iiinova — where your digital self keeps evolving

Reposted by Deelan Jariwala

James Franklin @franklinjamesl.bsky.social · 30/09/2026

Been a lot of storms since I ran 2026 AL/EP/CP verifications in early September. Results haven't changed much: GDMI still beating all the guidance for track and slightly better than OFCL on D4-5. For intensity, HCCA best, followed by OFCL and GDMI. Reminder that ECMWF is not in public ATCF decks.

154

Deelan Jariwala @wxtca.bsky.social · 09/09/2026

NBC Hannibal was probably my #2

000

Deelan Jariwala @wxtca.bsky.social · 07/09/2026

Does HCCA include GDMI this year?

100

Reposted by Deelan Jariwala

James Franklin @franklinjamesl.bsky.social · 06/09/2026

Took a quick look at how Google DeepMind (GDMI) is doing this year for the combined NWS basins AL/EP/CP. With the caveat that ECMWF is not in the data, GDMI for track blows everything else away, beating OFCL by 30% at 5 days! Nothing so dramatic for intensity, but still among the best guidance.

1278

Deelan Jariwala @wxtca.bsky.social · 07/09/2026

One of my favorite TV shows (of the handful I watched over the last few years...)

110

Reposted by Deelan Jariwala

Dr. Larissa @rissarwx.bsky.social · 01/08/2026

Final full descent sounding for this go around. Balloon is going back up and we'll keep doing that loop as long as Dolphin will let us.

25615

Deelan Jariwala @wxtca.bsky.social · 15/07/2026

Probably not a coincidence that thunderstorms are firing on the boundary associated with it, but I'm not sure its represented well on modeling so I have no real plots to show for it

180

Deelan Jariwala @wxtca.bsky.social · 12/06/2026

Half a million people watching a video who's title implies that NOAA/NHC/CSU are wrong about the hurricane season forecast is probably nothing

040

Reposted by Deelan Jariwala

Michael Fischer @mikefischerwx.bsky.social · 02/06/2026

It's the first day of hurricane season in the Atlantic. TC-ATLAS now features a real-time monitoring page with geostationary and microwave satellite data, real-time surface observations, and objective model forecasts from Google DeepMind. michaelfischerwx.github.io/TC-ATLAS/rea...

0217

Deelan Jariwala @wxtca.bsky.social · 16/05/2026

Feel like most of its too niche to find widespread use, which is totally fine because its hosted on github rather than a dedicated server, but I mean if it benefits literally anyone, that's better than no one. It's mostly for my own convenience, anyway. Also, I graduated

010

Deelan Jariwala @wxtca.bsky.social · 16/05/2026

I mostly lurk here (sorry, Twitter is much more entertaining to scroll + group chats), but I think this is a good platform to plug the webpage I had Claude put together to host a lot of the random projects I had lying around on my laptop: tcalert.github.io/cycloweb/

tcalert.github.io

CycloWeb

130

Reposted by Deelan Jariwala

Michael Fischer @mikefischerwx.bsky.social · 27/03/2026

Interested in hurricanes? I want to introduce TC-ATLAS: the Tropical Cyclone Analysis Tool for Live and Archived Structure. Explore live or past storms without writing a line of code or downloading a data file, building on the capabilities of TC-RADAR: michaelfischerwx.github.io/TC-ATLAS/ind...

47734

Deelan Jariwala @wxtca.bsky.social · 18/12/2025

Anecdotally, the GEFS corrected westward with Melissa well-before the GFS did. Not sure how much of a contribution that could be, but I'd imagine its something given the relatively low number of storms.

000

Reposted by Deelan Jariwala

James Franklin @franklinjamesl.bsky.social · 31/10/2025

Here are updated verifications for 2025, AL/EP combined, EMXI not available. Impressed that NHC is beating everything, including consensus, for intensity, and everything but Google DeepMind (GDMI) for track. Rough year for GFS and the statistical intensity aids. Banner year for GDMI.

