Importance of Meteorological Parameters and Airborne Conidia to Predict Risk of Alternaria on a Potato Crop Ambient Using Machine Learning Algorithms

Abstract

Secondary infections of early blight during potato crop season are conditioned by aerial inoculum. However, although aerobiological studies have focused on understanding the key factors that influence the spore concentration in the air, less work has been carried out to predict when critical concentrations of conidia occur. Therefore, the goals of this study were to understand the key weather variables that affect the hourly and daily conidia dispersal of Alternaria solani and A. alternata in a potato field, and to use these weather factors in different machine learning (ML) algorithms to predict the daily conidia levels. This study showed that conidia per hour in a day is influenced by the weather conditions that characterize the hour, but not the hour of the day. Specifically, the relative humidity and solar radiation were the most relevant weather parameters influencing the conidia concentration in the air and both in a linear model explained 98% of the variation of this concentration per hour. Moreover, the dew point temperature three days before was the weather variable with the strongest effect on conidia per day. An improved prediction of Alternaria conidia level was achieved via ML algorithms when the conidia of previous days is considered in the analysis. Among the ML algorithms applied, the CART model with an accuracy of 86% were the best to predict daily conidia level.

Keywords: Alternaria spp.; Solanum tuberosum; aerobiology; decision trees; early blight; k-nearest neighbor; machine learning; random forest; weather factors.

Figure 1

The weather conditions during the cropping seasons (2017–2021). The variables of daily mean relative humidity (RH, %) and accumulated rain (mm) are represented in Y1, and daily mean temperature (%) is shown in Y2. Days after emergence were the number of days after 50% of emergence. The emergence dates were 16 May (2017), 1 June (2018), 4 June (2019), and 10 June (2020 and 2021) (Table S1).

Figure 2

Daily conidia concentrations and phenological stages (vegetative, flowering, and senescence) in different cropping seasons. The emergence dates were 16 May 2017, 1 June 2018, 4 June 2019, and 10 June (2020 and 2021) (Table S1).

Figure 3

A graphical model showing the conditional dependence of different hourly weather variables and hourly concentrations of A. solani and A. alternata represented by the conidia variable. Variables that are connected by a line are conditionally dependent or correlated significantly and vice versa. The weather variables were relative humidity (RH), temperature (Temp), wind speed (Wind), dew temperature (DewTemp), leaf wetness (LW), solar radiation (Rad), and rainfall (Rain). The probability that conidia will be released is identified by the variable spore release (SR) and the probability that conidia will escape from the canopy is named escape (Escape).

Figure 4

Heatmap with Spearman correlations between hourly conidia concentration and weather parameters. The selected weather variables were relative humidity (RH), temperature (Temp), wind speed (Wind), dew temperature (DewTemp), leaf wetness (LW), solar radiation (Rad), and rainfall (Rain).

Figure 5

A graphical model showing the conditional dependence of weather variables and conidia concentrations at each hour in the day. Variables that are connected by a line are conditionally dependent or correlated significantly and vice versa. The weather variables were relative humidity (RH), temperature (Temp), wind speed (Wind), dew point temperature (DewTemp), leaf wetness (LW), solar radiation (Rad), and rainfall (Rain). The probability that conidia will be released is identified by the variable spore release (SR), and the probability that conidia will escape from the canopy is named escape (Escape).

Figure 6

Linear regression graphs with confidence bounds of solar radiation (Radiation) (a) or Relative humidity (b) and conidia. The grey area surrounding the regression line represents the 95% confidence interval.

Figure 7

Intra-diurnal variations among Alternaria conidia and the most influenced weather parameters, namely solar radiation (Radiation) (a) and relative humidity (b). Each bar represents each hour of 24 h in a day.

Figure 8

Graphical model showing conditional dependence/relationships between different daily weather variables and conidia on different days. The weather variables were temperature (Temp), dew temperature (DewTemp), relative humidity (RH), wind speed (Wind), spore escape (Escape), spore release (SR), leaf wetness (LW), solar radiation (SR), and rainfall (Rain). The variable (weather and conidia) numbers 1, 2, 3, and 4 indicate measurements taken 1, 2, 3, and 4 days ago, whereas those without numbers were measured or recorded on the current day.

Figure 9

Heatmap with Spearman correlations between daily conidia concentration and weather parameters. The selected weather variables were relative humidity (RH), temperature (Temp), wind speed (Wind), dew temperature (DewTemp), leaf wetness (LW), solar radiation (Rad), and rainfall (Rain). The variable (weather and conidia) numbers 1, 2, 3, and 4 indicate measurements taken 1, 2, 3, and 4 days ago, whereas those without numbers were measured or recorded on the current day.