HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ngwabie, N. M. Articles by Hinz, T. Search for Related Content PubMed Articles by Ngwabie, N. M. Articles by Hinz, T. Agricola Articles by Ngwabie, N. M. Articles by Hinz, T. Related Collections Biogeochemical Processes Air Pollution Animal Waste

Abundances and Flux Estimates of Volatile Organic Compounds from a Dairy Cowshed in Germany

^a Inst. of Environmental Physics, Univ. of Bremen, Bremen, Germany
^b Dep. of Agricultural Biosystems and Technology, Swedish Univ. of Agricultural Sciences, Alnarp, Sweden
^c Dep. of Atmospheric Sciences, Texas A&M Univ., College Station, TX 77843
^d Max Planck Inst. for Chemistry, Dep. of Atmospheric Chemistry, Mainz, Germany
^e Federal Agricultural Research Centre, Braunschweig, Germany

^* Corresponding author (schade{at}ariel.met.tamu.edu).

Received for publication September 29, 2006.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Animal husbandry and manure treatment have been specifically documented as significant sources of methane, ammonia, nitrous oxide, and particulate matter. Although volatile organic compounds (VOCs) are also produced, much less information exists concerning their impact. We report on chemical ionization mass spectrometry and photo-acoustic spectroscopy measurements of mixing ratios of VOCs over a 2-wk measurement period in a large cowshed at the Federal Agricultural Research Centre (FAL) in Mariensee, Germany. The high time resolution of these measurements enables insight into the sources of the emissions in a typical livestock management setting. During feeding hours and solid manure removal, large mixing ratio spikes of several VOCs were observed and correlated with simultaneous methane, carbon dioxide, and ammonia level enhancements. The subsequent decay of cowshed concentration due to passive cowshed ventilation was used to model emission rates, which were dominated by ethanol and acetic acid, followed by methanol. Correlations of VOC mixing ratios with methane or ammonia were also used to calculate cowshed emission factors and to estimate potential nationwide VOC emissions from dairy cows. The results ranged from around 0.1 Gg carbon per year (1 Gg = 109 g) for nonanal and dimethylsulfide, several Gg carbon per year for volatile fatty acids and methanol, to over 10 Gg carbon per year of emitted ethanol. While some estimates were not consistent between the two extrapolation methods, the results indicate that animal husbandry VOC emissions are dominated by oxygenated compounds and may be a nationally but not globally significant emission to the atmosphere.

Abbreviations: DMS, dimethylsulfide • PTR-MS, proton transfer reaction mass spectrometry • TMA, trimethylamine • VFA, volatile fatty acid • VOC, volatile organic compound

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Volatile organic compounds (VOCs) play central roles in atmospheric chemistry (i) through their reactions with the OH radical, (ii) by indirect production of ozone following their photochemical oxidation, and (iii) through their role in the production of secondary organic aerosols (Andreae and Crutzen, 1997). To fully understand the effects of these VOCs in their various roles on global climate and chemistry-climate feedbacks requires detailed knowledge of their budgets. Though much work has been done identifying sources and sinks of selected VOCs (e.g., Galbally and Kirstine (2002) for methanol; Guenther et al. (1995) for isoprene), new sources and sinks are continuously discovered and budgets revised, sometimes significantly. The significance of emissions of CH4, N2O, NH3, and particulate matter by livestock and their excretions for the atmospheric budget have already been recognized (NRC-Ad Hoc Committee on Air Emissions from Animal Feeding Operations, 2003). Although a variety of other trace gases, such as VOCs, have been observed in livestock emissions, their relative contribution to the atmospheric budget is largely unknown.

Intensive animal farming has been widely adopted to meet the demands, such as for milk and meat, of an increasing human population. The largest of these facilities, called concentrated animal feeding operations (CAFOs), are arguably profit-driven enterprises. Though historically associated with rural farmland, they are now often located near urban areas. The increasing number and size of livestock facilities and their proximity to human settlements has lead to increasing odor complaints from neighbors. This has prompted measurements to identify the odorants and to work out possible reduction strategies (McGinn et al., 2003; Rabaud et al., 2003; Spinhirne and Koziel, 2003; Spinhirne et al., 2003; Spinhirne et al., 2004; Hobbs et al., 2004; Filipy et al., 2006). Important odorants are amines, sulfides, phenols, and volatile fatty acids (VFAs).

In addition to the unpleasant odor from these facilities, some of the emitted compounds or their reaction products are toxic or carcinogenic in nature (e.g., phenolic species). As such, they are subject to government regulation specifying threshold concentration limits above which it is unsafe for humans and animals to be exposed. Trace VOC monitoring may be needed in animal facilities to ensure that these limits are not exceeded, thereby putting human and animal health at risk.

