Group 8. NMHC (Blake)

Members:

Gary Kleiman (MIT)
Nicola Blake (UCI)
Barkley Sive (UCI)
Oliver Wingenter (UCI)
Eric Saltzman (UM)

The group met to discuss NMHC results from ACE-1. NMHCs were measured on 3 platforms during ACE-1: C-130, Discoverer, and Cape Grim. The method employed aboard the C-130 was canister sampling followed by GC-FID analysis at the University of California, Irvine (UCI). Canister sampling was also used at Cape Grim followed by GC-MS analysis at CSIRO DAR. In-situ sampling from the bow (about 10m above sea level) with GC-FID analysis was employed aboard the Discoverer.

Comparison periods discussed by the group included a flyby of the Discoverer on December 5, however, the only way to include all 3 data sets is to consider the whole ACE-1 intensive period. During this period, 181 samples were analyzed on the Discoverer, about 1300 samples were collected on the C-130 with 720 samples taken in the marine boundary layer (MBL), and 9 samples were collected at Cape Grim during baseline conditions.


Table 1. Mean (in pptv) and standard deviation for ACE-1 intensive period
Compound Discoverer C-130 Cape Grim
MIT UCI CSIRO
Leg 2 MBL Baseline
N=181 n=720 N=9
Ethane 32027 16874
Ethyne 195 105
Ethene <3-8 5844
Propane 2830 124 2125
n-Butane 3148 <3-8 2125
Cis-2-Butene 109 <3 0
2-Methyl-Propene 70 (10-90) <3
Cyclo-Pentane 10-20 <3

In general, the average concentrations of hydrocarbons during the ACE-1 intensive were very low and consisted principally of C2 and C3 compounds, mostly within the concentration ranges previously observed in clean air for mid to high southern latitudes. However, Table 1 shows that there was significant disagreement for most gases, notably longer chain photochemically reactive C4 hydrocarbons reported by the MIT group. The December 5 flyby MIT and UCI produced results that were very similar to the median values shown in Table 1.

The UCI values exhibited low variability. The overall standard deviations in Table 1 include the effect of a gradual seasonal decline observed for gases such as ethane, propane, and ethyne associated with increasing spring concentrations of OH radicals.

Cape Grim reported very low variable ethane values. They have already identified that this was caused by cryogenic trapping problems, which affect the retention of ethane and its reproducibility. Inefficient cryo-trapping may also account for the low ethyne values. The relatively high ethene mixing ratios may be associated with growth of alkenes in the sampling canisters, which is a common problem encountered with this technique. The Cape Grim measurements were preliminary results from their newly reinstated NMHC sampling program which is still making improvements to their sampling methodology, standard preparation, and analytical methods.

The MIT group on the Discoverer also had problems cryo-trapping the lightest hydrocarbons so only report values for C3 and above. Median values for these gases were higher and the variability was greater than for the UCI results. No discernible seasonal concentration trend was observed. The high mixing ratios for the reactive C4 NMHCs, particularly 2-methyl-propene, were the cause for most concern to the group as even very low levels of these gases can significantly perturb photochemical calculations for the MBL. It was suggested that blank tests on the analytical system employed during ACE-1 may be helpful. Members of the group stated that they had previously found 2-methyl-propene to be common contaminant associated with various moisture and other traps such as those employed by the MIT group.

Laboratory comparisons between the MIT and UCI groups have already started as part of the NOMHICE intercomparison program. The lab-based phases that have been completed so far do not fully account for the observed differences between the two groups. A further phase still in the planning stage will comprise side-by-side field comparison which may help to clarify the situation.

The UCI group have performed a wide range of tests on their sampling and analytical system and implemented various procedures to minimize canister contamination in order to make high precision measurements in the clean remote troposphere. However, this information needs to be reported in a comprehensive analytical paper. Barkley Sive has undertaken to write such a paper in the near future but it will not be submitted as part of the ACE-1 special issue.

Gary Kleiman has undertaken to go back and perform a series of test on the analytical system that was used during the ACE-1 cruise. He will also look at the consistency between the air and water measurements made during that period.