SOURCE APPORTIONMENT OF HUMAN EXPOSURE TO PARTICULATES IN MUMBAI, INDIA Paper # 1035 Milind M. Kulkarni Civil Engineering Department, Sardar Patel College of Engineering, Andheri(West) Mumbai 400058 INDIA Phone: +91 022 26289777, 26241451 Extn. 221 Fax.: +91 022 26237819 Email: milind.kulkarni@vsnl.com ABSTRACT Mumbai is considered as the commercial capital of India and is one of the important cities in the world. The previous studies have shown that particulate matter is the major air pollutant for Mumbai. In this study, which was funded by AICTE, Government of India, human exposure to Respirable Particulate Matter(PM5) was assessed for respondents from the selected population subgroups viz. Low Income Group (LIG) and Middle Income Group (MIG). The respondents were monitored for 48 hour integrated exposure with the help of personal samplers (SKC make). Activity diary was recorded and questionnaire survey was carried out to collect information about workplace and home characteristics. The sampling period was October 2002 to January 2003. Each respondent was sampled for once or twice a week. For LIG respondent, the average personal exposure to RPM was 186 µg/m3 (n=8, Std.Dev.=184) whereas for MIG respondent, it was 73 (n=14, Std. Dev.=39). The Indian National Ambient Air Quality (NAAQ) standard was exceeded by 86% for LIG respondent. The respirable dust collected on polycarbonate membrane filter was digested and further analyzed for metals Copper, Manganese, Lead, Sodium, Chromium, Calcium, Potassium, Zinc, Iron and Arsenic using Atomic Absorption Spectrophotometer. The average concentration of Lead of 0.284 µg/m3 is much below Indian NAAQ standard of 1. The data on concentration of these metals were used for Source Apportionment using Factor Analysis technique. The analysis resulted in three factors (represented by groups of metals associated with the source) related to industrial, residential and crustal sources for all samples combined. Separate factor analysis resulted in two and three sources for MIG and LIG respondents respectively. The third source for the LIG respondent represented metals due to residential sources such as burning of low quality fuel. The ambient air concentration for PM10 for the area was 291 µg/m3 (n=10). Keywords: Personal exposure, Factor analysis, Respirable Particulate Matter, LIG, MIG INTRODUCTION Of late deteriorating air quality in urban areas has become a major public concern in most of the countries. The majority of South Asian cities suffer from extremely high levels of urban air pollution, particularly in the form of small particles. Region-wide, urban air pollution is estimated to cause 250,000 deaths and billions of cases of respiratory illnesses every year. 1 A study indicated that approximately 50,000 pre-mature deaths occur annually due to PM10 (particles smaller than 10 microns) pollution in India.2 Ambient air quality is routinely monitored by regulatory authorities in the major urban areas. However, there is a little data available on the personal exposure to air pollution which is closely associated with the health risk than ambient air quality. Assuming that the air pollution concentrations are homogeneous both in space and in time, within each microenvironment, the integrated exposure for an individual (i) is computed as the sum of the product of the pollutant level in each microenvironment(j) and the time the individual spent in the microenvironment. J Ei = ∑ Cj tij ………….. (1) 1 Where Ei = integrated exposure of an individual(i) over the time period Cj = pollutant level experienced in microenvironment (j) tij = time spent by individual (i) in microenvironment (j) J = total number of microenvironments occupied by individual(i) over the time period Average exposure of a person is closely related to the integrated exposure and can be computed by dividing it by the averaging time. Average exposure is also called as the time weighted concentration and is measured in µg/m3. Different average times are used in the calculation of exposure to air pollution. For example, total daily exposure often refers to 24-hours exposure. Author of this paper conducted one of the first studies of its kind in India in which personal exposure to various population subgroups in urban areas was measured.3 Source apportionment identifies different sources of air pollution and their relative contribution. There have been a few studies on source apportionment of ambient particulates.4 Recently, a study5 was conducted by the Georgia Institute of Technology, United States in collaboration with National Physical Laboratory, the Indian Institute of Technology, Mumbai and the National Environmental Engineering Research Institute, India in which analysis of ambient PM2.5 (particles smaller than 2.5 microns) was carried out using chemical mass balance receptor modeling. The results of this study indicate that