IMPROVED PROCEDURE FOR OBTAINING STATISTICALLY VALID PARAMETER ESTIMATES FROM SOIL RESPIRATION DATA

A method to obtain statistically reliable estimates of parameters of microbial activity in soil is presented. Most traditional methods of measuring soil respiration involve the analysis of curves of CO2 evolution obtained by accumulating discrete measurements of the CO2 over time. Such analyses can be misleading because they do not provide a statistically valid way of estimating error and result in non-random, autocorrelated residuals. We have developed a method that uses non-linear regression analysis of discrete, non-accumulated data, thus alleviating most of the problems associated with the analysis of cumulative data.