Factors controlling evaporation and energy partitioning beneath a deciduous forest over an annual cycle

https://doi.org/10.1016/S0168-1923(00)00124-6Get rights and content

Abstract

The energy balance components were measured above the ground surface of a temperate deciduous forest over an annual cycle using the eddy covariance technique. Over a year, the net radiation at the forest floor was 21.5% of that above the canopy, but this proportion was not constant, primarily because of the distinct phenological stages separated by the emergence and senescence of leaves. The dominant response to seasonal changes in net radiation was through corresponding changes in the sensible heat flux, and both net radiation and sensible heat flux peaked just before leaf emergence. Evaporation at the forest floor was typically less than 0.5 mm per day, and unlike sensible heat flux, was not closely coupled to seasonal changes in net radiation. Instead, evaporation at the forest floor responded primarily to rapid changes in litter water content. Forest floor evaporation was limited by the water-holding capacity of litter, and when the atmospheric demand was large, the litter layer dried on the time scale of several hours. After this rapid period of drying, net radiation and sensible heat flux dominated the energy budget.

When leaves were present during the growing season, the sensible and latent energy fluxes at the forest floor were less than 10% of the total canopy fluxes, and the mean Bowen ratio was similar to that above the canopy. However, during the dormant season, the controls of the energy budget at the forest floor largely determine the whole canopy fluxes. On an annual basis, the fluxes from the forest floor are roughly 15–22% of those above the canopy and the evaporation was 86 mm.

Introduction

Energy fluxes and evapotranspiration above vegetated surfaces depend not only on climatological and biophysical controls in the vegetative canopy but also on the available energy and energy partitioning beneath the canopy. Fluxes from a forest floor and the relative contribution of the forest floor to whole canopy fluxes vary considerably, from generally low in dense coniferous forests (Schaap and Bouten, 1997) and during the growing season in deciduous forests (Moore et al., 1996), to much greater in more open canopies (Lafleur, 1992, Kelliher et al., 1997, Kelliher et al., 1998, Baldocchi et al., 2000). The presence and density of the canopy modify a number of factors that influence energy availability and partitioning at the forest floor. These factors include solar and net radiation, soil water content, wind speed, temperature and humidity beneath the canopy, as well as the quantity and quality of decomposing litter, which can act as a mulching layer (Kondo and Saigusa, 1992) largely decoupling the surface litter layer from the soil. Deciduous forests of the southeast US are unique in having a dense canopy during the growing season, but are essentially open during the dormant season. Therefore, characterizing the magnitude, partitioning and controls of energy fluxes and forest floor evaporation in deciduous forests requires data sets that extend over a complete dormant and growing season cycle.

Although leaf area influences energy availability and partitioning at the forest floor, some evidence suggests that forest floor evaporation is often largely decoupled from net radiation (Baldocchi and Meyers, 1991, Schaap and Bouten, 1997), even during the dramatic changes associated with leaf emergence and senescence in a deciduous forest (Moore et al., 1996). Soil water content near the upper horizon (Kelliher et al., 1986) and leaf litter water content (Schaap and Bouten, 1997, Baldocchi et al., 2000) have also been suggested as controls on forest floor evaporation in other studies, but litter quantity and composition can also be important, along with the presence of ground-dwelling and water-absorbing vegetation, such as moss and lichens (Lafleur, 1992, Kelliher et al., 1998). There is scarce information detailing the hourly to annual interaction between forest floor litter water content and evaporation in deciduous forests.

Baldocchi and Meyers (1991) and Baldocchi and Vogel (1996) have presented energy budget data under the canopy of a temperate deciduous forest using the eddy covariance technique, but only during short-term studies. Moore et al. (1996) presented forest floor fluxes in a deciduous forest in the northeastern US over parts of an annual cycle. The forest in the present study is located in a much warmer climate, has almost no snow cover, and has an extended growing season and a greater leaf area compared to that reported in Moore et al. (1996). We also expect soil type and litter quality and composition at the forest floor to vary between the two sites.

In this study, we discuss the energy budget beneath a temperate deciduous forest in the southeastern US over an annual cycle using the eddy covariance technique. The daily and annual fluxes beneath the canopy will be compared with the whole canopy fluxes measured above the canopy. In addition to showing the annual magnitude of energy flux densities, we discuss the likely biophysical controls on forest floor evaporation at this site. Specifically, our goals are (1) to show the seasonality and magnitude of energy fluxes beneath a deciduous forest over an annual cycle, emphasizing differences between the dormant and growing seasons, (2) to determine relative importance of forest floor fluxes with those above the canopy, and (3) to determine the role of climate forcing and litter water content on energy partitioning, particularly evaporation.

