Maximum decomposition limits of forest litter types: a synthesis

Abstract

We used long-term litter decomposition data to estimate final decomposition levels using an asymptotic function. The estimated final limit values for decomposition were compared with available chemical data for the different litter types. A total of 41 limit values were estimated from as many different decomposition studies, and 20 different litter types were incubated in eight different forest systems. The limit values estimated varied with litter type. They ranged from about 35 to about 100% decomposition and were linearly related to the initial litter concentrations of N, Mn, and Ca in the newly shed litter. For these three nutrients, there are causal relationships to lignin degradation and to the lignin-degrading microbial community. Using all available data, we made simple and multiple linear regressions and obtained a negative linear relationship between limit value and initial N concentration (R(adj)(2) = 0.451; n = 41; p < 0.001). For needle litter alone, we obtained a better relationship (R(adj)(2) = 0.492; n = 23; p < 0.001). Manganese concentrations in litter gave a positive relationship (R(adj)(2) = 0.372; n = 25; p < 0.001), with a clear improvement when needle litter was tested for alone (R(adj)(2) = 0.512; n = 16; p < 0.001). Calcium alone gave a barely significant relationship. When combining nutrients in multiple linear relationships we obtained high R(2) values, indicating that the models were good. Thus for all sites and litter types, N, Mn, and Ca combined gave an R(adj)(2) value of 0.640 with n = 25 (p < 0.001). All needle litters combined gave an R(adj)(2) of 0.745 (n = 16; p < 0.001). The significance of this finding is discussed.