Skip to contents

luxR

Underwater Light Analysis and Visual Ecology

luxR provides tools for quantifying and modelling underwater light environments. It propagates spectral irradiance through the water column via Beer-Lambert attenuation, converts between energy and photon-flux units, and estimates photoreceptor excitation and visual contrast.

R CMD check

Installation 

# Development version from GitHub:
# install.packages("remotes")
remotes::install_github("JohnKirwan/luxR")

Quick start 

library(luxR)

# Load a coastal irradiance profile as a lux_spectrum object
x <- from_naples("0m")
print(x)
#> <lux_spectrum> irradiance [umol/m2/s/nm] | 352.5-862.5 nm, 5 nm bins (103 pts)

# Convert units
x_W <- convert_unit(x, "W/m2/nm")

# Compute PAR and lux
par_irradiance(x_W)    # µmol m⁻² s⁻¹
irradiance2lux(x_W)    # lux

# Propagate to depth using a Jerlov Type II water model
x_prop <- x_W[350, 700]       # bundled Jerlov Kd is supported here
lam   <- x_prop$lambda
Kd_II <- jerlov_Kd("II", lam)
specs <- propagate_spectrum(x_prop, Kd_II, from = 0,
                            to = c(0, 5, 10, 20, 50))
plot(specs)

# Colour discrimination at depth
grey_R <- lux_spectrum(rep(0.3, length(lam)), lam,
                       quantity = "reflectance", unit = "dimensionless")
red_R  <- lux_spectrum(0.2 + 0.25 * exp(-((lam - 625)/35)^2), lam,
                       quantity = "reflectance", unit = "dimensionless")

colour_jnd(
  stim1    = reflectance_to_radiance(grey_R, specs[["0"]])$E,
  stim2    = reflectance_to_radiance(red_R,  specs[["0"]])$E,
  lambda   = lam,
  species  = "Danio rerio",
  receptor = c("S-cone", "M-cone", "L-cone")
)

Default colour models use only cited chromatic receptor combinations from species_channels. Rods, melanopsin channels, and incomplete receptor sets are not silently treated as additional colour channels.

Propagation model and limits

luxR’s principal modeling tier is wavelength-resolved Beer–Lambert attenuation with a diffuse attenuation coefficient, Kd(λ). It assumes Kd is constant over the modeled path and propagates wavelength bins independently. It does not solve an angular radiance field, multiple scattering, or processes that move energy between wavelengths; use HydroLight, EcoLight, or WASI when those are required.

The bundled Jerlov table covers eight optical water types from 350–700 nm. jerlov_Kd() fails outside that range by default; constant endpoint extension must be explicitly requested and is recorded in result metadata. There is no universal supported maximum depth—the chosen Kd must remain representative of the full path. Inverse propagation is a homogeneous-column reconstruction and can strongly amplify measurement noise.

Key features

Area Functions
Spectral class lux_spectrum(), convert_unit(), as_lux_spectrum()
Depth propagation propagate_spectrum(), band_irradiance(), fit_Kd(), jerlov_Kd()
Photometrics par_irradiance(), irradiance2lux(), scotopic_lux()
File I/O from_naples(), from_solar(), read_trios(), from_trios(), read_ocean_optics(), from_ocean_optics()
Visual ecology govardovskii_template(), quantum_catch(), colour_jnd()
Visibility scenarios secchi_depth(), visual_range(), detection_range(), detectability()
Interactive app run_app()

Interactive explorer 

run_app()   # requires the 'shiny' package

run_app() launches a bundled Shiny application with guided tabs including:

  • Depth Propagation — pick a built-in spectrum or upload explicitly calibrated Ocean Optics spectral irradiance, attenuate it through a Jerlov water type, and download the per-depth lux, photon-flux, calibration, and provenance table. TriOS radiance is deliberately excluded because it cannot be converted to irradiance without angular measurements.
  • Species Perception — photoreceptor quantum catch for a chosen species under a selected bundled light-source condition at that source’s stated reference depth.
  • Colour discrimination — Vorobyev-Osorio JND between two reflectances under the in-water light field.
  • Visibility — Secchi depth, photic depth, and horizontal visual range from the water’s diffuse attenuation.
  • Detection — explicitly unvalidated contrast-threshold distance scenarios; these are teaching estimates, not predictions of actual organismal detection range.

Documentation

Full function reference and four vignettes are on the pkgdown site:

  • Light in the water column — field data loading, PAR, Kd fitting, Jerlov propagation, photic depth
  • From photons to perception — unit conversions, quantum catch, colour discrimination at depth
  • Seeing through water — an end-to-end object-detection scenario with explicit propagation and visual-model assumptions
  • Polarization of the underwater light field — the scope, validation, and use of the tunable Rayleigh-like polarization approximation

License

GPL (>= 3)