Assessing Litterfall Trap Bias
Forest Service scientists determined several factors that may cause bias while sampling litterfall in a forest, leading to over- or under-representation of the species present in the surrounding vegetation. Before this study, the few studies dealing with patterns of litterfall dispersal and collection provided contradictory results, so Forest Service scientists examined the correspondence between litterfall samples and standing vegetation in the Luquillo Experimental Forest in Puerto Rico.
The scientists compared litterfall and standing vegetation in a subtropical wet forest in Puerto Rico at three spatial scales—forest, sampling blocks, and plots—to determine the effect of tree height, crown area, leaf size, and distance to traps on litterfall species composition; to determine if representativeness of litterfall samples was affected by litter traps' central or lateral location relative to vegetation plots; and, to gain insights on scaling up data of litterfall from sampling plots into forest stand scale.
Higher tree height and a wider crown were more important in determining the relative abundance of species in litterfall samples than trees at a closer location to litter traps. Central and lateral traps provided equally representative samples at the forest scale. Correlations between litterfall and plant species' relative abundances at the scale of plots showed that central traps better represented the surrounding vegetation than lateral traps. When comparing vegetation community with litterfall across scales, similarity decreased from the scale of forest to that of plots.
Leaf size had a slight effect on litterfall composition suggesting that big and heavy leaves could be under-represented in litterfall samples. Due to the wide range of horizontal mobility of leaf litter, traps were not necessarily collecting leaf litter from the immediate vicinity. Therefore, care should be taken when scaling up from small to intermediate sampling units.
Forest Service Partners