This folder includes the code written in Matlab and developed by B.Martinez in order to reproduce the figure 8 of the manuscript entitled 
'The nonlinear embeddings method for long-term vegetation dynamics' and published in the
ISPRS Journal of Photogrammetry and Remote Sensing 2025.

The following content is included:
1:'COMPUTE_PIXEL_NDVI_2001_2022_NLE_BEAST_MRA.m', this is the code for reproducing Figure 8 of manuscript.
2: Data folder contains two files (.mat) with the NDVI and Quality Control time series, respectively. Each file contains 8 time series.
3: BEAST folder contains the code in Matlab for the computation of the BEAST method. This is call from the main code (1) and it must 
   be included in the working directory or preferences path in Matlab. 
4: NLE folder contains the code in MAtlab for the computation of the embedding trend. This is call from the main code (1) and it must 
   be included in the working directory or preferences path in Matlab.

The code provides the slope of the trend along with its associated error por 8 selected pixels in Spain described in Table 1.

The code plots the original time series along with the trend components associated to NLE, BEAST and MRA methods. 
The slope magnitude and its error for the three methods is also written in the plot.
The plot obtained in this code agrees with Figure 8 although it does not exactly reproduce the same layout. 
It has to be mentioned that the BEAST results are slighty different due
the Bayesian averaging scheme used by the BEAST model (see section 3.4 of the manuscript). 

Created by Beatriz Martínez Díaz on 2025-04-02.
Departament de Física i Termodinàmica. Facultat de Física
Universitat de València
email: beatriz.martinez@uv.es
Copyright (c) 2025 Beatriz Martínez
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%