[ncl-talk] Grainy image

[Mary Haley]

Hi Tim,

I echo what Rick says about not being certain without seeing the data.

Is your data a set of random points, or on a gridded structure, or
something else? If it's on a gridded structure, then what happens if you
don't set trGridType at all?

I'm also interested to know the structure of your lat/lon data and how
dense the values are. Does your data contain any missing values? I don't
think this is the issue here, but it is important to be aware of this.

One thing I like to do for problem plots like this is to zoom in on the
area with the problem, and then plot dots where the lat/lon locations are
just to see if there's anything going on with regard to the density of the
locations and/or potential missing values.

If you look at this page:

http://www.ncl.ucar.edu/Applications/datagrid.shtml

you will see some examples of using gsn_coordinates to plot the lat/lon
locations of your data, and also use different colors for where the data is
missing or not.

In particular, datagrid_6.ncl might be useful if you have curvilinear data.

http://www.ncl.ucar.edu/Applications/datagrid.shtml#ex6

If you have random data, then contour1d.ncl on the same same might be
useful.

--Mary


On Thu, Jan 10, 2019 at 10:07 AM Rick Brownrigg <brownrig at ucar.edu> wrote:

> Hi Tim,
>
> I don't know for certain without looking at the data, but I have to
> imagine its manifest in the data itself, and that in those regions you
> mentioned the probability is hovering right around 50%.  The color contrast
> between 40% and 50% is rather abrupt, relative to the rest of the
> successive colors, which I think draws attention to the fluctuations. There
> also appears to be similar fluctuations going on in Montana and Minnesota,
> but its harder to see because those colors "closer" to each other,
> perceptually.
>
> FWIW...
> Rick
>
>
> On Thu, Jan 10, 2019 at 9:52 AM Tim Risner <trisner at awis.com> wrote:
>
>> I'm curious as to what may be causing a grainy section in Central
>> Mississippi and SE Ohio of the attached image.
>> I'm using "RasterFill" and cnRasterSmoothingOn is set to True.
>> trGridType of all types causes it.
>> Data is pretty uniform in those regions.
>>
>>
>> *Tim Risner*
>> *Operational Meteorologist*
>> *(888) 798-9955 ext 1006*
>>
>> <http://www.awis.com>[image:
>> https://www.facebook.com/AWISWeatherServices/]
>> <https://www.facebook.com/AWISWeatherServices/>
>> <https://twitter.com/awisweather>
>> _______________________________________________
>> ncl-talk mailing list
>> ncl-talk at ucar.edu
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20190110/a200e335/attachment.html>