[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2002":3,"related-2002":103},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"primary_topic":22,"secondary_topics":28,"classification":40,"search_phrases":74,"slug":75,"view_count":15,"doi":76,"paper":77,"created_at":102},2002,"Identifying the influencing factors of lake surface greenhouse gas concentrations using sentinel-2 remote sensing and machine learning: A case study of bird Island, Qinghai Lake","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-67206-5","Alpine lakes on the Qinghai-Tibet Plateau are important regions for studies of the regional carbon cycle and are highly sensitive to climate change. However, the variation patterns of lake surface greenhouse gases and their environmental regulation mechanisms remain unclear. In this study, Bird Island of Qinghai Lake was selected as the study area. Based on the measured concentrations of lake surface CO 2 , CH 4 , and H 2 O corresponding to Sentinel-2 satellite overpass periods throughout 2021, combined with Sentinel-2 remote sensing imagery and ERA5-Land meteorological data, Random Forest (RF) and Extreme Gradient Boosting (XGBoost) models, together with the SHAP method, were used to identify and predict the influencing factors and their contribution characteristics to the variations in lake surface CO 2 , CH 4 , and H 2 O. The results showed that CO 2 , CH 4 , and H 2 O exhibited distinct seasonal variations. The annual variation in CO 2 concentration generally showed a “V”-shaped pattern, with concentrations during the growing season being lower than those during the non-growing season. The annual variation in CH 4 concentration was relatively small and exhibited a “wave”-shaped pattern. The H 2 O concentration showed a distinct unimodal pattern, and its high-value period was generally consistent with the low-value period of CO 2 . Temperature and MNDWI were the main predictive factors affecting the variations in CO 2 , CH 4 , and H 2 O concentrations, while H 2 O concentration was also affected by Radiation. This study combined observational data, remote sensing imagery, and machine learning, providing new insights into the environmental predictive factors of greenhouse gas concentrations in lakes under climate change.","青藏高原高山湖泊是研究区域碳循环的重要区域，对气候变化高度敏感。然而，湖泊表层温室气体的变化规律及其环境调控机制仍不清楚。本研究以青海湖鸟岛为研究区，基于2021年全年与Sentinel-2卫星过境时段对应的湖泊表层CO₂、CH₄和H₂O实测浓度，结合Sentinel-2遥感影像和ERA5-Land气象数据，采用随机森林（Random Forest, RF）和极端梯度提升（Extreme Gradient Boosting, XGBoost）模型，并借助SHAP方法，识别和预测了影响湖泊表层CO₂、CH₄和H₂O变化的因素及其贡献特征。结果表明，CO₂、CH₄和H₂O表现出明显的季节变化。CO₂浓度的年变化总体呈“V”形模式，生长季浓度低于非生长季。CH₄浓度的年变化幅度相对较小，呈“波浪”形模式。H₂O浓度呈明显的单峰模式，其高值期与CO₂的低值期总体一致。温度和归一化差异水体指数（MNDWI）是影响CO₂、CH₄和H₂O浓度变化的主要预测因子，而H₂O浓度还受到辐射的影响。本研究将观测数据、遥感影像和机器学习相结合，为气候变化下湖泊温室气体浓度的环境预测因子提供了新的见解。",null,"Scientific Reports","2026-09-07T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"研究青海湖温室气体，非农业领域，不相关。",[19],{"name":10,"url":6},[21],"遥感",{"code":23,"slug":24,"name":25,"path":26,"parent_code":27},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture","B",[29,35],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":41,"locked":14,"version":42,"method":43,"evidence":44},"classified","2026-10-22.2","rules",{"candidates":45,"agricultural_evidence":48},[46,53,60,66],{"code":36,"score":47,"supported":48,"title":49,"abstract":51,"tags":52},15,true,[50],"remote sensing",[21,50],[21],{"code":23,"score":54,"supported":48,"title":55,"abstract":57,"tags":59},14,[56],"greenhouse",[58,56],"温室",[],{"code":30,"score":54,"supported":48,"title":61,"abstract":63,"tags":65},[62],"machine learning",[64,62],"机器学习",[],{"code":67,"score":68,"supported":48,"title":69,"abstract":70,"tags":73},"H7",6,[],[71,72],"气候变化","climate change",[],[],"Identifyingtheinfluencingfactorsoflakesu-2002","10.1038\u002Fs41598-026-67206-5",{"doi":76,"openalex_id":78,"authors":79,"venue":10,"cited_by_count":15,"oa_url":94,"card":95,"direction":99,"ingested_from":101},"W7211873544",[80,82,85,87,90,92],{"name":81,"orcid":9},"Xingyue Li",{"name":83,"orcid":84},"Yu