Abstract
Current soil organic matter (SOM) models are empirical in nature by employing few conceptual
SOM pools that have a specific turnover time, but that are not measurable and have no direct
relationship with soil structural properties. Most soil particles are held together in aggregates and
the number, size and stability of these aggregates significantly affect the size and amount of
organic matter contained in these aggregates, and its susceptibility to decomposition. While it
has been shown that soil aggregates and their dynamics can be measured directly in the
laboratory and in the field, the impact of soil aggregate dynamics on SOM decomposition has not
been explicitly incorporated in ecosystem models. Here, we present AggModel, a conceptual and
simulation model that integrates soil aggregate and SOM dynamics. In AggModel, we consider
unaggregated and microaggregated soil that can exist within or external to macroaggregated soil.
Each of the four aggregate size classes contains particulate organic matter and mineral-associated
organic matter fractions. We used published data from laboratory incubations to calibrate and
validate the biological and environmental effects on the rate of formation and breakdown of
macroaggregates and microaggregates, and the organic matter dynamics within these different
aggregate fractions. After calibration, AggModel explained more than 70% of the variation in
aggregate masses and over 90% of the variation in aggregate-associated carbon. The model
estimated the turnover time of macroaggregates as 32 days and 166 days for microaggregates.
Sensitivity analysis of AggModel parameterization supported the notion that macroaggregate
turnover rate has a strong control over microaggregate masses and, hence, carbon sequestration.
In addition to AggModel being a proof-of-concept, the advantage of a model that is based on
measurable SOM fractions is that its internal structure and dynamics can be directly calibrated
and validated by using experimental data. In conclusion, AggModel successfully incorporates the
explicit representation for the turnover of soil aggregates and their influence on SOM dynamics
and can form the basis for new SOM modules within existing ecosystem models.