Comparison of colorimetric and ICP determination of phosphorus in Mehlich3 soil extracts

Soil test laboratories are moving away from colorimetry to inductively coupled plasma (ICP) spectroscopy to quantify phosphorus (P) is soil test extracts. The use of ICP has advantages of being quicker and provides the ability to quantify P and other plant nutrients in one analytical process. The adoption of ICP for P analysis has come with some controversy because the P value with ICP is not always comparable with the colorimetric P value that has been historically used to set up fertilizer P recommendations. Soil testing laboratories at the University of Kentucky in Lexington and Princeton received ICP spectrometers for soil testing. Data are presented on the comparison of Mehlich3 soil test P using ICP versus colorimetry. The overall comparison of the two analytical techniques resulted in a linear slope of 1.02 and R-2 of 0.987 in Lexington for 1118 samples and a linear slope of 0.968 and R-2 of 0.978 in Princeton for 418 samples. The scatter of the data around a 1 : 1 line was compared with normal laboratory error. In the agronomically significant range from 0 to 30 mg dm(-3), 5.5% of all the data showed ICP greater than colorimetry above a 4 x standard deviation (4 x SD) limit, and 0.2% showed ICP less than colorimetry below the limit. Further research is needed to determine if the higher ICP results are due to higher P bioavailability or analytical interferences. Information is also presented on normal laboratory error at the higher soil test P level of 200mg dm(-3) being used to assess environmental risk of P runoff in Kentucky. Considering normal laboratory error, the 4 x SD range for 200 mg dm(-3) is 157-243 mg dm(-3). The use of a single soil test value to determine environmental risk should be approached with caution because it does not account for routine laboratory error.