----------------------------------------------------------------- Readme file to accompany COBRA 2003 flask aircraft data v. 080214 ----------------------------------------------------------------- TERMS AND CONDITIONS (Adapted from NOAA and NACP data policy) ------------------------------------------------------------ Use of these data in any part implies an agreement on the part of the user that individuals and/or institutions responsible for contributing to data sets used must be specifically cited in addition to a general citation of the NACP greenhouse gas database. The COBRA 2003 flask aircraft data set should be cited as follows: 1) Kort, E. A., J. Eluszkiewicz, B. B. Stephens, J. B. Miller, C. Gerbig, T. Nehrkorn, B. C. Daube, J. O. Kaplan, S. Houweling, and S. C. Wofsy (2008), Emissions of CH4 and N2O over the United States and Canada based on a receptor-oriented modeling framework and COBRA-NA atmospheric observations, Geophys. Res. Lett., 35, L18808, doi:10.1029/2008GL034031. 2) Miller, S. M., D. M. Matross, A. E. Andrews, D. B. Millet, M. Longo, E. W. Gottlieb, A. I. Hirsch, C. Gerbig, J. C. Lin, B. C. Daube, R. C. Hudman, P. L. S. Dias, V. Y. Chow, and S. C. Wofsy, Sources of carbon monoxide and formaldehyde in North America determined from high-resolution atmospheric data, Atmos. Chem. Phys. Discuss., 8, 11395-11451, 2008. NACP investigators will include an acknowledgement in each publication or presentation arising from participation in NACP. The wording shall be similar to the following: "This study was part of the North American Carbon Program." Data providers and funding agencies may request additional acknowledgements. Upon publication of results, investigators should send the NACP Office an electronic copy of the publication. USE OF DATA These data are made freely available to the public and the scientific community in the belief that their wide dissemination will lead to a greater understanding and new scientific insights. The availability of these data does not constitute publication of the data. We rely on the ethics and integrity of the user to assure that the source(s) receive fair credit for their work. If the data are obtained for potential use in a publication or presentation, the source(s) should be informed at the outset of the nature of this work. If the source's data are essential to the work, or if an important result or conclusion depends on their data, co-authorship may be appropriate. This should be discussed at an early stage in the work. Manuscripts using the source's data should be sent to the source(s) for review before they are submitted for publication so we can ensure that the quality and limitations of the data are accurately represented. RECIPROCITY AGREEMENT Use of these data implies an agreement to reciprocate. Laboratories making similar measurements agree to make their own data available to the general public and to the scientific community in an equally complete and easily accessible form. Modelers are encouraged to make available to the community, upon request, their own tools used in the interpretation of the source data, namely well documented model code, transport fields, and additional information necessary for other scientists to repeat the work and to run modified versions. Model availability includes collaborative support for new users of the models. COBRA 2003 PROJECT PURPOSE & DESCRIPTION ---------------------------------------- The program was coordinated with the Fluxnet Canada Research Network (FCRN) and Environment Canada. The PIs for the project are: Steven C. Wofsy, Harvard University, swofsy@fas.harvard.edu James W. Elkins, NOAA, James.W.Elkins@noaa.gov Britton B. Stephens, NCAR/UCAR , stephens@ucar.edu John B. Miller, NOAA, John.B.Miller@noaa.gov The Terrestrial Ecology Program (director, Dr. Diane E. Wickland) of NASA's Earth Science Enterprise provided the principal funding for this airborne campaign, with additional resources from NOAA-CMDL and NOAA-OGP, NASA's Upper Atmosphere program, and from NSF (Atmospheric Chemistry Program and NCAR). The objective of COBRA 2003 is to test concepts for observations and data assimilation to determine the large scale