New method for establishing a network of operational warning of Septoria leaf blotch disease in winter wheat

A mechanistic model, PROCULTURE, based on commonly available
meteorological data and assessing in real time the risk of progression of
septoria leaf blotch disease on winter wheat has been developed in Belgium
and the Grand-Duchy of Luxemburg (GDL) to limit fungicide use. However,
the reliability of meteorological stations used for the warning system varies
according to the distance to the fields. A weather analysis based on the Fourier
transform highlighted a great difference in the intraday variation between two
sites in the GDL (Everlange and Reuland). The correlation between these two
sites is very high for the hourly temperature (R = 0.96), and for the hourly
relative humidity (RH) (R = 0.86), (P < 0.05). However, the intraday variation
(<11 hours) highlights contrasts for a given meteorological parameter. Hence,
the correlation between temperature or RH decreased respectively from 0.96
to 0.43 and from 0.86 to 0.30. The comparison between infection conditions
given by PROCULTURE using the Fourier transform, shows: (i) a positive
but weak correlation between temperature at Reuland and Everlange (R =
0.64), (ii) a good correlation between RH for these two sites (R = 0.86), and
(iii) a contrasted difference for rain (R = 0.27), (P < 0.05). This Fourier
transform based method enables to take into account the RH and temperature
variation related to topography levels in the warning system and to understand
and explain the variation in disease expression between a plateau and a valley
bottom or between North and South slopes.