NCAR/NSF C-130 Chemistry SO2 and DMS Data

Dataset Contact Information:
Alan Bandy
Dept. of Chemistry
Philadelphia, Pa, 19104
Telephone: (215) 895-2640
Facsimile: (215) 895-1980

Drexel University Sulfur Dioxide and Dimethyl Sulfide data for RICO

Two atmospheric pressure ionization mass spectrometers (APIMS) were deployed on the NCAR C-130 for sampling sulfur dioxide (SO2) and dimethyl sulfide (DMS). Both instruments were located in the middle of the aircraft and use independent sampling lines through the aircraft belly port just behind the same plate where the radiometers were located. The inlet tubing was 0.50 inch od FEP Teflon, which were heated above ambient temperatures below the cabin floor. Subsampled air was passed through a Nafion dryer before entering the mass spectrometers. Each instrument had isotopically labeled internal standard continuously added to the air sampling manifold. A further description of the instruments and their performance have been published (Thornton, Donald C., Alan R. Bandy, Fang H. Tu, Byron W. Blomquist, Glenn M. Mitchell, Wolfgang Nadler, Donald H. Lenschow, Fast airborne sulfur dioxide measurements by atmospheric pressure ionization mass spectrometry (APIMS), J. Geophys. Res., 107, 4632, doi:10.1029/2002JD002289R, 2002; Bandy, Alan R., Donald C. Thornton, Fang H. Tu, Byron W. Blomquist, Wolfgang Nadler, Glenn M. Mitchell, Donald H. Lenschow, Determination of the vertical flux of dimethyl sulfide by eddy correlation and atmospheric pressure ionization mass spectrometry (APIMS), J. Geophys. Res., 107, 4743, doi:10.1029/ 2002JD002472R, 2002.).


Sulfur Dioxide
The sampling rate for the APIMS was set for 20 milliseconds (ms) with equal time spent sampling the ambient SO2 ions and the isotopically labeled internal standard ions. The ambient SO2 concentrations were computed point wise using internal standard concentration and flow rate data also sampled at the start of each sample period. This produced an ambient SO2 high rate (HRT) data set of 25 samples per second (sps). The SO2 data low rate (LRT) dataset contain 1 second integration of the HRT data.

There are no SO2 data for C-130 flights RF01-RF03 because of contamination of the air sampling inlet and Nafion dryer during the landing, and possibly on the ground, on the arrival at Antigua in early December 2004. Flights RF04 to RF07 were impaired by low efficiency in the Nafion dryer. Data were salvaged for RF04 and RF06. For RF08 no data were obtained due to a total failure of the dryer under the high liquid water content conditions. Between RF08 and RF09 the sampling configuration was modified to improve the Nafion dryer operation. The data for RF09 to RF19 were of high quality. The HRT data contained an undetermined source of high frequency variation that can be greatly reduced with a simple 3 point moving average filter. Blank corrections for both the ambient and internal standard ion signals were determined using a copper tube as an annular denuder for the subsampled flow going to the mass spectrometer. The 1/e response time for the denuder was 0.5 second. Based on the response of the internal standard the denuder was >99% effective.


Dimethyl Sulfide
The sampling rate for the APIMS was set for 20 milliseconds (ms) with equal time spent sampling the ambient DMS ions and the isotopically labeled internal standard ions. The ambient DMS concentrations were computed point wise using internal standard concentration and flow rate data also sampled at the start of each sample period. Because of undetermined noise source the data were filtered with a 25 point moving average. This produced an ambient DMS low rate (LRT) dataset with a 1 second integration time.

An undetermined source of background signal for both the ambient ion and the internal ion channels for DMA were encountered. In previous field sampling this background did not occur. The problem was treated by using ultrapure air to exclude ambient air from being sampled to determine the blank for the ambient ion channel. Background for the internal standard was determined by intermittently turning off the internal standard addition to the air sampling manifold. The background signals were pressure altitude dependent so that frequent blank determinations were performed at as many constant flight altitudes as possible. An algorithm was developed for each flight to fit the blank corrections as a function of pressure altitude. Consequently, the corrections for DMS during vertical profiles may not be as accurate as for the level flight legs.