there is no single dominant source, but sources differ by location and season among the three Indian cities examined. However, no study has been carried out in India on the source apportionment of personal exposure to particulates which makes this study first of its kind. Source apportionment will quantify the contributions of various indoor and outdoor sources to the personal exposure by statistical analysis and this is a major input for policy makers. In this study, personal or human exposure to Respirable Particulate Matter(PM5) was assessed for respondents from the selected population subgroups viz. Low Income Group (LIG) and Middle Income Group (MIG). The respondents were monitored for 48 hour integrated exposure with the help of personal samplers (SKC make). Activity diary was recorded and questionnaire survey was carried out to collect information about workplace and home characteristics. The sampling period was October 2002 to January 2003. Each respondent was sampled for once or twice a week. The respirable dust collected on polycarbonate membrane filter was digested and further analyzed for various metals using Atomic Absorption Spectrophotometer. The data on concentration of these metals were used for Source Apportionment using Factor Analysis which is a statistical technique. However, the respirable dust was not analyzed for nonmetals and organic components in this study. To summarize objectives of this study are as follows: 1. To assess the human exposure to particulates for the selected population subgroups in Mumbai. 2. To assess the human exposure to toxic/nontoxic metals. 3. Source apportionment of Human exposure to particulates using factor analysis. 4. To identify the risk factors and propose guidelines for policy making. STUDY DESIGN Personal exposure monitoring studies involve considerable expenditure in terms of manpower and equipment. Therefore, a careful study design is imperative. Area of study The study is conducted in Mumbai, the commercial capital of India, a city with a total population exceeding 14 million and located on the west coast of India. In this study, Andheri, a western suburb of Mumbai has been selected as the area of study because with a typical mix of roads with high traffic density, industrial estates and residential areas, it represents the city. The study was conducted in two phases. In the first phase, the concentration of respirable particulate matter RPM (PM5) was measured with the help of personal samplers at breathing level at following locations selected on the basis of sources:1. Campus of Sardar Patel College of Engineering: To represent the background air quality. 2. Apana Bazar Traffic Junction, Andheri(West): To represent air pollution due to automobile exhaust and resuspension of dust from roads. 3. Hotel Sea View, Juhu Beach: To represent natural/marine sources 4. Indoor home: To represent indoor residential sources 5. Personal Exposure The possibility of locating a station at an industrial area was also explored. However, most of the industries are small scale and they are interspersed in the residential areas. In one area demarcated for industries, most of the establishments were of commercial nature without any specific air pollution source. Therefore, no location could be set up to represent industrial sources. In the second phase, personal exposure was monitored for the selected respondents in the study area. Selection of air pollutant Respirable particulate matter (RPM), Suspended Particulate Matter (SPM), Sulphur dioxide (SO2) and Nitrogen dioxide (NO2) are the criteria air pollutants which are routinely monitored by the regulatory authorities in India. However, in this study RPM is selected due to its high levels in Mumbai 6 and its adverse impact on health as discussed above. In RPM, sampling can be done for three different sizes: viz. 2.5, 5 and 10 microns. In this study, particles less than 5 micron size or PM5 are sampled. Apart from being respirable, comparatively easy availability of PM5 samplers is also one of the reasons. They are widely used in measuring the occupational exposure to RPM all over the world. A study has developed correlation between concentration of different sizes of particulates. 