A temporally and spatially explicit, data-driven estimation of airborne ragweed pollen concentrations across Europe.

Ongoing and future climate change driven expansion of aeroallergen-producing plant species comprise a major human health problem across Europe and elsewhere. There is an urgent need to produce accurate, temporally dynamic maps at the continental level, especially in the context of climate uncertainty. This study aimed to restore missing daily ragweed pollen data sets for Europe, to produce phenological maps of ragweed pollen, resulting in the most complete and detailed high-resolution ragweed pollen concentration maps to date. To achieve this, we have developed two statistical procedures, a Gaussian method (GM) and deep learning (DL) for restoring missing daily ragweed pollen data sets, based on the plant's reproductive and growth (phenological, pollen production and frost-related) characteristics. DL model performances were consistently better for estimating seasonal pollen integrals than those of the GM approach. These are the first published modelled maps using altitude correction and flowering phenology to recover missing pollen information. We created a web page (http://euragweedpollen.gmf.u-szeged.hu/), including daily ragweed pollen concentration data sets of the stations examined and their restored daily data, allowing one to upload newly measured or recovered daily data. Generation of these maps provides a means to track pollen impacts in the context of climatic shifts, identify geographical regions with high pollen exposure, determine areas of future vulnerability, apply spatially-explicit mitigation measures and prioritize management interventions.

Explainable AI for unveiling deep learning pollen classification model based on fusion of scattered light patterns and fluorescence spectroscopy.

Pollen monitoring have become data-intensive in recent years as real-time detectors are deployed to classify airborne pollen grains. Machine learning models with a focus on deep learning, have an essential role in the pollen classification task. Within this study we developed an explainable framework to unveil a deep learning model for pollen classification. Model works on data coming from single particle detector (Rapid-E) that records for each particle optical fingerprint with scattered light and laser induced fluorescence. Morphological properties of a particle are sensed with the light scattering process, while chemical properties are encoded with fluorescence spectrum and fluorescence lifetime induced by high-resolution laser. By utilizing these three data modalities, scattering, spectrum, and lifetime, deep learning-based models with millions of parameters are learned to distinguish different pollen classes, but a proper understanding of such a black-box model decisions demands additional methods to employ. Our study provides the first results of applied explainable artificial intelligence (xAI) methodology on the pollen classification model. Extracted knowledge on the important features that attribute to the predicting particular pollen classes is further examined from the perspective of domain knowledge and compared to available reference data on pollen sizes, shape, and laboratory spectrofluorometer measurements.

Towards European automatic bioaerosol monitoring: Comparison of 9 automatic pollen observational instruments with classic Hirst-type traps.

To benefit allergy patients and the medical practitioners, pollen information should be available in both a reliable and timely manner; the latter is only recently possible due to automatic monitoring. To evaluate the performance of all currently available automatic instruments, an international intercomparison campaign was jointly organised by the EUMETNET AutoPollen Programme and the ADOPT COST Action in Munich, Germany (March-July 2021). The automatic systems (hardware plus identification algorithms) were compared with manual Hirst-type traps. Measurements were aggregated into 3-hourly or daily values to allow comparison across all devices. We report results for total pollen as well as for Betula, Fraxinus, Poaceae, and Quercus, for all instruments that provided these data. The results for daily averages compared better with Hirst observations than the 3-hourly values. For total pollen, there was a considerable spread among systems, with some reaching R2 > 0.6 (3 h) and R2 > 0.75 (daily) compared with Hirst-type traps, whilst other systems were not suitable to sample total pollen efficiently (R2 < 0.3). For individual pollen types, results similar to the Hirst were frequently shown by a small group of systems. For Betula, almost all systems performed well (R2 > 0.75 for 9 systems for 3-hourly data). Results for Fraxinus and Quercus were not as good for most systems, while for Poaceae (with some exceptions), the performance was weakest. For all pollen types and for most measurement systems, false positive classifications were observed outside of the main pollen season. Different algorithms applied to the same device also showed different results, highlighting the importance of this aspect of the measurement system. Overall, given the 30 % error on daily concentrations that is currently accepted for Hirst-type traps, several automatic systems are currently capable of being used operationally to provide real-time observations at high temporal resolutions. They provide distinct advantages compared to the manual Hirst-type measurements.

