[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"daily-2026-10-01":3},{"date":4,"title":5,"highlights":6,"content":13,"items":14},"2026-10-01","农业农村日报：智慧农业与绿色低碳前沿",[7,8,9,10,11,12],"可解释植物健康监测可穿戴系统问世，知识增强方法弥合精准农业“可解释性鸿沟”，为田间诊断提供新工具。","欧洲化肥价格2021—2024年经历第二轮大涨大跌，研究揭示投入品用量与能源需求的调整节奏“削减快、恢复慢”。","碳标签与碳交易被用于海南山地传统水稻保护，探索以市场化激励替代单一生态补偿、守住农业生物多样性。","树莓派轻量深度学习模型结合PhenoCam图像实现玉米生育期自动分类，低成本方案有望替代人工田间巡查。","空天地一体化网络综述提出SDN\u002FNFV智能资源管理，支撑智慧农业与智慧育种走向分布式数据闭环。","多篇研究聚焦遥感：橡胶林碳储量建模、Sentinel-2土壤退化评估、西班牙干旱产量响应及塞内加尔小农户作物制图。","本期汇编12篇论文，覆盖植物健康可穿戴监测、化肥能源价格周期、碳标签与碳交易、轻量深度学习作物分类、空天地一体化网络、橡胶林碳建模、遥感土壤退化与干旱评估、小农户作物制图、畜牧AI环境价值及气候价值分叉模型等方向。\n\n---\n*本日报内容整理自公开来源，学术论文元数据来自 OpenAlex 等开放接口；外文资料已译为中文，翻译与摘要仅供参考；引用与决策请以官方原文与正式出版物为准。*",[15,116,185,275,368,441,533,607,691,762,817,878,937,1028,1090],{"id":16,"title":17,"url":18,"summary":19,"summary_zh":20,"content":21,"source_name":22,"source_url":18,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":26,"score":27,"score_detail":28,"sources":36,"tags":38,"primary_topic":44,"secondary_topics":50,"classification":51,"search_phrases":72,"slug":75,"view_count":76,"doi":77,"paper":78,"created_at":115},3864,"A Knowledge‐Augmented Wearable System for Interpretable Plant Health Monitoring","https:\u002F\u002Fdoi.org\u002F10.1002\u002Ffle2.70020","ABSTRACT In precision agriculture, modern sensing technologies can autonomously collect and process vast plant data using statistical or machine‐learning models. However, an “interpretability gap” persists: these outputs often remain black‐box correlations rather than interpretable physiological insights, which means users cannot validate their credibility, trace causal links to plant physiology, or translate them into actionable decisions. To bridge this gap, we propose a knowledge‐augmented wearable system for interpretable plant health monitoring (K‐Wear). By coupling a soft wearable sensing platform with a knowledge‐guided reasoning framework, K‐Wear leverages a large language model to transform raw sensing signals into mechanistic diagnoses and recommendations for proactive management. In simulated evaluations, the system achieved 99% diagnostic accuracy and first correctly detected plant stress on day 4.95 on average. In a real 14‐day potassium‐deficiency monitoring experiment, K‐Wear successfully identified early stress signals on day 4 and generated targeted intervention recommendations, enabling timely treatment and recovery of plant health.","在精准农业中，现代传感技术能够利用统计或机器学习模型自主采集并处理海量植物数据。然而，“可解释性鸿沟”依然存在：这些输出往往仍是黑箱式的相关性结果，而非可解释的生理学洞见，这意味着用户无法验证其可信度、追溯其与植物生理的因果联系，或将其转化为可操作的决策。为弥合这一鸿沟，我们提出了一种用于可解释植物健康监测的知识增强可穿戴系统（K‐Wear）。通过将柔性可穿戴传感平台与知识引导的推理框架相结合，K‐Wear利用大语言模型将原始传感信号转化为机理诊断和主动管理建议。在模拟评估中，该系统达到了99%的诊断准确率，并最早在平均第4.95天正确检测到植物胁迫。在一项为期14天的真实缺钾监测实验中，K‐Wear成功在第4天识别出早期胁迫信号，并生成有针对性的干预建议，从而使植物健康得到及时治疗和恢复。",null,"FlexTech","2026-09-29T00:00:00Z","论文",10,true,84,{"impact":29,"substance":30,"depth":31,"authority":32,"freshness":33,"relevant":34,"comment":35},22,23,18,13,8,1,"将可穿戴传感与大模型知识推理结合，实现可解释的植物健康诊断，方法新颖、实验数据扎实，对智慧农业感知决策有参考价值。",[37],{"name":22,"url":18},[39,40,41,42,43],"智慧农业","农业人工智能","大模型","可穿戴传感","植物健康监测",{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":52,"locked":53,"version":54,"method":55,"evidence":56},"classified",false,"2026-10-22.2","rules",{"candidates":57,"agricultural_evidence":26},[58,64],{"code":45,"score":33,"supported":26,"title":59,"abstract":60,"tags":63},[],[61,62],"机器学习","large language model",[40,41],{"code":65,"score":66,"supported":26,"title":67,"abstract":68,"tags":71},"I1",7,[],[69,70],"精准农业","precision agriculture",[39],[73,74],"K-Wear 可穿戴 植物健康监测","大语言模型 植物胁迫 诊断","K-Wear可穿戴植物健康监测-3864",4,"10.1002\u002Ffle2.70020",{"doi":77,"openalex_id":79,"authors":80,"venue":22,"cited_by_count":106,"oa_url":18,"card":107,"direction":113,"ingested_from":114},"W7214803187",[81,83,85,87,90,92,95,97,100,103],{"name":82,"orcid":21},"Xingxing