[Firex-aq] FIREX-AQ Trajectory-based smoke ages

[Christopher Holmes]

All,

Smoke ages derived from trajectory analysis are now available in the FIREX-AQ data archive at Langley (data ID firexaq-FSU-smokeage). They are available as 1-second ICARTT files, so that they can be included in merge files. Here is some information that will be useful to users. Ages are provided for all western fires and selected eastern fires that produced plumes that were sampled in multiple downwind transects. Ages for small eastern fires are still coming.

The ages account for vertical plume rise and advection. The advection component is from back trajectories (HYSPLIT) using three different meteorological fields (HRRR 3km, NAM CONUS nest 3km, GFS 0.25°). The vertical rise time is based on the height of trajectory above the surface at the point where it most closely approaches the source fire, assuming a mean vertical rise speed (12 m/s for pyroCb, 7 m/s otherwise).  The variables that I expect to be most widely used are
·         smoke_age: The mean of ages from the meteorological fields (without wind-speed correction described next). Ages from all three meteorological models are averaged, except in cases where a model predicted transport that was inconsistent with the actual conditions. The smoke_agemethod variable indicates which fields were used at each point in time.
·         smoke_age_corr: same as smoke_age, except that the advection component of the ages are corrected or scaled so that the model wind speed matches the wind speed observed by the aircraft at the time of smoke measurement.
·         smoke_age_unc: uncertainty in the smoke age due to: spread in age estimates between the three meteorological models, discrepancy between observed and modeled wind speeds, distance by which the trajectory misses the fire source, multiple possible sources of the smoke (selected fires with large area or multiple fire fronts), and range of likely updraft speeds.
·         For anyone interested in seeing the trajectories or disaggregating the age calculation into component parts and their uncertainties, I can provide more detailed files.

Here are some key differences between the trajectory-based age estimates and those based on mean wind and provided in Shuka’s fire flags.
·         Trajectories provide age estimates for some smoke that has no reported ages in the fire flags (e.g. aged Williams Flats, aged Tucker, Sheep fires).
·         Trajectory ages account for wind speed variations in space and time during a flight, to the extent that they are captured by the meteorological models. The ages in fire flags assume a uniform wind field that varies with altitude (derived from measurements aboard the aircraft). Variations in wind speed can be important, particularly during the day-to-night transition and in the presence of terrain.
·         Trajectory-based smoke ages can vary across a transverse plume transect, while mean wind-based ages do not. Both trajectory-based and mean wind-based ages vary along longitudinal plume transects.
·         Despite sometimes large differences in estimated ages, the trajectory and mean-wind ages are strongly correlated overall across the campaign (R^2 ≈ 0.8).

Let me know if you have any further questions about using this data.

Best,
Chris




Christopher D. Holmes, PhD

Werner A. and Shirley B. Baum Associate Professor of Meteorology
Earth, Ocean and Atmospheric Science

Florida State University



EOAS Room 6035
phone: 850-645-0972

https://acgc.eoas.fsu.edu<https://urldefense.proofpoint.com/v2/url?u=https-3A__acgc.eoas.fsu.edu&d=DwMGaQ&c=WO-RGvefibhHBZq3fL85hQ&r=XTJBopAzwSsAgEljIk6g1TdJEZxb_bVLEjVNpBxfBIw&m=IFh98_0Q-hq9C6UPKlz1O2oYuFtabQ2LN0jxIH-X7AQ&s=d8jPLlmUtu_QrD1LcbyyE-FZh1aFEJlOVC9jA9pb4Ug&e=>



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://espo.nasa.gov/pipermail/firex-aq/attachments/20200829/41ab1030/attachment-0001.html>