Science of the Total Environment 896 (2023) 165337 Contents lists available at ScienceDirect Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv Influence of synoptic meteorology on airborne allergenic pollen and spores in an urban environment in Northeastern Iberian Peninsula ⁎ Marta Alarcón a, , Raül Rodríguez-Solà b, M. Carmen Casas-Castillo c, Francisco Molero d, Pedro Salvador d, Cristina Periago a, Jordina Belmonte e,f a Departament de Física, EEBE, Universitat Politècnica de Catalunya·BarcelonaTech, Eduard Maristany 16, 08019 Barcelona, Spain b Departament de Física, ETSEIB, Universitat Politècnica de Catalunya·BarcelonaTech, Diagonal 647, 08028 Barcelona, Spain c Departament de Física, ESEIAAT, Universitat Politècnica de Catalunya·BarcelonaTech, Colom 1, 08222 Terrassa, Spain d Departamento de Medio Ambiente, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (CIEMAT), Avenida Complutense 22, 28040 Madrid, Spain e Departament de Biologia Animal, Biologia Vegetal i Ecologia, Facultat de Biociències, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain f Institut de Ciència i Tecnologia Ambientals (ICTA-UAB), Universitat Autònoma de Bellaterra, 08193 Bellaterra, Spain H I G H L I G H T S G R A P H I C A L A B S T R A C T • Six synoptic situations explain most weather variability in Iberian Peninsula. • Airborne pollen/spore dynamics in Barce- lona is influenced by synoptic meteorol- ogy. • There is one synoptic pattern where Platanus shows always high allergy risk values. • The start of pollination of Urticaceae was advanced under certain weather patterns. • Alternaria has a greater presence under the most frequent summer scenario. A R T I C L E I N F O A B S T R A C T Editor: Anastasia Paschalidou The influence of the most frequent patterns of synoptic circulation on the dynamics of airborne pollen/spores recorded at the Barcelona Aerobiological Station (BCN) was analysed. Six pollen types (Platanus, Cupressaceae, Olea, Poaceae, Keywords: Urticaceae and Amaranthaceae), and one fungal spore (Alternaria) were selected for their high allergenic effect in sen- Meteorological patterns sitive people. Six synoptic meteorological patterns were identified through cluster analysis of sea level pressure fields Synoptic situation as the main responsible of the weather conditions in the Iberian Peninsula. The local meteorological conditions in Bar- Environmental health Air pollution celona associated with each one of the synoptic types were also stablished. Different statistical methods were applied Aerobiology to analyse possible relationships between concentrations and timing of the recorded aerobiological particles and spe- Western Mediterranean cific synoptic types. The study, focused in the 19-year period 2001–2019, shows that one of the scenarios, frequent in winter and linked to high stability and air-mass blockage, registered the highest mean and median values for Platanus and Cupressaceae, but it was not very relevant for the other taxa. It was also this scenario that turned out to be the most influent on the pollination timing showing a significant influence on the start occurrence of Urticaceae flowering and on the peak date of Platanus. On the other hand, the most frequent synoptic type in the period, relevant in spring and summer, was linked to sporadic episodes of levels considered to be of high risk of allergy to Platanus, Poaceae, and Urticaceae pollen, and Alternaria fungal spore. This synoptic pattern, characterized by the presence of the Azores an- ticyclone and the Atlantic low located in the north of the United Kingdom, was associated with high temperatures, ⁎ Corresponding author. E-mail addresses: marta.alarcon@upc.edu (M. Alarcón), raul.rodriguez@upc.edu (R. Rodríguez-Solà), m.carmen.casas@upc.edu (M.C. Casas-Castillo), f.molero@ciemat.es (F. Molero), pedro.salvador@ciemat.es (P. Salvador), cristina.periago@upc.edu (C. Periago), jordina.belmonte@uab.cat (J. Belmonte). http://dx.doi.org/10.1016/j.scitotenv.2023.165337 Received 10 March 2023; Received in revised form 3 July 2023; Accepted 3 July 2023 Available online 4 July 2023 0048-9697/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 low relative humidity and moderate winds from the NW in Barcelona. The identification of an interaction between syn- optic meteorology and pollen/spore dynamics will allow better abatement measures, reducing adverse health effects on sensitive population. 1. Introduction transport in the Western Mediterranean is highly influenced by the complex meteorology that characterize the region. Different factors such as the orog- Biological aerosols have been the subject of increasing interest because raphy close to the entire coastline, the westerly wind regime, the subtropi- of their impact on air quality and population health (Sauliene et al., 2015; cal high-pressure regime, mesoscale phenomena and Mediterranean Aerts et al., 2021). This impact is especially relevant in urban environ- advections contribute to this complexity (Lana et al., 2021). Several studies ments, since a relationship has been observed between air pollution, air- have shown the occurrence of transport patterns that favour the arrival of borne pollen, and allergenic effects. For instance, epidemiological studies pollen in Catalonia, region located on the northeast of the Iberian Peninsula in Canada showed that the allergenic effect of ambient aeroallergens was (IP), from specific source areas in Central Europe (Belmonte et al., 2000; significantly higher after exposure to air pollutants, increasing the effect Belmonte et al., 2008; Fernández-Llamazares et al., 2012), Pyrenees of pollen on symptoms (Cakmak et al., 2012). A relationship was found be- (Izquierdo et al., 2017) and northwest of the IP (Alarcón et al., 2022). tween asthma hospitalizations and exposure to tree pollen and fine partic- The knowledge of the synoptic atmospheric circulations leading these ulate matter (PM2.5, median aerodynamic diameter less than or equal to transport patterns can help to interpret pollen dynamics. 2.5 μm), and exposure to grass pollen and PM10 particulate matter (median Other studies carried out in the Mediterranean region have shown that aerodynamic diameter less than or equal to 10 μm). Furthermore, it was synoptic meteorological patterns (SMP) play a significant role in the forma- found in France and USA that exposure to ozone increases the frequency tion and regional transport of several pollutants like O3, NOx and particu- of respiratory allergies (Riediker et al., 2001). On the other hand, different late matter (Russo et al., 2014) and in the development of high air studies analysed the impact on health of exposure to pollen based on differ- pollution episodes and new particle formation processes (Salvador et al., ent health indicators. For example, in the city of Paris, they found that up to 2021). Furthermore, the relationship between synoptic atmospheric circu- the threshold of 30 Pollen m−3 of grasses, the number of hospitalizations lations and the transport of African dust towards south European countries for asthma in the total population increases significantly, then the effects re- has been demonstrated (Russo et al., 2020; Salvador et al., 2014). Transport main stable regardless of pollen concentration levels above this threshold routes have often been linked to cyclonic activity over the Mediterranean (Caillaud et al., 2014). The same study found effects on allergic rhinitis and to the presence of high-pressure systems over North Africa (Àvila and asthma that appeared for low values of grasses (< 10 Pollen m−3) et al., 2007; Avila and Alarcón, 1999; Escudero et al., 2005). Recent studies due to a very gradual release of these pollen grains before the pollination have characterized the transport of pollen/spores towards the European season; concluding that the entire pollen season of grasses could be consid- continent during African dust outbreaks (Grewling et al., 2019; López- ered risky (Caillaud et al., 2014). Orozco et al., 2020). A study at the European continental-scale showed evidence of a trend Multivariate techniques such as cluster analysis and principal compo- towards an increase in atmospheric pollen in Europe, including highly aller- nent analysis have been widely used in studies related to air pollution. genic taxa such as Alnus, Ambrosia, Artemisia, Betula, Corylus, Cupressaceae, For instance, cluster analysis supported the classification of rains for prov- Olea, Poaceae (Ziello et al., 2012). These trends could not be attributed to enance to identify the source areas of pollutants in precipitation in a Cata- rising temperatures, but experimental studies suggested that they might lan rural site (NE Spain) (Avila and Alarcón, 1999). Another study be influenced by anthropogenic increases in the greenhouse gas CO2 obtained a classification of the synoptic patterns associated with the epi- (Rogers et al., 2006). However, another continental-scale study in North sodes of extreme rain recorded at the Ebro Observatory in the period America found that the impact of CO2 concentrations was important, but 1905–2003 (Pérez-Zanón et al., 2018). Principal Component Analysis of temperature had a much stronger effect on pollen variability (Anderegg the pressure and geopotential fields revealed the synoptic configurations et al., 2021). Both factors, temperature and CO2 concentration, are syner- that caused the entry of extremely cold and hot air masses into the IP gistically connected and further global increase in both is projected, leading (Mohammed et al., 2018). In the recent years, similar techniques based to further increases in pollen concentrations and greater exposure of the on clusters of circulation types have been applied to interpret pollen levels population to pollen allergens. More research is needed in this area given in a mountain valley in Central Italy (Pace et al., 2022). Ojrzynska et al. the negative consequences for public health. (2020) have studied the influence of synoptic and local meteorology on The emission of pollen depends on characteristics of the environment the Betula and Alnus pollen concentrations in Wroclaw, Poland. (soil composition, moisture content, vegetation cover), and on local atmo- Paschalidou et al. (2020) have applied a method of synoptic circulation- spheric conditions such as temperature, humidity and wind speed (Rojo to-environment classification to explain the relationship between different et al., 2015). Many studies have concluded that temperature is the main weather patterns and pollen concentrations in Thessaloniki (Greece). meteorological variable that influences timing and pollen levels In this work we focused on six pollen types and one fungal spore abun- (Veriankaitė et al., 2012; Majeed et al., 2018). Recent studies found little dant in the study area and with impact on the well-being and health of sen- but significant influence of moderately extreme rainfall on later pollination sitized people due to their allergenic capacity (Ojeda et al., 2018). (Kirchner et al., 2022; Rodríguez-Solà et al., 2022). These studies showed Consequently, its emission and subsequent presence in the air is highly con- that an increase in precipitation in the previous winter appeared to be pos- ditioned by the local weather. However, transport on a