3 The respirable dust was digested and analyzed for metals like Copper, Manganese, Lead, Sodium, Chromium, Calcium, Potassium, Zinc, Iron, Arsenic, Cadmium and Nickel Selection of respondents The study was carried out with respondents from middle and low income groups. The respondents were employees working with the organization conducting study viz. Sardar Patel College of Engineering (SPCE). The personal exposure to RPM was monitored for following two respondents representing middle and low-income groups: 1. Dr. Milind M. Kulkarni Designation: Professor Residence: Dhake Park, Jogeshwari (East), Mumbai Work-place: SPCE, Andheri(West), Mumbai Distance from workplace = 3 km Mode of commuting: Car Approximate commuting time = 40 minutes Middle Income Group = 40 2. Shankar Anant Pujari Designation: Lab Attendant Residence: School Committee Chawl, Andheri(E), Mumbai Work-place: SPCE, Andheri(West), Mumbai Distance from workplace = 2 km Mode of commuting: Walk Approximate commuting time = 40 minutes Low Income Group Sampling time The sampling time for each respondent was 48 hours or two consecutive days. Ideally 24 hour sampling period is enough to measure daily integrated exposure which is a suitable indicator of health risk. However, to get more mass of dust collected for better accuracy 48 hour sampling period is selected. This will also help in the detection of metals many of which are below detection limits. Monitoring period For the first phase (source location monitoring), the monitoring period was from March 21 to June 02, 2002. On an average each location was monitored once in a week. Monitoring period for second phase(respondent personal exposure monitoring) was from October 23, 2002 to January 23, 2003. Each respondent was monitored for once a week on an average. Questionnaire Survey A questionnaire survey was carried out to collect other data such as housing characteristics, fuel use, socio-economic status etc. An activity diary was also maintained. The correlation between this data and the personal exposure is explored. MATERIALS AND METHODS Sampling and Analysis Method Personal exposure to RPM is measured by personal sampler (SKC, USA make). The personal sampler consists of a diaphragm pump and is operated on rechargeable batteries. It is a compact device and can be worn by the respondent. The personal sampler is fitted to the waist belt of the respondent and the cyclone can be clipped to the shirt collar. The inlet of the cyclone is kept near the breathing level of the respondent. The Aluminum cyclone is a lightweight respirable dust sampler that is used with a filter loaded into a three piece filter cassette. The respirable particles are collected on the filter for analysis while larger particles fall into the grit pot and are discarded. In this study 37 mm cyclone with 37 mm three piece cassette are used. So, the size of filter is also 37 mm. Polycarbonate membrane filters with a pore size of 0.8 µm are used. Before and after sampling, the filter is kept in airtight cassette to avoid absorption of moisture. The flow rate is maintained at1.9 lit/min. The flow rate is calibrated before and after sampling along with cyclone loaded with filter. If the flow rate differs by more than 5% after sampling, the concerned observation is considered as void. Sampling protocol The sampling was started at the workplace in the morning of the first day. After duty hours, both the respondents carried the sampler to their residences with the charged pump and the sampling was carried out in the night either in the living room or the bedroom. The cycle was repeated for the next day. The sampler should not be kept near the window or the stove. Polycarbonate filter papers were used for the sampling and the rate of air flow through the cycle was set at 1.9 lit/min. The personal sampler pump used for finding RPM concentration works at a stretch for about 10 to 12 hours. After this, its battery gets exhausted and recharging is necessary before further usage. Each respondent conducted sampling for two whole days (48 hours). On the first day the cyclone is attached to the pump which runs for 12 hours after which it is kept for charging during night time. The same cyclone is then attached to another freshly charged pump. The procedure is repeated for the second day and thus 48 hours of sampling is done. Digestion of respirable dust In order to determine the concentration of metals, the respirable dust collected on the Plolycarbonate filter is digested for further analysis. The procedure used for digestion is as follows: The 37 mm diameter filter was kept in a Teflon beaker of 200 ml capacity. It gets dissolved in 5 ml of Chloroform. To digest the dust refluxing was carried out by filling the beaker with nitric acid (5 ml), hydrochloric acid (5 ml), perchloric acid (2 ml) at 80 to 90 0 C till the dust is completely digested to incipient dryness. To the cooled mass are added 2-3 drops of concentrated nitric acid and clear solution was made up to volume by N/10 hydrochloric acid. A blank was prepared following the same procedure using identical quantity of unexposed filter and acid. Factor Analysis for Source Apportionment Factor analysis attempts to identify underlying variables, or factors, that explain the pattern of correlations within a set of observed variables. Factor analysis is often used in data reduction to identify a small number of factors that explain most of the variance observed in a much larger number of manifest variables. Factor analysis can also be used to generate