The entire country is grimacing under an autumn rainy spell. Crop damage, in particular, is growing. In Hongseong, South Chungcheong, famous for garlic, 519 mm of rain has fallen since last month. That is twice the average autumn precipitation. A Japanese research team has found scientific evidence for the autumn rainy season.
A team led by Tetsuya Takemi of Kyoto University said on the 16th in the international journal Scientific Reports that the physical law that "heavy rain intensifies as temperatures rise" shows up clearly in actual climate data.
Simply put, the warmer the air, the more water vapor it can hold, which increases the likelihood of rain pouring down all at once. This finding is meaningful in that it scientifically proves in numbers the common-sense experience that "it rains more on hot days."
The law the team verified is called the Clausius–Clapeyron relationship. According to it, when the temperature rises by 1 degree Celsius, the amount of water vapor that air can hold increases by about 7%. This phenomenon, known as the "7% rule," appears most distinctly at temperatures around 25 degrees.
However, until now, there had been few studies that precisely examined how much and in what way water vapor changes in each layer of the atmosphere. Most studies looked only at the water vapor right above the ground or the average for the entire atmosphere, and they could not properly analyze changes by altitude due to a lack of measurement data.
Also, because commonly used relative humidity varies greatly depending on the maximum amount of water vapor air can hold, even a small temperature error reduces accuracy. The team instead used specific humidity, which indicates the actual amount of water vapor in 1 kg of air, to analyze how changes in water vapor affect heavy rainfall.
The team divided Japan into seven regions including coastal and inland areas, and analyzed observational data from 1951 to 2010 and future climate projection data from 2051 to 2110. They used the Asian Precipitation—Highly-Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) and Japan's Database for Policy Decision-Making for Future Climate Change (d4PDF), increasing reliability through 50 and 90 repeated experiments, respectively.
The analysis found that even in the current climate, most regions show a tendency for heavy rainfall intensity to increase as temperatures rise. In a future climate scenario assuming an average temperature increase of 4 degrees, this relationship becomes clearer, with heavy rainfall intensity increasing by an average of more than 7% per degree. This suggests a greatly increased likelihood of "water bombs" across all of East Asia, including the Korean Peninsula, not just Japan.
To identify the cause more specifically, the team revalidated the distribution of atmospheric water vapor from 1980 to 2022 using NASA's Modern-Era Retrospective analysis for Research and Applications, version 2 (MERRA-2). They found that in the lower atmosphere about 1.5 km above ground, water vapor increases the most as temperatures rise. In this layer, the amount of water vapor increased by an average of about 8.3% for every 1-degree rise in temperature. At higher altitudes, this relationship barely appeared.
The team explained that this study shows that extreme heavy rainfall due to climate change will be commonly intensified across Japan. In particular, as warm and humid days with sufficient water vapor become more frequent, the frequency and scale of heavy rains and floods will both grow, they projected.
Takemi said, "This analysis is the first study to quantitatively show the vertical structure of atmospheric dynamics that causes extreme precipitation," adding, "It will serve as important scientific grounds for designing climate adaptation plans tailored to regional characteristics."
Because this study is based on medium-resolution data with a 20 km grid, the team plans to elucidate the mechanism of heavy rainfall by combining more precise climate models with cloud microphysics data. Cloud microphysics data precisely show how many tiny water droplets and ice crystals are gathered within clouds and how they move.
References
Scientific Reports (2025), DOI: https://doi.org/10.1038/s41598-025-22287-6