Chen","https:\u002F\u002Forcid.org\u002F0000-0002-9910-6942",{"name":86,"orcid":9},"Zhen Chen",{"name":88,"orcid":89},"Lei Li","https:\u002F\u002Forcid.org\u002F0000-0003-4368-8637",{"name":91,"orcid":9},"Dong Han",{"name":93,"orcid":9},"Kelong Chen","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-67206-5_reference.pdf",{"tldr":96,"method":97,"finding":98,"direction":99,"opportunity":100},"利用Sentinel-2遥感与机器学习识别青海湖鸟岛湖面温室气体浓度的影响因素。","结合Sentinel-2影像、ERA5-Land气象数据，使用RF和XGBoos","CO2呈V型季节变化，CH4呈波浪形，H2O呈单峰；温度和MNDWI是主要影响因素。","农业遥感与作物表型","可延伸至其他高原湖泊或内陆水体，结合更多环境变量和深度学习模型，提高温室气体浓度预测精度。","openalex","2026-09-09T23:30:23.971965Z",{"total":68,"page":104,"page_size":68,"items":105},1,[106,165,256,329,434,507],{"id":107,"title":108,"url":109,"summary":110,"summary_zh":111,"content":9,"source_name":112,"source_url":109,"published_at":113,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":114,"score_detail":115,"sources":121,"tags":123,"primary_topic":128,"secondary_topics":129,"classification":131,"search_phrases":149,"slug":152,"view_count":15,"doi":153,"paper":154,"created_at":164},5570,"Evaluation of Counterfactual Explanation Robustness for Computer Vision Agriculture Yield-Forecast Models Under Satellite Cloud-Mask Gaps: An Interpretability Reliability Bound for Cooperative Storage Decisions","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545079","Counterfactual explanations are increasingly proposed as the interface through which cooperative storage managers interrogate computer-vision yield-forecast models, yet their reliability under the satellite cloud-mask gaps that routinely afflict agricultural remote sensing has not been bounded. The bound is derived from a Lipschitz continuity assumption on the forecast model and a masking-operator perturbation analysis, and it is expressed as a function of the cloud-mask gap fraction, the model's local Lipschitz constant, and the counterfactual search tolerance. A replication-based evaluation of the before-after-control-impact estimator was conducted to characterise the operating characteristics of the design under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), with power estimated across 2000 design cells. The reliability bound degrades non-linearly as the gap fraction rises, and that the intraclass correlation of 0.64 implies that cooperative-level aggregation of counterfactual recourse is necessary to preserve explanation stability. The contribution is a decision-relevant reliability bound that storage cooperatives can use to set a maximum tolerable cloud-mask gap before counterfactual recourse is suspended.","反事实解释日益被提出作为合作式仓储管理者质询计算机视觉产量预测模型的接口，然而在农业遥感中经常出现的卫星云掩膜缺口条件下，其可靠性尚未得到界定。该界基于预测模型上的Lipschitz连续性假设和掩膜算子扰动分析推导得出，并表示为云掩膜缺口比例、模型局部Lipschitz常数和反事实搜索容差的函数。对前后对照影响估计量进行了基于重复的评估，以表征该设计在给定方差结构（组间标准差=6.0，残差标准差=4.5，组内相关系数=0.64）下的运行特征，并在2000个设计单元中估计了检验效能。可靠性界随缺口比例上升呈非线性退化，且组内相关系数0.64意味着需要对反事实救济进行合作社层面的聚合，以保持解释的稳定性。本文的贡献在于提供了一个与决策相关的可靠性界，仓储合作社可据此设定在暂停反事实救济之前可容忍的最大云掩膜缺口。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",67,{"impact":116,"substance":117,"depth":118,"authority":119,"freshness":13,"relevant":104,"comment":120},8,20,17,12,"该论文提出反事实解释在卫星云掩膜缺口下的可靠性边界，方法新颖且结论可用于合作社决策，但属细分方法学研究，产业影响有限。",[122],{"name":112,"url":109},[124,125,126,21,127],"智慧农业","农业人工智能","产量预测","可解释性",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[130],{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":132},{"candidates":133,"agricultural_evidence":48},[134,140,144],{"code":30,"score":119,"supported":48,"title":135,"abstract":137,"tags":139},[136],"computer vision",[138],"计算机视觉",[125],{"code":36,"score":116,"supported":48,"title":141,"abstract":142,"tags":143},[],[21,50],[126,21],{"code":145,"score":104,"supported":14,"title":146,"abstract":147,"tags":148},"I1",[],[],[124],[150,151],"农业人工智能 产量预测 可解释性 智慧农业","农业人工智能 产量预测","农业人工智能产量预测可解释性智慧农业-5570","10.5281\u002Fzenodo.23545079",{"doi":153,"openalex_id":155,"authors":156,"venue":112,"cited_by_count":15,"oa_url":109,"card":159,"direction":99,"ingested_from":101},"W7223114002",[157],{"name":158,"orcid":9},"Kadiatou