README

This R package, cercospoRa, takes steps to automate the epidemiological modelling of Cercospora leaf spot (CLS) epidemics in Cercospora beticola in sugar beet farms. The package contains functions to use remotely sensed spatially explicit empirical data to estimate leaf area index (LAI) when the crop canopy closes. Crop canopy closure is one key variable in estimating when growers need to be vigilant, and commence scouting their crops for CLS disease incidence. Once CLS epidemic incidence shows at lease 50 % of leaves with one or more lesions, ‘epidemic onset’ is deemed to have occured. This negative prognosis model attempts to estimate the earliest possible date when ‘epidemic onset’ is likely to occur, based on weather conditions. Farmers can be reasonably confident an epidemic has not occured prior to the dates estimated by this model.

Once epidemic onset has occured a decision to protect their crop with fungicides is needed to prevent yield limiting Cercospora leaf spot epidemic.

cercospoRa uses functions described by Wolf and Verreet (2005) “Factors Affecting the Onset of Cercospora Leaf Spot Epidemics in Sugar Beet and Establishment of Disease-Monitoring Thresholds” Phytopathology

Installation

Getting started

Format weather data

library(data.table)
library(cercospoRa)

# Inspect raw weather station data
head(cercospoRa::weathr)

weathr is a data.table containing weather data recorded at a sugar beet field trial observing the spread and severity of C. beticola.

# Make a copy of the data so we can modify the formatting without affecting the 
#  raw data
wthr <- data.table(weathr)

# Format times to POSIXct time with UTC timezone
wthr[,Time := as.POSIXct(paste0(Datum, " ",Stunde,":00"),tz = "UTC")]

# Nominate Latitude and Longitude location of the weather station. 
# While not needed in cercospoRa some plant disease models will use location to 
#  decide the closest weather station to pull weather from
wthr[, c("lon","lat") := list(9.916,51.41866)]

# Weather is hourly and will error if we don't specify a standard deviation of 
#  weather direction. This is intentional to force the user to decide how variable
#  the wind direction data could be.
wthr[, wd_std := 20]

# Remove all data after September as it contains missing data
wthr <- wthr[Datum < as.POSIXct("2022-10-01")]

# Set NA wind speed values to zero
wthr[is.na(WG200),WG200 := 0]

# Set NA wind direction values to 20 degrees. Wind is not important for this model
wthr[is.na(WR200),WR200 := 20]

# format_weather() formats weather data to 
#  hourly and checks for missing data or any issues that may cause downstream faults
#  in the model.
wthr <- 
  format_weather(wthr,
                      POSIXct_time = "Time",
                      time_zone = "UTC",
                      temp = "T200",
                      rain = "N100",
                      rh = "F200",
                      wd = "WR200",
                      ws = "WG200",
                      station = "Station",
                      lon = "lon",
                      lat = "lat",
                      wd_sd = "wd_std",
                      data_check = FALSE # this stops the function from checking for faults
                      )

Calculate the proportional progress towards an epidemic

# Resistant cultivar
calc_epidemic_onset(c_closure = as.POSIXct("2022-07-30"),
                    weather = wthr,
                    cultivar_sus = 3)
# Susceptible cultivar                    
calc_epidemic_onset(c_closure = as.POSIXct("2022-07-30"),
                    weather = wthr,
                    cultivar_sus = 6)

This returns a POSIXct date for the onset of an epidemic for the susceptible and more resistant cultivar. If the input weather data does not provide a window where a epidemic onset date is met, the proportional progress towards an epidemic is returned.

calc_epidemic_onset() is a wrapper for calc_DIV() which implements the effect of the cultivar susceptibility on the epidemic onset date. calc_DIV() however returns a data.table detailing the daily contribution towards the “daily infection values” (Wolf and Verreet, 2005) and is agnostic of cultivar susceptibility.
For more detailed output of daily infection values call calc_DIV()

Calculate daily infection values

calc_DIV(dat = wthr)

This produces a data.table detailing the daily infection value for each day using the method described in Wolf and Verreet (2005).

Note: Missing humidity values do not prevent the model from running and these days are assumed to not progress the model.

Notes for contributors

The main branch is the production branch and is the version available on CRAN. The main branch is locked, please contribute to the dev branch.
The plus_racca (development) branch also includes functions to recreate other C. beticola mechanistic models published by Racca and Jörg (2007) and auxiliary functions which might be helpful for future versions.

References

Wolf, P. F., & Verreet, J. A. (2005). Factors Affecting the Onset of Cercospora Leaf Spot Epidemics in Sugar Beet and Establishment of Disease-Monitoring Thresholds. Phytopathology, 95(3), 269-274.

Racca, P., and Jörg, E. (2007). CERCBET 3 – a forecaster for epidemic development of Cercospora beticola. EPPO Bulletin, 37(2), 344-349.