生物多样性强烈的时空尺度依赖性和多层次性决定了生物多样性现状与变量的分析需要在不同生态系统进行多空间尺度、全面和连续的监测。因此,构建生物多样性研究监测网络是生物多样性保护和研究的基础工作。近年来,对地观测组织-生物多...生物多样性强烈的时空尺度依赖性和多层次性决定了生物多样性现状与变量的分析需要在不同生态系统进行多空间尺度、全面和连续的监测。因此,构建生物多样性研究监测网络是生物多样性保护和研究的基础工作。近年来,对地观测组织-生物多样性观测网络(GEO BON)、亚太生物多样性监测网络(APBON)等全球、区域以及国家尺度的生物多样性监测网络蓬勃发展。中国陆续在国家尺度上建立了针对生态系统和物种的长期监测网络,其中,中国生物多样性监测与研究网络(China Biodiversity Observation and Research Network,Sino BON)于2013年启动建设,在我国主要生态系统和环境梯度设置30个监测主点和60个监测辅点,目前已建成10个专项网对动物、植物和微生物进行监测,并建立了以数据标准与汇交、近地面遥感为核心的综合监测中心。Sino BON打造了从地下、地面到森林林冠的多尺度、多类群(功能群)以及多营养级交互为重点的监测与研究平台,为理解生物多样性变化趋势及其驱动因素、研究生物多样性维持机制,以及国家履行《生物多样性公约》、保护生物多样性和生物资源提供详实可靠的生物多样性变化数据。为进一步支撑国家生物多样性治理能力、深化全球多样性保护合作,我国生物多样性监测亟需在监测技术、监测区域、数据标准、综合信息平台等方向谋求更大的发展。展开更多
生态位模型通过拟合物种分布与环境变量之间的关系提供物种空间分布预测,在生物多样性研究中有广泛应用。激光雷达(LiDAR)是一种新兴的主动遥感技术,已被大量应用于森林三维结构信息的提取,但其在物种分布模拟的应用研究比较缺乏。本研...生态位模型通过拟合物种分布与环境变量之间的关系提供物种空间分布预测,在生物多样性研究中有广泛应用。激光雷达(LiDAR)是一种新兴的主动遥感技术,已被大量应用于森林三维结构信息的提取,但其在物种分布模拟的应用研究比较缺乏。本研究以美国加州内华达山脉南部地区的食鱼貂(Martes pennanti)的分布模拟为例,探索Li DAR技术在物种分布模拟中的有效性。生态位模型采用5种传统多类分类器,包括神经网络、广义线性模型、广义可加模型、最大熵模型和多元自适应回归样条模型,并使用正样本–背景学习(presence and background learning,PBL)算法进行模型校正;同时对这5种模型使用加权平均进行模型集成,作为第6个模型。此外,一类最大熵模型也被用于模拟该物种的空间分布。模型的连续输出和二值输出分别使用AUC(area under the receiver operating characteristic curve)以及基于正样本–背景数据的评价指标F_(pb)进行评价。结果表明,仅考虑气候因子(温度和降水)时,7个模型的AUC和F_(pb)平均值分别为0.779和1.077;当考虑Li DAR变量(冠层容重、枝下高、叶面积指数、高程、坡度等)后,AUC和F_(pb)分别为0.800和1.106。该研究表明,Li DAR数据能够提高食鱼貂空间分布的预测精度,在物种分布模拟方面存在一定的应用价值。展开更多
Accurate estimates of forest aboveground biomass(AGB)are essential for global carbon cycle studies and have widely relied on approaches using spectral and structural information of forest canopies extracted from vario...Accurate estimates of forest aboveground biomass(AGB)are essential for global carbon cycle studies and have widely relied on approaches using spectral and structural information of forest canopies extracted from various remote sensing datasets.However,combining the advantages of active and passive data sources to improve estimation accuracy remains challenging.Here,we proposed a new approach for forest AGB modeling based on allometric relationships and using the form of power-law to integrate structural and spectral information.Over 60 km^(2) of drone light detection and ranging(LiDAR)data and 1,370 field plot measurements,covering the four major forest types of China(coniferous forest,sub-tropical broadleaf forest,coniferous and broadleaf-leaved mixed forest,and tropical broadleaf forest),were collected together with Sentinel-2 images to evaluate the proposed approach.The results show that the most universally useful structural and spectral metrics are the average values of canopy height and spectral index rather than their maximum values.Compared with structural attributes used alone,combining structural and spectral information can improve the estimation accuracy of AGB,increasing R^(2) by about 10%and reducing the root mean square error by about 22%;the accuracy of the proposed approach can yield a R^(2) of 0.7 in different forests types.The proposed approach performs the best in coniferous forest,followed by sub-tropical broadleaf forest,coniferous and broadleaf-leaved mixed forest,and then tropical broadleaf forest.Furthermore,the simple linear regression used in the proposed method is less sensitive to sample size and outperforms statistically multivariate machine learning-based regression models such as stepwise multiple regression,artificial neural networks,and Random Forest.The proposed approach may provide an alternative solution to map large-scale forest biomass using space-borne LiDAR and optical images with high accuracy.展开更多
Accurate, efficient, and timely yield estimation is critical for crop variety breeding and management optimization. However, the contributions of proximal sensing data characteristics(spectral, temporal, and spatial) ...Accurate, efficient, and timely yield estimation is critical for crop variety breeding and management optimization. However, the contributions of proximal sensing data characteristics(spectral, temporal, and spatial) to yield estimation have not been systematically evaluated. We collected long-term, hypertemporal, and large-volume light detection and ranging(Li DAR) and multispectral data to(i) identify the best machine learning method and prediction stage for wheat yield estimation,(ii) characterize the contribution of multisource data fusion and the dynamic importance of structural and spectral traits to yield estimation, and(iii) elucidate the contribution of time-series data fusion and 3 D spatial information to yield estimation. Wheat yield could be accurately(R^(2)= 0.891) and timely(approximately-two months before harvest) estimated from fused Li DAR and multispectral data. The artificial neural network model and the flowering stage were always the best method and prediction stage, respectively. Spectral traits(such as CIgreen) dominated yield estimation, especially in the early stage, whereas the contribution of structural traits(such as height) was more stable in the late stage. Fusing spectral and structural traits increased estimation accuracy at all growth stages. Better yield estimation was realized from traits derived from complete 3 D points than from canopy surface points and from integrated multi-stage(especially from jointing to heading and flowering stages) data than from single-stage data. We suggest that this study offers a novel perspective on deciphering the contributions of spectral, structural, and timeseries information to wheat yield estimation and can guide accurate, efficient, and timely estimation of wheat yield.展开更多
