It looks like you're new here. If you want to get involved, click one of these buttons!

- All Categories 2.2K
- Applied Category Theory Course 343
- Applied Category Theory Exercises 149
- Applied Category Theory Discussion Groups 48
- Applied Category Theory Formula Examples 15
- Chat 475
- Azimuth Code Project 108
- News and Information 145
- Azimuth Blog 148
- Azimuth Forum 29
- Azimuth Project 190
- - Strategy 109
- - Conventions and Policies 21
- - Questions 43
- Azimuth Wiki 708
- - Latest Changes 700
- - - Action 14
- - - Biodiversity 8
- - - Books 2
- - - Carbon 9
- - - Computational methods 38
- - - Climate 53
- - - Earth science 23
- - - Ecology 43
- - - Energy 29
- - - Experiments 30
- - - Geoengineering 0
- - - Mathematical methods 69
- - - Meta 9
- - - Methodology 16
- - - Natural resources 7
- - - Oceans 4
- - - Organizations 34
- - - People 6
- - - Publishing 4
- - - Reports 3
- - - Software 20
- - - Statistical methods 2
- - - Sustainability 4
- - - Things to do 2
- - - Visualisation 1
- General 39

Options

I ran across another ENSO index, called BEST for Bivariate EnSo Timeseries

- Info: http://www.esrl.noaa.gov/psd/people/cathy.smith/best/
- Data: http://www.esrl.noaa.gov/psd/people/cathy.smith/best/enso.ts.1mn.txt

This is a good one for doing machine learning on because it is relatively free from noise and shows little by way of a trend. It is all oscillations.

The machine learning finds the usual QBO forcing period of around 28 months, and a Mathieu-like modulation of 9 to 12 year periods. It also finds a characteristic period of a little over 4 years, spanning an interval running back to 1880.

The top chart has double the complexity fit as the second chart. Both correlation coefficients are above 0.85.

## Comments

Hello Paul

But how do they post-process these, if they apply any filters the forecast algorithms' accuracies wildly vary!

If you filter the noise, and work with trend, you could get awesome forecasts, for almost any time-series

`Hello Paul But how do they post-process these, if they apply any filters the forecast algorithms' accuracies wildly vary! If you filter the noise, and work with trend, you could get awesome forecasts, for almost any time-series`

No trend in this.

`No trend in this. ![best_fit](http://imageshack.com/a/img903/3773/5YZFMR.gif)`

And the machine learning model can detect the deviations from the intrinsic ENSO, which not surprisingly are the result of transient cooling shortly after the three massive volcanic eruptions of the latter part of the 20th century -- Agung, El Chichon, and the pair of Pinatubo and Cerro Hudson in 1991.

Is it a forecast?

Is it a model?

Is it overly filtered?

Does it help us understand and reveal the true nature of ENSO?

Yes to the last one.

`And the machine learning model can detect the deviations from the intrinsic ENSO, which not surprisingly are the result of transient cooling shortly after the three massive volcanic eruptions of the latter part of the 20th century -- Agung, El Chichon, and the pair of Pinatubo and Cerro Hudson in 1991. ![volcanos](http://imageshack.com/a/img538/2687/0yKCsR.gif) Is it a forecast? Is it a model? Is it overly filtered? Does it help us understand and reveal the true nature of ENSO? Yes to the last one.`