The Swedish Environmental Protection Agency, the unit for Air Quality and Climate Change, is responsible for the national air quality and precipitation monitoring in rural background areas. This report presents the results from the activities within the National monitoring programme for air pollutants regarding measurements (performed by IVL, SU, SLU and SMHI respectively) until 2015 and modelling (performed by SMHI) until 2015 for ground level ozone and 2014 for other pollutants. Furthermore, the National monitoring programme includes some activities on Svalbard, but these results are not presented in this report. January 1, 2016 the monitoring activities at the EMEP site Vavihill, Söderåsen in Skåne, were moved to Hallahus, 1 km north of the former location.
For most of the air pollutants monitored the situation has improved significantly since the measurement started between 15 and 35 years back, regarding air concentrations as well as deposition in the rural background. The pollution load is in general decreasing the further north one goes.
For most of the components for which there are environmental quality standards and environmental objectives, the concentrations are well below the limit and target values. The concentrations of ground-level ozone exceed the air quality standard for health. For ground-level ozone, particles and benzene (in urban background air) there is a risk for concentration levels above the specifications of the environmental objectives.
Acidifying and eutrophying substances
The concentration of pollutants in air and precipitation has during the last decades decreased significantly due to international agreements to reduce emissions of e.g. sulphur and nitrogen oxides.
Air
A comparison between the measured concentrations of sulphur dioxide (SO2) in air in the 1980's and in the 2010's shows that the annual average concentration has been reduced by between 88 and 95% at the EMEP sites. When comparing the Swedish regions, it becomes clear that the concentrations in the 2010's were lowest in northern Sweden, whereas the concentrations in the southeastern and southwestern Sweden were a comparable levels. An exception is 2014 when the concentration in northern Sweden was almost on the same level as in the rest of Sweden. The reason for this is believed to be air transport of sulphur from the outbreak at Holuhraun in Iceland. In Sweden's southwestern and southeastern regions the reduction of SO2 in air has been larger compared to northern Sweden. Similar results were seen also for concentrations of sulphur in deposition for which the decrease of sulphur in southern Sweden was greater compared to northern Sweden. Annual average concentrations of sulphate in the air (SO4) in air, measured at the EMEP sites have fallen by between 70 and 80% from the 1980's to the 2010's.
A comparison between the measured concentrations in air of nitrogen dioxide (NO2) in the 1980's and the 2010's shows that the annual average concentrations at the EMEP sites have decreased by 50% in southern Sweden (Vavihill, Rörvik/Råö and Aspvreten) and with more than 70% in northern Sweden (Bredkälen). The comparison between the Swedish regions shows that the levels during the 2010’s have been the highest in southwest Sweden and lowest in northern Sweden. The concentrations in southeastern Sweden lie between.
Annual average concentrations of nitrate (NO3) in the air, measured at the EMEP sites during 1986-2015, show a reduction of approximately 50% in Vavihill and Aspvreten and 35 % at Rörvik / Råö. The nitrate concentration in Bredkälen is for the whole time period much lower compared with the other EMEP sites, but even here the concentrations have decreased significantly. The annual average of ammonium (NH4) in the air has at the EMEP sites declined by 50-60% during the same time period.
Concentrations of Cl, Na, Mg, Ca, and K in air have been measured at the EMEP sites from 2009 to 2015. Generally the annual average concentrations have been highest at Råö and lowest at Bredkälen.
Modelling the concentration of different substances in air is achieved by combining observations with model data, in the MATCH Sweden system. Measurements and model data complement each other in order to achieve a better knowledge. With the model system, concentration levels can be divided into contributions from Swedish and foreign emission sources (long-range transport).
Modelling results (on a regional scale) show that for reduced nitrogen (the sum of ammonium and ammonia) the total yearly average concentration varied between 1.19 μg N/m3 in the south of Sweden, and 0.04 μg N/m3 in the most northern parts of the country in 2014. For nitrogen dioxide the highest concentrations were modelled in the larger urban areas and the concentration varied between 0.06 and 2.31 μg N/m3. The air concentration of SO2 were modelled to vary between 0.03 and 0.81 µg S/m3 in Sweden, and the highest values were seen in the larger urban areas and along the coast of Norrland. During 2014 the Swedish emissions were calculated to have caused on average 27% of the total concentration of SO2 over Swedish land areas. For NO2 and reduced nitrogen the equivalent percentage was estimated to 48% and 37%.