Section snippets

General forest characteristics

Micrometeorological and flux measurements were made below and above the canopy of a temperate deciduous forest in Oak Ridge, TN (35°57′30″N, 84°17′15″W, 365 m asl) continuously during 1998. The canopy height is approximately 26 m above the surface and the maximum leaf area is approximately 6.0. The forest is a mixed deciduous stand dominated in the overstorey by oak, maple and hickory. The stand is over 50 years old, having regenerated from agricultural land. The instruments were located on a

Energy balance closure and spectral analysis

Two checks on the quality of eddy covariance data are energy balance closure (Baldocchi et al., 1988) and spectral analysis (Baldocchi and Meyers, 1991). The hourly energy balance at the forest floor (LE+H versus RnfG) during the dormant season is shown in Fig. 2. The slope is less than 1 (0.80), similar to the above-canopy fluxes (Wilson and Baldocchi, 2000). The intercept of 1.98 W m−2 was not significantly different from 0 and the coefficient of determination (r2) was 0.90. Although there

Energy balance closure

The slope of the energy balance closure was less than 1 above (Wilson and Baldocchi, 2000) and below the canopy. Slopes less than 1 are typical for many sites (Aubinet et al., 1999). A lack of closure at the forest floor was partially due to high frequency losses of water vapor fluxes (up to 10% of total) and resulting underestimates of H+LE. Lack of closure at the forest floor can also result from spatial heterogeneity in solar and net radiation at the forest floor (Baldocchi et al., 2000).

Acknowledgements

This work was funded by a grant from NASA/GEWEX and the US Department of Energy (Terrestrial Carbon Program) and is a contribution to the AmeriFlux and FLUXNET projects. PJH was sponsored by the Program for Ecosystem Research, Environmental Sciences Division, Office of Health and Environmental Research, US Department of Energy under contract No. DE-ACO5-84OR21400 with Lockheed Martin Energy Research Corporation. Mark Brewer and Mark Hall provided field and laboratory assistance. We thank Tilden

References (33)

  • D.L. Auble et al.

    An open path, fast response infrared absorption gas analyzer for H2O and CO2

    Boundary Layer Meteorol.

    (1992)
  • D.D. Baldocchi

    Flux footprints under forest canopies

    Boundary Layer Meteorol.

    (1997)
  • D.D. Baldocchi et al.

    Trace gas exchange at the floor of a deciduous forest I. Evaporation and CO2 efflux

    J. Geophys. Res.

    (1991)
  • D.D. Baldocchi et al.

    Energy and CO2 flux densities above and below a temperate broad-leaved forest and a boreal pine forest

    Tree Physiol.

    (1996)
  • D.D. Baldocchi et al.

    Measuring biosphere–atmosphere exchanges of biologically related gases with micrometeorolgical methods

    Ecology

    (1988)
  • Baldocchi, D.D., Law, B.E., Anthoni, P.M., 2000. On measuring and modeling energy fluxes above the floor of a...
  • Cited by (139)

    • Evaluating surface renewal models for estimating sensible heat flux above and within a coffee agroforestry system

      2021, Agricultural and Forest Meteorology
      Citation Excerpt :

      Micrometeorological methods: (i) measure area-average evapotranspiration (or latent heat flux, λE) at spatial scales much greater than can be sampled with sapflow and micro-lysimeter methods; (ii) do not disturb the vegetation, soil, and litter layer; (iii) integrate evaporation from soil and weeds; (iv) can be carried out continuously; and (v) require relatively little effort (Baldocchi et al., 1988; Wilson et al., 2001). Although micrometeorological methods were developed for use in the atmospheric surface layer above the vegetation, the eddy covariance and surface renewal methods have been proven to work well within vegetation canopies and the roughness sublayer above (Baldocchi and Meyers, 1991; Paw U et al., 1995; Katul et al., 1996; Chen et al., 1997a, b; Baldocchi et al., 2000; Wilson et al., 2000; Holwerda and Meesters, 2019). The eddy covariance method determines λE as the covariance between turbulent fluctuations of vertical wind speed and water vapor.

    View all citing articles on Scopus
    View full text