We hypothesized that VOC emissions from animal husbandry included more than the previously identified odorants, and our objective was to carry out measurements inside animal buildings for quantification of exposure, and for flux estimation to provide data for future national emissions reporting requirements. Here, we discuss only the results from the dairy cowshed measurements.

	Materials and Methods

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
A selection of VOCs was measured using a proton transfer reaction mass spectrometer (PTR-MS) in different animal shelters (dairy cows, sheep, and pigs) of the Federal Agricultural Research Centre (FAL) in Niedersachsen, Germany. Based on correlations between the VOCs and methane or ammonia with documented annual fluxes for Germany (Daemmgen, 2004), the emission rates of several VOCs were estimated. The daily fluxes of some VOCs were modeled from the emission profiles in the cowshed from which annual fluxes were estimated, and the values compared to those obtained via the correlation method.

Experimental Sites
The major portion of measurements was made in a dairy cowshed at an FAL branch in Mariensee near Neustadt-Hannover, Germany. Preliminary experiments were conducted in a tie-stall barn with slatted concrete floors and slurry manure treatment containing up to 16 fistulated dairy cows at the FAL main location in Braunschweig, Germany. The larger barn in Mariensee was chosen based on previous experience from a larger EU project. It uses a typical three double-row tie-stall setting with solid manure treatment (straw litter) on sloped concrete floors. The straw/manure mixture was cleaned out twice a day (?5 am and 2 pm), after which fresh straw was provided. The main compartment of the stable was 40 x 22 x 4 m and housed 120 adult milk cows. It was ventilated through windows on the long sides and flow-adjustable, forced vents in the roof, with ventilation mostly passive during the winter months. Measurements were performed in December 2004 when the animals were permanently indoors. Weather conditions during the measurements, as acquired from a German weather service station 20 km to the southeast of Mariensee, included ambient temperatures between –2 and 8°C, and wind speeds between 2 and 10 m s^–1 generally blowing from southwesterly to northwesterly directions, resulting in perpendicular to parallel flows with the shed orientation, respectively. A calm situation was encountered during the third measurement day and high wind speeds during a frontal passage in the second week.

Animals and Diet
The shed's milk cows had an average weight of 650 kg in December 2004. They were fed and milked with a moving milker twice a day at the same times their manure was cleaned away. The cows received a typical corn- and grass-silage dominated diet of an average amount of 18.7 kg dry matter per cow per day. The fodder was composed of 33% corn silage, 30% grass silage, 15% high protein source (i.e., rape expeller), 13% pressed beet pulp, and 9% barley straw for crude fiber. An individual cow's daily milk production during December 2004 was on average 25 kg.

Experimental Setup
A schematic of the setup is shown in Fig. 1 . Air was pulled at a constant 12 L min^–1 either from inside or outside of the cowshed to the PTR-MS and photo-acoustic monitoring devices using a membrane pump. The source of the sample air was chosen in a pre-programmed way by switching a three-way, PFA Teflon solenoid valve controlled by the PTR-MS computer. Both sampling lines from inside and outside of the cowshed (0.95-cm OD PFA Teflon) were 20-m long (source to PTR-MS/photo-acoustic spectrometer), and air from inside the barn was sampled from a central location at a height of 3 m. This location was very close to one of the passive roof ventilators, and should generally be well-mixed air representative of the emissions composition of the shed as a whole. The sampling line for air from outside of the cowshed ran to a point half a meter above the roof of the shed. Although air from the roof of the shed is unavoidably influenced by emissions from the shed itself, this position represented the closest natural source of "clean" air that could be used for comparison with that in the shed itself. Both shed air and outside air sampling inlets were equipped with 1 to 2 µm Teflon PTFE filters that removed particles from the air stream before analysis for VOCs. The high flow rate of the air ensured that it spent a very short time in the tubing (?3 s) before chemical analysis. The PTR-MS measurement cycle was slightly less than 2 min with dwell times on individual masses between 0.5 and 2 s. Air was exchanged in the photo-acoustic spectrometer approximately every 120 s. Shed air was sampled for 50 min at the top of each hour while outside or reference air was sampled for the last 10 min of each hour, corresponding to five cycles. More time was given to the shed air so as to cover all short-term fluctuations in the shed. To complement air measurements, a temperature sensor was placed near the center of the shed at a height of 3 m and recorded temperatures at a 120-s time interval.

View larger version (20K): [in this window] [in a new window]   Fig. 1. Experimental setup for monitoring gaseous emissions in animal housing (PFA, perfluoroalkoxy; PTR-MS, proton transfer reaction mass spectrometer).

[1]

The resulting mixture of primary (H3O+) and newly protonated VOC product ions (VOCH+) is sampled through a pinhole into a quadrupole mass spectrometer. Protonated VOCs are separated according to their mass-to-charge ratio and sensitively detected using a secondary electron multiplier (SEM). In our experiments, the drift tube was operated at 2 mbar pressure, a drift voltage of 500 V, and a temperature of 50°C.

