U R G E N T P R O B L E M S OF I M P R O V I N G
AIR POLLUTION
MONITORING
BACKGROUND
SYSTEMS
M. E. B E R L Y A N D , N. SH. V O L B E R G , R. F. L A V R I N E N K O a n d E . N. R U S I N A Voeikov Main Geophysical Observatory, USSR
Abstract. For more than 12 years, systematic observations of background air pollution have been carried out in accordance with the WMO Programme using the network of USSR stations located in sparsely populated settlements and in a number of neighbouring cities. The parameters involved include spectral radiation measurements, determination of chemical composition of precipitation and the concentrations of a number of atmospheric pollutants. Analysis of the data obtained allows conclusions to be drawn on the capabilities of the current system and to evaluate methods of improving it. In order to further improve the monitoring system, it is recommended that the system should perform the same observations on air pollution and precipitation as carried out by other international and national programs, and also to create centralized laboratories to deal with the analysis of samples from these monitoring stations. Additionally, solid sorbents are emerging as an effective means of sampling certain air pollutants. They may be sent by post, they increase the accuracy of measurements and allow air sampling intervals of up to 7-10 days, thus synchronizing this period with the interval of precipitation sampling.
1. Introduction
The activities which started in the early 1970s within the WMO BAPMoN program and aimed at creating a global background air pollution monitoring system, have been continuously expanding. In the USSR, a network of stations has been established where systematic observations under this program have been carried out for more than 12 years. The stations are located in different geographical zones, some of them in sparsely populated settlements, others (their 'counterparts') in neighbouring cities with no large industrial enterprises. The organization of monitoring according to the principle of counterpart-stations, may facilitate the assessment of local anthropogenic impact on the parameters under study [1, 7]. There are two pairs of stations in the European territory of the USSR (ETU), one of them in the north-east near Syktyvkar and the other is situated in one of the Soviet Baltic republics; one more baseline station is located near the city of Pyatigorsk (Caucasus) at a height of 2 000 m. There are also two pairs of stations in the Asian part of the USSR near the cities of Kurgan (Trans-Ural region) and Irkutsk (East Siberia) and a station in Turukhansk (West Siberia) within the Arctic circle. The monitoring program includes spectral radiation measurements, analysis of precipitation chemistry and the concentrations of selected atmospheric pollutants. Observations under the program are also carried out in a number of Soviet biosphere reserves; and, in addition, the chemical composition of precipitation sampled at a specialized meteorological network since 1985-60 is also determined. Analysis of the data obtained provides information on the background values of a number of air Environmental Monitoring and Assessment 11: 269-278, 1988. 9 1988 Kluwer Academic Publishers. Printed in the Netherlands.
270
M. E. BERLYAND El" AL.
pollution parameters, and also allows evaluation of the capacities of the monitoring system, as well as the means to improve it. 2. Improvement of Methods and Means of Observations
The need to improve this program results from the necessity to take measurements which are often close to the threshold sensitivity of analytical equipment. Furthermore, one of the major goals of monitoring is to detect any temporal trends in the parameters measured. Such trends are likely to be weak and, thus, accuracy of measurement and advanced data processing methodology are needed. Solar radiation measurements of aerosol turbidity are performed in the USSR using a standard line-operated actinometer M-3 equipped with five wide-band light filters. The use of wide-band filters decreases the accuracy of single measurements. Nevertheless, the the methodology devised for the processing of data bases brings about a considerable reduction in the measurement error of the parameters studied [12]. Analysis reveals that the values obtained for the aerosol optical depth Xsoo for the wavelength 500 nm are in good agreement with data provided by a reference spectrometer (Foisper actinometer with interference filters). Comparison of actinometric instruments used for background observations in CMEA member countries [5] also shows a good correlation with the data obtained by means of the M-3 actinometer and Volz photometer recommended by WMO. To increase the sensitivity and accuracy of measurements of spectral turbidity parameters in situations of high transmission values, as observed at high-mountain stations, one can use the model two-channel filter pyrheliometer designed at the Main Geophysical Observatory (MGO) [13]. The pyrheliometer is almost 3 times more sensitive than that of the Foisper actinometer and of an order of magnitude higher than that of the M-3 actinometer. In addition, it provides absolute values of direct solar radiation. Activities are also in progress on the determination of precipitation chemistry. To determine heavy metals in precipitation, MGO has devised a flotation absorption method of analysis which produces a considerable increase in the coefficient of concentration and reduces the duration of analysis. It is primarily devised for the determination of toxic trace elements, such as Pb, Cd, Hg and As. Measurement of the concentrations of a number of pollutants in the atmospheric surface layer have recently been started at the network of background stations. In such cases, when concentrations are determined at a fixed level, it is necessary to measure very small amounts of pollutants. In this connection new sensitive methods have been devised in MGO to determine pollutant concentrations by air sampling with absorption instruments based on solid-state film sorbents [4, 8]. The latter are glass tubes containing a layer of glass beads which is impregnated by a specially selected solution before the analysis. Their advantages are easy preparation, small size and low weight, complete desorption of substances absorbed in the process of sampling, no risk of sample contamination by impurities contained in the sorbent.