Dong",{"name":84,"orcid":21},"Yihan Luo",{"name":86,"orcid":21},"Jiang Qin",{"name":88,"orcid":89},"Jie Ye","https:\u002F\u002Forcid.org\u002F0000-0001-7728-3588",{"name":91,"orcid":21},"Yihui Fan",{"name":93,"orcid":94},"Menghao Pu","https:\u002F\u002Forcid.org\u002F0009-0003-1421-6785",{"name":96,"orcid":21},"Zisheng Zong",{"name":98,"orcid":99},"Guoqiang Xu","https:\u002F\u002Forcid.org\u002F0009-0007-5481-1422",{"name":101,"orcid":102},"Han Ding","https:\u002F\u002Forcid.org\u002F0000-0003-2905-6369",{"name":104,"orcid":105},"Zhigang Wu","https:\u002F\u002Forcid.org\u002F0000-0002-3719-406X",0,{"tldr":108,"method":109,"finding":110,"direction":111,"opportunity":112},"提出知识增强可穿戴系统K-Wear，将植物传感信号转化为可解释诊断与建议。","软可穿戴传感平台+知识引导推理框架+大语言模型。","模拟诊断准确率99%，真实钾缺乏实验第4天检出早期胁迫并给出干预建议。","智慧农业 \u002F 农业物联网","可探索知识增强LLM推理在更多作物胁迫类型与田间长期监测中的泛化能力。","农业人工智能与决策模型","openalex","2026-09-30T23:30:55.022226Z",{"id":117,"title":118,"url":119,"summary":120,"summary_zh":121,"content":21,"source_name":122,"source_url":119,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":26,"score":123,"score_detail":124,"sources":129,"tags":131,"primary_topic":137,"secondary_topics":143,"classification":144,"search_phrases":166,"slug":169,"view_count":170,"doi":171,"paper":172,"created_at":184},3816,"Cut fast, restore slowly: fertiliser and energy demand in European agriculture through the 2021–2024 input price cycle, with evidence from the 2008 episode","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1976070","Introduction The EU fertiliser price index rose from 129 in 2021 to 244 in 2022 (2020 = 100) and fell back to 152 by 2024, the second such cycle in 15 years. This study investigates how much input use was reduced when prices spiked, how much was restored when they fell, and whether slow restoration reflects a ratchet effect or lagged symmetric adjustment to a price level that never returned to its base. Methods Three sources are combined: the regional economic accounts for agriculture at previous-year prices, yielding price and volume indices for five intermediate inputs in 210 NUTS 2 regions of 20 countries (3,060 observations, 2021–2024, with price variation in 300 country-input-year cells); national statistics on mineral nitrogen and phosphorus consumption; and spliced national fertiliser price indices for 1996–2025. Fixed-effects regressions compare inputs within the same region and season; sign-split and error-correction regressions on a 24-country panel for 2002–2024 test whether the within-cycle pattern generalises. Results The short-run price–volume elasticity is −0.33 (standard error 0.07), indistinguishable across inputs and far from unity. Nitrogen use fell by 13.3 log per cent in 2022, and only 21 per cent of it had been restored by 2024, despite prices having returned three quarters of the spike; regions that cut fertiliser more deeply recorded lower crop output, an association that the 2022 drought confounds. Within the cycle, the sign-split regressions detect a response to price increases and a weaker one to decreases, a borderline difference sensitive to the inferential method. Over 2002–2024, when 42 per cent of price changes were decreases, the difference is insignificant in every specification that contains the error-correction term, which is strongly present (adjustment speed 0.28 per year, 0.17 to 0.18 after bias correction, bootstrap p = 0.001) and survives relative prices, year effects and standard errors robust to cross-country dependence. Discussion Slow restoration is consistent with lagged symmetric adjustment to a still-elevated price level; the data provide no convincing evidence of a ratchet effect. The 2022 cut was nevertheless contemporaneous and twice as deep as in 2008–2009. An illustrative scenario places 2025 nitrogen use 8 to 12 per cent below 2021.","引言 欧盟肥料价格指数从2021年的129上升至2022年的244（2020年=100），到2024年回落至152，这是15年来的第二个此类周期。