Wakene Negassa, Crop and Soil Sciences, Michigan State University, East Lansing, MI and Alexandra Kravchenko, Crop and Soil ciences, Michigan State University, East Lansing, MI
Accumulation of particulate organic matter (POM) is an early indication of carbon sequestration in agroecosystems. X-ray computed tomography (CT) is a new opportunity to study POM in soil aggregates in a non-destructive way, however; its application needs validation of regular POM measurement. Such validation can be difficult because size of individual soil aggregates or soil samples used for CT is more than 100 folds smaller than that used for regular POM analysis. The objectives of the present study are to (i) evaluate performance and reliability of POM determination methods as a function of sample size (ii) compare mass and volume based POM determinations in small samples using the loss on ignition approach (iii) compare measured POM with that obtained by analyzing 3D X-ray CT images of intact soil aggregates from contrasting land use systems. The concentration of POM recovered from the smallest soil sample size (0.25 g) was the highest of all sample sizes used in the present study (0.5 to 25 g) regardless of variation in land use systems. Such significant variation could be attributed to lack of complete dispersion of small sample sizes in small solution (1 ml). In general, the results indicate that sample sizes and pretreatments significantly affect the concentration of POM, and that a combination of POM recovered from small sample size, gray-scale value range, statistical and geostatistical filters can successfully identify POM on CT images of soil aggregates.