AI-GEOSTATS: determinsitic models - residual kriging
- Hello Yetta, hello all
Yetta! i don't know whether you remeber; you sent me your tm-report in march
and i wish to thank you very much! I tried to answer you but may be your
e-mail does not work?!
As i am not an expert in geostatistics i would like to ask some questions
and would be glad if someone could bring me on the right way!
What i got and want to do:
I have data from different campaigns between 1980 and 2000 in a
morphological heterogenous (elevation) region. I want to assess large scale
(1:500.000) spatial variability of heavy metals in upper soil layers due to
atmogenous depostion. Before making spatial estimations i have to homogenize
the data (=> pattern free data!). i don't want to assess a spatial
(coordinate dependend) drift!
What i did up to now:
i built up one deterministic (pattern-)model for each element via multiple
linear regression analysis (SPSS). The pattern-models should explain
variations in element-concentrations due to different binding/retention
capacities of different soil types (c-content, pH), morphology (elevation at
sample site) and (linear?) decreasing deposition rates since 1980 (year of
taking the sample). As the data are "lognormal" i used a ln-transformation.
1. When calculating multiple linear regression (pattern-)model i did
not include xy-coordinates of the sample sites as aditional predictor
variables allthough there is considerable correlation between coordinates
and heavy metal contents (but there is no trend in variograms!). is that ok?
2. (backtransformed) semivariograms of the pattern-models still have a
"good shape" suggesting a spatial/random variability. shouldn't they behave
like "white noise"?
3. how do i get backtransformed residuals from ln-transfortmed data ?
when i backtransform ln-residuals of the regression/pattern-models negative
ln-values are just interpreted as positive values below 1,0. is it possible
to get residuals by substracting (backtransformed) predicted values
(=exp(regression model)) from the original (= untransformed) values?
4. simultanous estimation of drift and semivariogram with a single
realization is rigorously not possible. Are there some (simple)
aproximations for the semivariogram? What about the iterative solution for
the simultaneaous inference of drift and semivariogram, how does it work in
I know there are a lot of questions but i hope you can give me some answers
Bayerisches Geologisches Landesamt
Heßstr. 128, D-80797 München
* 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