I am curious to see if others have run into this: Using the option to calculate a mean concentration for an irregular block, I get an estimate of the mean that ranges from 3876 - 4084 ppm, depending upon how I discretize the irregular block (I tried spacing points at 50, 25 and 10 feet). The simple mean of the 236 samples that are located within this 'block', which is actually an entire "superfund" site in the U.S., is 1662 ppm. The distribution of the 236 samples across the site provides reasonably good spatial coverage; the area-weighted mean (using the area of Thiessen polygons to weight the samples) is 1583 ppm. The means reported by Gstat then appear to be too high. Furthermore, if I compare the estimated means that Gstat provides for various sizes of irregular blocks with the mean of point kriging values that Gstat provides for the same blocks (and same discretization), I find the results compare favorably up to an irregular block that is approximately 1/2 the size of the entire site that I am working with. The area of the entire site is approximately 55 acres. Has anybody else run into this? I would also like to hear from anyone who has investigated the relationship between the discretization of the irregular blocks and the estimated means/variances; can anyone suggest a rule of thumb to use, for example, that is based on the area/dimensions of the irregular block? ************************************************** William C. Thayer, P.E. Environmental Science Center Syracuse Research Corporation 6225 Running Ridge Road North Syracuse, NY 13212-2510 phone: (315) 452-8424 fax: (315) 452-8090 email: [EMAIL PROTECTED] **************************************************
