Forest loss as a driver of food web change
The first part of our findings suggests that allochthony did not change over time, so we focussed on TP for our second research question. We found that the mean pairwise difference in TP(Δ \(\overset{\overline{}}{\text{TP}}\)) was best predicted by the ratio of forest loss:gain in the immediate upstream area( \(\text{ΔF}_{\text{sub}.\text{ratio}}\)) (Table 2). Further, slope coefficients describing the relationships betweenΔ \(\overset{\overline{}}{\text{TP}}\) and candidate predictors were negative for all six forest loss metrics, and positive for predictors unrelated to forest loss (Fig. 4).