Communicating agrometeorological information to farming ... and Forest Meteorology 2000... · Communicating agrometeorological information to farming communities A. Weissa;, L. Van

Agricultural and Forest Meteorology 103 (2000) 185–196

Communicating agrometeorological informationto farming communities

A. Weissa,∗, L. Van Crowderb, M. Bernardica School of Natural Resource Sciences, University of Nebraska, Lincoln, NE, USA

b Extension, Education and Communication Service (SDRE), Research, Extension and Training Division, (SDR) Food and AgricultureOrganization of the United Nations (FAO), Rome, Italy

c Environment and Natural Resources Service (SDRN), Research, Extension and Training Division (SDR), Food and Agriculture Organizationof the United Nations (FAO), Rome, Italy

Abstract

Agrometeorological information, used for decision making, represents part of a continuum; at the other end is scientificknowledge and understanding. Other components of this continuum are the collection of data and transforming data intouseful information. Information has value when it is disseminated in such a way that the end-users get the maximum benefit inapplying its content. This paper explores the potential of the new information and communications technologies to improve theaccess to agrometeorological information. The Internet will play an important role in the collection and transfer of information.In developing countries, Multi-Purpose Community Telecentres (MCTs) will be the equivalent of an information supermarket.Radio can be used to transfer information from MCTs to rural areas. Using response farming as an example, a prototypeinformation system that can have wide applicability is suggested. Procedures on evaluating the impact of agrometeorologicalinformation are provided. Future concerns about the information needs of diverse end-users, information on a fee basis, and thetraining needs of end-users and intermediaries are discussed. Although modern technology has improved agrometeorologicalinformation and increased the number of end-users, continued improvements are necessary to ensure that the content ofthe information is adequate to fulfill the requirements of the farming communities. © 2000 Elsevier Science B.V. All rightsreserved.

Keywords:Agrometeorology; Information; Communication; Farmers; Technology

1. Introduction

Information can be identified as the cornerstone tosuccessful farming in the 21st Century. Agriculturalproducers already know that information is importantand valuable. Many farmers in developed countriesalready pay for information ranging from updates onweather conditions, soil and nutrient status, pest man-

∗ Corresponding author. Tel.:+1-402-472-6761;fax: +1-402-472-6614.E-mail address:[email protected] (A. Weiss)

agement reports and recommendations, to advice onwhat genetic seed line to plant for various field condi-tions. The price that farmers pay for information can bethought of in cost per hectare. Farmers will judge theexpected benefit of acquiring information against thecost of implementing a recommendation, e.g., a pes-ticide application. Technology that provides informa-tion to a producer, which would cost more than the costof the recommendation (a pesticide application), willhave a very limited chance of being readily accepted.

It must also be considered that timely availabilityand appropriate use of agrometeorological information

0168-1923/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved.PII: S0168-1923(00)00111-8

186 A. Weiss et al. / Agricultural and Forest Meteorology 103 (2000) 185–196

are vital to successful farming operations. In order forthis information to be considered as a resource it mustbe shown to have value to the farmers as well as to thevarious agencies that provide support to the develop-ment, generation, and dissemination of this informa-tion. The benefits of agrometeorological informationneed to be communicated to the many potential usersin farming communities, especially those in develop-ing countries where agricultural production is low andfood insecurity is prevalent.

Agrometeorological information is part of the con-tinuum that begins with scientific knowledge andunderstanding and ends with evaluation of the in-formation. Scientific knowledge and understandingtranscend national borders, cultures, and societies. Forexample, the C3 photosynthetic pathway is the samein Argentina as it is in Canada. To use this knowledgeand understanding to generate agrometeorologicalinformation is a function of the user community.While the problems may be similar across differentcommunities, they differ due to the various degree ofavailability of human, financial and natural resources.Major components that also may differ between de-veloped and developing countries are: capabilities forcollection of data, changing data into useful infor-mation, dissemination of information, and evaluationof information. Collection of data and changing datainto useful information has been discussed by Do-riaswamy et al. (2000), Maracchi et al. (2000) andHoogenboom and Hammer (2000).

The objective of this paper is to address methods ofdisseminating and evaluating agrometeorological in-formation and to provide examples of these processes.Although concentrating on the developing countries,the ideas presented here are equally valid for thedeveloped world, as the underlying concepts are thesame while the implementation is different. The paperconcludes by raising and responding to some criticalquestions about communicating agrometeorologicalinformation in the future.

2. User community, data, and information

We begin by defining the ‘user community’. Theuser community for agrometeorological data and in-formation can be understood in its broadest sense tocover the ‘spectrum from institutions and governments

to farmers at the subsistence level; the methodologiesand products used to reach the different categories ofusers vary greatly’ (Gommes, 1997).

The content of agrometeorological information can-not be standardized because of the diverse nature ofthe user community. Depending on its purpose, thecontent of information can be related to: early warn-ing advice to prevent famine crisis; the developmentof agricultural planning policies; the development ofnational climate policies as a follow-up of the KyotoProtocol on Climate Change; special advisories pro-vided to farmers through the national extension ser-vice; general advisories accessed by farmers directlythrough the electronic media; and to general issuesdisseminated to the public.

It is important to distinguish between ‘data’ and‘information’. Data are digits with the appropriateunits, e.g., MJ/m2 day, degree Celsius, mm of rain,which in these cases represent incoming solar radia-tion, air temperature, and precipitation. These digitsmay come from observations or predictions. By them-selves, they are not useful for making decisions.

