Artificial neural network models of relationships betweenCladosporiumspores and meteorological factors in Szczecin (Poland)

Abstract

Moulds are common aeroallergens and Cladosporium is considered to be one of the more prevalent examples of these. Aiming to reduce the risk for allergic individuals, we constructed predictive models for the fungal spore circulation in Szczecin, Poland. Monthly forecasting models were developed for the airborne spore concentrations of Cladosporium, the most abundant fungal taxa in the area. Aerobiological sampling was conducted over 2004-2007, using a Lanzoni trap. Simultaneously, the following meteorological parameters were recorded: daily level of precipitation, maximum and average wind speed, relative humidity, maximum, minimum, average and dew point temperature. The original factors, as well as with lags (up to 3 days), were used as the explaining variables. Due to non-linearity and non-normality of the data set, Cladosporium spore concentrations were predicted using artificial neural networks (ANN). The final model included classification (spore presence or absence) and regression for spore seasons with log(x+1) transformed Cladosporium spore concentration. All variables, except average wind speed 3 days previously and precipitation (on the same day and with lags) were factors important in classification. In the regression model for spore seasons, close relationships were noted between Cladosporium spore concentration and dew point temperature the previous day, humidity, and minimum and maximum temperatures. Additionally, time series prediction was performed. Our study shows that ANN models are applicable to forecasting Cladosporium spore concentration with high accuracy.

Dew point temperature affects ascospore release of allergenic genus Leptosphaeria

“…(Grinn-Gofroń and Strzelczak 2008a ; Astray et al 2010 ; Kumar et al 2013 ), Cladosporium spp. (Grinn-Gofroń and Strzelczak 2008b ), and to some extent also for Ganoderma spp. (Kasprzyk et al 2011 ; Sadyś et al 2016 ).…”

Effects of meteorological factors on the composition of selected fungal spores in the air

“…Most of these studies were based on simple descriptive statistics, such as Pearson or Spearman’s correlation coefficients, or on multiple technique, such as the Duncan multiple range test and multiple regression model (Katial et al 1997 ; Angulo-Romero et al 1999 ; Mitakakis et al 2001 ; Troutt and Levetin 2001 ; Stennett and Beggs 2004 ). For the three types of spores ( Alternaria , Cladosporium and Ganoderma ), the predictive models were developed using advanced statistical methods like: artificial neural networks (ANN) and multivariate regression trees (MRT) (Grinn-Gofroń and Strzelczak 2008a , 2008b , 2009 , 2011 ). All these studies put emphasis on the statistical analysis of the correlation between the level of concentration of particular fungal spore types and weather parameters, however do not examine the complex composition of spores and its dependence on meteorological factors.…”

A fungal spore calendar for the atmosphere of Szczecin, Poland

“…In Szczecin, there have been few studies on atmospheric mycobionta. Aeromycological research in outdoor environments has been carried out in Szczecin only for selected allergenic taxa (Alternaria, Cladosporium, Ganoderma, Didymella) [1][2][3][4]. The main goal of this study was to create a spore calendar for 2013 and show the percentage of all identified fungal spores.…”