With a penetration capacity through vegetation, passive microwave provides an opportunity to estimate vegetation water content (VWC) through vegetation optical depth (VOD) and vegetation structure parameter b(p). However, accurately modeling b(p) is challenging as it depends on diverse vegetation characteristics that are difficult to measure directly. Consequently, b(p) is typically prescribed as a constant based on land cover to avoid complexity. This simplification can introduce significant uncertainty in VWC/VOD estimation, particularly for crops, whose vegetation structure and phenology vary considerably during the growing season. To address this limitation, this study explored multiple empirical yet physically interpretable models to optimize b(p) estimation for improved retrievals of VWC/VOD and soil moisture. Using the Passive Active L-band Sensor (PALS) dataset from the SMAPVEX16-MB campaign, we first simulated b(p) for canola, wheat and soybean via the L-band Microwave Emission of the Biosphere (L-MEB) with calibrated surface roughness and scattering albedo. Then, we established multilinear regression, stepwise regression, and Random Forest (RF) algorithms to estimate b(p) using screened input features, including biomass, leaf area index (LAI), VWC, height, phenology and brightness temperature (TB). These screened features, which are common across different crop types, exhibit high correlation with b(p) for both h- and v- polarizations. The established three algorithms provided robust estimation of b(p) parameter, with strong correlations (R >= 0.88) and low errors (RMSE <= 0.05 and ubRMSE <= 0.05) to the simulated b(p). Among the three algorithms, the RF and multilinear regression with vegetation-type variables performed the best. However, RF is a black-box model, and detailed vegetation type data is not available at large scales. Compared with multilinear regression, stepwise regression achieved similar R, RMSE, and ubRMSE, while also reducing the collinearity among input variables. Our study deepened the understanding of the b(p) as a function of vegetation type, growth and polarization. For wheat and soybean, the b(p) at v-polarization (b(v)) shows greater variability and is higher than b(p) at h-polarization (b(h)). Besides, b(h) was mainly driven by biomass, VWC and LAI, while b(v) was dominated by biomass and VWC. Our study provides an insight in adaptive b(p) parameter estimation, with the potential to improve VWC estimation at different scales and to enable the accurate separation of vegetation and soil contributions in the mixed signature.

Accurate individual tree crown (ITC) segmentation is essential for quantifying forest fine carbon stocks. However, deep learning segmentation methods face prohibitive annotation costs while unsupervised algorithms struggle in structurally complex forests. This study proposes a semi-supervised framework integrating unsupervised pseudo-label generation, deep instance segmentation, and staged fine-tuning to overcome these limitations. The framework transforms outputs from marker-controlled watershed (MCWS) and region-growing (itcSegment) algorithms into initial training targets to obtain generalizable instance segmentation representations, then refines Mask R-CNN models with minimal manual annotations to enhance structural awareness of instance boundaries and semantic context understanding, significantly reducing labeling dependency while improving robustness. Validated across three structurally heterogeneous Chinese fir stands spanning age (11-67 years), density (450-2500 stems & sdot;ha-1), and elevation (192-1047 m) gradients, our UAV RGB-based framework achieved consistent superiority over LiDAR and fused inputs, with spectral-textural features proving dominant for boundary delineation. It attained F1-scores of 0.826 (young, undulating terrain), 0.836 (mature, high-density), and 0.711 (over-mature, occluded) using only 40 % expert annotations (0.6 personnel-hours), representing a 0.42 average improvement over MCWS/itcSegment baselines. The designed staged fine-tuning strategy effectively mitigated pseudo-label error propagation, while expanding this annotation effort to 70-100 % yielded marginal accuracy gains, demonstrating exceptional efficiency in leveraging minimal supervision. Based on the segmented crowns, structural parameters were extracted with high fidelity: crown diameter (R2 > 0.76, rRMSE < 11.3 %) and crown area (R2 > 0.88, rRMSE < 16.4 %). This approach reduces annotation demands while maintaining robustness across forest heterogeneity, enabling operationally scalable solutions for precision forestry.

Precision forestry requires smart monitoring tools that deliver operational estimates of aboveground carbon (AGC) stocks to support sustainable plantation management. While UAV-LiDAR technology enables highresolution forest structure characterization, practitioners lack evidence-based guidance on selecting appropriate modeling approaches that balance accuracy, interpretability, and processing efficiency for real-world deployment. This study evaluated four UAV-LiDAR-based modeling strategies for reference AGC estimation across 33 field plots in Chinese fir (Cunninghamia lanceolata) plantations, providing practical guidance for technology selection. We compared individual-tree models (height-only H; height-crown diameter H-CD), a stand parameter model (mean height-stem density H-N), and a point cloud metrics model (PCM) using highdensity LiDAR data (1500-2000 pts center dot m-2) and concurrent field measurements. Stand-level models significantly outperformed individual-tree approaches in reproducing the reference AGC, revealing a fundamental scaling effect with direct operational implications. The exponential H-N model achieved the highest predictive performance against the reference (R2 = 0.912, RMSE = 12.184 Mg C center dot hm-2), while the segmentation-free PCM performed robustly (R2 = 0.861). Critically, LiDAR-derived stem density showed no significant difference from field measurements (p = 0.177), supporting its use in stand-level modeling within this allometric context. Based on these empirical findings, we propose a scenario-specific selection strategy for applying these modeling pathways under a common derived reference: (1) for individual-tree management objectives, H-CD is applicable but users must account for higher uncertainty; (2) for stand-level carbon accounting, H-N provides optimal agreement with the reference and clear interpretability; (3) for dense stands where individual tree detection fails, PCM offers a robust, scalable alternative. This study bridges remote sensing technology and operational forestry by clarifying the relative strengths of each UAV-LiDAR-based modeling pathway against a consistent allometric reference, thereby supporting data-driven method selection and enabling managers to deploy UAV-LiDAR-based monitoring matched to their specific application scenarios.

Addressing climate change requires both global cooperation and local action. Mountainous rural areas, with their significant carbon sink potential, play a crucial yet under-explored role in achieving carbon neutrality. However, it is not clear how far these villages are from carbon neutrality. This study proposes a novel carbon neutrality assessment framework specifically designed for mountainous villages, integrating carbon emission reduction, sink, and community engagement to assess and guide local carbon neutrality efforts. Case studies in Baizhang Town and its six surrounding villages showed that the town and four villages achieved carbon neutrality, illustrating the framework's effectiveness in reducing emissions and enhancing sink. Theoretical analysis further supports the framework by providing insights into its applicability and scalability for rural communities. The framework offers a practical tool for local communities to manage carbon emissions and implement sustainable practices. This framework can be adapted to other rural areas, offering a model for carbon neutrality efforts across diverse mountainous regions.

L-band vegetation optical depth (VOD) from NASA's Soil Moisture Active Passive (SMAP) mission is widely used to study carbon, water, and energy exchange. Its retrieval relies on the tau-omega radiative transfer model, constrained by Normalized Difference Vegetation Index (NDVI) climatology and a constant surface roughness parameter (Hrp). However, constant Hrp fails to capture dynamics from agricultural practices and weather, while NDVI saturates under dense canopies and cannot directly reflect vegetation structure. To address these limitations, we propose a new algorithm that incorporates leaf area index (LAI) climatology and its dynamic Hrp into the constrained multi-channel algorithm (CMCA) that can account for smooth temporal vegetation variations, with the final aim to improve VOD retrievals by optimizing Hrp parameterization and prior information of VOD. Based on Passive Active L-band Sensor (PALS) data from SMAPVEX16-MB, we first evaluated four Hrp parametrization schemes across three well-developed algorithms (the regularized dual-channel algorithm (RDCA), the multitemporal dual-channel algorithm (MT-DCA), and CMCA). The schemes include two constant models based on surface geometry, one dynamic model based on LAI and brightness temperature (TB), and the original algorithm parameterization. Then, we assessed measured LAI and NDVI climatology as prior information for constraining VOD, demonstrating that measured LAI provides more accurate VOD estimates due to its superior characterization of vegetation water content (VWC) and biomass dynamics. Synergistically upscaling field-based LAI and Hrp to satellite footprint-scale consistently enhanced the three algorithms, though to varying degrees, with retrieved dynamic Hrp via VOD determined by LAI climatology. We found that the CMCA performs best, with the correlation coefficients (R) between VOD and VWC/biomass across the three crops ranging from 0.73 to 0.88 and 0.84-0.88, respectively. This improvement stems from the better ability of LAI climatology to capture VWC and biomass variations than NDVI climatology, the physical constraints in CMCA, and the dynamic Hrp that characterizes surface roughness. The proposed algorithms were validated against vegetation water content and biomass measurements from SMAPVEX12. Results suggest that LAI climatology can significantly improve VOD retrieval from SMAP observations: R values between VOD and VWC/biomass were 0.71-0.81/0.68-0.90 for the improved algorithm versus 0.38-0.81/0.37-0.94 for the original. Overall, this study highlights the importance of incorporation LAI and dynamic roughness parameter for improved VOD retrieval, which is essential for carbon cycle studies dependent on accurate vegetation dynamics.

Moso bamboo forests have a high capacity for carbon sequestration and are associated with greenhouse gas emissions, but are facing abandonment due to rising labor costs and falling prices of bamboo products. Due to lower carbon fixation and ecosystem degradation of abandoned bamboo forest, transformation (thinning and replanting with tree species) has recently been used to address the issue but its effect on soil N2O fluxes is unknown. In this study, a 25-month field experiment was conducted with four management treatments (light, moderate and heavy strip transformation and abandonment management), using intensive management as a control, to investigate the effects of abandonment and its transformation on soil N2O fluxes in shallow soils of an abandoned Moso bamboo forest. The results revealed that the highest N2O emissions occurred in the intensive management control, while the lowest values were observed under the abandonment management. Furthermore, compared with the control, forest transformation with heavy, moderate and light intensities and abandonment management lowered annual cumulative soil N2O emissions by 7 %, 12 %, 14 %, and 20 %, respectively, in the first year, and by 6 %, 14 %, 17 %, and 22 %, respectively, in the second year. Regardless of the treatment, soil N2O emissions were correlated positively with soil temperature, and the concentrations of NO3--N, NH4+-N, microbial biomass C and N, and water-soluble organic C and N (P < 0.05), but negatively with soil water-filled porosity (P < 0.01). The increased N2O emissions in the forest transformation treatments had mainly resulted from the elevated soil temperature and increased concentrations of labile C and N. The study suggests that the light strip transformation, due to its lower N2O emissions during the first two years of the transformation, is a favorable practice for managing abandoned Moso bamboo forests.

Regulating carbon emissions during landscape maintenance is crucial for increasing net carbon sequestration in urban green spaces. This research focuses on balancing water and energy resource conservation with increasing carbon sequestration. We introduced an integrated water-energy-carbon (WEC) fluxes framework to evaluate the net carbon sequestration of green spaces and a water-energy-carbon sustainability index (WECSI) was developed to assess overall sustainability, emphasizing the carbon sequestration potential (CSP). Taking five types of urban green spaces at Zhejiang Agriculture and Forestry University as a case study, we observed significant differences in the WEC fluxes among the green space types. Specifically, mainly arbors and closed green space (AC) types had greater CSP due to scale effects, whereas mainly successional short grass and open green space (SGO) types were at risk of becoming net carbon sources. The WECSI analysis revealed the difficulties in simultaneously achieving water conservation, energy efficiency, and net carbon sequestration, with an average sustainability score of 0.57. To maximize CSP in urban green spaces, scenario analysis indicated that low-carbon irrigation practices could increase CSP by up to 25 %, whereas biomass energy from garden waste could reduce irrigation-related carbon emissions by 19%. These findings provide a strong foundation for optimizing urban green space management to maximize CSP through WEC fluxes regulation.

Global and local efforts have each made significant contributions to advancing sustainable development, yet systematic research on the role of global strategies in guiding local practices remains scarce. This study takes the Yangtze River Delta urban agglomerations as a case study and introduces an innovative framework that integrates global perspectives with local practices. By conducting a comparative analysis of ecosystem service (ES) bundles under the climate change scenario and a local scenario group, the study identifies critical pathways for optimizing future spatial planning. The findings suggest that merely increasing ESs supply is not the primary objective of the local spatial management, addressing key trade-offs among ESs should take precedence. For instance, promoting sustainable agriculture can mitigate conflicts between ESs, reduce boundaries between woodland and cropland, foster synergies among multiple ESs, and alleviate the imbalance in ecological development between the northern and southern regions. In highly urbanized areas, the growth of urban green spaces can also contribute positively to sustainable development. This framework not only bridges the gap between global strategies and local sustainable development practices but also expands the application of ES bundles in spatial planning and management. It offers new theoretical insights and practical solutions for achieving sustainability.

Unmanned aerial vehicle (UAV)-captured RGB imagery, with high spatial resolution and ease of acquisition, is increasingly applied to individual tree crown detection (ITCD). However, ITCD in dense subtropical forests remains challenging due to overlapping crowns, variable crown size, and similar spectral responses between neighbouring crowns. This paper investigates to what extent the ITCD accuracy can be improved by using dual-seasonal UAV-captured RGB imagery in different subtropical forest types: urban broadleaved, planted coniferous, and mixed coniferous–broadleaved forests. A modified YOLOv8 model was employed to fuse the features extracted from dual-seasonal images and perform the ITCD task. Results show that dual-seasonal imagery consistently outperformed single-seasonal datasets, with the greatest improvement in mixed forests, where the F1 score range increased from 56.3%–60.7% (single-seasonal datasets) to 69.1%–74.5% (dual-seasonal datasets) and the AP value range increased from 57.2%–61.5% to 70.1%–72.8%. Furthermore, performance fluctuations were smaller for dual-seasonal datasets than for single-seasonal datasets. Finally, our experiments demonstrate that the modified YOLOv8 model, which fuses features extracted from dual-seasonal images within a dual-branch module, outperformed both the original YOLOv8 model with channel-wise stacked dual-seasonal inputs and the Faster R-CNN model with a dual-branch module. The experimental results confirm the advantages of using dual-seasonal imagery for ITCD, as well as the critical role of model feature extraction and fusion strategies in enhancing ITCD accuracy.

Forest transformation can markedly impact soil greenhouse gas emissions and soil environmental factors. Due to increasing labor costs and declining bamboo prices, the abandonment of Moso bamboo forests is sharply escalating in recent years, which weakens the carbon sequestration capacity and decreases the ecological function of forests. To improve the ecological quality of abandoned Moso bamboo forests, transformations of abandoned bamboo forests have occurred. However, the impact of such transformations on N2O emissions remains elusive. To bridge the knowledge gap, we conducted a 23-month field experiment to compare the effects of various forest management practices on soil N2O emissions and soil environmental factors in abandoned Moso bamboo forests in subtropical China. These practices included uncut abandonment as a control, intensive management, three intensities (light, moderate, and heavy) of strip clear-cutting with replanting local tree species, and clear-cutting with replanting transformation. During the experimental period, the mean soil N2O flux in abandoned Moso bamboo forests was 13.2 +/- 0.1 mu g m(-2) h(-1), representing a 44% reduction compared to intensive management forests. In comparison to the uncut control, light, moderate, and heavy strip clear-cutting and clear-cutting transformations increased soil N2O emission rates by 20%, 43%, 64%, and 94%, respectively. Soil temperature (69-71%), labile C (2-6%) and N (3-8%) were the main factors that explain N2O emissions following the transformation of abandoned Moso bamboo forests. Additionally, replanting could decrease soil N2O emissions by increasing the contribution of soil moisture. Overall, the light strip clear-cutting transformation is suggested to convert abandoned Moso bamboo forests to mitigate N2O emissions.

