Lord have mercy 1039 banana high. Pants tight
-
Hello, please take a minute to check out our awesome content, contributed by the wonderful members of our community. We hope you'll add your own thoughts and opinions by making a free account!
CustomWX Graphic Solution!
- Thread starter Ollie Williams
- Start date
Here's a gif from the Blizzard of 1996.
The infamous winter storm from 2011.
Winter storm of '88. Beautiful looking storm!
Winter storm of 2016.
All time favorite
Maps from the December storm of 2018. 
I just plotted some GFS maps going back to a storm in 2017 (The Infamous RDU gradient storm). Here's where the run happened: https://southernwx.com/community/threads/1-7-17-1-10-17-winter-storm.53/page-12
Last edited:
I just created a nice ensemble member panel graphic. I plan to use my graphics during certain storms. These will be useful if I wanted to zoom into a certain state rather than with the map restrictions most sites have.
Last edited:
Here's an example of the zoom capability.
I just finished a GEFS mean graphic.
Webberweather53
Meteorologist
Made some graphics in python using the African Rainfall Climatology Version 2 (ARCv2) Dataset w/ 0.1*0.1 deg grid spacing.
The last decade of May 1-June 26 precipitation anomalies plus this year. This year is on a whole other planet
May-October Monthly precipitation climatology.
This dataset nearly broke python, 2.5*10^9 data points is a lot to handle haha
Here's a daily standardized anomaly time series I created in matlab. North Africa has seen above normal precipitation basically every day since the beginning of May, little sign of that changing anytime soon
The last decade of May 1-June 26 precipitation anomalies plus this year. This year is on a whole other planet
May-October Monthly precipitation climatology.
This dataset nearly broke python, 2.5*10^9 data points is a lot to handle haha
Here's a daily standardized anomaly time series I created in matlab. North Africa has seen above normal precipitation basically every day since the beginning of May, little sign of that changing anytime soon
Nice work Eric! I’m curious what you did to calculate the average climatology? Was it already a product, or did you have to calculate it separately? I’ve looked into calculating the CFSR climatology and had no luck. How did you go about downloading it, did you write it to download from the Thredds server, whenever I try, my computer locks up.
Webberweather53
Meteorologist
I had to calculate it separately within my code using a few for loops. It honestly wasn't too bad.
I downloaded each May-Oct precip file separately from LDEO's website & concatenated them along the time dimension to produce one big 5 GB file which then I used to create daily & monthly climatologies. I also occasionally flipped back & forth between python & matlab to make some of my graphics (because some matrix operations are easier in matlab)
Webberweather53
Meteorologist
Everywhere in green is experiencing record high precipitation for May 1-June 26th according to ARCv2. ~25% of the grid boxes in this domain had records >1 mm/day, ~15% >2 mm/day.
Last edited:
Webberweather53
Meteorologist
One more N Africa precip graphic:
Today, I've been playing around creating windows in C++ (A programming language). I've been learning C++ for a while, because unlike Python, which is great for simple graphics, and calculations, C++ is developed for speed and size. Most applications are created in C++, such as video games, or weather applications such as Baron Lynx or WSI Max. My dream has been to create a broadcast application similar to the ones you see on the news, but for personal users. The closest anyone has come to that has been WSV3, but they only include maps, my dream would be to add on headers, and forecast graphics, similar to the ones I make in Powerpoint. Although it is much harder than it sounds. This is what I managed to accomplish.
Last edited:
I just created a fun little program in C++
.
.
VegasEagle
Member
How hard to learn c++
It’s easy if you follow tutorials. Granted It was easier because I understood Python in which the concepts are universally the same. C++ is one of the tougher ones, as far as syntax (the way it’s written), which is really the only reason people say it's hard. However, it’s not too hard to get used to as long as you practice. This guy does a great job teaching it: .
I do want to mention that what I posted is a different aspect of C++ programming, which deals with graphics/Directx/HLSL, a lot of the stuff is much different/tougher, which I'm currently learning. This guy that I'm following does teach Directx along with C++ concepts, although for someone new to the language, I think it would be difficult to understand. Here's the link:
VegasEagle
Member
Thank you!
