Spatial and temporal temperature variations exist in a compost pile. This study demonstrates that systematic temperature sampling of a compost pile, as is widely done, tends to underestimate these variations, which in turn may lead to false conclusions about the sanitary condition of the final product. To address these variations, a proper scheme of temperature sampling needs to be used. A comparison of the results from 21 temperature data loggers randomly introduced into a compost pile with those from 20 systematically introduced data loggers showed that the mean, maximum and minimum temperatures in both methods were very similar in their magnitudes. Overall, greater temperature variation was captured using the random method. In addition, 95% of the probes introduced systematically had attained thermophilic sanitation conditions (≥ 55°C for three consecutive days), as compared to 76% from the group that were randomly introduced. Furthermore, it was found that, from a statistical standpoint, readings from at least 47 randomly introduced temperature loggers are necessary to capture the observed temperature variation. Lastly, the turning of the compost pile was found to increase the chance that any random particle would be exposed to the temperature ≥ 55°C for three consecutive days. One turning was done during the study, and it increased the probability from 76% to nearly 85%. Using the Markov chain model it was calculated that if five turnings had been implemented on the evaluated technology, the likelihood that every particle would experience the required time-temperature condition would be 98%.
Keywords: Compost engineering; Pathogen inactivation; Temperature monitoring.
Copyright © 2014 Elsevier Ltd. All rights reserved.