Reponse to Interactive comment on “Reassessing the ratio of glyoxal to formaldehyde as an indicator of hydrocarbon precursor speciation”

Abstract. The yield of formaldehyde (HCHO) and glyoxal (CHOCHO) from oxidation of volatile organic compounds (VOCs) depends on precursor VOC structure and the concentration of NOx (NOx = NO + NO2). Previous work has proposed that the ratio of CHOCHO to HCHO (RGF) can be used as an indicator of precursor VOC speciation, and absolute concentrations of the CHOCHO and HCHO as indicators of NOx. Because this metric is measurable by satellite, it is potentially useful on a global scale; however, absolute values and trends in RGF have differed between satellite and ground-based observations. To investigate potential causes of previous discrepancies and the usefulness of this ratio, we present measurements of CHOCHO and HCHO over the southeastern United States (SE US) from the 2013 SENEX (Southeast Nexus) flight campaign, and compare these measurements with OMI (Ozone Monitoring Instrument) satellite retrievals. High time-resolution flight measurements show that high RGF is associated with monoterpene emissions, low RGF is associated with isoprene oxidation, and emissions associated with oil and gas production can lead to small-scale variation in regional RGF. During the summertime in the SE US, RGF is not a reliable diagnostic of anthropogenic VOC emissions, as HCHO and CHOCHO production are dominated by isoprene oxidation. Our results show that the new CHOCHO retrieval algorithm reduces the previous disagreement between satellite and in situ RGF observations. As the absolute values and trends in RGF observed during SENEX are largely reproduced by OMI observations, we conclude that satellite-based observations of RGF can be used alongside knowledge of land use as a global diagnostic of dominant hydrocarbon speciation.

The sources of errors in satellite measurements are numerous, including uncertainties in absorption cross sections, the computation of the air mass factor, instrumental uncertainty (e.g., wavelength calibration), potential interferences from other compounds, and low signal to noise. However, seasonal averaging helps to reduce these errors. Assuming a 15% systematic uncertainty and following the formulation thoroughly explained in Vrekoussis et al. (2010), (section 4.3.1), the average error in satellite R GF over the SE US is 0.005, which is 18% of the average R GF value observed in this region. This is now discussed in section 2.2.
The error calculated for the pixel over the Kisatchie forest (0.0047) is not larger than errors in other pixels (figure at right: flight track is shown in black, forest in grey). The reviewer is correct that at this scale, pixel-to-pixel variation in R GF is mostly associated with noise. Therefore, the retrievals shown here cannot distinguish the local influence of the forest.
This is now discussed in section 3.5 2) Please cite, and compare your results with the study of Lee et al. (1998) which also provided vertical profiles of formaldehyde, glyoxal and other OVOCs over a BVOC-rich area in the Southeast US. Please provide a plot of the mixing ratios of HCHO and CHOCHO instead of (or in addition to) the profiles given in molec/cm3 (Fig. 8). This would facilitate comparison with previous studies. The Lee et al. study also found slightly higher RGF values in the FT compared to the BL. The possible source for the apparent additional source of glyoxal in the FT is unknown, as pointed out in this manuscript, but it has been hypothesized that the heterogeneous oxidation of aerosols might release glyoxal and other OVOCs, as a possible explanation for high CHOCHO in the FT over the Tropical Pacific (Volkamer et al., 2015). The ozonolysis of fatty acids has indeed been found to be a source of glyoxal and other compounds (Zhou et al., 2014).
We now include a quantitative comparison of our measurements and the Lee et al. (1998) measurements in section 3. Figure 8 has been remade to show HCHO and CHOCHO in units of ppb. Comparison with R GF vertical profiles observed by Lee et al. (1998) and Li et al. (2014) is now included in section 3.5. Section 3.5 also now mentions the possibility of heterogernous oxidation of aerosols as a CHOCHO source in the F.T.

Minor comments 2.2, p. 6244: Provide some discussion of the uncertainties in the satellite retrievals
See response to major comment #1.
p. 6246, line 2: which monoterpenes are emitted by longleaf pines? Frankin and Snyder (1971) mention alpha-pinene and 1-pinene, but there should be more recent studies. This is relevant as there might possibly be large differences between the glyoxal yields of different monoterpenes.
We now cite a more recent study which details the relative emissions rates of speciated monoterpenes from longleaf pines (Pinus palustris). The emission rate of β-pinene is the largest, approximately 30% greater than the α-pinene emission rate. All other monoterpenes emissions are at least an order of magnitude lower (Geron et al., 2000). The modeled relative abundance of HCHO and CHOCHO from the oxidation of α-pinene and β-pinene is included in Table 2.
p. 6248, l. 6: "Isoprene is still likely the dominant OVOC precursor": true, but aren't there means to prove that hypothesis?
Proving this requires modeling the complete HCHO and CHOCHO budgets, which is beyond the scope of this work. However, we now state that isoprene is a much larger source of OH reactivity than anthropogenic VOCs. This strongly supports our conclusion that isoprene is likely the largest source of both HCHO and CHOCHO.
p. 6248, l. 14-16: Yes, ISOPOOH can interfere with MVK+MACR measurement, but this does not weaken the argument that oxidation occurs faster in the plume, since ISOPOOH is also isoprene oxidation product.
If ISOPOOH creates a positive bias MVK+MACR measurement, the artifact would be larger in the low-NO x areas, artificially increasing the (MVK+MACR)/isoprene ratio observed outside of the plume. Because (MVK+MACR)/isoprene is higher inside the plume, any interference would not affect the conclusion that oxidation occurs faster in the plume. This is now stated in the manuscript.
p. [6249][6250] and While ethene and ethyne are not expected to contribute significantly to the HCHO and CHOCHO budgets, they are now included in Table 2 and discussed in section 3.4. By providing results for increasing length of alkane (ethane to butane) and also bond order (ethane to ethyne), we provide one example of the effect of precursor structure on resultant R GF .
We thank the reviewer for the careful reading. All minor/technical remarks have been addressed.