Evaluation of High-Volume Air Samplers for
Ambient Air Quality Monitoring in Real Atmosphere
Ritesh K. Singh*, D.S. Ramteke, G. H.
Pandya, M. N. Wahale, S.R. Verma, Y. J. Baseshankar
Environmental Impacts and Risk Assessment
Division, National Environmental Engineering Research Institute, Nehru Marg,
Nagpur -440020, India.
*Corresponding Author E-mail: riteshsingh248@gmail.com
ABSTRACT:
Quality Assurance/Quality Control techniques have been studied in
order to establish an appropriate level of quality control (QC) during ambient
air quality monitoring in real atmosphere. QA/QC study of four high-volume air
samplers was carried out at the roof of an ambient air quality site during the
Environmental Impact Assessment study. Suspended Particulate Matter (SPM),
Sulphur dioxide (SO2), Oxides of Nitrogen (NOx) was
continuously monitored for 24 hours for 8 days. The average SPM, SO2
and NOx measured by an identical pair of samplers ranged between
79.98 to 82.62 µg/m3; 30.40 to 36.75 µg/m3 and 31.63 to 36.36 µg/m3.
The standard deviation and coefficient of variation for SPM, SO2 and
NOx were between 13.59-16.69; 8.34-10.14; 7.81-10.01 and
16.45-20.66; 22.71-33.35; 24.68-30.80 respectively. The day-to-day variation
and sampler-to-sampler variation was also studied and control chart was
prepared for accuracy and precision of the measurements. The QA/QC data
generated in the field will be used to monitor reproducibility (precision) of
the sampling methods and the accuracy of measurement. Proper quality assurance
and quality control (QA/QC) procedures helps in achieving reproducible and
reliable results during the ambient air quality monitoring. The reproducibility
of the samplers for all the parameters is found to be satisfactory.
KEYWORDS: QA/QC, High Volume Samplers, Ambient Air
Quality, EIA, Reproducibility, Control Charts
INTRODUCTION
Air pollution has become a matter of grave
concern, particularly in mega-cities and urban areas, where the situation is
alarming and becoming more and more severe day by day. Ambient Air Quality
Monitoring (AAQM) in the vicinity of industrial development site is an
indispensable part of Environmental Impact Assessment studies.
Quality Assurance (QA) and Quality Control
(QC) comprise of the complete sequence of activities, which assures that a
measurement meets defined standards of quality with a stated level of
confidence. Proper QA/QC practice is necessary to ensure data integrity and guarantee
the data quality required for meeting the overall objectives of monitoring
effort.1 Quality Assurance (QA) in air pollution measurements
requires a set of operating procedures for sample collection and analysis to
produce data of known and defensible quality. In order to confirm that such
procedures are working properly, a field evaluation sampling program is
essential.
Evaluation study of high volume air
samplers was carried out to sample a real atmosphere for Suspended Particulate
Matter (SPM), Sulphur dioxide (SO2) and Oxides of Nitrogen (NOx).
It is a normal practice to have a known
sample to test the methodology of the analysis during any Quality Assurance
program.2 In the present study it was difficult to have an
externally supplied standard dust atmosphere, which could be used for test
purpose. Hence it was decided to make measurements by sampling a real
atmosphere at one of the ambient air quality monitoring site. The paper reports
the results of evaluation of high volume air samplers over period of time and
discusses the results in terms of statistical analysis to support the QA
procedures.
MATERIALS AND METHODS:
Experimental Set Up
The experiment is carried out at one of the
ambient air quality monitoring location to maintain uniformity in measurements.
A large area on the roof of a single storied building was selected and four
samplers were assembled at this location so that the same atmosphere is sampled
simultaneously. The site was easily accessible for calibration, starting and
closing of the samplers daily. The samplers were placed on a single row with 5m
spacing between them. Care was taken that no sampler was placed closer to the
edge of the roof than 5m. Thus, avoiding turbulence that might exist at the
edge or the corner of the building.
The samplers were started simultaneously
and the initial flow on the manometer was recorded. The starting time of the
instruments was also recorded. At the end of 24 hours, final flow rates and
closing time was recorded. The filters and impingers were carefully removed and
new filter and impingers were installed. Entire sampling procedure was then
repeated each day for each sampler. All the filters and impingers were brought
to the laboratory, conditioned for 24 hour and then analyzed.
