Department of Environmental Engineering, Engineering Faculty, Universidad Tecnológica de Bolívar, Cartagena, Colombia.
Institut de Ciència i Tecnologia Ambientals (ICTA), Universitat Autònoma de Barcelona, Barcelona, Spain.
Int J Biometeorol. 2019 Dec;63(12):1541-1553. doi: 10.1007/s00484-019-01767-1. Epub 2019 Aug 3.
Alternaria and Cladosporium are the most common airborne fungal spores responsible for health problems, as well as for crop pathologies. The study of their behavior in the air is a necessary step for establishing control and prevention measures. The aim of this paper is to develop a logistic regression model for predicting the daily concentrations of airborne Alternaria and Cladosporium fungal spores from meteorological variables. To perform the logistic regression analysis, the concentration levels are binarized using concentration thresholds. The fungal spore data have been obtained at eight aerobiological monitoring stations of the Aerobiological Network of Catalonia (NE Spain). The meteorological data used were the maximum and minimum daily temperatures and daily rainfall provided by the meteorological services. The relationship between the meteorological variables and the fungal spore levels has been modeled by means of logistic regression equations, using data from the period 1995-2012. Values from years 2013-2014 were used for validation. In the case of Alternaria, three equations for predicting the presence and the exceedance of the thresholds 10 and 30 spores/m have been established. For Cladosporium, four equations for the thresholds 200, 500, 1000, and 1500 spores/m have been established. The temperature and cumulative rainfall in the last 3 days showed a positive correlation with airborne fungal spore levels, while the rain on the same day had a negative correlation. Sensitivity and specificity were calculated to measure the predictive power of the model, showing a reasonable percentage of correct predictions (ranging from 48 to 99%). The simple equations proposed allow us to forecast the levels of fungal spores that will be in the air the next day, using only the maximum and minimum temperatures and rainfall values provided by weather forecasting services.
交链孢霉和链格孢霉是引起健康问题和作物病害的最常见空气传播真菌孢子。研究它们在空气中的行为是制定控制和预防措施的必要步骤。本文的目的是建立一个逻辑回归模型，用于预测气象变量对空气中交链孢霉和链格孢霉真菌孢子日浓度的影响。为了进行逻辑回归分析，使用浓度阈值将浓度水平二值化。真菌孢子数据是在西班牙东北部加泰罗尼亚空气生物学监测网络的 8 个空气生物学监测站获得的。使用的气象数据是气象服务提供的每日最高和最低温度以及日降雨量。通过逻辑回归方程对气象变量与真菌孢子水平之间的关系进行建模，使用的是 1995-2012 年的数据。2013-2014 年的数据用于验证。在交链孢霉的情况下，建立了三个预测存在和超过阈值 10 和 30 孢子/m 的方程。对于链格孢霉，建立了四个预测阈值 200、500、1000 和 1500 孢子/m 的方程。最后 3 天的温度和累积降雨量与空气中的真菌孢子水平呈正相关，而当天的降雨量则呈负相关。计算了灵敏度和特异性来衡量模型的预测能力，显示出合理的正确预测百分比（范围从 48%到 99%）。所提出的简单方程可以使用气象预报服务提供的最高和最低温度以及降雨量值来预测第二天空气中的真菌孢子水平。