High-throughput maize phenotyping at both organ and plant levels plays a key role in molecular breeding for increasing crop yields. Although the rapid development of light detection and ranging(Li DAR) provides a new ...High-throughput maize phenotyping at both organ and plant levels plays a key role in molecular breeding for increasing crop yields. Although the rapid development of light detection and ranging(Li DAR) provides a new way to characterize three-dimensional(3 D) plant structure, there is a need to develop robust algorithms for extracting 3 D phenotypic traits from Li DAR data to assist in gene identification and selection. Accurate 3 D phenotyping in field environments remains challenging, owing to difficulties in segmentation of organs and individual plants in field terrestrial Li DAR data. We describe a two-stage method that combines both convolutional neural networks(CNNs) and morphological characteristics to segment stems and leaves of individual maize plants in field environments. It initially extracts stem points using the Point CNN model and obtains stem instances by fitting 3 D cylinders to the points. It then segments the field Li DAR point cloud into individual plants using local point densities and 3 D morphological structures of maize plants. The method was tested using 40 samples from field observations and showed high accuracy in the segmentation of both organs(F-score =0.8207) and plants(Fscore =0.9909). The effectiveness of terrestrial Li DAR for phenotyping at organ(including leaf area and stem position) and individual plant(including individual height and crown width) levels in field environments was evaluated. The accuracies of derived stem position(position error =0.0141 m), plant height(R^(2)>0.99), crown width(R^(2)>0.90), and leaf area(R^(2)>0.85) allow investigating plant structural and functional phenotypes in a high-throughput way. This CNN-based solution overcomes the major challenges in organ-level phenotypic trait extraction associated with the organ segmentation, and potentially contributes to studies of plant phenomics and precision agriculture.展开更多
Background:Continuing controversy exists in diferent guidelines’recommendations regarding whether total thyroidectomy(TT)or lobectomy is the optimal surgery for patients with low-risk papillary thyroid carcinoma(PTC)...Background:Continuing controversy exists in diferent guidelines’recommendations regarding whether total thyroidectomy(TT)or lobectomy is the optimal surgery for patients with low-risk papillary thyroid carcinoma(PTC).Diverse perceptions of the risk of completion TT after lobectomy are the main debate between guidelines and institutions.Methods:Patients who underwent thyroidectomy and prophylactic central lymph node dissection for≤4 cm PTC(January 2007 to December 2020)by high-volume surgeons were included.Patients with preoperatively known highrisk characteristics or suspicious bilateral multifocality were excluded.The pathological fndings were defned as the risk stratifcations of completion TT from low to high to evaluate which initial surgical procedure could allow more patients to meet the criteria of optimal surgical extent.Results:Of 4965 consecutive patients met lobectomy criteria as the initial operation.Aggressive histological subtypes were found in 2.5%of patients,T3b disease in 1.1%,T4 disease in 3.1%,LNs involved≤5 in 29.5%,LNs involved>5 in 3.1%,and incidental bilateral multifocality in 7.9%.According to our defned risk stratifcation system,TT and lobectomy would be considered the optimal initial procedure in 12.0%and 67.2%PTC patients with a tumor≤1 cm and 28.7%and 36.6%in the 1-4 cm groups in our real-world cohort,respectively.Conclusion:Lobectomy alone,as an initial procedure,could allow more low-risk PTC patients with a tumor either≤1 cm or 1-4 cm to achieve the optimal surgical extent.Moreover,surgeons should balance the high-risk characteristics and complication risks during surgery to re-evaluate surgical decision-making.展开更多
Accurate quantification of tree populations within regions is critical for evaluating forest ecosystem conditions and developing effective forest management strategies[1].High-quality tree census data,collected throug...Accurate quantification of tree populations within regions is critical for evaluating forest ecosystem conditions and developing effective forest management strategies[1].High-quality tree census data,collected through field surveys and remote sensing technologies,is fundamental to China's sustainable development and environmental conservation initiatives.This data facilitates the monitoring of forest structural changes,carbon sequestration dynamics,and ecosystem health evaluations.Notably,China maintains the world's largest managed forest area,achieved through comprehensive national afforestation and reforestation programs[2,3].Consequently,precise tree enumeration is crucial for formulating effective forest management policies,monitoring and protecting wildlife habitats,and preventing natural disasters in China.展开更多
Canopy properties (e.g. canopy structure and spectral variables) strongly influence forest above-ground biomass (AGB). However, the importance of these canopy properties in driving AGB in natural forests, especially r...Canopy properties (e.g. canopy structure and spectral variables) strongly influence forest above-ground biomass (AGB). However, the importance of these canopy properties in driving AGB in natural forests, especially relative to other drivers such as plant species diversity and environmental conditions, remains poorly understood. We assessed the relative importance of canopy properties (structure and spectral variables) and plant species diversity (multidimensional diversity metrics and trait composition) in regulating AGB along environmental gradients (topography and soil nutrients) in a temperate forest in Northeast China, using UAV-based LiDAR and hyperspectral data. We found that the explanatory power of environmental conditions, plant species diversity, canopy spectral properties and canopy structure on temperate old-growth forests AGB was 3.8%, 8.0%, 4.1% and 13.3%, respectively. AGB increased with increasing canopy height and structural complexity. Canopy spectral diversity was a better predictor of AGB than traditional diversity metrics in old-growth forests. Canopy spectral composition also played an important role in explaining AGB in the secondary forests. In addition, plant phylogeny, functional diversity and the community-weighted mean of acquisitive traits had significant direct positive effects on AGB. Finally, topography and soil nutrient content indirectly influenced AGB through canopy properties and plant species diversity. Our study highlights the key role of canopy properties in influencing AGB. For future monitoring, regular monitoring with spectral and LiDAR data should be emphasized to provide real-time insights for forest management.展开更多