regional scale and itively correlated with increased pollen counts for herbaceous taxa in subse- also over long distances is not negligible. Our starting hypothesis is that quent seasons, whereas for the arboreal taxa the proportion of positive the levels we measure in Barcelona (BCN) are influenced by both synoptic correlations was lower. and local meteorology. The question we want to answer is whether there is After pollen is released into the atmosphere, its fate is determined by a direct link between certain atmospheric circulation patterns and the levels mesoscale and synoptic scale mechanisms that govern the regional and and timing of airborne pollen/spore in the Mediterranean city of Barcelona. long-range atmospheric transport (Adams-Groom et al., 2017). An impor- The study comprises 19 years of daily airborne pollen/spore data at the tant fraction of the released pollen settles near the emission sources, but an- BCN aerobiological station. A robust classification of the main synoptic other part can be transported by air masses and cause allergies to sensitive weather patterns that occurred during that period (2001–2019) over the people in more or less distant regions (Alarcón et al., 2022; Belmonte et al., IP has also been used. The classification process of sea level pressure daily 2008; Izquierdo et al., 2011, 2017; Sofiev et al., 2013). Atmospheric data fields used to generate these patterns, as well as the validation of 2 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 this process, has been previously published (Salvador et al., 2021). The spe- period 2001–2019, following the European aerobiological standards cific objectives here were: (1) Establish the local meteorological character- (Galán et al., 2014). istics linked to each synoptic meteorological pattern (SMP) and the type of Airborne daily samples were obtained from Hirst sampler (Hirst, 1952) weather they give in Barcelona. (2) Determine the influence of SMPs on the and analysed using the standardized Spanish method (Galán Soldevilla concentration and timing (start, end and duration of the pollination and et al., 2007). Six pollen taxa and one fungal spore of high interest due to peak date) of the taxa under study. their abundance in the atmospheric pollen and spore spectrum and aller- genic significance were considered in this study. Some selected pollen 2. Methodology types were from trees: Platanus (PLAT), Cupressaceae (CUPR), Olea (OLEA); other from herbaceous plants: Poaceae (POAC), Urticaceae 2.1. Area of study (URTI), Amaranthaceae (AMAR); and the fungal spore was Alternaria (ALTE). Barcelona is a Mediterranean city, located in the region of Catalonia, The results obtained in the analysis at the light microscope of the sam- northeast of IP (Fig. 1). The vegetation in the area is typically Mediterra- ples from the Hirst trap (pollen/spore counts) are converted into daily con- nean, with evergreen oaks (Quercus ilex) forests, mixed with deciduous centrations thanks to the conversion factor resulting from the suction oaks (Q. pubescens, Q. cerrioides) in the wetter areas and abundantly power of the trap, the proportion of sample analysed, and the optical values mixed with Pinus halepensis and Pinus pinea. Maquis are also present and of the microscope used. Thus, for each studied pollen/spore type one con- some cultivations persist. Two important rivers at the north and the south centration value per day of the period 2001–2019 is considered. These of the city add the typical plants of genus Salix and Populus. In the city, a daily concentrations are the basis for establishing, by summing up all the long list of ornamental and ruderal species adds their pollen to the airborne values of the year, the Annual Pollen Integral (APIn) per taxon and year. spectra and between them we find the ones selected for this study: In this paper the Main Pollen/Spore Season (MPS) of each pollen/spore Cupressaceae, Olea and Platanus as ornamental trees, Poaceae making type and year has been defined (Andersen, 1991; Gehrig and Clot, 2021) part of the grasses in the city and the surroundings, and Amaranthaceae as the period between the date when the sum of daily mean pollen concen- and Urticaceae as ruderal plants. See detailed information in: https:// trations reaches 2.5 % of the APIn (Start), and the date when the sum land.copernicus.eu/pan-european/corine-land-cover reaches 97.5 % of the APIn (End). Then, the pollen/spore parameters in- The weather in Catalonia is characterized by processes related to circu- cluded in this study were: the dates of Start and End of the MPS; the date lation patterns in mid-latitudes, but also by subtropical phenomena. Its cli- of the Peak, or date of the daily maximum concentration in a year; and matic diversity is greatly influenced by its proximity to the Mediterranean the Seasonal Pollen Integral (SPIn) or Seasonal Spore Integral (SSIn), or Sea and the complex orography of the region (Clavero et al., 1996; Lana sum of the daily pollen or spore concentrations between the Start and et al., 2020; Llabrés-Brustenga et al., 2020). The Pyrenees, along the End dates (Galán et al., 2017). north border between France and Catalonia, act as a barrier in many parts Based on the intensity of the recorded pollen and spore concentrations of the country against the northern and north-western flows (Izquierdo and their impact on the allergic population, each day was classified, for et al., 2017). In the same way, the effects of the Mediterranean disturbances each pollen/spore type, into 5 levels of risk of allergy, being 0 the lowest are limited to the coastal regions and have a lesser effect inland because the concentration and risk, and 4 the highest concentration and maximum Littoral and Pre-littoral chains, parallel to the coastline from north to south, risk. The concentration of particles assigned to the allergenicity risk levels reduce their effects. Likewise, the passage of the westerly winds that come differ from one taxon to another (Docampo et al., 2007; Galán et al., from the Atlantic to the Mediterranean coast through the IP weakens the ef- 2010). We have grouped levels 0, 1, 2 into a single group LC (low pollen/ fect of the Atlantic disturbances, reducing the intensity of the rains, being spore concentration) and levels 3 and 4 into the single group HC (high pol- only North-western Catalonia significantly affected. Consequently, the len/spore concentration). LC for Platanus, Cupressaceae and Olea include coastal and pre-coastal zones, where Barcelona is located, are governed daily concentrations from 0 to 50 pollen m−3, for Poaceae, Urticaceae mainly by Mediterranean advection, with the effect of breezes also being and Amaranthaceae from 0 to 8 pollen m−3 and for Alternaria from 0 to very important. Barcelona has a typically Mediterranean climate, character- 30 spores m−3. HC are the concentrations over the LC upper threshold. ized by mild temperatures and infrequent but sometimes torrential rains, concentrated mainly in autumn and spring. In the 19-year period studied 2.3. Synoptic classification here, the average annual rainfall in BCN was 620 mm, with an average an- nual temperature of 16 °C, and relative humidity of 68 %. Non-hierarchical k-means cluster analysis was the technique employed by Salvador et al. (2021) for classifying into similar groups, reanalysis data 2.2. Pollen record fields of sea level pressure (SLP) at 12 UTC in the period 2001–2019. First, SLP global fields at 12 UTC derived from the National Centre for Environ- Airborne pollen data were produced by the Aerobiological Network of mental Prediction/National Centre for Atmospheric Research (NCEP/ Catalonia (Xarxa Aerobiològica de Catalunya, XAC) at Barcelona (BCN, NCAR) Reanalysis dataset (Kalnay et al., 1996) provided by NOAA/OAR/ 41° 23′ 37.42” N, 2° 09′ 53.72″ E, 93 m a.s.l.) (Fig. 1) over the 19-year ESRL PSD, USA were obtained for all the days of the period of study. Data were distributed in each field in a 2.5° latitude x 2.5° longitude global grid. Those data contained in the region located between 50°W-50°E and 20°N-70°N were downloaded, resulting in 861 SLP grid points for each daily field. Belis et al. (2019) recommend performing classifications of syn- optic meteorological patterns (SMPs) using SLP data fields, for discriminat- ing periods of high atmospheric stability. Under such stagnant meteorological conditions, concentration of pollen and spores from local plants and trees can reach very high levels. The k-means cluster is one of the statistical methods most widely used for classification of atmospheric circulation patterns. It comprises 5 stages (Belis et al., 2019; Salvador et al., 2021): Stage 1: define an initial partitioning of the SLP fields: k daily fields rep- resenting different representative synoptic meteorological situations in Fig. 1. Geographical location of the Barcelona aerobiological station. the study area were selected as initial cluster centers. In this study, the 3 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 most frequent atmospheric circulation patterns over the Iberian Penin- meteorology associated with each SMP. The median and average concen- sula were selected. They were identified from previous studies and trations as well as the Pollen/Spore Integral values corresponding to the used as initial conditions in the clustering procedure. dates of each synoptic meteorological pattern (SMP) in Barcelona were also computed. Besides, Pearson and Spearman correlations between the Stage 2: calculate the change in the clustering criterion that result from number of days with a given SMP during the Main Pollen/Spore Season changes in membership and reassign SLP fields. Hence, the euclidean (MPS) and the Seasonal Pollen/Spore Integral (SPIn/SSIn) were obtained. distance from each SLP field to each cluster-center k was calculated In addition, two different statistical approaches, described in the follow- for every grid-point value of their 861 SLP observations and summed. ing sections, were applied. Finally, the SLP field was assigned to the cluster with the smallest total distance from its cluster center. 