本研究考察了价格飙升时投入品使用量减少了多少，价格回落后又恢复了多少，以及恢复缓慢反映的是棘轮效应，还是对从未回归基期的价格水平的滞后对称调整。方法 本研究综合了三个数据来源：按上年价格计算的区域农业经济账户，得出20个国家210个NUTS 2区域五种中间投入品的价格和数量指数（3，060个观测值，2021—2024年，价格变异来自300个国家—投入品—年份单元）；矿物氮和磷消费量的国家统计数据；以及1996—2025年拼接的国家肥料价格指数。固定效应回归比较同一区域和季节内的投入品；对2002—2024年24国面板数据的符号分解回归和误差修正回归检验周期内模式是否具有普遍性。结果 短期价格—数量弹性为−0.33（标准误0.07），各投入品之间无显著差异，且远非单位弹性。2022年氮使用量下降了13.3个对数百分点，尽管价格已从峰值回落了四分之三，到2024年也仅恢复了21%；削减肥料更深的区域录得较低的作物产量，但这一关联受到2022年干旱的混淆。在周期内，符号分解回归检测到对价格上涨的响应以及对价格下降的较弱响应，这一差异处于临界水平且对推断方法敏感。在2002—2024年间，当42%的价格变化为下降时，在包含误差修正项的所有设定中该差异均不显著；误差修正项显著存在（调整速度为每年0.28，偏差校正后为0.17至0.18，自助法p=0.001），且在相对价格、年份效应和对跨国依赖稳健的标准误下依然成立。讨论 恢复缓慢与对仍然偏高的价格水平的滞后对称调整相一致；数据未提供令人信服的棘轮效应证据。然而，2022年的削减是同期发生的，且幅度是2008—2009年的两倍。一个示例性情景预测2025年氮使用量将比2021年低8%至12%。","Frontiers in Sustainable Food Systems",83,{"impact":31,"substance":30,"depth":125,"authority":126,"freshness":127,"relevant":34,"comment":128},19,14,9,"基于210个欧盟NUTS 2区域面板数据，量化化肥涨价期投入削减与恢复的不对称性，结论扎实、政策参考价值高，值得进入每日精选。",[130],{"name":122,"url":119},[132,133,134,135,136],"粮食安全","欧盟农业","化肥价格","氮肥减量","投入品价格",{"code":138,"slug":139,"name":140,"path":141,"parent_code":142},"H7","climate","气候适应与防灾减灾","\u002Ftopic\u002Fenvironment\u002Fclimate","H",[],{"status":52,"locked":53,"version":54,"method":55,"evidence":145},{"candidates":146,"agricultural_evidence":26},[147,154,161],{"code":138,"score":148,"supported":26,"title":149,"abstract":150,"tags":153},6,[],[151,152],"干旱","drought",[],{"code":155,"score":156,"supported":26,"title":157,"abstract":158,"tags":160},"F3",3,[],[159],"肥料",[],{"code":162,"score":34,"supported":53,"title":163,"abstract":164,"tags":165},"A8",[],[],[132],[167,168],"欧盟 化肥价格 氮肥用量","NUTS 2 农业投入品 价格弹性","欧盟化肥价格氮肥用量-3816",2,"10.3389\u002Ffsufs.2026.1976070",{"doi":171,"openalex_id":173,"authors":174,"venue":122,"cited_by_count":106,"oa_url":119,"card":178,"direction":182,"ingested_from":114},"W7214770501",[175],{"name":176,"orcid":177},"Aras Yolusever","https:\u002F\u002Forcid.org\u002F0000-0001-9810-2571",{"tldr":179,"method":180,"finding":181,"direction":182,"opportunity":183},"研究欧洲农业在2021-2024化肥价格周期中投入品削减与恢复的不对称性。","用NUTS2区域农业经济账户、氮磷消费和化肥价格指数做固定效应与误差修正回归。","短期价格弹性-0.33，2022年氮肥用量降13.3%，到2024年仅恢复21%，无棘轮效应。","农业绿色发展与碳","可探究投入品恢复滞后对作物产量和农业碳排放的长期影响，以及区域差异化补贴政策。","2026-09-30T23:30:07.311518Z",{"id":186,"title":187,"url":188,"summary":189,"summary_zh":190,"content":21,"source_name":122,"source_url":188,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":26,"score":191,"score_detail":192,"sources":194,"tags":196,"primary_topic":203,"secondary_topics":209,"classification":220,"search_phrases":246,"slug":249,"view_count":170,"doi":250,"paper":251,"created_at":274},3815,"Carbon labels and carbon trading as market-based incentives for conserving traditional upland rice: a case study from Hainan, China","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1969345","Introduction Traditional crop varieties that underpin agricultural biodiversity are rapidly disappearing as their cultivation becomes economically marginalized. Public ecological compensation has proven effective in incentivizing on-farm conservation, but it faces long-term fiscal sustainability constraints. Market-based carbon mechanisms including carbon labeling and carbon trading are increasingly proposed as complementary incentives, but real-world documented cases in smallholder traditional agroecosystems remain extremely limited. This study addresses this gap by analyzing an operational pilot practice launched for the globally important Shanlan upland rice system in Hainan, China. Methods Drawing on program documents, field interview records, and official verification data from the UNDP-GEF C-SAP3 2020–2025 project, this study systematically evaluates the full operational logic, implementation process and actual performance of the combined carbon label and carbon trading mechanism. A