Calibration
In air quality monitoring the pollutants is
determined per unit volume of air as standard conditions and flow measurement
plays a vital role, as it is inversely proportional to the pollutant
concentration and directly proportional to the sampling time. The total volume
of air sampled is calculated as:
Total Volume = Flow rate X Total sampling time
Therefore the possibility for cumulative
error depends upon the variations in flow rate with time and total time of
sampling.
In High Volume samplers, orifice is used to
measure airflow rate passing through the sampler. Usually a Rota meter or manometer
across the orifice is employed to measure pressure drop and this pressure drop
is correlated to the volumetric flow rate. Because only a small portion of the
total air sampled passes through the Rota meter/ manometer during measurement,
therefore, it must be calibrated against actual airflow with the orifice
calibration unit (OCU). The OCU in turn is calibrated against the positive
displacement primary standard.
In order to calibrate the flow measuring
system of a High-Volume sampler it was necessary to vary the flow rate over the
working range of the sampler. This was done by using perforated plates in place
of the filter to vary resistance to flow during sample collection. A U-tube
water manometer was connected across the orifice in the High Volume Air
sampler. The Orifice was calibrated with a top loading device using a series of
perforated plates. The top loading device was kept over the stainless steel
mesh of the hopper and tightly fitted with the help of gaskets. A resistance
plate with minimum number of holes was first kept in the cylindrical portion of
top loading device. The nozzle of this device was attached to a manometer. The
pressure drop across the orifice was measured with another water manometer.
When the sampler was started air was drawn through the orifice at the top of
calibrator and passes through the resistance plates and subsequently through
the orifice in the sampler. The difference in pressure levels was recorded by
two manometers for each resistance plate tested. From the standard calibration
chart of the calibrator, flow rates were calculated for different resistance
plates. The results were in turn used to calculate the flow rates for orifice
used in the sampler.
RESULTS AND DISCUSSION:
The variability in ambient air quality monitoring
data will be mainly arising due to instrumental and operational error. During
the monitoring of environmental parameters such as SPM, SO2 and NOx
by high volume samplers in ambient air, the instrumental error comes from
differences in sampler performance (Motor speed, shelter geometry, flow rate
variations) and difference in filter quality. The operational error/measurement
errors comes in either initial or final weighing, loss of collected sampling
during handling, error in recording flow meter readings or calibration and
sample analysis.
The above results clearly point out that
the performance of all the samplers is identical and one can choose any one of
them as reference. The monitoring results thus obtained, helps in achieving
reproducible and reliable data during the ambient air quality monitoring.
It is well known that the SPM, SO2 and
NOx concentrations very much depend upon the flow rate at which the
samples are collected. Hence during the sampling flow rate of each sampler was
also monitored continuously. Table -4 summarizes the machine flow rate, SO2
flow rate and NOx flow rate variation between four samplers
over a period of 8 days. The average machine flow rate, SO2 flow
rate and NOx flow rate between samplers over the study period varied
between 1.19 to 1.24, 0.47 to 0.54 and 0.46 to 0.53 m3/ min with
standard deviations varying between 0.02 to 0.05, 0.02 to 0.04 and 0.01 to 0.03
respectively. The average machine flow rate, SO2 flow rate and NOx
flow rate within days varied between 1.19 to 1.23, 0.50 to 0.52 and 0.47 to
0.49 m3/ min with standard deviations varying between 0.03 to 0.04,
0.03 to 0.04 and 0.03 to 0.06 respectively. The Relative Standard Deviation for
machine flow rate, SO2 flow rate and NOx flow rate for
each sampler varied between 2.10 to 3.05, 5.54 to 8.50 and 5.46 to 11.69
respectively. All the samplers show a good correlation coefficient (0.99).
Another likely measurement affecting the
concentrations of SPM is the weight of dust collected by each sampler over a
period of 8 days. The weight of dust recorded for each sampler is summarized in
Table 5. The mean weight of dust collected during the study period between
samplers varied between 0.1182 to 0.1198 grams. The standard deviation for each
sampler was fairly consistent between 0.0277 to 0.0311, the relative standard
deviation varies between 23.43 to 26.00. The Correlation coefficient (0.98) of
dust measured between all the samplers shows good correlation.
