Compost Sampling for Nutrient and Quality Parameters: Variability of Sampler, Timing and Pile Depth

Abstract

In order to establish analysis-based quality guidelines for composts, it is necessary to have information about the variability of these traits as dependent on sampling methods. Composts exhibit significant spatial, site and time-related variability. In a preliminary survey of commercial and home users of compost in New York State, respondents indicated strong interest in test quality of purchased composts. However, little work has been done examining the consistency of reported values in composts sampled and shipped to laboratories. We evaluated within-farm and between-sampler variability of test traits at 7 farm compost sites. In one study we compared farmer and extension agents as separate samplers of the same compost piles. In the next study we examined repeated sampling by same sampler visiting the site a second time. Finally, we compared depth-based sampling variability at 3 selected farm compost sites. Analysis of variance was employed to distinguish effects of samplers, sampling timing and farms. The data showed very small differences in test values due to samplers for all parameters except weed seed and fecal coliforms. Geometric mean transformation eliminated fecal counts as a source of significant variability. Repeated sampling after two weeks of matured compost indicated no appreciable differences between points of time except for weed seeds. The sampling-depth study revealed significant variation of several selected test parameters attributed to pile stratification, and the variables most affected were moisture, potassium and weed seeds. Weed seed testing may not be useful for a quality parameter unless methodological variance is better understood. Overall, biological parameters varied less than chemical traits. The study underscores that prior to establishing numerical quality guidelines the preferred compost pile sampling protocol should be very carefully described.