Integration of reference data from different Rapid-E devices supports automatic pollen detection in more locations.

Pollen is the most common cause of seasonal allergies, affecting over 33 % of the European population, even when considering only grasses. Informing the population and clinicians in real-time about the actual presence of pollen in the atmosphere is essential to reduce its harmful health and economic impact. Thus, there is a growing network of automatic particle analysers, and the reproducibility and transferability of implemented models are recommended since a reference dataset for local pollen of interest needs to be collected for each device to classify pollen, which is complex and time-consuming. Therefore, it would be beneficial to incorporate the reference dataset collected from other devices in different locations. However, it must be considered that laser-induced data are prone to device-specific noise due to laser and detector sensibility. This study collected data from two Rapid-E bioaerosol identifiers in Serbia and Italy and implemented a multi-modal convolutional neural network for pollen classification. We showed that models lost their performance when trained on data from one and tested on another device, not only in terms of the recognition ability but also in comparison with the manual measurements from Hirst-type traps. To enable pollen classification with just one model in both study locations, we first included the missing pollen classes in the dataset from the other study location, but it showed poor results, implying that data of one pollen class from different devices are more different than data of different pollen classes from one device. Combining all available reference data in a single model enabled the classification of a higher number of pollen classes in both study locations. Finally, we implemented a domain adaptation method, which improved the recognition ability and the correlations of transferred models only for several pollen classes.

Why should we care about high temporal resolution monitoring of bioaerosols in ambient air?

This is the first time that atmospheric concentrations of individual pollen types have been recorded by an automatic sampler with 1-hour and sub-hourly resolution (i.e. 1-minute and 1-second data). The data were collected by traditional Hirst type methods and state-of the art Rapid-E real-time bioaerosol detector. Airborne pollen data from 7 taxa, i.e. Acer negundo, Ambrosia, Broussonetia papyrifera, Cupressales (Taxaceae and Cupressaceae families), Platanus, Salix and Ulmus, were collected during the 2019 pollen season in Novi Sad, Serbia. Pollen data with daily, hourly and sub-hourly temporal resolution were analysed in terms of their temporal variability. The impact of turbulence kinetic energy (TKE) on pollen cloud homogeneity was investigated. Variations in Seasonal Pollen Integrals produced by Hirst and Rapid-E show that scaling factors are required to make data comparable. Daily average and hourly measurements recorded by the Rapid-E and Hirst were highly correlated and so examining Rapid-E measurements with sub-hourly resolution is assumed meaningful from the perspective of identification accuracy. Sub-hourly data provided an insight into the heterogenous nature of pollen in the air, with distinct peaks lasting ~5-10 min, and mostly single pollen grains recorded per second. Short term variations in 1-minute pollen concentrations could not be wholly explained by TKE. The new generation of automatic devices has the potential to increase our understanding of the distribution of bioaerosols in the air, provide insights into biological processes such as pollen release and dispersal mechanisms, and have the potential for us to conduct investigations into dose-response relationships and personal exposure to aeroallergens.

The influence of source maps on SILAM performance in modeling ragweed pollen concentrations in the area of a major European source.

The Pannonian Plain is one of the centers of ragweed distribution in Europe. The province of Vojvodina (Serbia) is located on the southern part of the Pannonian Plain, representing a highly infested region. In this study, we have used the SILAM atmospheric dispersion model to simulate ragweed pollen concentrations during the season 2016 in the Vojvodina region. SILAM was tested with three different source maps of ragweed distribution in Vojvodina only: (1) map used in operational SILAM, which was calibrated with the SILAM model and observations, (2) map derived using "top-down" approach with land cover data inventory, and (3) map obtained with "top-down" approach using crop classification from the satellite data. Additionally, the sensitivity studies were done using two modified maps to study the effect of the source strength and long-range transport. Results of simulations were validated with the bi-hourly, daily, and seasonal pollen concentrations measured at five stations in Vojvodina. Overall Pearson correlation coefficients were 0.51 (Map 1), 0.50 (Map 2), and 0.42 (Map 3), while debiased scores were 232.95 pollen m-3 (Map 1), 245.59 pollen m-3 (Map 2), and 258.24 pollen m-3 (Map 3). Even though Vojvodina is in the area of a major European source, regional transport of ragweed pollen from a few hundred kilometers of the surrounding area was important in explaining the presence of pollen in the afternoon hours, although it could not completely explain total pollen quantity. The results confirmed that it is vital to calibrate source maps using atmospheric dispersion model with the observed pollen data.