Precipitation and Deposition
Wet deposition of sulphur in Sweden was significantly higher in 2014 compared to 2015. This can mainly be explained by the high sulphur emissions from the volcano eruption in Iceland between August 2014 and February 2015. Also the wet deposition of inorganic nitrogen was significantly higher in 2014 compared to 2015, primarily in southwestern Sweden. The highest wet deposition of inorganic nitrogen was just above 20 kg/ha in the most southern part of Sweden. In 2015 the highest wet deposition of inorganic nitrogen was measured at the southwest coast of Sweden, with just below 15 kg/ha.
A statistical trend analysis for wet deposition has been made for the years 2000-2015. Averages for measurements separated for three different areas in Sweden (North, South-East and South-West) are used in the trend analysis. The monitoring stations included in the analysis in the different areas are the stations that have full data coverage during all years. No aspects of the stations representability in the different areas have been included in the analysis.
No statistically significant trend in any of the three analysed areas was obtained for the amount of precipitation. Wet deposition of sulphur has decreased significantly by 51-65% over the period 2000-2015 for all three areas in Sweden. The largest decline of sulphur deposition was insouthwestern Sweden and lowest in northern Sweden. The hydrogen ion deposition, which may be used as a measure of the acid load, has also declined in all regions since 2000. No statistically significant change was obtained for chloride deposition in any area 2000-2015.
Wet deposition of inorganic nitrogen (nitrate + ammonium nitrogen) decreased significantly during the period 2000-2015 in northern Sweden (29%) and southwestern Sweden (24%), while no significant changes were obtained for southeastern Sweden. The wet deposition of ammonium nitrogen showed no statistically significant change in any of the three areas during 2000-2015. However, the wet deposition of nitrate nitrogen decreased by 41% in southwestern Sweden, by 35% in southeastern Sweden and by 34% in northern Sweden since 2000.
The model system divides the total deposition into wet and dry. The share of the wet to the total deposition was 71% for sulphur (sum of sulphur dioxide and sulphate excluding contribution from sea salt), 79% for reduced (the sum of ammonium and ammonia) and 78% for oxidized nitrogen (sum of NO, NO2, HNO3, PAN, N2O5, NO3- salts and organic NO3- among other substances) in 2014. The deposition for different land use types can also be calculated.
The modelling shows that the deposition of reduced nitrogen is highest in the southern parts of Sweden, and is lowest in northern Sweden. In 2014 the deposition was increased compared to 2013, and it ranged between 35 and 820 mg N/m2. The Swedish contribution was marginally higher in 2014, indicating that most of the increase was due to long-range transport. A similar pattern was seen for oxidised nitrogen, for which the deposition varied between 40 and 690 mg N/m2 in 2014. For sulphur (sea salt not included) the total deposition varied between 100 and 600 mg S/m2, which was an increase compared to 2013. The volcanic eruption at Bardarbunga in Iceland contributed to the long-range transport of sulphur, especially in mid-Sweden. In 2014 the Swedish emissions caused on average 5% of the deposition of sulphur (sea salt not included) in Swedish land areas. The corresponding number for oxidized and reduced nitrogen was 6% and 15% respectively.
Heavy metals
The concentrations of heavy metals in air and in precipitation are lower in Sweden than in some other comparable countries. This can be explained by Sweden’s northerly position and the relative low use of fossil fuels for electricity and heath production. The highest yearly average concentrations of lead, nickel, cadmium and arsenic in the air in southern Sweden are ten times lower than the threshold values given in the EU directives 2004/107/EC and 2008/50/EC. The situation with mercury is slightly different, since this metal predominantly occurs as an elemental gas in the atmosphere. Due to its long atmospheric residence time it is more or less evenly distributed in the northern hemisphere. Nowadays the concentration levels of mercury in air and in precipitation in southern Sweden are similar to that of many other European countries.
Persistent organic substances
The concentration of PCBs and chlorinated pesticides in background air has generally declined since the start of the measurements in 1996, the decline has however in recent years levelled off. The air concentrations of PAHs, PCBs and DDTs are generally higher in southern Sweden compared to northern Finland, while α-HCH and chlordanes are at the same level both in the south and in north. The same pattern also applies to the atmospheric deposition.
