Re: AI-GEOSTATS: choosing grid resolution

[Duane Marble]

There is also a fair amount of literature on the Modifiable Areal Unit Problem (MAUP) that goes back for several decades. The role of spatial scale in the analysis can be a critical one with not only changes in cell size but also rotations, etc., applied to the basic grid causing shifts in correlation behavior. Duane Marble On 3/14/2011 7:30 PM, Younes Fadakar wrote: Dear Raul, Any changes in the size {<GS>support</GS>}, shape {<GS>anisotropy</GS>} etc of cells (grids) in cell-based analysis including from spatial analysis (GIS etc) and geostatistics (grided variogram, kriging etc) need to be considered carefully while they can provide very different correlation structure and so result in very different interpretations. Geostatistically speaking, the support effect should be considered initially very well. The support can be simply defined as how your choice (e.g. cell) is being represented with the samples inside it (e.g., number of samples per cell). There are some methods you will find them interesting, I think: For an evaluation of the dispersion of data into cells, consider Morisita (Morishita) factor (e.g., from Kanevsky); Also consider declusterting methods (e.g. from Clayton); ... Best Regards, Younes yfa.st...@ymail.com http://alghalandis.com From: Raul Sierra Alcocer <raul.sierra.alco...@gmail.com> To: ai-geostats@jrc.it Sent: Tue, 15 March, 2011 3:49:31 AM Subject: AI-GEOSTATS: choosing grid resolution Dear list, I am a PhD student in Computer Science new to spatial analysis, and I am working in a data mining platform at my institute that will provide some tools for spatial data mining. It is basically analysis of spatial features using rectangular grids. My question is if there are any studies about the effects of grid resolution in the results of correlation analysis between spatial features. When I say effects, I mean how the choice of cell size affects the statistical significance of the results. Thanks in advance, Ra�l Sierra. --------------------------------------------------------------- Centro de Ciencias de la Complejidad, Universidad Nacional Autonoma de M�xico. + + To post a message to the list, send it to ai-geost...@jrc.ec.europa.eu + To unsubscribe, send email to majordomo@ jrc.ec.europa.eu 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/   -- Dr. Duane F. Marble Email: dmarble at OregonFast.net 2226 Primrose Lane Telephone: 541.902.8837 Florence, OR 97439 Cell: 541.991.1730 Emeritus Professor of Geography -- The Ohio State University Courtesy Professor of Geosciences -- Oregon State University An early comment on the proper use of GIS technology: "I warne yow wel, it is no childes play" Originally from Chaucer, but also quoted by me in 1967 in a monograph on software tools for computational geography + + To post a message to the list, send it to ai-geost...@jrc.ec.europa.eu + To unsubscribe, send email to majordomo@ jrc.ec.europa.eu 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/