Grasslands store approximately one-third of terrestrial carbon(C)with most of it located belowground as soil organic carbon(SOC).Preserving and restoring SOC is essential for sustaining grassland ecosystem health and ...Grasslands store approximately one-third of terrestrial carbon(C)with most of it located belowground as soil organic carbon(SOC).Preserving and restoring SOC is essential for sustaining grassland ecosystem health and mitigating climate change.However,the effectiveness of soil management is constrained by limited understanding of where,how much,and how additional C can be stabilized in mineral-associated forms.Here,we combined a 2000-km field survey across temperate grasslands in China with machine learning to map the spatial distribution of mineralogical C deficit,defined as the unfilled capacity for long-term C stabilization through the formation of mineral-associated organic carbon(MAOC).We further conducted a^(13)C-labeled laboratory incubation experiment to identify key drivers of SOC formation efficiency.Random forest analysis revealed that mineralogical C deficit was primarily controlled by fine particle content,followed by mean annual temperature(MAT).Structural equation modeling showed that human disturbance indirectly reduced the deficit by reducing fine particles,while MAT increased it by altering soil chemistry and reducing plant cover.The largest deficits occurred in degraded and arid regions,totaling 0.78±0.08 Pg C within the top 15 cm of soil.Isotope tracing further demonstrated that MAOC formation efficiency declined with increasing C deficit,as high-deficit soils were also biogeochemically degraded,characterized by lower SOC,nitrogen content,and microbial biomass,which limit C stabilization.Collectively,our study provides a spatially explicit assessment of soil C sequestration potential,highlights how degradation constrains SOC formation,and identifies priority areas for targeted restoration and climate mitigation.展开更多
Vegetation maps are important sources of information for biodiversity conservation,ecological studies,vegetation management and restoration,and national strategic decision making.The current Vegetation Map of China(1:...Vegetation maps are important sources of information for biodiversity conservation,ecological studies,vegetation management and restoration,and national strategic decision making.The current Vegetation Map of China(1:1000000)was generated by a team of more than 250 scientists in an effort that lasted over 20 years starting in the 1980s.However,the vegetation distribution of China has experienced drastic changes during the rapid development of China in the last three decades,and it urgently needs to be updated to better represent the distribution of current vegetation types.Here,we describe the process of updating the Vegetation Map of China(1:1000000)generated in the 1980s using a‘‘crowdsourcing-change detection-classification-expert knowledge"vegetation mapping strategy.A total of 203,024 field samples were collected,and 50 taxonomists were involved in the updating process.The resulting updated map has 12 vegetation type groups,55 vegetation types/subtypes,and 866 vegetation formation/sub-formation types.The overall accuracy and kappa coefficient of the updated map are 64.8%and 0.52 at the vegetation type group level,61%and 0.55 at the vegetation type/subtype level and 40%and 0.38 at the vegetation formation/sub-formation level.When compared to the original map,the updated map showed that 3.3 million km^2 of vegetated areas of China have changed their vegetation type group during the past three decades due to anthropogenic activities and climatic change.We expect this updated map to benefit the understanding and management of China’s terrestrial ecosystems.展开更多
Ecological resources are an important material foundation for the survival,development,and self-realization of human beings.In-depth and comprehensive research and understanding of ecological resources are beneficial ...Ecological resources are an important material foundation for the survival,development,and self-realization of human beings.In-depth and comprehensive research and understanding of ecological resources are beneficial for the sustainable development of human society.Advances in observation technology have improved the ability to acquire long-term,cross-scale,massive,heterogeneous,and multi-source data.Ecological resource research is entering a new era driven by big data.Traditional statistical learning and machine learning algorithms have problems with saturation in dealing with big data.Deep learning is a method for automatically extracting complex high-dimensional nonlinear features,which is increasingly used for scientific and industrial data processing because of its ability to avoid saturation with big data.To promote the application of deep learning in the field of ecological resource research,here,we first introduce the relationship between deep learning theory and research on ecological resources,common tools,and datasets.Second,applications of deep learning in classification and recognition,detection and localization,semantic segmentation,instance segmentation,and graph neural network in typical spatial discrete data are presented through three cases:species classification,crop breeding,and vegetation mapping.Finally,challenges and opportunities for the application of deep learning in ecological resource research in the era of big data are summarized by considering the characteristics of ecological resource data and the development status of deep learning.It is anticipated that the cooperation and training of cross-disciplinary talents may promote the standardization and sharing of ecological resource data,improve the universality and interpretability of algorithms,and enrich applications with the development of hardware.展开更多