The VOC mixing ratios were inferred from SEM count rates either by addition of a calibration gas containing known mixing ratios of a variety of species, or were estimated based on knowledge of the ion-molecule reaction kinetics occurring in the ion drift tube. Dynamic dilutions of a µmol mol^–1 standard in nitrogen (Air Products GmbH, Hattingen, Germany) into ambient air were used to quantify methanol, acetaldehyde, ethanol, acetone, isoprene, and benzene. For those species found in the barn air but for which a calibration standard was unavailable, VOC drift-tube number densities were estimated using Eq. [2]. Mixing ratios were obtained by dividing this value by the total molecular number density in the drift tube, which is obtained through routine pressure measurements.

[2]

Here, t is the residence time of the ions in the drift tube in seconds, k is the ion/molecule reaction rate coefficient for a given ion/VOC pair in cm3 molecule^–1 s^–1, and values i(H3O+) and i(VOCH+) correspond to ion count rates corrected for their transmission through the mass spectrometer using values provided by the instrument manufacturer. Reaction times are calculated as described in Lindinger et al. (1998). Rate coefficients were obtained by the method of Su and Chesnavich (1982) using dipole moments and polarizabilities given by Zhao and Zhang (2004). Regardless of the method used to obtain mixing ratios, it is assumed that the product ion signal used for calculation is produced exclusively by the reaction between H3O+ and a unique VOC precursor. Calculations based on ion-molecule reaction kinetics generally assume that no fragmentation of the product ion occurs as a result of this reaction. For known fragmentations, fragment ion signals were included in the calculation as part of the product ion signal. When all assumptions concerning ion origin are satisfied, the overall accuracy of the measurement is dependent on the accuracy of the calculated reaction time, the rate coefficient, the ion signals themselves, and applied transmission factors, or in cases where calibration gas was available, on the accuracy of the gas standard measurement. When sampling an unknown mixture of VOCs in a cowshed, the verity of assumptions and the instrument accuracy cannot be rigorously assessed without independent measurement techniques, including ion signal background determinations. However, measurement precision or scatter of the PTR-MS technique is typically £30% (Lindinger et al., 1998) and changes with ion source conditions, magnitude of measured ion signals, and set dwell times for various mass measurements. In summary, reported mixing ratios should be regarded as upper limits, with higher accuracies in the case of calibrated species, such as methanol and ethanol.

Photo-acoustic Multigas Monitor
The concentrations of water vapor, methane, nitrous oxide, ammonia, and carbon dioxide were monitored at 2-min intervals with a photo-acoustic spectrometer (Brüel and Kjaer model 1302, INNOVA AirTech Instruments, Bellerup, Denmark). Working on the principle of infra-red absorption spectroscopy, the isolated air sample was irradiated by modulated light from a heated wire filament using a filter-selected wavelength for each species. The modulated pressure change due to absorption in the sample generated an acoustic signal, which was proportional to the concentration of the absorber in the cell. The same instrument has been used simultaneously with a Fourier transform infra-red spectrometer for measurements in animal shelters by Hinz and Linke (1998), with deviations in measured concentrations of ammonia and carbon dioxide within ±2.5% and detection limits of 0.15 and 3 µmol mol^–1, respectively. The factory calibration of the instrument, which included water-vapor interferences, was left unchanged for the experiments reported here. Recalibration of the instrument at the FAL in Braunschweig (Hinz and Linke, 1998) in 2005 showed no significant deviations, hence no corrections were made.

	Results and Discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Proton Transfer Reaction-Mass Spectrometry Data Volatile Organic Compound Assignment
A short series of experiments was conducted before the measurements in Mariensee to determine which chemicals might normally be found at elevated mixing ratios in a cowshed as compared to ambient air. Aside from the location and particular characteristics of the housing, the experimental setup for these measurements in Braunschweig was identical to that already described for Mariensee. To survey cowshed-affected chemicals, full mass scans (ranging from 20 to 210 amu) and a variety of selected ions in this same range were monitored for several days. A t test of means was performed comparing ion signals observed in shed air to those in ambient air using P < 0.05 as a cutoff to determine whether there was an observable difference between them. Those mass-to-charge ratios showing significant deviation between cowshed and outside air were targeted for more complete measurements in Mariensee where operations were more representative of the dairy industry as a whole.

