1 Introduction

Alternaria is certainly one of the most allergenic spores, causing respiratory diseases in humans (D’Amato et al. 1997; Gioulekas et al. 2004). In addition, Alternaria can be the cause of different plant diseases, with negative effects on agriculture (fruit trees, potatoes and tomatoes among the others) and on natural vegetation (Harteveld et al. 2014; Gur et al. 2017; Escuredo et al. 2011; Jambhulkar et al. 2016).

The aim of the study is to describe the major features of the current distribution of Alternaria spores in the Emilia-Romagna Region, located in the southern part of the Po Valley, in Northern Italy. The aerobiological network managed by the Regional Environmental Agency in Emilia-Romagna (Arpae-ER) is dedicating a substantial effort to the monitoring of some fungal spores since the end of the 1990s.

Daily concentration data of Alternaria spores are available from Arpae-ER aerobiological network, and the monitoring activity dates back to the end of the 1990s in 10 stations of the network (PC1: Piacenza; PR2: Parma; RE1: Reggio Emilia; MO1: Modena; BO1: Bologna; FE1: Ferrara; RA3: Ravenna; FO1: Forlì; FO2: Cesena; RN1: Rimini): the availability of such data allows to set up a detailed analysis of the presence of Alternaria spores in different climatic and environmental settlements inside Emilia-Romagna territory, following aerobiological and phytopathological aspects related to the presence of these spores.

Another focus of this study is related to a preliminary analysis of long-term trends for Alternaria spores, which can effectively been supported by the 19-year database (1999–2017) of daily spores concentration in most of the stations. To our knowledge, only few studies have been focused on the analysis of long-term trends for Alternaria, as well as for other fungal spores. Decreasing trends for Alternaria concentration are shown in Corden et al. (2003) for two UK cities and in Grinn-Gofron et al. (2016) for two Polish cities. Also in Damialis et al. (2015), where 14 fungal taxa are examined in a city in Greece, decreasing concentration is found for 11 taxa; in addition, a later onset of the spore season together with a shortening of the season length is found, although only few significant trends are present.

2 Methods

Daily concentration of Alternaria spores collected in the 10 monitoring stations belonging to Arpae-ER aerobiological network has been used in order to compute the average daily concentration of Alternaria in each location and to define some seasonal indicators to synthesize the main characteristics of the spore season.

The selected indicators are: the annual spore integral, which is obtained summing the daily concentration of Alternaria spores during the whole year; the peak value, which is the highest daily concentration during the year; the peak date, which is the day of the year when the peak value occurs; the starting and ending date of the spore season, which represent the day when the 5% (start) and 95% (end) of the annual spore integral are reached; finally, the number of days between these two calendar dates is the length of the spore season.

Time series of the previous indicators are submitted to different statistical tests. First of all, the annual median of the time series for each indicator in the period 1999–2017 is computed, in order to point out the average (climatic) conditions in each monitoring station. A geographical analysis focused on the distribution of the median of the seasonal indicators is performed. Such analysis is meant to point out the geographical coherence of Alternaria seasonal characteristics throughout Emilia-Romagna territory, according to the statistical relationship between the indicators and the geographical coordinates of the stations (latitude and longitude).

The statistical index used to outline the geographical analysis is the Moran’s index (Moran 1950), which measures the degree of spatial autocorrelation.

In addition, a trend analysis of the 1999–2017 annual time series is carried out computing the nonparametric Kendall rank correlation coefficient τ (Kendall 1976) in order to evaluate the monotonic increasing/decreasing trend in the time series (together with its statistical significance level, p level). In addition, the Theil–Sen estimator is computed in order to quantify the unit (annual) variation in each station over time (Theil 1950; Sen 1968).

The median of the distribution of Kendall τ coefficients in the 10 monitoring stations is computed for each seasonal indicator, and the Mann–Whitney test (1947) is used in order to check whether the median values are different from zero, being then a reasonable approximation of the presence of a prevalent trend (positive or negative) for the specific indicator.

All the statistical analyses have been carried out using R software (R Core Team 2018).

3 Results

Alternaria spores are present in the atmosphere from May to November, and null concentrations appear only in winter (Fig. 1). During summer, daily concentration of Alternaria spores shows two distinct peaks almost in all the stations: the first is in June and the second in September. On average, the latter peak is more pronounced at least in the stations located in the western and central Po Valley.

