GEOSTATS: Assessing Aggregation Algorithms
- I am relatively new to the field of geo-statistics and would like to
request more expert advice on the topic of assessing aggregation
algorithms for watershed health rankings.
I am currently embarking on thesis work centering on the Environmental
Protection Agency's (EPA) Index of Watershed Indicators (IWI). The IWI
ranks the 2,108 watersheds (USGS 8-digit hyrologic unit code "Cataloging
Unit") in the U.S for 15 different categories plus an overall score that
combines the 15 categories.
I will analyze the effectiveness of various aggregation algorithms in
determining the overall score (combination of the 15 categories). I am
trying to establish a reference or method for gaging effectiveness of
algorithms. I have 2 preliminary ideas:
1) Compare corelations of overall rankings with proximity of neighboring
I am thinking that the closer together the centroids of watersheds are,
the closer their overall rankings would be. The degree of corelation of
these variables could serve as a measure of an aggregation algorithm's
2) Compare overall rankings with other national rankings of the health of
water in watersheds.
The problem with this method is the lack of any other such national
rankings, to my knowledge. [Does anyone else know of other similar
A possible alternative to this approach could be to survey some experts in
the field as to their epxert evaluation of watershed health for a few
sample watersheds. These expert evaluations could be used to gage the
effectiveness of various aggregation algorithms.
Input on any of the above topics is greatly appreciated.
Thank you very much. Sincerely,
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!