Issue link: http://read.dmtmag.com/i/55888
pixel. The Kernel radius also depends on the confi- dence of the hazards, in accordance with their type. For example, mine fields have a higher confidence than dan- gerous areas, so the mine-field Kernel radius is lower. The Kernel method has a lot of advantages, includ- ing the ability to process heterogeneous data and provide short processing times—a few minutes of cal- culation for 5,000 hazards, upon an area of 1 million square kilometers with a 20-kilometer Kernel search radius and a 200-meter output pixel size. The density raster in Figure 1 was obtained from Afghan data. The interpolation results in a smoothed density map (grid) displaying the hazard contamina- tion, while not showing all details. Red values cor- respond to a high probability of contamination and yellow values to a low probability. There are no known hazards in white areas. A tool was developed within ArcGIS Engine to auto- mate the process of creating and representing rasters from data in MySQL databases. The tool currently is being interfaced with IMSMANG . When countries feel comfortable sharing them, the density rasters will be published on a GIS server. Maps and Web services will be provided to key users from GICHD to partners and eventually the entire Internet community. The sensitivity and confidentiality of the data will prevent them from being shown at scales higher than 1:2 million. Exposed Populations and Operational Difficulty Showing hazards and field activity doesn't provide a complete vision of humanitarian mine action, so it would be interesting to combine and compare those rasters with global datasets to determine in which areas populations are most exposed and better model the operational difficulties of clearance. An ongoing research process explores the possibility of exploiting at a worldwide scale—and with a resolu- tion higher than a few square kilometers—datasets for population, infrastructure, hydrology, climate, eleva- tion, slope, soils, etc. Encouraging tests have been made, and first results already can be shown (see Figures 2 and 3). Other datasets are being evaluated for their potential use. "The Mine Action Toolbar is a central resource for key spatial functions used in mine action that dramati- cally simplifies access to the most common geospatial needs in the community," notes Noah Klemm, director, International and Homeland Security Programs, FGM Inc. "The toolbar exposes field users with limited ArcGIS experience to the most-useful GIS functions in the ArcGIS desktop products." Customized Tools More countries are using ArcGIS Desktop, but the users have limited training, so some customized, user-friendly ArcMap tools were developed, including the following: Figure 3. Combining and comparing hazards with global datasets can provide a more complete vision of mine action. Soil characteristics such as cation-exchange capacity, texture, bulk density and available water-storage capacity are being investigated for their potential influence on the operational difficulty of de-mining. 28 GEO W ORLD / FEBRUAR Y 2O12 in the same toolbar. MySQL databases, creation of closed polygons from a list agement, vector and raster clipping, etc. (see Figure 4). Figure 4. The Mine Action Toolbar provides standard tools for printing and designing layouts as well as customized tools in direct connection with user needs. FAO/IIASA/ISRIC/ISSCAS/JRC 2009. HARMONIZED WORLD SOIL DATABASE (VERSION 1.1) International GIS