Grassland is one of the largest terrestrial biomes,providing critical ecosystem services such as food production,biodiversity conservation,and climate change mitigation.Global climate change and land-use intensificati...Grassland is one of the largest terrestrial biomes,providing critical ecosystem services such as food production,biodiversity conservation,and climate change mitigation.Global climate change and land-use intensification have been causing grassland degradation and desertification worldwide.As one of the primary medium for ecosystem energy flow and biogeochemical cycling,grassland carbon(C)cycling is the most fundamental process for maintaining ecosystem services.In this review,we first summarize recent advances in our understanding of the mechanisms underpinning spatial and temporal patterns of the grassland C cycle,discuss the importance of grasslands in regulating inter-and intra-annual variations in global C fluxes,and explore the previously unappreciated complexity in abiotic processes controlling the grassland C balance,including soil inorganic C accumulation,photochemical and thermal degradation,and wind erosion.We also discuss how climate and land-use changes could alter the grassland C balance by modifying the water budget,nutrient cycling and additional plant and soil processes.Further,we examine why and how increasing aridity and improper land use may induce significant losses in grassland C stocks.Finally,we identify several priorities for future grassland C research,including improving understanding of abiotic processes in the grassland C cycle,strengthening monitoring of grassland C dynamics by integrating ground inventory,flux monitoring,and modern remote sensing techniques,and selecting appropriate plant species combinations with suitable traits and strong resistance to climate fluctuations,which would help design sustainable grassland restoration strategies in a changing climate.展开更多
Plant growth rhythm in structural traits is important for better understanding plant response to the ever-changing environment.Terrestrial laser scanning(TLS)is a well-suited tool to study structural rhythm under fiel...Plant growth rhythm in structural traits is important for better understanding plant response to the ever-changing environment.Terrestrial laser scanning(TLS)is a well-suited tool to study structural rhythm under field conditions.Recent studies have used TLS to describe the structural rhythm of trees,but no consistent patterns have been drawn.Meanwhile,whether TLS can capture structural rhythm in crops is unclear.Here,we aim to explore the seasonal and circadian rhythms in maize structural traits at both the plant and leaf levels from time-series TLS.The seasonal rhythm was studied using TLS data collected at four key growth periods,including jointing,bell-mouthed,heading,and maturity periods.Circadian rhythms were explored by using TLS data acquired around every 2 hours in a whole day under standard and cold stress conditions.Results showed that TLS can quantify the seasonal and circadian rhythm in structural traits at both plant and leaf levels.(1)Leaf inclination angle decreased significantly between the jointing stage and bell-mouthed stage.Leaf azimuth was stable after the jointing stage.(2)Some individual-level structural rhythms(e.g.,azimuth and projected leaf area/PLA)were consistent with leaf-level structural rhythms.(3)The circadian rhythms of some traits(e.g.,PLA)were not consistent under standard and cold stress conditions.(4)Environmental factors showed better correlations with leaf traits under cold stress than standard conditions.Temperature was the most important factor that significantly correlated with all leaf traits except leaf azimuth.This study highlights the potential of time-series TLS in studying outdoor agricultural chronobiology.展开更多
Forests of the Sierra Nevada(SN)mountain range are valuable natural heritages for the region and the country,and tree height is an important forest structure parameter for understanding the SN forest ecosystem.There i...Forests of the Sierra Nevada(SN)mountain range are valuable natural heritages for the region and the country,and tree height is an important forest structure parameter for understanding the SN forest ecosystem.There is still a need in the accurate estimation of wall-to-wall SN tree height distribution at fine spatial resolution.In this study,we presented a method to map wall-to-wall forest tree height(defined as Lorey’s height)across the SN at 70-m resolution by fusing multi-source datasets,including over 1600 in situ tree height measurements and over 1600 km^(2) airborne light detection and ranging(LiDAR)data.Accurate tree height estimates within these airborne LiDAR boundaries were first computed based on in situ measurements,and then these airborne LiDAR-derived tree heights were used as reference data to estimate tree heights at Geoscience Laser Altimeter System(GLAS)footprints.Finally,the random forest algorithm was used to model the SN tree height from these GLAS tree heights,optical imagery,topographic data,and climate data.The results show that our fine-resolution SN tree height product has a good correspondence with field measurements.The coefficient of determination between them is 0.60,and the root-mean-squared error is 5.45 m.展开更多
Improved monitoring and understanding of tree growth and its responses to controlling factors are important for tree growth modeling.Airborne Laser Scanning(ALS)can be used to enhance the efficiency and accuracy of la...Improved monitoring and understanding of tree growth and its responses to controlling factors are important for tree growth modeling.Airborne Laser Scanning(ALS)can be used to enhance the efficiency and accuracy of large-scale forest surveys in delineating three-dimensional forest structures and under-canopy terrains.This study proposed an ALSbased framework to quantify tree growth and competition.Bi-temporal ALS data were used to quantify tree growth in height(ΔH),crown area(ΔA),crown volume(ΔV),and tree competition for 114,000 individual trees in two conifer-dominant Sierra Nevada forests.We analyzed the correlations between tree growth attributes and controlling factors(i.e.tree sizes,competition,forest structure,and topographic parameters)at multiple levels.At the individual tree level,ΔH had no consistent correlations with controlling factors,ΔA andΔV were positively related to original tree sizes(R>0.3)and negatively related to competition indices(R<−0.3).At the forest-stand level,ΔH andΔA were highly correlated to topographic wetness index(|R|>0.7),ΔV was positively related to original tree sizes(|R|>0.8).Multivariate regression models were simulated at individual tree level forΔH,ΔA,andΔV with the R2 ranged from 0.1 to 0.43.The ALS-based tree height estimation and growth analysis results were consistent with field measurements.展开更多