hypotheses regarding causal mechanisms or to screen variables for subsequent analysis. Principal component and common factor analysis are often placed under the heading Factor Analysis. Although they are based on different mathematical models, they can be used on the same data and both produce similar results. These procedures are often used in exploratory data analysis to Study the correlation among a large number of inter-related quantitative variables by grouping the variables into a few factors; after grouping, the variables within each factor are more highly correlated with variables in that factor with variable in other factor. Interpret each factor according to the meaning of the variables. To find Outliers i.e. Data not conforming to the general trend. Steps in Factor Analysis There are three main steps in factor analysis: 1. The correlation matrix is computed. If the variable has very small correlations with others, you may consider eliminating it from the list of variables, as it is not significant. 2. The factor loadings are estimated. Here principal component analysis is used as a method for factor extraction. 3. The loadings are rotated to make them more interpretable. Rotation methods make the loading of each factor either large or small, not in between. Here, Varimax rotation is used. Data used in Factor Analysis In this study two respondents are monitored for personal exposure to RPM for the period of three months. Each respondent was monitored for once a week on an average. In all 22 valid samples were generated. Each sample filter was digested and analyzed for 12 metals. This data is used in the factor analysis. RESULTS AND DISCUSSION As discussed in the study design, the air monitoring was carried out in two phases. In the Phase I, the personal exposure is monitored at selected sites representing different sources such as traffic, natural, indoor, background etc along with the personal level monitoring. In Phase II, personal exposure to RPM was monitored for two respondents. One of the respondent was from the middle income group, whereas the other was from the low income group. Personal Exposure Monitoring at Selected Source Sites (Phase I) In this phase, though personal exposure to respondents was not measured, the term personal exposure monitoring is used to indicate that at selected sites, RPM (PM5) was monitored using personal samplers and hence the sample will represent the daily personal exposure at the particular site. The monitoring period was from March 21 to June 02, 2002 and the sampling period was 48 hours. On an average each location was monitored once in a week. Seven valid readings were obtained for each site. The results are summarized in Table 1. Table1. Average RPM (PM5) concentration at Selected Source Sites (n=7) Site Personal Indoor-home Traffic Junction Background Sea shore RPM (PM5) concentration in µg/m3 149.71 ± 83.68 125.68 ± 73.16 254.37 ± 73.85 150.62 ± 78.58 111.65 ± 33.73 The traffic junction site recorded the maximum whereas the sea shore site recorded the minimum RPM concentration. The indoor-home and the background concentration falls in between. Apparently, the air quality at the background site is affected by the traffic at the adjoining roads. Hence, the sea-shore site better represents the background air quality. The personal exposure falls between the indoor-home and traffic junction air quality. The excess exposure than indoor-home may be due to outdoor sources such as traffic. The Indian NAAQ for RPM(PM10) is 100 µg/m3. This exceeds at all the stations. The respirable dust samples collected at the source stations were digested and analyzed for metals by the Atomic Absorption Spectrophotometer. The results are as given in Table 2. Table 2. Average concentration of metals at selected source stations (n=7) Site Mn (µg/m3) (Mean ± std dev). Traffic 0.6213(1) Cd (µg/m3) (Mean ± std dev). 0.0033 ± 0.0020(3) Cr (µg/m3) (Mean ± std dev). 1.4299 ± 1.6352(3) Cu (µg/m3) (Mean ± std dev). 0.1355 ± 0.0717(3) Fe (µg/m3) (Mean ± std dev). 5.9302 ± 7.8116(7) Na (µg/m3) (Mean ± std dev). 1.4418 ± 0.8892(5) Pb (µg/m3) (Mean ± std dev). 0.5025 ± 0.3264(5) Zn (µg/m3) (Mean ± std dev). 