Data comprise concentration ratios for lead and 210Pb in terrestrial wildlife and plants, and residual maximum likelihood (REML) model values for taxons (order, family and genus). Pb was used as an example of the derivation and application of a REML model for terrestrial wildlife species. The model outputs consists of a mean value for each taxa on a common scale after REML adjustment (referred to as the ‘REML mean’) taking account of the random factor (i.e. site). The REML mean value represents a relative scaling value.
Funding for preparing this data set was provided by the TREE project (http://www.ceh.ac.uk/tree) funded by the NERC, the Environment Agency and Radioactive Waste Management Ltd. under the RATE programme.
Publication date: 2018-12-18
This dataset is part of the following
REML fitting was performed using the IBM SPSS statistical package (version 24) using ‘Linear Mixed Model’ analysis within which REML fitting is a standard option. Because of the data requirements of the REML analyses, as described above, then different amounts of data were available for the runs at genus, family and order levels. The Akaike information criterion (AIC) was used as an estimator of the relative quality of statistical models (the lower the AIC the better the model). To check that including site as a random factor resulted in a better model, all runs were repeated with site removed (ensuring that an intercept was included for the fixed parameter (i.e. taxon)).