Phenolic Composition Influences the Health-Promoting Potential of Bee-Pollen.

Information on compositional, nutritional and functional properties of bee-pollen, as a health-promoting food, is essential for defining its quality. Concerning the nutritional importance of phenolic compounds, the aim of this study was to determine the phenolic profile and antioxidant activity of twenty-four bee-pollen samples collected from different regions of Serbia. High-performance thin-layer chromatographic (HPTLC) fingerprinting was used for profiling of bee-pollen samples according to the botanical type. HPTLC hyphenated with image analysis and a pattern recognition technique confirmed the grouping of samples caused by the specific phenolic composition of pollens of different botanical origin. Flavonoid glycosides in bee-pollen samples were identified by applying ultra-high-performance liquid chromatography (UHPLC) coupled with linear ion trap-Orbitrap mass spectrometry (LTQ Orbitrap MS). Eight out of twenty-seven flavonol glycosides were identified in bee-pollen samples for the first time. All analyzed bee-pollen samples showed a high number of phenolic compounds which may have therapeutic potential.

Concomitant occurrence of anthropogenic air pollutants, mineral dust and fungal spores during long-distance transport of ragweed pollen.

Large-scale synoptic conditions are able to transport considerable amounts of airborne particles over entire continents by creating substantial air mass movement. This phenomenon is observed in Europe in relation to highly allergenic ragweed (Ambrosia L.) pollen grains that are transported from populations in Central Europe (mainly the Pannonian Plain and Balkans) to the North. The path taken by atmospheric ragweed pollen often passes through the highly industrialised mining region of Silesia in Southern Poland, considered to be one of the most polluted areas in the EU. It is hypothesized that chemical air pollutants released over Silesia could become mixed with biological material and be transported to less polluted regions further North. We analysed levels of air pollution during episodes of long-distance transport (LDT) of ragweed pollen to Poland. Results show that, concomitantly with pollen, the concentration of air pollutants with potential health-risk, i.e. SO2, and PM10, have also significantly increased (by 104% and 37%, respectively) in the receptor area (Western Poland). Chemical transport modelling (EMEP) and air mass back-trajectory analysis (HYSPLIT) showed that potential sources of PM10 include Silesia, as well as mineral dust from the Ukrainian steppe and the Sahara Desert. In addition, atmospheric concentrations of other allergenic biological particles, i.e. Alternaria Nees ex Fr. spores, also increased markedly (by 115%) during LDT episodes. We suggest that the LDT episodes of ragweed pollen over Europe are not a "one-component" phenomenon, but are often related to elevated levels of chemical air pollutants and other biotic and abiotic components (fungal spores and desert dust).

Near-ground effect of height on pollen exposure.

The effect of height on pollen concentration is not well documented and little is known about the near-ground vertical profile of airborne pollen. This is important as most measuring stations are on roofs, but patient exposure is at ground level. Our study used a big data approach to estimate the near-ground vertical profile of pollen concentrations based on a global study of paired stations located at different heights. We analyzed paired sampling stations located at different heights between 1.5 and 50 m above ground level (AGL). This provided pollen data from 59 Hirst-type volumetric traps from 25 different areas, mainly in Europe, but also covering North America and Australia, resulting in about 2,000,000 daily pollen concentrations analyzed. The daily ratio of the amounts of pollen from different heights per location was used, and the values of the lower station were divided by the higher station. The lower station of paired traps recorded more pollen than the higher trap. However, while the effect of height on pollen concentration was clear, it was also limited (average ratio 1.3, range 0.7-2.2). The standard deviation of the pollen ratio was highly variable when the lower station was located close to the ground level (below 10 m AGL). We show that pollen concentrations measured at >10 m are representative for background near-ground levels.

Ambrosia pollen source inventory for Italy: a multi-purpose tool to assess the impact of the ragweed leaf beetle (Ophraella communa LeSage) on populations of its host plant.

