving the skill in medium and 2 weeks to 1month monsoon forecasts. Therefore, ocean modelwith very high vertical resolution (1 m) in theupper ocean has to be coupled with a higherfrequency (1 hour) of interaction with atmospheremodel. Ocean data assimilation will be requiredto initialize the ocean model properly. Similarlya land surface model with its assimilation has tobe also coupled to form the full ‘Coupled EarthSystem Model’.

District Level Medium Range Weather Fore-cast

IMD implemented a Multi-model Ensemble(MME) based district level quantitative forecastsin the operational mode since 1 June 2008, as


technique was introduced in 2007. The 8predictors considered for the new ensembleforecast system are given in Table 1. For theApril forecast, the first 5 predictors given in thisTable 1 are used. For the updated forecast in June,last 6 predictors are used.

Dynamical (Numerical) Forecasting Systemfor Month and Season

The experiments of this study were carriedout with the National Centers for EnvironmentalPrediction (NCEP) seasonal forecasting model(SFM), which provides regional details for theGlobal Spectral Model (GSM). The SFM(Kanamitsu et al. 2002) uses the spectraltransform method to solve the primitive equations.The model is a non-hydrostatic global spectralmodel with a triangular truncation of 62 sphericalharmonics (T62), equivalent to a horizontalresolution of about 280 km was chosen. Themodel of this study used a vertical sigmacoordinate system which contains 28 layers. Thebasic model physical processes andparameterization schemes were developed fromreanalysis-II version of NCEP Medium RangeForecast model.

The ensemble simulation was carried out for31 year period from 1982 to 2012 to preparemodel climatology. Each forecast was initializedat first 10 days of each month and runcontinuously for next five months. The initialconditions corresponding to 0000Z UTC wereobtained from NCEP reanalysis. The forecastedSST from Coupled model, NCEP ClimateForecast System version 2 (CFSv2) were used as

SST condition. SST Boundary condition preparedby averaging of CFSv2 SST forecast for first 10days initial conditions of each month (Total 40members). For making real time forecast, initialand boundary conditions are taken from NOAAoperational Modal archival and distributionsystem.

Special Weather Forecast for ExtremeEvents

Special weather forecast for agricultureprovides the necessary meteorological input toassist farmers in making decisions. Therequirements for these special forecasts will varyfrom season to season and crop to crop. Specialforecast issued are as follows:

• Tropical storms (cyclones, hurricanes,typhoons, etc.) associated with high winds,flooding and storm surges

• Floods (other than those related to tropicalstorms) heavy rains due to monsoons, waterlogging and landslides

• Severe thunderstorms, hail storms, tornadoesand squalls

• Drought and heat waves• Cold spells, low temperature, frost, snow and

ice-storms• Dust storms and sand storms• Pest and diseases of crop and livestock

Early Warning Systems

IMD has developed early warning systems forextreme weather events like drought, flood,

Table 1. Details of the 8 predictors used for the new ensemble forecast system

Details of 8 Predictor Used for forecasts inpredictors

1 NW Europe Land Surface Air Temperature April2 Equatorial Pacific Warm Water Volume April3 North Atlantic Sea Surface Temperature April and June4 Equatorial SE Indian Ocean Sea Surface Temperature April and June5 East Asia Mean Sea Level Pressure April and June6 Central Pacific (Nino 3.4) Sea Surface Temp.Tendency June7 North Atlantic Mean Sea Level Pressure June8 North Central Pacific wind at 1.5 km above sea level June


This analysis would indicate qualitativelyretardation in the plants growth and so pooryields. Indirectly, this may also be helpful forirrigation scheduling, the amount and the time atwhich the water is badly needed by the plant.

IMD is also preparing Standard PrecipitationIndex maps which describe dry and wet periodsin the same way. SPI is an appropriate tool formonitoring agricultural drought but incombination with other information: waterbalance, crop condition, etc. Computation of SPIis done at a monthly time scale.