comparative analysis framework is constructed to compare the incentive strength, operation cost and institutional positioning of market-based carbon instruments and existing differentiated public ecological compensation schemes. Results The Shanlan rice conservation system has established a verified carbon sequestration rate of 2.1 t CO₂\u002Fha\u002Fyr, completed the first provincial carbon market transaction of RMB 530,000 in 2024, and officially released the dedicated Shanlan rice carbon label at the 2024 Hainan Winter Trade Fair. At the current carbon price of RMB 50\u002Ft, the annual carbon revenue per mu is only about RMB 7, accounting for roughly 3% of the existing public ecological compensation standard of RMB 200–220\u002Fmu. The carbon accounting parameter has already been embedded into the local ecological compensation calculation algorithm as an ecological service value component. Discussion The empirical results confirm that current carbon market instruments can only play a complementary rather than substitution role for public ecological compensation in traditional crop conservation scenarios, and their near-term value is more institutional than purely financial. The ambiguity between modeled sequestration parameters and actual measured conservation outcomes still limits the credibility of such market mechanisms. Three core propositions including incentive complementarity, measurement validation and long-term institutional stewardship are proposed to support future comparative research on similar agrobiodiversity conservation practices.","引言 支撑农业生物多样性的传统作物品种正因其种植在经济上被边缘化而迅速消失。公共生态补偿已被证明能有效激励农田就地保护，但面临长期财政可持续性约束。包括碳标签和碳交易在内的市场化碳机制日益被提议作为补充性激励手段，但在小农传统农业生态系统中，实际有据可查的案例仍然极为有限。本研究通过分析针对中国海南具有全球重要性的山栏稻旱作系统所开展的一项运行中的试点实践，来填补这一空白。方法 本研究依据联合国开发计划署—全球环境基金C-SAP3 2020—2025项目的项目文件、实地访谈记录和官方核证数据，系统评估碳标签与碳交易相结合机制的完整运行逻辑、实施过程和实际绩效。构建比较分析框架，对市场化碳工具与现有差异化公共生态补偿方案在激励强度、运行成本和制度定位方面进行比较。结果 山栏稻保护系统已确立经核证的碳汇速率为2.1吨二氧化碳\u002F公顷\u002F年，于2024年完成首笔省级碳市场交易53万元人民币，并在2024年海南冬交会上正式发布专用山栏稻碳标签。按当前碳价50元\u002F吨计算，每亩年度碳收益仅约7元人民币，约占现有公共生态补偿标准200—220元\u002F亩的3%。碳核算参数已作为生态服务价值组成部分嵌入地方生态补偿计算算法。讨论 实证结果证实，在当前传统作物保护情景下，碳市场工具只能对公共生态补偿发挥补充而非替代作用，其近期价值更多体现在制度层面而非纯粹财务层面。模型化碳汇参数与实际测得的保护成效之间的模糊性，仍限制此类市场机制的可信度。提出激励互补性、计量验证和长期制度管护三项核心命题，以支持未来比较研究。",82,{"impact":31,"substance":30,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":193},"以海南山栏稻为案例，首次量化碳标签与碳交易对传统作物保护的激励强度（约7元\u002F亩，仅为生态补偿3%），数据扎实、结论明确，对农业碳汇与生物多样性保护政策具参考价值。",[195],{"name":122,"url":188},[197,198,199,200,201,202],"农业碳汇","碳交易","生态补偿","碳标签","山栏稻","农业生物多样性",{"code":204,"slug":205,"name":206,"path":207,"parent_code":208},"A1","rice","水稻","\u002Ftopic\u002Ffield-crops\u002Frice","A",[210,215],{"code":211,"slug":212,"name":213,"path":214,"parent_code":142},"H4","carbon","农业碳与绿色低碳","\u002Ftopic\u002Fenvironment\u002Fcarbon",{"code":216,"slug":217,"name":218,"path":219,"parent_code":142},"H6","biodiversity","生态保护与生物多样性","\u002Ftopic\u002Fenvironment\u002Fbiodiversity",{"status":52,"locked":53,"version":54,"method":55,"evidence":221},{"candidates":222,"agricultural_evidence":26},[223,228,234,240],{"code":204,"score":224,"supported":26,"title":225,"abstract":226,"tags":227},11,[205],[205],[],{"code":216,"score":25,"supported":26,"title":229,"abstract":230,"tags":233},[],[231,232,217],"生物多样性","农业生态",[199],{"code":211,"score":33,"supported":26,"title":235,"abstract":236,"tags":239},[],[237,238],"碳汇","carbon sequestration",[197,198],{"code":241,"score":156,"supported":26,"title":242,"abstract":243,"tags":245},"G2",[],[244],"smallholder",[],[247,248],"海南 山栏稻 碳标签","山栏稻 碳交易 生态补偿","海南山栏稻碳标签-3815","10.3389\u002Ffsufs.2026.1969345",{"doi":250,"openalex_id":252,"authors":253,"venue":122,"cited_by_count":106,"oa_url":188,"card":269,"direction":182,"ingested_from":114},"W7214741801",[254,256,258,261,263,265,267],{"name":255,"orcid":21},"Qimin Cao",{"name":257,"orcid":21},"易伟鹏",{"name":259,"orcid":260},"Tao