I'm trying to construct a MODIS temporal series as shown in this thread:

Harmonic MODIS trend in Google Earth Engine?

But I'd like to plot, if possible, the values for trend and seasonality for each time in the time series, and the regression returns just one image as result.

I'd like to plot something like presented in the BFAST-R webpage (http://bfast.r-forge.r-project.org/), which the figure below was extracted.

BFAST attributes extraction

Can I extract this kind of information from resultant image given in GEE?

  • To clarify, are you asking about (1) how to find the breakpoints, (2) how to estimate the trend and seasonality for a time interval between breakpoints, or (3) do you already have breakpoints, trend & seasonality, and you are trying to produce similar plots? Note that stackexchange posts are best if they ask a single focused question. Mar 4, 2018 at 17:28
  • Hi Tyler. Actually my desire is (1) and (2), and the plot after that. If (2) is done like as presented by Nicholas on the Summit' 2017 (tutorials section in the playground), I think that I can reproduce it. My idea is: once I get the breakpoints, I could run the trend and seasonality extraction (as shown by Nich) and acquire one resultant scene as result (using given example, for instance). After that I should plot the results (and I don't know how to do it yet). Am I correct? Mar 5, 2018 at 17:40

2 Answers 2


If you want the values for trend and seasonality at each timepoint, you could just feed it the different series, iteratively, for timepoints 0-1, 0-2,0-3, etc.

The trend in your image looks like it's been broken at a small number of points.


I use Sepal for BFAST https://sepal.io/

Sepal = GEE+R

  • 1
    Welcome to GIS StackExchange! We are looking for longer answers that provide context or explanation. Please edit your response with additional information.
    – SMiller
    Feb 26, 2019 at 20:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.