Dear All, Thank you very much for your replies to my question concerning the relationship between the kriging error and the sample variance. Since my data is not from geology it might be useful to more fully explain it.
The data is from vector ecology, estimating the density of mosquito vectors of West Nile Virus in a very heavily visited (more than 100,000 visitors per month)256 acre site. The data is from only 12 light traps, a small number but they are relatively close together so sampling is considered dense given special problems of trap placement and manpower. The furthest distance between traps is only 900 meters. Spatial correlation is high as measured by Moran's I for total mosquito's and most of the individual species that make of the total. Variogram range is between 300 and 500 meters and follows flight range for the species concerned. Data is nominally normally distributed, but in the presence of strong spatial correlation, sample variance can be very high during some weeks. No trend or drift has been detected. The kriged estimates are used to concentrate spraying to areas of maximum density of putative vectors based on previous virus isolates by species. This has been much more effective than using just the light trap data, since modeled densities often suggest even higher densities away from the traps. Co-kriging with a layer created by combining vegetation , host availability and previous mosquito observations by artificial neural networking has improved cross validation and and lowered kriging error. In areas where the kriging error is higher than the sample variance, we have assumes the sample mean. any further comments that you have on this approach would be greatly appreciated. Sincerely Russell Barbour Ph.D. Research Associate in Applied Mathematics Vector Ecology Laboratory Yale School of Medicine 60 College St. Rm 600 New Haven CT. 06520 TEL: 203 785 3223 FAX 203 785 3604 email: [EMAIL PROTECTED] -- * To post a message to the list, send it to [EMAIL PROTECTED] * 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 [EMAIL PROTECTED] 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