Here, we produce Ambrosia pollen source inventories for Italy that focuses on the periods before and after the accidental introduction of the Ophraella communa beetle. The inventory uses the top-down approach that combines the annual Ambrosia pollen index from a number of monitoring stations in the source region as well as Ambrosia ecology, local knowledge of Ambrosia infestation and detailed land cover information. The final inventory is gridded to a 5 × 5-km resolution using a stereographic projection. The sites with the highest European Infection levels were recorded in the north of Italy at Busto Arsizio (VA3) (European Infection level 2003-2014 = 52.1) and Magenta (MI7) (European Infection level 2003-2014 = 51.3), whereas the sites with the lowest (i.e. around 0.0) were generally located to the south of the country. Analysis showed that the European Infection level in all of Italy was significantly lower in 2013-2014 compared to 2003-2012, and this decrease was even more pronounced at the sites in the area where Ophraella communa was distributed. Cross-validations show that the sensitivity to the inclusion of stations is typically below 1% (for two thirds of the stations) and that the station Magenta (MI7) had the largest impact compared to all other stations. This is the first time that pollen source inventories from different temporal periods have been compared in this way and has implications for simulating interannual variations in pollen emission as well as evaluating the management of anemophilous plants like Ambrosia artemisiifolia.

Aerobiology in the International Journal of Biometeorology, 1957-2017.

Aerobiology and biometeorology are related fields. Here we provide a broad review of aerobiology articles published in the International Journal of Biometeorology (IJB) over the past 60 years. We consider how the quantity of such work has varied over this period as well as which regions and countries have been the focus of such work, and where there is a relative paucity. We then focus on a number of highlights and themes in this research, including aerobiology and climate change and aerobiological modelling and forecasting. While much of the article focusses on airborne pollen research, we also discuss the extent to which other airborne organic particles such as fungal spores and bacteria have been the focus of research published in IJB. Also considered are knowledge gaps and research needs and priorities with respect to the field of aerobiology. While the IJB has been one of the main platforms for presenting aerobiological research over recent decades, the article highlights the need for the field of aerobiology to embrace new sampling technologies such as spectral analysis and next-generation sequencing to identify and quantify airborne biological particles.

The long distance transport of airborne Ambrosia pollen to the UK and the Netherlands from Central and south Europe.

The invasive alien species Ambrosia artemisiifolia (common or short ragweed) is increasing its range in Europe. In the UK and the Netherlands, airborne concentrations of Ambrosia pollen are usually low. However, more than 30 Ambrosia pollen grains per cubic metre of air (above the level capable to trigger allergic symptoms) were recorded in Leicester (UK) and Leiden (NL) on 4 and 5 September 2014. The aims of this study were to determine whether the highly allergenic Ambrosia pollen recorded during the episode could be the result of long distance transport, to identify the potential sources of these pollen grains and to describe the conditions that facilitated this possible long distance transport. Airborne Ambrosia pollen data were collected at 10 sites in Europe. Back trajectory and atmospheric dispersion calculations were performed using HYSPLIT_4. Back trajectories calculated at Leicester and Leiden show that higher altitude air masses (1500 m) originated from source areas on the Pannonian Plain and Ukraine. During the episode, air masses veered to the west and passed over the Rhône Valley. Dispersion calculations showed that the atmospheric conditions were suitable for Ambrosia pollen released from the Pannonian Plain and the Rhône Valley to reach the higher levels and enter the airstream moving to northwest Europe where they were deposited at ground level and recorded by monitoring sites. The study indicates that the Ambrosia pollen grains recorded during the episode in Leicester and Leiden were probably not produced by local sources but transported long distances from potential source regions in east Europe, i.e. the Pannonian Plain and Ukraine, as well as the Rhône Valley in France.

Risk of exposure to airborne Ambrosia pollen from local and distant sources in Europe - an example from Denmark.

