Considerations for Measuring Flux in Extremely Cold Environments

The LI-7500 has been used in eddy covariance measurements of net ecosystem exchange of CO2 and water vapor for nearly a decade, with excellent results from diverse environments all over the world. It has ample sensitivity and frequency response for such measurements, and has proven to be stable and robust. The LI-7500 can operate on solar power and requires little maintenance, which make it ideal for long-term studies at remote sites. Several studies have been published showing excellent agreement between eddy covariance flux measurements made with the LI-7500 and closed path instruments (e.g. Miller, et al., 2004; Haslwanter, et al., 2008; Clement, et al., 2009), and other work has confirmed the accuracy of the corrections proposed by Webb, et al. (1980) that are needed to account for air expansion and water vapor dilution due to sensible and latent heat fluxes (Leuning, et al., 1982; Ham and Heilman, 2003).

In recent years, a few reports have emerged describing small apparent CO2 uptakes where none were expected when fluxes were measured using open path LI-7500 instruments. Apparent uptakes were most significant when air temperatures were quite low, e.g. -10°C, or less (Amiro, et al., 2006; Lafleur and Humphreys, 2008; Jarvi, et al., 2009), becoming small at milder temperatures during the off-season (Grelle and Burba, 2007; Burba, et al., 2008; Ono, et al., 2008), and they have not been observed during the growing season even in cool northern or alpine environments (e.g. Amiro, et al., 2006, Haslwanter, et al., 2008).

Although it is not fully understood, studies suggest (Grelle and Burba, 2007; Burba, et al., 2008) the apparent uptake may be related to sensible heat entering the optical path from instrument surfaces warmed by energy dissipated in the electronics. The effect may also be influenced by other factors, such as instrument mounting orientation, air temperature, and wind speed.

An adjustment for instrument surface warming, applied in the Webb-Pearman-Leuning (WPL) term, often eliminates most of this apparent off-season CO2 uptake (Burba, et al., 2008). The correction is typically less than 0.025 mg CO2 m-2 s-1 (0.6 µmol CO2 m-2 s-1), which is an order of magnitude smaller than other necessary flux corrections, such as the WPL corrections and closed-path frequency response corrections, and similar in magnitude to open-path frequency response corrections. Adjustments for instrument surface heating can be applied to previously collected CO2 flux data using an approach developed from comparison of flux measurements made with open-path and closed-path instruments at a specific site (Jarvi et al., 2009), or using semi-empirical methods based on standard weather variables (Burba, et al., 2008). Adjustments for surface heating are not necessary for data collected during the growing season, and they will not significantly affect results if applied. The current physical and mathematical descriptions of the surface heating corrections should be considered preliminary and will benefit from further research.

Figure 1. Illustration of the typical magnitude of CO2 fluxes in summer (left) and winter (right) over a healthy ryegrass ecosystem. The correction term eliminates most signs of wintertime uptake. It has little or no impact on warm-season data.

Raw uncorrected flux
Flux after standard corrections for open-path frequency response (Moore, 1986, Boundary-Layer Meteorology, 37: 17-35)
Flux after standard Webb-Pearlman-Leuning density corrections (Webb et al., 1980, Quarterly Journal of Royal Meteorological Society, 106: 85-100)
Flux after correction for instrument surface heating (Burba et al., 2008, Global Change Biology, 14:1854-1876)

At LI-COR, we take measurement accuracy seriously. We have made both scientific and engineering efforts to understand, quantify, and mitigate this phenomenon; and we have published our results in international journals. We continue to partner with scientists around the world to “measure change in a changing world,” and we expect to find and resolve problems that emerge along the way, as we come to an ever deeper understanding of the world around us.

A series of experiments, which characterize the influence of surface-heating and explain the correction in detail, are described in the following documents: