Accurate estimation of forest carbon storage is essential for understanding the dynamics of forest resources and optimizing decisions for forest resource management. In order to explore the changes in the carbon storage of Pinus densata in Shangri-La and the influence of topography on carbon storage, two dynamic models were developed based on the National Forest Inventory (NFI) and Landsat TM/OLI images with a 5-year interval change and annual average change. The three modelling methods used were partial least squares (PLSR), random forest (RF) and gradient boosting regression tree (GBRT). Various spectral and texture features of the images were calculated and filtered before modelling. The terrain niche index (TNI), which is able to reflect the combined effect of elevation and slope, was added to the dynamic model, the optimal model was selected to estimate the carbon storage, and the topographic conditions in areas of change in carbon storage were analyzed. The results showed that: (1) The dynamic model based on 5-year interval change data performs better than the dynamic model with annual average change data, and the RF model has a higher accuracy compared to the PLSR and GBRT models. (2) The addition of TNI improved the accuracy, in which R² is improved by up to 10.48% at most, RMSE is reduced by up to 7.32% at most, and MAE is reduced by up to 8.89% at most, and the RF model based on the 5-year interval change data has the highest accuracy after adding TNI, with an R² of 0.87, an RMSE of 3.82 t-C·ha⁻¹, and a MAE of 1.78 t-C·ha⁻¹. (3) The direct estimation results of the dynamic model showed that the carbon storage of Pinus densata in Shangri-La decreased in 1987–1992 and 1997–2002, and increased in 1992–1997, 2002–2007, 2007–2012, and 2012–2017. (4) The trend of increasing or decreasing carbon storage in each period is not exactly the same on the TNI gradient, according to the dominant distribution, as topographic conditions with lower elevations or gentler slopes are favorable for the accumulation of carbon storage, while the decreasing area of carbon storage is more randomly distributed topographically. This study develops a dynamic estimation model of carbon storage considering topographic factors, which provides a solution for the accurate estimation of forest carbon storage in regions with a complex topography.

Forest carbon sinks are vital in mitigating climate change, making it crucial to have highly accurate estimates of forest carbon stocks. A method that accounts for the spatial characteristics of inventory samples is necessary for the long-term estimation of above-ground forest carbon stocks due to the spatial heterogeneity of bottom-up methods. In this study, we developed a method for analyzing space-sensing data that estimates and predicts long time series of forest carbon stock changes in an alpine region by considering the sample’s spatial characteristics. We employed a nonlinear mixed-effects model and improved the model’s accuracy by considering both static and dynamic aspects. We utilized ground sample point data from the National Forest Inventory (NFI) taken every five years, including tree and soil information. Additionally, we extracted spectral and texture information from Landsat and combined it with DEM data to obtain topographic information for the sample plots. Using static data and change data at various annual intervals, we built estimation models. We tested three non-parametric models (Random Forest, Gradient-Boosted Regression Tree, and K-Nearest Neighbor) and two parametric models (linear mixed-effects and non-linear mixed-effects) and selected the most accurate model to estimate Pinus densata’s above-ground carbon stock. The results showed the following: (1) The texture information had a significant correlation with static and dynamic above-ground carbon stock changes. The highest correlation was for large-window mean, entropy, and variance. (2) The dynamic above-ground carbon stock model outperformed the static model. Additionally, the dynamic non-parametric models and parametric models experienced improvements in prediction accuracy. (3) In the multilevel nonlinear mixed-effects models, the highest accuracy was achieved with fixed effects for aspect and two-level nested random effects for the soil and elevation categories. (4) This study found that Pinus densata’s above-ground carbon stock in Shangri-La followed a decreasing, and then, increasing trend from 1987 to 2017. The mean carbon density increased overall, from 19.575 t·hm⁻² to 25.313 t·hm⁻². We concluded that a dynamic model based on variability accurately reflects Pinus densata’s above-ground carbon stock changes over time. Our approach can enhance time-series estimates of above-ground carbon stocks, particularly in complex topographies, by incorporating topographic factors and soil thickness into mixed-effects models.