Barry",{"tldr":160,"method":161,"finding":162,"direction":99,"opportunity":163},"推导卫星云掩膜缺口下反事实解释可靠性的理论界，并评估其对合作社储粮决策的影响。","基于Lipschitz连续性假设与掩膜扰动分析推导可靠性界，并用2000个设计单","缺口比例上升使可靠性非线性下降；ICC=0.64表明需合作社层级聚合以保持解释稳定。","可研究不同作物区云掩膜统计特性下的自适应解释暂停阈值与聚合策略。","2026-10-22T11:20:39.240791Z",{"id":166,"title":167,"url":168,"summary":169,"summary_zh":170,"content":9,"source_name":171,"source_url":168,"published_at":172,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":173,"score_detail":174,"sources":178,"tags":180,"primary_topic":184,"secondary_topics":190,"classification":193,"search_phrases":210,"slug":213,"view_count":15,"doi":214,"paper":215,"created_at":255},5178,"Spatial Modeling of Topsoil Total Carbon and Texture from Spectral Characteristics of Bare Soil Surface Images of Landsat 5\u002F8 Using Machine Learning","https:\u002F\u002Fdoi.org\u002F10.1134\u002Fs1064229326601228","— This study examines the possibility for detecting bare soil surface (BSS), using remote sensing data and for applying this approach to model texture and total soil carbon in the topsoil. The study area comprises agricultural fields within the North Caucasus Federal Agricultural Research Center in the northeastern spurs of the Stavropol Upland. Landsat 5\u002F8 imagery and spectral indices (NDVI, NBR2, and BSI) were used to classify BSS. Optimal discrimination of BSS from green and non-photosynthetic vegetation was achieved, using a BSI threshold of >0.18 (F1-score = 0.98). Spatial modeling of topsoil properties was conducted by comparing five regression models (Linear Regression, K-Nearest Neighbors, Random Forest, Extreme Gradient Boosting, and Cubist). A linear regression model demonstrated the highest performance; key predictors included the mean and standard deviation of SWIR2. The model exhibits strong predictive capability for sand, silt, clay, and total carbon (adjusted R 2 = 0.82, 0.75, 0.78, and 0.74, respectively). The topsoil is characterized by the range of clay content from 5 to 30%, of silt content from 15 to 60%, of sand content from 20 to 80%, and of total carbon from 0.5 to 3%. Predictive maps of texture and total carbon derived from BSS reveal considerable local heterogeneity, with soils consisting of Chernozems and Phaeozems, influenced by parent rock variability. The modeling results highlight the high predictive power of the BSS spectral characteristics, even when used as a single predictor for digital soil mapping.","本研究探讨了利用遥感数据检测裸露土壤表面（BSS）的可能性，并将该方法应用于表层土壤质地和总碳的建模。研究区包括位于斯塔夫罗波尔高地东北支脉的北高加索联邦农业研究中心内的农田。利用Landsat 5\u002F8影像和光谱指数（NDVI、NBR2和BSI）对BSS进行分类。当BSI阈值>0.18时，实现了BSS与绿色植被及非光合植被的最优区分（F1分数=0.98）。通过比较五种回归模型（线性回归、K近邻、随机森林、极端梯度提升和Cubist）对表层土壤属性进行了空间建模。线性回归模型表现出最高的性能；关键预测变量包括SWIR2的均值和标准差。该模型对砂粒、粉粒、黏粒和总碳表现出较强的预测能力（调整R²分别为0.82、0.75、0.78和0.74）。表层土壤的黏粒含量范围为5%至30%，粉粒含量为15%至60%，砂粒含量为20%至80%，总碳含量为0.5%至3%。基于BSS生成的质地和总碳预测图揭示了显著的局部异质性，土壤由黑钙土和暗色始成土组成，受母岩变异性的影响。建模结果凸显了BSS光谱特征的高预测能力，即使仅作为数字土壤制图的单一预测变量使用亦然。","Eurasian Soil Science","2026-10-15T00:00:00Z",69,{"impact":119,"substance":117,"depth":175,"authority":176,"freshness":116,"relevant":104,"comment":177},16,13,"方法新颖、结论可靠，但属区域尺度研究，公共影响有限，适合作为技术类精选。",[179],{"name":171,"url":168},[64,21,181,182,183],"土壤碳","Landsat","数字土壤制图",{"code":185,"slug":186,"name":187,"path":188,"parent_code":189},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",[191,192],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":194},{"candidates":195,"agricultural_evidence":48},[196,201,206],{"code":185,"score":54,"supported":48,"title":197,"abstract":198,"tags":200},[186],[199,186],"土壤",[],{"code":30,"score":202,"supported":48,"title":203,"abstract":204,"tags":205},9,[62],[],[64],{"code":36,"score":116,"supported":48,"title":207,"abstract":208,"tags":209},[],[21,50],[182,21],[211,212],"Landsat 裸土 土壤碳","北高加索 土壤质地 遥感","Landsat裸土土壤碳-5178","10.1134\u002Fs1064229326601228",{"doi":214,"openalex_id":216,"authors":217,"venue":171,"cited_by_count":15,"oa_url":9,"card":250,"direction":99,"ingested_from":101},"W7222107492",[218,221,224,227,230,233,236,239,241,244,247],{"name":219,"orcid":220},"P. A. Braulov","https:\u002F\u002Forcid.org\u002F0000-0001-9919-7402",{"name":222,"orcid":223},"Pavel