0.2723 ± 0.1669(5) Background Not detected 0.0114 ± 0.0079(4) 1.5476 ± 1.1775(4) 0.0441 ± 0.0184(3) 4.5804 ± 6.2993(7) 1.3870 ± 0.5247(5) 0.9000 ± 0.1102(4) 0.1484 ± 0.0846(4) 2.5310(1) 1.7177 ± 0.0323(2) 0.1256 ± 0.0356(2) 0.0525 ± 0.0355(2) Not detected 0.0319(1) 2.6315 ± 1.4224(7) 10.8707 ± 11.5185(5) 4.8055 ± 5.1964(6) 1.0979 ± 0.8825(3) 1.5652 ± 1.1452(6) 1.5983 ± 1.3759(7) 1.0576 ± 0.4749(2) 0.5013 ± 0.2906(4) 0.2033 ± 0.2681(2) 0.1507 ± 0.1551(2) 0.1850 ± 0.1551(4) 0.2786 ± 0.3747(2) Personal 0.0091(1) Indoor-Home 0.0462(1) Sea shore 0.0785(1) 0.0061 ± 0.0026(3) 0.0045 ± 0(3) 0.0069 ± 0.0046(4) The results show some new trends. For example, Lead is supposed to be the tracer element for traffic. However, since last five years, unleaded petrol is served in India at all petrol pumps. Due to this, the traffic site is not associated with high levels of Lead. Apart from Na and Zn, the concentration of all other metals at sea-shore site is lower than other that of the Background site. Thus, the result of PM5 concentrations, that sea-shore site better represents background air quality, is confirmed by the results of metal analysis. It was found that manganese was rarely detected at any of the sites. Though cadmium was detected at all the sites, the concentration was very low and also there wasn’t much difference in the concentration to credit the cadmium concentration to one particular type of source. The concentration of chromium was also almost the same at every site but what is significant is the absence of chromium from the sea-shore site. This indicates that chromium contribution may be solely due to anthropogenic sources. The concentration of copper was the highest at the traffic junction, which may indicate its origin from auto exhaust. The concentration of iron was the same almost everywhere but it was also found that in couple of homes the concentration was very high. The concentration of sodium was almost the same everywhere but was slightly greater at the sea-shore site confirming the expectation that sodium has got a marine origin. The concentration of lead was almost the same at almost all the sites but was significantly low at the marine front underlining its anthropogenic origin. Amongst the metals, the Indian NAAQ standard is set only for lead, which is 1.0 3 µg/m . It has been observed that the personal exposure to lead (1.0576 µg/m3) is slightly exceeding the set standard and at all the other sites the concentration of lead conforms to the standards. The higher personal exposure to lead compared to other sites needs to be investigated. Personal Exposure Monitoring of Selected Respondents (Phase II) As discussed in the study design, personal exposure to RPM(PM5) was monitored for two respondents – one from middle and the other from lower income group. Both the respondents were working in the same organization. However, as the questionnaire survey revealed their housing characteristics were different. The MIG respondent was living in an apartment like building where the ventilation was better and LPG was used as the cooking fuel. On the other hand, the LIG respondent was residing in a chawl or slum like locality where ventilation was relatively poor and where along with LPG, kerosene and other low grade fuels were used for cooking. Table 3 shows the data-log of the gravimetric analysis for the determination of personal exposure to RPM. Table 3. Determination of personal exposure to Respirable Particulate Matter (PM5) SR.NO. Sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 M9_12 M20_1 S10_12 M19_12 M6_1 S6_1 S23_10 S2_12 S20_1 M16_12 M14_1 M22_1 S19_12 S24_12 M31_12 M30_10 M23_12 M29_11 S28_10 M23_10 M8_1 M12_12 W1 Gms 0.0384 0.03839 0.03827 0.03834 0.03864 0.03876 0.03769 0.03845 0.03804 0.03774 0.03848 0.03815 0.03844 0.03888 0.03823 0.03725 0.03854 0.03804 0.038 0.03416 0.03807 0.0385 W2 gms 0.03866 0.03874 0.03968 0.03883 0.03895 0.03995 0.03816 0.04063 0.0386 0.03791 0.03869 0.03832 0.03903 0.03909 0.03848 0.03762 0.03905 0.03853 0.03921 0.03498 0.03837 0.03869 W gms 0.00026 0.00035 0.00141 0.00049 0.00031 0.00119 0.00047 0.00218 0.00056 0.00017 0.00021 0.00017 0.00059 0.00021 0.00025 0.00037 0.00051 0.00049 0.00121 0.00082 0.0003 0.00019 T1 min 2803 2000 2641 2992 2080 1935 3040 1851 2158 2531 2716 2018 2395 2228 2644 2519 2552 2823 2042 2349 2615 2751 Q1 litre/min 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 1.912 Volume m3 5.360 3.824 5.050 5.720 3.977 3.699 5.812 3.540 4.126 4.840 5.193 3.858 4.580 4.260 5.055 4.816 4.880 5.397 3.904 4.492 5.000 5.260 Concentration µg/m3 48.51 91.53 279.21 85.66 77.95 321.71 80.87 615.82 135.72 35.12 40.44 44.06 128.82 49.30 49.46 76.83 104.51 90.79 309.94 182.55 60.00 36.12 In the sample column, M represents the sample corresponding to the MIG respondent whereas S represents the LIG respondents. The average personal exposure to RPM (PM5) is 110.91 µg/m3 . This exceeds the Indian NAAQS(PM10) of 100 µg/m3 by a factor of 1.11. If PM5 concentration is converted to corresponding PM10 concentration, this factor will increase further. However, in the case of MIG respondent, the average RPM (PM5) exposure was 73.11 µg/m3 and in the case of LIG respondent it was 240.17 µg/m3. This shows that due to their poor residential microenvironment as discussed above, the LIG respondents are exposed to higher dose of air pollutants. The filters with respirable dust were digested and analyzed for metals viz. Cu, Mn, Pb, Na, Cr, Ca, K, Zn, Fe, As, Ni and Cd. Out of these Nikel (Ni) and Cadmium (Cd) could not be detected. The results are shown in Table 4. Table 4. Personal