Advanced statistical models commonly applied in aerobiology cannot accurately predict the exposure of people to Ganoderma spore-relatedallergies
PBN-AR
Instytucja
Wydział Biologii (Uniwersytet im. Adama Mickiewicza w Poznaniu)
Informacje podstawowe
Główny język publikacji
en
Czasopismo
AGRICULTURAL AND FOREST METEOROLOGY
ISSN
0168-1923
EISSN
1873-2240
Wydawca
ELSEVIER SCIENCE BV
DOI
URL
Rok publikacji
2015
Numer zeszytu
Strony od-do
209-217
Numer tomu
 201
Liczba arkuszy
9,00
Słowa kluczowe
en
artificial neural network
basidiospore
ganoderma
inhalant allergy
multivariate regression tree
surface layer
Open access
Tryb otwartego dostępu
Otwarte czasopismo
Wersja tekstu w otwartym dostępie
Wersja opublikowana
Licencja otwartego dostępu
Inna
Czas opublikowania w otwartym dostępie
Razem z publikacją
Streszczenia
Język
en
Treść
The genus Ganoderma commonly comprises wood decay fungal species that actively produce a large number of spores liberated from basidiocarps thereby contributing significantly to fungal air spora worldwide. Basidiospores of Ganoderma incite disease of tree species, and constitute aeroallergens hazardous to people. Earlier forecasting models to predict high concentrations of Ganoderma spores pointed out the dew point temperature, maximum and average wind speed and precipitation as significant meteorological parameters that affect the basidiospore release. The main aim of this work was to study relationships between basidiospore counts and meteorological conditions and to verify whether regression models based on data collected at aerobiological studies accurately predict the risk of exposure of people with spore-related allergies. Basidiospores were captured over three autumn months from 2006 to 2008 using two Hirst-type volumetric spore traps with identical sampling protocols and evaluation methods. Daily spore concentrations were sampled and the dynamics of changes in Ganoderma spore concentrations sampled at heights used in standard aerobiological studies, located several meters above the ground level (a.g.l.), were compared with those recorded at the same site, but at the human respiration zone, much closer to the ground. Relationships between basidiospore concentrations and weather variables were investigated with the Spearman's rank correlation analysis. To reveal differences in meteorological parameters and Ganoderma spore content between consecutive years and months studied, the non-parametric Mann–Whitney U, Kruskal-Wallis and Dunn’s tests were applied. Furthermore Artificial Neural Networks and Multivariate Regression Trees were used, for which meteorological parameters were input variables, while Ganoderma basidiospore abundance was an output variable. Considerable differences were observed between Ganoderma spore concentrations at people’s respiratory zone and 18 a.g.l., a height used in standard aerobiological studies. At 1 m a.g.l. the concentrations were 1.2 to 6 times higher than those at 18 m a.g.l. Moreover, the dynamics of changes throughout spore trapping seasons were different; at 18 m a.g.l. they fluctuations were similar across all years sampled, whereas at 1 m a.g.l there were wide variations between years. The correlation between weather variables and concentrations of captured basidiospores at these levels was significant but rather low. The results questioned the usefulness of models based on spore samplings performed at several meters a.g.l. and suggested that the real numbers of basidiospores that are inhaled by people might depend on parameters that were as yet not included in the models.
Cechy publikacji
ORIGINAL_ARTICLE
Inne
System-identifier
550906
CrossrefMetadata from Crossref logo
Cytowania
Liczba prac cytujących tę pracę
Brak danych
Referencje
Liczba prac cytowanych przez tę pracę
Brak danych