The tentative association of ions showing up in cowshed-affected air with particular chemical species is an important point of discussion. As suggested earlier, while the PTR-MS instrument performs superbly for on-line monitoring, secondary techniques or hyphenation with gas chromatography are necessary for complete and unambiguous compound identification. In the absence of such instrumentation, assignments must be approached with some caution. Using gas chromatography and secondary measurement techniques, it has been shown that several ions associated with common VOCs are free from or have little interference from other ions when sampling air in the troposphere of rural and urban environments (de Gouw et al., 2003; Warneke et al., 2003). In making assignments for our measurements, we first assumed that the mixture of VOCs present in the animal housings was not substantially different from that of the urban or rural troposphere and that species previously reported to produce interference-free ions were likely also interference free in the cowshed. Previous reports that employed other techniques for compound identification in animal housings and that listed the discovered VOCs were also consulted for our chemical assignments (Hartung, 1998; Schiffman et al., 2001; Rabaud et al., 2002, 2003; McGinn et al., 2003; Spinhirne et al., 2003, 2004; Hobbs et al., 2004; Filipy et al., 2006). In some instances, the natural 13C abundance of an ion, measured at the m/z of the expected VOCH++1 can also provide insight into the precursors of an ion. For instance, several alcohols (CnH2n+2O) produce ions at the same m/z as their respective lower unsaturated organic acids (Cn-1H2(n-1)O2) but will produce a different 13C signature due to differing numbers of carbon atoms. When available and of sufficient intensity, measured 13C abundances of ions were compared to what might be expected based on assumption of the presence of a specific compound.

For this work, 16 VOCs were tentatively associated with particular m/z values according to the previously described criteria, and are listed in Table 1 . Ion signal intensities at m/z 33, 45, 47, and 59 were associated with methanol, acetaldehyde, ethanol, and acetone, respectively, and their mixing ratios were calculated based on results of standard gas dilution. Formic acid is not generally associated with dairy cow emissions in the literature, so the signal at m/z 47 was assigned to ethanol, a designation supported by an observed 13C isotopomer having an intensity 2.3% that of the parent 12C protonated product ion rather than 1.1%. It is interesting to note that high mixing ratios of methanol were observed in the cowshed, a compound that has not been previously reported. Other than modest contributions from the 16O17O+ isotopomer at m/z 33, the CH3OH2+ ion is generally free from interference. Here, the 16O17O+ isotopomer contribution was accounted for through measurement of m/z 32, and response due to protonated methanol derived from direct calibration with a standard gas.

The major odorants trimethylamine (TMA), assigned to m/z 60, dimethylsulfide (DMS), assigned to m/z 63, and VFAs have all previously been associated with animal husbandry. Characteristic masses of pure VFAs as measured by von Hartungen et al. (2004) were associated with their respective components. Strong correlations between the various VFA signals, an example of which is depicted in Fig. 2 , lend support to their assignments.

View larger version (21K): [in this window] [in a new window]   Fig. 2. Propionic acid/butyric acid correlation.

Mass 109, which also showed a significantly increased mixing ratio, was assigned to 4-methyl-phenol (p-Cresol), another commonly identified odorant. Lastly, we assigned a significantly enhanced signal at m/z 143 to nonanal, the emission of which from livestock has been reported by Rabaud et al. (2002) and Spinhirne et al. (2003, 2004).

Trace Gas Variability
Time series plots showed that methane, carbon dioxide, ammonia, and nitrous oxide, all of which have been previously investigated in livestock buildings (e.g., Berges and Crutzen, 1996; Bouwman et al., 1997; Koerkamp et al., 1998; McGinn et al., 2003; Yamaji et al., 2004) exhibited regular daily spikes. Methane and carbon dioxide showed a clear distinction between shed and outside air mixing ratios (not shown) indicating a large and constant cowshed production source, attributable to animal respiration. Nitrous oxide did not show a large fluctuation between shed and ambient air mixing ratios, which was in line with literature reports where little fluctuations have been recorded in cowsheds (Berges and Crutzen, 1996). Though ammonia had a large cowshed mixing ratio, the demarcation between shed and reference air mixing ratios was much less distinct. Due to ammonia's highly polar nature, which leads to wall adsorption, its true ambient level was rarely reached due to excessive "smearing," likely inside the photo-acoustic spectrometer under humid conditions (Hinz and Linke, 1998). Ammonia only gradually desorbed from the walls when clean reference air was sampled, thereby artificially elevating the signal of the outside air measurements. As this did not allow for correct "background" measurements, the ammonia level measured before "contaminating" the instrument was taken as reference.

Figure 3 depicts 2 days of the two week long campaign with the upper panel showing the temperature variation in the cowshed with a mean of 14 ± 1°C. The shed VOC mixing ratios also showed regular interval spikes, which appeared to decay exponentially although not always reaching steady state values before the next emission spike occurred. Figure 3 also highlights the difference between shed air and outside air-mixing ratios for the calibrated VOCs. This was largest for acetone, which indicates a constant production source in the cowshed, likely cow respiration as previously suggested (Spinhirne et al., 2003, 2004). It is important to briefly investigate the origin of the periodic large spikes in mixing ratios observed for this cowshed. Though it is conceivable that they arose as a result of changes in shed temperature or ventilation, instead they were rather strongly associated with daily cowshed operations. These included cleaning (solid manure removal) followed by feeding, performed at approximately 5 am and 2 pm each day. During this time, the straw-litter on the concrete floor was replaced manually and using machines while the animals were fed. The excellent correlation with cleaning/feeding at 5 am (solid vertical line) and 2 pm (dashed vertical line) in Fig. 3 strongly suggests that the spikes were a direct result of shed activities and not caused by temperature variations or other unknown factors.

