Skip to content

gdelannoy/obs_operator_calibration

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

obs_operator_calibration

WCM calibration

Description

This repository contains Matlab files to calibrate the Water Cloud Model (WCM), using NASA Land Information System (LIS) output. These scripts are used to calibrate the WCM in De Lannoy et al., (2024, JAMES) and de Roos et al., (2024, Biogeosciences).

  • WCM: four parameters, A, B, C, D, and optionally the observation error standard deviation can be calibrated. The initial values for A and B vary with the chosen incidence angle.
  • LIS output: containing surface soil moisture and LAI or biomass, as well as Sentinel-1 observations, collected during an open loop simulation. The output can be generated by any land surface or crop model, and has been tested for Noah-MP (De Lannoy et al., 2024) and AquaCrop (de Roos et al., 2024).
  • Optimization algorithm: four algorithms are tested, i.e. DREAMZS, PSO, SCE_gent, and SCEUA. Note that the latter two occasionally do not converge as well as the former.
  • Objective function (OF): for now only the Bayesian objective function has been fully tested. This consists of 2 penalty terms: the sum of squared errors (SSE) between observed and simulated backscatter, and the SSE between an initial constraint on the parameters and their estimates.

Contents

1. Main script CalWCM_main.m can be run directly in the Matlab interface. Turn on the debug option (inside the code), to see some time series results. CalWCM_main_PAR_CPU.m is the same script, but allows parallel processing. This script is called if the calibration is launched in the queue, e.g. as ./RUN/My_M_queued_WCM_irr.sh

2. Subroutines Optimization algorithms, functionalities to read model output, and WCM.

3. RUN Directory with example files on how to submit a job in the queue for fast parallel calibration, and combining the output from multiple processors into a single output file.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • MATLAB 97.7%
  • Shell 1.7%
  • Perl 0.6%