Staff Publications

  • external user (warningwarning)
  • Log in as
  • language uk
Record number 401099
Title Identifying and removing heterogeneities between monitoring networks
Author(s) Skoien, J.O.; Baume, O.P.; Pebesma, E.J.; Heuvelink, G.B.M.
Source Environmetrics 21 (2010)1. - ISSN 1180-4009 - p. 66 - 84.
Department(s) Land Dynamics
Soil Science Centre
Publication type Refereed Article in a scientific journal
Publication year 2010
Keyword(s) harmonization
Abstract There is an increased interest in merging observations from local networks into larger national and international databases. However, the observations from different networks have typically been made using different equipment and applying different post-processing of the values. These heterogeneities in recorded values between networks can lead to inconsistencies between different networks, and to discontinuities at the borders between regions if the observations are used as a source for interpolated maps of the process. Such discontinuities are undesirable, and could create difficulties in interpreting the maps by decision makers. In this paper, we present two variants of a method that can be used to identify and quantify differences between networks. The first variant deals with networks sharing the same region (usually multiple networks within a country) while the second variant deals with networks in neighbouring regions (usually networks in different countries). The estimated differences can be used to estimate individual biases for each network, which can be subtracted as a harmonization procedure. The method was applied to European gamma dose rate (GDR) measurements from May 2008 from the European Radiological Data Exchange Platform (EURDEP) database. Data from the Slovenian GDR network are used for an application of the first variant of the method whereas the complete dataset is used to illustrate the second variant. The results indicate that these two variants are able to identify and quantify biases reliably, and the interpolated maps after subtraction of the estimated biases appear more reliable than maps created on the basis of the recorded data
There are no comments yet. You can post the first one!
Post a comment
Please log in to use this service. Login as Wageningen University & Research user or guest user in upper right hand corner of this page.