2.5.1. Frequency comparison Stage 3: recalculate the cluster centers after all the SLP fields have been First, the time periods of likely occurrence (LOPs) of the Start, End and examined and assigned. The cluster centers were recalculated as the ar- Peak parameters for each taxon were defined as the period between the ear- ithmetic mean of all members of any cluster. liest observed date of occurrence and the latest one in the period Stage 4: repeat the steps 2 and 3 iteratively until no SLP field changes its 2001–2019. Then, we have considered two types of date samples, each cluster assignment. Then, composite synoptic maps could be obtained date with its corresponding SMP: one sample corresponds to the dates in which the occurrence of the parameters was observed (observational sam- by averaging all the SLP fields allocated in each group, grid-point by ple with 19 values for each parameter and taxon) while the other corre- grid-point. sponds to all the dates of the LOPs. Next, the percentage frequency of the Stage 5: The optimum number of clusters to be retained was selected ac- different SMPs in each sample is compared to assess the influence of each cording to the percentage change in within-cluster variance as a func- SMP on the occurrence of the pollen/spore time parameters. If the percent- tion of the number of clusters. This statistic increases abruptly as age of a specific SMP is the same or similar in both types of samples, it could clusters which are significantly different are joined. Moreover, a valida- be interpreted as that the SMP of the day did not influence the parameter to tion procedure on the resulting clusters was performed with the aim to occur. Instead, if the percentages are significantly different the SMP could assure their physical meaning. It was based on the analysis of heteroge- have influenced it. neous data sets of meteorological variables and atmospheric stability Let us denominate as P the occurrence percentage corresponding to a parameters, obtained from measurements performed in an instru- specific SMP in the observational sample for a time parameter and taxon mented tower, numerical models and radiosondes carried out in the (sample size of 19), and PLOP the percentage corresponding to the same SMP in the LOP sample for the same time parameter and taxon. For each central region of the Iberian Peninsula. taxon and parameter, the P and PLOP percentages corresponding to each SMP can be calculated as follows: This circulation classification methodology produced six different clus- ters each one representing a prevailing SMP over the IP. Detailed informa- Number of occurrences in the SMP P¼ x100 (1) tion on the characteristics of the reanalysis SLP data fields, the circulation 19 classification process and the validation procedure of the results, can be Number of occurrences in the SMP found in Salvador et al. (2021). The composite synoptic weather maps rep- PLOP ¼ x100 (2) Number of LOP dates x 19 resenting each SMP, which were obtained by averaging the SLP corre- sponding to the days of the different clusters and the main features that To compare these two percentages and assess the difference between characterize the SMPs are provided as Supplementary material (Fig. A.1). them, a statistical test has been applied. Comparing proportions between the two samples, the test determines if the percentage differences are statis- 2.4. Meteorological data tically significant or only a consequence of the variability derived from samples of finite sizes. To this end it has been constructed the statistic Z The influence of the 6 SMPs on the surface meteorological parameters as a variable whose values follow a standardized normal probability distri- registered in the Madrid metropolitan area, located in the center of the bution with a mean of 0 and a standard deviation of 1: IP, was analysed as part of the validation procedure in Salvador et al. (2021). In the present study, meteorological data were also used in order P  PLOP Z ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (3) to characterize each synoptic type with the local meteorology of the aerobi- PLOP ð100  PLOP Þ=N ological sampling site, in this case, the city of Barcelona, and thus represent the average meteorological conditions during each SMP. The data, pro- N is the number of years of the observational sample (19 years). For a vided by the Servei Meteorològic de Catalunya (SMC), were recorded in significance level of 95 % (p ≤ 0.05), the value of Z must exceed 1:96 the Fabra Observatory (41°25′N, 2°07′E, 415 m a.s.l.), at approximately to ensure that the percentages are significantly different. 5 km north of the pollen sampling site. Daily values of maximum tempera- ture (Tmax), minimum temperature (Tmin), mean temperature (Tmean), 2.5.2. Performance index precipitation (PREC), relative humidity (RH), solar irradiance (SI), mean In order to evaluate the association of each Pollen/Spore Integral to sea level pressure (SLP), and wind speed and direction at 10 m above each SMP a Performance Index (PI), defined as (Paschalidou et al., 2020): ground level were used. The available periods for the different variables " # and used in this study were: Tmax, Tmin, Tmean, 2001–2019 (with the ex- ðPollenðSporeÞ Integral in SMPÞ=ðPollenðSporeÞ Integral in the SeasonÞ PI ¼ 100x −1 ception of 10 days in May 2002 and 8 days in December 2003); PREC, Number of days in SMP=Number of days in the Season 2001–2019; RH, 2001–2019 (with the exception of November 2003); SI, ð4Þ 2002–2010; SLP, 2002–2019 (with the exception of 2007); wind, 2013–2019. where Pollen/Spore Integral in the Season is the sum of SPIn/SSIn of the 19-year period, and Pollen/Spore Integral in SMP is the sum of daily con- 2.5. Statistical methods centrations corresponding to each SMP along the 19-year period. When the values of PI approach 0, it can be interpreted as the daily Boxplots of the daily mean values of the meteorological variables show- mean pollen/spore concentration is uniformly distributed in different syn- ing the median and the interquartile range (range between the 25th and optic patterns. Instead, PI values >0 mean that higher daily pollen concen- 75th percentile) were obtained in order to characterize the local trations are more frequent under the specific SMP, and negative PI values 4 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 2. Summary of the daily dynamics of the mean (blue line) and maximum (orange line) pollen/spore concentration (Pollen m−3, Spore m−3) of the studied taxa along the year during the period 2001–2019 in Barcelona. 5 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Table 1 3.2. Synoptic meteorological patterns and local associated meteorology Number of days and the corresponding percentage with low (LC) and high (HC) pol- len/spore concentration during the period (shown in Table 1) for each taxon. The The dynamics of each SMP can be seen in Fig. 3, where the absolute fre- minimum and maximum concentrations considered here for allergenicity risk of quency throughout the year is shown. Boxplots of the time series of daily each taxon are also shown. mean values of the meteorological variables maximum temperature LC HC (Tmax), minimum temperature (Tmin), mean temperature (Tmean), rela- Min-Max Number of days (%) Min-Max Number of days (%) tive humidity (RH), solar irradiance (SI), mean sea level pressure (SLP), (Pollen·m−3) (Pollen·m−3) wind speed (W10m), total precipitation and average daily precipitation PLAT 0–50 895 (59) 50–3263 610 (41) for the different synoptic meteorological patterns (SMP) in the meteorolog- CUPR 0–50 4847 (90) 50–1895 568 (10) ical station of BCN are shown in Fig. 4. The wind rose plot for each synoptic OLEA 0–50 880 (82) 50–463 194 (18) meteorological pattern obtained from the wind data at 10 m a.g.l. is shown POAC 0–8 3092 (81) 8–136 702 (19) in Fig. 5. URTI 0–8 3701 (71) 8–132 1509 (29) AMAR 0–8 3718 (95) 8–45 183 (5) SMP-1 was the type of circulation that presented the lowest values of ALTE 0–30 3820 (68) 30–1081 1805 (32) Tmax, Tmin, Tmean, RH and SI, and the highest values of pressure and wind speed. Under the influence of this typically winter regime, NW and W flows were dominant in BCN mainly in dry sunny days, in an atmosphere mean that daily pollen concentrations are lower. Positive values of PI might of strong stability. It was absolutely absent from April to September (Figs. 3, be reflecting severe conditions of allergenic risk, while negative ones would 4, 5 and A.1). correspond to situations for which a lower pollen/spore concentration than In contrast, SMP-2 presented the highest values of temperatures and SI, the mean has been collected, and therefore, a less risk of allergenic condi- and the lowest values of RH, as well as light winds from the SW and W, tions. sometimes accompanied by precipitations of convective origin. This typical spring-summer pattern presents the best meteorological conditions to fa- 3. Results vour pollination. SMP-3 presented lower temperatures and solar radiation, but higher RH 3.1. Pollen parameters overview than SMP-2 and the lowest pressure of all SMPs. Frequent in spring and au- tumn, it produced episodes of low pressure above NW IP accompanied by Daily mean and daily maximum concentrations in BCN for the period storms and moderate/high westerlies (Figs. 3, 4, 5 and A.1). 2001–2019 are shown in the Fig. 2. SMP-4 was more frequent in autumn/winter than in spring/summer, As shown in Table 1, the number of days with LC in the period consid- and the associated temperatures in BCN were quite similar to those of ered is always higher than the number of days with HC, although this could SMP-3. It was characterized by the presence of an extensive anticyclone change in some cases if we do not consider the 0 concentration values or if over Eastern Europe and a northward displacement of the Atlantic low, we consider the exact pollination period of a particular year. The particles with SI, pressure and HR values higher than those of SMP-3. Winds were being able to cause allergenic troubles during a longer period were more moderate and did not show a predominant direction. Alternaria, followed (decreasing order) by Urticaceae, Poaceae, Platanus, SMP-5 was more frequent in spring/summer than in autumn/winter. It Cupressaceae, Olea and Amaranthaceae. was characterized by high or moderate temperatures and SI, and relatively The time periods of likely occurrence (LOPs) of the pollen/spore param- low RH values, prevailing smooth fluxes from the SW. This was the pattern eters Start, End and Peak for the studied taxa at the BCN station in the pe- with the highest accumulated precipitation and the highest mean value of riod 2001–2019 are listed in Table 2. daily precipitation, after SMP-4. SMP-2 represents the typical summer pat- As shown in Fig. 2 and Table 2, the taxa studied have a rather different tern in BCN, while SMP-5 together with SMP-3 would represent the preva- airborne dynamics not only when considering the moment of the year in lent SMPs occurring in spring. which the pollination/sporulation period occurs but also in the concentra- SMP-6 was a winter/autumn pattern, totally absent in summer, and tion values per day, which vary notably along the year and from year to gave rise to moderately low temperature and SI values, high pressures at year, these shown by the maximum daily concentrations (orange line in the surface level, and almost no precipitation in BCN. This pattern, which Fig. 2). Some pollen types are present in the air throughout the year like the SMP-1 resulted in high atmospheric stability associated with anticy- (Cupressaceae, Urticaceae and Amaranthaceae), other are more limited clonic weather. on time (Poaceae, Platanus and Olea,). Alternaria shows a yearly spread The mean value of the meteorological variables in each SMP during the sporulation. As also shown in Fig. 2 and Table 1, pollen from trees is pollination period of each taxon has been computed and represented in more abundant (Platanus peak mean values ranging 500 Pollen m−3 in a Fig. 6 for Platanus and Cupressaceae. On the left, wind speed versus pres- day but with an absolute maximum ranging 3.300 Pollen m−3; sure is represented, and, as a third variable, the difference between the av- Cupressaceae 500 and 2.000 Pollen m−3; and Olea 80 and 500 Pollen erage daily precipitation in each SMP and the average daily precipitation in m−3) than pollen from herbs (Poaceae 25 and 130 Pollen m−3; Urticaceae the 19-year