Agromet Advisory Service in India

Agrometeorological Advisory Service (AAS)rendered by India Meteorological Department(IMD), Ministry of Earth Sciences (MoES) inactive collaboration with ICAR, State AgricultureUniversities, State Department of Agriculture etc.is a step to contribute to weather informationbased crop/livestock management strategies andoperations dedicated to enhancing crop productionand food security. The main emphasis of theexisting AAS system is to collect and organizeclimate/weather, soil and crop information, andto amalgamate them with weather forecast toassist farmers in taking management decisions.This has helped to develop and apply operationaltools to manage weather related uncertaintiesthrough agro-meteorological applications forefficient agriculture in rapidly changingenvironments.

tropical cyclone, heat and cold wave, fog, hail.For drought monitoring and assessment, IMD hasestablished system for real-time monitoring ofrainfall on daily, weekly, monthly and seasonalbasis for the districts/ meteorologicalsubdivisions/ state in the country. IMD ispreparing indices based aridity anomaly maps onweekly basis that gives information about themoisture stress experienced by growing plant.

Fig. 2. Standardized precipitation index

Fig. 1. Weekly aridity anomaly map for AAS


The information support systems under AASinclude:

• Provision of weather, climate, crop/soil andpest disease data to identify biotic and abioticstress for on-farm strategic and tacticaldecisions.

• Provide district specific (Pan India) weatherforecast (rainfall, cloudiness, maximum/minimum temperature, wind speed, winddirection, maximum/minimum relativehumidity) up to 5 days with outlook forrainfall for remaining two days of a week.

• Translate weather and climate informationinto farm advisories using existing researchknowledge on making more efficient use ofclimate and soil resources throughapplications of medium range weatherforecast. A broad spectrum of advisoriesinclude weather sensitive farm operationssuch as sowing/ transplanting of crops,fertilizer application based on wind condition& intensity of rain, pest and disease control,intercultural operations, quantum and timingof irrigation using meteorological thresholdand advisories for timely harvest of crops.

• Introduction of the state-of-art technologiessuch as remote sensing, GIS and cropsimulation model based decision supportsystem for agro-meteorologists to adaptagricultural production systems to weather &climate variability and to the increasingscarcity of input such as water, seed, fertilizer,pesticide etc.

• Develop effective mechanism to on timedissemination of agromet advisories tofarmers,

• Effective training, education and extension onall aspects of agricultural meteorology.

Being a multidisciplinary and multi-institutional project, AAS project is beingimplemented through five tier structure to set updifferent components of the service spectrum. Itinclude meteorological (weather observing &forecasting), agricultural (identifying weathersensitive stress & preparing suitable advisoryusing weather forecast), extension (two way

communication with user) and informationdissemination (Media, Information Technology,Telecom) agencies.

Agrometeorological Support for Farm Man-agement

Weather based farm advisories as supportsystem has been organized after characterizationof agro-climate, including length of crop growingperiod, moisture availability period, distributionof rainfall and evaporative demand of the regions,weather requirements of cultivars and weathersensitivity of farm input applications. All thisknowledge is used as background information.Following are the ingredients of a typicalAgromet Advisory Bulletin to reap benefits ofbenevolent weather and minimize or mitigate theimpacts of adverse weather;

i. District specific weather forecast, inquantitative terms, for next 5 days for rainfall,cloud, max/min temperature, wind speed/direction and relative humidity, includingforewarning of hazardous weather eventslikely to cause stress on standing crop andsuggestions to protect the crop from them.

ii. Weather forecast based information on soilmoisture status and guidance for applicationof irrigation, fertilizer and herbicides etc.

iii. The advisories on dates of sowing/plantingand suitability of carrying out interculturaloperations covering the entire crop spectrumfrom pre-sowing to post harvest to guidefarmer in his day–to-day cultural operations.

iv. Weather forecast based forewarning systemfor major pests and diseases of principal cropsand advises on plant protection measures.

v. Propagation of techniques for manipulationof crop’s microclimate e.g. shading,mulching, other surface modification, shelterbelt, frost protection etc. to protect cropsunder stressed conditions.

vi. Advisory on contingency plan under extremeweather situations.

vii. Reducing contribution of agriculturalproduction system to global warming andenvironment degradation through judicious


management of land, water and farm inputs,particularly pesticides, herbicides andfertilizers.

viii.Advisory for livestock on health, shelter andnutrition.

