What does it really mean to be strong? Shaun Robinson, CEO of the Mental Health Foundation NZ, joins me to reflect on masculinity, compassion, and the quiet courage of authenticity. 🎧 Watch the full episode www.youtube.com/watch?v=Aifs...
youtube.com/shorts/3Yote...
youtube.com/shorts/3Yote...
Shaun Robinson Rethinking Strength — A Conversation on Masculinity and Authenticity
YouTube video by Authentic Leadership Podcast
youtube.com
October 29, 2025 at 2:52 AM
What does it really mean to be strong? Shaun Robinson, CEO of the Mental Health Foundation NZ, joins me to reflect on masculinity, compassion, and the quiet courage of authenticity. 🎧 Watch the full episode www.youtube.com/watch?v=Aifs...
youtube.com/shorts/3Yote...
youtube.com/shorts/3Yote...
📢The Anemoi training-ready ERA5 dataset is now openly available!
A ready-to-use, ML-optimised subset of ERA5 in cloud-friendly Zarr format, hosted by ECMWF and free under CC-BY-4.0.
Read the AIFS blog: ➡️ www.ecmwf.int/en/about/med...
A ready-to-use, ML-optimised subset of ERA5 in cloud-friendly Zarr format, hosted by ECMWF and free under CC-BY-4.0.
Read the AIFS blog: ➡️ www.ecmwf.int/en/about/med...
October 28, 2025 at 12:33 PM
📢The Anemoi training-ready ERA5 dataset is now openly available!
A ready-to-use, ML-optimised subset of ERA5 in cloud-friendly Zarr format, hosted by ECMWF and free under CC-BY-4.0.
Read the AIFS blog: ➡️ www.ecmwf.int/en/about/med...
A ready-to-use, ML-optimised subset of ERA5 in cloud-friendly Zarr format, hosted by ECMWF and free under CC-BY-4.0.
Read the AIFS blog: ➡️ www.ecmwf.int/en/about/med...
18z Euro AIFS is farther west with Melissa. I think this is the lowest pressure I've ever seen on the AIFS.
October 25, 2025 at 12:20 AM
18z Euro AIFS is farther west with Melissa. I think this is the lowest pressure I've ever seen on the AIFS.
In this moving conversation, Shaun Robinson, Chief Executive of the Mental Health Foundation of NZ, opens up about living with bipolar disorder, leadership, and the courage to be truly oneself. Full podcast visit www.youtube.com/watch?v=Aifs...
October 23, 2025 at 7:08 PM
In this moving conversation, Shaun Robinson, Chief Executive of the Mental Health Foundation of NZ, opens up about living with bipolar disorder, leadership, and the courage to be truly oneself. Full podcast visit www.youtube.com/watch?v=Aifs...
Over the past 30 days, Brazil #soybean regions have been substantially dry, with prod-wghtd precip only 82% of nrml. The 14-day AIFS, ECMWF, & GEFS fcst below nrml #Brazil soybean prod wghtd precip.
AIFS: 73% of nrml
ECMWF: 82% of nrml
GEFS: 63% of nrml
#oatt #AgWx #AgWeather
AIFS: 73% of nrml
ECMWF: 82% of nrml
GEFS: 63% of nrml
#oatt #AgWx #AgWeather
October 23, 2025 at 12:19 PM
Over the past 30 days, Brazil #soybean regions have been substantially dry, with prod-wghtd precip only 82% of nrml. The 14-day AIFS, ECMWF, & GEFS fcst below nrml #Brazil soybean prod wghtd precip.
AIFS: 73% of nrml
ECMWF: 82% of nrml
GEFS: 63% of nrml
#oatt #AgWx #AgWeather
AIFS: 73% of nrml
ECMWF: 82% of nrml
GEFS: 63% of nrml
#oatt #AgWx #AgWeather
Uncertainty is high, but with only two to three days to prepare we must consider the potential for a tropical cyclone hitting Hispaniola this week. Sixty percent of models this morning indicate this. See DeepMind AI (FNV3), American GFS, European AI (AIFS), and US Navy. 1/
October 20, 2025 at 8:18 PM
Uncertainty is high, but with only two to three days to prepare we must consider the potential for a tropical cyclone hitting Hispaniola this week. Sixty percent of models this morning indicate this. See DeepMind AI (FNV3), American GFS, European AI (AIFS), and US Navy. 1/
Since the default GEFS tracks from the NHC server continue to struggle identifying ensemble tracks for invest 98L, I swapped it for my in-house tracker that I use for the ICON & AIFS ensembles on my website:
polarwx.com/tropical/?st...
polarwx.com/tropical/?st...
