Re: Fw: AI-GEOSTATS: Small data set
- Dear Isobel,
I have received your analysis. I going to reproduce this analysis in details
as you suggest me.
Thank you very much for your kindness
----- Original Message -----
From: "Isobel Clark" <drisobelclark@...>
To: "Jos?_Quint?n_Cuador_Gil" <cuador@...>
Sent: Sunday, April 11, 2004 12:36 PM
Subject: Re: Fw: AI-GEOSTATS: Small data set
I have had a quick look at your data. You have
sufficient data for a geostatistical analysis, but you
have two problems which show up in these data files:
Data1: this large scale data file shows a very strong
trend accounting for around 99% of the variation
between the values. Your values are high in the
northwest and drop off consistently to the southeast.
Once the trend is removed, you have a slightly erratic
semi-variogram which can be modelled with a spherical
model, range of influence 3,000 (metres?). Sampling
should be at least 2,000 (metres?) coverage to get any
reliable mapping. Universal kriging is needed to
account for the trend. I used a search radius of
10,000 for UK because of needing to estimate the trend
Data2: this (geostatistical?) cross of values located
in the far south-east of your study area shows very
strange behaviour in several respects:
(a) the values seem to be clustered into three
groups, 4 samples around 71, 2 samples around
74 and a cohesive group in the 72.2-73 range.
(b) the general trend of data set 1 and the
values of the local samples in data set 1
are almost exactly 12 (units?) below the
average of the samples in data set 2.
In short, I like your Data1 samples but am very
worried about your Data2 samples in the same context.
I have attached a pdf file with some screen dumps of
part of my analyses. I also attach a copy of your data
which can be used with our demo software.
Download from http://geoecosse.bizland.com/softwares
You can reproduce most of what I have done this way. I
have combined the files with the samples identified by
'zone code' -- 1 for Data1, 2 for Data2.
III Simposio Internacional sobre Manejo Sostenible de Recursos Forestales,
Universidad de Pinar del Rio, Cuba 21-23 Abril del 2004.
Telef:(53)(82)779363, Fax:(53)(82)779353 email:fhernandez@...
* 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