Webberweather53
Meteorologist
I believe matlab was written in C++, matlab is unforuntately a software you'd have to pay for, but I personally love its simple graphics. Geospatial plotting is still better in python imo because matlab uses basemap, which appears somewhat archaic
Interesting, I didn't know Matlab was in C++.
Webberweather53
Meteorologist
Yeah that's what it was written! I've been working with matlab for about 5 years, I personally think the plotting functions are more intuitive than python and (for obvious reasons) it's much better & less work to handle matrix operations than say numpy. I personally prefer matlab over python when making simpler line plots, scatter plots, time series, bar graphs, histograms, etc.
Webberweather53
Meteorologist
Finally figured out how to plot streamfunction and velocity potential anomalies in python. This is what the composite for hyperactive Atlantic hurricane seasons looks like in the new ERA-5 reanalysis.
Positive streamfunction anomaly = high pressure (warm colors) & vis versa. Almost forgot to mention that this flips sign in the southern hemisphere! (May try to tweak my code s.t. it's a bit more intuitive to the average joe)
Positive 200 hPa velocity potential anomaly = convergence & sinking air (brown color) & vis versa.
Positive streamfunction anomaly = high pressure (warm colors) & vis versa. Almost forgot to mention that this flips sign in the southern hemisphere! (May try to tweak my code s.t. it's a bit more intuitive to the average joe)
Positive 200 hPa velocity potential anomaly = convergence & sinking air (brown color) & vis versa.
Last edited:
Webberweather53
Meteorologist
Also made my African precip graphic a look a bit more visually appealing.
[mention]Ollie Williams [/mention] [mention]Webberweather53 [/mention]
Ollie, I noticed you work with software programming. That’s good. I’m in school at UofM and about to either move my course to Cornel, or ECPI; around where I’m from (Greenville SC)
Did you use JavaScript/TypeScript/Node? Honestly, python is only used to analyze data. I’m learning HTML5, CSS3, and JavaScript. If you want to do any web development. My ultimate goal is to build a platform similar to Pivotal Weather, Weather Bell, and TT.
Generally, I’m doing this on my own. Most of this takes a team of programmers. I’m always looking for people with similar interest as myself. To bounce issues off one another.
Sent from my iPhone using Tapatalk
Ollie, I noticed you work with software programming. That’s good. I’m in school at UofM and about to either move my course to Cornel, or ECPI; around where I’m from (Greenville SC)
Did you use JavaScript/TypeScript/Node? Honestly, python is only used to analyze data. I’m learning HTML5, CSS3, and JavaScript. If you want to do any web development. My ultimate goal is to build a platform similar to Pivotal Weather, Weather Bell, and TT.
Generally, I’m doing this on my own. Most of this takes a team of programmers. I’m always looking for people with similar interest as myself. To bounce issues off one another.
Sent from my iPhone using Tapatalk
I only use Python for my graphics, I know Eric uses a bit of Matlab too which uses the same tools, but is a little faster. Python has a lot of libraries to plot really nice maps (Matplotlib, Pandas, Cartopy etc.). I know that is what all model sites use to make the maps. Then of course as you mentioned, they do use web design only for the user interface which is something I have not worked on personally, so I don't use JavaScript/TypeScript/Node or others. Eric may know more since he runs a site. There's also the data server-aspect which uses SQL, which I also know little about.
If you wanted to know as to how the Python plots are Unidata made a tutorial site that has basically all you need to know
Unidata Python Training
A one-stop shop for Python in atmospheric science and meteorology
unidata.github.io
I've personally always have wanted to create a model site, but have never gotten around to learning web development. Right now I'm focusing on making graphics with a broadcast flair.
As I promised to @NoSnowATL here's the '82 storm "Snow Jam" where the SE got 2 back-back board-wide storms.
Nice! Well done [mention]Ollie Williams [/mention]
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
NoSnowATL
Member
We so need this to happen again before I die!
UofM? Michigan? Miami?
Interesting, I'll post the vort maps soon.
Here are the vort maps that show the entire thing was one wave. Now that's how you get a big-dog!
Cool. Go Blue!!