exposure to metals through Respirable Particulate Matter in µg/m3 Sample M9_12 M20_1 S10_12 Cu µg/m3 0.252 0.999 0.501 Mn µg/m3 1.996 7.296 2.436 Pb µg/m3 0.000 0.549 0.475 Na µg/m3 102.332 202.694 107.525 Cr µg/m3 19.030 37.918 22.554 Ca µg/m3 71.866 41.056 51.465 K µg/m3 10.373 26.281 13.010 Zn µg/m3 4.030 6.172 4.851 Fe µg/m3 9.869 60.277 18.495 As µg/m3 5.466 14.540 35.683 M19_12 M6_1 S6_1 S23_10 S2_12 S20_1 M16_12 M14_1 M22_1 S19_12 S24_12 M31_12 M30_10 M23_12 M29_11 S28_10 M23_10 M8_1 M12_12 0.663 1.262 1.357 2.400 1.083 0.288 0.426 0.609 0.783 0.819 0.895 0.364 0.297 1.891 0.174 0.207 0.265 0.218 0.241 3.829 6.512 7.000 12.543 4.016 2.375 5.310 2.869 4.821 3.734 6.253 2.552 2.907 11.475 1.445 1.972 1.581 1.460 3.517 0.350 1.056 1.135 0.275 0.551 0.000 0.000 0.039 0.000 0.044 0.094 0.000 0.561 0.369 0.074 0.077 0.378 0.080 0.133 119.423 166.960 179.459 120.578 156.063 86.573 70.145 101.232 139.036 115.721 125.316 78.556 89.950 85.717 56.142 73.105 63.780 61.740 49.791 23.829 35.253 37.892 41.758 39.016 24.358 27.521 23.975 31.234 28.537 30.234 3.383 24.232 40.594 21.290 24.923 23.353 22.300 24.696 72.867 82.449 88.622 62.681 33.740 102.327 69.979 94.724 109.772 78.122 88.618 41.622 16.819 154.283 112.859 156.557 103.384 108.420 104.981 16.731 20.769 22.324 18.273 27.362 13.257 12.397 19.969 23.613 23.406 25.480 12.226 13.933 22.029 11.525 17.649 15.361 14.160 12.414 3.969 5.507 5.919 4.026 9.213 5.162 4.649 4.410 5.884 5.022 5.504 2.591 4.506 4.693 4.039 5.712 5.298 4.780 4.430 39.668 60.171 64.676 128.407 35.079 19.874 50.537 29.944 48.937 39.061 59.368 22.433 30.959 109.283 13.619 19.749 15.093 14.560 33.479 4.003 20.216 21.730 0.000 4.528 10.834 2.128 7.587 3.447 12.227 6.019 0.000 5.233 0.000 0.000 13.653 2.004 0.000 0.000 Mean SD 0.727 0.588 4.450 3.047 0.284 0.331 106.902 41.687 27.631 8.804 83.964 35.392 17.843 5.356 5.017 1.249 41.979 30.261 7.695 9.061 Factor Analysis SPSS software is used for factor analysis. Descriptive Statistics The table below shows the descriptive statistics for the variables, which are the metal concentrations in the respirable dust. Table 5: Descriptive statistics for metal concentration in µg/m3 (SPSS output) De scri ptive Statistics CU MN PB NA CR CA K ZN FE AS Mean .72697 4.44998 .28362 106.90170 27.63083 83.96424 17.84282 5.01660 41.97903 7.69533 St d. Deviat ion Analys is N .587872 22 3.047261 22 .331283 22 41.687424 22 8.804080 22 35.391805 22 5.355661 22 1.248900 22 30.261390 22 9.060756 22 For the 22 samples in this study, mean concentrations of Cu is 0.726μg/m3, Na is 106.9µg/m3 and so on. The standard deviation of Na is 41.687 while that of Cu is 0.587. Correlation Matrix The values in the Table 6 below show the correlations between the variables Table 6. Correlation matrix for variables (output) Correlation Matrix CU MN PB NA CR CA K ZN FE AS CU 1.000 .941 .440 .546 .787 -.027 .588 .193 .941 .044 MN .941 1.000 .349 .454 .765 .025 .501 .081 .991 -.049 PB .440 .349 1.000 .637 .517 -.260 .361 .389 .311 .523 NA .546 .454 .637 1.000 .606 -.387 .747 .533 .398 .480 CR .787 .765 .517 .606 1.000 .084 .714 .609 .751 .148 CA -.027 .025 -.260 -.387 .084 1.000 -.015 -.062 .049 -.160 K .588 .501 .361 .747 .714 -.015 1.000 .687 .486 .162 ZN .193 .081 .389 .533 .609 -.062 .687 1.000 .046 .250 FE .941 .991 .311 .398 .751 .049 .486 .046 1.000 -.098 AS .044 -.049 .523 .480 .148 -.160 .162 .250 -.098 1.000 These correlations indicate degree and nature of relation in the behavior of the variables. The absolute value indicates the degree of the correlation while the positive or negative sign indicates the nature of the correlation. Negative correlation between Cu and Ca indicates that high values of Cu are associated with low values of Ca, while positive correlation between Cu and Mn indicates that high values of Cu are associated with high values of Mn. Communalities For each variable, the communality is the proportion of the variance of that variable that can be explained by the common factors–the squared multiple correlation of the variable with the factors. In the principal components solution, all initial communalities are 1. After extraction, the estimate for each variable is the multiple R2 from regressing all other variables on that variable. Table 7 below gives the communalities for the variables. Table 7. Communalities for the variables Communalities CU MN PB NA CR CA K ZN FE AS Initial 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 Extraction .949 .980 .678 .849 .900 .628 .816 .876 .988 .578 Extraction Method: Principal Component Analysis. Communalities range from 0 to 1, with 0 indicating that common factors explain none of the variance of the variable and 1 that they explain all the variance. Total Variance Explained The Table 8, in the form of SPSS software output, shows statistics for each factor before and after the components are extracted. For principal components, the