Information is the result of utilizing a statisticalsimulation or other type of model, such that the outputcan be used to make a decision. For example, fromprecipitation data, probabilities of occurrence can bedeveloped, e.g., there is a 75% chance of rain occur-ring within the next 5 days; thus today may be a goodtime to plant. In this example, there is direct linkageto meteorological variables. For simulation models,meteorological inputs along with physiological typeinputs do not appear in the outputs. The resultingoutputs from these simulation models can be put ina form appropriate for the decision-maker, whetherit be a probability distribution of yields as a functionof irrigation amounts, a table of projected outcomesfor different input options such as plant populationdensity or fertilization amount, or the economic con-sequences of these different input options.

3. Dissemination of agrometeorologicalinformation

Agrometeorological information must be dissemi-nated in an optimum way based on the type of ad-vice required and the needs of the end user. Valuableagrometeorological information can be transmitted to

A. Weiss et al. / Agricultural and Forest Meteorology 103 (2000) 185–196 187

farmers through the extension services or through themedia. In developing countries, radio broadcasts ofagrometeorological information to rural farmers playsan important role to provide timely advice at the locallevel. In order to improve the use of this method ofcommunication among farming communities, in 1992,in Bamako (Mali) a meeting was organized by WMO,FAO and the Technical Center of Agricultural and Ru-ral Co-operation (CTA) on the use of rural radio totransmit advice to farmers (WMO, 1992).

Another way to disseminate information is through‘agrometeorological bulletins’. At present, most ofthese bulletins include just meteorological data. Mostof the times, agricultural information contained inthese bulletins is generated by interviews with farm-ers, particularly when the agrometeorologist is faraway from the area to be monitored. This kind offield information can be very useful, for example, toassess the predictions produced by crop simulationmodels and to report to central bodies about cropconditions. However, when the farmer makes use ofadditional inputs (e.g. fertilizer, pesticides, high po-tential varieties, etc), his own experience may not besufficient to take advantage of these inputs. In thissituation, the farmer would require additional advicebeyond that normally received. This information canbe supplied through the local extension services orspecialised advisory services, and if current trendscontinue, eventually for a fee.

4. Agrometeorological advisory services forresponse farming

A major factor affecting yields and production in de-veloping countries is the inter-seasonal weather varia-tion. There is a clear need to assist subsistence farmersin supplying information in order to adapt the agricul-tural system to increased weather variability. Farmersshould be able to plan their practices, given the re-cent past weather conditions, i.e. through contingencyplanning and response farming.

Important contributions to improve agriculturalproduction and food security in developing countriescan be achieved by ‘more efficient agrometeorologicaladvisory services to farmers to stabilize their yieldsthrough management of agroclimatic resources aswell as other inputs (fertilizer, pesticides)’ (Gommes,

1997). Advisory services to farmers can be improvedthrough close collaboration and co-ordination amongrelevant agencies and organizations, national exten-sion services, national agrometeorological services,specialized NGO communication outlets and farmingcommunities.

Response farming represents a set of techniques tooptimize farming practices as a function of currentenvironmental conditions under unusually dry condi-tions which are likely to lead to poor yields and nullifythe added economic return from the use of the inputs,or an increased effort to capture the benefits from un-usually favorable rains. In particular, ‘response farm-ing, requires co-ordination among these entities so thatexperimentation can be carried out under local pilotconditions before application on a wider scale’. A rec-ommended methodology involves the following steps:• Identify the main focus for agrometeorological ad-

visory services and response farming; determine thetechnical and operational details of pilot activities;

• Develop the decision tools, including suitable insti-tutional arrangements (national agrometeorologicalservices and extension agencies); briefing and train-ing will be required;

• Carry out statistical, agronomic and economic anal-yses of the response of selected crop productionsystems as a function of local decision parameters;

• Transmit the experimental advise to farmers throughselected channels and monitor implementation;compare output with and without advice;

• Make available phenology forecasts, yield forecasts,regular weather forecasts and price information tofarmers and monitor use in order to assess theirpotential for the farming community at large;

• Maintain regular contacts with media outlets andwork with them to adapt agrometeorological infor-mation to the appropriate formats and messages;

• Evaluate the pilot phase and if successful, makeplans to expand, i.e., decide on geographic area tocover, prepare documentation on the methodology,train extension technicians and others, determine themost appropriate media for channeling the advice;

• Institutionalize the working arrangements devel-oped during the pilot phase and subsequent phaseand prepare long-term plans to cover more areasand crops (Gommes, 1997).While the above methodology was developed within

the context of response farming, these concepts are


equally valid to address the informational needs ofother types of farming operations. The reason for thegenerality of this methodology lies in its basis in thescientific method which transcends geographic bound-aries. As noted earlier, the differences between scalesof operation are in availability of human, financial,and natural resources.

WHARF, an educational and scientific organizationfounded in 1984 in Davis, California, is dedicatedto extending ‘response farming’ worldwide (Stewart,1998). Response farming originated from research onrainfall behavior and its predictability in a croppingsystem project in Kenya sponsored by USDA and US-AID. The methodology also includes crop yield esti-mation procedures developed at UC Davis plus sub-sequent findings on rainfall behavior sponsored byWHARF.

In Kenya, the project was to design sustainable crop-ping systems for low-resource farmers in marginalrainfall zones characterized by great seasonal rainfallvariability, uncertainty and recurrent drought. Despitethe swing from large-scale high-tech irrigated farmingin the US to small-scale low-tech rainfed farming inKenya, crop interactions with the soil–water–weatherenvironment held to the same principles. However,great rainfall variation made it clear that no fixed setof cropping procedures could both minimize the cropfailure rate in poor seasons and produce above averageyields in good seasons. The need for flexibility in cropmanagement required some useful level predictabilityof seasonal rains. The selected predictor was the dateof onset of the rainy season. The question was, for anygiven crop(s), how do localized rainfall probabilitiesdiffer between seasons in which onset of rains is lateversus seasons when it is early?

