10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
The Extinction Coefficient of the Aerosol over the Denver Metropolitan Area: Comparison with a Historic Data Set
HELMUTH HORVATH, Paulus S. Bauer, University of Vienna, Faculty of Physics, AEP
Abstract Number: 940 Working Group: Remote/Regional Atmospheric Aerosol
Abstract Two identical series of optical measurements with the University of Vienna Telephotometer have been performed between July 23 and August 11 in the years of 1979 and 2017. Despite of being 38 years apart the same instrument and the same visibility targets have been used. During these 38 years the population of Colorado almost doubled from 2.8 to 5.6 Million, Boulder County's population increased from 130,000 to 300,000; Broomfield's population increased by a factor of more than 3, Denver City's population increased from 520,000 to 710,000, other surrounding communities also had a considerable population growth. Since human activities are also sources of particulate and gaseous emissions an increase in emissions has to be expected. On the other hand a series of new technologies has been introduced, e.g. exhaust gas cleaning of cars, thus a reduction in emissions per capita could be expected. Comparing the two datasets allows to see the result of the two counteracting effects.
The telephotometer was positioned on Gunbarrel Hill slightly north of Boulder and the spectral radiance of Mt. Morrison, slightly Southwest of Denver, was measured. Comparing the measured radiance with the radiance of the horizon behind the mountain, the extinction coefficient along the sight path can be determined. This sight path is between 100 and 800 m above the Greater Denver Area, and definitely also influenced by emissions in this area.
The weather conditions showed considerable variability, but this was the case in both years. Measurements performed during rain or fog (if possible at all) were excluded in the analysis reported below, so humidity growth of deliquescent particles needs not to be accounted for. Sometimes considerable haze screened the target, so optical measurements were not possible. In 1979 there was one day with considerable haze, so that no observations were possible. In 2017, there was also one day with considerable haze, but on about half of the days the extinction coefficient increased during the day and in the afternoon the target was obscured. This normally occurs when the mixing layer expands due to heating of the atmosphere by insolation.
The obscuration of the target makes a comparison difficult but still some conclusions can be drawn: Based on hourly measurements between 8:00 and 17:00, in 1979 the target was invisible 12 times during the measuring period. In 2017 it was obscured 29 times.
When being visible, the average measured extinction coefficient at 550 nm in 1979 was 46 Mm-1 with a standard deviation of 16 Mm-1. For 2017 the corresponding values are 44 Mm-1 ± 18 Mm-1, which can be considered the same value within the fluctuations. But this is only half of the story, since times where observations could not be performed were neglected. Therefore we included the data where no measurement could be performed as 100 Mm-1, which is the minimum extinction coefficient needed to obscure the target. Obviously its value could be larger, the averages reported below are minimum values. Using these values, we obtain for the 1979 period an average of 51 Mm-1 with a standard deviation of 22 Mm-1 and for the 2017 period 60 Mm-1 ± 29 Mm-1. This can also be considered as indication of an increase in atmospheric extinction coefficient from 1979 to 2017.
The measured aerosol extinction coefficient was wavelength dependent, for about 85 % of the measured spectral extinction coefficients the Angström approximation (the extinction coefficient being proportional to λα with α the Angström exponent) could be applied. The Angström exponent is an indication for the mean size of the particles, the smaller α, the smaller the particles.
For the measuring period in 1979 an average α of -0.87 with a standard deviation of 0.36 was found, for the 2017 period the average was α=-1.24 ± 0.3. Therefore the particles in the air of 1979 were larger than in 2017. This is reasonable, since e.g. particles emitted by modern cars have become much smaller over the years.