Guo","https:\u002F\u002Forcid.org\u002F0000-0003-0587-4762",{"name":262,"orcid":21},"Chuanliang Fu",{"name":264,"orcid":21},"Zhijun Zhang",{"name":266,"orcid":21},"Wen Zhang",{"name":268,"orcid":21},"Yujia Wu",{"tldr":270,"method":271,"finding":272,"direction":182,"opportunity":273},"分析海南山栏稻碳标签与碳交易试点，评估其作为传统作物保护市场激励的效果。","基于UNDP-GEF项目文件、访谈与核证数据，对比碳机制与公共生态补偿。","碳收益仅约7元\u002F亩，为公共补偿的3%，只能补充而非替代公共补偿。","小农户传统农业碳汇计量与碳市场激励的长期制度设计及实测验证研究。","2026-09-30T23:30:07.250289Z",{"id":276,"title":277,"url":278,"summary":279,"summary_zh":280,"content":21,"source_name":281,"source_url":278,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":26,"score":282,"score_detail":283,"sources":285,"tags":287,"primary_topic":291,"secondary_topics":295,"classification":297,"search_phrases":328,"slug":331,"view_count":332,"doi":333,"paper":334,"created_at":367},3842,"Raspberry Pi Lightweight Deep Learning Models Deployed for Corn Growth Stage Classification Using PhenoCam Images","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193337","Identifying crop growth stages is essential for comprehensive assessment that facilitates timely and precise agronomic management. However, conventional field-based manual scouting is labor-intensive, time-consuming, and lacks scalability. Digital field images with advanced models deployed on portable edge-computing devices offer an alternative solution for real-time crop monitoring and decision support. Therefore, in this study, the standard 8 corn growth stages were defined as 8 classes based on distinct visual canopy appearance patterns observed in PhenoCam (near-surface [proximal] remote sensing network camera) imagery collected from 10 PhenoCam sites. These classes do not correspond directly to standard agronomic vegetation and reproductive growth stages. For modeling, four lightweight deep learning (DL) models, ELiteCrop0, ELiteCrop1, ELiteCrop4, and MobNetCropV2, were developed and evaluated across five image vertical clipping levels (0–40%) using a supercomputer (NDSU CCAST). The optimized model was subsequently deployed on a single-board Raspberry Pi 5 computer for edge inference. Model training accounted for the majority of the total CPU time, exceeding 97%, while the testing times ranged from 0.01–0.12 min, enabling real-time applications. Among the models, ELiteCrop0 achieved the most balanced performance with a confusion-matrix diagonal ratio (CMDR) of 0.93, followed by ELiteCrop1 (CMDR = 0.92). Overall, model performance decreased with increasing vertical clipping; therefore, a moderate image clipping (0–10%) was recommended for improved computational efficiency. Analysis with a supercomputer produced an intrasite (same training sites) accuracy of 0.90–0.93 (Raspberry Pi: 0.78–0.81) and an intersite (unseen test sites) accuracy of 0.48–0.50 (Raspberry Pi: 0.41–0.43), indicating challenges with model generalization. Raspberry Pi processed ≈ 1000 images\u002Fmin at 68 °C, with a PhenoCam:Raspberry Pi processing ratio of 1:2500 at 50% capacity over an 8 h day. Overall, this study presents a scalable and cost-effective solution for real-time corn growth stage monitoring in precision agriculture, which could be extended to other crops.","识别作物生长阶段对于全面评估、促进及时精准的农艺管理至关重要。然而，传统的田间人工巡查劳动强度大、耗时且缺乏可扩展性。部署在便携式边缘计算设备上的数字田间图像与先进模型为实时作物监测和决策支持提供了替代方案。因此，本研究基于从10个PhenoCam（近地面[近端]遥感网络相机）站点采集的影像中观察到的不同冠层视觉外观模式，将标准的8个玉米生长阶段定义为8个类别。这些类别并不直接对应于标准的农艺营养生长和生殖生长阶段。在建模方面，开发了四种轻量级深度学习（DL）模型——ELiteCrop0、ELiteCrop1、ELiteCrop4和MobNetCropV2，并利用超级计算机（NDSU CCAST）在五种图像垂直裁剪水平（0–40%）下进行了评估。随后，优化后的模型被部署到单板Raspberry Pi 5计算机上进行边缘推理。模型训练占总CPU时间的大部分，超过97%，而测试时间为0.01–0.12分钟，支持实时应用。在各模型中，ELiteCrop0实现了最均衡的性能，混淆矩阵对角线比率（CMDR）为0.93，其次是ELiteCrop1（CMDR = 0.92）。总体而言，模型性能随垂直裁剪比例的增加而下降；因此，建议采用适度的图像裁剪（0–10%）以提高计算效率。使用超级计算机的分析得出站内（相同训练站点）准确率为0.90–0.93（Raspberry Pi：0.78–0.81），站间（未见测试站点）准确率为0.48–0.50（Raspberry Pi：0.41–0.43），表明模型泛化面临挑战。