sources and sinks of greenhouse gases and ozone-destroying gases from North America. The study will cover spatial scales from regional to continental. Gases of interest include CO2, CO, CH4, N2O, chlorofluorcarbons (CFCs), chlorinated solvents (CH3CCl3, CCl4 ), SF6, ozone (O3), and molecular hydrogen (H2). The capability to measure current and future sources and sinks of greenhouse gases and ozone-destroying chemicals represents a challenging scientific problem with important societal consequences. In order to predict future climate forcing and to project reocovery of stratospheric ozone from current levels of depletion, we must be able to determine the location and magnitude of emissions and understand them. But current models that simulate emissions of greenhouse gases do not represent current observations of concentrations across latitudes or vertical profiles of these gases within a factor of two. Transport of these gases into the middle and upper atmosphere is uncertain because convection, land ocean exchange, and other atmospheric processes are very difficult to model. The thickness of the atmospheric boundary layer decreases from daytime to nighttime, and transport processes also change seasonally. Diurnal and seasonal variations of atmospheric dynamics are correlated with variation of surface emissions, especially for CO2, making observations of trace gases very difficult to interpret. Previous studies have pointed out a large terrestrial sink for CO2 exists in North America. Systematic vertical profile data of the trace gases between 30 and 10,000 meters are required to critically test the models used in these analyses. A series of atmospheric experiments called the CO2 Budget and Regional Airborne Study with recent emphasis on North America is intended (COBRA-NA 2003) to take first steps in this direction. The experiment is anchored upon the extensive ground-based data from the National Oceanic and Atmospheric Administration/Climate Monitoring and Diagnostics Laboratory (NOAA/CMDL) Cooperative Station Networks for carbon cycle and halocarbon trace gases. The experiment will focus also on the extensive emission data at Harvard Forest, Massachusetts over the past 10 years by flying across the pollution plume from the New York City and Washington, D. C. metropolitan corridor. The intention is to estimate emissions using a data assimilation approach as well as by using a carbon monoxide (CO) as a reference for emission strength combined with the trace gas-CO correlations observed in COBRA-2003. This experiment is a continuation of an earlier COBRA experiment over a more limited region of North America in 2000. In that study significant deficiencies were discovered in the resolution and mass conservation properties of many existing data sets for analyzed meteorological fields. These issues currently limit the capability to accommodate the high degree of variability of local CO2 emissions (patchiness) and to use the information about sources contained in observed atmospheric variations. The current measurements will be analyzed using the advanced regional-scale meteorological model from scientists at the University of Sao Paulo (Brazil), who will collaborate with the team to apply their BRAMS model (Brazilian development of the Regional Atmospheric Mesoscale Modeling System (RAMS) model) for atmospheric dynamics to address the deficiencies of current meteorological models. The aircraft platform used in COBRA-NA 2003 is the University of North Dakota Cessna Citation II, a modified executive jet for atmospheric research (see picture above). The airborne experiment consists of a racetrack from Boulder to the west coast of the United States (Eureka, California) North to Campbell River British Columbia, across Canada to the east coast of North America (Yarmouth, NS), south to Portsmouth, NH, and back to Boulder (see map). Vertical profiles have been acquired over NOAA/CMDL cooperative stations located at Niwot Ridge, Colorado, Moody, Texas, Trinidad Head, and Harvard Forest. Other NOAA/CMDL stations and AmeriFlux and Fluxnet Canada sites along and outside the racetrack will provide important data on the boundary conditions for