AI-GEOSTATS The distance from the edge has an effect for sure (in an ideal situation). But if it's reflected in the actual measurements is hard to tell. Data was sampled according to a regular grid, so no real transect of a vegetation patch was made to see if the differences were significant. Given the conditions, the spatial location within a vegetation cover is hard to relate to the measurements I made. The distribution of the vegetation was patchy but the vegetation itself was also rather hetergeneous (with respect to the measurements).
Trend analysis seemed rather strange to me because there is no spatial relation (autocorrelation) between the indiviual points per block. Would it be useful? Koen > -----Original Message----- > From: Bill Northrop [mailto:[EMAIL PROTECTED] > Sent: maandag 22 mei 2006 15:38 > To: Hufkens Koen > Subject: RE: AI-GEOSTATS: question about averaging > > Hi there, > I am really in mining, but have you looked at whether the > distance from the edge of vegetation cover has any bearing on > values. ? Otherwise weighting according to coverage per block > will give you the best overall mean for the whole area. Did > you use the individual points per block for trend analysis ? > > Bill Northrop > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Hufkens Koen > Sent: Monday, May 22, 2006 11:03 AM > To: [email protected] > Subject: AI-GEOSTATS: question about averaging > > > AI-GEOSTATS > Dear list, > > I've little problem concerning the use of spatial information > when it comes to averaging data. > > I've measured 13 datapoints (a vegetation parameter) in a > regular grid as shown in the bottom part of this figure: > http://users.pandora.be/requested/images/site.jpg > > There is no spatial relation what so ever between them > (tested for it, in every possible way). So, just averaging > this data wouldn't be wrong. > > Now comes into play extra data collected in the field that > complicates it all. I also have real coverage for the data as > shown in this picture: > > http://users.pandora.be/requested/images/vegetation-cover.jpg > > (black is vegetation covered ground) > > Is there a way to merge the measured data with the coverage > data in a correct way? What I've done sofar is that I've > taken all the measurements (38x 13 datapoints) and made an > average (best fit for this population/group of data) and used > this average and the percentage cover within every 20x20m box > -as depicted in the second figure- to get a better estimate > then I would get by just averaging the measured data within > every 20x20m box. > > Is it wrong to do and if so, is there a way to incorporate > the extra information (being the coverage) to make my local > 20x20m averages more accurate? > > Best regards, > Koen > > -- > Koen Hufkens, Phds > Department of Biology > Research Group of Plant and Vegetation Ecology (PLECO) > University of Antwerp Universiteitsplein 1, B-2610 Wilrijk, > Belgium [EMAIL PROTECTED] > > + To post a message to the list, send it to [email protected] To > + unsubscribe, send email to majordomo@ jrc.it with no subject and > + "unsubscribe ai-geostats" in the message body. DO NOT SEND > Subscribe/Unsubscribe requests to the list As a general > service to list users, please remember to post a summary of > any useful responses to your questions. > + Support to the forum can be found at http://www.ai-geostats.org/ > + To post a message to the list, send it to [email protected] + To unsubscribe, send email to majordomo@ jrc.it with no subject and "unsubscribe ai-geostats" in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list + As a general service to list users, please remember to post a summary of any useful responses to your questions. + Support to the forum can be found at http://www.ai-geostats.org/