文摘生物多样性强烈的时空尺度依赖性和多层次性决定了生物多样性现状与变量的分析需要在不同生态系统进行多空间尺度、全面和连续的监测。因此,构建生物多样性研究监测网络是生物多样性保护和研究的基础工作。近年来,对地观测组织-生物多样性观测网络(GEO BON)、亚太生物多样性监测网络(APBON)等全球、区域以及国家尺度的生物多样性监测网络蓬勃发展。中国陆续在国家尺度上建立了针对生态系统和物种的长期监测网络,其中,中国生物多样性监测与研究网络(China Biodiversity Observation and Research Network,Sino BON)于2013年启动建设,在我国主要生态系统和环境梯度设置30个监测主点和60个监测辅点,目前已建成10个专项网对动物、植物和微生物进行监测,并建立了以数据标准与汇交、近地面遥感为核心的综合监测中心。Sino BON打造了从地下、地面到森林林冠的多尺度、多类群(功能群)以及多营养级交互为重点的监测与研究平台,为理解生物多样性变化趋势及其驱动因素、研究生物多样性维持机制,以及国家履行《生物多样性公约》、保护生物多样性和生物资源提供详实可靠的生物多样性变化数据。为进一步支撑国家生物多样性治理能力、深化全球多样性保护合作,我国生物多样性监测亟需在监测技术、监测区域、数据标准、综合信息平台等方向谋求更大的发展。
文摘生态位模型通过拟合物种分布与环境变量之间的关系提供物种空间分布预测,在生物多样性研究中有广泛应用。激光雷达(LiDAR)是一种新兴的主动遥感技术,已被大量应用于森林三维结构信息的提取,但其在物种分布模拟的应用研究比较缺乏。本研究以美国加州内华达山脉南部地区的食鱼貂(Martes pennanti)的分布模拟为例,探索Li DAR技术在物种分布模拟中的有效性。生态位模型采用5种传统多类分类器,包括神经网络、广义线性模型、广义可加模型、最大熵模型和多元自适应回归样条模型,并使用正样本–背景学习(presence and background learning,PBL)算法进行模型校正;同时对这5种模型使用加权平均进行模型集成,作为第6个模型。此外,一类最大熵模型也被用于模拟该物种的空间分布。模型的连续输出和二值输出分别使用AUC(area under the receiver operating characteristic curve)以及基于正样本–背景数据的评价指标F_(pb)进行评价。结果表明,仅考虑气候因子(温度和降水)时,7个模型的AUC和F_(pb)平均值分别为0.779和1.077;当考虑Li DAR变量(冠层容重、枝下高、叶面积指数、高程、坡度等)后,AUC和F_(pb)分别为0.800和1.106。该研究表明,Li DAR数据能够提高食鱼貂空间分布的预测精度,在物种分布模拟方面存在一定的应用价值。
基金supported by the Strategic Priority Research Program of Chinese Academy of Sciences(Grant No.XDA19050401)the National Natural Science Foundation of China(41871332,31971575,41901358).
文摘Accurate estimates of forest aboveground biomass(AGB)are essential for global carbon cycle studies and have widely relied on approaches using spectral and structural information of forest canopies extracted from various remote sensing datasets.However,combining the advantages of active and passive data sources to improve estimation accuracy remains challenging.Here,we proposed a new approach for forest AGB modeling based on allometric relationships and using the form of power-law to integrate structural and spectral information.Over 60 km^(2) of drone light detection and ranging(LiDAR)data and 1,370 field plot measurements,covering the four major forest types of China(coniferous forest,sub-tropical broadleaf forest,coniferous and broadleaf-leaved mixed forest,and tropical broadleaf forest),were collected together with Sentinel-2 images to evaluate the proposed approach.The results show that the most universally useful structural and spectral metrics are the average values of canopy height and spectral index rather than their maximum values.Compared with structural attributes used alone,combining structural and spectral information can improve the estimation accuracy of AGB,increasing R^(2) by about 10%and reducing the root mean square error by about 22%;the accuracy of the proposed approach can yield a R^(2) of 0.7 in different forests types.The proposed approach performs the best in coniferous forest,followed by sub-tropical broadleaf forest,coniferous and broadleaf-leaved mixed forest,and then tropical broadleaf forest.Furthermore,the simple linear regression used in the proposed method is less sensitive to sample size and outperforms statistically multivariate machine learning-based regression models such as stepwise multiple regression,artificial neural networks,and Random Forest.The proposed approach may provide an alternative solution to map large-scale forest biomass using space-borne LiDAR and optical images with high accuracy.
基金supported by the Jiangsu Agricultural Science and Technology Independent Innovation Fund Project (CX(21)3107)the National Natural Science Foundation of China(32030076)+4 种基金High Level Personnel Project of Jiangsu Province(JSSCBS20210271)China Postdoctoral Science Foundation(2021 M691490)Jiangsu Planned Projects for Postdoctoral Research Funds (2021K520C)Strategic Priority Research Program of the Chinese Academy of Sciences (XDA24020202)the Jiangsu 333 Program。
文摘Accurate, efficient, and timely yield estimation is critical for crop variety breeding and management optimization. However, the contributions of proximal sensing data characteristics(spectral, temporal, and spatial) to yield estimation have not been systematically evaluated. We collected long-term, hypertemporal, and large-volume light detection and ranging(Li DAR) and multispectral data to(i) identify the best machine learning method and prediction stage for wheat yield estimation,(ii) characterize the contribution of multisource data fusion and the dynamic importance of structural and spectral traits to yield estimation, and(iii) elucidate the contribution of time-series data fusion and 3 D spatial information to yield estimation. Wheat yield could be accurately(R^(2)= 0.891) and timely(approximately-two months before harvest) estimated from fused Li DAR and multispectral data. The artificial neural network model and the flowering stage were always the best method and prediction stage, respectively. Spectral traits(such as CIgreen) dominated yield estimation, especially in the early stage, whereas the contribution of structural traits(such as height) was more stable in the late stage. Fusing spectral and structural traits increased estimation accuracy at all growth stages. Better yield estimation was realized from traits derived from complete 3 D points than from canopy surface points and from integrated multi-stage(especially from jointing to heading and flowering stages) data than from single-stage data. We suggest that this study offers a novel perspective on deciphering the contributions of spectral, structural, and timeseries information to wheat yield estimation and can guide accurate, efficient, and timely estimation of wheat yield.