View larger version (55K): [in this window] [in a new window]   Fig. 3. Time series of calibrated volatile organic compounds (VOCs) in the cowshed.

Flux Estimate from Correlations
Correlation of VOC mixing ratios with any other trace gas whose flux has already been investigated in more detail has previously been used to estimate emission rates (Schade and Crutzen, 1995; Berges and Crutzen, 1996; Hobbs et al., 2004). Much work has been done to quantify emission rates of methane, ammonia, nitrous oxide, and carbon dioxide, and estimates of their fluxes for Germany can be readily obtained from the literature. Daemmgen (2004) calculated values of 0.97 Tg a^–1 for methane (from "enteric fermentation" and "manure management") and 0.18 Tg a^–1 for ammonia (from "manure management") from dairy cows in Germany before 2003. Where either ammonia or methane mixing ratio was significantly correlated with a VOC (R > 0.5), it was assumed that the VOC emission scales with that of ammonia or methane. In our case, mass emission ratios for VOCs to methane or VOCs to ammonia were calculated where good correlations where observed. These were then multiplied with the above methane (ammonia) mass release rates to get estimates of VOC fluxes into the lower troposphere. The respective emission ratio EV (here for methane) is given by Eq. [3]:

[3]

µ(VOC)shed – µ(VOC)Out is the enhancement of the VOC mixing ratio in µmol mol^–1 in the shed to that outside, and µ(CH4)shed - µ(CH4)Out is the enhancement of methane (or ammonia) mixing ratio in µmol mol^–1 to the ambient level. Note that this estimate does not require an absolute mixing ratio measurement, but rather only a correct relative one. Errors increase with a decrease in the shed-to-reference abundance, which was large for most VOCs and always very high for ammonia and methane.

While in some cases the correlations based on Eq. [3] were relatively poor, Table 2 lists the major VOCs for which strong covariances with methane or ammonia were found. The closest covariance was observed for those trace gases that presumably have the same main emission sources within the shed or even have the same or similar underlying biochemical production processes, such as acetic acid and methane from anaerobic fermentation. Table 3 compares emission ratios for some VOCs versus ammonia between our fieldwork, recent fieldwork by Mitloehner et al. (this issue) in California, and UK chamber measurements from liquid manure (Hobbs et al., 2004). Reasonable agreement was achieved between all three systems except for methanol (higher emissions from California dairy cows) and DMS (Table 5 in Hobbs et al., 2004), both of which were an order of magnitude different from our results. The DMS emissions estimate for cattle by Hobbs et al. (2004) was derived from previous ruminant breath studies, and the mismatch with our data may be explained by a higher sulfur content of the dairy cows' diet in Britain. Similarly, higher methanol emissions may be caused by a much higher pectin content in the California dairy cow diet, which is dominated by alfalfa containing 10 to 14% pectin versus a silage-dominated diet used in Mariensee that probably contained at most half as much pectin, provided essentially through the beet pulp (?5% in dry matter assuming 40% pectin in beet pulp) (Mertens, 2003; Martin and Mertens, 2005). If pectin is the main source of methanol production via its demethylation (Galbally and Kirstine, 2002), then increased pectin content of the feed will lead to higher methanol emissions from dairy cows.

[4]

x is the mixing ratio in µmol mol^–1 at time t, x0 the mixing ratio at the top of a concentration spike, xbg the background or reference mixing ratio, D the decay constant (dilution rate [h^–1]), t the step time (1, 2, 3...) being equivalent to 107 s, and P' the constant mixing ratio added from continuous emissions into the shed during one measurement cycle of 107 s. Using the value of P' from the nonlinear model fit, a constant mass flux P of the VOC during each period was calculated assuming an instantaneous dilution into the cowshed volume. The fit was done independently for both morning and afternoon spikes and generally showed r2 values between 0.9 and 0.95, reaching near steady-state values only in a few afternoon-night cases. The analysis could not be done for all the VOCs, as many had very uneven decay curves resulting in poor fits, mostly a result of intermediate emissions not accounted for by the model. The results of this analysis are summarized in Table 4 . Though morning and afternoon spike decays were co-varying, no significant dependencies of the dilution rates D were observed on ambient wind speed or direction. The D values varied only slightly between model fits but significantly between the listed species. This suggests that, although the turnover rate of the shed air was relatively stable from day to day, more than the considered processes may have been contributing to the VOC abundances in the shed, such as in-shed chemical reactions of the acids with ammonia. Production values P showed larger variations as is expected for the different compounds, sources, and production rates.

