An important basis for lime fertilisation is the recording of pH values. Many studies have shown that the pH value can vary greatly within a small area. Only through the development of a sensor by VERIS has it become possible to determine the pH value cheaply in a much higher sampling density than with the time and cost intensive laboratory method. With respect to their measurement principles, both methods differ fundamentally in that in the laboratory method an extraction medium is used. This causes the difference of the results between both methods to increase with decreasing pH values and makes it necessary to recalibrate the sensor values.

The subject of this paper consists of statements about agreement and precision of both methods using investigations of identical soil samples. Finally, a suggestion is given for a calibration of the sensor values, which makes the practical use of the sensor possible.

Authors who used an identically built sensor for their investigations used mainly the correlation coefficient r or R² as a quality criterion for the agreement of the results of both methods. This is, however, not sufficient for a comparison of methods since it only describes the relationship between two measurement parameters. The mean square error and the root mean square error is suitable for measuring the agreement. The results of the repeated sampling for the laboratory method show only an average quality for the repeatability (R²=0,77; MSE=0,23; RMSE=0,39). The results of the comparison of the sensor’s raw data with the laboratory method are inacceptable (R²=0,84; MSE=0,77; RMSE=0,82). Only through a post-calibration with the algorithm developed is it possible to reduce the error to an acceptable level (MSE=0,25; RMSE=0,32). Further statistical tests back up the suggested calibration. The developed approach could be tested on several practice fields.