URGENT PROBLEMS OF IMPROVING BACKGROUND AIR POLLUTION MONITORING SYSTEMS
271
Sampling on solid sorbents followed by photometric analysis is now becoming widely used in the USSR for urban air pollution monitoring, i.e. for monitoring at the impact level. Techniques have already been worked out to use this approach for the determination of SO2, H2S, NO2, HF, NH3, Hg, organic acids, etc. The sensitivity of the technique is very high, which allows its application, particularly after a certain modification at background stations, to determine SO2 and NO 2 concentrations. Examination of the technique has shown that when sampling 10 m 3 of air, the measurement sensitivity for SO2 is 0.05 gtg/m3, which is lower than the concentration of the pollutant in the cleanest areas over the ocean [12]. An intercomparison of the methods used for measuring background concentractions of SO2 in Bulgaria, GDR, Poland, USSR and Czechoslovakia was carried out in 1980 in Poland. Similar intercalibration exercises for NO2 were conducted in 1983 in GDR. In the USSR, air samples to be analyzed for SO2 and NO2 were collected in sorption tubes. In Bulgaria (for SO2), GDR, Poland and Czechoslovakia (for NO2), samples were taken in absorbing air bubblers. In Poland and Czechoslovakia impregnated filters were also used. Samples were analyzed by photometric techniques. When SO2 was determined according to the procedures developed by the USSR, Bulgaria and Czechoslovakia, pararosaniline and formaldehyde reactions were used. Procedures presented by Poland required thorin reactions. NO2 was determined using Griss-Salzman reaction. When air was analyzed for a given concentration, the results obtained by means of different methods showed a good agreement. At the same time, it was established that, given equal sampling periods, solid sorbent methods were more sensitive than other methods. In particular, NO 2 concentrations determined using solid sorbents, were of an order of magnitude lower than measured otherwise. The USSR methodology provided a detection limit of 0.3 ~tg/m 3 for air sample volume 2 m 3. The Polish and Czechoslovak detection limits were 3 and 4 ~tg/m 3, respectively. Measurements of sulphates according to the MGO procedure based on thermal decomposition of sulphates to sulfur dioxide and coulometric determination of the latter, have been started at background stations. The procedure is more sensitive than the widespread photometric thorin method and is much simpler, requiring only two operations instead of the 14 in the thorin technique. With the 103 m air sample, the sensitivity of the procedure is 2 ~g/m 3 and the measurement error is 15~ 3. Analysis of Results Statistical analysis of the monitoring data from the BAPMoN program identifies differences in background parameters due to the location of stations as well as a number of year-to-year variations. Some of the results obtained were given in [7] and other papers. The use of spectral radiation data from the USSR background stations in conjunction with material published by the International WMO Centre in Ashville [14] allows a comparison to be made between mean annual values of aerosol turbidity