M. Shilov","https:\u002F\u002Forcid.org\u002F0000-0003-4113-6396",{"name":225,"orcid":226},"R. G. Gadzhiumarov","https:\u002F\u002Forcid.org\u002F0000-0002-4220-623X",{"name":228,"orcid":229},"V. K. Dridiger","https:\u002F\u002Forcid.org\u002F0000-0002-0510-2220",{"name":231,"orcid":232},"E. V. Bogdan","https:\u002F\u002Forcid.org\u002F0009-0006-4781-5786",{"name":234,"orcid":235},"V. O. Megmerova","https:\u002F\u002Forcid.org\u002F0000-0001-8655-6437",{"name":237,"orcid":238},"Petr Tsymbarovich","https:\u002F\u002Forcid.org\u002F0000-0001-8591-2612",{"name":240,"orcid":9},"A. E. Shatunov",{"name":242,"orcid":243},"Pavel P. Fil","https:\u002F\u002Forcid.org\u002F0000-0002-9851-5381",{"name":245,"orcid":246},"D. E. Mitichkin","https:\u002F\u002Forcid.org\u002F0009-0006-2397-060X",{"name":248,"orcid":249},"Dmitriy S. Fomin","https:\u002F\u002Forcid.org\u002F0000-0003-3733-0284",{"tldr":251,"method":252,"finding":253,"direction":99,"opportunity":254},"用Landsat 5\u002F8裸土光谱指数识别裸土并机器学习建模表层土壤质地与全碳。","Landsat 5\u002F8影像、NDVI\u002FNBR2\u002FBSI指数、五种回归模型对比。","BSI阈值0.18识别裸土F1达0.98，线性回归最优，SWIR2均值与标准差为关键预测因子。","可探索多时相裸土合成与深度学习模型，提升数字土壤制图精度与跨区域迁移能力。","2026-10-17T23:30:31.485949Z",{"id":257,"title":258,"url":259,"summary":260,"summary_zh":261,"content":9,"source_name":262,"source_url":259,"published_at":263,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":264,"score_detail":265,"sources":268,"tags":270,"primary_topic":273,"secondary_topics":274,"classification":277,"search_phrases":303,"slug":306,"view_count":15,"doi":307,"paper":308,"created_at":328},5006,"Mapping Soil Salinity in Arid Agricultural Regions Using Multi-Source Remote Sensing and a Long-Range Dependency-Enhanced Deep Learning Framework","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fland15101966","Soil salinity estimation across arid agricultural regions is challenging due to pronounced environmental heterogeneity and complex dependencies among multi-source variables. This study proposes a Long-Range Dependency-Enhanced Network (LRD-Net) for regional-scale soil salinity estimation in multiple agricultural oasis regions using multi-source remote sensing and environmental covariates. A dataset comprising 180 field samples from three agricultural oasis regions in Xinjiang, China, was constructed with 25 covariates derived from spectral, topographic, climatic, and soil property data. LRD-Net consists of three stages: a multi-scale feature fusion backbone based on depthwise separable convolutions and cross-scale interaction; a Long-Range Dependency Enhancement Module (LDEM) with an asymmetric dual-branch architecture that captures forward and reverse-aware dependencies through reverse alignment and cross-directional fusion; and multi-scale attention refinement with a regression head for continuous salinity estimation. Under the region-stratified random-split evaluation, LRD-Net achieved an R2 of 0.905, an RMSE of 1.207 g·kg−1, and an MAE of 0.868 g·kg−1. Performance decreased under more spatially independent validation, with an R2 of 0.874 under spatial cross-validation and LORO test R2 values ranging from 0.739 to 0.801, indicating that prediction across spatially and geographically independent samples is more challenging than region-stratified random-split evaluation. The proposed framework effectively integrates multi-source information and captures complementary long-range dependencies, providing a potential approach for regional-scale soil salinity estimation within the studied arid agricultural oasis environments and supporting soil salinization monitoring and sustainable land management in the study region.","干旱农业区域土壤盐分估算面临显著的环境异质性和多源变量间复杂依赖关系的挑战。本研究提出了一种长程依赖增强网络（LRD-Net），利用多源遥感与环境协变量对多个农业绿洲区域进行区域尺度土壤盐分估算。研究构建了包含中国新疆三个农业绿洲区域180个野外样本的数据集，并衍生出25个协变量，涵盖光谱、地形、气候和土壤属性数据。LRD-Net由三个阶段组成：基于深度可分离卷积和跨尺度交互的多尺度特征融合骨干网络；具有非对称双分支架构的长程依赖增强模块（LDEM），通过反向对齐和跨方向融合捕获前向与反向感知依赖关系；以及多尺度注意力细化与回归头，用于连续盐分估算。