Affiliation:
1. Jones Center at Ichauway Newton Georgia USA
Abstract
AbstractForests are critical to water resources, but high evapotranspiration (ET) can reduce water yield. Thinning and prescribed fire reduce forest density and often reduce ET, promoting higher water yield. However, results from such treatments have been inconsistent, possibly because of unknown interactions among individual ET components. We compare water budget components of longleaf pine (Pinus palustris Mill.) woodlands with frequent prescribed fire to the water budget components of fire‐excluded stands. We hypothesized that fire exclusion would result in higher ET due to increased midstory transpiration (Et) and interception (Ei), and higher evaporation from litter (Ilitter). Reference plots were burned every two years while treatment plots had fire excluded for 15–20 years. Fire treatments were repeated in two sites representing a soil moisture gradient, noted as mesic and xeric. We measured woody Et using sap flux, and we modeled groundcover Et using physiological models. We measured Ei of canopy and groundcover layers, modeled Es litter biomass, and constructed a total component‐based water budget for each site and treatment. Compared with reference plots, midstory Et was 300%–800% higher in fire exclusion plots. Groundcover Et was ~80% less than reference treatments, countering the effects of midstory growth on total ET. Stand Ei followed similar trends, with groundcover Ei in reference plots countering the effects of midstory and litter Ei in fire exclusion plots. As expected, total ET in the xeric site was 18% higher in fire exclusion plots. However, ET in the mesic site was 16% lower in the fire exclusion plots due to high groundcover Et and Ei in reference plots. Thus, our results show that fire exclusion changes total forest ET, but the size and direction of the effect vary depending on the balance between midstory and groundcover transpiration and interception. These results highlight the importance of groundcover in ecosystem function in low‐density forests and may help explain inconsistent results from studies of water yields following thinning and fire. While prescribed fire is a valuable tool in forest management, we suggest that the effects of fire on ET are complex and require careful accounting of all water fluxes within a forest ecosystem.