Search

The Global Greenhouse Gas Reference Network Aircraft Program at NOAA has sampled ozone and other atmospheric trace constituents in North America for over a decade (2005-present). The method to derive tropospheric ozone climatology from the light aircraft measurements equipped with the 2B Technology instruments is described in this paper. Since ozone instruments at most of aircraft locations are flown once a month, this raises the question of whether the sampling frequency allows for deriving a climatology that can adequately represent ozone seasonal and vertical variability over various locations. Here we interpret the representativeness of the tropospheric ozone climatology derived from these under-sampled observations using hindcast simulations conducted with the Geophysical Fluid Dynamics Laboratory chemistry-climate model (GFDL-AM3). We first focus on ozone measurements from monthly aircraft profiles over the Front Range of Colorado and weekly ozonesondes launched in Boulder, Colorado. The climatology is presented as monthly values separated in 5th, 25th, 50th, 75th, 95th percentiles, and averaged at three vertical layers: lower (1.6–3 km), middle (3–6 km), and upper (6–8 km) troposphere. The aircraft-based climatology is compared to the climatology derived from the nearest located ozonesondes launched from Boulder, Colorado, from GFDL-AM3 co-sampled in time with in-situ observations, and from GFDL-AM3 continuous 3-h samples. Based on these analyses, we recommend the sampling frequency to obtain adequate representation of ozone climatology in the free troposphere. The 3-h sampled AM3 model is used as a benchmark reference for the under-sampled time series. We find that the minimal number of soundings required per month for the all altitude bins (1.6–3, 3–6, and 6–8 km) to sufficiently match the 95% confidence level of the fully sampled monthly ozone means vary between 3 and 5 sounding per month, except in August with a minimum of 6 soundings per month. The middle altitude bin required the least number of samplings per month. We determine the reasonably good agreement between the ozonesondes and aircraft measurements near Boulder suggest that valuable climatologies could be developed from the aircraft sites where no ozonesondes exist even though the aircraft measurements are more limited in number than the ozonesondes. When averaged over a number of years the aircraft data provide valuable information. More frequent sampling could tell us more but the measurements given would indicate that they can provide interesting climatological results.