October 20, 2025 at 12:11 AM
Since the default GEFS tracks from the NHC server continue to struggle identifying ensemble tracks for invest 98L, I swapped it for my in-house tracker that I use for the ICON & AIFS ensembles on my website:
polarwx.com/tropical/?st...
polarwx.com/tropical/?st...
My latest changes in the website were more "under the hood" - I revamped the tracking algorithm to better capture tropical waves before they become tropical cyclones, which is applied to both AIFS and ICON ensembles.
October 19, 2025 at 1:04 AM
My latest changes in the website were more "under the hood" - I revamped the tracking algorithm to better capture tropical waves before they become tropical cyclones, which is applied to both AIFS and ICON ensembles.
Technical note — neither ICON nor the AIFS had many members identify Tropical Storm Lorenzo with my strictly MSLP-based algorithm, so I switched to a 10m vorticity & MSLP tracking algorithm. This now picks up much more members than the previous method.
October 14, 2025 at 2:09 AM
Technical note — neither ICON nor the AIFS had many members identify Tropical Storm Lorenzo with my strictly MSLP-based algorithm, so I switched to a 10m vorticity & MSLP tracking algorithm. This now picks up much more members than the previous method.
Your next thought might be to look at AI weather models, since those have certain advantages over NWP models.
But even they are experiencing the same whiplash - notice how the AIFS over the last few runs keeps alternating between cutoff & no cutoff low in the Great Lakes:
But even they are experiencing the same whiplash - notice how the AIFS over the last few runs keeps alternating between cutoff & no cutoff low in the Great Lakes:
October 8, 2025 at 1:14 AM
Your next thought might be to look at AI weather models, since those have certain advantages over NWP models.
But even they are experiencing the same whiplash - notice how the AIFS over the last few runs keeps alternating between cutoff & no cutoff low in the Great Lakes:
But even they are experiencing the same whiplash - notice how the AIFS over the last few runs keeps alternating between cutoff & no cutoff low in the Great Lakes:
Mistakes happen - my ECMWF ensemble data feed accidentally replaced the AIFS ensemble data last night. This has now been fixed & I backfilled all AIFS ensemble & deterministic plots back to 10/4.
ICYMI: AIFS ensemble TC tracks are now live on PolarWx: polarwx.com/tropical
ICYMI: AIFS ensemble TC tracks are now live on PolarWx: polarwx.com/tropical
October 7, 2025 at 12:04 AM
Mistakes happen - my ECMWF ensemble data feed accidentally replaced the AIFS ensemble data last night. This has now been fixed & I backfilled all AIFS ensemble & deterministic plots back to 10/4.
ICYMI: AIFS ensemble TC tracks are now live on PolarWx: polarwx.com/tropical
ICYMI: AIFS ensemble TC tracks are now live on PolarWx: polarwx.com/tropical
The GFS, Euro and Euro AIFS are in good agreement regarding the east coast storm potential for this weekend. While it will likely be non-tropical (hence no mention in the NHC outlook), it will effectively have tropical storm conditions along the Carolina coast.
October 7, 2025 at 12:13 PM
The GFS, Euro and Euro AIFS are in good agreement regarding the east coast storm potential for this weekend. While it will likely be non-tropical (hence no mention in the NHC outlook), it will effectively have tropical storm conditions along the Carolina coast.
The 12z Euro AIFS continues to bring the system up the east coast this weekend.
October 6, 2025 at 6:37 PM
The 12z Euro AIFS continues to bring the system up the east coast this weekend.
Continuing down my tropical website checklist, AIFS ensemble tracks are now available on my website where storms can be identified & tracked in the model data.
Scroll over to the "Ensembles" tab for each storm to view AIFS ensemble maps if available:
polarwx.com/tropical/
Scroll over to the "Ensembles" tab for each storm to view AIFS ensemble maps if available:
polarwx.com/tropical/
October 6, 2025 at 3:05 AM
Continuing down my tropical website checklist, AIFS ensemble tracks are now available on my website where storms can be identified & tracked in the model data.
