RE: AI-GEOSTATS: Filtering the influence of a leaking component in a 5 componets decomposition process
If it's that clear a feature, can you model it with a polynomial surface and then subtract that from the data and work with the de-trended residuals? Tim Glover Senior Environmental Scientist - Geochemistry Geoscience Department Atlanta Area MACTEC Engineering and Consulting, Inc. Kennesaw, Georgia, USA Office 770-421-3310 Fax 770-421-3486 Email [EMAIL PROTECTED] Web www.mactec.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ricardo Centeno Sent: Thursday, September 13, 2007 10:33 AM To: ai-geostats@jrc.it Subject: AI-GEOSTATS: Filtering the influence of a leaking component in a 5 componets decomposition process Hello everyone, I don't have much experience in geostatistics and that's why I'm trying to get some advice from anyone with a better understanding in this matter. Even when I'm mentioning some specific terms they can be considered as any general variable, X and Y coordinates and a, b, c, d, e variables. I have a dataset (in my case seismic data) which is decomposed using the Gauss-Seidel method into five different components (source, receiver, shot, cmp and line). This decomposition is based on the contribution of each component to a final measured value (let's call it global amplitude) What I get at the end are five tables with XY coordinates and the corresponding value for each location (component amplitude). When I plot a map of XY coordinates vs the decomposed value (source) I can see that another component (cmp or geology) is leaking into the component I want to work on. This is, an underground river is interfering so much in the "global amplitudes", and the component of interest (source) is being affected by this strong response. Is nothing else but getting a map with green colors (because I'm not expecting much viriations in the component of interest) and a strong red lines with the shape of this river. What I want to do is to remove or filter this component (source) and make it statistically consistent with the rest of the survey. I have smoothed the data but this means spreading this undesired values around the survey, and also set a maximum limit allowed for my "component amplitudes" but then i'm taking away usefull values from other locations. Can anyone suggest a method to perform this filtering? is there any? something like location consistent statistical analysis? Please give me a hand with this... Thaks for your help, Ricardo + + To post a message to the list, send it to ai-geostats@jrc.it + 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 ai-geostats@jrc.it + 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/
AI-GEOSTATS: Filtering the influence of a leaking component in a 5 componets decomposition process
Hello everyone, I don't have much experience in geostatistics and that's why I'm trying to get some advice from anyone with a better understanding in this matter. Even when I'm mentioning some specific terms they can be considered as any general variable, X and Y coordinates and a, b, c, d, e variables. I have a dataset (in my case seismic data) which is decomposed using the Gauss-Seidel method into five different components (source, receiver, shot, cmp and line). This decomposition is based on the contribution of each component to a final measured value (let's call it global amplitude) What I get at the end are five tables with XY coordinates and the corresponding value for each location (component amplitude). When I plot a map of XY coordinates vs the decomposed value (source) I can see that another component (cmp or geology) is leaking into the component I want to work on. This is, an underground river is interfering so much in the "global amplitudes", and the component of interest (source) is being affected by this strong response. Is nothing else but getting a map with green colors (because I'm not expecting much viriations in the component of interest) and a strong red lines with the shape of this river. What I want to do is to remove or filter this component (source) and make it statistically consistent with the rest of the survey. I have smoothed the data but this means spreading this undesired values around the survey, and also set a maximum limit allowed for my "component amplitudes" but then i'm taking away usefull values from other locations. Can anyone suggest a method to perform this filtering? is there any? something like location consistent statistical analysis? Please give me a hand with this... Thaks for your help, Ricardo + + To post a message to the list, send it to ai-geostats@jrc.it + 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/
Re: AI-GEOSTATS: modelling and goodness of fit
Andrea We use Cressie's goodness of fit statistic which allows for number of pairs and other factors in semi-veriogram fitting. You can find a paper of his in Methematical Geology around 1992, or in his book. It is also illustrated in our free tutorial material at www.kriging.com Isobel Andrea Sciarretta <[EMAIL PROTECTED]> wrote: Hi, Im working on variogram modelling and in the majority of cases only R2 is available to evaluate the best-fit values, but in cases of non-linear functions, it is very criticized. Are there other standard methods to evaluate the goodness of a fit of a non-linear function (for example asymptotic confidence intervals) and how to calculate them, considering that the majority of geostatistical software do not perform any alternative coefficient? Thank you Andrea
Re: AI-GEOSTATS: modelling and goodness of fit
Andrea, have you considered maximum likelihood fitting? -- Edzer Andrea Sciarretta wrote: Hi, I’m working on variogram modelling and in the majority of cases only R2 is available to evaluate the best-fit values, but in cases of non-linear functions, it is very criticized. Are there other standard methods to evaluate the goodness of a fit of a non-linear function (for example asymptotic confidence intervals) and how to calculate them, considering that the majority of geostatistical software do not perform any alternative coefficient? Thank you Andrea + + To post a message to the list, send it to ai-geostats@jrc.it + 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/
AI-GEOSTATS: modelling and goodness of fit
Hi, I'm working on variogram modelling and in the majority of cases only R2 is available to evaluate the best-fit values, but in cases of non-linear functions, it is very criticized. Are there other standard methods to evaluate the goodness of a fit of a non-linear function (for example asymptotic confidence intervals) and how to calculate them, considering that the majority of geostatistical software do not perform any alternative coefficient? Thank you Andrea