View larger version (17K): [in this window] [in a new window]   Fig. 4. Sample emission model for an evening acetaldehyde spike. Open circles are measured mixing ratio, crosses represent individual mixing ratios added by the assumption of a bell-shaped emissions surge (offset in time for clarity). The dashed line shows the result of those emissions without, and the solid line the modeled result with shed air dilution.

It should be pointed out that in addition to the somewhat limited compound identification using the mass spectrometer alone, the use of emission factors from one shed only to extrapolate to regional and possible global fluxes is problematic as there are differences in animal nutrition, animal management systems, and various co-located sources within different cowsheds, which could lead to systematic flux differences. In addition to this, the CH4 and NH3 fluxes used for VOC flux extrapolation were for the entire sub-sections "manure management" plus "enteric fermentation" and for "manure management," respectively. They extend to both grazing and housed animals and all types of manure management systems (solid, liquid, bed-down). However, we used these fluxes with calculated emission factors from housed livestock only and only for a specific manure management system for our estimates. Therefore, the extrapolations should be viewed with caution, and our emissions estimates are semi-quantitative, and may have an associated error of a factor of two at best. Nevertheless, despite these quantitative uncertainties, (i) our measurements have identified the major VOCs emitted from dairy cows, including previously unidentified, but major, C1 to C3 alcohol species, and (ii) the ammonia and methane independent flux estimate from the spike decays is broadly consistent with the correlation result.

	Conclusions

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Numerous VOCs had increased mixing ratios in the cowshed, dominated by alcohols (ethanol, methanol, C3–C8 alcohols), followed by acetic acid and acetaldehyde, and included ketones (acetone, methyl ethyl ketone), amines (TMA), sulfides (DMS), aromatics (toluene), and higher volatile fatty acids (propionic, butyric, and valeric acids). Most importantly, we observed high levels of ethanol and methanol in the cowshed, the latter not previously having been associated with animal husbandry emissions. Confirmation of our results by similar measurements in California (Mitloehner et al., this issue) lend further support to this finding. Both ethanol and methanol are oxidized in the atmosphere to their respective aldehydes, which are both more harmful and affect the atmosphere's oxidizing capacity through their photolysis up into the upper troposphere (Tie et al., 2003).

The VOCs measured showed varying levels of correlation with methane or ammonia, which was explored to estimate nationwide and possible global emission rates. Emission rates were also modeled from the daily emission profiles in the sheds, and a broad consistency between the two extrapolation methods was achieved. The relatively close agreement of our measured emission ratios with those from California dairies and selected manure emission measurements from the UK indicates that the regional and global extrapolation of VOC emission rates may be reasonable despite large differences in animal management systems and feed nutritional value worldwide. However, the increase of methanol emissions with feed pectin content comparing German and Californian diets suggests that at least for some VOCs, emissions may strongly depend on nutritional regime, complicating global emissions extrapolation.

Our measurements further indicated that while there were constant emissions of trace gases related to respiration and the animal excreta, significant transient sources are the routine management activities inside the animal housings, in particular the mechanical removal of animal excrement and also the distribution of (fermenting) silage as fodder. If the latter could be stored for a shorter time than the common multi-month to year-long periods to limit its fermentation in the silo, and if excrement removal could be performed in a fashion that limits stirring, overturning, or otherwise perturbing the solid or liquid manure, this could be a large step toward VOC emissions abatement inside animal buildings, reducing animal and farm workers' exposure. For this particular barn, on comparing the measured mixing ratios of trace gases that could be identified and quantified to legal threshold limits, we observed that no maximum work place abundances were exceeded.

Further work needs to be done to identify particular sources of gaseous emissions from livestock that may contribute significantly to the total emission from animal sheds. The sources of interest are the animals themselves, their excrement, the animal fodder, and dust particles in animal sheds, which could serve as both VOC sources and sinks. Furthermore, the major differences in VOC emissions between animals fed different diets and between alternate manure management systems ought to be identified.

	ACKNOWLEDGMENTS

 
We thank John Burrows and our colleagues from the IUP in Bremen for hosting this research project. We are grateful to Mr. Zieseniss and Mr. Lindwedel of the FAL in Mariensee for coordinating measurements and to Mr. Olaf Schroeder of the FAL in Braunschweig for helping with the calibration of the photo-acoustic multigas monitor. Special thanks go to the agricultural workers of the FAL in Braunschweig and in Mariensee for their help and enthusiasm during these measurements. We also thank three anonymous reviewers for their helpful comments. This work was partially funded by the Bundesanstalt für Landwirtschaft und Ernährung (BLE) under project number 514–33.26/04HS006. N.M. Ngwabie thanks the German Research Society (DFG) for funding under project number SCHA922/2-1.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 