The support on above is rendered throughpreparing district specific agrometeorologicaladvisory bulletins by Agro-Meteorological FieldUnits (AMFUs) (Fig. 3). Inter-disciplinary groupof agricultural and extension specialists atAMFUs formulates weather based farm advisorybulletin. These bulletins contain location specificand crop specific farm level advisories tailored tomeet the farmers’ need. The bulletins are encodedin a format and language which is easy tocomprehend by the farmer in his decision makingprocesses. The agrometeorologists play a vitalrole in the encoding and decoding of the messagesfrom the meteorologists to the agricultural sector.Crop weather calendars initially developed byIMD is used in preparation of weather basedadvisory bulletin. Recently All India CoordinatedResearch Project on Agrometeorology(AICRPAm) units in different states coordinatedby CRIDA, Hyderabad have refined these

calendars including pest & disease componentwhich is very useful in crop specific advisorypreparation.

The Agro-met Advisory Bulletins are issuedat district, state and national levels to cater theneeds of local level to national level. The StateLevel bulletin is a composite of district bulletinshelping to identify the distressed district of thestate as well as plan the supply of appropriatefarm inputs such as seeds, irrigation water,fertilizer, pesticides etc. These bulletins are jointlyprepared by State Meteorological Centre of IMDand AMFUs and mainly used by StateGovernment functionaries. National AgrometAdvisory Bulletins are prepared by NationalAgromet Advisory Service Centre, IMD, Pune,using inputs from various states. This bulletinhelps identify stress on various crops for differentregions of the country and suitably incorporateadvisories. Ministry of Agriculture is prime userof these bulletins, which help take importantdecisions in Crop Weather Watch Group(CWWG) meetings at national level. The bulletinsare also used by a large number of other agenciesincluding fertilizer, pesticide industries.

Fig. 3. District level agro-met advisory service system


Agricultural scientists at AgrometeorologicalField Units have started using water balance, cropsimulation models as a decision support tool forhelping with weather forecast based farmmanagement decision making. Agro-meteorological Field unit can objectively assessthe impact of skipping irrigation at a particularphenophase of a crop on its dry matter yieldthough with some uncertainties.Agrometeorologists can consider many of thefactors involved, and answer the question with areasonable estimate. The crop models are alsobeing used as technique for prediction of differentphenophases and final yield.

System of Agrometeorological Advsiory Dis-semination

The task is to provide information to helpfarmers make the best possible use of weatherand climate information (Aggarwal, 2002).Critical factors for successful disseminationinclude relevance of information to weather andclimate sensitive decision making in agriculture,followed by good outreach and uptake at grassroot level. To ensure delivery of information tothe farming community, a multi-modedissemination system for agro-meteorologicaladvisories is essential in which besides theconventional modes (radio, television & printmedia), the emerging modes of communicationsuch as mobile phones and internet are alsodeployed. To smarten the service throughcontinuous up-gradation of service one needs backflow of information on quality and relevant ofinformation or demand for specific product, hencetwo way communication has to be the integralpart of the dissemination system.

AAS has considered different aspectspertinent to the flow and content of informationand accordingly evolved a strategy fordissemination of agro-meteorological information.Although concerted efforts are being made to setup two way communications, but as of now theinformation flow is largely one-way. Although,Agro-meteorological Field Units (AMFUs) havelimited interaction with the farmers, goodcommunication and working relationships have

been set up with the agricultural extension, KrishiVigyan Kendra (Agriculture Science Centres),Kisan (Farmer) Call Centre etc. to promoteparticipatory methods for interactions withfarmers. Information is disseminated throughmulti-modes of delivery including mass andelectronic media. It include, All India Radio,Television, Print Media (local news paper indifferent vernacular languages), internet (WebPages) as well as group and individualrelationships through email, telephone etc. Theagrometeorological bulletins always containdynamic information and hence, repetitivedissemination is being made. This reiterativeprocess also helps to address large temporal andspatial variability having significant influence ofweather & climate on agriculture.

The use of electronic media such as e-mail orthe Internet depend on the availability and accessof these methods to the users, which is pickingup in India particularly through initiative ofDepartment of Information technology, who is inthe process of setting up a very strong net workof common service centres (CSC). AAS is ascalable system, which can be incrementallydeveloped and extended to cover all the farmers(crops) of India in a cost-effective manner. Itenables the farmer to receive both crop andlocation-specific expert advice in a timelymanner. With the advent of computers andInternet, emphasis is often being given toelectronic communication systems. However, TVand radio services are still the best ways ofcommunicating advisories among rural people asthese are not only fast methods, but also largeand illiterate masses can be contacted by thesemethods. Broadcasting of advisories in vernacularlanguage provides an edge on other means ofcommunication. Under Integrated AgrometAdvisory Service (IAAS) scheme at IMD/MoESefforts are being made to strengthen the outreachof the agromet advisory as per the need of thefarmers. Under this project, advisories areprimarily disseminated to the farmers by massmode, outreach at village level and human facefor advisory dissemination. Advisories are beingdisseminated to farmers through the followingmulti-channel systems;


i. All India Radio (AIR) and Doordarshanii. Private TV and radio channelsiii. Newspaperiv. Mobile phone / SMSv. Internetvi. Virtual Academy / Virtual Universities /