initial and extraction statistics are always the same. After rotation the percentage of total variance accounted for by all the factors does not change. However, the percentage accounted for by each factor does change. The eigenvalues for the multivariate space of the original variables are ordered by size. Each value is the total variance explained by a factor. Table 8 Total Variance Explained Component 1 2 3 4 5 6 7 8 9 10 Total 5.019 2.049 1.174 .843 .437 .262 .108 6.466E-02 4.014E-02 4.212E-03 Initial Eigenvalues % of Variance Cumulative % 50.191 50.191 20.487 70.679 11.736 82.415 8.430 90.845 4.368 95.212 2.617 97.830 1.080 98.910 .647 99.556 .401 99.958 4.212E-02 100.000 Extraction Sums of Squared Loadings Total % of Variance Cumulative % 5.019 50.191 50.191 2.049 20.487 70.679 1.174 11.736 82.415 Rotation Sums of Squared Loadings Total % of Variance Cumulative % 3.808 38.080 38.080 2.465 24.654 62.734 1.968 19.681 82.415 Extraction Method: Principal Component Analysis. By default, SPSS computes as many components as there are eigenvalues greater than 1. One criterion for determining the number of useful factors for extraction is to exclude factors with variance less than 1 because they do no better than a single independent variable. Scree Plot As shown in Figure 1, Screen Plot is the display of eigenvalues in the above table plotted against their order (associated component). This is used to identify useful number of factors to be retained by looking for large values that separate well from smaller eigenvalues. Figure 1 Scree Plot 6 5 4 3 Eigenvalue 2 1 0 1 2 3 4 5 6 7 8 9 10 Component Number Component Matrix The Table 9 displays the coefficients (loadings) that relate the variables to the unrotated factors (components). Unrotated loadings and orthogonally rotated loadings are the correlations of the variables with the factors. Generally loadings less than 0.2 are suppressed. Table 9 Component Matrixa CR CU MN FE K NA PB AS CA ZN 1 .915 .888 .834 .808 .801 .795 .640 .262 .530 Component 2 3 .233 -.366 -.499 -.550 .386 .452 .443 .687 -.449 .505 -.267 .643 .583 Extraction Method: Principal Component Analysis. a. 3 components extracted. Rotated Component Matrix The objective of rotation is to make larger loadings larger and smaller loadings smaller than their unrotated values. This helps remove the loadings which do not contribute significantly to the solution. This enables us to name the factors more accurately. The results of the statistical analysis are shown in Table 10 below: Table 10 a Rotate d Com ponent Matrix FE MN CU CR ZN K CA AS PB NA 1 .991 .984 .944 .710 Component 2 .440 .310 .388 .205 .625 .925 .781 .211 .299 .345 .546 3 -.763 .684 .680 .633 Ex trac tion Met hod: Principal Component Analy sis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation c onverged in 5 iterations. Factor analysis on the combined sample of MIG and LIG respondents brings out three principal sources, which are as follows: Source 1: Fe, Mn, Cu and Cr Source 2: Zn, K Source 3: As, Pb, Na and Ca With reference to Table 2, Source 1 represents anthropogenic sources such as traffic and industry. This source is the most dominant source. Source 3 represents group of metals from natural sources such as marine, crustal and road dust. Source 2 represents metals emanating from residential sources such as burning low grade cooking fuel. Factor analysis on the MIG respondent samples resulted in two components as shown in Table 11. Table 11. Rotated Component Matrix for samples of MIG respondent Cu Mn Pb Component 1 0.902 0.906 0.292 2 0.264 0.205 0.736 Na Cr Ca K Zn Fe As 0.301 0.836 0.606 0.704 0.505 0.923 0.139 0.874 0.374 -0.573 0.539 0.532 0.155 0.932 The two sources brought out by the rotated component matrix are: Source 1: Fe, Mn, Cu, Cr Source 2: As, Na, Pb Source 1 represents anthropogenic sources such as traffic and industry whereas Source 2 represents natural sources such as marine, crustal and road dust. Factor analysis on the LIG respondent samples resulted in three components as shown in Table 12. Table 12. Rotated Component Matrix for samples of LIG respondent 1 Cu Mn Pb Na Cr Ca K Zn Fe As 0.941 0.982 0.302 0.761 -0.245 0.201 -0.339 0.992 -0.588 Component 2 0.304 0.157 0.939 0.826 0.410 -0.590 0.205 0.327 0.410 0.458 -0.207 0.862 0.83 0.487 -0.617 3 Source 1: Fe, Mn, Cu, Cr Source 2: Pb, Na Source 3: K, Zn In the case of LIG respondent, first two sources are same as in the case of MIG respondents. However, there is the additional third source which may be linked to the residential environment of the LIG respondent and to the use of low grade cooking fuels. CONCLUSIONS 1. Air monitoring at the selected source sites revealed that RPM(PM5) concentration is high and it exceeds the Indian NAAQ standard. The sea-shore site better represented the background air quality. Background site was apparently affected by anthropogenic