Extended droughts will be imposed on mature canopy
trees through the use of a reusable understory roof elevated above the
forest floor. Ground level tarpaulins, which act as a surface mulch
layer, have been shown to be ineffective at producing rapid soil drying.
The roof will be constructed from commercially available canopies joined
together and installed around the target trees with a custom designed tarp,
gutter and tree sleeve system.
The understory roof will be capable of removing 100 percent of throughfall and stem flow from the rooting zone of mature trees. Observations from the TDE experiment suggest that manipulations reaching beyond the tree crown should be adequate to contain the effective horizontal rooting distribution of the target mature canopy trees. Treatment canopies will be assigned to 8 of the 16 experimental trees in a random manner and the experimental unit for all analyses of variance or covariance will be the individual tree. The a priori tests suggested are: treated vs. untreated trees, Quercus vs. Liriodendron, and the interaction between species and treatment. Hydrologic modeling of the Walker Branch forest suggests that 100% throughfall exclusion starting on or about April 15 of a mean year would be required for a minimum of 40 days to be effective at producing soil water deficits sufficient to reduce plant water use, foliar photosynthetic function, and stem growth (Stage 1). The occurrence of acute drought conditions in late spring and early summer appears to be a prerequisite for substantial growth reductions in eastern forest trees Hanson and Weltzin (2000). Total exclusion for at least 70-80 days would be required to force premature leaf senescence and cavitation of the hydraulic pathways (Stage 2). We propose to induce increasing levels of acute moisture stress over three years to experimentally mimic this simulation.
Microclimate differences at the forest floor level should be minimized because the height of the canopies above the forest floor (2-3 meters) is sufficient to allow normal air movement. However, because stagnant conditions commonly occur in this area (especially at night) air and surface soil temperature monitoring of treated and untreated plots is being conducted.