Geographic information systems are becoming increasingly widespread as a tool for guiding decision-making in relief interventions. Geographic information systems (GIS) are computer-generated maps, built up with layers of information. The key components of a basic geographic database are land use, infrastructure and topography, on top of which information on other features, such as population movements, settlement patterns and accessibility, may also be layered.
The first step in approaching a GIS solution is to consider what information is needed. In food security, the starting-point is a geographical map of the region, indicating such features as slope, vegetation, water sources and international boundaries. Data relating to land use, rainfall, soil type, cultivation, settlements/population patterns and transport routes are then added. On the top layer, more precise information relates to such features as conflict areas or minefields. All of this can be called up as a specific image, or superimposed so that the resultant map reveals areas of high risk.Figure 1 and Figure 2 show a study of accessible ground water sources and land use which would be superimposed digitally.
Land use and cultivation patterns form the initial base of information necessary for devising a food security programme. However, food insecurity is a complex issue, and the critical aspects may not be the most readily apparent. Under-nutrition, for example, may be a problem of production or of access. Low productivity may derive from weather conditions, inadequate supplies for crops or livestock or policy decisions. Decreased income and lack of alternative employment may be a seasonal disturbance or a degradation of the environment. GIS mapping enables many other layers of data to be included for a more sophisticated analysis of the interrelationship of issues such as these.
GIS is particularly effective at revealing relationships between climate, meteorological conditions, water balance and crop forecasts. Data on early warning systems can be found on many internet sites, including the FAOs Global Information and Early Warning Systems (GIEWS) at www.fao.org/giews, and USAIDs Famine Early Warning System (FEWS) at www.fews.net. The European Commissions Joint Research Centre provides research on GIS applications in humanitarian relief, including the work of EURISY (www.eurisy.assoc.fr).
Minefields pose serious problems in more than 65 countries. In Afghanistan, for instance, an estimated 25% of agricultural land and 66% of grazing land is mined. Identifying fields so that they can be clearly marked, fenced off or cleared is thus another important use for GIS mapping, especially when linked to land reclamation. Important questions here may include which areas or sites have strategic significance?; where has land been disturbed?; where do factional territorial divisions lie?. The purpose of laying mines may also be important: is it defensive, for sabotage, or to terrorise a civilian population?
High-resolution satellite imagery from the internet (IKONOS, IRS) provides basic topological and statistical data for an initial framework for analysis. Key visual clues from satellite images include linear ploughshare patterns across patches of terrain, trenches, artificial embankments, fencing (especially along borders), evidence of military activity and seasonal variations in land use. However, satellite imagery and remote sensing tend to overestimate the extent of mined areas. Alternative survey methods must then be used to refine initial assessments. For studies on a smaller or very local scale, photos from aerial surveys are an alternative source. New technologies including radar and 3-D imaging for local surveying are being tested and refined. Ground surveys around the perimeter of the suspected minefield can then clarify or correct the maps compiled.
A mine action information system would need to include information on the minefields themselves (confirmed and suspected) and the local population, and thematic maps including socio-economic data and resources, with precise locations using the global positioning system (GPS). As minefields are of interest to the military, changes throughout the course of a conflict are often monitored and recorded, including reports from government defence agencies, refugees and other parties.
When addressing food security in regions where mines are a risk factor, a combination of data can be digitised and compared to coordinate demining programmes with food security plans. This gives a broad view of a regions status, and helps to fine-tune and prioritise interventions. The more data that is added, the greater potential use the final map will have, and the more thorough the decision-making process. In Figure 3, if this land use map were applied to demining, the darkly shaded patches would indicate minefield areas of greater density. The lighter shaded areas therefore would indicate the first choice for land clearance if the issue is ease and rapidity. If other data, for instance settlements, water sources and roads, were added, the field of choices would gradually narrow to the areas where clearing mines would have the greatest benefit in terms of access to productive land.
Data collection and implementation
The multiplicity of auxiliary data sources and the number of agencies working in the field make coordination a key obstacle to the effective use of GIS in food security and demining. Experience in Mozambique and Afghanistan has indicated that much information derived form local surveys is incomplete, inaccurate and not based on compatible standards. In conflict regions, it is also difficult to gain uniform access for thorough and timely mapping.
Since humanitarian interventions involve multi-sectoral assistance, it is possible to divide data collection up according to sectors, and assign these surveys to agencies primarily concerned with these sectors. A degree of overlap provides a control measure for assessing the accuracy of data inputs. Thus, infrastructure surveys include and overlap with water/sanitation surveys, and agricultural surveys include infrastructure and irrigation systems. Once an effort is made to corroborate satellite maps with data on the ground, the human factor can either introduce distortions or enhance precision. Developing a coordinated methodology is essential in accurate data collection for mapping.
Because accurate GIS mapping can require a vast array of data, often from official or technical sources not readily available in developing countries, there are concerns that it is not suitable for rapid assessments or for participatory methods. In fact, this is not necessarily so: maps available in the public domain on the internet provide much general information through aerial and satellite imagery. In matters relating to particular issues, like risk and resource mapping for food security or demining, local surveys are essential. These introduce potential inaccuracy due to variances in methodology, access, thoroughness and professional standards. But with adequate data input, GIS can reveal relationships not easily discerned through other survey methods, such as the link between the probability of crop failure and a communitys risk of food shortages.
Food security and demining are both linked to issues of land use, and are therefore ideal fields for the use of geographical information systems. Identifying land suitable for cultivation is one of the principal benefits of applying GIS in demining strategies. ESRI produces some of the leading software in GIS applications for resource conservation, urban planning, public health and disaster management. The coupling of satellite maps with local surveys provides an extremely useful framework for crafting a basic GIS of a site for projected humanitarian intervention. As more raw data are filled in to create new layers, the GIS model can be adapted and updated to provide information across many sectors, leading not only to more precise analysis, but also to better-coordinated programmes.
References and further reading
Getting To Know ArcView GIS (ESRI Press, 1999).
Laura Lang, Managing Natural Resources with GIS (ESRI Press, 1998).
UN Mine Action Service, The Application of Technology to Mine Action, MAS/IACG/1(98), 4 March 1998.
Charles Chopak, Food Security Early Warning: An Overview of Monitoring and Reporting, USAID, June 2000.
Reassessing the Impact of Humanitarian Mine Action, PRIO, 2000.
Sustainable Humanitarian Demining: Trends, Techniques and Technologies, Workshop Proceedings, Humanitarian Demining Information Center, Zagreb, Croatia, September 1997.
The Journal of Mine Action, www.maic.jmu.edu.
Human Rights Watch/International Campaign to Ban Landmines, Landmine Monitor Report 2001.