在区域分层随机划分评估下，LRD-Net取得了R²为0.905、RMSE为1.207 g·kg⁻¹、MAE为0.868 g·kg⁻¹的性能。在空间独立性更强的验证下性能有所下降，空间交叉验证R²为0.874，留一区域测试R²范围为0.739至0.801，表明跨空间和地理独立样本的预测比区域分层随机划分评估更具挑战性。所提出的框架有效整合了多源信息并捕获了互补的长程依赖关系，为研究区干旱农业绿洲环境下的区域尺度土壤盐分估算提供了潜在方法，并支持研究区土壤盐渍化监测与可持续土地管理。","Land","2026-10-14T00:00:00Z",72,{"impact":119,"substance":266,"depth":118,"authority":176,"freshness":202,"relevant":104,"comment":267},21,"方法新颖、数据扎实的遥感土壤盐渍化估算研究，对干旱区盐碱地监测与治理有参考价值，但属细分领域学术进展，公共影响有限。",[269],{"name":262,"url":259},[124,125,21,271,272],"土壤盐渍化","新疆农业",{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},[275,276],{"code":185,"slug":186,"name":187,"path":188,"parent_code":189},{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":278},{"candidates":279,"agricultural_evidence":48},[280,284,288,293,299],{"code":36,"score":47,"supported":48,"title":281,"abstract":282,"tags":283},[50],[21,50],[21],{"code":185,"score":54,"supported":48,"title":285,"abstract":286,"tags":287},[186],[199,186],[],{"code":30,"score":202,"supported":48,"title":289,"abstract":291,"tags":292},[290],"deep learning",[],[125],{"code":67,"score":294,"supported":48,"title":295,"abstract":296,"tags":298},3,[],[297],"干旱",[],{"code":145,"score":104,"supported":14,"title":300,"abstract":301,"tags":302},[],[],[124],[304,305],"新疆 农业绿洲 土壤盐渍化 遥感","LRD-Net 土壤盐分 估算","新疆农业绿洲土壤盐渍化遥感-5006","10.3390\u002Fland15101966",{"doi":307,"openalex_id":309,"authors":310,"venue":262,"cited_by_count":15,"oa_url":259,"card":323,"direction":99,"ingested_from":101},"W7221816183",[311,313,316,318,320],{"name":312,"orcid":9},"Mingjie Qian",{"name":314,"orcid":315},"Liangjun Wang","https:\u002F\u002Forcid.org\u002F0009-0006-8718-8326",{"name":317,"orcid":9},"Peng Wang",{"name":319,"orcid":9},"Hangyuan Liu",{"name":321,"orcid":322},"Weitao Chen","https:\u002F\u002Forcid.org\u002F0000-0002-6272-1618",{"tldr":324,"method":325,"finding":326,"direction":99,"opportunity":327},"提出LRD-Net深度学习框架，融合多源遥感与环境协变量，估算新疆干旱绿洲区土壤盐分。","构建180个样点、25个协变量的数据集，设计含长程依赖增强模块的多尺度网络。","区域分层随机划分下R²达0.905，但空间独立验证R²降至0.739-0.874，跨区域预测更难。","可探索跨区域迁移学习与不确定性量化，提升空间独立场景下盐分制图泛化能力。","2026-10-15T23:30:29.667342Z",{"id":330,"title":331,"url":332,"summary":333,"summary_zh":334,"content":9,"source_name":335,"source_url":332,"published_at":263,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":336,"score_detail":337,"sources":341,"tags":343,"primary_topic":345,"secondary_topics":346,"classification":353,"search_phrases":405,"slug":408,"view_count":104,"doi":409,"paper":410,"created_at":433},4988,"Hierarchical cross-modal spatiotemporal attention network for crop yield prediction in complex Himalayan agroecosystems","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1933033","Introduction Accurate crop yield prediction is essential for precision agriculture, climate-resilient farm management, and timely decision-making. However, yield prediction remains challenging because of complex, nonlinear, and multi-scale interactions among crop physiology, soil conditions, weather variability, and management practices. Conventional statistical, machine learning, and deep learning approaches, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory (LSTM) networks, often have limitations in capturing long-range temporal dependencies, hierarchical spatial relationships, heterogeneous data modalities, and predictive uncertainty. Methods This study proposes a Hierarchical Cross-Modal Spatio-Temporal Attention Network (HCM-STAN) for crop yield prediction using a unified attention-based framework for integrating heterogeneous agro-environmental data. The architecture incorporates four principal components: (i) hierarchical spatio-temporal attention operating across tree, block, orchard, and regional scales; (ii) a cross-modal attention module for adaptive fusion of IoT-based soil sensor observations, weather data, plant physiological measurements, and drone-based multispectral imagery; (iii) a soil-aware attention head that explicitly models soil memory, hysteresis effects, and