NGOsvii. Kisan Call Centres / ICAR and other related

Institutes / Agricultural Universities /Extension network of State

viii.Krishi Vigyan Kendra (KVKs)

For dissemination through SMS on mobilephone, an information platform has been created,which allows the existing Agromet Field Units(AMFUs) to provide the information in aconvenient and timely manner. The advisories arecrop and location specific and delivered withinactionable time to the farmers. Under IVR systemthe information from AMFUs for each state arecollected and then stored, and converted intovoice where the farmer would be calling andreceiving the desired information. IMD is sendingthese SMS to farmers with the help of manyprivate and government companies. IMD has sofar partnered with Reuter Market Light (RML),Handygo, NOKIA, IFFCO Kisan Sanchar Ltd.(IKSL), National Bank for Agriculture and Ruraldevelopment (NABARD), Maharashtra StateAgriculture Department (Government ofMaharashtra) and KISAAN SMS portal. Numberof farmers covered under SMS is presently 4.7million.

Capacity Building

Efforts made by Ministry of Earth Sciences,India Meteorological Department (IMD), IndianCouncil of Agriculture Research (ICAR), StateAgricultural Universities (SAUs), Union/StateDepartments of Agriculture and othercollaborative agencies through Agro-meteorological Advisory Service (AAS) havedemonstrated the role of weather forecast inincreasing overall preparedness of farmers,leading to substantially better outcomes.However, more efforts are needed to assist

farmers to further develop their adaptive capacitywith improved planning and better managementdecisions. Hence, more effective approaches todelivery of climate and weather information tofarmers through participatory, cross-disciplinaryapproach is being carried out through enhancingawareness of information user groups. It is donethrough organizing farmer’s awareness programs,roving seminar that brings together research anddevelopment institutions, relevant disciplines, andfarmers as equal partners to reap the benefits fromweather and climate knowledge. Given the currentconcerns with climate change and it’s impacts oncrop productivity, these programmes focus onneed to sensitize the farmers about the increasedweather variability, in their regions, and differentadaptation strategies that can be considered tocope with the extreme weather situations.

Such programs help increase the interactionbetween the local farming communities and theMeteorological Centres (MCs), Agro-Meteorological Field Units (AMFUs) and KrishiVigyan Kendras (KVKs). Considering above, alarge number of such seminars are organized indifferent agro-climates of the country to sensitizefarmers about the weather and climate informationand it’s applications in operational farmmanagement. These are jointly organized by IndiaMeteorological Department (Ministry of EarthSciences), Indian Council of AgriculturalResearch and State Agricultural Universities,Local NGOs and other stake holders.

In order to develop a local (village) level rainmeasuring network and improve the rainfalllinkage with the AAS system, 5 rain gauge madeof plastic are distributed to the selected group ofprogressive farmers during the meeting who aretrained to record and report the rainfallobservation to the concerned AMFUs who in turncommunicate the data to IMD. So far suchprogrammes have been organized at 104 AgrometField Units.

IMD has brought out Agromet Brochure indifferent languages highlighting the activities ofIntegrated Agromet Advisory Services for thebenefit of the Indian farmers. These are circulatedto the concerned organizations/institutes like


Ministry of Agriculture (Central & State), IndianCouncil of Agricultural Research (ICAR), StateAgricultural Universities, Regional ResearchInstitutes, Krishi Vigyan Kendras, Department ofSpace, NGOs and all other organization directlyand indirectly related to the agricultural servicesin the country. Besides, farmers are being awareof the service through participation of KisanMela, field visit, field demonstration, field day,farmer field school. During this programmefarmers are explained about various services insimple components and in local language and theimportance of the service, discussion with themwhat they already know about the skill todetermine where knowledge is missing andallowing the famers to record the weatherobservations using various weather instrumentsand familiarise the agro-meteorologist about theproblems faced by the farming community andexpectations from the service providers.

