Corrections

Date: June 15, 2024

Page: Sea Breeze Mechanics

Original text: "The specific heat capacity of water is approximately 4,186 J/kg·K. Typical dry soil is closer to 800 J/kg·K. This means land heats roughly five times faster than water."

Correction: The ratio of specific heat capacities (4186/800 ≈ 5.2) describes the amount of energy required to raise the temperature of each material by 1 K, not the rate of heating under solar irradiation. The actual heating rate depends on density, thermal conductivity, and solar penetration depth as well. For typical dry soil (density ~1,600 kg/m³) vs. water (density ~1,000 kg/m³), the volumetric heat capacity ratio is closer to 2:1, not 5:1. The text has been revised to clarify that specific heat capacity is only one factor in the differential heating story, and that the full explanation requires considering thermal conductivity and penetration depth as well.

We thank Dr. Yuki Tanaka (University of Tsukuba) for flagging this oversimplification.

Date: May 3, 2024

Page: Timing the Breeze

Original text: "The sea breeze reaches Ginza on roughly 60% of August afternoons. In the 2020s, our data suggests it's closer to 70%."

Correction: The 60% figure for the 1990s was based on a miscount of JMA AMeDAS data. After re-analysis with corrected station filtering, the 1990s frequency at Ginza was approximately 55%, not 60%. The 70% figure for the 2020s remains unchanged. The corrected text now reads: "In the 1990s, the sea breeze reached Ginza on roughly 55% of August afternoons. In the 2020s, our data suggests it's closer to 70%." This changes the implied increase from 10 percentage points to 15 percentage points.

Discovered during internal data audit. No external report.

Date: March 22, 2024

Page: Coastal Wards

Original text: "Odaiba is the coolest point in the 23 wards — often 4-5°C below Shinjuku and 1°C even below Koto."

Correction: The 1°C difference between Odaiba and Koto was based on a single-season dataset (August 2022). Full two-season data (August 2022 + August 2023) shows the mean difference is 0.6°C, not 1°C, with substantial day-to-day variation (standard deviation 0.8°C). The text has been revised to state "up to 1°C below Koto on strong breeze days" to better reflect the variability.

Discovered during seasonal data compilation.

Date: November 10, 2023

Page: Data Sources

Original text: "The Open-Meteo API provides model output at approximately 0.1° (11 km) spatial resolution."

Correction: Open-Meteo's global model resolution is approximately 0.25° (~25 km), not 0.1°. The 0.1° figure incorrectly referred to a higher-resolution nested domain that is not consistently used for Tokyo. The text has been corrected to state 0.25° (~25 km).

Flagged by a reader via the contact form. Thank you.

Date: September 5, 2023

Page: Sea-Land Front

Original text: "On August 12, 2023, the front crossed our Ginza sensor at 14:23 and reached Shimbashi at 14:41 — an 18-minute transit time, implying an advance speed of roughly 1.8 m/s."

Correction: The Shimbashi timestamp of 14:41 was a data logging error — the logger at Shimbashi experienced a 6-minute clock drift that was not corrected in the initial analysis. The corrected timestamp is 14:47, making the transit time 24 minutes and the advance speed 1.4 m/s. The narrative has been updated with the corrected timing.

Discovered during post-season data quality review.

Date: July 18, 2023

Page: About

Original text: "Marina cycled the entire Tokyo Bay coastline — all 180 kilometers from Choshi to Uraga — with a handheld anemometer."

Correction: The actual distance cycled was 174 kilometers, not 180. The 180 km figure included a 6-km detour to retrieve a forgotten data logger that was not part of the planned coastal route. The text has been corrected to "approximately 175 kilometers" to reflect the actual cycling distance.

Marina caught this herself while reviewing her Strava log for a talk.