Response farming analyses uses localized dailyrainfall records for farms, groups of farms, or avillage, or groups of records representing projects,regions, etc. Initial analyses determine an onset datefor each season of record, critical date separatingearly seasons from late seasons; reanalysis is thendone of the rainfall record for selected crops andtheir growing periods. Having decided on crops, va-rieties and acceptable planting criteria, the next stepis decide on slopes of plant rows, plant numbersand row spacing as well as fertilizer, which is es-sential to markedly increase yields in good rainfallyears.

Rainfall analyses reveal the critical date separatingearly from late seasons and guide in designing an earlycropping system with suitable modifications for thelate seasons. Full details are communicated to farmers,who prepare in advance to follow either plan as soonas the current season date of onset is revealed.

5. ICTs: access and applications in developingcountries

This paper looks specifically at how information andcommunication technologies (ICTs), and especiallycomputer-based and Internet communication tools,can be effectively used to reach farming communitieswith agrometeorological information for appropriatedecision-making. Information (its generation and dis-semination) and technology are closely linked. Thenew ICTs generally refer to computer-based technolo-gies and telecommunications. ICTs collect, process,store, retrieve, and disseminate data and informationusing microelectronics, optics, telecommunicationand computers. Talero and Gaudette (1996) viewICTs as hardware, software and media for collection,storage, processing, transmission and presentation ofinformation. Hamelink (1997) says that ICTs enablethe handling of information and facilitate differentforms of communication among participants, be-tween participants and electronic systems and amongelectronic systems.

Looking at the role of the Internet in agricultural de-velopment, Richardson (1997) concludes that ‘moderninformation and communication technologies usingmulti-media computer links and computer tools canpotentially benefit national agricultural knowledge andinformation systems and users in developing countries,including researchers, extension agents and farmers’.He also says that rural communities represent the ‘lastmile of connectivity’. While it is conventional wisdomthat ‘we live in the information age’, in an era charac-terized by information ‘superhighways’ that span theglobe, it is evident that there is a large gap betweenthe ‘information rich’ and the ‘information poor’.

Nevertheless, the Internet is becoming increasinglya dynamic market as a result of private-sector com-mercial undertakings in countries where monopolisticPublic Telecommunication Operators (PTOs) have lib-eralized Internet service provision (e.g., Tanzania and


Ghana). Forty-seven of the 54 African countries havesome form of Internet access in the capital cities, eitherthrough a local dial-up, store and forward e-mail witha gateway to the Internet, or through a full leased-lineservice. Of the 300 or so Internet Service Providers(ISPs) on the continent, about 200 offer full Internetservices; the highest concentration is in South Africa,with over 70 ISPs (Jensen, 1998).

Some observers see the lack of telecommunicationinfrastructure as a possible advantage in the long run.The low-level of infrastructure may mean that whennew telecommunication infrastructure is installed, itwill be digital and possibly wireless, from the begin-ning. Since Internet services rely heavily on the extentof network digitization, developing countries just start-ing Internet services may be able to quickly developdigital networks. This has been the case in countriessuch as Botswana, Gambia, Mauritius and Rwanda,where a large percentage of the lines are digital (Man-nisto et al., 1998).

While it is clear that it will be many years beforethe majority of rural people in Africa, and elsewherein the developing world, have a reasonable level ofaccess to even basic telephone services, there aremany recent technological developments which canmake telecommunication services more widely avail-able. For example, Jensen and Richardson (1998)recently assessed the potential of new developmentsin the use of radio frequencies for ‘wireless weaves’that can help close the gaps in rural telecommuni-cation coverage in Africa. Such an approach couldcombine land-based infrastructure (copper and fibreoptic cables) with technologies such as Very SmallAperture Terminal (VSAT) satellite systems, digitalcellular phone systems and point-to-point rural radiosystems that make use of advances in cellular phonestandards. The use of VSAT offers reasonably highbandwidth (64Kbps–2Mbps) and substantially lowercosts than internationally leased circuits supplied bymost PTOs. In Ghana, Uganda, Tanzania and Mozam-bique, where government regulation has been relaxed,various VSAT-based ISPs have sprung up.

‘Telecentres’ are increasingly being seen as ameans to provide a wide range of telecommunicationservices to rural residents through a single accesspoint. Multi-Purpose Community Telecentres (MCTs)are being established in various countries in Africa,Asia and Latin America by the International Telecom-

munication Union (ITU), with various national andinternational partners. Located in a shared rural com-munity facility, MCTs can offer telecommunicationservices such as telephone, fax, e-mail and Internetaccess along with training and support in their use.In the design of MCTs, attention is being given tospecific applications and content for several sectors,for example, health, education and agriculture.

In Mali, FAO is helping to develop agriculturaland rural development applications for an MCT inTimbuktu. Depending on the results of a survey ofthe information needs of local rural groups, applica-tions may include marketing information, specializedinformation (e.g., weather, environmental protection,disease and pest monitoring, water management),agricultural products and services trading networksand information networks linking research and exten-sion agencies, agricultural colleges and even farmerorganizations. There are plans to link the MCT tocommunity radio stations and provide training inadapting information from the telecentre for broadcastto wider audiences in local languages.