• Andreae, M.O., and P.J. Crutzen. 1997. Atmospheric aerosols: Biogeochemical sources and role in atmospheric chemistry. Science 276 :1052–1058.[Abstract/Free Full Text]
• Berges, M.G.M., and P.J. Crutzen. 1996. Estimates of global N2O emissions from cattle, pig, and chicken manure, including a discussion of CH4 emissions. J. Atmos. Chem. 24 :241–269.
• Bouwman, A.F., D.S. Lee, W.A.H. Asman, F.J. Dentener, K.W. VanderHoek, and J.G.J. Olivier. 1997. A global high-resolution emission inventory for ammonia. Global Biogeochem. Cycles 11 :561–587.[CrossRef][Web of Science]
• Buhr, K., S. van Ruth, and C. Delahunty. 2002. Analysis of volatile flavour compounds by proton transfer reaction-mass spectrometry: Fragmentation patterns and discrimination between isobaric and isomeric compounds. Int. J. Mass Spectrom. 221 :1–7.[CrossRef]
• Daemmgen, U. 2004. Nationaler Inventarbericht 2004-Berichterstattung unter der Klimarahmenkonvention der Vereinten Nationen-Teilbericht für die Quellgruppe Landwirtschaft. Sonderheft 260. Landbausforschung Völkenrode, Braunschweig, Germany.
• de Gouw, J., C. Warneke, T. Karl, G. Eerdekens, C. van der Veen, and R. Fall. 2003. Sensitivity and specificity of atmospheric trace gas detection by proton-transfer-reaction mass spectrometry. Int. J. Mass Spectrom. 223 :365–382.[CrossRef]
• Filipy, J., B. Rumburg, G. Mount, H. Westberg, and B. Lamb. 2006. Identification and quantification of volatile organic compounds from a dairy. Atmos. Environ. 40 :1480–1494.
• Galbally, I.E., and W. Kirstine. 2002. The production of methanol by flowering plants and the global cycle of methanol. J. Atmos. Chem. 43 :195–229.[CrossRef]
• Guenther, A., C.N. Hewitt, D. Erickson, R. Fall, C. Geron, T. Graedel, P. Harley, L. Klinger, M. Lerdau, W.A. McKay, T. Pierce, B. Scholes, R. Steinbrecher, R. Tallamraju, J. Taylor, and P. Zimmerman. 1995. A global model of natural volatile organic compound emissions. J. Geophys. Res. 100(D5):8873–8892.
• Hartung, J. 1998. Nature and amount of aerial pollutants from livestock buildings. Dtsch. Tieraerztl. Wochenschr. 105 :213–216.[Medline]
• Hinz, T., and S. Linke. 1998. A comprehensive experimental study of aerial pollutants in and emissions from livestock buildings: Part 1. Methods. J. Agric. Eng. Res. 70 :111–118.[CrossRef]
• Hobbs, P.J., J. Webb, T.T. Mottram, B. Grant, and T.M. Misselbrook. 2004. Emissions of volatile organic compounds originating from UK livestock agriculture. J. Sci. Food Agric. 84 :1414–1420.[CrossRef]
• Koerkamp, P., J.H.M. Metz, G.H. Uenk, V.R. Phillips, M.R. Holden, R.W. Sneath, J.L. Short, R.P. White, J. Hartung, J. Seedorf, M. Schroder, K.H. Linkert, S. Pedersen, H. Takai, J.O. Johnsen, and C.M. Wathes. 1998. Concentrations and emissions of ammonia in livestock buildings in Northern Europe. J. Agric. Eng. Res. 70 :79–95.[CrossRef]
• Lindinger, W., A. Hansel, and A. Jordan. 1998. On-line monitoring of volatile organic compounds at pptv levels by means of proton-transfer-reaction mass spectrometry (PTR-MS)-Medical applications, food control, and environmental research. Int. J. Mass Spectrom. 173 :191–241.[CrossRef]
• Martin, N.P., and D.R. Mertens. 2005. Reinventing alfalfa for dairy cattle and novel uses. Proc. of the 35th California Alfalfa and Forage Symp., 12–14 Dec. 2005, Visalia, CA. Dep. of Agronomy and Range Science Cooperative Ext., Univ. of California, Davis, CA.
• McGinn, S.M., H.H. Janzen, and T. Coates. 2003. Atmospheric ammonia, volatile fatty acids, and other odorants near beef feedlots. J. Environ. Qual. 32 :1173–1182.[Abstract/Free Full Text]
• Mertens, D.R. 2003. Nutritional implications of fiber and carbohydrate characteristics of corn silage and alfalfa hay. p. 94–107. In Proc. of the 38th Pacific Northwest Animal Nutrition Conf., 7–9 Oct. 2003, Portland, OR.
• NRC-Ad Hoc Committee on Air Emissions from Animal Feeding Operations. 2003. Air emissions from animal feeding operations: Current knowledge, future needs. National Academies Press, Washington, DC.
