Readme file for the final hydrocarbon dataset, Blodgett Forest 1998, ponderosa pine plantation hc98.1 & .2 VERSION 1, finished July 14, 2000 by Gunnar Schade University of California at Berkeley Department of Environmental Science, Policy and Management Ecosystem Science Division 250 Hilgard Hall Berkeley, CA 94720-3110 ph.: 1-510-643-6449 Fax: 1-510-643-5098 e-mail: gws@nature.Berkeley.edu this file written/modified: July 14, 2000 ---------------------- General description: This dataset contains over 20 compounds, analyzed from 10th July 1998 to 30th October 1998 at the Goldstein group Blodgett Forest research tower site. The site is described in Lamanna and Goldstein [JGR, 1999], and Goldstein et al. [Agr. For. Meteorol., 2000]. Measurements have been performed on an hourly basis with only little downtime. The principles of the hydrocarbon instrument are described in Lamanna and Goldstein [JGR, 1999], and modifications for the 1998 measurements are described in Schade et al. [GRL, 1999]; (additional manuscript in preparation). Mixing ratios were measured at two levels above the ponderosa pine plantation (gradient approach for flux measurement). ** This dataset contains mixing ratios measured at the lower level, ** approximately 2.5m above the average tree height (4m). During the measurement period, the trapping material was changed once at the beginning of August from being Carbopack B only (11.5 mg), to glassbeads (8 mg), Carbopack B (11 mg), and Carbosieve SIII (7 mg). Data before that date (3rd of August) are dubious for certain compounds as described below. Some have been set "NA". Chromatographic columns have been cut several times to remove degraded parts. We switched from DB-Wax (60m, 0.5um film, 0.32mm diameter) to Rtx-Wax columns (same but wide-bore: 0.53mm) in mid-September. That switch enabled the quantification of two anthropogenic VOCs, described below. All identification has been carried out by retention time and is therefore not 100% conclusive. Compounds which were not identified are listed as unknowns ("un") and are numbered. They will appear in order of retention. A "best guess" on what a special compound might be is given where appropriate. ** All chromatograms have been integrated manually for proper identification and quantification. Human error, however, is unavoidable. ** ** Zero mixing ratios were not set "NA". ** VERSION .1 contains all measurements in their original time line VERSION .2 is identical to .1 but with all data coerced into a continuous hourly time line ------ Individual compounds (calibrated compounds have name of Standard in paranthesis): 0. "time": Note that the time line is based on local summer time (PDT), and that it has to be shifted forward by one hour to match with local standard time (PST) 1. "nonpolars" (OVOC): Carbopack B did not trap these to a major extent. Data has been set "NA" for that time period. The peak consists most probably of methane, ethane, propane, butane (maybe pentane) and ethene (maybe propene). Precision: +-15% (from butane and pentane). --> significant Blank 2. "isoprene" (S1, OVOC): probably the most reliable compound measured (precision +-3%), stable throughout the whole dataset, minor Blank appearance, high accuracy. 3. "un4": an anthropogenic compound, too small in abundance to be relevant. However, it can be used as a tracer. Identified only after day 261. 4. "MTBE": This compound, that has the heptane retention time, was identified as MTBE with an authentic standard in 1999; its FID response was estimated theoretically from comparison with isoprene, as done for other uncalibrated compounds (see below). Mixing ratios should be viewed with some caution, and are probably an upper estimate because there may be other compound(s) eluting at the same retention time. However, MTBE is clearly of anthropogenic origin, and therefore a good tracer for anthropogenic emissions. It was not identified before changing the trapping materials (hardly separated from acetaldehyde). Precision: +-5%. 5. "acetaldehyde" (OVOC): having poor detector response and a background in the blank, mixing ratios should be treated with special caution; precision is +-100% untill the 3rd of August,+-25% in August and +-10% thereafter. --> significant Blank 6. "acetone" (S1, OVOC): mixing ratios are in doubt before August 3, due to poor results of the standard additions. As acetone eluted as a broad peak, the "noise" is higher than for compounds producing bell-shaped peaks. Precision is +-15%, improving towards October. 7. "butanal" (OVOC): small, but consistent peak; precision ~+-1% (1). 8. "MACR" (S1, OVOC): methacrolein, major isoprene oxidation product; second most precise compound measured consistent throughout October; precision +-5%. *** Important: Rather than the actually measured response factor, a theoretical response factor based one the relative FID response of MACR versus isoprene was used for calibration! Multiply all values by 1.2 to get the originally used MACR, MVK, and un7 values. *** 9. "un7": it is unclear whether this is an anthropogenic compound. It definitely shows some similarity to isoprene, and, due to its abundance, 3-methyl-furan can be considered a likely candidate. The response factor was taken to be equal to the theoretical one of MACR. Note: It turned out in 1999 that 3-methyl-furan elutes together with methanol and MEK on a similar analysis setup. 