Prediction of the birch pollen season characteristics in Cracow, Poland using an 18-year data series

The impact of data assimilation into the meteorological WRF model on birch pollen modelling

We analyse the impact of ground-based data assimilation to the Weather Research and Forecasting (WRF) meteorological model on parameters relevant for birch pollen emission calculations. Then, we use two different emission databases (BASE - no data assimilation, OBSNUD - data assimilation for the meteorological model) in the chemical transport model and evaluate birch pollen concentrations. Finally, we apply a scaling factor for the emissions (BASE and OBSNUD), based on the ratio between simulated and observed seasonal pollen integral (SPIn) to analyse its impact on birch concentrations over Central Europe. Assimilation of observational data significantly reduces model overestimation of air temperature, which is the main parameter responsible for the start of pollen emission and amount of released pollen. The results also show that a relatively small bias in air temperature from the model can lead to significant differences in heating degree days (HDD) value. This may cause the HDD threshold to be attained several days earlier/later than indicated from observational data which has further impact on the start of pollen emission. Even though the bias for air temperature was reduced for OBSNUD, the model indicates a start for the birch pollen season that is too early compared to observations. The start date of the season was improved at two of the 11 stations in Poland. Data assimilation does not have a significant impact on the season's end or SPIn value. The application of the SPIn factor for the emissions results in a much closer birch pollen concentration level to observations even though the factor does not improve the start or end of the pollen season. The post-processing of modelled meteorological fields, such as the application of bias correction, can be considered as a way to further improve the pollen emission modelling.

Aerobiological modeling I: A review of predictive models

The present work is the first of two reviews on applied modeling in the field of aerobiology. The aerobiological predictive models for pollen and fungal spores, usually defined as predictive statistical models, will, amongst other objectives, forecast airborne particles' concentration or dynamical behavior of the particles. These models can be classified into Observation Based Models (OBM), Phenological Based Models (PHM), or OTher Models (OTM). The aim of this review is to show, analyze and discuss the different predictive models used in pollen and spore aerobiological studies. The analysis was performed on published electronic scientific articles from 1998 to 2016 related to the type of model, the taxa and the modelled parameters. From a total of 503 studies, 55.5% used OBM (44.8% on pollen and 10.7% on fungal spores), 38.5% PHM (all on pollen) and 6% OTM (5.4% on pollen and 0.6% on fungal spores). OBM have been used with high frequency to forecast concentration. The most frequent model of OBM was linear regression (18.5% out of 503) on pollen and artificial neural networks (4.6%) on fungal spores. In the PHM, the principal use was to characterize the main pollen season (flowering season) based on the model of growth degree days. Finally, OTM have been used to estimate concentrations at unmonitored areas. Olea (14,5%) on pollen and Alternaria (4,8%) on fungal spores were the taxa most frequently modelled. Daily concentration was the most modelled parameter by OBM (25.2%) and season start day by PHM (35.6%). The PHM approaches include greater model diversity and use fewer independent variables than OBM. In addition, PHM show to be easier to apply than OBM; however, the wide range of criteria to define the parameters to use in PHM (e.g.: pollination start day) makes that each model is used with a lesser frequency than other models.

Changes in qualitative and quantitative traits of birch (Betula pendula) pollen allergenic proteins in relation to the pollution contamination

Birch (Betula pendula) pollen causes inhalant allergy in about 20% of human population in Europe, most of which is sensitive to the main birch allergen, Bet v1. The aim of the study was to find out (i) whether and how the analysed birch individuals differ in regard to composition of individual subunits of pollen proteins and to protein content in these subunits; (ii) whether the level of particulate matter relates to concentration of Bet v1 allergen. Study was performed in Southern Poland, in 2017-2019. Pollen material was collected at 20 sites, of highly or less polluted areas. Protein composition was analysed by SDS-PAGE, while the concentration of Bet v1 was evaluated by ELISA. The obtained results were estimated at the background of the particulate matter (PM10) level and the birch pollen seasons in Kraków. The electrophoregrams of pollen samples collected at different sites showed huge differences in staining intensities of individual protein subunits, also among important birch allergens: Bet v1, Bet v2, Bet v6 and Bet v7. The level of Bet v1 was significantly higher in the pollen samples collected at the more polluted sites. While the birch pollen allergenic potential is determined, the both pollen exposure and the content of the main allergenic components should be considered, as factors causing immunological response and clinical symptoms manifestation in sensitive individuals.

A statistical model for predicting the inter-annual variability of birch pollen abundance in Northern and North-Eastern Europe

The paper suggests a methodology for predicting next-year seasonal pollen index (SPI, a sum of daily-mean pollen concentrations) over large regions and demonstrates its performance for birch in Northern and North-Eastern Europe. A statistical model is constructed using meteorological, geophysical and biological characteristics of the previous year). A cluster analysis of multi-annual data of European Aeroallergen Network (EAN) revealed several large regions in Europe, where the observed SPI exhibits similar patterns of the multi-annual variability. We built the model for the northern cluster of stations, which covers Finland, Sweden, Baltic States, part of Belarus, and, probably, Russia and Norway, where the lack of data did not allow for conclusive analysis. The constructed model was capable of predicting the SPI with correlation coefficient reaching up to 0.9 for some stations, odds ratio is infinitely high for 50% of sites inside the region and the fraction of prediction falling within factor of 2 from observations, stays within 40-70%. In particular, model successfully reproduced both the bi-annual cycle of the SPI and years when this cycle breaks down.