Raspberry Pi在68 °C下处理约1000张图像\u002F分钟，在8小时工作日50%容量下，PhenoCam与Raspberry Pi的处理比为1:2500。总体而言，本研究为精准农业中的实时玉米生长阶段监测提供了一种可扩展且经济高效的解决方案，并可扩展至其他作物。","Remote Sensing",81,{"impact":31,"substance":29,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":284},"该研究将轻量深度学习模型部署于树莓派实现玉米生育期实时识别，方法新颖、数据扎实，对精准农业边缘监测有参考价值。",[286],{"name":281,"url":278},[39,40,288,289,290],"边缘计算","玉米","遥感",{"code":292,"slug":293,"name":289,"path":294,"parent_code":208},"A3","maize","\u002Ftopic\u002Ffield-crops\u002Fmaize",[296],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"status":52,"locked":53,"version":54,"method":55,"evidence":298},{"candidates":299,"agricultural_evidence":26},[300,307,313,317,323],{"code":45,"score":31,"supported":26,"title":301,"abstract":303,"tags":306},[302],"deep learning",[304,302,305],"深度学习","decision support",[40],{"code":292,"score":308,"supported":26,"title":309,"abstract":311,"tags":312},15,[310],"corn",[289,310],[289],{"code":65,"score":66,"supported":26,"title":314,"abstract":315,"tags":316},[],[69,70],[39],{"code":318,"score":66,"supported":26,"title":319,"abstract":320,"tags":322},"I3",[],[290,321],"remote sensing",[290],{"code":324,"score":76,"supported":26,"title":325,"abstract":326,"tags":327},"I5",[],[288],[288],[329,330],"Raspberry Pi 玉米 生长阶段","PhenoCam 玉米 深度学习","RaspberryPi玉米生长阶段-3842",5,"10.3390\u002Frs18193337",{"doi":333,"openalex_id":335,"authors":336,"venue":281,"cited_by_count":106,"oa_url":278,"card":361,"direction":365,"ingested_from":114},"W7214787420",[337,340,343,346,349,352,355,358],{"name":338,"orcid":339},"Anne Joice","https:\u002F\u002Forcid.org\u002F0000-0001-7499-9222",{"name":341,"orcid":342},"Humeera Tazeen","https:\u002F\u002Forcid.org\u002F0000-0003-2411-0725",{"name":344,"orcid":345},"Talha Tufaique","https:\u002F\u002Forcid.org\u002F0000-0002-4490-5341",{"name":347,"orcid":348},"Mathala Juliet Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-3886-2429",{"name":350,"orcid":351},"Cannayen Igathinathane","https:\u002F\u002Forcid.org\u002F0000-0001-8884-7959",{"name":353,"orcid":354},"Nitin Rai","https:\u002F\u002Forcid.org\u002F0000-0002-9350-9601",{"name":356,"orcid":357},"Craig W. Whippo","https:\u002F\u002Forcid.org\u002F0000-0002-7626-6919",{"name":359,"orcid":360},"David Walter Archer","https:\u002F\u002Forcid.org\u002F0000-0002-4816-7040",{"tldr":362,"method":363,"finding":364,"direction":365,"opportunity":366},"用PhenoCam图像训练轻量深度学习模型，在树莓派上实时分类玉米8个生长阶段。","基于10个PhenoCam站点图像，开发4种轻量DL模型，在树莓派5上边缘推理。","ELiteCrop0表现最佳（CMDR=0.93），站点内精度0.90-0.93，但站点间泛化差（0","农业遥感与作物表型","站点间泛化差是主要瓶颈，可研究跨站点迁移学习或域适应以提升模型鲁棒性。","2026-09-30T23:30:29.168194Z",{"id":369,"title":370,"url":371,"summary":372,"summary_zh":373,"content":21,"source_name":374,"source_url":371,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":26,"score":282,"score_detail":375,"sources":377,"tags":379,"primary_topic":382,"secondary_topics":386,"classification":387,"search_phrases":414,"slug":417,"view_count":170,"doi":418,"paper":419,"created_at":440},3823,"Space–Air–Ground Integrated Networks for Smart Agriculture and Smart Breeding: A Review of SDN\u002FNFV-Enabled Intelligent Resource Management","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fs26196181","Smart agriculture and smart breeding are evolving from isolated sensing toward geographically distributed, long-term, data-driven closed-loop management. Dispersed farms and breeding sites, UAV and remote-sensing phenotyping, real-time field control, edge inference, and digital-twin synchronization impose heterogeneous demands on coverage, latency, bandwidth, reliability, computing, and energy. Space–air–ground integrated networks (SAGINs) combine satellites, unmanned aerial vehicles\u002Fhigh-altitude platforms, and terrestrial networks for wide-area coverage and elastic access, but introduce dynamic topologies, heterogeneous multi-domain