Larix gmelinii is the fourth most important tree species in China and a typical zonal climax species in the cold temperate region, with high ecological and resource value. However, intensive logging, high-density afforestation, and insufficient scientific management have led to overly dense, homogeneous, and unstable plantations, severely limiting productivity. To clarify the mechanisms by which structural dynamics regulate productivity, we established a space-for-time sequence (T1–T3, T2-D, CK) under a consistent early-tending background. Using the “1 + 4” nearest-neighbor framework and six spatial structural parameters, we developed tree and forest spatial structure indices (TSSI and FSSI) and integrated nine structural–functional indicators for multivariate analysis. The results showed that TSSI and FSSI effectively characterized multi-level stability and supported stability classification. Along the stand-age gradient, structural stability and spatial use efficiency improved significantly, with FSSI and biomass per hectare (BPH) increasing by 91% and 18% from T1 to T3, though a “structural improvement–functional lag” occurred at T2. Moderate thinning markedly optimized stand configuration, reducing low-stability individuals from 86.45% in T1 to 42.65% in T2-D, while DBH, crown width, FSSI, and BPH (229.87 t·hm⁻²) increased to near natural-forest levels. At the tree scale, DBH, tree height, crown width, and TSSI were positive drivers, whereas a high height–diameter ratio (HDR) constrained growth. At the stand scale, canopy density, species richness, and mean DBH promoted FSSI and BPH, while mean HDR and stand density imposed major constraints. A critical management window was identified when DBH < 25 cm, HDR > 10, and TSSI < 0.25 (approximately 10–30 years post-planting). We propose a stepwise, moderate, and targeted thinning strategy with necessary underplanting to reduce density and slenderness, increase diameter and canopy structure, and enhance diversity, thereby accelerating the synergy between stability and productivity. This framework provides a practical pathway for the scientific management and high-quality development of L. gmelinii plantations.

Nature reserves serve as core spatial units for maintaining regional ecological security and biodiversity. Owing to their high ecosystem integrity, extensive vegetation cover, and low levels of disturbance, they play a crucial role in sustaining ecological processes and ensuring functional stability. Taking the Giant Panda National Park (GPNP), which spans the provinces of Gansu, Sichuan, and Shaanxi in China, as the study region, the vegetation net primary productivity (NPP) during 2001–2023 was simulated using the Carnegie–Ames–Stanford Approach (CASA) model. Spatial and temporal variations in NPP were examined using Moran’s I, Getis-Ord Gi* hotspot analysis, Theil–Sen trend estimation, and the Mann–Kendall test. In addition, the Optimal Parameters-based Geographical Detector (OPGD) model was applied to quantitatively assess the relative contributions of natural and anthropogenic factors to NPP dynamics. The results demonstrated that: (1) The mean annual NPP within the GPNP reached 646.90 gC·m⁻²·yr⁻¹, exhibiting a fluctuating yet generally upward trajectory, with an average growth rate of approximately 0.65 gC·m⁻²·yr⁻¹, reflecting the positive ecological outcomes of national park establishment and ecological restoration projects. (2) NPP exhibits significant spatial heterogeneity, with higher NPP values in the northern, while the central and western regions and some high-altitude areas remain at relatively low levels. Across the four major subregions of the GPNP, the Qinling has the highest mean annual NPP at 758.89 gC·m⁻²·yr⁻¹, whereas the Qionglai–Daxiaoxiangling subregion shows the lowest value at 616.27 gC·m⁻²·yr⁻¹. (3) Optimal NPP occurred under favorable temperature and precipitation conditions combined with relatively high solar radiation. Low elevations, gentle slopes, south facing aspects, and leached soils facilitated productivity accumulation, whereas areas with high elevation and steep slopes exhibited markedly lower productivity. Moderate human disturbance contributed to sustaining and enhancing NPP. (4) Factor detection results indicated that elevation, mean annual temperature, and land use were the dominant drivers of spatial heterogeneity when considering all natural and anthropogenic variables. Their interactions further enhanced explanatory power, particularly the interaction between elevation and climatic factors. Overall, these findings reveal the complex spatiotemporal characteristics and multi-factorial controls of vegetation productivity in the GPNP and provide scientific guidance for strengthening habitat conservation, improving ecological restoration planning, and supporting adaptive vegetation management within the national park systems.

