**4. Discussion**

As it was mentioned above, a long-term series of bacteria and yeas<sup>t</sup> concentration observations have not been presented in the literature for other regions of the world; therefore, the results presented in Figure 5 refer only to the south of Western Siberia. A similar result is described in [73] for 2014–2018 for fungi yearly averaged concentrations in the air in France: the decrease is about 1.5 times. At the same time, an increase in the average annual concentrations of fungi of the genera *Aspergillus* and *Penicillium* in Derby, UK, in 1970–2003 was recorded in [74]. As for the intra-annual dynamics of changes in the concentrations of bacteria and yeast, there is information in the literature, although such long-term observations that are presented in this paper are also absent in the literature. It should be noted that for di fferent microorganisms (strains, genera, kingdoms), these dynamics can vary significantly. In particular, in many cases, the concentration of culturable bacteria in a warm season is higher than in a cold one [75–86], but there are also opposite examples [80,87–91].

The noted decrease in the average annual concentrations of culturable microorganisms in the south of Western Siberia is probably associated with the ongoing climate changes, which are manifested in a change in the location and power of bioaerosol sources, pathways of predominant transport of bioaerosols in the atmosphere, etc.

There is also no information in the literature about the change over time of microorganism characteristics that are included in the index of potential pathogenicity of culturable microorganisms in the atmospheric aerosol and in the index of their sustainability in the atmospheric aerosol to adverse environmental factors. Similar comprehensive studies have not been conducted previously.

With regard to antibiotic resistance, the literature presents data indicating that there is a tendency to an increase in the number of bacteria resistant to the action of antibiotics over time, see, for example, [92–96]. In the collections of bacteria isolates collected over a long period of time, the latest trend can be traced reliably [97,98]. Moreover, more and more isolates with multiple antibiotic resistances are detected in nature [99–102].

It should be noted that while for ground-based observation points, the obtained data directly provide an assessment of the hazard to humans of microorganisms in the air. Let us remind the reader that a large portion of aerosol particles in the on-ground atmospheric layer originated from local sources. It is di fficult to relate the results to any particular place on the Earth's surface for high-altitude observations. Bioaerosols found in high-altitude samples of atmospheric air originated mainly from long-distance sources and can, due to the stochastic nature of their movement in the atmosphere, reach the surface at completely di fferent points. In addition, during the transfer process, some microorganisms can simply be inactivated, or at least lose their ability to grow under cultivation conditions.

It should be noted that it is not easy to build the dynamics of the monthly change of all the above-mentioned indices because of their very high variability and a small number of identified microorganisms for some months. Very weak dependencies can be observed only.

All the results presented in Section 3 have very high variability for both the average annual values of integral indices and the values of these indices for individual isolates. This is due to the properties of an atmospheric aerosol, which is a complex mixture of particles of various compositions. These particles enter the atmosphere from various terrestrial sources of inorganic, organic or biological nature, and are also formed in the atmosphere during nucleation processes. While in the aerosol, particles are continuously transformed under the influence of changing temperature and humidity, due to the presence of volatile compounds in it, capable of entering into chemical interactions with particles under the influence of solar radiation, etc. These factors can lead to inactivation of microorganisms in the particles. It is also necessary to consider the processes of particle deposition (sedimentation, washout by precipitations). Because of this, the concentration and composition of an aerosol in the atmosphere can vary greatly in samples that are close in time or in distance. For example, in [65], it was mentioned that two air samples taken during aircraft sounding of the atmosphere (a di fference of 1000 m in height and 15 minutes in time) contain radically di fferent compositions of microorganisms. This also leads to the fact that, if the average annual dependencies of the total concentration of microorganisms shown in Figure 5 have a relatively small dispersion, the attempt to construct similar dependencies for individual phylums and lower taxonomic groups was unsuccessful due to the extremely large concentration dispersion [103]. All of the above is particularly pronounced in Siberia with its sharply continental climate, many local sources of bioaerosols (forests, meadows, swamps, industrial production) and the prevailing direction of the winds, which bring a lot of dust (and soil microorganisms to it) from north-western Kazakhstan to Western Siberia. Moreover, the authors are not sure that an increase in the number of samples analyzed for each month (season) will noticeably a ffect a decrease in the variance of the result and, therefore, an increase in the reliability of the dependencies identified in this article.

The approach proposed in [28] made it possible, for the first time, to quantitatively compare the potential danger to humans of the totality of microorganisms in various samples of atmospheric air. Using this unique approach, it was possible to assess the change in potential danger to humans over 10 years of observation at di fferent points in Western Siberia. Currently, the authors are not aware of any other approaches that allow such estimates.

At the same time, the authors do not pretend to the universality of the method used in this article for assessing the hazard to humans of culturable bacteria and fungi located in atmospheric aerosol. The algorithm for estimating integral indices proposed in [28] is not exhaustive and, probably, there are still some characteristics of cultivated microorganisms (morphological, cultural, biochemical, and biophysical) that refine the constructed integral indices. In addition, the weight values of the various

characteristics of microorganisms by which integral indicators are calculated may also need to be clarified, and finally, an approach similar to that proposed for bacteria and yeas<sup>t</sup> was not developed for micromycetes. Therefore, the results obtained in this article reflect the hazard to humans of only bacteria and yeas<sup>t</sup> present in the aerosol, but not for all culturable microorganisms. In addition, not all microorganisms are easily cultivated under standard conditions. The method used simply cannot be implemented for unculturable microorganisms.
