Tree Cover, Forest Definitions, and Trees Outside Forests: A National Assessment of China

Forests play a pivotal role in carbon sequestration, biodiversity conservation, and climate change mitigation, making accurate and comparable forest monitoring essential for global change research and climate policy. However, forest area estimates remain highly uncertain due to substantial inconsistencies among widely used definitions, including those adopted by FAO FRA, UNFCCC, and national forest inventories. Moreover, conventional forest assessments predominantly focus on closed forests, systematically neglecting Trees Outside Forests (TOF), such as scattered or linear trees embedded in agricultural and urban landscapes. This omission limits our understanding of national tree resources and introduces biases in carbon accounting.

Recent advances in high-resolution remote sensing provide new opportunities to overcome these limitations. In this study, we leverage the 1-m resolution Global Canopy Height Model (GCHM) to conduct a national-scale, definition-consistent assessment of tree cover across China, explicitly incorporating TOF. Tree cover was derived by aggregating 1-m canopy height data to 10-m resolution, enabling the detection of both continuous forests and sparse individual trees. Forest distributions were extracted under three commonly used definitions (FAO FRA, China’s National Forest Inventory, and UNFCCC) using definition-specific thresholds for tree cover, tree height, and minimum patch size. TOF were identified as tree-covered areas not meeting the forest definition of China’s NFI. We further analyzed the spatial distribution, structural characteristics, and landscape patterns of TOF across major land-use types, ecological zones, and geomorphological regions using landscape metrics that quantify fragmentation, connectivity, and aggregation.

Our results demonstrate that forest area estimates in China are highly sensitive to forest definitions. Forest area derived from the UNFCCC and China’s NFI definitions is comparable (148.37 and 145.79 Mha, respectively), whereas the FAO FRA definition yields a substantially lower estimate (85.72 Mha, approximately 58% of the former). Despite these differences, all definitions consistently identify Northeast China, Southwest China, and southeastern mountainous regions as core forest areas, while pronounced discrepancies emerge in fragmented landscapes of southern hills and ecological transition zones. TOF cover 49.21 Mha, accounting for 5.1% of China’s land area and 26.9% of total tree cover. TOF are dominated by low canopy cover and are widely distributed across agricultural and urban regions. In several provinces with low forest coverage, TOF contribute more than 50% of total tree cover, indicating that they constitute the dominant form of tree-based vegetation outside traditional forest areas.

By explicitly integrating TOF into a national-scale tree cover framework, this study reveals a substantial yet previously underrecognized component of China’s terrestrial carbon sink. The findings highlight the limitations of binary forest–non-forest classification systems and demonstrate the necessity of incorporating TOF into forest monitoring, land-use planning, and carbon accounting. This work provides a strengthened basis for reducing uncertainties in national carbon sink assessments and supports more effective climate mitigation strategies.