In many cases, the effectiveness of ICTs foragrometeorological information dissemination can beenhanced by linking them to other communicationmedia, especially media which are more accessible tofarmers such as rural radio. In this way, a ‘multipliereffect’ can be achieved.

Changes in radio technology will have an impor-tant impact on the dissemination of information. Lowpowered radio stations that can broadcast a signalwithin a radius of 20 km are available at the cost ofa moderately priced PC. Radios that do not dependon batteries or line voltage are ideally suited for usein remote locations. These radios have a built in gen-erator that is operated by a crank mechanism, onerevolution is equivalent to approximately 45 min ofplaying time. A major limitation of this radio is thecost, currently around US$ 100.

The FAO has established Internet information andcommunication networks among farmers in Chile andin Mexico (Richardson, 1997). Through informationcentres located at the offices of farmer organizationsand NGOs, the networks provide farmers with dataon crops, international crop status, regional, nationaland international marketing information, weather andtechnical crop production information. The most im-portant aspect of this initiative is its specific attention


to the assessment of local information needs and theprovision of training the staff of farmer organizationsand NGOs to help them develop the skills necessaryto access, analyze, adapt and disseminate informationthat is locally relevant.

6. Communicating agrometeorologicalinformation

The advent of ICTs has impacted many aspectsof agrometeorology, the collection and archivingof meteorological data and the generation of cli-mate related information being just one importantexample. In the developed countries, networks ofautomatic weather stations have existed since theearly 1980s (Hubbard et al., 1983). Data from thesenetworks have been used in a wide variety of ap-plications, e.g., phenological predictions, irrigationscheduling, yield predictions, and pest managementstrategies. In 1980, the University of California es-tablished the Statewide Integrated Pest ManagementProject (UC-IPM) to develop and promote the useof integrated, ecologically sound pest managementprogrammes (http://www.ipm.ucdavis.edu/). The Cal-ifornia Weather database (http://www.ipm/ucdavis/WEATHER/weather1.html) stores both current andhistorical daily weather data for approximately 350weather stations throughout California. This databaseconsists of data from: automated weather stations,part of the CIMIS (California Irrigation and Manage-ment Information System), operated by the CaliforniaDepartment of Water Resources; volunteer observers;and weather stations operated by NOAA.

The poor telecommunication infrastructure in mostdeveloping countries, and in particular in remote ru-ral areas, makes the use of receive-only Intranetsand receive-only PCs for ‘data-casting’, or the trans-mission of data at regular intervals via satellite, anattractive option. The Inmarsat-D system provides agood example of data casting. The Inmarsat organi-zation, located in London, is the provider of satellitetelephone services to over 70 countries. Only a smallproportion of the capacity of the system is used totransmit digital information, for example, about mar-ket prices and weather conditions. To receive theinformation, all that is needed is a small palm-sizedreceiver connected to a PC, which stores the incom-

ing data. In a similar fashion, farming advisory ser-vices, or even farmers through their organizations orthrough telecentres, could access agrometeorologicalinformation with small dish antennas and a PC. Theinformation and communication technology is avail-able in many cases; what are lacking are agromete-orological data-casting services that convert data intosite-specific information that farmers can use for agri-cultural decision-making, including response farming.

The following examples from a variety of sourcesillustrate the different approaches that can be used tocommunicate agrometeorological information. It is notan exhaustive list. No doubt given the rapid changesin ICTs, some of these examples will be outdated ina short time, while new applications of ICTs will bedeveloped and accepted.

The California PestCast system is a joint effort ofthe UC-IPM Project, the US Environmental ProtectionAgency, and the California Department of PesticideRegulation (http://www.ipm.ucdavis.edu/DISEASE/california pestcast.html). The overall goal of this ef-fort is to expand the application of computer basedcrop disease forecasting in order to reduce unneces-sary pesticide usage. There are currently 15 diseasemodels available for use in fruits, vegetables, andturf. Weather data for use in these disease forecast-ing models comes from automated stations located inthe appropriate fields. Data are transmitted either bytelephone, cellular phone, or radio telemetry and canbe reported on intervals of 15 min, hourly, or daily.This database is different than the California WeatherDatabase, previously discussed.

Several meteorological and climatological databasescan be easily accessed from the Internet. An ex-tensive list can be found at the Internet site of theUsenet newsgroup sci.geo.meteorology (http://www.scd.ucar.edu/dss/faq/). The High Plains Climate Cen-ter (HPCC) was formed in 1987 with the mission toincrease the use and availability of climate data in theHigh Plains region of the United States. The HPCC’sactivities include operating the Regional AutomatedWeather Monitoring system; conducting region-widesoil moisture and drought studies; developing con-nections with other climate centers; and developingcomputer software for the summarization and dis-semination of important climate-related information(http://hpccsun.unl.edu). The Register of EcologicalModels (REM) at the University of Kassel (Germany)


is a meta-database for existing mathematical modelsin ecology (http://dino.wiz.uni-kassel.de/ecobas.htm).In South Africa, daily weather data can be retrieved fora large number of stations from the Internet using theweb address (http://www.sawb.gov.za/www/climate/bull.html).

From the FAO Internet site, the Global Informationand Early Warning System on Food and Agricul-ture (GIEWS) provides extensive agricultural andclimatic information for most of the countries inAfrica, including crop production areas, crop cal-endars, and satellite images (http://www.fao.org/WAICENT/ faoinfo/ economic/giews/ english/ giewse.htm).