Background and Aims: Tea plantations are frequently given substantial quantities of nitrogen fertilizers. However, there is the potential for considerable nitrogen loss to occur. This study assesses the nitrogen retention of acidic tea plantation’s soil and the role of biochar in improving nitrogen dynamics, highlighting the need for innovative technologies to streamline and enhance nitrogen supply management. Methods: Adopting a modified two-week aerobic incubation and ion-exchange membrane technology, this research offers a novel approach to evaluate soil nitrogen supply and to monitor the nitrogen dynamics of tea plantation soil following early-summer supplementary fertilization. Results: The study revealed that the surface soil of tea plantation had the ability to provide 48 mg N·kg -1 soil as inorganic nitrogen for 130 days. The utilization of a small amount of biochar (10 t·ha -1 ) had no impact on the soil's effective nitrogen availability. Nonetheless, the application of biochar at rates of 20 and 30 t·ha -1 resulted in a significant enhancement in soil effective nitrogen availability as measured using ion exchange membranes, with an increase of 65%–81%. Furthermore, the utilization of biochar-based organic fertilizers, when used at appropriate rates, has the potential to enhance the availability of nitrogen in the soil, thereby increasing its effectiveness. Conclusion: The study's findings underscore the efficacy of the employed methodologies in capturing the nuanced impact of biochar on nitrogen retention and availability in tea plantation soils. The use of aerobic incubation and ion-exchange membrane technology has proven effective in elucidating the potential of biochar to significantly improve nitrogen dynamics.

The light detection and ranging (LiDAR) systems mounted on unmanned aerial vehicle (UAV) platforms can provide high-density point cloud data for accurate individual tree detection and segmentation, which is needed for precision forestry. Individual trees can be detected and segmented based on tree trunk detection. It is a challenging task in forests characterized by high understory vegetation and varying point densities of trunks caused by obstruction from the upper canopy. We propose an approach to detect tree trunks and segment individual trees from UAV-LiDAR data. First, a trunk point distribution indicator (TPDI) was used to detect potential tree trunk positions (PTPs). Then random sample consistency (RANSAC)-based 3D line fitting was applied to each PTP to differentiate tree trunks from understory vegetation. Finally, a trunk-based region-growing segmentation method was applied to segment individual trees, and the result was refined through analysis of crown shape and vertical profiles. The approach was tested at three study sites in Eucalyptus plantations, which were characterized by overlapping crowns and relatively high understory vegetation. F-scores ranging from 0.920 to 1.000 were derived in 12 plots, and the accuracies increased with the tree heights. The comparative shortest-path algorithm for tree trunk detection and segmentation was applied for comparison and derived much lower F-scores (0.526–0.867). The proposed approach was also evaluated by replacing TPDI with a similar indicator. The comparison result indicated that the proposed approach was especially advantageous in forests characterized by relatively low tree heights and high understory vegetation.

Rural wetlands are complex landscapes where rivers, croplands, and villages coexist, making water quality monitoring crucial for the well-being of nearby residents. UAV-based imagery has proven effective in capturing detailed features of water bodies, making it a popular tool for water quality assessments. However, few studies have specifically focused on drone-based water quality monitoring in rural wetlands and their seasonal variations. In this study, Xiangfudang Rural Wetland Park, Jiaxin City, Zhejiang Province, China, was taken as the study area to evaluate water quality parameters, including total nitrogen (TN), total phosphors (TP), chemical oxygen demand (COD), and turbidity degree (TUB). We assessed these parameters across summer and winter seasons using UAV multispectral imagery and field sample data. Four machine learning algorithms were evaluated and compared for the inversion of the water quality parameters, based on the situ sample survey data and UAV multispectral images. The results show that ANN algorithm yielded the best results for estimating TN, COD, and TUB, with validation R2 of 0.78, 0.76, and 0.57, respectively; CatBoost performed best in TP estimation, with validation R2 and RMSE values of 0.72 and 0.05 mg/L. Based on spatial estimation results, the average COD concentration in the water body was 16.05 ± 9.87 mg/L in summer, higher than it was in winter (13.02 ± 8.22 mg/L). Additionally, mean TUB values were 18.39 Nephelometric Turbidity Units (NTU) in summer and 20.03 NTU in winter. This study demonstrates the novelty and effectiveness of using UAV multispectral imagery for water quality monitoring in rural wetlands, providing critical insights into seasonal water quality variations in these areas.