the flow of air through the area of the racetrack and for surface exchange fluxes. There have also been also excursions over the Pacific Ocean from California, and the Atlantic Ocean from Pease, New Hampshire to Sable Island to examine coastal influence of marine air and the mixing processes between maritime and continental air, and intensive regional flights in Oklahoma and New England. The two bases of operations will be one at the Jefferson County Airport (Jeffco) in Broomfield, Colorado, and the other at Pease (Portsmouth), New Hampshire. Test flights started on May 22, 2003 at Jeffco Airport and the first racetrack started there on May 26. The second racetrack started from Jeffco on June 19 and until June 28. Measurements of CO2, CO, H2O, aerosol size and concentration were measured at 1 Hz on the aircraft by scientists from Harvard University and University of North Dakota. Scientists from the Halocarbons and other Atmospheric Trace Species Group (HATS) group in NOAA/CMDL operated an in situ gas chromatograph (GC) . The HATS airborne GC, Airborne Chromatograph for Atmospheric Trace Species (ACATS-IV), measures twelve different trace gases: nitrous oxide (N2O), SF6, CFC-12 (CCl2F2), and halon-1211 (CBrClF2), once every 70 seconds; and hydrogen (H2), CO, methane (CH4), CFC-11 (CCl3F), CFC-113 (CClF2-CCl2F), chloroform (CHCl3), methyl chloroform (CH3CCl3), and carbon tetrachloride (CCl4) once every 140 seconds. A whole air sampler (WAS) for flasks was operated by researchers from the Cooperative Institute for Research in the Environmental Sciences (Univ. of Colorado), the National Center for Atmospheric Research, and Scripps Institution for Oceanography. The WAS collects air in glass flasks to be measured for a suite of trace gases at Scripps, NOAA/CMDL, and CU/INSTAAR. These include CO2, CO, H2, CH4, N2O, and SF6 concentrations, and ratios of O2/N2, Ar/N2, and isotopic C13/C12 and O18/O16 in CO2, and C13/C12 in CH4. GENERAL ------- Two types of flasks were filled on the MEDUSA system. This system fills 1.5 L glass flasks at a flow rate of ~ 3.3 SLPM flow and controlled pressure, usually set to 1 atm. The air is dried using a two-stage dry-ice trap. SIO Flasks (4-digit flask IDs beginning with '1') These flasks were analyzed at Scripps using a LiCor CO2 analyzer and a Mass Spec that measured O2/N2, Ar/N2, and CO2/N2. They were first run in a PURGE mode that maintained a constant pressure in the flask by replacing the sample air with purge gas. They were then analyzed in a SNIFF mode where a small amount of air was pulled from the flask. By comparing the SNIFF and PURGE CO2 concentrations to the known CO2 value in the purge gas, a DILUTION factor was calculated. These flasks were then analyzed at GMD for CO2, CO, CH4, N2O, H2, and SF6 and at INSTAAR for 13CO2/12CO2 and C18O16O/C16O16O. These values were then adjusted using the DILUTION factor calculated at Scripps and the known values for these species in the purge gas, using: Back-calculated Value = [ Measured - Purge Gas * ( 1 - DILUTION ) ] / DILUTION The SIO flasks have viton o-rings which lead to large positive offsets in the CO measurements and these flasks have been flagged by NOAA. For CO plotting and analyses, only use GMD flask data. For H2, in COBRA-2000 we observed a 21 +/- 10 ppb offset between Scripps and GMD PFPs that suggested a potential viton effect, however storage tests at GMD in the 80s showed no viton effect on H2 (though not all vitons are the same). The H2 values in the Scripps flasks have not been flagged, but caution should be exercised in interpreting them. For other GMD/INSTAAR species (CO2, CH4, N2O, SF6, and 13C) measured in both Scripps flasks and PFPs during the 2000 campaign there was no evidence of bias in the Scripps flasks (18O also looked good, but because it is unreliable in PFPs, no comparison could be made). The difference between Scripps-measured CO2 and back- calculated CO2 from the GMD measurements was +0.06 +/- 0.12 ppm, comparable to the uncertainty in the individual analyses themselves. NOAA GMD Flasks (4-digit flask IDs beginning with '9') These flasks were not analyzed at Scripps, and thus have DILUTION values of 1. They were