Scientists estimate sunlight in 18th and 19th century Tokyo using historical diaries

The amount of sunlight, or solar radiation, that a location receives makes a big impact on weather conditions, crop success, rainfall and overall climate trends. Today, instruments called pyrheliometers are used to carefully calculate how much sunlight occurs at a particular location, but these instruments weren’t available to quantify sunlight until 1838, and automated pyrheliometers weren’t invented until the early 20th century.

Without these measurements, scientists don’t have accurate, consistent solar radiation records for any location prior to the early 1900s. Despite this, climate scientists could generate a wealth of information by correlating local changes in solar radiation with downstream events, such as crop failure or changes in longer-term weather patterns.

To address this lack of pre-20th century solar radiation records, scientists from the Research Organization of Information and Systems (ROIS), Rissho University and Tokyo Metropolitan University used historical diaries to estimate sunlight levels in Tokyo long ago. In doing so, the team identified long-term changes in solar radiation over time and attempted to correlate these changes with climate and other societal impacts that occurred in Tokyo during the same time frame.

The researchers published their study on December 1 in the journal Climatic Change.

“Although historical diaries contain daily weather descriptions, these qualitative records have rarely been converted into quantitative meteorological variables at monthly resolution. In this study, we developed and validated a method to reconstruct long-term solar radiation using daily weather descriptions from historical diaries and established a continuous solar radiation record for Tokyo extending back to 1720,” said Mika Ichino, project researcher at the Center for Open Data in the Humanities in the Joint Support-Center for Data Science Research at ROIS in Tokyo,  Japan and first author of the research paper. 

The team used daily qualitative records describing weather from historical diaries in Tokyo to construct an estimate of local solar radiation. Daily weather diaries were customary in Japan as far back as the eighth century, and were often kept by aristocrats, officials and literate farmers. Many diaries have since been digitized in the Historical Weather Database of Japan (HWDB), providing 60 published daily weather records accessible to scientists.

The researchers standardized weather descriptions in historical diaries and Japan Meteorological Agency (JMA) official sunshine duration records, which date back to the 1870s, providing approximately 140 years of reliable data. The team then reconstructed the sunshine levels in Tokyo from 1720 to 1912, a time period that overlaps the Little Ice Age.
“We demonstrate[d] that qualitative weather descriptions in historical diaries can be systematically translated into quantitative solar radiation estimates with high monthly-scale reliability. Using this approach, we reconstructed approximately 300 years of solar radiation variability for Tokyo, revealing significant low-insolation episodes during the 1780s and 1830s that coincided with severe famines,” said Ichino.

Low-insolation episodes are characterized by significant reductions in solar radiation, which can have a profound effect on crop productivity, weather patterns and climate. The team used the Ishikawa Diaries to generate their sunlight estimates and validated the quantification of the diary data by comparing it to official JMA records. The two long-term fluctuations in solar radiation that correlated with famine were presumably due to cool summers and decreased agricultural productivity. 

Beyond the confines of Tokyo, the research team sees additional uses for their quantitative modeling of qualitative historical weather data. “Our ultimate goal is to build a spatially and temporally extended dataset of reconstructed solar radiation and to clarify how seasonal and interannual climate variability influenced agricultural production and societal change in preindustrial Japan. More broadly, we aim to provide a methodological framework that can be applied to historical documents worldwide to better understand long-term climate–society dynamics,” said Ichino.

Kooiti Masuda from the Geo-Environmental Science Department at Rissho University in Kumagaya, Japan and Takehiko Mikami from Tokyo Metropolitan University in Tokyo, Japan also contributed to this research.

This study was supported by JSPS KAKENHI (grant numbers 17540410, 18H03794, 20K01152,
21H03776, 22H04938, 25K04619) and ROIS-DS-JOINT (grant numbers 032RP2020, 027RP2021, 031RP2021, 041RP2022, 043RP2022, 044RP2023, 060RP2023).

About the Research Organization of Information and Systems (ROIS)
ROIS is a parent organization of four national institutes (National Institute of Polar Research, National Institute of Informatics, the Institute of Statistical Mathematics and National Institute of Genetics) and the Joint Support-Center for Data Science Research. It is ROIS's mission to promote integrated, cutting-edge research that goes beyond the barriers of these institutions, in addition to facilitating their research activities, as members of inter-university research institutes.