Why Linear Rainbands Form: Simulating the Drivers of Disastrous Downpours

Fascinated by his course “Theory of Environment Sensing Technology,” I decided to join Prof. Yoshiaki Miyamoto’s Seminar on meteorology in the Fall Semester of my second undergraduate year. I have always been interested in environmental issues, and I find it fascinating to explore global environmental challenges through the lens of everyday natural phenomena. 

Rainfall, a phenomenon familiar to us, caught my attention after I joined the Seminar. I became particularly drawn to linear rainbands, which account for about two-thirds of all extreme rainfall. My interest became personal when I got caught in a heavy downpour caused by a linear rainband during a trip to Kumamoto. That experience fueled my determination to uncover the mechanisms behind how they form and why they stall.

Thanks to enhanced observational technology and more accurate numerical weather models, we can now predict linear rainbands more precisely than before. With the phenomenon frequently featured in the media lately, and given concerns about the rising frequency of weather disasters driven by climate change, improving forecasting accuracy has become an urgent social issue. I find it deeply rewarding to work on research with the chance to solve problems as closely related to our daily lives as rainfall.

The goal of my research is to figure out what triggers and sustains linear rainbands so we can help improve forecasting accuracy. To this end, one thing I look at is the interaction between atmospheric instability—how easily convection occurs—and water vapor flux convergence.

Water vapor flux convergence indicates how much water vapor increases or decreases due to wind convergence, the phenomenon that occurs when air flowing from different directions converges into a specific area. Assuming the law of conservation of mass, wind convergence forces the air upwards, creating an updraft. The key here is to observe the movement of water vapor along with the wind movement. Water vapor flux convergence is a metric that captures how much water vapor gets carried upward by wind. I study it closely as it plays an important role in analyzing how water vapor condenses in the upper atmosphere and forms clouds.

While water vapor flux convergence is known to have a strong correlation with rainfall, its connection with linear rainbands remains unclear. Understanding how it contributes to the development and persistence of linear rainbands will lead to better forecasting and ultimately help prevent weather disasters.

Numerical simulations using weather models are crucial for uncovering the mechanisms behind linear rainbands. Because these models solve governing equations on a computer using observational data, they can calculate weather variables that physical observations alone cannot capture. They also allow us to finely tune spatial and temporal resolutions to match the specific phenomenon we are studying. This flexibility is vital for analyzing linear rainbands; they are difficult to reproduce without setting spatial resolution precisely, and they form and persist as a result of mixed weather factors, including variables we can’t derive from observational data. In my study, I simulated and reproduced the atmospheric environment of the linear rainband that occurred in August 2021 using SCALE, a weather model developed through a RIKEN initiative that Prof. Miyamoto was a part of. 

Predicting linear rainbands is extremely difficult. The accuracy rate as of 2024 is only about 10% at the prefectural level. This is because numerical prediction models have low resolution, and also because we still don’t know many aspects of how linear rainbands develop and persist. Recently, six quantitative conditions for predicting linear rainbands have been introduced. Nevertheless, reducing the frequent false alarms—cases where a linear rainband is predicted but does not develop—remains a major challenge.

The dilemma is that if we tighten the criteria to reduce false alarms, we risk more missed forecasts, cases where a linear rainband occurs despite not being predicted. Because the phenomenon is triggered by interconnected factors, pinpointing the exact conditions is inherently challenging.

Currently, I’m testing a hypothesis that the conditions to form typhoons are applicable to linear rainbands. During my undergraduate years, this research was selected for the Yamagishi Student Project Support Program for two years. Research grants, like Taikichiro Mori Memorial Research Grants, are also available in the Graduate School of Media and Governance. Being able to independently drive research forward while receiving financial support is a major advantage of studying at SFC.

Alongside my research, I work part-time at a weather startup where Prof. Miyamoto serves as CTO. Although the business focuses on aviation, I can apply insights from my research to the work. Gaining real-world, practical experience that cannot be acquired through research alone is invaluable to shape my future career.

While my life currently revolves around research, I didn’t start out aspiring to be a researcher back in high school. I chose SFC because I wanted to learn a wide variety of things without being restricted by the traditional boundaries between the humanities and sciences. SFC is an excellent environment even for those who haven't decided on a specific path yet. The main reason I joined Prof. Miyamoto’s Seminar in the first place was simply because he is such a kind person. Right now, I’m really enjoying finding issues and digging into them at my own pace.

Met Lab.

Research Fields: Meteorology and Climate Science