Fig. 1
figure 1

Daily average concentration of Alternaria spores (expressed in spores/m3) in the monitoring station of Arpae-ER network

Full size image

The median values of the Alternaria season indicators in each station of the Arpae-ER aerobiological network are given in Table 1; medians are computed in the 1999–2017 period. As for the magnitude of the peak, highest values (above 1000 spore/m3) are observed in the westernmost stations (PC1, PR2 and RE1) and in FE1.

Table 1 Median of the distribution of spore season indicators in each station of the Arpae-ER aerobiological network for the period 1999–2017
Full size table

Alternaria season indicators show high level of spatial autocorrelation: Moran’s indices associated with all indicators (except for ending date) are statistically significant (p level < 0.01 for starting and peak date of the season, for spore integral and peak value; p level < 0.05 for the season length).

The Kendall τ coefficients and the Theil–Sen estimator of the time series of each seasonal indicator for Alternaria spores in the period 1999–2017 are given in Table 2.

Table 2 (a) Kendall τ coefficient; (b) Theil–Sen estimator in each station for the different indicators of Alternaria season for the period 1999–2017
Full size table

Stations of PC1, FE1 and RA3 show the most relevant signals, the first two stations for spore season indicators and the latter for spore production indicators. In general, only few coefficients reaching statistical significance are present in Table 2a; nevertheless, many stations are characterized by relevant trends for spore season indicators, with advancing starting dates (in eight stations; two of them are statistically significant), delaying ending dates and lengthening of seasonal duration as a consequence (in eight stations; four of them are significant, for both indicators).

Therefore, the overall situation in Emilia-Romagna shows that major changes in Alternaria season are mainly related to variations of the ending date and the length of the season, for which the median of Kendall τ coefficients distribution is significantly positive (p level < 0.01). In addition, the median is significantly negative (p level < 0.05) for the starting date of the season.

An increase of 1–2 day/year for the Alternaria season length and a delay of about 1 day/year of the ending date of the season can be assessed considering the values of Theil–Sen estimator (Table 2b). Also the advance in the starting date (about 0.7 day/year) contributes to the lengthening of the season.

As for spore production indicators, the median of Kendall τ coefficients is negative (− 0.22 for peak value, p level < 0.05, and − 0.23 for spore integral, not significant). However, the amount of spores is generally decreasing in most of the stations, with a reduction of about 30–40 spore m−3 per year for the peak value and of more than 400 spore per year (with a maximum of 1200 spore per year) for spore integral.

4 Conclusions

The main features of Alternaria spore season show a relevant geographical coherence in the Emilia-Romagna territory. The Alternaria seasonal indicators are characterized by a high level of spatial autocorrelation for spore production indicators as well as for the onset and peak of the spore season.

The trend analysis shows a general decrease in the amount of Alternaria spores, but statistical significance is reached only locally. In addition, a longer duration of the sporulation period (statistically significant in almost half of the stations) is also present in the analysis, with a simultaneous advance of the starting date and delay of the ending date. The latter signal is more pronounced, being statistically significant in a larger number of stations.

The resulting net effect of the previously outlined signals related to the presence of Alternaria in the Emilia-Romagna territory can be summarized as longer spore seasons, with delaying ending date and advancing onset of Alternaria presence with smaller spore amount, both in terms of the peak values and the spore integral.

The comparison of previous results with those emerging from the few other studies focusing on long-term trends for fungal spores in the European area shows remarkable similarities. The most important ones are related to spore season indicators, for which tendencies to earlier start and peak dates as well as to a longer season are in close agreement with the Polish study (Grinn-Gofron et al. 2016), in which the time window (2004–2013) is quite similar to that of present study (1999–2017). Results in Thessaloniki (Damialis et al. 2015) specific for Alternaria spores, which contribute to 10% of the whole fungal spore spectrum in the study period (1987–2005), do not show any relevant trend.

As for spore production, some differences emerge, since both Polish and UK studies (Corden et al. 2003) indicate a decrease in the spore amount (both annual integral and peak value) in the coastal stations (Szczecin and Cardiff, respectively) and an increase in the inland stations (Cracow and Derby, respectively). However, negative signs for the tendencies of spore production indicators are almost ubiquitous in Emilia-Romagna stations, thus supporting the generally decreasing trends for Alternaria spore amount in the area, even though the territory is certainly characterized by a relevant agricultural vocation, which some studies associated with the atmospheric diffusion of Alternaria spores.