period (balls size), in blue positive values and, in grey, negative 25 and 100 Pollen m−3; and Amaranthaceae 10 and 45 Pollen m−3). values. On the right, solar irradiance as a function of mean temperature is Alternaria ranges between 100 and 1.000 Spore m−3 per peak day. represented, and the difference between the average relative humidity Table 2 Earliest and latest observed days (day of the year and date) in which the Start, End and Peak pollen/spore parameters occurred, in the period 2001–2019 at the BCN station. The last column indicates the average length (number of days) of the main pollen/spore season (MPS/MSS). Start End Peak Length Earliest Latest Earliest Latest Earliest Latest Days PLAT 67 (3/7/16) 90 (3/31/05) 103 (4/13/14) 257 (9/14/05) 75 (3/16/07) 96 (4/5/05) 84 CUPR 8 (1/8/13) 37 (2/16/10) 146 (4/27/11) 354 (12/21/13) 39 (2/8/04) 86 (3/27/05) 296 OLEA 106 (4/16/01) 133 (5/12/16) 159 (6/8/17) 212 (7/31/18) 125 (5/5/03) 165 (6/14/07) 58 POAC 58 (2/27/06) 112 (4/11/12) 254 (9/11/07) 295 (10/22/05) 128 (5/8/15) 188 (7/7/19) 195 URTI 20 (1/20/07) 76 (3/17/10) 285 (10/11/04) 349 (12/15/19) 89 (3/30/06) 212 (7/31/01) 279 AMAR 37 (2/6/07) 116 (4/26/02) 287 (10/14/03) 320 (11/16/09) 123 (5/3/15) 258 (9/15/01) 204 ALTE 18 (1/18/11) 87 (3/28/15) 317 (11/13/01) 354 (12/19/04) 64 (3/4/08) 312 (11/8/01) 298 6 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 3. Absolute frequency of each SMP along the year in the period 2001–2019. and its value in each SMP as third variable. Both taxa presented similarities, and solar irradiance. However, SMP-3 showed differences in precipitation e.g., SMP-1 was the coldest and driest synoptic type. SMP-2 also presented and relative humidity, with higher values in the Cupressaceae period, and similar characteristics in both pollination periods with high temperatures higher solar irradiance in the Platanus period. 7 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 4. Boxplots of daily mean values of the meteorological variables showing the median and the interquartile range (range between the 25th and 75th percentile) corresponding to the dates of each synoptic meteorological pattern (SMP) in Barcelona. 8 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 5. Wind rose plot for each synoptic meteorological pattern (SMP) obtained from the wind data at 10 m a.g.l. registered in the meteorological station of Barcelona. The mean value of the meteorological variables in each SMP during the 2 was also the circulation type that presented the highest concentrations pollination period of the other taxa is shown in Supplementary material for Amaranthaceae. (Olea/Poaceae in Fig. A.2, Urticaceae/Amarantaceae in Fig. A.3 and Cupressaceae, with the longest pollination period, has an interannual Alternaria in Fig. A.4). It is remarkable the similarities between Urticaceae, pollen season from October to April. Unlike Urticaceae and Poaceae, also Poaceae and Amarantaceae. There are differences only in SMP-1 and SMP-3 with long pollination periods, it was during the most frequent synoptic for precipitation, more abundant in the Urticaceae period for which the types in autumn/winter (SMP-1, SMP-6 and SMP-3) that the highest daily wind is also slightly higher. The greatest differences between the taxa in mean values of Cupressaceae pollen were recorded. The lowest daily the local meteorology linked to the SMPs correspond to the pollination pe- mean values of Cupressaceae occurred in SMP-2, which, however, was riod of Olea, in which the coldest synoptic type, with the least insolation the most frequent synoptic type for this taxon (Table 3). The spring/sum- and the highest precipitation, is SMP-5, being SMP-6 the windiest pattern mer pollen Olea presented an Integral value of zero during SMP-1 and the with the lowest relative humidity (Supplementary material, Fig. A.2). highest median values corresponded to SMP-3. For this taxon, the highest daily mean concentration corresponded to SMP-6 (Table 3), which was 3.3. Pollen/spore concentration and synoptic types. Correlations the synoptic pattern with the lowest relative humidity and higher wind The also spring pollen Platanus presented only 12 cases (out of a total of Boxplots with the median pollen/spore concentrations (Pollen m−3) 1505) in SMP-1, but all of them with very high values (mean 1094 Pollen and the interquartile range (range between the 25th and 75th percentile) m−3), in the range of high allergenicity risk (Table 1); it was followed by of the daily pollen/spore time series registered during each synoptic pat- SMP-6, SMP-3,and SMP-4 with also high daily concentrations (mean values tern, for the Main Pollen/Spore Season (MPS) of each taxon, are shown in of 390, 312, and 175 Pollen m−3, respectively), included in the HC. The Fig. 7. The number of cases N (days with measured concentration during fungal spore Alternaria showed the highest values of median and mean dur- the MPS of each taxon), the percentage of HC days with respect to the ing SMP-2 episodes. total days, the Pollen/Spore Integral, and the mean daily concentration in Significant correlations have been found between the number of days each synoptic pattern are shown in Table 3. The most abundant synoptic with a given SMP in the Main Pollen/Spore Season and the Pollen/Spore In- type (Supplementary material, Fig. A.1), SMP-2, presented the highest tegral in the period, SPIn/SSIn. Table 4 shows the Pearson and Spearman total number of cases for all taxa. The highest median and mean values correlation coefficients obtained for the level of significance p ≤ 0:1. It is for Urticaceae corresponded to SMP-5 followed by SMP-2, and the lowest remarkable the fact that all the significant correlations were negative. values corresponded to SMP-4. In the case of Poaceae, also the highest SMP-2, the most frequent SMP, correlated with 5 out of the 7 taxa, some mean values were in SMP-5 and SMP-2, but the median value in SMP-2 with long pollination periods, i.e. Cupressaceae, Urticaceae and Poaceae; slightly exceeded that of SMP-5, while SMP-1 was the synoptic type with and other with the shortest ones, i.e. Platanus and Olea. The fungal spore the lowest median and frequency for this taxon. On the other hand, SMP- Alternaria, with the longest period of presence in the air, showed no 9 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 6. Mean value of the meteorological variables in Barcelona in each synoptic meteorological pattern during the Main Pollen Season of Platanus and Cupressaceae for the period 2001–2019. Left: wind speed vs. pressure; balls size represents the difference between average daily precipitation in the SMP and average daily precipitation in the 19- year period (blue and grey, positive and negative values, respectively). Right: same than left for the variables: mean temperature, solar irradiance and relative humidity. correlation with any of the synoptic types. Platanus is the pollen that the dates 64 and 312, multiplied by 19 years) (Table 2). Fig. 9 shows the showed more correlations: with SMP-2, SMP-4 and SMP-5. Cupressaceae percentage differences P  PLOP for each taxon, parameter and SMP. The correlated with SMP-2 and SMP-4, while Poaceae correlated with SMP-2 statistic Z of Eq. (3) has been applied to determine if the difference of and SMP-3. Urticaceae and Olea only had significant correlations with these percentages is statistically significant or not. For a significance level SMP-2. It should also be noted that if we go up to the significance level of of 95 % (p ≤ 0:05) the sample percentages are significantly different if p ≤ 0:05, only Cupressaceae and Platanus showed significant correlations. the value of Z exceeds 1:96. As an example, Table 5 shows the Z values These correlations were with the circulation types SMP-2 and SMP-5 obtained for the parameter Start of Urticaceae. (Fig. 8). It only exceeds the threshold value of 1:96 for SMP-1 (2.52). There- The number of days with LC (low pollen/spore concentration) and HC fore, it can be stated that the number of times for which the parameter (high pollen/spore concentration) in each synoptic meteorological pattern Start occurred in a SMP-1 day is higher than the expected number, in rela- is shown in Table 3. Platanus pollen had the highest proportion of HC tion to the percentage of SMP-1 days in the corresponding LOP. Instead, days in most of the patterns. For this taxon, during SMP-1, HC occurred negative values of Z would indicate that the time parameter occurred less on 100 % of the days, near 90 % during SMP-6, over 60 % during SMP-3 than expected under the considered SMP, i. e., the SMP would disfavour and SMP-4, and around 30 % during the other patterns. The HC percentages the occurrence of the time parameter. were also high for Urticaceae, with over 33 % of HC days under SMP-2 and Asterisks in Fig. 9 show all the cases with significant percentage differ- SMP-5. Poaceae also presented high concentrations for these same synoptic ences after applying the test Z for each taxon and time parameter. In all types. The fungal spore Alternaria, which is present every day of the year, these cases Z resulted positive, i. e., the specific parameter occurred more registered the highest number of days under SMP-2, being the HC percent- frequently than expected for the indicated SMPs. One of these cases is the age of 45 % under this synoptic pattern. aforementioned Start of Urticaceae for the SMP-1, with a percentage of SMP-2 is the circulation pattern concentrating the highest proportion of PLOP ¼ 7% among the dates of the corresponding LOP while the percentage HC days (Alternaria, Urticaceae, Poaceae and Amaranthaceae), followed by among the observed dates is P ¼ 21%, which is the maximum percentage SMP-1, SMP-3 and SMP-5. None of the taxa showed a higher proportion of among the Start dates of Urticaceae. According to this, the type of synoptic HC days under SMP-4 and SMP-6. pattern had an influence on the Start occurrence of Urticaceae, and SMP-1 was the most influential. 3.4. Relationship between SMPs and pollen/spore parameters. Frequency com- The End parameters of Cupressaceae, Platanus, Amaranthaceae and parison Alternaria show percentage differences for diverse SMPs. In dates with these SMP the main pollen/spore season of the mentioned taxa has more To evaluate the influence of each SMP on the occurrence of pollen/ frequently ended than expected considering the observed occurrence prob- spore time parameters, a comparison has been done between the P percent- ability P. The greatest percentage differences corresponded to dates with age and PLOP percentage for each SMP. While the samples of observed dates SMP-2 for Amaranthaceae (P ¼ 32% vs. PLOP ¼ 13%) and dates with of occurrence have a size of 19 days, one per year, the size of the samples SMP-5 for Alternaria (P ¼ 32% vs. PLOP ¼ 14%). The case of dates with corresponding to each LOP is much longer, ranging from 418 days for the SMP-4 for Cupressaceae is debatable because this significant difference is Peak of Platanus (22 days between the first (75) and last (96) dates, multi- due to two anomalous years for which the End occurred much earlier plied by 19 years) to 4731 days for the Peak of Alternaria (249 days between than commonly: day 146 in 2011 (May 26) and day 149 in 2019 (May 10 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 7. Boxplots of the time series of daily pollen/spore concentrations (Pollen m−3 and Spore m−3) during the Main Pollen/Spore Season showing the median values (line inside the box) and the interquartile range (range between the 25th and 75th percentile, limit of the boxes) registered in each synoptic meteorological pattern (SMP) in Barcelona for the time period 2001–2019. 