Crop Yield Forecasting

Reliable and timely forecast of agriculturalcommodity provides important and useful inputsfor proper planning. This enables planners anddecision makers to predict how much to importin case of shortfall or optionally, to export incase of surplus. It also enables government to putin place strategic contingency plans forredistribution of food during times of famine. InIndia, there is also growing need for micro levelplanning and particularly the demand for cropinsurance, which increases the need for field levelyield statistics. Forecasting Agricultural Outputusing Space, Agrometeorology and Land basedobservations (FASAL) involves developingeconometric, remote sensing and agromet basedmodel to generate multiple crop yield forecasts atnational, state and district level starting with cropsowing to end of season for 11 major kharif andrabi crops viz., Rice, Jowar, Maize, Bajra, Jute,Ragi, Cotton, Sugarcane, Groundnut, Rapeseed &Mustard and Wheat.

IMD is implementing Agromet component ofthe scheme in coordination with 23 State AgrometCentres (SAMCs), 46 principal Agro-Met FieldUnits (AMFUs) and IASRI in the country to

develop agromet models and issue in-season cropyield forecast based on statistical and cropsimulation models. This information is sharedwith Mahalanobis National Crop Forecast Centre(NCFC), New Delhi.

Future Needs and Strategies

Though Agro met Advisory service is beingprovided efficiently at district level in the country,still there is need to strengthen in terms ofobservation, seamless weather forecast,manpower, real time information flow, researchand development (R&D), dissemination etc. Thereis need to develop methodologies for remotesensing and conventional data merging. Concertedefforts are needed for ground based datacollection, satellite data collection, GIS softwareapplications, operational applications ofmeteorological satellite data, weather radars andthe monitoring of cropping season bymeteorological and remote sensing data to equipAAS units to generate better advisories. Thoughdistrict level medium range weather forecast isbeing prepared for agromet services, there is anurgent need to develop and issue high resolutionaccurate weather forecasts to help tailor crop planfor the farmers at village level. Thus there is aneed to scale up the service from district to blocklevel with dissemination up to village level acrossthe country to meet the end user’s requirement.Such idea leads to take up the new concept ofestablishing the Gramin Krishi Mausam Sewa(GKMS) scheme under Integrated AgrometAdvisory Services in the country during the XIIthFive Year Plan. Some of the important activitiesbeing planned under GKMS are as follows.

Seamless Weather Forecast

a. Block level forecast

Though district level medium range weatherforecast is being prepared for agromet services,there is an urgent need to prepare and issue highresolution (9 km × 9 km or finer using WRFmodel) accurate weather forecast at sub-districtlevel. Thus it is proposed to generate block levelforecast during XIIth Five Year Plan. On pilot


mode block level weather forecast for theparameters like rainfall, maximum and minimumtemperature, cloud amount, maximum andminimum relative humidity, wind speed and winddirection for 342 blocks in the countryconsidering one district per state is beingdeveloped. This procedure would be followeduntil the bias removal techniques and highresolution (10-12 km) GCM becomes operational.Neural networks would be tried for MMEforecasts and Kalman filter would be tried formaximum and minimum temperature as the biasremoval techniques. As such, this forecast hasbeen started for 37 selected districts in thecountry. By the end of XIIth FYP, block levelforecast along with advisories will be scaled tocover the maximum districts in the country.

b. Extended range forecast

b.1.Under National Monsoon Mission set up byMinistry of Earth Sciences, Indian Instituteof Tropical Meteorology (IITM), Pune incollaboration with IMD and National Centrefor Medium Range Weather Forecasting(NCMRWF) has started experimental forecastfor monsoon rainfall during next four pentads(up to 20 days) issued every week. IMD hasstarted National Agromet Advisory bulletinsbased on Extended Range Weather Forecastduring southwest monsoon 2013 to fulfil theneeds of farmers and other users. Bulletin wasprepared and issued for the next fortnight withupdate on every Friday. Successfulimplementation of this initiative would helpthe farmer to get more lead time to managehis weather based farm management.

b.2.In another effort entitled “Development andApplication of Extended Range ForecastSystem (ERFS) for Climate Risk Managementin Agriculture” supported by Ministry ofAgriculture, Government of India, the keyobjectives to develop the experimentalextended range test forecast from month toseason scale for Indian region and itsapplication for the development of climaterisk management tools to address these riskswith the involvement of agricultural scientists,

farmers and service support agencies at fieldlevels. This project integrates riskmanagement and climate science research,involving leading institutions of India andabroad. The project adopts a very strongdemonstration approach, focusing on selectdistricts in nine states that face significantlivelihood impacts due to variability in thesouthwest monsoon: Andhra Pradesh, Gujarat,Maharashtra, Madhya Pradesh, Orissa,Himachal Pradesh, Rajasthan, Tamil Nadu,and Uttarakhand.

