Hi- No one responded as of 1500 on 28 Jan so a few ideas: I assume you have irregular data and are using a TIN to generate a grid. Since a point in a TINMessage 1 of 2 , Jan 28, 2004View SourceHi-
No one responded as of 1500 on 28 Jan so a few ideas:
I assume you have irregular data and are using a TIN to generate a grid.
Since a point in a TIN triangle is a linear combination of the 3 points the
linear estimation variance using the variogram of the TIN X Y Z points and
weights should work as error estimator.
Possible pit falls
1.) If you are modeling topography it is often non covariance stationary.
2.) If the xyz data are contour lines you should honor the contours when you
3.) Three points are probably less than optimal for min estimation variance.
A cross validation where you drop out a point, TIN with nearby data and read
the z at the x y of the thrown out point and compare to actual z at that
point should also work if the data isn't on contour lines. (and cross
validation isn't too hard or time consuming in arcGis).
Hope this helps
Robert (Bob) L. Sandefur PE
Senior Geostatistician / Reserve Analyst
200 Union Suite G-13
303 472-3240 (cell) <-best choice
303 716-1617 ext 14
From: ai-geostats-list@... [mailto:ai-geostats-list@...]On Behalf Of
Sent: Tuesday, January 27, 2004 3:20 AM
Subject: AI-GEOSTATS: request - accurancy TIN
I must generate a DEM using the TIN interpolator, using geostatistical
analyst extension of ArcGis software can I define the DEM accurancy (raster
output not only thr RMS error) like I do with the kriging interpolator?
if it is not possible tell me a method to do this,which software can I use?
thank you for your availability
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org