Decomposition patterns for foliar litter - A theory for influencing factors.

For decomposing foliar litter in forest systems the pattern for accumulated mass loss vs time varies among litter species and environments. Thus, decomposition may be complete (100% accumulated mass loss) or the rate may cease and proceed at such a low rate that a limit value for decomposition may be estimated. Such limit values have been found to range from 42% accumulated mass loss to 100%, resulting in stable fractions of 58 and 0%, respectively. To support a discussion about pattern we need causal relationships and I have used a conceptual model with three decomposition stages as a reference and discuss that in relation to two nutrients that influence the degradation of lignin and lignified tissue. Whereas a high nitrogen (N) concentration may have a stimulating effect on degradation of holocellulose in the early stage the same high N concentration may retard the degradation of lignified tissue in the late stage and the higher the N concentration the stronger the retardation. Through the enzyme manganese peroxidase (MnP), manganese (Mn) stimulates lignin degradation and thus the lignified fraction of litter. A high level of Mn appears to support a further degradation of the lignified tissue and thus results in a high limit value, whereas a high N has been related to a low value. The aim behind this review is to: (i) identify specific factors that may influence the pattern of accumulated mass loss and thus the type of regression model that describes the data; and (ii) present a possible climate influence on decomposition pattern. To this purpose we need to discuss a possible variation in pattern following the variation in concentrations of the two nutrients N and Mn. Concentrations of N and Mn in newly shed litter are both influenced by site climate, with N increasing and Mn decreasing with increasing mean annual temperature (MAT) and increasing annual actual evapotranspiration (AET). Thus, climate becomes at least an indirect regulating factor for decomposition pattern. Two main regression models are used to describe the patterns, namely the single exponential and a function separating a readily decomposed and a stable fraction, by estimating a limit value (asymptotic function). For the discussion on decomposition pattern a model parameter was used, namely estimated limit values with a so far observed range in variation from 42 to 100%. The stabilized litter fraction defined by the limit value may be used for estimating the accumulation rate of stable carbon (C) in soil. [ABSTRACT FROM AUTHOR]