— Have any seasons greatly surprised you?
— Yes, quite a few to be honest. For example, in 2012 there was so much pollen that it was collected in several layers in our samples. There were incredible concentrations of pollen. There were 20,000 pollen grains per cubic meter of air per day, while the average is 3,000–4,000 pollen grains per cubic meter... Analyzing one sample usually takes an hour or two, depending on the intensity of pollen release. During that remarkable season, we spent the entire day analyzing samples and had to take turns doing so.
In 2020, we started seeing pollen as early as January. This was because pollination in Europe had begun earlier than usual, and we detected these pollen grains in Russia due long-distant transport. Interestingly, this was around the same time the COVID-19 outbreak occurred.
Before the pandemic even began, a group of researchers from various countries including Germany, Sweden, the Netherlands, Switzerland, and the UK published a comprehensive study in the journal Allergy. Scientists tried to understand how atmospheric pollen impacts the population generally, not only people who suffer from allergies. Imagine you're not allergic, but you're in an area with a high pollen count. Will it have any effect on you? It turns out that it will indeed have an effect. Primarily, it leads to a decrease in the production of type I and III interferons. As a result, your antiviral immunity weakens, making you more susceptible to viruses.
And then the coronavirus struck! That got aerobiologists all excited, as it was a unique opportunity to test their hypothesis in real-world conditions. Particularly since, as I mentioned earlier, the 2020 pollination season in Europe started unusually early. For instance, hazel trees began to release pollen as early as December. As COVID-19 spread, scientists started collecting data, attempting to correlate pollen concentration peaks with increases in disease rates. And it all fell into place! It turns out that spikes in pollen concentration are indeed followed by increases in disease rates but with a delay of three to four days.
— How often do you actually get to test your models in real life and see everything align and come together so perfectly?
— The models we're trying to build largely rely on meteorological data and forecasts. So, much like with weather forecasts, we seldom achieve a hundred percent accuracy. However, in aerobiology, we have a fair ability to predict the onset of the pollen season. Based on accumulated positive temperatures, we can estimate with an accuracy of about two or three days when a specific plant will start to release pollen.
— It's just like weather forecasts then.
— Yes, indeed. There is an 80% chance that tomorrow will be the same as today, but if it were always so, nothing would ever change. Therefore, predicting the next day's concentration is challenging. We can probably forecast global trends. For instance, we can anticipate a lot of pollen arriving from somewhere tomorrow based on the movement of air masses. Such models do exist. For example, the Finnish Meteorological Institute's model, SILAM. But even it can sometimes be significantly off. The issue is that it's well adapted to Scandinavia and adjusted to that specific region, but making predictions for Russia's territory is somewhat more difficult. There aren’t many locations where constant monitoring is conducted and where there are long-term observation series and phenological data as well. We do some small-scale modeling for local conditions on our own, but it's nothing compared to SILAM.