root-zone dynamics; and (iv) an uncertainty-aware prediction module combining Monte Carlo dropout and quantile regression to generate calibrated prediction intervals. The framework was evaluated using a multi-source dataset collected from Himalayan apple orchards during 2023–2026, comprising 1,247 monitored trees across three commercial orchards. Results The proposed HCM-STAN achieved an R 2 of 0.89 for four-week-ahead crop yield prediction, representing a 15.6% improvement over the CNN-LSTM baseline. Furthermore, zero-shot transfer to mango orchards achieved an R 2 of 0.67, corresponding to approximately 75% of the performance obtained under full training. Discussion The results demonstrate that integrating hierarchical spatio-temporal representation learning, cross-modal attention, soil-aware modelling, and uncertainty estimation can effectively address the heterogeneous and dynamic nature of crop yield prediction in complex mountain agroecosystems. The observed zero-shot transfer performance further indicates the potential of HCM-STAN to generalize across crop systems without crop-specific retraining. Overall, the proposed framework provides a promising foundation for reliable, uncertainty-aware yield forecasting and can support precision agriculture, climate-smart farm management, and data-driven decision-making in heterogeneous agroecosystems.","引言 准确的作物产量预测对于精准农业、气候适应性农场管理以及及时决策至关重要。然而，由于作物生理、土壤条件、天气变异和管理措施之间复杂、非线性且多尺度的相互作用，产量预测仍然具有挑战性。传统的统计方法、机器学习以及深度学习方法，包括卷积神经网络（CNN）、循环神经网络（RNN）和长短期记忆（LSTM）网络，在捕捉长程时间依赖关系、层次化空间关系、异构数据模态以及预测不确定性方面往往存在局限。方法 本研究提出了一种用于作物产量预测的层次化跨模态时空注意力网络（HCM-STAN），采用统一的基于注意力的框架来整合异构农业环境数据。该架构包含四个主要组成部分：（i）在树、地块、果园和区域尺度上运行的层次化时空注意力；（ii）用于自适应融合基于物联网的土壤传感器观测、天气数据、植物生理测量和无人机多光谱影像的跨模态注意力模块；（iii）显式建模土壤记忆、滞后效应和根区动态的土壤感知注意力头；（iv）结合蒙特卡洛 dropout 和分位数回归以生成校准预测区间的不确定性感知预测模块。该框架使用2023—2026年在喜马拉雅苹果园采集的多源数据集进行了评估，该数据集包含三个商业果园中1，247棵监测树木。结果 所提出的 HCM-STAN 在提前四周作物产量预测中达到了0.89的 R²，相较于 CNN-LSTM 基线提升了15.6%。此外，零样本迁移至芒果园达到了0.67的 R²，相当于完整训练下所获性能的约75%。讨论 结果表明，整合层次化时空表征学习、跨模态注意力、土壤感知建模和不确定性估计，能够有效应对复杂山地农业生态系统中作物产量预测的异构性和动态性。观察到的零样本迁移性能进一步表明，HCM-STAN 具有在无需作物特定再训练的情况下跨作物系统泛化的潜力。总体而言，该 p","Frontiers in Plant Science",81,{"impact":338,"substance":339,"depth":338,"authority":54,"freshness":202,"relevant":104,"comment":340},18,22,"提出分层跨模态时空注意力网络，融合物联网土壤传感、气象、生理与无人机多光谱数据，苹果园四周产量预测R²达0.89并实现芒果园零样本迁移，方法新颖、数据扎实，对智慧农业产量预报有参考价值。",[342],{"name":335,"url":332},[124,125,126,344,21],"多模态融合",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[347,352],{"code":348,"slug":349,"name":350,"path":351,"parent_code":27},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit",{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":354},{"candidates":355,"agricultural_evidence":48},[356,362,370,376,383,387,394,400],{"code":30,"score":357,"supported":48,"title":358,"abstract":359,"tags":361},11,[],[360,64,62,290],"深度学习",[125,344],{"code":348,"score":202,"supported":48,"title":363,"abstract":364,"tags":369},[],[365,366,367,368],"苹果","芒果","orchard","apple",[],{"code":36,"score":116,"supported":48,"title":371,"abstract":372,"tags":375},[],[373,374],"无人机多光谱","multispectral",[126,21],{"code":145,"score":377,"supported":48,"title":378,"abstract":379,"tags":382},7,[],[380,381],"精准农业","precision agriculture",[124],{"code":185,"score":68,"supported":48,"title":384,"abstract":385,"tags":386},[],[199,186],[],{"code":388,"score":68,"supported":48,"title":389,"abstract":390,"tags":393},"I5",[],[391,392],"物联网","IoT",[],{"code":395,"score":294,"supported":48,"title":396,"abstract":397,"tags":399},"H6",[],[398],"农业生态",[],{"code":67,"score":294,"supported":48,"title":401,"abstract":402,"tags":404},[],[403],"气候适应",[],[406,407],"喜马拉雅 苹果园 产量预测","HCM-STAN 作物产量预测","喜马拉雅苹果园产量预测-4988","10.3389\u002Ffpls.2026.1933033",{"doi":409,"openalex_id":411,"authors":412,"venue":335,"cited_by_count":15,"oa_url":332,"card":426,"direction":432,"ingested_from":101},"W7221795850",[413,416,418,420,422,424],{"name":414,"orcid":415},"Shabir Ahmad Mir","https:\u002F\u002Forcid.org\u002F0000-0003-2311-218X",{"name":417,"orcid":9},"Umarfarooque