Do different methods used to estimate soil phosphorus availability across Europe give comparable results?

Each European country is using its own method for the determination of phosphorus availability to plants, together with an appropriate interpretation scheme of the P status and fertilizer recommendations. In order to compare systems, a soil exchange program was organized: 16 P methods were compared on 135 soils from 12 countries. The amount of extracted P decreased in the order P-total > P-oxal. > P-AL > P-Me3 > P-Bray > P-AAEDTA, P-DL, P-CAL > P-Olsen > P-paper (strip), P-AAAc, P-Morgan > P-H2O, P-CO2, P-CaCl2, Isotopically exchangeable P was also measured. A large variability was observed in the results obtained by laboratories using the same method, thus demonstrating the great importance of an identical lab procedure as a prerequisite to any comparison. The traditional correlation/regression approach revealed its limitations when applied to non-homogeneously distributed data and was replaced by more robust techniques that showed laboratory differential bias and confidence intervals of the log-transformed values. Even though all the methods reacted in the same way to increasing amounts of added P in several trials, there were wide differences between results obtained with different methods. The interpretation schemes for P status were also compared and revealed that about 50% of the tested soils were P-deficient. This observation appears not to be in line with a generally high P fertilization during the last decades in Europe and should lead to a better evaluation of the plant-available soil phosphorus.