Show HN: Making pro weather data free for Hurricane Dorian

73 points by duanem ↗ HN
I have added OpenZones to Flowx, my Android weather app[0]. OpenZones are regions covering disaster zones where all the pro data in the app is freely available.

I have created an 1200km diameter OpenZone over Florida with Hurricane Dorian approaching. So if you add Florida to the app, you can view radar, NOAA's GFS, NAM and HRRR models, CMC GDPS and RDPS models, and the DWD (Germany) ICON model. There are also predicted hurricane tracks from the NOAA and CMC ensemble models which I find extremely valuable to predict the possible paths of Dorian. This YouTube video[1] shows an example of the hurricane tracks.

The reasoning behind OpenZones is that meteorological organizations solve weather simulations and release data to the public with the aim to reduce damage, injury and fatalities. It's only right to support this purpose and open up all data in disaster zones.

The idea for OpenZones came from the fires (Camp Fire) in California last year. I was part way through adding HRRR smoke simulation data to Flowx when a user asked if I could release the smoke data early for the California fires. I did so in the free version of the app and then rolled it into the pro version after the fires.

Sorry there is no Apple version of the app.

[0] https://play.google.com/store/apps/details?id=com.enzuredigital.weatherbomb

[1] https://www.youtube.com/watch?v=h5do7dYfYfQ

12 comments

[ 2.1 ms ] story [ 34.6 ms ] thread
This is cool! I will download.

I also have a Dorian-related Android app to share. I use the barometers in phones to measure surface pressure with All Clear:

https://play.google.com/store/apps/details?id=com.allclearwe...

Here is a graph of Hurricane Florence last year: https://www.allclearweather.com/hurricane-florence. I will do a much more geographical and full analysis for Dorian. The raw data from barometers in phones is messy, and requires significant bias correction and quality control, but one day could be assimilated into weather models like WRF and improving forecast accuracy.

I believe the hurricane tracks is a solution to the problem outlined in the HN post 3 days ago on "Misinterpreting hurricane forecasts"[0].

The hurricane tracks are a new addition to Flowx but I believe it can be improved a lot. For example, you can color the dots by the hurricane category or wind speed. Better still, draw a heat map around each dot would generate a probability map for the hurricane.

[0] https://news.ycombinator.com/item?id=20833030

Not a bad idea but only if the tracks are independent. Was reading something saying that these tracks aren’t different random realizations of the same simulation (in which case what you’re saying is a good way to estimate the posterior of the hurricane track — it’s just Monte Carlo) but that these tracks come from many different types of simulations, some with higher quality than others and they’re likely all correlated in some weird way. If it’s the latter case there has to be some care and thought put into combining the predictions into a single estimate of the posterior.
The initial conditions of the weather simulation, which is based on measured data and previous simulation results, are perturbed to within a reasonable range to represent the error in the measurements. Because of the equations being solved and numerical methods being used, these small changes can cause large change on long term predictions.

The tracks come from one original simulation. If you look at the start of the tracks closely, you'll see the eye of the hurricanes diverge in a regular grid, so I'm guessing they are perturbing the direction and speed of the bulk movement of mass - I would have to research more.

So the initial conditions and early stages of the simulations are correlated, but each simulation is solved independently. The question is how long into the future can you consider each track less-correlated.

Given the met organizations release probability envelopes based on these track means it's a reasonable approach.

What I really want to do is weight the predicted wind speed with the probability map. I don't know if this will work. But you can imagine if a hurricane has a 75% change of going well into the continent it would weaken and cause less damage, but if the other 25% of the tracks go over sea and hit an island, it'll likely strengthen and cause more damage.

In any case, all the above will be interesting.

How does this post not qualify as 'spam'?
It's informational, for the free version of the app, and you have to click on it to read it. Spam is a term for unsolicited commercial messages sent indiscriminately. Still marketing arguably, but then so is a lot of things that get posted to HN.
Thanks. I was going through my memory of previous "Show HN" posts I've read and many are products.

If it's any consolation, we don't have a marketing department or budget :-)

I did read the guidelines before posting and I honestly don't know how to quantify what is spam.

I don't believe it's spam since there are new two features, OpenZones and hurricane tracks, that are interesting and useful to people, especially with the hurricane. I'm quite proud of these developments. The OpenZones meet public good while still having a way to cover costs. The hurricane tracks visualization is new and unique and addresses the problem highlighted in the "Misinterpreting hurricane forecasts"[0]. I believe these are appropriate for "Show HN".

I have many other ideas to develop for Flowx, it would be sad if they could not be shown here because it's in a product.

BTW, the guidelines say this about spam: "Please don't complain that a submission is inappropriate. If a story is spam or off-topic, flag it. Don't feed egregious comments by replying; flag them instead. If you flag, please don't also comment that you did."

[0] https://news.ycombinator.com/item?id=20833030

Because OP is showing something that he's made, and that's exactly what Show HN is for.
Exactly, the post is absolutely fine. Also, Duane is acting in good spirit.
I've been using your app for awhile now! It's quite good. OpenZones is a fantastic idea.