Momin",{"name":419,"orcid":9},"Sudeep Kumar E",{"name":421,"orcid":9},"Itfaq Ahmad Mir",{"name":423,"orcid":9},"Ab Rouf Khan",{"name":425,"orcid":9},"Ghulam Mohammad Mir",{"tldr":427,"method":428,"finding":429,"direction":430,"opportunity":431},"提出分层跨模态时空注意力网络HCM-STAN，融合多源农业数据预测喜马拉雅山区苹果产量。","用IoT土壤、气象、生理和无人机多光谱数据，结合分层时空与跨模态注意力及不确定性","四周前产量预测R²达0.89，比CNN-LSTM提升15.6%，零样本迁移芒果R²为0.67。","农业人工智能与决策模型","可探索跨作物零样本迁移的领域自适应机制，并验证模型在更多山地农业生态系统的泛化能力。","智慧农业 \u002F 农业物联网","2026-10-15T23:30:08.750140Z",{"id":435,"title":436,"url":437,"summary":438,"summary_zh":439,"content":9,"source_name":112,"source_url":437,"published_at":440,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":441,"score_detail":442,"sources":444,"tags":446,"primary_topic":449,"secondary_topics":450,"classification":452,"search_phrases":480,"slug":483,"view_count":15,"doi":484,"paper":485,"created_at":506},4505,"Application of Deep Learning Techniques for Vineyard Mapping and Characterization Using Multispectral UAV Imagery: A Case Study from Northern Greece","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23232412","Developments in unmanned aerial vehicle (UAV) technology and the rise of Deep Learning (DL) have drastically boosted the capacities of precision viticulture to monitor the environments of farms in a scalable and detailed manner. Vineyard mappings and extraction of vine-specific features are crucial for yield estimation, plant health assessment, and management decisions. This study examines the application of Deep Learning-based image analysis techniques for vineyard mapping and characterization using multispectral UAV image data (Deep Learning). The proposed approach integrates spatial and spectral information to delineate vineyard structures and capture key canopy characteristics relevant to precision agriculture. Multimodal datasets, including RGB and multispectral bands, facilitate improved discrimination between vines and surrounding elements. Geospatial analysis and data processing are further performed through ArcGIS Pro, enabling spatial validation of results, identifying extracted features, and creating appropriate interpretative thematic layers. The combination of UAV imagery, DL tools and GIS based analysis offers a strong and flexible workflow that can be adjusted for different grape cultivars, growth conditions and circumstances. Overall, the study demonstrates the potential benefits of the rigid union of UAV-based remote sensing with Deep Learning and GIS techniques to create a flexible, operational, and data-driven framework for vineyard management. Based on this, sustainable and fruitful management practices can be elaborated, thus optimizing resources and vineyard management decisions in precision viticulture,underlining the importance of such a technological integration for monitoring and managing modern vineyards.","无人机（UAV）技术的发展与深度学习（DL）的兴起，极大地提升了精准葡萄栽培以可扩展且精细的方式监测农场环境的能力。葡萄园制图与葡萄植株特异性特征的提取对于产量估算、植株健康评估和管理决策至关重要。本研究探讨了基于深度学习的图像分析技术在多光谱无人机影像数据（深度学习）用于葡萄园制图与表征中的应用。所提出的方法整合了空间与光谱信息，以 delineate 葡萄园结构并捕捉与精准农业相关的关键冠层特征。多模态数据集（包括RGB和多光谱波段）有助于更好地区分葡萄藤与周围元素。通过ArcGIS Pro进一步进行地理空间分析与数据处理，从而实现结果的空间验证、提取特征的识别以及适当的解释性专题图层的创建。无人机影像、深度学习工具与基于GIS的分析相结合，提供了一套强大且灵活的工作流程，可针对不同葡萄品种、生长条件和环境进行调整。总体而言，本研究展示了将基于无人机的遥感与深度学习和GIS技术 rigidly 结合，以创建灵活、可操作且数据驱动的葡萄园管理框架的潜在优势。在此基础上，可制定可持续且富有成效的管理实践，从而优化精准葡萄栽培中的资源与葡萄园管理决策，凸显了这种技术整合对现代葡萄园监测与管理的重要性。","2026-10-08T00:00:00Z",68,{"impact":119,"substance":338,"depth":175,"authority":176,"freshness":202,"relevant":104,"comment":443},"希腊北部葡萄园的多光谱无人机深度学习制图案例，方法新颖但属区域性研究，对国内智慧农业有参考价值。",[445],{"name":112,"url":437},[124,447,125,21,448],"无人机","葡萄园管理",{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},[451],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":453},{"candidates":454,"agricultural_evidence":48},[455,459,463,467,473],{"code":36,"score":338,"supported":48,"title":456,"abstract":457,"tags":458},[374],[21,50,374],[21],{"code":30,"score":47,"supported":48,"title":460,"abstract":461,"tags":462},[290],[360,290],[125],{"code":145,"score":377,"supported":48,"title":464,"abstract":465,"tags":466},[],[380,381],[124],{"code":348,"score":68,"supported":48,"title":468,"abstract":469,"tags":472},[],[470,471],"葡萄","grape",[],{"code":474,"score":68,"supported":48,"title":475,"abstract":476,"tags":479},"I4",[],[477,478],"空间分析","GIS",[],[481,482],"希腊 