The dynamics of the Corylus, Alnus, and Betula pollen seasons in the context of climate change (SW Poland)

The changes in the main features of early spring tree or shrub pollen seasons are important due to the significant impact on the occurrence of pollen-related allergy symptoms. This study shows the results of pollen monitoring for a period of eleven years (2003-2013) using a Burkard volumetric spore trap. The main characteristics of the hazel, alder, and birch pollination season were studied in Wrocław (SW Poland). The statistical analyses do not show a significant trend of annual total pollen count or shift in timing of the pollen season in the period of analysis. The research confirms a great impact (at the statistically significant level of 0.05) of the heat resources on pollination season (the value of the correlation coefficient ranges from -0.63 up to -0.87). Meteorological variables (e.g. sum of temperature for selected period) were compiled to 5-year running means to examine trends. Changes in the pollination period features due to climate change including both timing and intensity of pollen productivity, would have important consequences for allergy sufferers.

Forecasting model of Corylus, Alnus, and Betula pollen concentration levels using spatiotemporal correlation properties of pollen count

The aim of the study was to create and evaluate models for predicting high levels of daily pollen concentration of Corylus, Alnus, and Betula using a spatiotemporal correlation of pollen count. For each taxon, a high pollen count level was established according to the first allergy symptoms during exposure. The dataset was divided into a training set and a test set, using a stratified random split. For each taxon and city, the model was built using a random forest method. Corylus models performed poorly. However, the study revealed the possibility of predicting with substantial accuracy the occurrence of days with high pollen concentrations of Alnus and Betula using past pollen count data from monitoring sites. These results can be used for building (1) simpler models, which require data only from aerobiological monitoring sites, and (2) combined meteorological and aerobiological models for predicting high levels of pollen concentration.

Predicting daily ragweed pollen concentrations using Computational Intelligence techniques over two heavily polluted areas in Europe

Forecasting ragweed pollen concentration is a useful tool for sensitive people in order to prepare in time for high pollen episodes. The aim of the study is to use methods of Computational Intelligence (CI) (Multi-Layer Perceptron, M5P, REPTree, DecisionStump and MLPRegressor) for predicting daily values of Ambrosia pollen concentrations and alarm levels for 1-7 days ahead for Szeged (Hungary) and Lyon (France), respectively. Ten-year daily mean ragweed pollen data (within 1997-2006) are considered for both cities. 10 input variables are used in the models including pollen level or alarm level on the given day, furthermore the serial number of the given day of the year within the pollen season and altogether 8 meteorological variables. The study has novelties as (1) daily alarm thresholds are firstly predicted in the aerobiological literature; (2) data-driven modelling methods including neural networks have never been used in forecasting daily Ambrosia pollen concentration; (3) algorithm J48 has never been used in palynological forecasts; (4) we apply a rarely used technique, namely factor analysis with special transformation, to detect the importance of the influencing variables in defining the pollen levels for 1-7 days ahead. When predicting pollen concentrations, for Szeged Multi-Layer Perceptron models deliver similar results with tree-based models 1 and 2 days ahead; while for Lyon only Multi-Layer Perceptron provides acceptable result. When predicting alarm levels, the performance of Multi-Layer Perceptron is the best for both cities. It is presented that the selection of the optimal method depends on climate, as a function of geographical location and relief. The results show that the more complex CI methods perform well, and their performance is case-specific for ≥2 days forecasting horizon. A determination coefficient of 0.98 (Ambrosia, Szeged, one day and two days ahead) using Multi-Layer Perceptron ranks this model the best one in the literature.

Predicting tree pollen season start dates using thermal conditions

Thermal conditions at the beginning of the year determine the timing of pollen seasons of early flowering trees. The aims of this study were to quantify the relationship between the tree pollen season start dates and the thermal conditions just before the beginning of the season and to construct models predicting the start of the pollen season in a given year. The study was performed in Krakow (Southern Poland); the pollen data of Alnus, Corylus and Betula were obtained in 1991-2012 using a volumetric method. The relationship between the tree pollen season start, calculated by the cumulated pollen grain sum method, and a 5-day running means of maximum (for Alnus and Corylus) and mean (for Betula) daily temperature was found and used in the logistic regression models. The estimation of model parameters indicated their statistically significance for all studied taxa; the odds ratio was higher in models for Betula, comparing to Alnus and Corylus. The proposed model makes the accuracy of prediction in 83.58 % of cases for Alnus, in 84.29 % of cases for Corylus and in 90.41 % of cases for Betula. In years of model verification (2011 and 2012), the season start of Alnus and Corylus was predicted more precisely in 2011, while in case of Betula, the model predictions achieved 100 % of accuracy in both years. The correctness of prediction indicated that the data used for the model arrangement fitted the models well and stressed the high efficacy of model prediction estimated using the pollen data in 1991-2010.