Just to reinforce what Edzer said about projections - convert your longitude and latitude data into a planar projection first - UTM should be OK though check your data doesn't cross zones. If you don't do this the stats will not produce meaningful results as degrees are not a correct spatial measure. Even for very small areas the shape and hence the interrelationship between data is very different for long lat and metres.
Martin Roseveare ______________________________________ ArchaeoPhysica Ltd. Reconnaissance & Geophysics for Archaeology Telephone: +44 (0) 7050 369789 E-mail: [EMAIL PROTECTED] Website: http://www.archaeophysica.co.uk ______________________________________ This e-mail is intended only for the addressee named above and may contain confidential or privileged information. If you receive this e-mail by mistake please advise the sender and destroy it without further disclosure of its content. Unless otherwise stated no opinions expressed in this e-mail should be regarded as representative of any policy of ArchaeoPhysica Ltd. ----- Original Message ----- From: "Marcelo Alexandre Bruno" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Wednesday, October 22, 2003 12:03 PM Subject: AI-GEOSTATS: geoR and memory > Dear list-members > > I try to run geoR for generate interpolation maps. > My data is: x(longitud, decimal degree),y(latitud, > decimal degree),z(Sa, acoustic parameter, limits range > is 0 to 1000000, extremely skewed, represent > backscatering echoes signals of fish schools). The > dataset is 3007 nautic miles of sampling! > > The analysis is run according to steps of Dr. Paulo > Justiano's homepage > http://www.est.ufpr.br/~paulojus/geoR/geoRdoc/geoRintro.html#starting > . > When kriging run, consume all memory of system, > (R.bin with only geoR consumes 86.8mb!! in Athlon > xp1.6GHz, Linux redhat 9.0, 250mb + swapp 541mb). > Is there something wrong (possible newbie question) > or its possible vectorizing computations to > reduce memory consume? > > > "The necessary reliance on a n by n matrix, where n > represents the number of observations, greatly impedes > traditional spatial statistics from handling larger > sample sizes." published from: > Pace, R. Kelley, and Ronald Barry, "Fast CARs," > Journal of Statistical Computation and Simulation, > Volume 59, Number 2, 1997, p. 123-147. > > This is the case?? > (Apologies for my long msg.) > Thanks for all. > > ===== > ## ~~~~~~~ Oceanslogo ~~~~~~~ ## > # Marcelo Alexandre Bruno > # Linux User: 124592 > # Pss-graduagco Oceanografia Biolsgica > # FUNDACAO UNIV. FEDERAL do RIO GRANDE > # Departamento de Oceanografia > # Lab. de Tecnologia Pesqueira e Hidroaczstica > # AV. ITALIA km 8 s/n - CARREIROS > # 96201-900 (0xx53) 2336528 > # Rio Grande - RS - BRAZIL > ## ---------------------------------------- ## > > Yahoo! Mail - o melhor webmail do Brasil > http://mail.yahoo.com.br > > -- > * To post a message to the list, send it to [EMAIL PROTECTED] > * 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 [EMAIL PROTECTED] 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 > -- * To post a message to the list, send it to [EMAIL PROTECTED] * 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 [EMAIL PROTECTED] 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
