Ordinary kriging trend removal

644
3
11-16-2021 01:25 AM
JCRogier
New Contributor

Hi, 

 

I was wondering something about my understanding of Ordinary kriging and the implementation in ArcGIS Pro. As I have understood the concept, the trend is modelled in the same process as with kriging the residuals. The same holds for Universal Kriging and Kriging with external drift, which would also model the trend in the same process (regression kriging in contrast would require to model the drift externally). If so, this would mean that I would not expect to see the possibility of trend removal in the ArcGIS application. However, this option is provided. So now  I think that I might be missing a key concept. Could someone explain why the option for trend removal is given ? 

 

Best,

JCRogier

0 Kudos
3 Replies
DanPatterson
MVP Esteemed Contributor

did you see some of the help topics on this subject? For example

Understanding transformations and trends—ArcGIS Pro | Documentation


... sort of retired...
0 Kudos
JCRogier
New Contributor

Thank you for your response Dan! 

Yes, I did see this documentation. However, I still can't grasp the idea completely. If I understand the information in the link you provided correctly, the trend removal is only applicably on the second variable (i.e., when using cokriging)?

 

0 Kudos
curtvprice
MVP Esteemed Contributor

According to the first table on that help page, ordinary kriging does support trend removal. However, if you have known trends or other issues, I recommend looking into EBK which is designed to automatically do a lot of the work required for a valid kriging analysis. It also provides a lot of diagnostics the other methods do not.

0 Kudos