Matlab code for the retrievals of SPM and its associated uncertainties
This module provides a package of Matlab functions to derived SPM and associated uncertianties from field measurement of remote sensing reflectance at any spectral resolution or radiometer sensor. This is likely not the fastest implementation possible for the MW algorithm. However, it is easy to use and handles the processing of either multispectral and hyperspectral remote sensing reflectance data.
To make easier the understanding of the algorithm package, a sample of water-leaving reflectance (RW) its standard deviation, wavelengths, and field meadured SPM was set: sample_data.mat. In addition the function 'RW2SPM.m' calls all the input information necessary and the sequence of functions to run the scripts.
Because radiometric field measurements are most likely to be made available as water-leaving reflectance (RW) we suggest the use of the function 'RW2rrs.m' to convert the measured water-leaving reflectance (RW) to below water remote sensing reflectance (rrs).
In sequence the inputs necessary for for the main function 'MW_algorithm.m' are the ranges of shape parameters (S and Y) and the mass-specific coefficientes (a_nap at 443 nm, a_nap at 750 nm, and b_bp at 700 nm), the temperature at which RW was measured, and the saturation threshold (Q_filter)
The 'MW_algorithm.m' function will run in two steps: the first step estimating SPM and uncertaintyes and the second step constraining shape parameter and mass-specific coefficients.
For use applied to satellite remote sensors find: https://github.com/OceanOptics/MW_algorithm_satellite/
For questions regarding the script implementation or to suggest changes to improve its functionality, please contact Juliana Tavora at juliana.tavora@maine.edu; j.tavora@utwente.nl