53613

Reposted by Deelan Jariwala

James Franklin @franklinjamesl.bsky.social · 13/10/2025

With about 80% of the precincts reporting I've seen enough. GDMI - Google DeepMind is going to win the seat for best track model in 2025. The race for best intensity model is still too close to call, but GDMI is right there with the consensus and OFCL. Quite a remarkable campaign.

24116

Reposted by Deelan Jariwala

Philippe Papin @pppapin.bsky.social · 14/08/2025

Yesterday, #Podul moved over #Taiwan as a 95 kt #typhoon. A unique feature seen w/ TCs near Taiwan is a blowup of seemingly cold convection (<-80C) as core moves over the island. Turns out it's mostly thick cirrus w/o much sfc precip due to vertically propagating mountain waves. Diagnosed below⤵️

2328

Deelan Jariwala @wxtca.bsky.social · 14/08/2025

Further west ensemble members have stronger ridging over the eastern half of North America. However, members with stronger ridging in this region also tend to have a more intense system over the southwestern Atlantic! HAFS-A shows that this is a product of flow becoming more WNW with height.

031

Deelan Jariwala @wxtca.bsky.social · 14/08/2025

Are those what became these plots? The TC real-time model products are severely underrated imo

160

Reposted by Deelan Jariwala

Philippe Papin @pppapin.bsky.social · 10/08/2025

I'd like list a couple pet peeves of mine that regular folks & professional mets have done this #hurricane season that is poor form: (1) Naming storms before they are actually named. (2) Sharing deterministic model forecasts (including AI models) beyond 5-7 days. Allow me a mini-rant to explain:🧵

49829

Reposted by Deelan Jariwala

Philippe Papin @pppapin.bsky.social · 27/06/2025

😬 yeah unfortunately ATMS does not have enough horizontal res needed to resolve inner core features in TCs like #Erick. Nice TS Storms post by Naufal Razin shows footprint of SSMIS imagers on overall TC microwave coverage. Microwave coverage peaked in 2015, w/ SSMIS ~60% of current coverage data.

Average 24h microwave coverage of TCs since 1987 with SSMIS contribution in red

34717

Deelan Jariwala @wxtca.bsky.social · 28/04/2025

Aaaand I was immediately blocked for this quote-retweet. Really weird behavior, lol

020

Deelan Jariwala @wxtca.bsky.social · 28/04/2025

These are important constraints and many people discussed these during the previous hurricane season. I would say it's mildly concerning that a scientist would recognize these issues and proceed to use the model regardless as the basis of their forecast, but that's just my opinion.

110

Deelan Jariwala @wxtca.bsky.social · 28/04/2025

With MDR SSTs exceeding virtually every year in the historical record in 2024, it shouldn't be surprising that the model spit out an unusually high value. This is *exacerbated* by AGW, as the model uses mean SSTs out to 2007, as opposed to anomalies!

110

Deelan Jariwala @wxtca.bsky.social · 28/04/2025

This is a pretty strange and uncalled-for reaction to someone making a valid point about that seasonal forecast! A statistical model, as described in Kozar et al. 2012 and used in the UPenn forecast, will tend to struggle when provided with data outside of its training dataset (I.E. extrapolation).

220

Reposted by Deelan Jariwala

Paul Voosen @voosen.me · 11/04/2025

BREAKING from @science.org: The Trump admin is seeking to kill nearly all climate research at NOAA, its climate science agency. Its near-final budget proposal would end all NOAA research labs, academic institutes, and regional climate centers. And it wants to fully end the NOAA Research division.

science.org

Trump seeks to end climate research at premier U.S. climate agency

White House aims to end NOAA’s research office; NASA also targeted

11621351653

Deelan Jariwala @wxtca.bsky.social · 22/03/2025

On the bright side, I guess this is some free extra-early public commentary to help guide future improvements :)

000

Deelan Jariwala @wxtca.bsky.social · 22/03/2025

Out of curiosity -- what's the process of developing these new graphics? Do they come about by recommendations of social scientists primarily or forecasters? I know meteorologist do a lot on the communications side but at the end of the day, that's not really the degree's expertise.

110

Deelan Jariwala @wxtca.bsky.social · 08/03/2025

Do you have a version without the paywall?