Scroll over to the "Ensembles" tab for each storm to view AIFS ensemble maps if available:
polarwx.com/tropical/
Scroll over to the "Ensembles" tab for each storm to view AIFS ensemble maps if available:
polarwx.com/tropical/
An interesting artifact of some AI emulator models I just noticed - the AIFS seems to leave a "trail" of lower MSLP and geopotential heights in the wake of Hurricane Humberto's track, which lingers at those gridpoints long after the storm has passed:
September 27, 2025 at 2:39 PM
An interesting artifact of some AI emulator models I just noticed - the AIFS seems to leave a "trail" of lower MSLP and geopotential heights in the wake of Hurricane Humberto's track, which lingers at those gridpoints long after the storm has passed:
Partly since I’m short on free time to do these all at once, and partly out of an interest in user feedback, I’ll put this out as a poll — what feature should I add to my tropical realtime site next?
1) full superensemble viewer
2) AIFS ensemble tracks
3) recon data viewer
1) full superensemble viewer
2) AIFS ensemble tracks
3) recon data viewer
September 25, 2025 at 12:42 AM
Partly since I’m short on free time to do these all at once, and partly out of an interest in user feedback, I’ll put this out as a poll — what feature should I add to my tropical realtime site next?
1) full superensemble viewer
2) AIFS ensemble tracks
3) recon data viewer
1) full superensemble viewer
2) AIFS ensemble tracks
3) recon data viewer
So the models seem to think there may be a tropical cyclone or two in the Atlantic this coming week.
September 24, 2025 at 1:15 AM
So the models seem to think there may be a tropical cyclone or two in the Atlantic this coming week.
12z Euro AIFS ensembles are continuing to show development chances in both the NW Caribbean and SW Atlantic.
September 20, 2025 at 10:32 PM
12z Euro AIFS ensembles are continuing to show development chances in both the NW Caribbean and SW Atlantic.
Latest Euro AIFS ensembles are shown below.
September 20, 2025 at 1:13 PM
Latest Euro AIFS ensembles are shown below.
🌡La très nette hausse de la température de la masse d'air à Paris (jusqu'à 20 °C à 850 hPa) devrait être suivie d'une chute probablement spectaculaire en début de semaine prochaine avec 15 °C de moins et un temps passagèrement pluvieux. C'est ce qu'indique en tout cas l'ensemble AIFS @ecmwf.int
September 17, 2025 at 7:47 AM
🌡La très nette hausse de la température de la masse d'air à Paris (jusqu'à 20 °C à 850 hPa) devrait être suivie d'une chute probablement spectaculaire en début de semaine prochaine avec 15 °C de moins et un temps passagèrement pluvieux. C'est ce qu'indique en tout cas l'ensemble AIFS @ecmwf.int
Its not the first time models teased this but now may be the time for a substantial #cyclone in the Australian region. EPS in particular has near unanimous support and all major global models + AIFS are on board. Currently not really expecting a landfall but we will see #tropicswx
January 15, 2025 at 10:08 AM
Its not the first time models teased this but now may be the time for a substantial #cyclone in the Australian region. EPS in particular has near unanimous support and all major global models + AIFS are on board. Currently not really expecting a landfall but we will see #tropicswx
FYI Experimental: AIFS (ECMWF) ML model, 2nd Feb
January 18, 2025 at 10:06 PM
FYI Experimental: AIFS (ECMWF) ML model, 2nd Feb
My colleagues published the AIFS ensemble forecast! 📉🌤️
Learn about AIFS-CRPS, an ensemble forecasting model trained using the Continuous Ranked Probability Score.
Learn about AIFS-CRPS, an ensemble forecasting model trained using the Continuous Ranked Probability Score.
My colleagues published the AIFS ensemble forecast! 📉🌤️
It’s pretty neat for precision in forecasting, blending statistical rigor and domain knowledge with cutting-edge AI techniques!
amplt.de
April 29, 2025 at 3:58 PM
My colleagues published the AIFS ensemble forecast! 📉🌤️
Learn about AIFS-CRPS, an ensemble forecasting model trained using the Continuous Ranked Probability Score.
Learn about AIFS-CRPS, an ensemble forecasting model trained using the Continuous Ranked Probability Score.