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

Spatial Autocorrelation

In addition to autocorrelation exercises, some exercises on similarity coefficients are included into this topic. Similarities and distances are the typical source data for Mantel test and Mantel correlogram, which is characterizing autocorrelation.
Not answered
The answer needs to be updated
ID P Question or exercise
3966 3 The attached file contains data of 100 forest sites observed in Saare County in 2002: coordinates of observation sites, coverage of tree species (from kuusk to muu), stand total coverage (katvus), land cover type (maakate) and reflectance values measured by the Landsat ETM+ (TM1 ... TM8).
  1. The stand composition of closer stands is more similar. Up to which distance the simlarity holds?
  2. How strong and how significant is the Mantel correlation between distance and stand composition within radius 500 m and within 1000 m?
  3. Up to which distance is the correlation statistically significant at p < 0.05?
3967 2 The attached file contains precipitation in May measured in the Baltic weather observation stations in different years.
  1. Which p< 0.01 level spatio-temporal patterns are evident in these data?
  2. Add the 3D autocorrelogram.
  3. Would you use this autocorrelogram when predicting precipitation amount for May 2028 in Tallinn knowing only the corresponding value for Tartu in 2017? If yes, then explain how.
3988 3 The attached file contains long term mean precipitation in autumn (September, October and November).
  1. Calculate autocorrelogram from the total autumn precipitation amount depicting spatial relationship in 10 km distance zones up to distance 500 km and add the autocorrelogramm to the answer.
  2. At which distances is the autocorrelation statistically significant (p < 0.01)?
  3. Add the correlogram and some thoughts how to climatically interpret the relationship.