29). In the rest of the years the End always occurred after summer, the ear- year than the natural year. In this case, the series of End dates don't show out- liest in day 283 (October 10). Because of these two anomalous Ends, the LOP liers, but significant percentage differences appear for SMP-2 and SMP-6. of Cupressaceae has been enlarged in 137 days, with consequences in the per- Considering the observed occurrence probability PLOP , in dates with SMP-6 centages of each SMP in this period. If these two years are not considered, the main pollen season of Cupressaceae has more frequently ended than ex- none of the SMPs show significant percentage differences. Another possible pected, being less frequently ended under SMP-2 dates. consideration to make is to take an interannual season pollen for this taxon, For Platanus, a higher percentage of End occurred under SMP-6 dates, since it adjusts better from October of the previous year to April of the current and a higher percentage of Peak occurred during SMP-1 dates. Even though 11 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Table 3 pollination days in each SMP (see Table 3) besides the proportions that ap- Number of days with measured concentration during the pollination/sporulation pear in Eq. (4). For instance, the positive values of PI found for Urticaceae, period of each taxon (N), percentage of HC days with respect to the total days in Poaceae and Amaranthaceae during SMP-2 have diverse significance in the SMP (HC/N), Pollen/Spore Integral (units: Pollen/Spore*day m−3), and mean terms of allergenic risk. Although Table 3 shows mean concentration of 8, daily concentration (units: Pollen/Spore m−3), in each SMP. 6 and 3 Pollen m−3 respectively, which can seem low, the respective num- SMP-1 SMP-2 SMP-3 SMP-4 SMP-5 SMP-6 ber of HC days under this SMP-2 are 33 %, 21 % and 5 % On the other hand, N 12 649 210 188 373 73 for the interpretation of the high positive values of PI found for HC / N (%) 100 26 62 57 34 88 Cupressaceae and Platanus in SMP-1 (295 and 512 respectively) and for Pollen Integral 13,132 81,880 65,545 32,972 47,816 28,450 Olea in SMP-6 (107) (Fig. 10), it has to be kept in mind that even though PLAT Daily mean 1094 126 312 175 128 390 mean daily pollen concentrations are important in SMP-1 for Cupressaceae N 97 1927 802 869 1225 495 HC / N (%) 37 5 18 9 8 22 (98 Pollen m−3) and Platanus (1094 Pollen m−3), and in SMP-6 for Olea (58 Pollen Integral 9542 28,095 30,944 21,967 23,399 22,954 Pollen m−3) (Table 3), the pollen concentrations in these synoptics situa- CUPR Daily mean 98 15 39 25 19 46 tions last few days: around 5 days per year in the case of Cupressaceae N 0 543 101 67 354 9 and <1 day per year for Platanus and Olea. But, when they occur, the con- HC / N (%) 0 17 32 12 18 11 Pollen Integral 0 14,821 4318 1455 10,231 526 centrations are HC in 100 % of the cases for Platanus, 37 % for Cupressaceae OLEA Daily mean 0 27 43 22 29 58 and 11 % for Olea. N 6 1779 387 490 1022 110 However, the value of the index PI is very significant for Cupressaceae HC / N (%) 0 21 16 9 21 6 and Platanus in SMP-3 and SMP-6 (53 and 85 for Cupressaceae and 75 Pollen Integral 5 10,601 1607 1471 5796 216 and 114 for Platanus), since the pollen concentrations registered in these POAC Daily mean 1 6 4 3 6 2 N 71 2054 749 793 1145 398 synoptic situations are much more frequent, especially for Cupressaceae HC / N (%) 14 33 28 20 33 17 with >42 days per year in SMP-3 (18 % HC days) and >26 days per year Pollen Integral 298 16,766 5231 4514 9586 1907 in SMP-6 (22 % HC days) (Table 3). For Platanus, 88 % of the days under URTI Daily mean 4 8 7 6 8 5 SMP-6 showed HC and 57 % under SMP-3. Moreover, the negative values N 7 1824 484 511 918 157 HC / N (%) 0 5 5 4 4 2 of PI for these two taxa in the SMP-2 and SMP-5 situations are also very re- Pollen Integral 4 4644 951 911 1973 194 markable because these are the most frequent synoptic situations during AMAR Daily mean 1 3 2 2 2 1 their main pollen season (26 % and 34 % of HC days for Platanus and 5 % N 106 2090 836 878 1218 497 and 8 % for Cupressaceae, respectively). This result is in accordance with HC / N (%) 16 45 22 22 32 18 the finding of a negative correlation between the Pollen Integral collected Spore Integral 2050 80,041 18,763 18,654 37,156 10,125 ALTE Daily mean 19 38 22 21 31 20 during the Cupressaceae and Platanus main pollen seasons and their occur- rences (see Table 4 and Fig. 8). The large negative values of PI in the SMP-1 synoptic situation for the expected probability was very low in both cases (PLOP ¼ 1%), the ob- Poaceae (−84) and Amaranthaceae (−73) indicate the almost absence of served probability was much higher (P ¼ 11%) since two times of the pollen in these situations, and the complete absence for Olea (−100). total of 19 years the End of Platanus occurred in SMP-6, and also two For Alternaria the values of PI in all the considered SMPs range between times of the 19 years the Peak day was SMP-1. −34 (SMP-1, with the lowest N and only 16 % of HC days) and + 29 (SMP- 2, with the highest N and 45 % of HC days). The spore concentrations are 3.5. Performance index more uniformly distributed among the different synoptic meteorological situations. The association of Pollen/Spore Integral to each SMP has been assessed using the Performance Index (PI) of Eq. (4). This PI index has been calcu- 4. Discussion lated for each SMP and taxon (see Fig. 10). PI index is useful to show, for each taxon, which SMPs comprise the days with the highest concentrations, The possible connection between the most frequent synoptic situations suggesting that some meteorological conditions might result more severe in affecting the Iberian Peninsula and the daily concentrations and timing of terms of risk of allergy. Anyway, this index must be interpreted carefully, some selected airborne allergenic pollen and spore taxa in the city of Barce- bearing in mind the daily concentrations and the number of LC and HC lona has been analysed. To this end, we have used several complementary Table 4 Significant correlations at p ≤ 0:1 and correlation coefficients, Pearson's r and Spearman's ρ, between the number of days with a given SMP during the Main Pollen Season and the corresponding Pollen/Spore Integral (SPIn/SSIn). All correlations are negative. SMP-2 SMP-3 SMP-4 SMP-5 PLAT Pearson p = 0.008 p = 0.085 p = 0.031 r = −0.591 r = −0.406 r = −0.497 Spearman p = 0.021 p = 0.084 p = 0.04 ρ = −0.526 ρ = −0.407 ρ = −0.474 CUPR Pearson p = 0.007 p = 0.014 r = −0.596 r = −0.556 OLEA Pearson p = 0.075 r = −0.418 POAC Pearson p = 0.096 p = 0.055 r = −0.393 r = −0.447 Spearman p = 0.09 p = 0.039 ρ = −0.399 ρ = −0.476 URTI Pearson p = 0.068 r = −0.427 Spearman p = 0.024 ρ = −0.515 AMAR Spearman p = 0.084 ρ = −0.407 12 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 8. Seasonal Pollen Integral SPIn vs. number of days in the synoptic meteorological pattern (SMP) for the taxa with significant correlations at the level p ≤ 0:05. statistical methodologies that have provided different results, coherent values during SMP-6 (Supplementary material, Fig. A.2) could explain why among them. The synoptic types, obtained by cluster analysis of the surface the highest daily mean concentration is found in this pattern. pressure fields, showed a clear seasonality. The scenario described as SMP- The winter SMP-1 scenario, which is not very relevant for the rest of the 1 is dominant in winter, along with SMP-6 which is also frequent in fall. In taxa, is, however, the one that registers the highest mean and median summer and spring SMP-2 predominates. The SMP-5 pattern occurs in all values for Cupressaceae and Platanus (Table 3, Fig. 7), consistent with the seasons, and similarly SMP-3 and SMP-4, but to a much lesser extent in very high values also obtained of the Performance Index (PI, Fig. 10). summer the last two. This seasonality, added to the phenology of the taxa This is remarkable considering that SMP-1 is the least frequent type. The studied here, makes it difficult to interpret possible relationships. The reason for the high concentration of these taxa could be, added to the explo- taxon with the shortest pollination period is Olea followed by Platanus, sive pollination they show, that the configuration linked to SMP-1 causes which flower in summer and spring, respectively. On the other hand, blocking of air-mass and high stability over the IP and the Mediterranean Urticaceae, Poaceae and Amaranthaceae pollen types are present in spring, Sea due to the position of the Azores anticyclone, displaced in front of the summer and autumn, while Cupressaceae pollen and Alternaria spore are IP, and the deep low at the latitude of Iceland, extending to the east (Supple- present throughout the year. The interannual pollen seasons, as is the mentary material, Fig. A.1). During atmospheric blocking situations, high case of the Cupressaceae, represent an added difficulty. Another special pressure zones remain stationary for several days, giving rise to stagnation case is the bimodal patterns of the pollen and spore curves, such as the of air mass, accumulation of pollutants, including pollen and spores, and, case of Amaranthaceae, Alternaria or even Urticaceae pollen (Picornell eventually, causing serious air pollution situations (Otero et al., 2022). et al., 2022). All these particularities in pollen dynamics add complexity Platanus and Cupressaceae also showed both the highest daily mean con- to the analysis of the influence of synoptic meteorological conditions. The centration in SMP-1 and SMP-6 (Table 3). This coincidence could be due preliminary results obtained from the descriptive statistics attributed to to the similarity of the meteorological conditions in their respective pollina- SMP-2 an important influence on the pollen/spore concentration levels tion periods (Fig. 6). The conditions in SMP-1, SMP-2, SMP-5, SMP-4 were measured in Barcelona. This is not surprising since this scenario is linked similar, although with higher precipitation and relative humidity during to high temperatures, low precipitation, and moderate north-westerly SMP-4 in Cupressaceae. This presence of precipitation could explain the winds (Figs. 4 and 5), a favourable meteorology for pollination. Moreover, favouring of the End of Cupressaceae obtained for this synoptic pattern it is the SMP presenting the highest absolute frequencies along spring and (Fig. 9). The main differences are observed during SMP-6, with higher rel- summer. However, in subsequent analyses with the other methods, it has ative humidity in Cupressaceae and higher temperature and solar irradi- been seen that this pattern is not as influential. Nevertheless, SMP-2 is the ance in Platanus, and during SMP-3, with higher precipitation and relative most frequent pattern in absolute terms (one third of the total days in a humidity in the Cupressaceae period and