Environmental Sustainability in Practice

Spatial Autocorrelation

The concept of spatial autocorrelation (also known as spatial association or spatial dependence) is related to Tobler’s First Law of Geography which states that “everything is related to everything else, but near things are more related than distant things”. Simply put, spatial autocorrelation is the degree to which characteristics at one location are similar (or dissimilar) to those at nearby locations. Spatial autocorrelation provides important information about underlying processes that result in certain landscape patterns.

There are three general possibilities relating to spatial autocorrelation:
  • Positive autocorrelation: nearby locations are likely to be similar to one another.
  • Negative autocorrelation: observations from nearby locations are likely to be different from one another.
  • Zero autocorrelation: no spatial effect is discernable, and observations seem to vary randomly through space.
One of the oldest and most popular indicators of spatial autocorrelation is known as Moran’s index. The values in Moran’s index range between -1 and +1.
  • Positive values = high spatial autocorrelation (clustered)
  • Negative values = low spatial autocorrelation (dispersed)
  • 0 = no spatial autocorrelation (random) 

This page has paths: