Premium
Mass Loss and Nutrient Changes in Decomposing Upland Oak and Mesic Mixed‐Hardwood Leaf Litter
Author(s) -
Kelly J. M.,
Beauchamp J. J.
Publication year - 1987
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1987.03615995005100060038x
Subject(s) - litter , hardwood , nutrient , watershed , environmental science , agronomy , zoology , forestry , botany , geography , biology , ecology , machine learning , computer science
Nylon net litterbags (30 by 30 cm) containing ∼10 g of upland oak leaf litter were collected at 30‐d intervals over a 1110‐d period from two upland oak ( Quercus sp.)‐mixed hardwood stands on the Cumberland Plateau of Tennessee. Statistically significant differences ( P ≤0.01) in losses of mass, N, K, Ca, and S were found between sites, whereas P and Mg losses did not differ. Litterbags containing mesic mixed‐hardwood leaf litter were collected at 30‐d intervals over a 810‐d period from mesic mixed‐hardwood stands located on the same watershed study sites as the upland oak stands. Statistical comparisons indicated significant differences in mass, N, P, K, Ca, Mg and S loss between sites. Constant rate projections of the time required for the upland oak litter weight to reach 10% of the original bag content based on 365, 730, and 1095 d of data were 960, 1530, and 1770 d, respectively, at the Camp Branch Watershed and 870, 1230, and 1410 d at the Cross Creek Watershed. Estimates of decomposition times derived from constant rate projections for both cover types were considerably less than observed values. In an attempt to obtain more realistic projections of weight and nutrient loss, several mathematical models commonly used to examine decomposition data were evaluated. A modified exponential model gave the best data fit and provided estimates of 3240 and 2370 d as the time required to reach the 10% mass‐remaining value for the upland oak litterbags at Camp Branch and Cross Creek, respectively. Estimates for the mesic mixed‐hardwoods were 1590 and 2490 d at Camp Branch and Cross Creek, respectively. The weight of most nutrients in the litterbags was modeled with limited success by the modified exponential model. However, N and particularly P weights could not be described by the model since they generally increased with time ove rthe study period. The results of this analysis, in addition to suggesting that a longer time is required for decomposition than is normally projected, also suggest that a single model may not be appropriate to describe all response variables. Data presented here along with the modeling work also reiterate the fact that a decay rate derived from a single species or group of species must be applied with caution over larger landscape units because of the potential differences both within and among sites.