These forecast products include both type offorecasts viz., deterministic as well as probabilisticforecasts. Methods used to generate thedeterministic rainfall and mean temperatureforecast include Singular Value Decomposition(SVD) based Regression, Supervised PrincipalComponent Regression, Canonical CorrelationAnalyses. In these regression techniques,predictors (independent variables) are differentgeneral circulation models’ rainfall andtemperature products while the predictand(dependent variables) is IMD rainfall andtemperature. All these above stated forecasts werecombined using some statistical techniques andhas been converted in probabilistic forecast to seethe chance of occurrence of particular event.Probabilistic forecast is made in three categoriesviz., Below Normal, Near Normal and AboveNormal. These categories are based on theprevious year’s observed data set (IMD’s griddeddata). The experimental forecast products thusgenerated since monsoon 2009 were disseminatedto all the selected demonstration sites throughIMD for its verification and application purposes.The performance of experimental test forecastproducts were evaluated at each selected pilotdemonstration sites with the observed data oftemperature and rainfall during the month.Performance of real time experimental forecastfor summer monsoon seasonal rainfall (JJAS) isshown in Fig. 4.

Climatic Risk Management

The idea of informed decision making atfarm/ farming system level gains more reality if


Fig. 4. Performance of real time experimental forecast for summer monsoon

integrated with reliable seamless weather forecastduring the crop season. Crop simulation modelsand DSS are being increasingly used in decisionmaking for risk minimization in agriculture. Thesemodels need daily weather data for crop season.The major challenge in linking seasonal forecastto crop model is mismatch between spatio-temporal scale of forecast and crop model. Onthe one hand, these forecasts are available atbigger spatial domain for a month or season andon the other, crop simulation models can be runat point scale or smaller grid space. Forecastneeds to be downscaled spatially and temporally.Crop model based generated viable managementoptions against a given forecast require frequentvalidation through farmers participatory meetings.In order to make use of extended range forecastfor decision making on strategic agricultureplanning and risk management and integrate withongoing advisory service system, a range ofclimatic risk management tools were developed.

1. Extended/ seasonal climate forecast.

2. Climatic risk matrices: Climate risk analysisfor crop under various farming situationsshowing the impact of soil (type and depth),time of planting and weather requirement andextremes at various phenopahese.

3. Stochastic Weather generator: Downscalingcan be achieved by linking a seasonal forecast

with a weather generator. Nonhomogeneoushidden Markov model (NHMM) or HMM areused as a promising approach to constructingmulti-station weather generators.

4. Crop simulation model and Decision supportsystem: Crop models embedded in DecisionSupport System for Agro-technology Transfer(DSSAT), InfoCrop, APSIM etc. are process-oriented, management-level model of cropgrowth and development and can be linked todaily sequence of desegregated weather datato generate crop growth performance againsta number of weather realizations.

5. Farmers participatory meeting: A very criticalprocess to validate/refine the above findingsas viable options for farm management.

Pilot study demonstrated use of monthlyforecast on rainfall and temperature byestablishing an end-to-end information generationand application. This was implemented withparticipation of agricultural universities/researchcentres, service support agencies at field levelsand farmers for better crop planning andmanagement in the following areas:

• Land preparation• Nursery raising• Irrigation scheduling• Urea application to rainfed crops in normal

rain forecast


• Frequent intercultural operation to conservesoil moisture

• Contingency plan under delayed monsoononset rains

• Temperate apple crops- Chilling hours to helpskip hormonal spray for bud breaking/ flowersetting

• Nomadic farming hills- sheep, goat• Livestock- malnutrition and disease infection

under deficit rainfall

Future plan includes the integration of ERFSexperimental system in IMD for itsoperationalization, generation of seamless weatherforecast for agriculture risk management, expandto all parts of the country and R&D for continuousimprovement in weather forecast products atfortnightly to monthly scale.

eAgromet System

As there is a need to develop a computer-based decision support system and to automatethe process of advisory bulletin preparation anddissemination, IMD in collaboration with IIIT,Hyderabad is generating a software forpreparation of advisory bulletin under the scheme“IT-based eAgromet software for automated AASbulletin preparation” to prepare district/blocklevel AAS bulletins. It simplifies the preparationof district level AAS bulletin as it acts as anorganizing tool for weather forecasts and agrometadvisories with the help of repository of past AASbulletins. A server has been launched by IIIT forthe said purpose. Number of AMFU scientists hasalready been trained and remaining AMFUscientists will be trained in utilization of thesoftware and it will be implemented in GKMS.