Accurate estimation of forest tree species diversity at regional scales is essential for biodiversity monitoring, forest resource management, and ecological conservation. Because tree species differ in canopy spectral responses and phenological dynamics, multi-season remote sensing observations can provide critical information for characterizing species composition and diversity patterns. However, the potential contribution of seasonal image features to improving remote-sensing-based tree species diversity estimation has often been insufficiently considered. In this study, the Yichun forest region in Heilongjiang Province, northeastern China, was selected as the study area. Sentinel-1, Sentinel-2, and topographic data were integrated to extract multi-seasonal spectral, vegetation index, texture, radar, and topographic features. The Boruta algorithm was used for feature selection, and random forest (RF), extreme gradient boosting (XGBoost), k-nearest neighbor (KNN), support vector regression (SVR), Bayesian regularized neural network (BRNN), and Stacking ensemble learning were developed to estimate and map Richness, Shannon, and Gini–Simpson indices. The results showed that: (1) Sentinel-2 optical features were the primary information source for tree species diversity estimation, topographic factors further improved model performance, and Sentinel-1 radar features mainly provided complementary structural information; (2) seasonal remote sensing features differed in their predictive ability, with Richness performing better in spring, while Shannon and Gini–Simpson achieved higher accuracy in winter. The four-season fusion scenario produced the highest accuracy for all three indices, with optimal R² values of 0.51, 0.63, and 0.57, respectively; (3) the Stacking ensemble generally improved estimation accuracy and model stability, although the optimal model differed among diversity indices, with Stacking, SVR, and RF performing best for Richness, Shannon, and Gini–Simpson, respectively; and (4) summer Sentinel-2 NDVI, GNDVI, and NDWI contributed strongly to all three indices, elevation was particularly important for Richness, and winter vegetation indices and autumn red-edge bands and texture features were also informative for Shannon and Gini–Simpson. These findings indicate that integrating multi-seasonal remote sensing features and multi-source data using machine learning models can effectively improve forest tree species diversity estimation, providing technical support for regional forest biodiversity monitoring and precision forest management.

Forest inventory compartments provide extensive and management-relevant reference information for satellite-based tree species mapping, but their dominant-species attributes are defined at the stand level rather than for individual image pixels. Existing applications commonly derive training samples from compartment centres or assign polygon labels to enclosed pixels, which may introduce boundary effects, uneven class representation, and disproportionate contributions from individual compartments. However, the intermediate step of converting inventory polygons into spatially controlled pixel-level reference samples has received comparatively limited attention. Here, we developed a geometry-constrained reference sample construction framework that integrates interior-position screening, class balancing, source compartment contribution control, and spatial-spacing constraints. The framework was evaluated for mapping Korean pine, larch, white birch, and spruce in a temperate mixed forest in northeastern China using Sentinel-1/2 time series and ancillary predictors. Predictor–classifier combinations were selected using compartment-grouped out-of-fold evaluation, sampling workflows were compared on 60 independently withheld compartments, and the final map was further assessed using 306 independent reference points. Relative to centroid sampling, the geometry-constrained workflow increased compartment-level macro-F1 from 0.612 to 0.709. XGBoost with optical time series and ancillary predictors achieved the best development-set performance, while inclusion of the complete Sentinel-1 time series provided no further gain. The final model achieved an overall accuracy of 0.827 and a macro-F1 of 0.820 on the independent reference points. Aggregation of 10 m predictions further enabled compartment-level characterization of mapped dominant species, dominance strength, and mixing intensity. These results demonstrate that reference sample construction is a consequential step in tree species mapping from polygon-based forest inventories and provide a practical approach for linking pixel-level remote sensing classification with forest management units.