Cluster Calculations

This tool can be used if you have a big 1D geophysical model dataset and you need to group all information into categories/clusters. The calculation is iterative where the sum of distances is minimized between datapoints and their corresponding centroid that later forms the basis of the clusters. The number of clusters need to be predefined to make the calculation where each datapoints continually will be assigned to its closest centroid. In this case the closest centroid may consist of commonalities away from other clusters.

Requirements:

• 1D geophysical models

1. Open profile and get a data overview in order to evaluate the number of clusters needed.

• A layered profile showing 1D PACES data.

2. “Themes” menu –> “Cluster 1D Models…”.

• In “Models” different 1D geophysical data can be chosen. Here for PACES data.
• In previous profile PACES data was assesed according to the number of clusters. 10 clusters were chosen for all layers.

3. The resultant cluster information is added to the topplot.

• The volume of clusters can be observed in the range of 4-6 clusters for the PACES data.

1. Repeat step 1.2. and pick “Region”.

• The resultant clusters can be observed on the map within the region and the legend in the layer manager .