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA24020202)the National Key Research and Development Program of China(2017YFA0604300)+2 种基金the National Natural Science Foundation of China (U1811464 and 41875122)the Western Talents(2018XBYJRC004)the Guangdong Top Young Talents(2017TQ04Z359)。
文摘High-throughput maize phenotyping at both organ and plant levels plays a key role in molecular breeding for increasing crop yields. Although the rapid development of light detection and ranging(Li DAR) provides a new way to characterize three-dimensional(3 D) plant structure, there is a need to develop robust algorithms for extracting 3 D phenotypic traits from Li DAR data to assist in gene identification and selection. Accurate 3 D phenotyping in field environments remains challenging, owing to difficulties in segmentation of organs and individual plants in field terrestrial Li DAR data. We describe a two-stage method that combines both convolutional neural networks(CNNs) and morphological characteristics to segment stems and leaves of individual maize plants in field environments. It initially extracts stem points using the Point CNN model and obtains stem instances by fitting 3 D cylinders to the points. It then segments the field Li DAR point cloud into individual plants using local point densities and 3 D morphological structures of maize plants. The method was tested using 40 samples from field observations and showed high accuracy in the segmentation of both organs(F-score =0.8207) and plants(Fscore =0.9909). The effectiveness of terrestrial Li DAR for phenotyping at organ(including leaf area and stem position) and individual plant(including individual height and crown width) levels in field environments was evaluated. The accuracies of derived stem position(position error =0.0141 m), plant height(R^(2)>0.99), crown width(R^(2)>0.90), and leaf area(R^(2)>0.85) allow investigating plant structural and functional phenotypes in a high-throughput way. This CNN-based solution overcomes the major challenges in organ-level phenotypic trait extraction associated with the organ segmentation, and potentially contributes to studies of plant phenomics and precision agriculture.
基金supported by grants from the“Ten Thousand People Plan”of Yunnan Province-Medical Experts Project[RLCRC20210412].
文摘Background:Continuing controversy exists in diferent guidelines’recommendations regarding whether total thyroidectomy(TT)or lobectomy is the optimal surgery for patients with low-risk papillary thyroid carcinoma(PTC).Diverse perceptions of the risk of completion TT after lobectomy are the main debate between guidelines and institutions.Methods:Patients who underwent thyroidectomy and prophylactic central lymph node dissection for≤4 cm PTC(January 2007 to December 2020)by high-volume surgeons were included.Patients with preoperatively known highrisk characteristics or suspicious bilateral multifocality were excluded.The pathological fndings were defned as the risk stratifcations of completion TT from low to high to evaluate which initial surgical procedure could allow more patients to meet the criteria of optimal surgical extent.Results:Of 4965 consecutive patients met lobectomy criteria as the initial operation.Aggressive histological subtypes were found in 2.5%of patients,T3b disease in 1.1%,T4 disease in 3.1%,LNs involved≤5 in 29.5%,LNs involved>5 in 3.1%,and incidental bilateral multifocality in 7.9%.According to our defned risk stratifcation system,TT and lobectomy would be considered the optimal initial procedure in 12.0%and 67.2%PTC patients with a tumor≤1 cm and 28.7%and 36.6%in the 1-4 cm groups in our real-world cohort,respectively.Conclusion:Lobectomy alone,as an initial procedure,could allow more low-risk PTC patients with a tumor either≤1 cm or 1-4 cm to achieve the optimal surgical extent.Moreover,surgeons should balance the high-risk characteristics and complication risks during surgery to re-evaluate surgical decision-making.
基金supported by the National Key Research and Development Program of China(2022YFF1300203)the National Natural Science Foundation of China(42371329 and 32301285)。
文摘Accurate quantification of tree populations within regions is critical for evaluating forest ecosystem conditions and developing effective forest management strategies[1].High-quality tree census data,collected through field surveys and remote sensing technologies,is fundamental to China's sustainable development and environmental conservation initiatives.This data facilitates the monitoring of forest structural changes,carbon sequestration dynamics,and ecosystem health evaluations.Notably,China maintains the world's largest managed forest area,achieved through comprehensive national afforestation and reforestation programs[2,3].Consequently,precise tree enumeration is crucial for formulating effective forest management policies,monitoring and protecting wildlife habitats,and preventing natural disasters in China.