resources, and complex cross-layer orchestration. Focusing on intelligent SAGIN resource management enabled by software-defined networking (SDN) and network function virtualization (NFV), this review examines controller placement, NFV\u002Fservice function chain orchestration, SDN\u002FNFV cooperation, learning-driven optimization, and security mechanisms. Representative studies are compared by objectives, decision variables, mechanisms, applicability boundaries, and engineering costs. These mechanisms are connected to smart agriculture and smart breeding through ubiquitous connectivity, edge computing, task offloading, phenotyping, and digital-twin closed loops, while distinguishing experimentally supported agricultural evidence from SAGIN-oriented architectural inference. Future research should move beyond single-metric optimization toward joint evaluation of state freshness, decision latency, reconfiguration cost, learning and security overhead, and agronomic outcomes to improve deployability, robustness, and verifiability of cross-domain resource management under real deployment conditions across heterogeneous agricultural environments.","智慧农业与智慧育种正从孤立感知向地理分布式、长期、数据驱动的闭环管理演进。分散的农场与养殖场、无人机与遥感表型分析、实时田间控制、边缘推理以及数字孪生同步，对覆盖范围、时延、带宽、可靠性、计算和能量提出了异构需求。天地空一体化网络（SAGIN）融合卫星、无人机\u002F高空平台与地面网络，实现广域覆盖和弹性接入，但引入了动态拓扑、异构多域资源和复杂的跨层编排。本综述聚焦于软件定义网络（SDN）与网络功能虚拟化（NFV）赋能的智能SAGIN资源管理，考察控制器部署、NFV\u002F服务功能链编排、SDN\u002FNFV协同、学习驱动的优化以及安全机制。代表性研究从目标、决策变量、机制、适用边界和工程成本等方面进行比较。这些机制通过泛在连接、边缘计算、任务卸载、表型分析和数字孪生闭环与智慧农业和智慧育种相联系，同时区分有实验支持的农业证据与面向SAGIN的架构推断。未来研究应超越单一指标优化，转向对状态新鲜度、决策时延、重构成本、学习与安全开销以及农艺结果的联合评估，以提升异构农业环境下真实部署条件中跨域资源管理的可部署性、鲁棒性和可验证性。","Sensors",{"impact":31,"substance":29,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":376},"发表于核心期刊的系统性综述，梳理SDN\u002FNFV赋能的空天地一体化网络资源管理及其在智慧农业与智能育种中的应用，学术价值高但属架构性推演，产业落地影响有限。",[378],{"name":374,"url":371},[39,40,288,380,381],"智能育种","空天地一体化",{"code":324,"slug":383,"name":384,"path":385,"parent_code":49},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",[],{"status":52,"locked":53,"version":54,"method":55,"evidence":388},{"candidates":389,"agricultural_evidence":26},[390,396,401,405,410],{"code":65,"score":31,"supported":26,"title":391,"abstract":393,"tags":395},[392],"smart agriculture",[39,394,392],"数字孪生",[39],{"code":324,"score":66,"supported":26,"title":397,"abstract":398,"tags":400},[],[288,399],"edge computing",[288],{"code":318,"score":156,"supported":26,"title":402,"abstract":403,"tags":404},[],[290],[],{"code":406,"score":34,"supported":53,"title":407,"abstract":408,"tags":409},"F2",[],[],[380],{"code":45,"score":34,"supported":53,"title":411,"abstract":412,"tags":413},[],[],[40],[415,416],"SAGIN 智慧农业 资源管理","SDN NFV 智能育种","SAGIN智慧农业资源管理-3823","10.3390\u002Fs26196181",{"doi":418,"openalex_id":420,"authors":421,"venue":374,"cited_by_count":106,"oa_url":371,"card":435,"direction":111,"ingested_from":114},"W7214724892",[422,424,427,430,433],{"name":423,"orcid":21},"Yixiang Zhao",{"name":425,"orcid":426},"Bo Li","https:\u002F\u002Forcid.org\u002F0000-0003-4326-4519",{"name":428,"orcid":429},"Chenglin Xu","https:\u002F\u002Forcid.org\u002F0000-0001-9133-613X",{"name":431,"orcid":432},"Jiao Zhang","https:\u002F\u002Forcid.org\u002F0000-0002-9456-3992",{"name":434,"orcid":21},"L. Wang",{"tldr":436,"method":437,"finding":438,"direction":111,"opportunity":439},"综述SDN\u002FNFV赋能的空天地一体化网络智能资源管理，并关联智慧农业与智慧育种。","文献综述，比较控制器部署、NFV编排、学习优化与安全机制。","现有研究多单指标优化，需联合评估状态新鲜度、时延、重构成本与农艺效果。","可研究面向农业异构环境的跨域资源管理联合优化与真实部署验证，兼顾农艺产出。","2026-09-30T23:30:09.636727Z",{"id":442,"title":443,"url":444,"summary":445,"summary_zh":446,"content":21,"source_name":447,"source_url":444,"published_at":448,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":449,"score_detail":450,"sources":452,"tags":454,"primary_topic":459,"secondary_topics":460,"classification":465,"search_phrases":500,"slug":503,"view_count":34,"doi":504,"paper":505,"created_at":532},3830,"Current