Where and how many trees should be thinned in a pure managed forest to improve forest quality and increase ecological benefits are important forest questions. In this study we address such challenges by providing a novel framework for planning thinning operations through Unmanned Aerial Vehicle (UAV) remote sensing techniques, which can not only obtain forest attributes of its entire stand with spatial properties, but also optimize the selection of thinning areas, thinning intensities and cut-trees. This study helps to reduce the costs of time-consuming and laborious ground investigations. The framework was demonstrated by applying it into a subtropical Chinese fir plantation in southeastern China. Results showed that RGB images acquired by a low-cost UAV have great potential in depicting forest structure. The overall accuracy of the individual tree detection in the case study was 85.19 % ± 0.48 %. The overall accuracy and the intersection over union of the non-crown area extraction were 94.94 % and 82.65 %, respectively. For the two determined thinning areas, 19.5 % and 14.3 % crown density were required to thin in the primary and secondary regions, respectively. In addition, the top-down perspective of UAV remote sensing makes up for the limitations of the bottom-up perspective of traditional forestry. The framework can act as a basic model for forest managers to modify and expand for customizing detailed thinning guidelines.

Due to the rapid development of society and economy, the contradiction between the increasing demand for energy and the crisis of energy utilization in various countries has been intensified in the past decades. In order to alleviate the current situation of energy shortage and adapt to the global climate change, the development and utilization of renewable energy such as forestry biomass energy have attracted the attention of all countries. This article summarizes the development, status, and potential of China’s forestry biomass energy and analyzes the importance of China’s forestry biomass energy. It is expected to provide not only a certain reference for market development and policy adjustment in developing countries, but also the effective data support and scientific basis for subsequent research.

[目的]比较不同间伐强度下杉木Cunninghamia lanceolata人工林土壤呼吸速率,解释影响土壤呼吸速率的主要因子,为森林经营及碳汇管理提供科学依据.[方法]采用随机区组设计,设置对照(间伐0％)、中度间伐(间伐45％)、重度间伐(间伐70％)3种间伐处理,采用静态箱-气象色谱法对杉木人工林土壤呼吸速率进行短期原位监测.[结果]间伐显著增加了杉木林土壤呼吸速率(P＜0.05),与对照相比,中度和重度间伐的土壤呼吸速率分别增加了23.30％和44.94％.土壤呼吸速率与5 cm 土壤温度呈显著指数相关,而与土壤含水量无关.不同间伐处理下,土壤呼吸温度敏感性系数(Q10)为1.77～2.16,对照处理下Q10 最高,间伐降低了杉木林土壤呼吸的温度敏感性.杉木人工林土壤呼吸速率与土壤水溶性有机碳、微生物生物量碳和易氧化有机碳呈显著正相关(P＜0.01).[结论]间伐初期,间伐对杉木林土壤呼吸速率有促进作用,且随着间伐强度的增加而增加.土壤温度是土壤呼吸速率变化的主要影响因子,土壤活性有机碳是重要因子.

[目的]研究雄安新区建区前后土地利用变化和生态质量动态变化,以期为雄安新区各行业建设和管理提供理论依据和数据支撑.[方法]以雄安新区3个县为研究区,基于2000、2017和2020年3期土地利用数据,利用遥感生态指数(remote sensing ecological index,RSEI)模型评价生态质量的时空变化特征,研究土地利用变化对生态质量的影响,探讨土地利用变化的生态效应.[结果]①传统城镇化建设阶段(2000-2017年)土地利用类型主要由耕地向建设用地和未利用地转移,新区建设阶段(2017-2020年)土地利用类型主要由耕地和未利用地向林地转移;②2000-2017年雄安新区RSEI均值从0.62下降至0.57,生态质量总体有所退化;2020年RSEI均值增加至0.59,生态质量轻微改善;③土地利用与生态质量的分布和变化在空间上基本一致,林地、耕地面积的增减对生态质量的变化影响显著.在当前建设水平和1km空间尺度下,未利用地转耕地面积比每增加10％,则生态质量改善的面积比约增加13％;未利用地转林地面积比每增加10％,则生态质量改善的面积比约增加7％.[结论]雄安新区应维护林地和耕地质量和数量,改善雄安新区生态质量,从而有效促进雄安新区可持续健康发展.