• Rabaud, N.E., S.E. Ebeler, L.L. Ashbaugh, and R.G. Flocchini. 2002. The application of thermal desorption GC/MS with simultaneous olfactory evaluation for the characterization and quantification of odor compounds from a dairy. J. Agric. Food Chem. 50 :5139–5145.[CrossRef][Medline]
• Rabaud, N.E., S.E. Ebeler, L.L. Ashbaugh, and R.G. Flocchini. 2003. Characterization and quantification of odorous and non-odorous volatile organic compounds near a commercial dairy in California. Atmos. Environ. 37 :933–940.
• Schade, G.W., and P.J. Crutzen. 1995. Emission of aliphatic-amines from animal husbandry and their reactions-Potential source of N2O and HCN. J. Atmos. Chem. 22 :319–346.[CrossRef]
• Schiffman, S.S., J.L. Bennett, and J.H. Raymer. 2001. Quantification of odors and odorants from swine operations in North Carolina. Agric. For. Meteorol. 108 :213–240.[CrossRef]
• Singh, H.B., L.J. Salas, R.B. Chatfield, E. Czech, A. Fried, J. Walega, M.J. Evans, B.D. Field, D.J. Jacob, D. Blake, B. Heikes, R. Talbot, G. Sachse, J.H. Crawford, M.A. Avery, S. Sandholm, and H. Fuelberg. 2004. Analysis of the atmospheric distribution, sources, and sinks of oxygenated volatile organic chemicals based on measurements over the Pacific during TRACE-P. J. Geophys. Res. 109, D15S07, doi:10.1029/2003JD003883.
• Spanel, P., Y.F. Ji, and D. Smith. 1997. SIFT studies of the reactions of H3O+, NO+, and O2(+) with a series of aldehydes and ketones. Int. J. Mass Spectrom. 165 :25–37.[CrossRef]
• Spinhirne, J.P., and J.A. Koziel. 2003. Generation and calibration of standard gas mixtures for volatile fatty acids using permeation tubes and solid-phase microextraction. Trans. ASAE 46 :1639–1646.
• Spinhirne, J.P., J.A. Koziel, and N.K. Chirase. 2003. A device for non-invasive on-site sampling of cattle breath with solid-phase microextraction. Biosyst. Eng. 84 :239–246.[CrossRef]
• Spinhirne, J.P., J.A. Koziel, and N.K. Chirase. 2004. Sampling and analysis of volatile organic compounds in bovine breath by solid-phase microextraction and gas chromatography-mass spectrometry. J. Chromatogr. A. 1025 :63–69.[CrossRef][Medline]
• Su, T., and W.J. Chesnavich. 1982. Parameterization of the ion-polar molecule collision rate-constant by trajectory calculations. J. Chem. Phys. 76 :5183–5185.[CrossRef]
• Tie, X., A. Guenther, and E. Holland. 2003. Biogenic methanol and its impacts on tropospheric oxidants. Geophys. Res. Lett. 30:1881, doi:10.1029/2003GL017167.
• von Hartungen, E., A. Wisthaler, T. Mikoviny, D. Jaksch, E. Boscaini, P.J. Dunphy, and T.D. Mark. 2004. Proton-transfer-reaction mass spectrometry (PTR-MS) of carboxylic acids-Determination of Henry's law constants and axillary odour investigations. Int. J. Mass Spectrom. 239 :243–248.[CrossRef]
• Warneke, C., J.A. De Gouw, W.C. Kuster, P.D. Goldan, and R. Fall. 2003. Validation of atmospheric VOC measurements by proton-transfer-reaction mass spectrometry using a gas-chromatographic preseparation method. Environ. Sci. Technol. 37 :2494–2501.[Medline]
• Warneke, C., J. Kuczynski, A. Hansel, A. Jordan, W. Vogel, and W. Lindinger. 1996. Proton transfer reaction mass spectrometry (PTR-MS): Propanol in human breath. Int. J. Mass Spectrom. Ion Process. 154 :61–70.[CrossRef]
• Yamaji, K., T. Ohara, and H. Akimoto. 2004. Regional-specific emission inventory for NH3, N2O, and CH4 via animal farming in south, southeast, and East Asia. Atmos. Environ. 38 :7111–7121.
• Zhao, J., and R.Y. Zhang. 2004. Proton transfer reaction rate constants between hydronium ion (H3O(+)) and volatile organic compounds. Atmos. Environ. 38 :2177–2185.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ngwabie, N. M. Articles by Hinz, T. Search for Related Content PubMed Articles by Ngwabie, N. M. Articles by Hinz, T. Agricola Articles by Ngwabie, N. M. Articles by Hinz, T. Related Collections Biogeochemical Processes Air Pollution Animal Waste