Extract AUDPC Summary as data.table — audpc

Extracts the Area Under Disease Progress Curve (AUDPC) from an epicrop simulation output along with spatial (lat/lon) and temporal (emergence) metadata. Returns a compact summary table suitable for analysis, mapping, or reporting.

Usage

audpc_dt(x)

Arguments

x: An epicrop.sim object from seir(), disease model functions (bacterial_blight(), leaf_blast(), etc.), or run_epicrop_model() with output = "full".

Value

A data.table::data.table() with one row containing:

lat: Latitude (double; NA if not available)
lon: Longitude (double; NA if not available)
AUDPC: Area under disease progress curve (double)
emergence: Emergence date (character, "YYYY-MM-DD")

Table is keyed on emergence for efficient operations.

Output Structure

The returned data.table::data.table() includes:

lat: Latitude (if available in the model object)
lon: Longitude (if available in the model object)
AUDPC: Area under disease progress curve (numeric)
emergence: Emergence date as character (YYYY-MM-DD)

Columns are present only if the corresponding data exist in the input object. The table is keyed on emergence for efficient sorting and joining.

Use Cases

Batch processing: Quickly summarise results from run_epicrop_model()
Mapping: Combine with spatial data for disease risk mapping
Reporting: Generate summary tables for publications or extension
Analysis: Join with cultivar, management, or climate data

Author

Adam H. Sparks, adamhsparks@gmail.com

Examples

# Example 1: Extract AUDPC summary from single model run
wth <- get_wth(
  lonlat = c(121.25562, 14.6774),
  dates = "2000-06-30",
  duration = 120L
)

sim <- bacterial_blight(wth, emergence = "2000-06-30")
summary_table <- audpc_dt(sim)
summary_table
#> Key: <emergence>
#>        lat      lon    AUDPC  emergence
#>      <num>    <num>    <num>     <char>
#> 1: 14.6774 121.2556 37.74063 2000-06-30
# Output includes lat, lon, AUDPC, emergence

# Example 2: Batch summarise multiple emergence dates
emergence_dates <- seq(as.Date("2000-06-15"), as.Date("2000-07-15"), by = 7)
summaries <- lapply(emergence_dates, function(ed) {
  sim <- bacterial_blight(wth, emergence = as.character(ed))
  audpc_dt(sim)
})
#> Error in .window_weather(wth = wth, emergence = emergence, duration = duration): Emergence date not found in weather data
#> ✖ Emergence 2000-06-15 not in range 2000-06-30--2000-10-28
summary_all <- data.table::rbindlist(summaries)
#> Error: object 'summaries' not found
summary_all
#> Error: object 'summary_all' not found

# Example 3: Compare multiple locations and emergence dates
locations <- data.frame(
  name = c("IRRI", "Manila"),
  lon = c(121.25562, 120.985),
  lat = c(14.6774, 14.6042)
)

emergence_dates <- c("2000-06-30", "2000-07-15")

results <- expand.grid(
  location = locations$name,
  emergence = emergence_dates,
  stringsAsFactors = FALSE
)

results$AUDPC <- mapply(function(loc_name, em_date) {
  loc <- locations[locations$name == loc_name, ]
  wth <- get_wth(
    lonlat = c(loc$lon, loc$lat),
    dates = em_date,
    duration = 120L
  )
  sim <- leaf_blast(wth, emergence = em_date)
  get_audpc(sim)
}, results$location, results$emergence)

results
#> # Data frame like object (class data.frame) 3 x 4:
#>  │location│emergence │AUDPC
#>  │<chr>   │<chr>     │<dbl>
#> 1│IRRI    │2000-06-30│   41
#> 2│Manila  │2000-06-30│   41
#> 3│IRRI    │2000-07-15│   38
#> 4│Manila  │2000-07-15│   38
# Identify highest risk scenario
worst <- results[which.max(results$AUDPC), ]
sprintf("Highest risk: %s, emergence %s, AUDPC = %.1f",
        worst$location, worst$emergence, worst$AUDPC)
#> [1] "Highest risk: IRRI, emergence 2000-06-30, AUDPC = 40.8"

# Example 4: Export for mapping or further analysis
wth <- get_wth(
  lonlat = c(121.25562, 14.6774),
  dates = "2000-06-30",
  duration = 120L
)
sim <- brown_spot(wth, emergence = "2000-06-30")
summary_dt <- audpc_dt(sim)