higher insolation in the Platanus year), and it is also the most frequently occurring during the pollination pe- period (Fig. 6). riod of all taxa. The performed analyses show that SMP-2 is especially rele- The analysis of the SPIn and SSin reflected in Table 4 and Fig. 8 reduces vant for those taxa with long pollination periods such as Amaranthaceae, the possible masking effect that seasonality of the variables can introduce. Urticaceae, Poaceae and Alternaria. Instead, for the only pollen that polli- It is noteworthy that no significant positive correlations at p ≤ 0:1 level nates throughout the year, Cupressaceae, the median and mean values are were found between the number of days that a given synoptic pattern ap- low during the SMP-2 scenario. These contrasts could be explained from pears during the pollination period of the different taxa and the SPIn and the characteristics of the local meteorology associated with each pollination SSin. The few significant correlations obtained were all negative period and SMP. In the case of Urticaceae and Poaceae, both with long pol- (Table 4). In the case of Platanus and Cupressaceae we found negative cor- lination periods spanning from February to November, the coincidence in relations with SMP-2 and SMP-5 with a higher significant level (p ≤ 0:05) the high mean daily values observed during SMP-2 and SMP-5 (Table 3) (Fig. 8). These results agree with those obtained applying the Performance is consistent with the similarity in local meteorology associated to their Index (PI) in which negative values were obtained for these taxa during Main Pollen Season (Supplementary material, Fig. A.2 and A.3). In the SMP-2 and SMP-5, indicating lower concentrations than expected of case of Olea, the taxon that differs the most from the others in the local me- Platanus and Cupressaceae pollen during these scenarios. There is also a teorology associated with its MPS, the low relative humidity and high wind commonality between the different methods in describing SMP-1 scenario 13 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 9. Differences of sample proportions P  PLOP of each SMP for the different time parameters (Start, End and Peak) and taxon. P corresponds to the total of observed dates of occurrence of the time parameters, PLOP collects all the dates belonging to the likely occurrence periods (LOPs) of each parameter. Asterisks indicate the cases for which the differences are statistically significant. as the least relevant in the concentrations of Urticaceae, Olea, Poaceae and Amaranthaceae. In reference to Cupressaceae and Platanus, the PI method highlights the high presence of these taxa in the SMP-3 and SMP-6 scenar- Table 5 ios, in addition to the aforementioned SMP-1, in agreement with the high Number of occurrences and percentage P (%) of each SMP for the observed Start mean and median concentration values obtained. dates of Urticaceae in the period 2001–2019 (19 days), number of days and percent- The Performance Index complements the information provided by age PLOP ð%Þ of each SMP for all the days of the corresponding LOP (57 days: from Fig. 7 in the sense that PI involves a sort of “pollen intensity” (Pollen Inte- 20 to 76) in the period 2001–2019 (19 years), and value of the statistic Z comparing gral Season/Number of days in the season), and compares this “pollen in- both percentages. tensity” for the days of each SMP with the averaged “pollen intensity”. Number of Number of LOP dates in P PLOP Z Fig. 10 shows how in some cases the “pollen intensity” is much higher occurrences 19 years (%) (%) in 19 years (Total 1083 days) under some SMP than the mean. The comparison involves intensity rather than simple concentration, which can be useful in relation to high aller- SMP-1 4 72 21 7 2.52 SMP-2 1 154 5 14 −1.12 genic situations. For instance, Platanus in SMP-1 appears very high in SMP-3 3 227 16 21 −0.55 both Figs. 7 and 10. However, while Cupressaceae in SMP-1 achieves simi- SMP-4 2 244 11 23 −1.25 lar values than in SMP-3 and SMP-6 in Fig. 7, Fig. 10 is spotlighting how SMP-5 4 213 21 20 0.15 Cupressaceae pollen intensity is much higher under SMP-1 than under SMP-6 5 173 26 16 1.23 SMP-3 and SMP-6. 14 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Fig. 10. Performance Index (PI) for each SMP and taxon. Regarding the fungal spore Alternaria, the PI method does not highlight occurrence (LOPs) of the time parameters Start, End and Peak have re- any of the synoptic patterns, except for the positive value of SMP-2. There- vealed that SMP-1 is the most likely to initiate the Urticaceae pollination fore, it attributes similar concentration values in the different synoptic as well as to provide the date of the peak for Platanus. On the other hand, types, but higher in SMP-2, which is consistent with the fact that the the scenarios that favour the End are SMP-4 for Cupressaceae, SMP-2 for mean and median Alternaria concentration values are the highest in this Amaranthaceae, and SMP-6 for Platanus (Fig. 9). pattern that accumulates 45 % of days of high allergy risk (HC). It should Platanus is the most abundant ornamental tree in Barcelona and pre- also be noted that the SMP-2 and SMP-5 scenarios have registered an im- sents significant allergy problems every spring (Puiggròs et al., 2015). portant percentage of days HC in Platanus, Poaceae, Urticaceae and The Platanus tree produces large amounts of pollen during its pollination Alternaria (Table 3). period with an estimate of 3-10 × 106 pollen grains per inflorescence As it is well known, the phenological parameters of plants, mainly (Vrinceanu et al., 2021; Tormo Molina et al., 1996) and a threshold of clin- woody plants, depend not only on the current year, but also on the condi- ical symptoms in sensitized persons of 50 Pollen m−3 (Cariñanos et al., tions of the previous year. It is especially important for the start date of 2020). These concentrations are widely exceeded by the values registered the pollen season, which is a very complex process also based on past con- in Barcelona. Platanus pollen has been identified by the European Academy ditions, mainly in perennial plants (Campoy et al., 2011; Luedeling et al., of Allergy and Clinical Immunology (EAACI) as an important allergen re- 2013). Despite the fact that the current synoptic condition is not the deter- sponsible for respiratory symptoms and was introduced as mandatory for mining factor of the phenological moment of the plant, we found it interest- pollen allergen screening in patients with possible allergic rhinitis in ing to look for possible relationships and include here those results with Europe (Bousquet et al., 2012). The SMP-1 scenario presented high mean statistical significance (Fig. 9). The determination of time periods of likely values of this pollen, all them corresponding to high allergy risk (HC), 15 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 concentrated in only 102 days in the 19-year period studied, and it was also for the peak day of the episode showed low pressures over the IP and the most probable scenario for the peak day. Considering that pollen parti- the Western Mediterranean while high pressures moved northward, cles behave in the atmosphere in a similar way than the particulate matter over the British Islands, a scenario that is similar to that corresponding of the atmospheric aerosol (Sofiev et al., 2006, 2013), this result would co- to the SMP-5 type, linked in the present study to high concentrations of incide with the high values of daily mean particle number concentrations Platanus pollen in Barcelona. Therefore, the use of sea level pressure found by Salvador et al. (2021) in the urban background of the city of fields to characterize SMP seems to be a good option for this type of Madrid for this same synoptic type (SMP-1). In the analysis of this series, study, instead of trajectories or geopotential height fields at a given also in concordance with our results for Olea pollen, SMP-6 favoured the ac- pressure level, useful to identify long-range transport episodes. They cumulation of pollutants, gases and particulate matter, emitted from local correctly depict the atmospheric circulation at relatively high tropo- sources, and consequently the generation of high concentrations in the spheric levels in most cases. But they are not appropriate for describing Madrid metropolitan area (Salvador et al., 2021). the main meteorological features that drive surface atmospheric circula- We have not found a different behaviour in both concentrations and tions (Belis et al., 2019), which have been the most influential regarding timing, depending on whether the pollen types are from trees or herbaceous pollen concentration in BCN. plants, neither for the spore Alternaria, except for the case of the most signif- icant SMPs influencing their presence in the air. For the three arboreal pol- 5. Summary and conclusions len types studied, scenarios SMP-3 and SMP-6 were relevant and also scenario SMP-1 for Platanus and Cupressaceae. For the herbaceous pollen In this study we have established relationships between the occurrence types and the spore, scenarios SMP-2 and SMP-5 were relevant (except of specific synoptic meteorological patterns and pollen/spore dynamics in Amaranthaceae with only SMP-2). an urban environment, focusing on the seven most allergenic taxa abundant The physical mechanisms that determine the influence of synoptic in the study area. The city of Barcelona has been chosen as representa- meteorology on airborne pollen are complex and involve processes tive of an urban environment under a Mediterranean climate and due ranging from the micro-scale to the synoptic scale. In this study we to its location in an emplacement with a great diversity of flora and have focused on taxa of local origin, but long-distance contributions vegetation. Cluster analysis of the mean sea level pressure fields has pro- can also be important. In previous works, these external contributions vided six synoptic meteorological scenarios, which we have character- were studied through the analysis of pollen types that are scarce in the ized from the values of the local meteorological variables. The study region, such as Betula, Fagus and Ambrosia. We were able to asso- different statistical analyses that we have applied to the pollen/spore se- ciate important peaks of Ambrosia with air-mass trajectories coming ries obtained in Barcelona during the period 2001–2019 show that from North-eastern and Southern Europe, entering Catalonia through synoptic meteorology may be a relevant factor in the measured concen- the northern Mediterranean region (Fernández-Llamazares et al., trations of some pollen/spores. In addition, and although without a di- 2012). In the case of Fagus and Betula, episodes of long-range transport rect cause-effect relationship, it can also influence the flowering from Central Europe were identified, showing back-trajectories timing. The following points can be highlighted: bypassing the Pyrenees through from its easternmost region (Alberes) • The synoptic type SMP-1, which occurred mainly in winter, coincided or crossing them through the valley of the Garonne river (Belmonte with the highest concentrations of Cupressaceae and Platanus pollen. et al., 2008; Izquierdo et al., 2011, 