The pesticides, aldrin, heptachlor, diuron, atrazine and isoproturon, which only are measured at Råö, have only been detected occasionally in air and deposition samples. Endosulfan (α- and β-endosulfan, endosulfan sulfate) have been detected in all of the air and deposition samples from both Råö and Pallas.
BDE (47, 99 and 100) has declined in both air and deposition and the levels are in the same range at all sites; Råö, Aspvreten and Pallas. BDE-85, 153, 154, 209 and HBCDD have only been detected occasionally.
The dioxin/furans concentrations in air are generally higher at the Swedish west coast compared to the Swedish east coast, while the levels of chlorinated paraffins (SCCPs) are higher at Aspvreten compared to Råö.
The long term monitoring program gives the possibility to follow up measures and bans. Although the use of PCBs was banned long time ago, they still occur in air from background areas. The decrease in PCB levels is slow, which shows that the PCBs are stored in the communities and ecosystems. As regards e.g. PBDEs, there is a marked decline in the levels, which shows the effect of the ban of these chemicals in the EU.
Plant protection products (pesticides)
Higher concentrations and a larger number of different pesticides were found in precipitation collected at Vavihill in the very south of Sweden, compared to precipitation collected at Aspvreten (situated just south of Stockholm). Differences in findings between the sites can be explained by the closeness of Vavihill to more intense agricultural areas, both in Sweden and on the European continent. A substantial portion (close to 50%) of the pesticides occurring in precipitation are no longer applied within Sweden, high-lightening the importance of a trans-boundary transport, also of some modern pesticides. Average deposition of pesticides at Vavihill has varied between 100 and 650 mg/ha, month (10-65 µg/m2, month). The deposition at Aspvreten has been one tenth of that at Vavihill. Substances found in air samples are to a large extent the same as those found in precipitation. In a special project we examined whether there are pesticides that are mainly transported in air bound to particles. The results show that a significant proportion of the substances that are currently permitted for use could be detected in the filter rather than in the adsorbent normally analysed. The project shows that it is important in the future to include analysis also of the filter material within the long-term air-monitoring program for pesticides.
Volatile organic components (VOC)
Highest concentrations of the VOCs were measured in November to Mars, i.e. during the coldest period of the year. The seasonal variations are probably due to higher emissions from combustion processes wintertime combined with a lower mixing layer in the lower part of the atmosphere during the same period of the year. No specific seasonal variation in the distribution of the VOCs was detected in 2009-2015, the most volatile substances accounted for the largest share in all seasons.
The environmental quality standard (EQS) for benzene is 5 µg/m3 as an arithmetic annual average. With the guidance of 85 weekly measurements of benzene at the background site Råö and 58 weekly average values in urban background air in Gothenburg (roof level), the EQS was not exceeded at these sites during 2009-2015.
Rapport C 224 - Nationell luftövervakning – Sakrapport med data från övervakning inom Programområde Luft t.o.m. 2015
19
For 1,3-butadiene and benzene there are clarifications to the national environmental objectives (butadiene 0.2 µg/m3 and benzene 1 µg/m3 as annual averages). The average concentrations of more than 6 000 hourly data during 2009-2015 in urban background air in Gothenburg, were <0.1 µg/m3 for 1,3-butadiene and 0.9 µg/m3 for benzene. These results indicate a risk that the environmental objective for benzene is exceeded on an annual basis.
Particles
The concentration of PM10 in the regional background is about 15 µg/m³ in southern Sweden (Vavihill and Råö), about 8 in central Sweden (Aspvreten and Norr Malma) and 3 -4 in the north (Bredkälen). The concentration of PM2.5 in the regional background is 7 -9 µg/m³ in southern Sweden (annual mean, Vavihill), 5-7 µg/m³ in central Sweden (Råö and Aspvreten) and about 2 in the northern part of the country (Bredkälen). Also in urban background air the average concentration of PM2.5 decreases northwardly, and is in the south of Sweden (Burlöv, Stockholm) of the same order as in the regional background. In the northern part of the country (Umeå) the concentration levels are somewhat higher in urban than in rural areas. The average exposure indicator shows that Sweden achieves the requirements set by the EU at an acceptable level of exposure.
At Aspvreten in central Sweden, where measurements of PM10 has been going on since 1990, the level has dropped from almost 20 to 7 -9 µg/m³ today. At Vavihill in Skåne, where measurements began in 2000, and Råö in the Gothenburg area (start of measurement in 2007), there is no clear trend.
