To the list of courses || RASA2018 || To the theme || Estonian

Exercise 3947. Points 3, theme: Local Statistics (categorical variable)

Open exercise
Landscape diversity can be measured by the number of different land cover categories in a fixed size area or in certain radius around a location.
  1. Is landscape diversity according to the number of Corine land cover categories larger in drumlin areas or in the glacial accumulated mounds in Estonia? Present the mean number of Corine land cover categories in 1 km radius in both above mentioned landscape types.
  2. Is this difference in mean values statistically significant? Present the number of locations used and the significance p value.
  3. How to explain the difference?


How to craeate random locations to a masked in area is described in the theme Creating point patterns.
The number of random locations should be at least 1000 to get reliable estimates from a sample.
When the random locations are generated, copy these to Excel and add a column with unique ID values. The SDC demands an ID vcalue for each location.
  • To calculate local statistics in different radii select Point data → Local statistics.
  • Corine 2006 land cover data are freely usable in the SDC. Students of this subject can log in and use more data layers.
  • Open Difference tests when the number of land cover categories is calculated.
  • Select U test and t test, two samples and insert the number of categories around locations in drumlin areas into one cell and the number of categories around locations in the glacial accumulated mounds to the other cell.
  • Find the mean values and significance (p) values from the results.
Log in to send your results and to see the expected answer and responses from other students.