"best dealiasing method' and 'missing fourdd function' ??

[Redmund Nacario]

Hello all,

I'm a newbie in dealiasing radial velocity in pyart.
Can you please help me choose the best method or function in dealiasing radial velocity?
The realibility of the method or algorithm is also my concern.
Also, I had noticed that the fourdd function is missing in the pyart version '".5.0.dev+1567932".

Your help will be greatly appreciated :)

thank you,

Sincerely,

Redmund Nacario

[Jonathan Helmus]

Redmund,

    Py-ART has three functions which can dealias radial velocities:

pyart.correct.dealias_fourdd : Uses the FourDD algorithm to unfold velocities from a volume consisting of PPI scans.  Described in detail in James and Houze, JTech, 2001.

pyart.correct.dealias_unwrap_phase : Unfold velocities using a multi-dimensional phase unwrapping procedure used in image processing.  Should work on PPI, RHI and VPT scans.  Works best when all regions in the volume are connected (few bad gates) and have high reliability.  Better algorithms are chosen by the function when the number of rays in each sweep are the same and the sweeps are aligned. 

pyart.correct.dealias_region_based : Dealias velocities by finding regions of similar velocities and unfolding these regions against each other iteratively.  Works on PPI, RHI and VPT scans.  Deals well with gaps in data.  Does not deal well with noise and artefacts which should be identified prior to unfolding and excluded from processing using a gatefilter.

Personally, I find the region based method the best, but my opinion is biased as I designed and wrote the algorithm.  A great feature of Py-ART is that you can try out all three methods and compare the results.  Try for yourself on a few test cases and see which routine works best for your data!

As for the missing fourdd function, when you built Py-ART was the TRMM RSL library installed and available?  If it was not a warning message was printed to the screen but you may have missed it with all the text scrolling by.  FourDD makes use of RSL and is required for use.  You can check the availability of RSL using the command:

python -c "import pyart; pyart._debug_info()"

TRMM RSL is listed on the first line of the optional dependencies. For example on my system:

$ python -c "import pyart; pyart._debug_info()"
Py-ART version: 1.5.0.dev+1843b09

---- Dependencies ----
Python version: 2.7.10
NumPy version: 1.9.2
SciPy version: 0.15.1
matplotlib version: 1.4.3
netCDF4 version: 1.1.8

---- Optional dependencies ----
TRMM RSL version: v1.48
CyLP: Available
PyGLPK version: 4.45
CVXOPT version: 1.1.6
basemap version: 1.0.7
nose version: 1.3.4

Instructions for installing RSL can be found at the libraries website (http://trmm-fc.gsfc.nasa.gov/trmm_gv/software/rsl/).  If you install RSL in location other than the default you must specify this location using the RSL_PATH environmental variable when building Py-ART.  The section on optional dependencies in the README provides a bit more details on this process (http://trmm-fc.gsfc.nasa.gov/trmm_gv/software/rsl/).

Cheers,

    - Jonathan Helmus

--
You received this message because you are subscribed to the Google Groups "Py-ART Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pyart-users...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[wubean]

Hello Jonathan Helmus,

         I find  the region-based method which designed and wrote by you works well. Thank you very much for the algorithm you provided！Can you provide a detailed introduction or a paper to the dealias_region_based algorithm. Through it, I can understand this algorithm better , thanks a lot ! 

在 2015年8月25日星期二 UTC+8下午9:44:10，Jonathan Helmus写道：

[Jonathan Helmus]

I've given a number of presentation of the region based dealiasing routine which can be found at  https://github.com/jjhelmus/presentations. 

Cheers,

    - Jonathan Helmus

[Scott Collis]

Thanks Jonathan! 

In addition Bobby Jackson has been working on the “Out by one nyquist” issue using a sounding or previous volume.. We should be able to get this back into Py-ART core soon (we really have not stopped being busy since the Radar Conference!)

--

Scott Collis

[wubean]

Dear Jonathan Helmus，

       Thank you for all your assistance！

       

        Yours Sincerely,

         wubean

在 2017年10月16日星期一 UTC+8下午11:06:47，Jonathan Helmus写道：