BACKGROUND: Ambrosia artemisiifolia L. is a noxious invasive alien species in Europe. It is an important aeroallergen and millions of people are exposed to its pollen. OBJECTIVE: The main aim of this study is to show that atmospheric concentrations of Ambrosia pollen recorded in Denmark can be derived from local or more distant sources. METHODS: This was achieved by using a combination of pollen measurements, air mass trajectory calculations using the HYPLIT model and mapping all known Ambrosia locations in Denmark and relating them to land cover types. RESULTS: The annual pollen index recorded in Copenhagen during a 15-year period varied from a few pollen grains to more than 100. Since 2005, small quantities of Ambrosia pollen has been observed in the air every year. We have demonstrated, through a combination of Lagrangian back-trajectory calculations and atmospheric pollen measurements, that pollen arrived in Denmark via long-distance transport from centres of Ambrosia infection, such as the Pannonian Plain and Ukraine. Combining observations with results from a local scale dispersion model show that it is possible that Ambrosia pollen could be derived from local sources identified within Denmark. CONCLUSIONS: The high allergenic capacity of Ambrosia pollen means that only small amounts of pollen are relevant for allergy sufferers, and just a few plants will be sufficient to produce enough pollen to affect pollen allergy sufferers within a short distance from the source. It is necessary to adopt control measures to restrict Ambrosia numbers. Recommendations for the removal of all Ambrosia plants can effectively reduce the amount of local pollen, as long as the population of Ambrosia plants is small.

The effect of changes to the method of estimating the pollen count from aerobiological samples.

Pollen data have been recorded at Novi Sad in Serbia since 2000. The adopted method of producing pollen counts has been the use of five longitudinal transects that examine 19.64% of total sample surface. However, counting five transects is time consuming and so the main objective of this study is to investigate whether reducing the number to three or even two transects would have a significant effect on daily average and bi-hourly pollen concentrations, as well as the main characteristics of the pollen season and long-term trends. This study has shown that there is a loss of accuracy in daily average and bi-hourly pollen concentrations (an increase in % ERROR) as the sub-sampling area is reduced from five to three or two longitudinal transects. However, this loss of accuracy does not impact on the main characteristics of the season or long-term trends. As a result, this study can be used to justify changing the sub-sampling method used at Novi Sad from five to three longitudinal transects. The use of two longitudinal transects has been ruled out because, although quicker, the counts produced: (a) had the greatest amount of % ERROR, (b) altered the amount of influence of the independent variable on the dependent variable (the slope in regression analysis) and (c) the total sampled surface (7.86%) was less than the minimum requirement recommended by the European Aerobiology Society working group on Quality Control (at least 10% of total slide area).

The occurrence of Ambrosia pollen in Rzeszów, Kraków and Poznan, Poland: investigation of trends and possible transport of Ambrosia pollen from Ukraine.

Previous studies have shown that ragweed pollen arrives in Poland from sources in the south, in Slovakia, the Czech Republic, Hungary and Austria. It is likely that ragweed pollen also arrives from sources in the southeast (e.g. Ukraine). This hypothesis was investigated using 13 years of pollen data and back-trajectory analysis. Ambrosia pollen data were collected at three sites in Poland, Rzeszów, Kraków and Poznan. The amount of ragweed pollen recorded at Rzeszów was significantly higher than in Poznan and Kraków. This can be related to either a higher abundance of local populations of Ambrosia in south-east Poland or the proximity of Rzeszów to foreign sources of ragweed pollen. The combined results of pollen measurements and air mass trajectory calculations identified plumes of Ambrosia pollen that were recorded at Rzeszów, Kraków and Poznan on 4 and 5 September 1999 and 3 September 2002. These plumes arrived at the pollen-monitoring sites from an easterly direction, indicating sources of Ambrosia pollen in eastern Poland or Ukraine. This identifies Ukraine as a possible new source of ragweed pollen for Poland and therefore an important source area of Ambrosia pollen on the European Continent.

Airborne grass and ragweed pollen in the southern Panonnian Valley--consideration of rural and urban environment.

The aims of this study were to describe and compare the characteristics of grass and ragweed airborne pollen in rural and urban areas in the southern Panonnian Valley. Airborne pollen data were collected by using Hirst type volumetric samplers simultaneously in rural and urban localities. If rural and urban environment are considered, both grass and ragweed daily pollen concentrations showed a significant degree of association. Observed parameters (pollen index, maximum daily concentration, number of days during which the pollen is recorded in the air and start day of main pollen season), showed year-to-year variations for both grass and ragweed aeropollen. Average values of these parameters were higher in the rural environment, but the difference was statistically significant only for grass pollen index. Such a low difference indicates the possibility for conducting dose response clinical trials based on data obtained from one sampling station. The least year-to-year variations as well as the least difference between rural and urban environment, have been observed in the case of start date of the MPS. Such a situation suggests the possibility for using data obtained in one type of environment for the development of long-term forecast models for an entire region.