Search Results

You are looking at 1 - 1 of 1 items for :

  • Author or Editor: William M. Frank x
  • Journal of Atmospheric and Oceanic Technology x
  • Refine by Access: All Content x
Clear All Modify Search
John M. Frank
William J. Massman
Edward Swiatek
Herb A. Zimmerman
, and
Brent E. Ewers


Sonic anemometry is fundamental to all eddy-covariance studies of surface energy and ecosystem carbon and water balance. Recent studies have shown that some nonorthogonal anemometers underestimate vertical wind. Here it is hypothesized that this is due to a lack of transducer and structural shadowing correction. This is tested with a replicated intercomparison experiment between orthogonal (K-probe, Applied Technologies, Inc.) and nonorthogonal (A-probe, Applied Technologies, Inc.; and CSAT3 and CSAT3V, Campbell Scientific, Inc.) anemometer designs. For each of the 12 weeks, five randomly selected and located anemometers were mounted both vertically and horizontally. Bayesian analysis was used to test differences between half-hourly anemometer measurements of the standard deviation of wind (σ u , συ, and σ w ) and temperature, turbulent kinetic energy (TKE), the ratio between vertical/horizontal TKE (VHTKE), and sensible heat flux (H). Datasets were analyzed with various applications of transducer shadow correction. Using the manufacturer’s current recommendations, orthogonal anemometers partitioned higher VHTKE and measured about 8%–9% higher σ w and ~10% higher H. This difference can be mitigated by adding shadow correction to nonorthogonal anemometers. The horizontal manipulation challenged each anemometer to measure the three dimensions consistently, which allowed for testing two hypotheses explaining the underestimate in vertical wind. While measurements were essentially unchanged when the orthogonal anemometers were mounted sideways, the nonorthogonal anemometers changed substantially and confirmed the lack of shadow correction. Considering the ubiquity of nonorthogonal anemometers, these results are consequential across flux networks and could potentially explain half of the ~20% missing energy that is typical at most flux sites.

Full access