2017; Alarcón et al., 2022). The syn- SMP-1 was the less frequent of the patterns (4 %) and it was associated optic situations corresponding to the Ambrosia episodes were character- to atmospheric blockage on the Iberian Peninsula and situations of ized by the presence of low pressures over the IP extending towards the great atmospheric stability, favouring low temperatures, solar radiation Eastern Mediterranean, and high-pressure systems over Scotland and and humidity and northwesterly winds in Barcelona. Instead, SMP-1 Eastern Europe. This scenario would correspond to the SMP-4 of the was also linked to the lowest values of the other pollen/spores. In partic- present study. In the case of Betula and Fagus, the synoptic situation ular, the pollen types Poaceae, Amaranthaceae and Olea are absent when showed high-pressure systems over the British Islands and the presence this synoptic pattern occurs, partly due to the phenology of the species of low pressures over Eastern Mediterranean. In the present study these contained in these taxa. situations would correspond to types SMP-4 and SMP-6. However, these • The synoptic situation, SMP-3, with the anticyclone over the Azores and two synoptic types, SMP-4 and SMP-6, have turned out to be the least in- the Atlantic low displaced to the north of the UK, which characterizes fluential in the present study that deals with taxa predominantly of local the weather in Barcelona with low pressures, low temperatures and origin. This could indicate that the local meteorology linked to each syn- solar radiation, and moderate winds from the west and southwest, was ac- optic type is the most determining factor, rather than its associated companied by high concentrations of Cupressaceae pollen, which is a tree transport pattern. Indeed, relative humidity and temperature are two that pollinates throughout the year although mainly in winter. variables that influence the release of pollen into the atmosphere: low • The most frequent synoptic type, especially in spring and summer, SMP-2, relative humidity combined with high temperature favours the increase produced sporadic episodes with concentrations considered to be of high of airborne pollen. The duration of the insolation also influences the re- risk of allergy of the herbs Urticaceae and Poaceae, and Alternaria fungal lease of pollen and the wind speed is a determining factor in its transport spore. Also, Amaranthaceae showed the highest percentage of days with and dispersal (Alba et al., 2000). The effect of rain simultaneous to pol- concentrations of allergy risk under this scenario. lination has mainly a washing effect by reducing the amount of pollen • SMP-1 was the most likely to initiate the Urticaceae pollination (Start) as grains (Rodríguez-Solà et al., 2022). In another study we analysed the well as to provide the date of the Peak for Platanus. dynamics of the hourly concentrations during the episode correspond- • The scenarios that favoured the End of pollination are SMP-4 for ing to the absolute maximum of Total pollen recorded in Barcelona in Cupresaceae, SMP-2 for Amaranthaceae, and SMP-6 for Platanus. 2015, in which 80 % of its total concentration was due to Platanus and • The Alternaria spore was present in all synoptic types, although in higher Pinus pollen (Sicard et al., 2016, 2021). The analysis of the samples in- concentrations during SMP-2 episodes. dicated that the correlation coefficient between the daily mean temper- ature and the daily total pollen concentration was 0.95, indicating a strong dependence of pollen release upon temperature. The correlation Knowledge of the relationship between synoptic meteorology and air- coefficient between the daily mean relative humidity and the daily Total borne pollen/spore dynamics is useful to anticipate situations that may pollen concentration was negative (−0.18) but much lower than the compromise health of the population and for the adoption of prevention one for temperature. The correlation coefficient between daily wind measures by the competent authorities. speed and Total pollen concentration was 0.82, indicating a strong de- Supplementary data to this article can be found online at https://doi. pendence of pollen release also upon wind speed. The synoptic chart org/10.1016/j.scitotenv.2023.165337. 16 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 CRediT authorship contribution statement European Union Luxembourg (doi:10.2760/439106, JRC117306. ISBN 978-92-76- 09001-4). Belmonte, J., Vendrell, M., Roure, J.M., Vidal, J., Botey, J., Cadahía, A., 2000. Levels of Am- Marta Alarcón: methodology, original draft preparation, writing- brosia pollen in the atmospheric spectra of Catalan aerobiological stations. Aerobiologia reviewing. 16, 93–99. https://doi.org/10.1023/A:1007649427549. Belmonte, J., Alarcón, M., Avila, A., Scialabba, E., Pino, D., 2008. Long-range transport of Raul Rodríguez-Solà: methodology, figures, writing- reviewing. beech (Fagus sylvatica L.) pollen to Catalonia (north-eastern Spain). Int. J. Biometeorol. Maria del Carme Casas-Castillo: methodology, figures, writing- 52, 675–687. https://doi.org/10.1007/s00484-008-0160-9. reviewing. Bousquet, J., Heinzerling, L., Bachert, C., Papadopoulos, N.G., Bousquet, P.J., Burney, P.G., Canonica, G.W., Carlsen, K.H., Cox, L., Haahtela, T., et al., 2012. Practical guide to skin Francisco Molero: synoptic meteorological patterns, writing- reviewing. prick tests in allergy to aeroallergens. Allergy 67, 18–24. https://doi.org/10.1111/j. Pedro Salvador: synoptic meteorological patterns, writing- reviewing. 1398-9995.2011.02728.x. Cristina Periago: writing- reviewing. Caillaud, D., Toloba, Y., Raobison, R., Besancenot, J.P., Thibaudon, M., Martinc, S., Segala, C., 2014. Impact sanitaire des pollens: revue des études épidémiologiques. Rev. Mal. Respir. Jordina Belmonte: pollen data, aerobiology, writing- reviewing. 31, 142–149. https://doi.org/10.1016/j.rmr.2013.09.017. Cakmak, S., Dales, R.E., Frances Coates, F., 2012. Does air pollution increase the effect of Data availability aeroallergens on hospitalization for asthma? J. Allergy Clin. Immunol. 129, 228–231.. https://www.researchgate.net/publication/51755064. Campoy, J.A., Ruiz, D., Egea, J., 2011. Dormancy in temperate fruit trees in a global warming Data will be made available on request. context: a review. Sci. Hortic. 130, 357–372. https://doi.org/10.1016/j.scienta.2011.07. 011. Cariñanos, P., Ruiz-Peñuela, S., Valle, A.M., de la Guardia, C.D., 2020. Assessing pollination Declaration of competing interest disservices of urban street-trees: the case of London-plane tree (Platanus x hispanica Mill. ex Münchh). Sci. Total Environ. 737 (139722). https://doi.org/10.1016/j. The authors declare that they have no known competing financial inter- scitotenv.2020.139722. Clavero, P., Martín-Vide, J., Raso, J., 1996. Atles climàtic de catalunya: Termoplu-viometria. ests or personal relationships that could have appeared to influence the Departament de Medi Ambient, Generalitat de Catalunya, pp. 1–42. work reported in this paper. Docampo, S., Recio, M., Trigo, M.M., Melgar, M., Cabezudo, B., 2007. Risk of pollen allergy in Nerja (southern Spain): a pollen calendar. Aerobiologia 23, 189–199. https://doi.org/10. 1007/s10453-007-9063-1. Acknowledgements Escudero, M., Castillo, S., Querol, X., Avila, A., Alarcón, M., Viana, M.M., Alastuey, A., Cuevas, E., Rodríguez, S., 2005. Wet and dry African dust episodes over eastern Spain. J. Geophys. We acknowledge the financial support of the European Commission for Res. Atmos. 110, D18S08. https://doi.org/10.1029/2004JD004731. ENV4-CT98-0755; the Spanish Government for AMB97-0457-CO7-021, Fernández-Llamazares, A., Belmonte, J., Alarcón, M., López-Pacheco, M., 2012. Ambrosia L. in Catalonia (NE Spain): expansion and aerobiology of a new bioinvader. Aerobiologia REN2001-10659-CO3-01, CGL2004-21166-E, CGL2005-07543/CLI, 28, 435–451. https://doi.org/10.1007/s10453-012-9247-1. CGL2009-11205, CGL2012-39523-C02-01/CLI, CGL2012-39523-C02-01, Galán Soldevilla, C., Cariñanos González, P., Alcázar Teno, P., Domínguez Vilches, E., 2007. CGL2012-39523-C02-02, CGL2016-75996-R, CTM2017-89565-C2-1-P, Manual de Calidad y Gestión de la Red Española de Aerobiología. Servicio de Publicaciones. Universidad de Córdoba. CTM2017-89565-C2-2, grant PID2020-117873RB-I00 funded by MCIN/ Galán, I., Prieto, A., Rubio, M., Herrero, T., Cervigón, P., Cantero, J.L., Gurbindo, M.D., AEI/ 10.13039/501100011033 and CONSOLIDER GRACCIE; the Catalan Martínez, M.I., Tobías, A., 2010. Thorax 2010 (65), 398–402. https://doi.org/10.1136/ Government agency AGAUR for 2002SGR00059, 2005SGR00519, thx.2009.118992. Galán, C., Smith, M., Thibaudon, M., Frenguelli, G., Oteros, J., Gehrig, R., et al., 2014. Pollen 2009SGR1102, and 2014SGR1274; Sociedad Española de Alergología e monitoring: minimum requirements and reproducibility of analysis. Aerobiologia 30, Inmunología Clínica (SEAIC); Laboratorios LETI PHARMA; Societat 385–395. https://doi.org/10.1007/s10453-014-9335-5. Catalana d'Al·lèrgia i Immunologia Clínica (SCAIC); Stallergenes Ibérica; Galán, C., Ariatti, A., Bonini, M., et al., 2017. Recommended terminology for aerobiological studies. Aerobiologia 33, 293–295. https://doi.org/10.1007/s10453-017-9496-0. S.A.; J Uriach y Compañia; S.A. This research contributes to the “María de Gehrig, R., Clot, B., 2021. 50 years of pollen monitoring in Basel (Switzerland) demonstrate Maeztu” Programme for Units of Excellence of the Spanish Ministry of the influence of climate change on airborne pollen. Front. Allergy 2. https://doi.org/ Science and Innovation (CEX2019-000940-M). The authors gratefully ac- 10.3389/falgy.2021.677159. Grewling, Ł., Bogawski, P., Kryza, M., Magyar, D., Šikoparija, B., Skjøth, C.A., Udvardy, O., knowledge the Servei Meteorològic de Catalunya the data used in this pub- Werner, M., Smith, M., 2019. Concomitant occurrence of anthropogenic air pollutants, lication. mineral dust and fungal spores during long-distance transport of ragweed pollen. Envi- ron. Pollut. 254, 112948. https://doi.org/10.1016/j.envpol.2019.07.116. References Hirst, J.M., 1952. An automatic volumetric spore trap. Ann. Appl. Biol. 39, 257–265. https:// doi.org/10.1111/j.1744-7348.1952.tb00904.x. Izquierdo, R., Belmonte, J., Àvila, A., Alarcón, M., Cuevas, E., Alonso-Pérez, S., 2011. Source Adams-Groom, B., Skjøth, C.A., Baker, M., Welch, T.E., 2017. Modelled and observed surface areas and long-range transport of pollen from continental land to Tenerife (Canary soil pollen deposition distance curves for isolated trees of Carpinus betulus, Cedrus Islands). Int. J. Biometeorol. 55, 67–85. https://doi.org/10.1007/s00484-010-0309-1. atlantica, Juglans nigra and Platanus acerifolia. Aerobiologia 33, 407–416. https://doi. Izquierdo, R., Alarcón, M., Mazón, J., Pino, D., De Linares, C., Aguinagalde, X., 2017. Are the org/10.1007/s10453-017-9479-1. Pyrenees a barrier for the transport of birch (Betula) pollen from Central Europe to the Aerts, R., Bruffaerts, N., Somers, B., Demoury, C., Plusquin, M., Nawrot, T.S., Hendrickx, M., Iberian Peninsula? Sci. Total Environ. 