In Brazil, two systems have been developed bythe National Institute of Meteorology (INMET) fordissemination of meteorological and agrometeoro-logical information. They are ‘VISUAL TEMPO’and ‘VISUAL CLIMA’. The first system allowsthe user to have access, through different modali-ties (BBS or Internet), to real-time meteorologicalinformation as weather forecast or satellite im-agery. Through the second system, the user canhave access to the agrometeorological informationas published in the dekadal and monthly bulletin.The software to get access can be downloaded free(http://www.inmet.gov.br/frameset.htm).

Also in Brazil, the National Confederation of Agri-culture (CNA), in collaboration with the Council ofthe Small and Mid-Size Enterprise Supporting Ser-vice (SEBRAE), has implemented an Internet systemcalled ‘SIAGRO’ providing useful information aboutprices, weather, databases on rural legislation and cropand animal protection laws and measures. INMETsupplies weather information. The system, at present,can reach almost 2000 small communities in ruralareas supplying information at the national level. Inthe future, it will supply information at regional level(http://www.siagro.com.br/siagro/ClimaTempo.html).

The private sector has a role in the disseminationof information as witnessed by the formation of Ade-semi Communications International (formerly knownas African Communications Group) and reported inthe Wall Street Journal of 25 September 1998. It has‘55 employees and nearly 1000 wireless pay phonesin Tanzania, with many more to come. It has won asecond license, in Ghana, and a move into Sri Lanka isimminent. Just as exciting, the company is developing

wireless kiosks for Internet access and will host Webpages enabling artisans, farmers and other small-timeentrepreneurs to set up shop in a global marketplace.’

Creative use of ICTs allowed for the developmentof a new business model. ‘The model is ingeniouslysimple: communication by voice-mail only’. Thecompany sets up wireless public pay phones at dras-tically lower cost than burying copper cable. It pro-vides voice-mail accounts and pagers that announceincoming messages. The recipient leaves voice mailin return, using a phone card. The whole package ispriced for people making US$ 200 a month. ‘Virtualphone service’, the company calls it.

An example of generating, disseminating, and eval-uating agrometeorological information is described inWeiss and Kerr (1989) concerning Cercospora leafspot, CLS, (Cercospora beticola Sacc.). CLS can bea serious disease of sugar and table beets. Duringthe 1980s, CLS caused significant reductions in sugaryields in several irrigated sugar beet production areasalong the North Platte River in western Nebraska. Anadvisory system to predict if fungicide treatment wasnecessary was developed in 1986. A survey on the ef-fectiveness of these advisories was carried out afterbeet harvest in November 1987.

In the past few years, CLS has increased in severity.Recently, the local sugar company, which purchasesthe beets and processes them into sugar, bought tem-perature and relative humidity sensors and data log-gers to use in the prediction of this disease. One of themost severe outbreaks of CLS occurred in 1998, andthe advisories were well received by growers and thesugar company. A source of funding for informationof immediate economic importance is the companythat buys the commodity since it is in the company’sbest interest to have the highest quality product.

The ‘AgroExpert’ Disease Forecasting System de-veloped by Adcon Telemetry GmbH (http://www.adcon.at/Products/AgroExpert.html) is a complex sys-tem intended to reduce the amount of chemicals usedin the treatment of plant diseases. Basically, the sys-tem uses climatic data which is processed accordingto rules developed by plant protection researchers, toestablish the optimum time for chemical treatments.The system has been used in northern Europe for thepast 5 years. It employs a network of solar-poweredweather stations to monitor rainfall, humidity, temper-ature, leaf wetness and other factors. Every 15 min,


data are transmitted by radio to a PC at the basestation where it is compared with a model of the con-ditions, which favor attack, by specific diseases. Thesystem issues recommendations to spray at the opti-mum time when the chemical will be most effective.Farmers can be contacted by phone or pager, or canaccess the system directly via a PC and modem.

This technology will next be introduced in Spainand Sicily, where traditional farming is practiced onsmall (1–15 ha) farms. Farmers who are already pro-ducing for international markets, and can afford to in-vest in the system and have the most to gain from usingit, will be targeted first. Local technology relay centres(TRCs), set up by Agricultural Data Management Srlin Sicily and Inforural in Spain, each with expertisein farm sector software development, will manage theinstallation and validation of the pilot networks andwill disseminate results to stimulate farmer ‘take-up’of this technology.

Recently, the use of NOAA satellite data, to pro-duce imagery at 1.1 km resolution, which is receivedin both South Africa and Zimbabwe, has significantlyimproved prospects for providing relevant informa-tion on crop production conditions at the sub-nationallevel. More detailed data with spatial resolutions from1 km to 20 m will be available from satellites to belaunched in the near future. Access to these data, incombination with information from other sources, isessential for extension workers, researchers, NGOsand farming communities. The challenge lies inextracting the information from regional technicaldatabases and national early warning systems andadapting it into formats and styles that are relevantto the information needs of these users. The reversechallenge is to resolve the logistical and method-ological problems of retrieving, in a timely manner,farming results and experiences from these users inorder to improve the value of the information. Surveytechniques is one method to address this concern, andwill be addressed later in this paper.

Under a FAO technical assistance project, a wellequipped, decentralized system for agrometeorologi-cal and remote sensing data handling, processing andanalysis, as well as information product generationcompatible with relevant background databases, isoperational in southern Africa. Activities are carriedout by trained staff in the Regional Early WarningUnit for Food Security and the National Early Warn-

ing Systems for Food Security within SADC (SouthAfrica Development Community) countries. Facilitiesare being established in Harare, Zimbabwe, for thedirect acquisition of Meteosat data to support rainfallmonitoring across the region. The use of geographicinformation systems and the FAO-GIEWS (GlobalInformation and Early Warning System on Food andAgriclture) workstation for handling analysis of datafrom different sources on a common geographic basisexists at the national level within SADC countries.

