[Thread Prev][Thread Next][Index]

Re: [ferret_users] Problem with detrending spatial rainfall



Hi Deni,

This is called anomaly, which only contains temporal variation but without mean and trend. So negative is correct. If you don’t want to see them, I guess to plus the mean back will help a bit. To plus the qhat at the last time pint will solve this problem, though I don’t know its physical meaning.

Tony

Sent from my iPhone

On Feb 10, 2019, at 6:26 PM, Satyesh Ghetiya <satyeshghetiya@xxxxxxxxx> wrote:

Hi Deni,
If I am correct, actually when we detrend , in almost all data sets, negative values will definitely come. 
In your case, detrend is rain minus 'qhat (kind of mean)'. (so the mean of detrend is about 0, which means detrend is about half times +ve and half -ve).  So what happens is that the signal (detrended) will now fluctuate about 0, 0 being midway. So we are removing the trend and hence only time variable component remains whose mean is about 0. 

Regards,
G Satyesh

On Sun, Feb 10, 2019 at 1:43 PM Deni Okta Lestari <deniokta.lestari28@xxxxxxxxx> wrote:
Dear ferret user,

I’m trying to detrending for 3D rainfall data. As we all know, the rainfall data has zero value. So, If I try to calculating following this archive https://ferret.pmel.noaa.gov/static/FAQ/analysis/regressions.html :

yes? USE rainfall.nc
yes? LET p = t[gt=rain]
yes? LET q = rain
yes? SET GRID q
yes? GO regresst

yes? LET/TITLE="rainfall - mean, trend" detrended = rain - qhat

the "zero value” also calculated. It will cause the detrended value become -(negative). Any experince for this case?

Hope to get suggestion soon.

Best regards,
Deni
 


--
G Satyesh


[Thread Prev][Thread Next][Index]
Contact Us
Dept of Commerce / NOAA / OAR / PMEL / Ferret

Privacy Policy | Disclaimer | Accessibility Statement