Analyzing the Inconsistencies Among Five Chinese Forest Age Data Sets and Producing an Integrated Data Set

Abstract Accurate forest age data is essential for carbon stock quantification, ecosystem monitoring, and sustainable forest management. However, significant inconsistencies persist among existing forest age products in China, undermining the reliable assessment of forest dynamics and evidence‐based management strategies. Leveraging China's Ninth National Forest Inventory field data (2912 plots), we conducted a comparative analysis of five forest age data sets and evaluated the accuracy of fused forest age data set. Our findings reveal that 37.9% of forested areas show substantial discrepancies and 15.7% exhibit extreme discrepancies. Although forest cover in extremely high mountain regions is limited, data set inconsistencies are particularly pronounced. Additionally, areas of significant divergence are predominantly concentrated in mid and high mountainous regions, where major natural forests are distributed. Then we evaluated the performance of random forest (RF) regression algorithms and geographically weighted regression (GWR) in synthesizing existing forest age data sets for forest age estimation. The RF‐based integration approach outperformed GWR by effectively synthesizing complementary strengths of existing data sets, achieving superior model accuracy, a 47.4% improvement of R 2 , a 19.7% decrease of RMSE, and a 75.2% decrease of Bias over the best‐performing of existing products. The integrated forest age map provides enhanced reliability for national‐scale carbon budget modeling, while offering critical baseline data for optimizing forest management policies and carbon neutrality roadmaps.