575, 1183–1196. https://doi.org/10.1016/j. 2021. Tree pollen allergy risks and changes across scenarios in urban green spaces in scitotenv.2016.09.192. Brussels, Belgium. Landsc. Urban Plan. 207, 104001. https://doi.org/10.1016/j. Kalnay, E., Kanamitsu, M., Kistler, R., et al., 1996. The NCEP/NCAR 40-year reanalysis pro- landurbplan.2020.104001. ject. Bull. Am. Meteorol. Soc. 77, 437–471. https://doi.org/10.1175/1520-0477(1996) Alarcón, M., Periago, C., Pino, D., Mazón, J., Casas-Castillo, M.C., Ho-Zhang, J.J., De Linares, 077%3C0437:TNYRP%3E2.0.CO;2. C., Rodríguez-Solà, R., Belmonte, J., 2022. Potential contribution of distant sources to air- Kirchner, R., Casas-Castillo, M.C., Rodríguez-Solà, R., Alarcón, M., Periago, C., De Linares, C., borne Betula pollen levels in northeastern Iberian Peninsula. Sci. Total Environ. 818. Belmonte, J., 2022. Comparative study between the effects of autumn and winter rainfall https://doi.org/10.1016/j.scitotenv.2021.151827. on aerobiological variables in the NE of the Iberian Peninsula. In: Mensink, C., Jorba, O. Alba, F., Díaz de la Guardia, C., Comtois, P., 2000. The effect of meteorological parameters on (Eds.), Air Pollution Modeling and its Application XXVIII. ITM 2021. Springer Proceed- diurnal patterns of airborne olive pollen concentration. Grana 39, 200–208. https://doi. ings in Complexity. Springer, Cham. https://doi.org/10.1007/978-3-031-12786-1_36. org/10.1080/00173130051084340. Lana, X., Rodríguez-Solà, R., Martínez, M.D., Casas-Castillo, M.C., Serra, C., Kirchner, R., Anderegg, W.R.L., Abatzoglou, J.T., Anderegg, L.D.L., Bielory, L., Kinney, P.L., Ziska, L., 2021. 2020. Multifractal structure of the monthly rainfall regime in Catalonia (NE Spain): eval- Anthropogenic climate change is worsening North American pollen seasons. PNAS 118. uation of the non-linear structural complexity. Chaos Int. J. Nonlinear Sci. 30 (7), 3117. https://doi.org/10.1073/pnas.2013284118. https://doi.org/10.1063/5.0010342. Andersen, T.B., 1991. A model to predict the beginning of the pollen season. Grana 30 (1), Lana, X., Casas-Castillo, M.C., Rodríguez-Solà, R., Serra, C., Martínez, M.D., Kirchner, R., 269–275. https://doi.org/10.1080/00173139109427810. 2021. Rainfall regime trends at annual and monthly scales in Catalonia (NE Spain) and Avila, A., Alarcón, M., 1999. Relationship between precipitation chemistry and meteorologi- indications of CO2 emissions effects. Theor. Appl. Climatol. 146 (3–4), 981–996. cal situations at a rural site in NE Spain. Atmos. Environ. 33, 1663–1677. https://doi.org/ https://doi.org/10.1007/s00704-021-03773-z. 10.1016/S1352-2310(98)00341-0. Llabrés-Brustenga, A., Rius, A., Rodríguez-Solà, R., Casas-Castillo, M.C., 2020. Influence of re- Àvila, A., Alarcón, M., Castillo, S., Escudero, M., Garcıía Orellana, J., Masqué, P., Querol, X., gional and seasonal rainfall patterns on the ratio between fixed and unrestricted mea- 2007. J. Geophys. Res. 112, D05210. https://doi.org/10.1029/2006JD007153. sured intervals of rainfall amounts. Theor. Appl. Climatol. 140 (1), 389–399. https:// Belis, C.A., Favez, O., Mircea, M., Diapouli, E., Manousakas, M.-I., Vratolis, S., Gilardoni, S., doi.org/10.1007/s00704-020-03091-w. Paglione, M., Decesari, S., Mocnik, G., Mooibroek, D., Salvador, P., Takahama, S., López-Orozco, R., Hernández-Ceballos, M.A., Galán, C., García-Mozo, H., 2020. Atmospheric Vecchi, R., Paatero, P., 2019. European guide on air pollution source apportionment pathways and distance range analysis of castanea pollen transport in southern Spain. For- with receptor models - revised version 2019. EUR 29816 EN publications Office of the ests 11, 1092. https://doi.org/10.3390/f11101092. 17 M. Alarcón et al. Science of the Total Environment 896 (2023) 165337 Luedeling, E., Guo, L., Dai, J., Leslie, C., Blanke, M.M., 2013. Differential responses of trees to Rojo, J., Rapp, A., Lara, B., Fernández-González, F., Pérez-Badia, R., 2015. Effect of land uses temperature variation during the chilling and forcing phases. J. Agric. Meteorol. 181, andwind direction on the contribution of local sources to airborne pollen. Sci. Total Envi- 33–42. https://doi.org/10.1016/j.agrformet.2013.06.018. ron. 538, 672–682. https://doi.org/10.1016/j.scitotenv.2015.08.074. Majeed, H.T., Periago, C., Alarcón, M., Belmonte, J., 2018. Airborne pollen parameters and Russo, A., Trigo, R.M., Martins, H., Mendes, M.T., 2014. NO2, PM10 and O3 urban concentra- their relationship with meteorological variables in NE Iberian Peninsula. Aerobiologia tions and its association with circulation weather types in Portugal. Atmos. Environ. 89, 34, 375–388. https://doi.org/10.1007/s10453-018-9520-z. 768–785. https://doi.org/10.1016/j.atmosenv.2014.02.010. Mohammed, A.J., Alarcón, M., Pino, D., 2018. Extreme temperature events on the Iberian Russo, A., Sousa, P.M., Durão, R.M., Ramos, A.M., Salvador, P., Linares, C., Díaz, J., Trigo, Peninsula: statistical trajectory analysis and synoptic patterns. Int. J. Climatol. 38, R.M., 2020. Saharan dust intrusions in the Iberian Peninsula: predominant synoptic con- 5305–5322. https://doi.org/10.1002/joc.5733. ditions. Sci. Total Environ. 717. https://doi.org/10.1016/j.scitotenv.2020.137041. Ojeda, P., Sastre, J., Olaguibel, J.M., Chivato, T., 2018. Alergológica 2015: a national survey Salvador, P., Alonso-Pérez, S., Pey, J., Artíñano, B., de Bustos, J.J., Alastuey, A., Querol, X., on allergic diseases in the adult Spanish population. J. Investig. Allergol. Clin. Immunol. 2014. African dust outbreaks over the western Mediterranean Basin: 11-year characteri- 28, 151–164. https://doi.org/10.18176/jiaci.0264. zation of atmospheric circulation patterns and dust source areas. Atmos. Chem. Phys. 14, Ojrzynska, H., Bilinska, D., Werner, M., Kryza, M., Malkiewicz, M., 2020. The influence of at- 6759–6775. https://doi.org/10.5194/acp-14-6759-2014. mospheric circulation conditions on Betula and Alnus pollen concentrations in Wrocław, Salvador, P., Barreiro, M., Gómez-Moreno, F.J., Alonso-Blanco, E., Artíñano, B., 2021. Synop- Poland. Aerobiologia 36, 261–276. https://doi.org/10.1007/s10453-020-09629-9. tic classification of meteorological patterns and their impact on air pollution episodes and Otero, N., Jurado, O.E., Tim Butler, T., Henning, W., Rust, H.W., 2022. The impact of atmo- new particle formation processes in a south European air basin. Atmos. Environ. 245. spheric blocking on the compounding effect of ozone pollution and temperature: a https://doi.org/10.1016/j.atmosenv.2020.118016. copula-based approach. Atmos. Chem. Phys. 22 (1905–1919), 2022. https://doi.org/ Sauliene, I., Sukiene, L., Kainov, D., Greiciuviene, J., 2015. The impact of pollen load on qual- 10.5194/acp-22-1905-2022. ity of life: a questionnairebased study in Lithuania. Aerobiologia 32 (2). https://doi.org/ Pace, L., Incerto, A., Curci, G. (Unpublished). Synoptic and Local Meteorology Influence on 10.1007/s10453-015-9387-1. Pollen Concentration in a Mountain Valley in Central Italy: Hints for Localization of Sicard, M., Izquierdo, R., Alarcón, M., Belmonte, J., Comerón, A., Baldasano, J.M., 2016. Main Sources, 31 August 2022, PREPRINT Research Square. https://doi.org/10.21203/ Near-surface and columnar measurements with a micro pulse lidar of atmospheric pollen rs.3.rs-2004496/v1. in Barcelona, Spain. Atmos. Chem. Phys. 16, 6805–6821. https://doi.org/10.5194/acp- Paschalidou, A.K., Psistaki, K., Charalampopoulos, A., Vokou, D., Kassomenos, P., Damialis, 16-6805-2016. A., 2020. Identifying patterns of airborne pollen distribution using a synoptic climatology Sicard, M., Jorba, O., Ho, J.J., Izquierdo, R., De Linares, C., Alarcón, M., Comerón, A., approach. Sci. Total Environ. 714, 136625. https://doi.org/10.1016/j.scitotenv.2020. Baldasano, J.M., Belmonte, J., 2021. Measurement report: characterization of the vertical 136625. distribution of airborne Pinus pollen in the atmosphere with lidar-derived profiles – a Pérez-Zanón, N., Casas-Castillo, M.C., Peña, J.C., Aran, M., Rodríguez-Solà, R., Redaño, À., modeling case study in the region of Barcelona, NE Spain. Atmos. Chem. Phys. 21, Sole, G., 2018. Analysis of synoptic patterns in relationship with severe rainfall events 17807–17832. https://doi.org/10.5194/acp-21-17807-2021. in the Ebre Observatory (Catalonia). Acta Geophys. 66, 405–414. https://doi.org/10. Sofiev, M., Siljamo, P., Ranta, H., Rantio-Lehtimaki, A., 2006. Towards numerical forecasting 1007/s11600-018-0126-1. of long-range air transport of birch pollen: theoretical considerations and a feasibility Picornell, A., Rojo, J., Trigo, M.M., Ruiz-Mata, R., Lara, B., Romero-Morte, J., Serrano-García, study. Int. J. Biometeorol. 50, 392–402. https://doi.org/10.1007/s00484-006-0027-x. A., Pérez-Badía, R., Gutiérrez-Bustillo, M., Cervigón-Morales, P., Ferencova, Z., Morales- Sofiev, M., Belmonte, J., Gehrig, R., Izquierdo, R., Smith, M., Dahl, A., Siljamo, P., 2013. A re- González, J., Sánchez-Reyes, E., Fuentes-Antón, S., Sánchez-Sánchez, J., Dávila, I., view of production, release, distribution and health impact of allergenic pollen. In: Oteros, J., Martínez-Bracero, M., Galán, C., García-Mozo, H., Alcázar, P., Fernández, S., Bergman, K.-C., Sofiev, M. (Eds.), Airborne Pollen Transport. Springer, pp. 127–159. González-Alonso, M., Robles, E., de Zabalza, A.P., Ariño, A.H., Recio, M., 2022. Environ- https://doi.org/10.1007/978-94-007-4881-1_5. mental drivers of the seasonal exposure to airborne Alternaria spores in Spain. Sci. Total Tormo Molina, R., Muñoz Rodríguez, A., Silva Palacios, I., Gallardo López, F., 1996. Pollen Environ., 153596. https://doi.org/10.1016/j.scitotenv.2022.153596. production in anemophilous trees. Grana 35, 38–46. https://doi.org/10.1080/ Puiggròs, A., Muñoz-Cano, R., Roger, Reig A., Raga, E., Belmonte, J., Valero, A., 2015. Prev- 00173139609430499. alence of sensitization to pollen from trees planted in Barcelona city. J Investig Allergol Veriankaitė, L., Šaulienė, I., Bukantis, A., 2012. Evaluation of meteorological parameters in- Clin Immunol 25, 150–151. fluence upon pollen spread in the atmosphere. J. Environ. Eng. Landsc. Manag. 19, Riediker, M., Monn, C., Koller, T., Stahel, W., Wüthrich, B., 2001. Air pollutants enhance 5–11. https://doi.org/10.3846/16486897.2011.557252. rhinoconjunctivitis symptoms in pollen-allergic individuals. Ann. Allergy Asthma Vrinceanu, D., Berghi, O.N., Cergan, R., Dumitru, M., Ciuluvica, R.C., Giurcaneanu, C., Immunol. 87, 311–318. https://www.researchgate.net/publication/11667587. Neagos, A., 2021. Urban allergy review: allergic rhinitis and asthma with plane tree sen- Rodríguez-Solà, R., Casas-Castillo, M.C., Ho-Zhang, J.J., Kirchner, R., Alarcón, M., Periago, C., sitization (review). Exp. Ther. Med. 21. https://doi.org/10.3892/etm.2021.9706. De Linares, C., Belmonte, J., 2022. A study on correlations between precipitation ETCCDI Ziello, C., Sparks, T.H., Estrella, N., Belmonte, J., Bergmann, K.C., et al., 2012. Changes to air- and airborne pollen/fungal spore parameters in the NE Iberian Peninsula. Int. borne pollen counts across Europe. PLoS One 7 (4), e34076. https://doi.org/10.1371/ J. Biometeorol. 66, 1173–1187. https://doi.org/10.1007/s00484-022-02267-5. journal.pone.0034076. Rogers, C.A., Wayne, P.M., Macklin, E.A., Muilenberg, M.L., Wagner, C.J., et al., 2006. Inter- action of the onset of spring and elevated atmospheric CO2 on ragweed. (Ambrosia artemisiifolia L.) pollen production. Environ. Health Perspect. 114, 865–869. https:// doi.org/10.1289/ehp.8549. 18