Under separate FAO technical assistance, a Re-gional Centre for Communication for Development(RCCD) has also been established in Harare. Thecentre is capable of providing high-quality advisoryand training services to development workers (e.g.,extension agents) in SADC member countries. Ithas a professional staff that can design and delivertraining in participatory communication skills andin multi-media message design, including the use ofICTs. The next objective is now to link the agrom-eteorological data and information systems and GISservices to RCCD in order to provide informationon agricultural production conditions and food secu-rity outlook to a range of potential users, includingresearch institutions, extension services, NGOs andmembers of the farming community.

7. Evaluating the impact of information andinformation delivery systems

Evaluation of the impact of information deliverysystems can be done by surveys and by the use of focusgroups and innovative end-users. The specific surveyinstrument or the techniques for gathering informationmay differ from community to community, but the goalis essentially the same: to evaluate the impact of theinformation and the information delivery system andto have a quantifiable basis to improve the system.

It is necessary to explain the purpose and impor-tance of the survey or information gathering technique.This justification may be in the form of a cover letterthat accompanies the survey. Or it can be the intro-ductory remarks at a meeting of a focus group.

Confidentiality must be stressed in any surveyproject. Numbered surveys are associated with namesand addresses (mailing lists) of potential respondentsin order to keep track of who responded to the first


mailing of a survey and to target the names for a sec-ond mailing, if necessary. Once the study, based onsurvey results, has been completed all materials relat-ing to the mailing lists and surveys must be destroyed.

Survey results must be communicated to thenecessary agencies that developed and funded the sur-vey for evaluation of current programs and for futureplanning. The results may be published in the appro-priate scientific journals and shared with colleaguesat scientific meetings. The initial respondents shouldnot be forgotten in this communication process. Byproviding a summary of the survey results to the re-spondents, not only are the survey developers givingfeedback to those who participated in the survey, theyare developing a marketing tool. The feedback maybe used to develop interest, support, and future usersof the information dissemination system.

Some of the questions mentioned here may not berelevant to a region, other questions may have to beadded, and some questions may have to be deleted.The following questions serve as an example.• Location: Answers to this question give an idea of

the demographics of the end user.• For pest surveys: Occurrence of pest in the last sev-

eral years. Response to this question gives an ideaof locations where pest has occurred and if theselocations fit into what ‘experts’ have predicted ornoted. Ability to identify pest. Provide several de-scriptors. If the end user can’t identify pest or pestsymptoms, then this is an area of further educationthat fits into the training needs of end users.

• Has the respondent used the information deliverysystem?

• Has the information been helpful? At this pointin the survey for the remaining relevant questions,there should be several options such as ‘stronglyagree’, ‘agree’, ‘neutral’, ‘disagree’, and ‘stronglydisagree’.If yes, how has the information been helpful? (List

several key components of the information deliverysystem; e.g., timeliness, caused the end user to thinkdifferently about the situation, easy to understand thedifferent recommendations and their consequences,easy to implement recommendations).

If no, how has the information not been helpful?(List same options as mentioned earlier.) Possibly theend user knew how to deal with the management sit-uation and the information was not needed.

• What feature did the user like most about theinformation delivery system? List several options,respondent can also fill in blanks.

• What feature did the user like least about theinformation delivery system? List several options,respondent can also fill in blanks.

• What improvements would the user like to see inthe system? Leave blank spaces to fill in responses.Responses may indicate areas for future research.

• Would the user be willing to pay for this informa-tion?

If no, what improvements in the system wouldmake users change their minds? Leave blank spacesto fill in responses. Again, responses may indicateareas for future research or changes in dissemina-tion.

• How should the cost of this information deliverysystem be supported? Rate percentage of cost tobe borne by each entity. Choices should includegovernment, university, farmers association, privateindustry, and others sources.

8. Some critical questions

Continued improvement in communicating agrom-eteorological information to farming communities re-quires addressing the following critical questions.

How can diverse types of agrometeorological databe integrated into useful information that respondsto the often-dissimilar application needs of farmingcommunities?

Agrometeorologists change data into informationthrough whatever tools are available, currently throughthe use of computer-based technologies. In addressingthe first question, we must begin by considering thetraining of agrometeorologists. Training in this areashould include an appreciation of the complex inter-actions of biotic and abiotic factors on plants and ani-mals. This addition to the traditional training of agrom-eteorologists is necessary to provide a larger perspec-tive to the transformation of data into useful informa-tion.

As plants and animals go through their life cycles,they encounter different meteorological conditions andare subjected to different kinds of biotic and abioticstresses. While the same meteorological variables aremeasured (e.g., temperature, solar radiation, precipi-


tation, etc.) throughout these life cycles, their applica-tions to the different potential problems change. It isnecessary to know when these different stress relatedproblems can occur and to evaluate these situationsbased on current and predicted meteorological condi-tions. For example, plant now or wait a week in an-ticipation of rain, will it be necessary to irrigate in thenext week, should a preventative measure against pestsbe made, the impact on quality of harvesting now orwaiting a week, and contracting grain before harvest.

What types of information are needed by diversegroups of end-users and, given their different farm-ing, socio-economic and cultural systems, which arethe appropriate communication technologies to reachthem?

