Glasius, M., Konggaard, P., Stubkjær, J., Bossi, R., Hertel, O., Ketzel, M., Wåhlin, P., Schleicher, O. & Palmgren, F. 2007. Research Note No. 235, 42 pp.
Recent Danish as well as international studies have shown that wood burning may lead to increased levels air pollution levels of particles and different types of organic compounds in the vicinity of the sources. The air pollution constitutes a health risk for the local population but this risk is poorly quantified. Furthermore, the number of available studies is relatively limited with regard to pollutant emissions and concentrations related to wood burning. This report summarizes a number of studies which have the aim of contributing to the current understanding of the air pollution problem related to the use of wood stoves and boilers. The investigations include studies of emissions and resulting outdoor pollutant levels in two villages: Gundsømagle and Vindinge. Furthermore investigations of indoor-outdoor levels have been performed in two single-family detached houses in <st1:City w:st="on"><st1:place w:st="on">Randers</st1:place></st1:City> .
Measurements of emissions were carried out for 19 private chimneys in Gundsømagle. From these chimneys flue gas was sampled and subsequently analysed for particles, PAH and dioxins. The results revealed that the emissions vary several orders of magnitude between different stoves. In average the results showed an emission of 6.2 g/kg burned wood, but the figures varied between 0.1 and up to 23.5 g/kg wood. Measurement errors cannot be excluded for the lowest values since the analysis method is encumbered with uncertainty. Two specific measurements are exceptional with very low values of 0.1 and 0.2 g/kg wood, respectively. The measurements of PAH showed in average an emission of 57 ng/kg wood, but also in this case large variations where found with values in the range of 4 to 380 ng/kg wood; this is equivalent to a factor 90 between lowest and highest value. The measurements of dioxins were made in international toxicity equivalents I-TEQ, and again the results revealed large variations in emissions. On average the results showed an emission of 19 ng/kg wood, but the results varied between 0.027 and 140 ng/kg wood – this means a factor 5000 between highest and lowest value. Generally the results show that few single sources contribute with the main part of the emission. A significant reduction in the emission could thus be obtained by a reduction in emissions from the most polluting sources.
In Vindinge measurements showed an increased concentration of the outdoor particle pollution compared to the background levels, where natural gas is installed in the single-family detached houses. For PM2.5 the diurnal mean concentrations were in average 1.2 mg/m3 above background level, and during evening hours the average concentrations were 2.6 mg/m3 above background level. The increased levels can primarily be attributed to emissions related to wood burning. The traffic in the local area is limited and no other local sources, e.g. other types of power and heat production or industry, are present. The PAH content in particles was increased by 5.1 ng/m3, to about twice the background level. The situation is similarly for the contents of levoglucosan and mannosan, which both are tracers for wood smoke. For these compounds the levels in Vindinge were in the range 313±237 ng/m3, while the background level was found to be 168±89 ng/m3; in this connection levoglucosan is by far the largest contributor to the measured mass. However, large temporal variations in concentrations where observed for both PAH and levoglucosan/mannosan.
The particle pollution from wood stoves in Vindinge and Gundsømagle was calculated with the Gaussian plume model OML-Multi. For both areas it was assumed that the stoves had a four hour daily activity with a constant emission. The emission factor used corresponds to that used in national emission inventories. This is a relatively high value compared to the results of emission measurements within the current study. For Gundsømagle the model results are in good agreement with the observations, whereas the calculated levels for Vindinge are about 75 % higher than the measurements. However, the results depend very much on the emission factor and the assumed daily stove activity, and the overestimation in Vindinge probably occurs because the chosen combination of the two variables is less representative here. For Vindinge an additional calculation was performed, based on information about wood burning activity obtained from a questionnaire handed out to the local population. The result was, however, an even more pronounced overestimation, probably due to a crude classification of periods with wood burning activity. This crude classification has most likely led to an overestimation of the number of hours with wood burning. Generally the calculations showed that OML-Multi is suitable for estimation of concentrations related to wood burning from single-family detached houses, but in order to obtain good results a highly detailed mapping of the sources is necessary.
Indoor-outdoor measurements of PM2.5, levoglucosan and soot were carried out in two single-family detached houses in <st1:City w:st="on"><st1:place w:st="on">Randers</st1:place></st1:City> (one with and one without wood stove). For the house with wood stove it was investigated whether the indoor levels were related to use of own wood stove. For PM2.5 other activities in the house made it difficult to establish such a relation. Measurements of levoglucosan showed that the house with wood stove had increased levels compared with outdoor levels. Increased levels may thus be associated with use of wood stove in the house. Just after upstart of the wood stove, levoglucosan is not released. Levoglucosan is thus a poor tracer in this situation. For soot, however, increased levels are observed immediately after upstart of the stove and subsequently the levels decrease. The decrease is believed to be associated the exchange of air related to the stove, which drags cleaner air from outside into the house. Levoglucosan is also measured in the house without wood stove, which shows that particles from outdoor are transported into the house.
Full report in pdf-format (in Danish; 1,649 kB)
