Analyzing Air Pollutant Concentration and Dosage Data

Simplified Methods for Statistical Interpretation of Monitoring Data

“…The actual frequency distributions show no negative concentration values and are usually asymmetrical with the longer tails for the values above the median concentration. The lognormal distribution has been shown to describe more accurately pollutant concentrations by Zimmer and Larsen, 2 Air Quality Section, 3 Larsen, Zimmer, et al, 4 Larsen, 5 and many others. The standard statistical calculations also may be applied to a lognormal distribution if we first transform all the values of concentrations to their logarithms (base 10).…”

“…4 What concentration is exceeded by the highest 1% of the sample population? Answer: A straight line is drawn on Figure 2 from 1% on the right scale through the center scale at SGD = 1.75.…”

Surface Air Pollutant Concentration Frequency Distributions: Implications for Urban Modeling

“…This absence of detailed knowledge of the source strength, pollution, and wind data at the same location, makes it difficult to determine accurately the normalizing constant KQ = K' in Eq. (1). Figure 2 shows the annual hourly wind speed distribution U for San Francisco in 1966 and 1970 comprised of 8639 and 8622 hourly averaged data points, respectively.…”

Significance of Sampling Time in Air Monitoring

“…They also showed that the maximum concentrations plotted linearly against sampling time on log-log paper. In a later more complete study Larsen, et al 3 provided extensive tabulations of their results for all the pollutants. Figure 1 presents a plot from these data in a somewhat different form.…”