基金supported by the National Key Research and Development Program of China (2023YFE0124300, 2022YFF1300501)the CAS (Chinese Academy of Sciences) Project for Young Scientists in Basic Research (YSBR-108)+1 种基金the National Natural Science Foundation of China (32301344)LiaoNing Revitalization Talents Program (XLYC2402003)。
文摘Canopy properties (e.g. canopy structure and spectral variables) strongly influence forest above-ground biomass (AGB). However, the importance of these canopy properties in driving AGB in natural forests, especially relative to other drivers such as plant species diversity and environmental conditions, remains poorly understood. We assessed the relative importance of canopy properties (structure and spectral variables) and plant species diversity (multidimensional diversity metrics and trait composition) in regulating AGB along environmental gradients (topography and soil nutrients) in a temperate forest in Northeast China, using UAV-based LiDAR and hyperspectral data. We found that the explanatory power of environmental conditions, plant species diversity, canopy spectral properties and canopy structure on temperate old-growth forests AGB was 3.8%, 8.0%, 4.1% and 13.3%, respectively. AGB increased with increasing canopy height and structural complexity. Canopy spectral diversity was a better predictor of AGB than traditional diversity metrics in old-growth forests. Canopy spectral composition also played an important role in explaining AGB in the secondary forests. In addition, plant phylogeny, functional diversity and the community-weighted mean of acquisitive traits had significant direct positive effects on AGB. Finally, topography and soil nutrient content indirectly influenced AGB through canopy properties and plant species diversity. Our study highlights the key role of canopy properties in influencing AGB. For future monitoring, regular monitoring with spectral and LiDAR data should be emphasized to provide real-time insights for forest management.
基金supported by the National Natural Science Foundation of China(32125025,32330066,and 32101307).
文摘Grasslands store approximately one-third of terrestrial carbon(C)with most of it located belowground as soil organic carbon(SOC).Preserving and restoring SOC is essential for sustaining grassland ecosystem health and mitigating climate change.However,the effectiveness of soil management is constrained by limited understanding of where,how much,and how additional C can be stabilized in mineral-associated forms.Here,we combined a 2000-km field survey across temperate grasslands in China with machine learning to map the spatial distribution of mineralogical C deficit,defined as the unfilled capacity for long-term C stabilization through the formation of mineral-associated organic carbon(MAOC).We further conducted a^(13)C-labeled laboratory incubation experiment to identify key drivers of SOC formation efficiency.Random forest analysis revealed that mineralogical C deficit was primarily controlled by fine particle content,followed by mean annual temperature(MAT).Structural equation modeling showed that human disturbance indirectly reduced the deficit by reducing fine particles,while MAT increased it by altering soil chemistry and reducing plant cover.The largest deficits occurred in degraded and arid regions,totaling 0.78±0.08 Pg C within the top 15 cm of soil.Isotope tracing further demonstrated that MAOC formation efficiency declined with increasing C deficit,as high-deficit soils were also biogeochemically degraded,characterized by lower SOC,nitrogen content,and microbial biomass,which limit C stabilization.Collectively,our study provides a spatially explicit assessment of soil C sequestration potential,highlights how degradation constrains SOC formation,and identifies priority areas for targeted restoration and climate mitigation.
基金supported by the Strategic Priority Research Program of Chinese Academy of Sciences(XDA19050401)Maps in this article were reviewed by Ministry of Natural Resources of the People’s Republic of China(GS(2020)1044)。
文摘Vegetation maps are important sources of information for biodiversity conservation,ecological studies,vegetation management and restoration,and national strategic decision making.The current Vegetation Map of China(1:1000000)was generated by a team of more than 250 scientists in an effort that lasted over 20 years starting in the 1980s.However,the vegetation distribution of China has experienced drastic changes during the rapid development of China in the last three decades,and it urgently needs to be updated to better represent the distribution of current vegetation types.Here,we describe the process of updating the Vegetation Map of China(1:1000000)generated in the 1980s using a‘‘crowdsourcing-change detection-classification-expert knowledge"vegetation mapping strategy.A total of 203,024 field samples were collected,and 50 taxonomists were involved in the updating process.The resulting updated map has 12 vegetation type groups,55 vegetation types/subtypes,and 866 vegetation formation/sub-formation types.The overall accuracy and kappa coefficient of the updated map are 64.8%and 0.52 at the vegetation type group level,61%and 0.55 at the vegetation type/subtype level and 40%and 0.38 at the vegetation formation/sub-formation level.When compared to the original map,the updated map showed that 3.3 million km^2 of vegetated areas of China have changed their vegetation type group during the past three decades due to anthropogenic activities and climatic change.We expect this updated map to benefit the understanding and management of China’s terrestrial ecosystems.
基金supported by the Strategic Priority Research Program of Chinese Academy of Sciences(Grant No.XDA19050401)the National Natural Science Foundation of China(Grant Nos.31971575&41871332)。
文摘Ecological resources are an important material foundation for the survival,development,and self-realization of human beings.In-depth and comprehensive research and understanding of ecological resources are beneficial for the sustainable development of human society.Advances in observation technology have improved the ability to acquire long-term,cross-scale,massive,heterogeneous,and multi-source data.Ecological resource research is entering a new era driven by big data.Traditional statistical learning and machine learning algorithms have problems with saturation in dealing with big data.Deep learning is a method for automatically extracting complex high-dimensional nonlinear features,which is increasingly used for scientific and industrial data processing because of its ability to avoid saturation with big data.To promote the application of deep learning in the field of ecological resource research,here,we first introduce the relationship between deep learning theory and research on ecological resources,common tools,and datasets.Second,applications of deep learning in classification and recognition,detection and localization,semantic segmentation,instance segmentation,and graph neural network in typical spatial discrete data are presented through three cases:species classification,crop breeding,and vegetation mapping.Finally,challenges and opportunities for the application of deep learning in ecological resource research in the era of big data are summarized by considering the characteristics of ecological resource data and the development status of deep learning.It is anticipated that the cooperation and training of cross-disciplinary talents may promote the standardization and sharing of ecological resource data,improve the universality and interpretability of algorithms,and enrich applications with the development of hardware.
基金supported by the National Natural Science Foundation of China (32125025 and 31988102)the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA23080301 and XDA26010303)JMG acknowledges the support of the Israel Science Foundation (1796/19).