The information needed for diverse groups ofend-users growing the same crops or raising the sameanimals is basically the same. The differences arisefrom the human and financial resources availableto implement this information and the methods ofinformation dissemination. These differences mustbe considered in designing any information dis-semination system. For those who have access, theInternet is a major source of data and informationon all aspects of agricultural production. Data andinformation on the Internet are available from uni-versities, industry, and advisory services, sometimesfor a fee. At present, many national and sub-nationalagrometeorological services use the Internet to dis-seminate very useful information, e.g., real-timecrop information such as water requirements, ad-vice on pest disease, etc. Another example of usefulagrometeorological information provided throughInternet can be found at the Provincial Agromete-orological Centre of San Michele all’Adige (Italy):http://www.ismaa.it/html/ita/meteo/agri.html.

As we have seen, however, the availability of In-ternet hosts is extremely limited in many developingcountries. ICTs offer great hope for innovative dissem-ination of agrometeorological information to farm-ing communities. In most cases at present, however,reaching the majority of farmers in developing coun-tries with agrometeorological information, especiallysmall-scale farmers in remote areas, will require spe-cial targeted communication efforts that utilize localmedia such as radio.

Given diminishing public support for agriculturaladvisory services, what alternatives exist for the com-

munication of agrometeorological information and un-der which circumstances can it be provided on a feebasis?

First, as much as possible, agricultural advisoryservices must accurately document the added valueand impacts of these services. This should be done asa proactive rather than a reactive position. The doc-umentation should include the monetary value of theinformation, if used properly. Just because accurate in-formation is disseminated doesn’t mean that it will beused or used properly. Included in this documentationshould be detailed descriptions on changes in positivebehaviour (impacts), e.g., higher yields due to im-proved forecasts of the beginning of the rainy season,uses of techniques that minimize chemical applica-tions and ground water pollution, or limit soil erosion,based on disseminated information. This documenta-tion may positively influence government funding.

The information must be prepared by agrometeo-rologists in a way that the majority of users will easilyunderstand it. Then it can be adapted and sent to keycommunication outlets such as radio, television, news-papers, bulletins, specialized information networksand web sites for broad-scale as well as targeted dis-semination. From the perspective of these media out-lets, the information must also have a value, a differentvalue than intended for the end users. Various types ofusers may buy the products advertised by the differentmedia outlets or subscribe to information services.

In this era of tighter government funding for agricul-ture, how can agrometeorological based informationbe effectively disseminated? As noted previously, thisinformation must first have an added value; if it has avalue, then there will be a demand for it. For exam-ple, the German Weather Service (DWD) has formedpublic/private partnerships to provide different detailsof agrometeorological information at different pricesto a wide range of users, (Kruger and Dommermuth,1999). In this process, the Department of Agrome-teorology was reorganized into the Business Unit ofAgriculture.

Munthali (1999) addressed some of these issuesin a survey of users of agrometeorological informa-tion in Malawi. Commercial farms, which have welltrained staff and access to ICT, were candidates forspecialized information on a fee basis. At the otherend of the farming spectrum, small holders would re-quire the interaction of extension personnel to obtain


the necessary information. In addressing diminishingpublic support for meteorological services, it was sug-gested that information be treated as a commodity,and marketing plans be developed for user fees, wherepossible.

The earlier mentioned discussion has tacitly as-sumed that a fee for information implies an exchangeof money for a service. This situation may not beappropriate in many developing nations. Fee can alsobe thought of as an exchange of one service for an-other service, such as collecting weather or crop dataand transmitting these data in a timely fashion to anagrometeorological center. In return, the provider ofthese data could receive management recommenda-tions not normally provided in a general agriculturalweather forecast. For a system based on an exchangeof one service for another service to be successful,both sides have to receive value for their service. Forthe provider of data, it may mean higher yields, whilefor the agrometeorological center, it may mean datathat are needed to help form a governmental policy.

FRIENDS (Farming and Rural Information Exper-tise and Dissemination Services) is an EU-fundedproject that provides remote Greek agricultural com-munities with information and expertise, Thompson(1998). One group of users is ‘access point users’from the Greek Union of Young Farmers (UYF). Theuse of access points, which are workstations mannedby site managers, helps to overcome the problem thatthe majority of Greek farmers lack basic computerskills. Site managers help visiting farmers get theinformation they want to retrieve from the system.

FRIENDS collects, maintains and distributes awide variety of agricultural information services.These services range from national weather, EU leg-islation and daily agricultural market prices in Athensto online-consulting and transactions. FRIENDS actsas the information channeler — aggregating, filtering,packaging then disseminating it to the farmers andother users.

A key success factor has been ‘one-stop shopping’for agricultural information — FRIENDS aggregatesagricultural information from many sources and savesusers valuable time by aggregating these sources inone format (WebPages) and presenting them in oneplace (the FRIENDS website). The most basic valueFRIENDS adds to its information services is intelli-gently filtering the information it provides.

What are the training needs of end-users and of thevarious intermediaries that provide them with advisoryservices?

Before the training needs of end-users and interme-diaries can be addressed, the question of motivationfor users to access and use the information should bediscussed. While altruistic values to use the informa-tion can be a source of motivation, e.g., sustainabilityof the environment, they will probably have minimalsuccess. Farming is a long-term business, and like anybusiness, its practitioners want to be successful — mo-tivation should be based on sustainable profitability.Given this perspective, what information is needed?What information is already available? What infor-mation needs to be provided? Of the information thatneeds to be provided, what information is of primaryimportance, secondary importance, etc? Addressingthese questions requires assessments of the informa-tion needs and resources of specific groups within thediverse user community. In order to facilitate the com-munication of information to the user community, so-cial scientists should interact with agrometeorologiststo provide a structure for the information that is suitedto the target audience.