272
M. E. BERLYA.ND ET AL.
m
~udaM (~) TSOO~,0,~0
Turke g (G
~ndonesco(()
0400
AIR POLLUTION
MONITORING
BACKGROUND
SYSTEMS
M. E. B E R L Y A N D , N. SH. V O L B E R G , R. F. L A V R I N E N K O a n d E . N. R U S I N A Voeikov Main Geophysical Observatory, USSR
Abstract. For more than 12 years, systematic observations of background air pollution have been carried out in accordance with the WMO Programme using the network of USSR stations located in sparsely populated settlements and in a number of neighbouring cities. The parameters involved include spectral radiation measurements, determination of chemical composition of precipitation and the concentrations of a number of atmospheric pollutants. Analysis of the data obtained allows conclusions to be drawn on the capabilities of the current system and to evaluate methods of improving it. In order to further improve the monitoring system, it is recommended that the system should perform the same observations on air pollution and precipitation as carried out by other international and national programs, and also to create centralized laboratories to deal with the analysis of samples from these monitoring stations. Additionally, solid sorbents are emerging as an effective means of sampling certain air pollutants. They may be sent by post, they increase the accuracy of measurements and allow air sampling intervals of up to 7-10 days, thus synchronizing this period with the interval of precipitation sampling.
1. Introduction
The activities which started in the early 1970s within the WMO BAPMoN program and aimed at creating a global background air pollution monitoring system, have been continuously expanding. In the USSR, a network of stations has been established where systematic observations under this program have been carried out for more than 12 years. The stations are located in different geographical zones, some of them in sparsely populated settlements, others (their 'counterparts') in neighbouring cities with no large industrial enterprises. The organization of monitoring according to the principle of counterpart-stations, may facilitate the assessment of local anthropogenic impact on the parameters under study [1, 7]. There are two pairs of stations in the European territory of the USSR (ETU), one of them in the north-east near Syktyvkar and the other is situated in one of the Soviet Baltic republics; one more baseline station is located near the city of Pyatigorsk (Caucasus) at a height of 2 000 m. There are also two pairs of stations in the Asian part of the USSR near the cities of Kurgan (Trans-Ural region) and Irkutsk (East Siberia) and a station in Turukhansk (West Siberia) within the Arctic circle. The monitoring program includes spectral radiation measurements, analysis of precipitation chemistry and the concentrations of selected atmospheric pollutants. Observations under the program are also carried out in a number of Soviet biosphere reserves; and, in addition, the chemical composition of precipitation sampled at a specialized meteorological network since 1985-60 is also determined. Analysis of the data obtained provides information on the background values of a number of air Environmental Monitoring and Assessment 11: 269-278, 1988. 9 1988 Kluwer Academic Publishers. Printed in the Netherlands.
270
M. E. BERLYAND El" AL.
pollution parameters, and also allows evaluation of the capacities of the monitoring system, as well as the means to improve it. 2. Improvement of Methods and Means of Observations
The need to improve this program results from the necessity to take measurements which are often close to the threshold sensitivity of analytical equipment. Furthermore, one of the major goals of monitoring is to detect any temporal trends in the parameters measured. Such trends are likely to be weak and, thus, accuracy of measurement and advanced data processing methodology are needed. Solar radiation measurements of aerosol turbidity are performed in the USSR using a standard line-operated actinometer M-3 equipped with five wide-band light filters. The use of wide-band filters decreases the accuracy of single measurements. Nevertheless, the the methodology devised for the processing of data bases brings about a considerable reduction in the measurement error of the parameters studied [12]. Analysis reveals that the values obtained for the aerosol optical depth Xsoo for the wavelength 500 nm are in good agreement with data provided by a reference spectrometer (Foisper actinometer with interference filters). Comparison of actinometric instruments used for background observations in CMEA member countries [5] also shows a good correlation with the data obtained by means of the M-3 actinometer and Volz photometer recommended by WMO. To increase the sensitivity and accuracy of measurements of spectral turbidity parameters in situations of high transmission values, as observed at high-mountain stations, one can use the model two-channel filter pyrheliometer designed at the Main Geophysical Observatory (MGO) [13]. The pyrheliometer is almost 3 times more sensitive than that of the Foisper actinometer and of an order of magnitude higher than that of the M-3 actinometer. In addition, it provides absolute values of direct solar radiation. Activities are also in progress on the determination of precipitation chemistry. To determine heavy metals in precipitation, MGO has devised a flotation absorption method of analysis which produces a considerable increase in the coefficient of concentration and reduces the duration of analysis. It is primarily devised for the determination of toxic trace elements, such as Pb, Cd, Hg and As. Measurement of the concentrations of a number of pollutants in the atmospheric surface layer have recently been started at the network of background stations. In such cases, when concentrations are determined at a fixed level, it is necessary to measure very small amounts of pollutants. In this connection new sensitive methods have been devised in MGO to determine pollutant concentrations by air sampling with absorption instruments based on solid-state film sorbents [4, 8]. The latter are glass tubes containing a layer of glass beads which is impregnated by a specially selected solution before the analysis. Their advantages are easy preparation, small size and low weight, complete desorption of substances absorbed in the process of sampling, no risk of sample contamination by impurities contained in the sorbent.
