The use of relaxed eddy accumulation to measure biosphere-atmosphere exchange of isoprene and other biological trace gases

The micrometeorological flux measurement technique known as relaxed eddy accumulation (REA) holds promise as a powerful new tool for ecologists. The more popular eddy covariance (eddy correlation) technique requires the use of sensors that can respond at fast rates (10 Hz), and these are unavailable for many ecologically relevant compounds. In contrast, the use of REA allows flux measurement with sensors that have much slower response time, such as gas chromatography and mass spectrometry. In this review, relevant micrometeorological details underlying REA are presented, and critical analytical and system design details are discussed, with the goal of introducing the technique and its potential applications to ecologists. The validity of REA for measuring fluxes of isoprene, a photochemically reactive hydrocarbon emitted by several plant species, was tested with measurements over an oak-hickory forest in the Walker Branch Watershed in eastern Tennessee. Concurrent eddy covariance measurements of isoprene flux were made using a newly available chemiluminesence instrument. Excellent agreement was obtained between the two techniques (r ² = 0.974, n = 62), providing the first direct comparison between REA and eddy covariance for measuring the flux rate of a reactive compound. The influence of a bias in vertical wind velocity on the accuracy of REA was examined. This bias has been thought to be a source of significant error in the past. Measurements of normalized bias () alone would lead us to think that a large potential error exists at this site. However, with our isoprene data and through simulations of REA with fast-response H₂O and CO₂ data, we conclude that accurate REA flux measurements can be made even in the presence of a bias in w. Received: 26 March 1997 / Accepted: 14 April 1998     

Composition and structure of a chronosequence of young, mixed-species forests in southeastern Ohio, USA

Clearcutting, a commonly used silvicultural practice in southeastern Ohio, often results in a forest stand with a different species composition than the parent stand. The time frames during which shifts in species composition occur on different sites are unclear. While some studies have documented species composition at specific points in time, none have attempted to examine differences throughout the first decades of stand development. This study focused on the early successional dynamics of young, mixed-species forests of southeastern Ohio. Species compositions were examined across a chronosequence of sixteen stands that developed following clearcutting. Stand ages ranged from six to 26 years. The sample was limited to dry-mesic hardwood forests on southerly aspects and on soils derived from residuum or colluvium. Across the chronosequence, stand density ranged from 17 636 stems ha^-1 at age 6 to 2759 stems ha^-1 at age 26, and basal area ranged from 8.2 m² ha^-1 to 22.1 m² ha^-1. Clumps comprised a substantial portion of the total stand density and basal area. At age 6–8 years after clearcutting, clumped stems accounted for 35.1% of the density and 48.2% of the basal area. At age 26 years, these proportions were 25.7% and 29.4%, respectively. Clumped stems were significantly larger (p<0.05) than non-clumped stems at each age group except 26 years. Total Quercus spp. density was greatest at age 6–8 years (3386 stems ha^-1), and least at age 26 years (581 stems ha^-1). When considered as a proportion of the total stand, however, the proportion was relatively stable, averaging 21.3%. However, importance value (IV=[relative density + relative basal area]/2) of Quercus in the upper canopy (dominant and codominant crown classes) was twice as much (72%) at age 26 years compared to age 6–8 years (35%). Quercus prinus L. was the major species across the chronosequence. For all age groups except 18–20 years. Q. prinus IV was the highest of any individual species in the upper canopy, and it ranged from 27 in the youngest stands (6–8 years) to 69 in the oldest stand (26 years). Within the intermediate crown class, the IV of Q. prinus equaled or exceeded those of all other species, except for the 18–20 year age group where it was second to A. rubrum. Quercus alba L. and Quercus velutina Lam. were minor components at age 26 years, although they dominated a comparison sample of six mature stands of the same ecosystem type. Liriodendron tulipifera L. was abundant 6–8 years after clearcutting, but nearly absent at age 26 years. Acer rubrum L. was the major species in both the intermediate and overtopped crown classes throughout the chronosequence. As gaps in the canopy occur. A. rubrum may become a more common species within the dominant-codominant crown class.