10. "MEK?" (OVOC): poor peak shape and coelution with methanol make this compound very difficult to quantify; attempts to use these results is unadvisable though the data set has been cleaned from too unreasonable values! Precision: +-7% (1). --> varying blank from methanol 11. "methanol" (OVOC): unreliable due to incomplete desorption and appearance in the Blanks. Therefore, mixing ratios appear much too low. Values left in the data set are those that correlate with acetone, as appropriate in comparison with 1999 results. Methanol mixing ratio variation may be considered qualitatively correct. 12. "MVK": methyl-vinyl-ketone, major isoprene oxidation product; "noise" comes from interference with the double peak before and increases towards the end of the measurement period. The compound is not calibrated but FID- response is assumed to equal that of MACR. Unequivocally identified with an authentic sample. 13. "pentanal" (OVOC): small, but consistent peak(s) (coelution with pentanone); integrated and calibrated as one peak; precision +-4% (1). 14. "alpha-pinene" (S2): local emission compound; mixing ratios are in doubt before August 3, due to possible breakthrough. There is a possibility of its production from beta-pinene on the microtraps during sampling or desorption (Jim Greenberg, NCAR, pers. comm.). Precision: +-8%. 15. "toluene" (Scotty): small, but consistent peak; most reliable, calibrated anthropogenic compound, precision +-6% (1). Contamination has mostly been removed but may still be visible in August. A mean mixing ratio of 10 ppt has been substracted from all data due to a varying blank of 5 to 20 ppt (range). 16. "MBO" (MBO): methylbutenol, dominant local emission compound; very "noisy", due to the bad peak shape (fronting + tailing); however, stable calibration throughout the measurement period (precision: +-5%). 17. "hexanal" (OVOC): small, but consistent peak (dominant alkanal/aldehyde); precision +-2% (1). 18. "beta-pinene": local emission compound (dominant terpene); there is a possibility of its conversion to alpha-pinene on the microtraps during sampling or desorption (Jim Greenberg, NCAR, pers. comm.). Uncalibrated, but unequivocally identified with an authentic sample. 19. "ethylbenzene" (Scotty): too small in abundance to be relevant; consistent aromatic, however, after toluene; precision +-6% (1). 20. "p-xylene" (Scotty): too small in abundance to be relevant; precision +-6% (1). 21. "3-carene" (S2): local emission compound (highest precision among the terpenes, +-4%); unreliable untill 3rd of August due to possible breakthrough; coelutes with m-xylene, which, however, is neglibile. 22. "monoterpeneX": small, but consistent terpene, identified due to its similarity with the other terpenes during the rain events. Most probable candidate is myrcene (but could also be beta-phellandrene), both known terpene compounds/emissions in/from ponderosa pine trees. 23. "o-xylene" (Scotty): too small in abundance to be relevant; precision +-6% (1). --> minor blank 24. "d-limonene" (S2): local emission compound (minor terpene); highly reactive with ozone; unreliable untill 3rd of August due to possible breakthrough; precision +-14% (July throughout October). --> varying, but small blank (1) NOTE: Where precisions are very good (e.g. butanal), they are based more or less only on one calibration line and are therefore not representative for the whole measurement period!! -------- Standards (manufacturer's accuracy in parenthesis): S1 (Scott-Marrin): acetaldehyde (+-5%), MACR (+-10%), heptane (+-2%), alpha-pinene (+-2%), acetone (+-2%), isoprene (+-2%) balance: nitrogen S2 (Scott-Marrin): heptane, 3-carene, alpha-pinene, d-limonene (unanalyzed Standard; alpha-pinene and heptane are consistent with S1). balance: nitrogen MBO (Scott-Marrin): 2-methyl-3-buten-2-ol (+-2%) balance: nitrogen OVOC (Scott-Marrin; mixing ratios assigned by NCAR and highly uncertain for the aldehydes and alcohols): butane, isoprene, acetaldehyde, acetone, butanal, MACR, MEK, methanol, benzene, ethanol, pentanal, 2-pentanone, hexanal, n-butanol balance: nitrogen Scotty (Scott Specialty Gases small tank): benzene, toluene, ethylbenzene, p-, m-, and o-xylene (all +-10%) balance: nitrogen -------- References: In-situ measurements of C2-C10 VOCs and OVOCs above a Sierra Nevada Ponderosa Pine Plantation, Mark S. Lammana and Allen H. Goldstein, Journal of Geophysical Research - Atmospheres, 104, 21247-21262, 1999 Effects of climate variability on the carbon dioxide, water, and sensible heat fluxes above a ponderosa pine plantation in the Sierra Nevada (CA), A.H. Goldstein, N.E. Hultman, J.M. Frachebaud, M.R. Bauer, J.A. Panek, M. Xu, A. Guenther, and W. Baugh, Agricultural and Forest Meteorology, 101, 113-129, 2000 Are monoterpene emissions influenced by humidity? Gunnar W. Schade, Allen H. Goldstein, and Mark S. Lamanna Geophysical Research Letters, 26(14), 2187-2190, 1999