1 · The argument
A threshold, a flat top, and an objection
Past 28.25 °C, warmth stops mattering
The test is too weak to tell
Re-measure, then test out of sample
2 · The data
Every Atlantic track, coloured by the water beneath it
The week-before sea-surface temperature under every named Atlantic storm, 1982–2026
3 · The replication
Their 1982–2005 result holds up on a better ocean
Each storm's peak wind against the warmest water it met on the way there
One dot per storm. Warmest water = highest 1° cell, week-before OSTIA SST at any 6-hourly position up to the peak (MKD's definition). Lines: least-squares fits above 28.25 °C for each era. Hover a dot for the storm.
4 · Out of sample
Twenty-one more seasons tilt the flat top
How fast peak wind rises with SST among storms that met at least a given temperature
Least-squares slope of peak wind (m s⁻¹ per °C) on the warmest water met, for storms whose warmest water was at or above each threshold. Filled markers: p < 0.05. Hover for N and p.
5 · The threshold
Warm water is close to a requirement, and there is more of it
The 28.25 °C pool grew, and the majors that never reached it
More storms reach warm water; no larger share of them becomes major
Storms whose warmest water met was ≥ 28.25 °C. Open diamonds: the values MKD published from weekly OISST for their two eras.
6 · Emanuel's objections, tested
Four objections, one per panel
If peaks tracked potential intensity exactly, how often would MKD's test notice?
Hours from the warmest water met to the peak
All storms, 1982–2026, in 6-hour bins. Green: peak within 18 h of the warmest water, about Emanuel’s response time. Hover a bar for its count.
Where majors met their warmest water, and where they peaked
How much potential intensity rises per degree of SST
ERA5 monthly potential intensity (kt per °C). Across space: along the storm tracks, in three SST bands. Through time: the Aug–Oct mean over the main development region, 1979–2025.
Main-development-region SST and Atlantic power dissipation, smoothed
Aug–Oct ERSST v5 over 6–18°N, 20–60°W, and the Atlantic power dissipation index, both smoothed 1-3-4-3-1 (Emanuel's filter) and shown as standard scores on one axis. Shaded: years after Emanuel's comment.
7 · What it adds up to
Both sides were partly right, and the record since 2005 decides between them
Method & data
How the numbers were made
References
- Michaels, P. J., P. C. Knappenberger and R. E. Davis (2006). Sea-surface temperatures and tropical cyclones in the Atlantic basin. Geophys. Res. Lett., 33, L09708. PDF
- Emanuel, K. (2007). Comment on "Sea-surface temperatures and tropical cyclones in the Atlantic basin" by Patrick J. Michaels et al. Geophys. Res. Lett., 34, L06702. PDF
- Michaels, P. J., P. C. Knappenberger and R. E. Davis (2007). Reply to comment by Kerry Emanuel. Geophys. Res. Lett., 34, doi:10.1029/2006GL027527 (not yet reviewed here).
- Good, S. et al. (2020). The current configuration of the OSTIA system for operational production of foundation sea surface temperature and ice concentration analyses. Remote Sensing, 12, 720.
- Huang, B. et al. (2017). Extended Reconstructed Sea Surface Temperature, version 5 (ERSSTv5). J. Climate, 30, 8179–8205.
- Lin, J., R. Rousseau-Rizzi, C.-Y. Lee and A. Sobel (2023). An open-source, physics-based, tropical cyclone downscaling model with intensity-dependent steering (FAST). J. Adv. Model. Earth Syst., 15.