Forest Area and Carbon Stock in Living Biomass Relationship Analysis Through Clustering and Polynomial Regression Methods

This study shows the relationship between forest area and carbon stock in living biomass across 15 major CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emitting countries during 2010 to 2020. The testing of assumptions revealed a right skewed distribution in the data leading to the adoption of an analytical approach combining K-Means clustering and nonparametric analysis. Data from the Food and Agriculture Organization (FAO) is analyzed to understand forest carbon sequestration patterns. Polynomial regression shows a strong fit (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.9763) despite distribution challenges with model complexity. The K-Means clustering provides complementary information by identifying four different groups of countries based on their forest area and carbon storage characteristics (Dunn Index = 0.6859, Silhouette Score = 0.83). The clustering approach which does not rely on distributional assumptions effectively categorized countries with similar forest carbon relationships revealing significant variations in carbon storage capacity among different countries. The finding illustrates that Brazil and Russia show a notable high carbon sequestration potential due to their extensive forest coverage. This research provides valuable information through a robust analytical framework that accounts for the characteristics of the data.