Alerte rapide

Early Warning

Four key points to any successful early warning system

An effective system should have at least four key components:

a network of regular surveillance
rapid transmission of data from the field to decision-makers who should be able to access it easily
geo-referenced field data (i.e. data collected with GPS coordinates) should be fully managed and analyzed within a geographic information system (GIS)
summaries, warnings and other outputs should be simple, clear and well-targeted for those who need early warning. Continue for more details on each of these components

The Desert Locust early warning system

The Locust Group at FAO Headquarters in Rome operates an early warning system to alert countries about the development of Desert Locust plagues. During calm periods, Desert Locust infestations are usually present somewhere within about 16 million square km of desert in 25 countries between West Africa and India. During plagues, the number of countries and the size of the area that can be potentially affected doubles, representing about 20% of the Earth's land mass

A network of linked national locust centres

One of the keys to effective monitoring of a mobile pest is to encourage affected countries to communicate and work together, to exchange information and to share experiences in early warning. The Desert Locust Information Service which operates the locust early warning system at FAO Headquarters manages an Internet-based group of national locust information officers. This is a simple mechanism for these officers to keep in contact with each other, to share problems and solutions, and to keep informed about the latest developments that affect their work.

The flow of data in the Desert Locust early warning system

Locust-affected countries are responsible for monitoring the Desert Locust situation in their own country. Each country has a Locust Control Centre within their Ministry of Agriculture. Well-trained staffs from the Centre undertake surveys in four-wheeled drive vehicles to look for green vegetation in the desert and check to see if locusts are present. The survey results are collated at the Centre by a designated Locust Information Officer who manages and analyzes the data which is used for planning and forecasting. The data is also sent to FAO DLIS for further analysis and forecasting.

Getting complete data from remote areas in real time

Until recently, one of the major hurdles in effective early warning was the collection and recording of accurate and complete data in the field and its subsequent transmission to the National Locust Centre in near real time. If there are delays in receiving this data, then its value declines and well-informed decisions and plans cannot be made. FAO DLIS developed, in collaboration with Novacom (France), a handheld device for field officers to enter and send geo-referenced data in real time. This device is called eLocust2. The field officer enters and saves the data into a rugged handheld device which automatically determines the coordinates of the location of the survey or control operation. With a press of a button, the officer sends this data via satellite to the National Locust Centre where it is received as an email attachment, downloaded, decoded and imported into a geographic information system (GIS). The GIS is used to manage and analyze the data.

Data must be easily accessible by decision-makers

The eLocust2 system offers the added advantage that locust information officers and national locust directors can easily access eLocust2 via the Internet at any time, day or night. In this way, staff in the National Centre can see a map of where the teams are in the field, how well they are working, and what there are finding. Gaps in survey coverage can be identified easily and quickly corrected. It is not necessary for the Directors to be very computer literate because the system is simple and easy to use.

RAMSES: a geographic information system (GIS) used at National Locust Centres

FAO, in collaboration with NRI (UK), has developed a GIS for national locust information officers to use for the management and analysis of field results and of locust and environmental data. eLocust2 data from the field are automatically imported into RAMSES where it is checked and corrected if necessary. The data can be compared with earlier or historical data, with rainfall reports, meteorological information and satellite imagery. In this way, the current situation can be assessed accurately and quickly. Based on this assessment, well-informed decisions can be taken regarding survey and control operations, and the need for additional resources. Data from RAMSES are exported and sent to FAO DLIS for further analysis, forecasting and early warning.

SWARMS: a sophisticated GIS used operationally at FAO HQ for locust early warning

FAO, in collaboration with NRI (UK) and the University of Edinburgh (UK), has developed a GIS for the management and analysis of locust and environmental data. SWARMS has been used operationally for locust early warning since 1996 by the Locust Forecasting Officer in DLIS at FAO Headquarters. SWARMS was one of the first GIS developed for operational monitoring. Its databases contain historical data that covers nearly 100 years. SWARMS is under constant revision and updating in order to take advantage of the latest technological developments.

Additional tools are used by FAO DLIS for analysis and forecasting

FAO DLIS relies on several specialized tools to supplement its analysis and forecasts. For example, a Trajectory Model (developed by Meteo Consult) estimates the source and destination of swarm migrations forward and backward in time. Rainfall estimates and MODIS satellite imagery provided by IRI at Columbia University (New York, USA) are used to try to understand where it has rain and where vegetation is green in the desert. Seasonal predictions of temperature and rainfall six months in advance are analyzed. Another model is used that estimates the developmental times of locust eggs and hoppers. The use of these tools in combination with the GIS allows the best possible analysis from which forecasts and early warning can be issued.

Well-targeted outputs must be simple and clear for early warning

The best data collection, transmission, management and analysis will be useless unless they can be transformed into clear, simple products that are easily understood by decision makers and others who must react to early warnings. DLIS produces a situation summary and 6-week forecast every month that is distributed by email and Internet. The bulletin is written in a way which provides an increasing amount of details, starting with a single paragraph, then the first page and followed by individual country summaries and forecasts. The monthly bulletin is supplemented by alerts, warnings and updates. The latest information about the locust situation can be found on regularly updated Locust Watch web pages. A simple colour-coded system has been developed to indicate the threat level. This system is used for all of the various outputs produced by FAO DLIS.

Early warning information should be available in different ways and formats

In addition to the bulletins, updates, alerts and warnings issued by FAO DLIS, anyone who has Internet access can create a custom map that shows the latest locust situation or one in the past. This is useful for presentations and for those that prefer information on maps rather than in bulletins. The system is called DL MAPPER and it is on the Locust Watch web site.

Early warning information should be available in different ways and formats
In addition to the bulletins, updates, alerts and warnings issued by FAO DLIS, anyone who has Internet access can create a custom map that shows the latest locust situation or one in the past. This is useful for presentations and for those that prefer information on maps rather than in bulletins. The system is called DL MAPPER and it is on the Locust Watch web site.

Effective and reliable early warning must be sustainable and relies on everyone's cooperation

In addition to the four key components of an early warning system, there are several other aspects that should be considered within the context of an early warning program at the national level: (1) individuals should be energetic, curious and dedicated, (2) field teams need to be well trained in collecting, recording and transmitting data, (3) they also need to be well equipped and be given the tools to do so (GPS, vehicles), (4) financial support must be provided on a regular basis by the Government, (5) early warning has to develop into a routine activity, (6) everyone involved in early warning needs to receive feedback if they are expected to continue their efforts and maintain interest, and (7) it is the affected country that owns its data.

Back to Activities