analyzed at GMD for CO2, CO, CH4, N2O, H2, and SF6 and at INSTAAR for 13CO2/12CO2 and C18O16O/C16O16O. From 030525 through 030619, the flasks were filled to a nominal pressure of 760 torr. On June 20, in Edmonton, the rotor in the MEDUSA Valco valve was changed from Type E material to Type M material. The new rotor also had a smaller bore. As a result, the flasks were filled to a nominal pressure of 900 torr from 030621 to 030628. This change in fill pressure does not appear to have affected the measurements. On several early flights, the dry ice trap plugged, resulting in very low flask pressures. Also, during several later flights a failing pressure transducer led to variable flask pressures. For the flasks with pressures that could not be analyzed at Scripps, some were sent on to GMD and INSTAAR for analysis, while others were refilled later in the campaign. The continuous in situ data was merged with the flask data using a weighted average, where the weighting is determined by a lag time and an e-folding mixing time to account for the mixing of air in the flask. Currently the lag time is fixed at 3 seconds and the mixing time for each flask is calculated from the flask fill pressure and flow rate. These mixing times range from 22-27 seconds. The difference between Scripps-measured CO2 and the merged Harvard CO2 values was +0.46 +/- 0.81 ppm. The difference between GMD-measured CO (in GMD flasks only) and the merged Harvard CO values was +5.2 +/- 7.8 ppb. VARIABLE NAMES -------------- YYYYMMDD : Year, month, and day of sampling. doy : Absolute day of year UTC : Coordinated Universal Time, in seconds from midnight FLAGS : General flag; see Flags section below. flask.ID : Flask ID indicating either SIO flask (begins with 1) or NOAA flask (begins with 9) sample.order : Sample fill order valve.time : UTC time of flask closure in seconds since the midnight before the plane took off. Due to 22-27 second flask integration, this is not the representative time of the sample. For plotting and analyses, use the "UTC" variable above. flt.num : Flight number, ranges between 1 and 3 profile.num : Daily vertical profile number, ranges from 1 to 13 latitude : Latitude, in degrees longitude : Longitude, in degrees altitude : Altitude from the POS, in meters press.altitude : Pressure altitude, in meters rad.altitude : Radar altitude, in meters roll : Aircraft roll angle, in degrees pitch : Aircraft pitch angle, in degrees heading : Aircraft heading angle, in degrees alpha.angle : Alpha (attack) angle, in degrees beta.angle : Beta (slideslip) angle, in degrees track : Aircraft track angle from POS, in degrees counter : Condensation nuclei counter (V) airT.C : Air temperature, in degrees celsius dew.pt_EGG : Dew point from EGG, in degrees celsius dew.pt_TDL : Dew point from TDL (Laser Hygrometer), in degrees celsius frost.pt_TDL : Frost point from TDL (Laser Hygrometer), in degrees celsius THETA : Potential temperature, in Kelvin THETAe : Equivalent potential temperature, in Kelvin static.press : Static pressure, in mbar cabin.press : Cabin pressure, in mbar pitot.press : Pitot pressure from nose probe, in mbar ground.speed : Aircraft ground speed from POS, in m/s air.speed : True air speed, in m/s wind.m : Horizontal wind speed based on nose probe, in m/s wind.d : Horizontal wind direction from nose probe (true direction from which it blows) , in degrees wind.u : u-component of wind vector (positive North, negative South) wind.v : v-component of wind vector (positive East, negative West) pushgas.ID : flag to indicate which of two cylinders was used in purge analysis dilution : dilution factor calculated for purge analysis APO : Atmospheric Potential Oxygen in per meg calculated as O2N2P + 1.1 / 0.2095 * ( CO2P - 363.29 ) tau : e-folding mixing time used to average in-situ data O2.flow : O2 flow rate (V) O2.press1 : O2 pressure one (V) O2.press2 : O2 pressure two (V) GMDvSNIFF : difference between CO2_GMD and CO2S BCvPURGE : difference between CO2_NOAA and CO2P PCTHUCO2 : Percent of weighted averaging period in which Harvard CO2 data is available SIOvHUCO2 : CO2P - CO2_LICOR GMDvHUCO : CO_NOAA - CO_VUV GMDvHUCO2 : CO2_NOAA - CO2_LICOR ArN2P : Ar/N2 ratio in per meg from Scripps purge analysis ArN2S : Ar/N2 ratio in per meg from Scripps sniff analysis C13_INST : d13C in CO2 in per mil as measured at INSTAAR C13_VPDB : back-calculated INSTAAR d13C in CO2 in per mil on VPDB scale CH4_GMD : Methane concentration in ppb as measured at GMD CH4_GMD.flag : Methane data flag (see DATA FLAGS section below for more information) CH4_NOAA : back-calculated GMD CCH4 concentration in ppb on NOAA2004 scale CO_GMD : Carbon monoxide concentration in ppb as measured at GMD CO_GMD.flag : Carbon monoxide data flag (see DATA FLAGS section below for more information) CO_NOAA : back-calculated GMD carbon monoxide concentration in ppb on NOAA2000 scale CO_VUV : Carbon monoxide mixing ratio from Harvard VUV instrument, in ppb; in-situ reading taken at time of flask collection CO_VUV.10sec : Carbon monoxide mixing ratio, time averaged over 10 seconds (moving block average), in ppb CO2_GMD : Carbon dioxide concentration in ppm as measured at GMD CO2_GMD.flag : Carbon dioxide data flag (see DATA FLAGS section below for more information) CO2_LICOR : Carbon dioxide from Harvard CO2 instrument, in ppm; in-situ reading taken at time of flask collection. CO2_NOAA : back-calculated GMD CO2 concentration in ppm on NOAA2007 scale CO2P : Carbon dioxide concentration in ppm from Scripps purge analysis CO2S : Carbon dioxide concentration in ppm from Scripps sniff analysis H2_GMD : Hydrogen concentration in ppb as measured at GMD H2_GMD.flag : Hydrogen data flag (see DATA FLAGS section below for more information) H2_NOAA : back-calculated GMD H2 concentration in ppb on NOAA1996 scale H2O_EGG : Mixing Ratio of H2O from EG&G, in g/kg H2O_TDL : Mixing Ratio by weight from TDL (Laser Hygrometer), in ppmw N2O_GMD : Nitrous oxide concentration in ppb as measured at GMD N2O_GMD.flag : Nitrous oxide data flag (see DATA FLAGS section below for more information) N2O_NOAA : back-calculated GMD N2O concentration in ppb on NOAA2006 scale O18_INST : d18O in CO2 in per mil as measured at INSTAAR O18_VPDB : back-calculated INSTAAR d18O in CO2 in per mil on VPDB-CO2 scale O2N2P : O2/N2 ratio in per meg from Scripps purge analysis O2N2S : O2/N2 ratio in per meg from Scripps sniff analysis O3_2B : Ozone from 2B monitor, in ppb; in-situ reading taken at time of flask collection SF6_GMD : Sulfur hexafluoride concentration in ppt as measured at GMD SF6_GMD.flag : Sulfur hexafluoride data flag (see DATA FLAGS section below for more information) SF6_NOAA : back-calculated GMD SF6 concentration in ppt on NOAA2000 scale MISSING VALUES -------------- For files in .RData format, missing values are represented by "NA". For files in .csv format, missing values are represented by "NaN" DATA FLAGS ---------- Two sets of flags have been applied to the data, general flags and GMD flags. 1) General flags (in column titled FLAGS) (counts in parentheses) 0: No known issues with data (257) 1: Soft flag - at Scripps, Ar/N2 during purge on the mass spec or CO2 during sniff on the LiCor had a high standard deviation (34) 2: Bad GMD and INSTAAR data - error in SIO sniff, GMD analysis, or leak during transit (11) 3: Bad GMD data - flasks run at GMD on day when MAGICC Valco valve was failing (31) 2) GMD flags (in columns titled "species.flag") NOAA ESRL uses a 3-column quality control flag where each column is defined as follows: column 1 REJECTION flag. An alphanumeric other than a period (.) in the FIRST column indicates a sample with obvious problems during collection or analysis. This measurement should not be interpreted. column 2 SELECTION flag. An alphanumeric other than a period (.) in the SECOND column indicates a sample that is likely valid but does not meet selection criteria determined by the goals of a particular investigation. column 3 COMMENT flag. An alphanumeric other than a period (.) in the THIRD column provides additional information about the collection or analysis of the sample. For plotting and analysis of GMD data, use all except flasks with a general flag of 2 or 3, OR with a non "." character in the first position of the GMD flag. For plotting and analysis of INSTAAR data, use all except flasks with a general flag of 2.