FOSS4G-Asia 2023 Seoul

Estimation of Vineyard Suitable Areas Using Machine Learning and Data PNG
11-30, 15:10–15:30 (Asia/Seoul), Seoul Archive

The quality of wine grapes is heavily influenced by environmental conditions such as geology, meteorology, geomorphology, and more. Therefore, it's necessary to consider these factors when selecting the location of a vineyard and the specific variety of wine grapes to be cultivated. However, these selections are often based on the experience of the farmers, making it a barrier for newcomers and raising concerns about potential economic losses from inappropriate choices of vineyard locations and varieties. In this research, we attempt to estimate suitable areas for wine grape cultivation using machine learning, utilizing the distribution of vineyards as supervised data. The evaluation areas include the surroundings of Ueda, Toumi, and Saku City in Nagano Prefecture. The locations of vineyards were identified through field surveys and interview surveys, and the data were aggregated on a 3rd level Japanese Standard Grid. The environmental factors used for evaluation include geology, slope angle, slope direction, average annual temperature, lowest annual temperature, and highest annual temperature. These factors are publicly available in Data PNG format, and a Python script was created to aggregate these data at the 3rd Standard Grid. By incorporating the spatial distribution of existing vineyards as the supervised data and analyzing various environmental factors, this approach aims to provide a more systematic and accessible way to identify optimal vineyard locations, thereby reducing reliance on individual farmer experience.

Nobusuke Iwasaki is actively engaged in research related to land use changes and Satoyama landscapes, utilizing FOSS4G. He is also involved in initiatives related to open data, making research findings accessible to a wide audience. He works on the dissemination of FOSS4G to efficiently utilize such data. As a part of his achievements, he has made public systems, such as the Historical Agro-Environment Browsing System and Vineyards The Vineyard Suitability Assessment System.