Loading ...
Sorry, an error occurred while loading the content.

Re: [Re: AI-GEOSTATS: Cokriging]

Expand Messages
  • Gregoire Dubois
    The following paper might also be interesting (it is also stored on the AI-GEOSTATS web site, topic: online papers) Performance comparison of geostatistical
    Message 1 of 1 , Jan 16, 2003
    • 0 Attachment
      The following paper might also be interesting (it is also stored on the
      AI-GEOSTATS web site, topic: online papers)

      "Performance comparison of geostatistical algorithms for incorporating
      elevation into the mapping of precipitation"

      Author; Pierre Goovaerts

      REFERENCE:

      Proceedings of, the IV International Conference on GeoComputation, was Mary
      Washington College in Fredericksburg, VA, USA, 25-28 July 1999.

      Full paper at

      http://www.geovista.psu.edu/sites/geocomp99/Gc99/023/gc_023.htm

      Abstract

      This paper presents three geostatistical algorithms for incorporating a
      digital elevation model into the spatial prediction of rainfall: simple
      kriging with varying local means, kriging with an external drift, and
      colocated cokriging. The techniques are illustrated using annual and monthly
      rainfall observations at 36 climatic stations in a 5,000 km2 region of
      Portugal. Cross validation is used to compare the prediction performances of
      the three geostatistical interpolation algorithms with the straightforward
      linear regression of rainfall against elevation and three univariate
      techniques: Thiessen polygon, inverse square distance, and ordinary kriging.

      Larger prediction errors are obtained for the two algorithms (inverse square
      distance, Thiessen polygon) that ignore both elevation and rainfall records at
      surrounding stations. The three multivariate geostatistical algorithms
      outperform other interpolators, in particular linear regression, which
      stresses the importance of accounting for spatially dependent rainfall
      observations in addition to the colocated elevation. Last, ordinary kriging
      yields more accurate predictions than linear regression when the correlation
      between rainfall and elevation is moderate (less than 0.75 in the case study).
      "

      Hope this helps,

      Gregoire




      "RGN" <tmalvic@...> wrote:
      > Dear Mr. Silva,
      >
      > I think your problem with (un)expected results for different interpolation
      > methods is not so rare. I had similar problem with (too) small differences
      > between IDW and Kriging/Cokriging solutions. If you are interested about it
      > you can find question and answers stored in ai-geostats mailing list in
      > archive 11/2002 (group.yahoo.com/group/new_ai_geostats/ - question in mail
      > no.756, complete answers in mail no. 761).
      >
      > Maybe only could add from my experience that Cokriging can be sometimes
      > replaced with Kriging with External Drift. Of course carefully, because as
      I
      > understood in theory of Cokriging secondary variable describes behaviour of
      > primary, and in KED secondary variable has influence on primary.
      >
      > But, this mailing list includes many experts with much more experience and
      > knowledge then I have and probably you will find more useful information in
      > ai-geostats mailing archives.
      >
      > Best regards,
      > Tomislav Malvic
      > ------------------------------------
      > MSc. Tomislav Malvic, Grad. Eng. of Geol.
      > INA-Industry of Oil plc. (Naftaplin)
      > Subiceva 29, 10000 Zagreb, CROATIA
      >
      >
      >
      >
      >
      >
      >
      >
      >
      >
      >
      >
      >
      >
      > ----- Original Message -----
      > From: Alvaro Silva
      > To: ai-geostats@...
      > Sent: Tuesday, January 14, 2003 1:02 PM
      > Subject: AI-GEOSTATS: Cokriging
      >
      >
      > Dear all,
      >
      > I'm comparing some different methods in temperature estimation, i.e. idw,
      > spline, ordinary and universal kriging and cokriging. I'm doing it in
      > geostatistical analyst (ARCVIEW 8.1), nevertheless the low errors (by
      x-val)
      > presented by all the methods, some give better results. The universal
      > cokriging (covariable altitude) is the one with low errors and r nearest to
      > 1, but the map is very close to the other methods's maps. It doesn't reveal
      > the conexion with altitude like it should (the regression coeficcient
      > between temperature and altitude is high 0.88). The DEM has 1km resolution,
      > but the cokriging's map is smooth like the ones from other methods. What
      can
      > be wrong? I have already reduced the search elypse in the variable altitude
      > to the minimum and the points to include also (so i'm doing something like
      > colocalized cokriging), but the map doesn't reveal much spacial
      variability.
      >
      > Thanks for your attention
      >
      > Álvaro
      >
      >
      >
      > José Álvaro Mendes Pimpão Alves Silva
      > Geógrafo - Técnico de SIG Geographer - GIS Technician
      > Departamento de Clima e Ambiente Atmosférico Climate Department
      > Instituto de Meteorologia Portuguese Meteorological Institute
      >
      > Rua C do Aeroporto
      > 1749 - 047 Lisboa
      > Portugal
      >
      > Tel: (+351) 218483961
      > Fax: (+351) 218402370
      > Email: alvaro.silva@...
      >
      >
      > --
      > * 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



      --
      * 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
    Your message has been successfully submitted and would be delivered to recipients shortly.