URGENT PROBLEMS OF IMPROVING BACKGROUND AIR POLLUTION MONITORING SYSTEMS
271
Sampling on solid sorbents followed by photometric analysis is now becoming widely used in the USSR for urban air pollution monitoring, i.e. for monitoring at the impact level. Techniques have already been worked out to use this approach for the determination of SO2, H2S, NO2, HF, NH3, Hg, organic acids, etc. The sensitivity of the technique is very high, which allows its application, particularly after a certain modification at background stations, to determine SO2 and NO 2 concentrations. Examination of the technique has shown that when sampling 10 m 3 of air, the measurement sensitivity for SO2 is 0.05 gtg/m3, which is lower than the concentration of the pollutant in the cleanest areas over the ocean [12]. An intercomparison of the methods used for measuring background concentractions of SO2 in Bulgaria, GDR, Poland, USSR and Czechoslovakia was carried out in 1980 in Poland. Similar intercalibration exercises for NO2 were conducted in 1983 in GDR. In the USSR, air samples to be analyzed for SO2 and NO2 were collected in sorption tubes. In Bulgaria (for SO2), GDR, Poland and Czechoslovakia (for NO2), samples were taken in absorbing air bubblers. In Poland and Czechoslovakia impregnated filters were also used. Samples were analyzed by photometric techniques. When SO2 was determined according to the procedures developed by the USSR, Bulgaria and Czechoslovakia, pararosaniline and formaldehyde reactions were used. Procedures presented by Poland required thorin reactions. NO2 was determined using Griss-Salzman reaction. When air was analyzed for a given concentration, the results obtained by means of different methods showed a good agreement. At the same time, it was established that, given equal sampling periods, solid sorbent methods were more sensitive than other methods. In particular, NO 2 concentrations determined using solid sorbents, were of an order of magnitude lower than measured otherwise. The USSR methodology provided a detection limit of 0.3 ~tg/m 3 for air sample volume 2 m 3. The Polish and Czechoslovak detection limits were 3 and 4 ~tg/m 3, respectively. Measurements of sulphates according to the MGO procedure based on thermal decomposition of sulphates to sulfur dioxide and coulometric determination of the latter, have been started at background stations. The procedure is more sensitive than the widespread photometric thorin method and is much simpler, requiring only two operations instead of the 14 in the thorin technique. With the 103 m air sample, the sensitivity of the procedure is 2 ~g/m 3 and the measurement error is 15~ 3. Analysis of Results Statistical analysis of the monitoring data from the BAPMoN program identifies differences in background parameters due to the location of stations as well as a number of year-to-year variations. Some of the results obtained were given in [7] and other papers. The use of spectral radiation data from the USSR background stations in conjunction with material published by the International WMO Centre in Ashville [14] allows a comparison to be made between mean annual values of aerosol turbidity
272
M. E. BERLYA.ND ET AL.
m
~udaM (~) TSOO~,0,~0
Turke g (G
~ndonesco(()
0400