It is observed that irrespective of the
flow rate measured and the dust deposited, the SPM, SO2 and NOx
concentrations for all the samplers are comparable
Percent Difference
When two independent measurements of the
same characteristics (SPM, SO2 and NOx) are available, it
is convenient to express the results in terms of percent difference.1,2
Hence the SPM, SO2 and NOx results of the first two
Hi-Vol samplers were further analyzed by calculating the Relative Deviation
percent as follows
Relative Deviation (%) = (x-y)/ (x+y) .200
Where x and y are SPM, SO2 and
NOx concentrations observed with Sampler I and sampler II. The
results are summarized in Table-1 to Table-3. The Relative percent Deviation
SPM, SO2 and NOx varies between -4.57 to 13.09, -6.70 to
13.41 and -5.83 to 32.59 respectively.
Another test of significance for bias is to
check the sign of differences of SPM, SO2 and NOx between samplers.
If all differences are positive or negative then one has a considerable doubt
about lack of bias as it would be expected that on an average two would be +ve
and two –ve. if no bias were present. Since out of 8 values 2 or 3 are positive
one infers that there is a bias in Day to Day measurements. This was further
analyzed by constructing Control charts for the paired samplers.
Control Charts
The control charts provide a tool for
distinguishing the pattern of indeterminate (random) variations from
determinate (assignable cause) variation.3 One goal of using a
Control Chart is to achieve and maintain process stability. Process stability
is defined as a state in which a process has displayed a certain degree of
consistency in the past and is expected to continue to do so in the future.
This consistency is characterized by a stream of data falling within control
limits based on plus or minus 3 standard deviations (3 sigma) of the
centerline.4
The accuracy control charts were
constructed for determining the upper and lower control limits as well as upper
and lower warning limits for the SPM, SO2 and NOx
measured by two identical samplers over a period of 8 days simultaneously. The
step by-step construction of control charts require calculation of Mean `X based on duplicate or paired sets of
results.5
The Upper and Lower control limits were
calculated using the following relations:`
Figures 1 to 3 are graphical representation of accuracy control
chart for mean ( ) variation of SPM, SO2 and NOx plotted
on the vertical axis and the horizontal axis representing the Time in Days. It
is observed that the mean SPM, SO2 and NOx concentrations
very much depend on the atmospheric condition on a particular day. Except on 1st
Day (SO2 and NOx), 4th Day (SPM) and 8th
Day (SPM), the concentrations were outside the upper and lower control limits.
The study indicates that the reproducibility in SPM, SO2 and NOx
measurements not only depends on the performance of the sampler but also on the
atmospheric conditions on a particular Day.
Figure 1 : Variation of SPM during Study
period
Figure 2 : Variation of SO2
during Study period
Figure 3 : Variation of NOx
during Study period
3CONCLUSIONS:
The comparative evaluation of high volume
air samplers in real atmosphere is an essential part of Quality Assurance
program before baseline Ambient Air Quality Monitoring in EIA Studies. A number
of steps are involved in measurements right from filter and absorbing media
preparation to calibration of samplers, flow rates etc. and needs to be
optimized to generate quality data which are defendable. It is observed that
irrespective of the flow rates and amount of dust deposited and gases absorbed,
SPM, PM10, SO2 and NOx measured by all the
samplers are comparable. Application of statistical tools to calculate the
control limits and confidence interval of the measurements is useful.
Reliability in measurements can be achieved by adopting proper quality
assurance and quality control (QA/QC) procedures.
ACKNOWLEDGEMENTS:
The authors would like to thank Dr. S. R.
Wate, Director, NEERI, Nagpur for the encouragement in the present work and
permission to publish the work.
REFERENCES:
1. Dixon, W.F, and Massey,JR. F.J. Introduction
to Statistical Analysis, 2nd edition, McGraw Hill, N.Y.,1957, P-124.
2. Youdon, W.J.
Statistical Methods for Chemists, John Wiley and Sons, Inc., New York, 1951.
3. Grant, E.J., and
Leavenworth, R.S., Statistical Quality Control , Fourth Edition, McGraw
-Hill Book Co., New York,1972.
4. Gitlow, H., Gitlow,
S., Oppenheim, A., and Oppenheim, R., Tools and Methods or the Improvement of
Quality. Homewood, IL: Richard D. Irwin, Inc., 1989.
5. Wheeler, D.J., and
Chambers, D.S., Understanding Statistical Process Control (2nd Ed.). Knoxville,
TN: SPC Press., 1992.