sources. Site RPM (PM5) concentration in Personal Indoor-home Traffic Junction Background Sea shore µg/m3 149.71 ± 83.68 125.68 ± 73.16 254.37 ± 73.85 150.62 ± 78.58 111.65 ± 33.73 2. Metals’ concentrations were also lower at the sea-shore site. Analysis of metals’ concentrations in the respirable dust provided some guidelines for the source apportionment. Lead was not associated with the traffic site. Manganese, Chromium, Copper and Iron were linked with the anthropogenic sources such as traffic. Sodium was associated with marine source. Zinc can be associated with traffic as well as well as indoor residential site 3. The concentration of elements in the Respirable Particulate Matter (PM5) in µg/m3 is as follows: Element Cu Mn Pb Na Cr Ca K Zn Fe As MIG Samples (n =14) Average Std. Dev. 0.701 0.51 4.694 2.90 0.234 0.33 107.057 46.82 26.339 10.26 86.547 32.86 17.360 5.42 4.647 0.99 43.560 27.71 5.217 6.64 LIG Samples (n = 8) Average Std. Dev. 0.738 0.21 4.141 1.94 0.204 0.24 116.19 8.91 27.109 4.03 72.74 19.15 20.632 6.68 5.126 0.34 38.975 20.44 17.976 15.65 Combined Samples (n=22) Average Std. Dev. 0.727 0.59 4.450 3.05 0.284 0.33 106.90 41.69 27.631 8.80 83.96 35.39 17.843 5.36 5.017 1.25 41.979 30.26 7.695 9.06 The level of average Personal exposure to Lead is 0.284 µg/m3 which is below the standard of 1 µg/m3. 4. The microenvironment of a person has a significant effect on personal exposure to particulates. Factor analysis on the combined sample of MIG and LIG respondents brings out three principle sources. Source 1 corresponds to group of metals of industrial or anthropogenic(traffic) origin. Source 2 apparently correspond to group of metals emanating from residential sources such as heating, burning of low grade cooking fuel etc. Source 3 consists of metals which are crustal elements(Ca), elements from natural sources(Na) and which may also present in the resuspended road dust. Factor analysis for samples of MIG respondent separately brings out two principal sources. Source 1 corresponds to group of metals of anthropogenic origin such as traffic and industry. Source 2 corresponds to group of metals from natural sources such as marine, crustal and road dust. Factor analysis for samples of LIG respondent separately brings out three principal sources. Source 1 and 2 are the same as in the case of MIG respondent. However, there is an additional third source which corresponds to the residential environment of the LIG respondent. This source may be traced to burning of low grade cooking fuel. 5. The air pollution due to anthropogenic sources such as traffic, industries is the major source contributing to the personal exposure to respirable particulates. Therefore, efforts should be made to reduce air pollution due to these sources. In India, there are many old vehicles (more than 15 years old) still running on the road. These should be gradually phased out. Transportation system should be improved to reduce idling time of vehicles at the traffic signals. There are many small scale industries of unauthorized nature. They should be identified and brought under the purview of the regulatory authorities. Low income group population is exposed to additional dose of air pollutants due to their poor residential environment. Efforts should be made to spread the use of higher quality fuels such as LPG in these localities. Use of low grade cooking fuel such as kerosene, bio-fuel such as wood should be discouraged. Awareness about necessity of good ventilation should be increased. ACKNOWLEDGEMENT The financial support provided by the All India Council for Technical Education, New Delhi under Thrust Area Programme in Technical Educaion (File No. : 8018/RDII/TAP(289)/99-2000) is greatly acknowledged. The support provided by my students Mr. Zaheer Shaikh, Mr. Lokesh Padhye, Mr. A. Krishna Mohan, Miss Kavita Deshmukh, Mr. Hariharan Iyer, Mr. Anand Kulkarni and Miss Prashanti Shetty in the sampling and analysis work is also acknowledged. REFERENCES 1. The World Bank – South Asia Urban Air Quality Management Home Page. See http://www.worldbank.org/sarubanair (Accessed January 2005). 2. Brandon, C.; Homman, K. The cost of inaction: valuing economy-wide cost of environmental degradation in India. Paper presented at the United Nations University Conference on the Sustainable Future of the Global System, Tokyo, 16-18 October 1995 3. Kulkarni, M.M. Ph.D. Thesis, IIT Bombay, Mumbai, November, 1998 4. Sharma,V.K.; Patil, R.S. Monitoring and Modeling of atmoshpheric aerosols in an industrial region of Bombay, Environmental Technology, 13, 1043-1052 5. South Asia Urban Air Quality Management Briefing Note No. 14. The World Bank, August 2004 6. Kulkarni, M.M.; Patil R.S. Monitoring of daily integrated exposure of outdoor workers to respirable particulate matter in an urban region of India. J. Environmental Monitoring and Assessment, 1999, 56, 129-146.