葡萄园 无人机 深度学习","多光谱 UAV 葡萄园 制图","希腊葡萄园无人机深度学习-4505","10.5281\u002Fzenodo.23232412",{"doi":484,"openalex_id":486,"authors":487,"venue":112,"cited_by_count":15,"oa_url":437,"card":501,"direction":99,"ingested_from":101},"W7220993218",[488,490,493,495,498],{"name":489,"orcid":9},"Dimitra Passa",{"name":491,"orcid":492},"Spyridon E. Detsikas","https:\u002F\u002Forcid.org\u002F0000-0002-3960-9361",{"name":494,"orcid":9},"Georgios Nektarios Tselos",{"name":496,"orcid":497},"George P. Petropoulos","https:\u002F\u002Forcid.org\u002F0000-0003-1442-1423",{"name":499,"orcid":500},"Antonis Litke","https:\u002F\u002Forcid.org\u002F0000-0002-7658-2559",{"tldr":502,"method":503,"finding":504,"direction":99,"opportunity":505},"用多光谱无人机影像和深度学习实现葡萄园制图与冠层特征提取。","多光谱无人机影像、RGB与多光谱融合、深度学习、ArcGIS Pro空间分析。","UAV+DL+GIS可灵活构建葡萄园精准管理的数据驱动框架。","可探索跨品种、跨生长阶段的迁移学习与轻量化模型，提升葡萄园实时监测泛化能力。","2026-10-09T23:30:25.912664Z",{"id":508,"title":509,"url":510,"summary":511,"summary_zh":9,"content":9,"source_name":512,"source_url":510,"published_at":440,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":513,"score_detail":514,"sources":516,"tags":518,"primary_topic":521,"secondary_topics":526,"classification":529,"search_phrases":548,"slug":551,"view_count":104,"doi":552,"paper":553,"created_at":585},4500,"Integrating UAV-Based Remote Sensing and Deep Learning for Early Detection of Bacterial Leaf Blight in Rice","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-11163942\u002Fv1","Integrating UAV-Based Remote Sensing and Deep Learning for Early Detection of Bacterial Leaf Blight in Rice。Research Square","Research Square",74,{"impact":338,"substance":117,"depth":338,"authority":13,"freshness":116,"relevant":104,"comment":515},"将无人机遥感与深度学习结合用于水稻白叶枯病早期检测，方法有创新、结论具应用价值，但属预印本且为细分技术进展，适合进入主题聚合而非头条精选。",[517],{"name":512,"url":510},[124,125,519,21,520],"水稻","病害检测",{"code":522,"slug":523,"name":519,"path":524,"parent_code":525},"A1","rice","\u002Ftopic\u002Ffield-crops\u002Frice","A",[527,528],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"code":36,"slug":37,"name":38,"path":39,"parent_code":34},{"status":41,"locked":14,"version":42,"method":43,"evidence":530},{"candidates":531,"agricultural_evidence":48},[532,536,540,544],{"code":522,"score":119,"supported":48,"title":533,"abstract":534,"tags":535},[523],[523],[519],{"code":30,"score":119,"supported":48,"title":537,"abstract":538,"tags":539},[290],[290],[125],{"code":36,"score":119,"supported":48,"title":541,"abstract":542,"tags":543},[50],[50],[21],{"code":145,"score":104,"supported":14,"title":545,"abstract":546,"tags":547},[],[],[124],[549,550],"无人机遥感 水稻 白叶枯病","深度学习 水稻病害 早期检测","无人机遥感水稻白叶枯病-4500","10.21203\u002Frs.3.rs-11163942\u002Fv1",{"doi":552,"openalex_id":554,"authors":555,"venue":512,"cited_by_count":15,"oa_url":510,"card":9,"direction":99,"ingested_from":101},"W7221055917",[556,559,562,564,566,569,571,573,576,578,580,582],{"name":557,"orcid":558},"Carlito B. Balingbing","https:\u002F\u002Forcid.org\u002F0000-0002-3672-9253",{"name":560,"orcid":561},"Dmytro Chebotarov","https:\u002F\u002Forcid.org\u002F0000-0003-1351-9453",{"name":563,"orcid":9},"Ian  Carlo Perilla",{"name":565,"orcid":9},"Stephen Klassen",{"name":567,"orcid":568},"Nancy Castilla","https:\u002F\u002Forcid.org\u002F0000-0002-9143-4933",{"name":570,"orcid":9},"Raphael  Gabriel Carvajal",{"name":572,"orcid":9},"Kendrix Ortega",{"name":574,"orcid":575},"Joseph M. Sandro","https:\u002F\u002Forcid.org\u002F0000-0002-8311-2299",{"name":577,"orcid":9},"Nurmi Pangesti",{"name":579,"orcid":9},"Caesar Arloo R. Centeno",{"name":581,"orcid":9},"Arnel Rala",{"name":583,"orcid":584},"Li-Yu  Daisy Liu","https:\u002F\u002Forcid.org\u002F0000-0001-6997-8101","2026-10-09T23:30:18.390549Z"]