Prediction Equations of Phytoncide Concentration in Korean Pine (Pinus Koraiensis) Forest

Abstract

TPS 691: Methods of measurement, design and data analysis, Exhibition Hall, Ground floor, August 28, 2019, 3:00 PM - 4:30 PM Background: What is drawing attention is the development of technology to create forest as a healing space. Phytoncide is a representative forest healing factor and helps to improve human health and immunity, and is absorbed by the human body through respiratory system or skin in the forest atmosphere. In this study, the prediction model of phytoncide concentration was developed to identify the characteristics of stand density and to simulate phytoncide concentration in Korean pine forest to increase the occurrence of phytoncide beneficial to the human body. Methods: The collection of phytoncide was carried out every month until Dec. 2017 in Korean pine forest (Pocheon, Korea), and the measurement time was selected as 5th/day considering the visiting hours of the healing forest. Phytoncide sampling was performed using a collection set consists of an air pump and a Tenax TA tube. Micrometeorology elements (air temperature, relative humidity, wind and soil condition, solar radiation, and photosynthetic effective radiation (PAR)) were measured to secure basic environmental data in the target area simultaneously. Correlation analysis and multiple regression analysis were performed to develop a phytoncide prediction model using the collection data. Results: The early model showed that phytoncide concentration increased with increasing air temperature, PAR and relative humidity, and decreased with increasing wind speed, solar radiation and soil temperature(R2=0.61). Also, we proposed the simple model using three factors(air temperature, wind speed, relative humidity) which have high influence on concentration(R2=0.51) considering the over fitting the data and economics at the site. It is confirmed that the regression formula can be utilized because neither of the two equations has a clear tendency to find a residuals. Conclusions: These results contribute to predict the concentration of forest environment factors in connection with the mountain meteorology observation system and will be available as part of the forest forecast system in the future.