An improved water correction function for Picarro greenhouse gas analyzers
Friedemann Reum, Christoph Gerbig, Jost V. Lavric, Chris W. Rella, Mathias Göckede
Atmospheric Measurement Techniques Discussions
Measurements of dry air mole fractions of atmospheric greenhouse gases are widely used in inverse models of atmospheric tracer transport to quantify the sources and sinks of the gases. The measurements have to be calibrated to a common scale to avoid bias in the inferred fluxes. The World Meteorological Organization (WMO) has set requirements for the inter-laboratory compatibility of atmospheric greenhouse gas measurements to ±0.1&thinsp;ppm for CO<sub>2</sub> (Southern hemisphere
... thinsp;ppm) and to ±2&thinsp;ppb for CH<sub>4</sub>. An established series of devices for measurements of greenhouse gas (GHG) mole fractions are the trace gas analyzers manufactured by Picarro, Inc. These have been shown to deliver dry air mole fractions with accuracies within the WMO goals when trace gas signals are measured in wet air and the effects of water vapor are corrected for. Here, we report for the first time on sensitivity of the pressure inside the measurement cavity of Picarro GHG analyzers to water vapor. This sensitivity induces biases in the inferred dry air mole fractions of CO<sub>2</sub> and CH<sub>4</sub> if they are obtained using the traditional water correction function. To correct for the pressure effect, we add a pressure-related term to the traditional water correction function, and consider differences between the traditional and enhanced water correction function to be biases of the traditional model. The effect primarily affects low water vapor mole fractions from about 0.05 to about 0.5&thinsp;%, a domain that has gone undersampled in previous studies of the water correction for Picarro GHG analyzers. We observed biases up to about 40&thinsp;% of the WMO tolerances (80&thinsp;% for CO<sub>2</sub> in the southern hemisphere). The magnitude of the effect varied across instruments and appeared to be negligible for some, and our experimental results were more robust for CH<sub>4</sub> than for CO<sub>2</sub>. Thus, correction coefficients should be determined for each analyzer individually. Applying our enhanced water correction function improves the accuracy of measurements of dry air mole fractions of CO<sub>2</sub> and CH<sub>4</sub> in humid air with Picarro GHG analyzers on a scale important for keeping the measurement accuracy within the WMO requirements.