文摘Grassland is one of the largest terrestrial biomes,providing critical ecosystem services such as food production,biodiversity conservation,and climate change mitigation.Global climate change and land-use intensification have been causing grassland degradation and desertification worldwide.As one of the primary medium for ecosystem energy flow and biogeochemical cycling,grassland carbon(C)cycling is the most fundamental process for maintaining ecosystem services.In this review,we first summarize recent advances in our understanding of the mechanisms underpinning spatial and temporal patterns of the grassland C cycle,discuss the importance of grasslands in regulating inter-and intra-annual variations in global C fluxes,and explore the previously unappreciated complexity in abiotic processes controlling the grassland C balance,including soil inorganic C accumulation,photochemical and thermal degradation,and wind erosion.We also discuss how climate and land-use changes could alter the grassland C balance by modifying the water budget,nutrient cycling and additional plant and soil processes.Further,we examine why and how increasing aridity and improper land use may induce significant losses in grassland C stocks.Finally,we identify several priorities for future grassland C research,including improving understanding of abiotic processes in the grassland C cycle,strengthening monitoring of grassland C dynamics by integrating ground inventory,flux monitoring,and modern remote sensing techniques,and selecting appropriate plant species combinations with suitable traits and strong resistance to climate fluctuations,which would help design sustainable grassland restoration strategies in a changing climate.
基金the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA24020202)Plant Phenomics Research Program of Science and Technology Department of Jiangsu Province(No.BM2018001)Beijing Municipal Science and Technology Project(Z191100007419004).
文摘Plant growth rhythm in structural traits is important for better understanding plant response to the ever-changing environment.Terrestrial laser scanning(TLS)is a well-suited tool to study structural rhythm under field conditions.Recent studies have used TLS to describe the structural rhythm of trees,but no consistent patterns have been drawn.Meanwhile,whether TLS can capture structural rhythm in crops is unclear.Here,we aim to explore the seasonal and circadian rhythms in maize structural traits at both the plant and leaf levels from time-series TLS.The seasonal rhythm was studied using TLS data collected at four key growth periods,including jointing,bell-mouthed,heading,and maturity periods.Circadian rhythms were explored by using TLS data acquired around every 2 hours in a whole day under standard and cold stress conditions.Results showed that TLS can quantify the seasonal and circadian rhythm in structural traits at both plant and leaf levels.(1)Leaf inclination angle decreased significantly between the jointing stage and bell-mouthed stage.Leaf azimuth was stable after the jointing stage.(2)Some individual-level structural rhythms(e.g.,azimuth and projected leaf area/PLA)were consistent with leaf-level structural rhythms.(3)The circadian rhythms of some traits(e.g.,PLA)were not consistent under standard and cold stress conditions.(4)Environmental factors showed better correlations with leaf traits under cold stress than standard conditions.Temperature was the most important factor that significantly correlated with all leaf traits except leaf azimuth.This study highlights the potential of time-series TLS in studying outdoor agricultural chronobiology.
基金This study is supported by the National Science Foundation of China[project numbers 41471363 and 31270563]National Science Foundation[DBI 1356077]the USDA Forest Service Pacific Southwest Research Station.
文摘Forests of the Sierra Nevada(SN)mountain range are valuable natural heritages for the region and the country,and tree height is an important forest structure parameter for understanding the SN forest ecosystem.There is still a need in the accurate estimation of wall-to-wall SN tree height distribution at fine spatial resolution.In this study,we presented a method to map wall-to-wall forest tree height(defined as Lorey’s height)across the SN at 70-m resolution by fusing multi-source datasets,including over 1600 in situ tree height measurements and over 1600 km^(2) airborne light detection and ranging(LiDAR)data.Accurate tree height estimates within these airborne LiDAR boundaries were first computed based on in situ measurements,and then these airborne LiDAR-derived tree heights were used as reference data to estimate tree heights at Geoscience Laser Altimeter System(GLAS)footprints.Finally,the random forest algorithm was used to model the SN tree height from these GLAS tree heights,optical imagery,topographic data,and climate data.The results show that our fine-resolution SN tree height product has a good correspondence with field measurements.The coefficient of determination between them is 0.60,and the root-mean-squared error is 5.45 m.
基金This study is supported by the National Natural Science Foundation of China[project numbers 41471363 and 31270563]National Science Foundation[DBI 1356077]the USDA Forest Service Pacific Southwest Research Station.
文摘Improved monitoring and understanding of tree growth and its responses to controlling factors are important for tree growth modeling.Airborne Laser Scanning(ALS)can be used to enhance the efficiency and accuracy of large-scale forest surveys in delineating three-dimensional forest structures and under-canopy terrains.This study proposed an ALSbased framework to quantify tree growth and competition.Bi-temporal ALS data were used to quantify tree growth in height(ΔH),crown area(ΔA),crown volume(ΔV),and tree competition for 114,000 individual trees in two conifer-dominant Sierra Nevada forests.We analyzed the correlations between tree growth attributes and controlling factors(i.e.tree sizes,competition,forest structure,and topographic parameters)at multiple levels.At the individual tree level,ΔH had no consistent correlations with controlling factors,ΔA andΔV were positively related to original tree sizes(R>0.3)and negatively related to competition indices(R<−0.3).At the forest-stand level,ΔH andΔA were highly correlated to topographic wetness index(|R|>0.7),ΔV was positively related to original tree sizes(|R|>0.8).Multivariate regression models were simulated at individual tree level forΔH,ΔA,andΔV with the R2 ranged from 0.1 to 0.43.The ALS-based tree height estimation and growth analysis results were consistent with field measurements.
