Diagnose crop N status from biomass and N content (NNI classes)

Wraps compute_NNI and turns continuous NNI values into three discrete classes used for fertilisation diagnosis and for diagnostic maps in variable-rate applications.

Usage

diagnose_N_status(
  N_content,
  biomass,
  crop,
  curve = NULL,
  is_percent = FALSE,
  deficient_threshold = 0.9,
  excessive_threshold = 1.1
)

Arguments

N_content: Aboveground N concentration as a fraction of DM (e.g. 0.025 = 2.5pct) or percentage DM (pass is_percent = TRUE).
biomass: Aboveground dry biomass in t DM / ha (tonnes).
crop: Crop identifier, passed to critical_N_curve.
curve: Optional list with a and b (and optionally W_min) to override the default curve. Useful for custom local calibration.
is_percent: Logical. Is N_content supplied as a percentage (e.g. 2.5 for 2.5pct) rather than a fraction? Default FALSE (fraction). Auto-detected and warned if values > 1.
deficient_threshold: Lower bound of the optimal NNI band. Default 0.90.
excessive_threshold: Upper bound of the optimal NNI band. Default 1.10.

Value

A list with:

NNI: Continuous NNI (numeric or RasterLayer).
class: Integer class (1/2/3) or RasterLayer of classes.
labels: Character labels matching class.
summary: Counts / fractions per class (rasters only).
thresholds: The thresholds used.
curve: The critical-N curve parameters.

Details

Class codes (integer factor levels):

1 - deficient: NNI < deficient_threshold (default 0.90)
2 - optimal: deficient_threshold <= NNI <= excessive_threshold (default 1.10)
3 - excessive: NNI > excessive_threshold

The class raster can be fed straight into variable-rate logic: e.g. apply the full agronomic dose only where class = 1 or 2, zero dose where class = 3.

Examples

if (FALSE) { # \dontrun{
# Scalar diagnosis
diagnose_N_status(N_content = 2.8, biomass = 4,
                  crop = "maize", is_percent = TRUE)

# Pixel-wise diagnosis
d <- diagnose_N_status(N_content = n_raster, biomass = w_raster,
                       crop = "wheat", is_percent = TRUE)
raster::plot(d$class)
d$summary
} # }