Smart Dissemination System

IMD in collaboration with National RemoteSensing Centre (NRSC) is planning to usegeoportal ‘BHUVAN’ to establish direct (virtual)interaction between farmers and the scientists tosolve the field specific problems on real time basisby generating customized advisory and thus tominimize risk factor at farm level. Registered

farmers raise queries and upload the crop imageson operational farm activities to get specificadvice. As geoportal is GPS enabled, this willhelp to get the location specific feedback fromthe farmers as well as status of the crops, whichwill enhance the ability of the scientists to prepareagromet advisories more customized toindividual’s need. Also BHUVAN portal willprovide a platform for the dissemination of theexisting and proposed agrometeorologicalproducts over a wide area and for a large numberof users.

Use of Satellite Data to Derive New ProductIncluding Alerts

The regular and national scale agro-meteorological monitoring of the physiologicalprocesses and growth indicators require retrievalof basic land surface variables using spatialobservations from the suit of current and futureIndian geostationary satellite sensors. Satellitebased agrometeorological products and theinterpretation of the same in terms of crop andsoil moisture status will help the experts toaddress spatial variability in framing theadvisories and ultimately improve the quality ofthe advisories at higher resolution. A number ofnew products for agromet services are beinggenerated recently. IMD in collaboration withSpace Application Centre (SAC), Ahmedabad isusing the Normalised Difference Vegetation Index(NDVI), derived from INSAT 3A CCD data, foragromet advisories. IMD is developing thesoftwares “Operationalization of Satellite DataBased Soil Moisture Monitoring System in India”in collaboration with IIT, Roorkee and“Determination of Soil Moisture over India usingSpace Borne Passive Microwave Sensors on boardSMOS” in collaboration with ICRS, Jodhpur. Soilmoisture data generated using this software willbe used in water balance, irrigation schedulingand determination of sowing dates etc. for AAS.Besides, it is proposed to generate a number ofsatellite products in preparation of agrometadvisories including alerts to the farmers withrespects to extreme weather.


Activities under CREAM and R&D Relatedto Weather and Climate

Climate induced vulnerability of agriculturecause plateau in agriculture productivity in thecountry in recent past. Projected climate changecan influence pest and disease dynamics resultingin crop losses. An improved adaptation strategy,particularly in the rainfed areas where climatevariability is large, holds the key to improvingfood production for the growing population inIndia. It is important to address the climatic riskin agriculture, which requires development of riskmanagement tools and their integration in GKMS.A Cell for Research and Excellence in AgricultureMeteorology (CREAM) is proposed to beestablished at IMD New Delhi interfacing withIARI New Delhi for irrigated farming andCRIDA, Hyderabad for dry land farming.CREAM will attempt to channelize/operationalizethe R&D work for climate risk mitigation in thefarming sector in India and also help in RA-IIregion under WMO as Agromet Advisory Serviceof IMD is well recognized by world community.R&D work projected under CREAM willsynergize with the relevant programme/activitiesof CRIDA (AICRPAM and NICRA) foragricultural research programmes. Its objectiveincludes (i) to develop climatic risk managementtools, (ii) to help farmers cope with climate risksand uncertainties (iii) to help reduce thevulnerability of agroecosystems to climatevariability and change, (iv) to help achieve greaterefficiency in natural resource use in agriculture,(v) to establish a World Class TrainingProgramme in Agricultural Meteorology, (vi) toorganize periodic Symposia/Conferences onAgrometeorological Research and Applications.

