Formaldehyde over the central Pacific during PEM‐Tropics B

Formaldehyde, CH 2 O, mixing ratios are reported for the central Pacific troposphere from a series of 41 flights, which took place in March‐April 1999 as part of the NASA Pacific Exploratory Mission (PEM) ‐Tropics B mission. Ambient CH 2 O was collected in aqueous media and quantified using an enzyme‐derivatization fluorescence technique. Primary calibration was performed using aqueous standards and known flow rates. Occasionally, CH 2 O gas standard additions to ambient air were performed as a secondary calibration. Analytical blanks were determined by replacing ambient air with pure air. The estimated precision was ±30 pptv and the estimated accuracy was the sum of ±30 parts per trillion by volume (pptv) ±15% of the measured value. Approximately 25% of the observations were less than the instrumental detection limit of 50 pptv, and 85% of these occurred above 6 km. CH 2 O mixing ratios decreased with altitude; for example, near the equator the median value in the lowest 2 km was 275 pptv, decreased to 150 pptv by 6 km and was below 100 pptv above 8 km. Between 130 and 170 W and below 1 km, a small variation of CH 2 O mixing ratio with latitude was noted as near‐surface median mixing ratios decreased near the equator (275 pptv) and were greater on either side (375 pptv). A marked decrease in near‐surface CH 2 O (200 pptv) was noted south of 23°S on two flights. Between 3° and 23°S, median CH 2 O mixing ratios were lower in the eastern tropical Pacific than in the western or central Pacific; nominal differences were >100 pptv near the surface to ∼100 pptv at midaltitude to ∼50 pptv at high altitude. Off the coast of Central America and Mexico, mixing ratios as high as 1200 pptv were observed in plumes that originated to the east over land. CH 2 O observations were consistently higher than the results from a point model constrained by other photochemical species and meteorological parameters. Regardless of latitude or longitude, agreement was best at altitudes above 4 km where the difference between measured and modeled CH 2 O medians was less than 50 pptv. Below 2 km the model median was approximately 150 pptv less than the measured median.