SEARCH

SEARCH BY CITATION

Figure S1. Check for differences in analytical precision in soil pH measurement between the two NSI samplings (1978–1983 and 1994–2003). Stored samples from 100 of the original sites were re-analysed for soil pH in 2003. The figure shows good agreement between the original and re-analysed pH values across the full range of pH. The slope of the reduced major axis regression (Sokal & Rohlf, 1981) of these two variables is 1.05±0.06, close to 1.0. The concordance correlation (Lin, 1989), ρc, is large (0.96) indicating only small bias and random deviations between the two variables. Bias is quantified by Cb where ρc=Cbρ and ρ is the conventional correlation. If Cb=1 then there is no systematic deviation; here Cb was 0.9997.

Figure S2. Output of the disjunctive kriging of the spatial variation in measured pH. Left hand panel: local estimates of the variance in the rate of change in pH. Right hand panel: the conditional probability that the change in pH was negative.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
GCB_2135_sm_supplinfo.doc697KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.