Economic Impact Assessment

Impact assessment framework made byNational Centre for Agricultural Economics andPolicy Research (NCAP) was adapted to assesseconomic impact by taking field experimentsconducted from 2003-2007 at 15 locations spreadin country covering field crops, vegetables andhorticulture (Rathore and Maini, 2008). Forecastbased agro-advisories were issued by NCMRWF

to AAS users. An objective comparison betweenAAS and non AAS farmers at village level showsthat impact depends on the risk taking ability offarmers. In quantitative terms, study reports thatthe AAS farmers were able to reduce the cost ofcultivation by 2-5% except in case of fruits wherethe cost of cultivation has increased by 5-10%.This shows that the right selection of fertilizersand seed due to organization of awarenessprogramme in the village and spraying ofappropriate pesticide due to advisory saved inputcost. It is also observed that the yield increasedby almost 10-25% in most of the crops withmaximum increase in the fruit crops.

National Council of Applied EconomicResearch (NCAER), an independent agency,during 2009-2010 estimated that the economicbenefit from the use of weather information asthe product of the percentage of farmers receivinginformation, scenario-wise, times the percentageof farmers benefiting from the information timesaverage profit, crop-wise, attributable to weatherinformation times the total national production ofcrops. Conversion factors, crop-wise, were usedto convert farmers’ financial profits to economicprofits (NCAER, 2010). At present only 10 to 15percent of the farmers are benefitting from theSMS services. The economic profit estimates varybetween Rs. 50,000 crore (where currently 24percent farmers receive weather information byall means) to Rs. 211,000 crore (where all farmersreceive weather information).

In order to judge worthiness and role of theagromet service in improving the agriculturalproduction system and farm management, itseconomic impact will be assessed throughconducting extensive feedback surveys as guidedby an independent agency. Studies will beundertaken at different representative sitescovering principal crops. The study will compriseof kharif (monsoon) and rabi (winter) cropseasons including livestock. The objective of thestudies will be to observe whether weather basedadvisories have a positive impact on the overallyield and also help in reducing the cost ofcultivation.


Gramin Rainfall Monitoring Mission(GRaMM)

IMD with help of other Govt. organizationsis planning to monitor the rainfall at village levelin India by creating network of manual rainguagesunder the project called GRaMM (GraminRainfall Monitoring Mission). Village Panchayatoffice will be involved in this project for upkeep,measurement and communication of rainfall everyday. This village level raingauge network infuture will facilitate the existing gridded datamodelling framework to process and create thespatial continuous rainfall grids at every kilometreor even finer scale.

Rainfall Visualizer Tool

Rainfall visualizer is a simple tool developedunder ACCA project operating in Telanganaregion of Andhra Pradesh, which enables thefarmers to track the village level rainfall situationof current season in comparison with previousyears rainfall. By entering the current year’s dailyrainfall data farmers can observe the three curveswhich depicts current season rainfall progressingcurve compared to the wet, dry and last year’srainfall. They can easily understand the value ofcomparing the present rainfall situation with thegood, bad and normal situations from the point ofview of learning from their own past experience.For them it can be a useful tool to discuss thecurrent situation and regulate their actions basedon their past experience. The Visualizer provides

the farmers a tool to track the current rainfall toknow which particular trend (good, bad, ornormal) was most likely to follow. This in turnwould help them to decide on their irrigation,fertilizer and pesticide application planning in thecurrent season as well as prepare them for therabi season. The tool would help them especiallyin deciding the number of irrigations that may berequired so that in times of water shortage thefarmers could save water for critical irrigation.

ReferencesAggarwal, P.K. 2002. Climate Change and Agricul-

ture: Information Needs and Research Priorities.International Conference on Capacity Buildingfor Climate Change 21 October 2002.www.ficci.com/media-room/speeches-presenta-tions/2002/oct/oct-climate-aggarwal.ppt

Kanamitsu, M., Kumar, A., Ming, H., Jung, H., Kyung,J., Wang, W.I., Yang, F., Hong, S., Peng, P.,Chen,W., Moorthi, S. and Ji, M. 2002. NCEPDynamical Forecast System 2000., Bull. Am.Meteorol. Soc. 83: 1019-1037.

National Council of Applied Economic Research(NCAER). 2010. Impact assessment and eco-nomic benefits of weather and marine services.Supported by Ministry of Earth Sciences, NewDelhi.

Rathore, L.S. and Maini, P. 2008. Report on Economicimpact assessment of agrometeorological advi-sory service of NCMRWF. Ministry of EarthSciences, New Delhi.

weather information for sustainable agriculture in … information for sustainable agriculture in...

Documents