PM2.5 at Aspvreten since 1998 has fallen from 11 to 12 to about 6 µg/m³ today. Most of the decrease occurred in the period 2000 – 2005. The trend is similar at the other stations in Sweden.
The concentration of organic carbon (OC), in the PM10 fraction, was approximately 1.5 µg/m³ in the southern and central parts of Sweden (Vavihill and Aspvreten) with no clear seasonal variation. The monthly average concentration of elemental carbon (EC) is about 0.2 to 0.5 µg/m³ during winter and from 0.1 to 0.2 during the summer. There is no obvious trend of OC and EC since the start of the measurements in 2008. No measurements are made in Northern Sweden.
Soot has been measured with an indirect method as 'black smoke' (BS) at several background sites since the early 1980s. Since then, the concentration in southern Sweden decreased from 4-8 to approximately 1.5 µg/m³ today. In northern Sweden (Bredkälen) the concentration was approximately 1.5 in the 1980s and is below 1 µg/m³ today. Most of the decrease occurred in the 1980s and early 1990s. One reason that no reduction is seen thereafter may be that the levels are often below the detection limit of the measurement method.
Ground-level ozone
The concentration of ground-level ozone is largely determined by the meteorological conditions, and for the average annual level of ozone there is neither a clear trend in time nor a geographical gradient over the country. The number of episodes of high concentrations of ozone, though, is significantly higher in the southern part of Sweden than in the north, both as regards the 8 hour mean value (limit value for health) and AOT40 (limit value for vegetation). During 2014 and 2015 no hourly values above the information threshold (180 µg/m3) were observed. However, the environmental goal for hourly means (80 µg/m3) as well as the limit value for the 8 hour mean (120 µg/m3) were exceeded at many of the monitoring sites during these years.
Modelled daily exceedances of 70 μg/m3 of ground-level ozone show that year 2013 had more exceedances than 2014 and 2015 in all of Sweden. Most exceedances occurred in southern Sweden and in the inlands of Norrland. The trend of these exceedances during the years 1990-2013 is increasing across the whole country.
Between 2013 and 2015, daily exceedances of 120 μg/m3 occur most frequently in southern Sweden, and during 2013 also along the coast of Norrland. The highest number of exceedances occurred in Blekinge in 2014, with about 13 days exceeding 120 μg/m3. In 2015, most exceedances occurred in the south of Sweden and around Gothenburg. The yearly number of days with exceedances of 120 μg/m3 over the period 1990-2013 was lower in the northern parts of Sweden (up to two days) and higher in the south (4-15 days). This suggests that 2013 was a normal/low year for the highest concentrations, while 2014 was a year with higher but not extreme concentrations than normal, except for in Blekinge. The trend of the last 25 years suggests that the number of exceedances of 120 μg/m3 is decreasing in the south of Sweden. That 2014 had higher concentrations than 2013 and 2015 was a result of meteorological variability.
The indicator AOT40, which shows vegetation impact on crops and forest, was higher during 2013 than during 2014 and 2015, with the largest impact in the south, around Stockholm and in the Gävleborg region. AOT40 for crops in 2013 was close to the average of the period 1990-2013, whilst 2014 and definitely 2015 were lower than average. AOT40 for forest was higher in the north and lower or similar in the south in 2013 compared to the average. The trend over the last 25 years is decreasing everywhere except for the most northern part of Sweden, for both crops and forest impacts. The differences between the years 2013-2015 were caused by meteorological variability.
Stratospheric ozone
There is a large natural variation of ozone in the stratosphere from day to day and also over the year. These variations are mainly caused by large scale atmospheric transport mechanisms, but there is also a dependence on atmospheric chemistry. One of these chemically introduced factors is a long term decline of stratospheric ozone due to manmade ozone depleting substances. In global records of total ozone an indication of recovery can now be seen. However, this is often hard to see in data from specific stations due to the large natural variation and local conditions.
STRÅNG
The STRÅNG model that generates a number of radiation quantities over northern Europe has now produced data for a long period. An upgrade is under way and should be ready in 2017. For the last couple of years one can see the typical pattern with a strong latitudinal dependence, and that there also is relatively more radiation over the Baltic area compared to nearby similar latitudes.