9. Conclusions

The Internet will play a major role in agrometeo-rological information either through direct or indirectaccess. Richardson (1997) summarizes many websites that document examples of the application ofagrometeorological information. Also, Maracchi andDe Vincenzi (1995) lists important sites on the Inter-net where is possible to obtain meteorological dataand information for agrometeorology. In developingcountries, Multi-purpose Community Telecentres willbe the focal point for many types of information,some of which will come from the Internet. TheseTelecentres can be the source of agrometeorologicalinformation that can be disseminated over rural radiostations in local languages. Remote sensing data andgeographic information systems are being used insouthern Africa to provide information on agriculturalproduction conditions and food security.

While information can be disseminated, is it beingused? A methodology was presented to evaluate theimpact of information. This methodology can be ap-


plied as a written survey or through interviews. Ev-idence is presented that people would be willing topay a fee for agrometeorological information if it hasvalue, in both the developed and developing world. Insome cases, this fee may take the form of an exchangeof important information.

Critical questions have been raised but clear re-sponses are linked mainly to the scientific, technologi-cal, and social developments that will take place in the21st Century. It is clear that information and commu-nication technologies will improve in the future as willaccessibility. What will limit the generation and dis-semination of agrometeorological information in thefuture is the same that limits it today: the interactionof people, from scientist to extension worker, in thecontinuum from basic understanding to practical ap-plications. Thus, to prepare for the future now, wehave to better integrate the human capital available atall levels of organization. Specifically, we recommendthat information and communication technologies bea component of the training of agrometeorologists inorder to provide the best possible advice to farmers.

References

Doriaswamy, P.C., Pasteris, P.A., Jones, K.C., Motha, R.P.,Nejedlik, P., 2000. Techniques and methods for collection,distribution and analysis of agrometeorological data Agric. For.Meteorol. 103, 83–97.

Gommes, R., 1997. FAO’s experience in the provision ofagrometeorological information to the user community. In:Sivakumar, M.V.K., De, U.S., Sinha Ray, K.C., Rajeevan, M.(Eds.), User Requirements for Agrometeorological Services.Proceedings of an International Workshop, Pune, India,10–14 November 1997. India Meteorological Department,Shivajinagar, Pune, India, pp. 53–77.

Hamelink, C.J., 1997. New Information Technologies, SocialDevelopment and Cultural Change. United Nations ResearchInstitute for Social Development (UNRISD), Geneva,Switzerland, Discussion paper No. 86.

Hoogenboom, G., Hammer, G., 2000. Contribution of agrometeoro-logy to the simulation of crop production and its applications.Agric. For. Meteorol. 103, 137–157.

Hubbard, K.G., Rosenberg, N.J., Nielsen, D.C., 1983. Automatedweather data network for agriculture. J. Water ResourcesPlanning Manage. 109, 213–222.

Jensen, M., 1998. Internet connectivity in Africa — a status report.Available at http://demirrge.wn.apc.org/africa/afsra.htm.

Jensen, M., Richardson, D., 1998. Wireless weaves tolessen the gaps in rural telecommunication coverage inAfrica. In Richardson, D., Paisley, L. (Eds.), The FirstMile of Connectivity: Advancing Telecommunications forRural Development through a Participatory CommunicationApproach. FAO, Rome, Italy.

Kruger, J., Dommermuth, H., 1999. Operational aspects ofagrometeorology. CAgM Report 77, Contributions frommembers on operational applications in agrometeorology for theInternational Workshop ‘Agrometeorology in the 21st Century,Needs and Perspectives’, Accra, February 1999. WMO,Geneva.

Mannisto, L., Kelley, T., Petrazzini, B., 1998. Internet andglobal information structure in Africa. In: Ras-Work, T. (Ed.),Telecoms in Africa — Tam Tam to Internet. Mafube Publishing,Johannesburg, South Africa.

Maracchi, G., Perarnaud, V., Kleschenko, A.D., 2000. Applicationsof geographical information systems and remote sensing inagrometeorology. Agric. For. Meteorol. 103, 119–136.

Maracchi, G., De Vincenzi, M., 1995. Researching meteo data foragrometeorology on the Internet. CNR-IATA, Florence, Italy.

Munthali, C.K., 1999. The communication of agrometeorologicalinformation. CAgM Report 77, Contributions from memberson operational applications in agrometeorology for theInternational Workshop ‘Agrometeorology in the 21st Century,Needs and Perspectives’, Accra, February 1999. WMO,Geneva.

Richardson, D., 1997. The Internet and Rural and AgriculturalDevelopment. FAO, Rome, Italy.

Stewart J.I., 1998. WARF: World Hunger Alleviation throughResponse Farming. Davis, California.

Talero, E., Gaudette, P.H., 1996. Harnessing Information forDevelopment: A Proposal for a World Bank Group Strategy.The World Bank, Washington, DC, USA.

Thompson M., 1998. Friends in Greece: Commercializingand Agricultural Information System, in SD Dimensions,WebPages of the FAO Sustainable Development Department.http://www.fao.org/WAICENT/FAOINFO/SUSTDEV.

Weiss, A., Kerr, E.D., 1989. Evaluating the use of pest managementinformation by growers: an example using cercospora leaf spotof sugar beet. Appl. Agric. Res. 4, 168–172.

WMO, 1992. La radio rurale et la diffusion des infor-mations agrometeorologiques. Proceedings of an InternationalWorkshop, Geneva, Switzerland.

Documents

Communicating agrometeorological information to farming ... and Forest Meteorology 2000... · Communicating agrometeorological information to farming communities A. Weissa;, L. Van