Prediction of Soil Carbon Contents Using Smartphone Images and Multiple Regression Analysis

Abstract

Soil carbon is an important factor in the process of mitigating climate change and solving greenhouse gas problems. However, the previous technology for soil carbon content analysis required a lot of labor, time, and expensive equipment (i.e. an elemental analyzer). In this study, the disadvantages of previous analysis method were secured by using smartphone images and multiple regression analysis. To predict the soil carbon content, the color variables (e.g., RGB, CIE-L * a * b * , CIE-L * c * h * , and CIE-L * u * v * ), gravimetric water content, and bulk density were used as statistical data. After Pearson's correlation analysis, several variables that had high correlations were removed and then used. In addition, the result of variance inflation factor (VIF) analysis indicated that all variables should not cause multicollinearity problems. The predictive model was classified based on land use, and the predictive model for the entire sample was also derived. The adjusted coefficient of determination (Adj. R 2 ), root mean square error (RMSE), and mean absolute percentage error (MAPE) were used to verify the predictive model. When the verification method was substituted for each predictive model, the reliability of the predictive model classified based on land use was high. Therefore, in order to predict the carbon content in the agricultural soil, it is efficient to assign each prediction model after classifying agricultural land.

Quantifying soil organic matter for sustainable agricultural land management with soil color and machine learning technique

“…These findings indicate that an increase in the SOM content is associated with a decrease in the brightness and hue of soil color and an increase in the saturation of soil color. These results are consistent with previous studies, which have also emphasized the significant relationship between soil color and SOM (Fu et al, 2020;Kang et al, 2022;Ma et al, 2017;Sonn et al, 2022;J. Yang et al, 2021).…”

“…Yang et al, 2021). Additionally, previous studies have provided supportive evidence for the relationship between soil color and SOM (Fu et al, 2020;Heil et al, 2022;Kang et al, 2022;Sonn et al, 2022;Yang et al, 2021).…”

“…This study considered the wavelength range of 620-750 nm as the basis for the R value, while the G and B values corresponded to wavelength ranges of 495-570 nm and 450-475 nm, respectively, as proposed in Fu et al (2020). The averaged values of the R, G, and B parameters were transformed into color indices, such as B I , H I , and S I , as described in Table 1 (Fu et al, 2020;Heil et al, 2022;Kang et al, 2022;Sonn et al, 2022;Wu et al, 2017). Table 1 represents a brief description of the color indices and conversion formulas.…”

“…Soil organic matter (SOM), which comprises organic C, serves as a pivotal environmental factor crucial in assessing terrestrial ecosystem services (e.g., soil health, soil microorganism community) and striving for sustainable land management (Heil et al, 2022;Lal, 2016Lal, , 2020. Soil C has been recognized as a key factor by South Korea and the Sustainable Development Goals for 2023 in addressing climate change and combating major threats to soil degradation and food security (Castaldi et al, 2019;Kang et al, 2022;Sonn et al, 2022). Furthermore, the IPCC (2023) report described soil as a huge C storage, indicating that soil C sequestration is an important agenda for climate change mitigation.…”

“…Soil color acts as an indicator for various soil physical and chemical properties owing to its reliance on different soil attributes (Sonn et al, 2022). Notably, SOM is one of the parameters that can influence soil color, often resulting in darker soil color owing to the presence of concealed mineral elements (Kang et al, 2022;J. Yang et al, 2021).…”