by Mark MacAllister
June 21, 2002
Page 3 : Basic data analysis
The first step in analyzing incoming elephant location data is to convert the satellite reports to a useable format. We use a Java-based program to convert all data to Microsoft Excel worksheets. From there, we can list, edit, and sort location data. Collar number, date, time, latitude, longitude, and LC data are held in these spreadsheets.
We then build several GIS maps that provide a general outline of the location data in relation to Cameroon's national borders, neighboring nations, local villages, national parks, and so on. The goal at this stage is to understand the animal's basic range--that is, how much territory the animal may have covered--without adjusting for the accuracy (or inaccuracy) of the location data. Look at the Media Gallery to see an example of an initial GIS map.
Next, we split the data into groups based on each LC figure. In other words, we build a map showing all locations with an LC equal to "3", another map of all locations with an LC equal to "2", and so on. We can then group those maps and begin to discern a more accurate view of the elephant's actual locations. The map below shows Desiré's locations, but only in those instances where the LC lies between "0" and "3".
Several differences between the two maps are immediately apparent. Obviously, the biggest difference is that the second map has far fewer points. What might that indicate? Secondly, you will notice that the second map does not show Desiré entering the Central African Republic to the east; likewise, he does not appear to be straying as far to the west and southwest as indicated in the first map. What is important about those findings? What other differences between the two maps do you see?
Next Page : Determining homeranges: Overview
Pages: 1, 2, 3, 4, 5, 6, 7, 8
View printer-friendly version