020

Deelan Jariwala @wxtca.bsky.social · 29/01/2025

Closest to band 7 today -- it would be cool if you could visualize these with more modern colortables and whatnot but I don't think the actual data files are extant (or at least don't exist in a format that can easily be read)

130

Deelan Jariwala @wxtca.bsky.social · 29/01/2025

Shortwave right? It says 3.4-4.2 microns

120

Deelan Jariwala @wxtca.bsky.social · 29/01/2025

Satellite loop of Hurricane Carmen (1974) rapidly intensifying over the Gulf of Mexico, as well as its landfall and decay over Louisiana. These pictures, taken by SMS-1, are some of the earliest examples of enhanced longwave infrared imagery that I know of.

1102

Deelan Jariwala @wxtca.bsky.social · 29/01/2025

From what I was told, the NHC will be using Gulf of Mexico until given some sort of notice or directive to make the switch to Gulf of America (which they anticipate to be coming in the next few weeks). Not sure what the SPC is doing or if this is just a forecaster going rogue to spark controversy…

030

Deelan Jariwala @wxtca.bsky.social · 24/12/2024

Seems like it comes down to where the NPAC jet deposits a ridge in some regards (is it in the Gulf of Alaska poleward or the Western US/near the coast?). We'll see how this plays out but with the MJO crossing into the Pacific, I'd probably lean towards the ECMWF knowing the GFS's habits here.

010

Reposted by Deelan Jariwala

Philippe Papin @pppapin.bsky.social · 23/12/2024

Original thread worthy of an update to show evolution of this low on IR from 19 to 22 December. An attempt at #tropical #transition was made, but as it shed frontal features, marginal SSTs prolonged length it took for organized deep convection to develop, & window was too short for classification.

1174

Deelan Jariwala @wxtca.bsky.social · 20/12/2024

(This is going to sound really silly) I didn’t know about Lily until a month ago or so when I had a dream about a storm of that sort in the Central Atlantic in December which meandered for quite some time and got rather strong. @1900hurricane.bsky.social helped me identify it as Lily lol

120

Deelan Jariwala @wxtca.bsky.social · 20/12/2024

Right now, home in the Oranges for winter break, but only until the first week of January lol

110

Deelan Jariwala @wxtca.bsky.social · 20/12/2024

And not because I expect this to overperform...

120

Deelan Jariwala @wxtca.bsky.social · 20/12/2024

Fully expecting this to be the largest snowstorm I experience this year

200

Deelan Jariwala @wxtca.bsky.social · 17/12/2024

I’m liking this one a little better than the others I think

120

Deelan Jariwala @wxtca.bsky.social · 17/12/2024

Perhaps of the subtropical nature!

000

Deelan Jariwala @wxtca.bsky.social · 16/12/2024

Probably not likely to go anywhere interesting, but it’ll be interesting to watch one or more cut off lows in the central Atlantic over the next week as they meander over the subtropics south of strong ridging.

171

Deelan Jariwala @wxtca.bsky.social · 15/12/2024

Hello from an alien drone over New Jersey

1150

Deelan Jariwala @wxtca.bsky.social · 07/12/2024

Seems like the surface low has trouble staying beneath the upper trough on most guidance

020

Deelan Jariwala @wxtca.bsky.social · 07/12/2024

The result has been a prolonged period of easterly trade winds over the equatorial Pacific and westerlies over the Indian Ocean. The latter has resulted in significant cooling, as well as multiple tropical cyclones (including some more to come).

010

Deelan Jariwala @wxtca.bsky.social · 07/12/2024

Pretty cool to see the change in the global circulation in response to the strengthening La Niña here on a longer-term VP200 hovmoller. Previously, the MJO signal was pretty robust and circuited the world every month or so. Recently though, it’s been parked over the Indian Ocean.

120

Reposted by Deelan Jariwala

Sang-Ki Lee @sklee621.bsky.social · 30/11/2024

Storm-resolving climate models project a 36% increase in the year-to-year variability of Atlantic hurricane activity by the 2050s 🌊🧪: ocean2climate.org/2024/11/30/s...

36818