breakthroughs in biomass and carbon modelling of rubber plantations: A bibliometric analysis and literature review","https:\u002F\u002Fdoi.org\u002F10.14719\u002Fpst.15821","Rubber plantations (Hevea brasiliensis (Willd. ex A.Juss.) Müll.Arg.) are expanding tropical carbon (C) stores, storing an estimated 160–300 mg C ha-1 of total ecosystem C at maturity, but precise biomass and soil organic carbon (SOC) assessment remains difficult due to age disparities, historical land-use variations and data gaps. This review systematically analyses trends and research gaps in biomass accumulation of rubber systems. PRISMA 2020 guidelines were used to examine 104 studies in Scopus published between 1999 and 2025. The literature revealed a marked increase in research studies in China and Thailand post-2013 using a bibliometric approach presented by Biblioshiny. The most important methods are species-specific allometry (coefficient of determination (R2) > 0.9), age-based growth curves, remote sensing (RS) regressions (R2 0.2–0.7) and machine learning (ML) models like Random Forest\u002FXGBoost (R2 0.8–0.97, root mean square error (RMSE) 7–40 mg\u002Fha), which outperform simpler parametric models. Deep learning (DL) yields the highest accuracies (RMSE ~6 mg ha-1). Gaps persist in standardised protocols, South Indian datasets (e.g., Tamil Nadu\u002FWestern Ghats) and integrated lifecycle carbon accounting, including latex\u002FSOC. Forest-to-rubber conversion incurs net losses, while cropland-to-rubber\u002Fagroforestry offers gains. The authors recommend adopting harmonised Tier 2\u002F3 IPCC models, multi-decadal RS archives for reducing emissions from deforestation and forest degradation (REDD+) and ML extension to Indian smallholders. Policy integration with offset markets requires avoiding deforestation to realise climate-smart benefits.","橡胶林（Hevea brasiliensis (Willd. ex A.Juss.) Müll.Arg.）正在扩大热带碳（C）储量，据估计，成熟期生态系统总碳储量可达160–300 mg C ha⁻¹，但由于林龄差异、历史土地利用变化及数据缺口，精确评估生物量和土壤有机碳（SOC）仍然困难。本综述系统分析了橡胶系统生物量积累的趋势与研究空白。采用PRISMA 2020指南，对Scopus数据库中1999年至2025年间发表的104篇研究进行了审查。通过Biblioshiny文献计量方法分析发现，2013年后中国和泰国的研究数量显著增加。最重要的方法包括物种特异性异速生长方程（决定系数（R²）> 0.9）、基于林龄的生长曲线、遥感（RS）回归（R² 0.2–0.7）以及机器学习（ML）模型如随机森林\u002FXGBoost（R² 0.8–0.97，均方根误差（RMSE）7–40 mg\u002Fha），后者优于简单的参数模型。深度学习（DL）精度最高（RMSE约6 mg ha⁻¹）。在标准化方案、南印度数据集（如泰米尔纳德邦\u002F西高止山脉）以及涵盖乳胶\u002FSOC的综合生命周期碳核算方面仍存在空白。森林转橡胶林会导致净碳损失，而农田转橡胶林\u002F农林复合经营则带来碳增益。作者建议采用统一的IPCC第2\u002F3层级模型、多年代际遥感档案以支持减少毁林和森林退化所致排放（REDD+），并将机器学习应用扩展至印度小农户。与抵消市场的政策整合需要避免毁林，以实现气候智慧型效益。","Plant Science Today","2026-09-28T00:00:00Z",80,{"impact":31,"substance":29,"depth":31,"authority":32,"freshness":127,"relevant":34,"comment":451},"系统综述橡胶林生物量与碳建模研究进展，方法学与数据缺口梳理扎实，对农业碳汇与遥感监测领域有参考价值。",[453],{"name":447,"url":444},[61,455,197,456,457,458],"遥感监测","生物量估算","橡胶林","碳汇计量",{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},[461],{"code":318,"slug":462,"name":463,"path":464,"parent_code":49},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":52,"locked":53,"version":54,"method":55,"evidence":466},{"candidates":467,"agricultural_evidence":26},[468,475,480,484,491,496],{"code":469,"score":126,"supported":26,"title":470,"abstract":472,"tags":474},"I7",[471],"bibliometric",[473,471],"文献计量",[],{"code":45,"score":25,"supported":26,"title":476,"abstract":477,"tags":479},[],[304,61,478,302],"machine learning",[61],{"code":318,"score":33,"supported":26,"title":481,"abstract":482,"tags":483},[],[290,321],[456,455],{"code":485,"score":148,"supported":26,"title":486,"abstract":487,"tags":490},"H2",[],[488,489],"土壤","soil",[],{"code":241,"score":156,"supported":26,"title":492,"abstract":493,"tags":495},[],[494],"小农户",[],{"code":211,"score":34,"supported":53,"title":497,"abstract":498,"tags":499},[],[],[197],[501,502],"橡胶林 生物量 碳储量","橡胶园 遥感 机器学习","橡胶林生物量碳储量-3830","10.14719\u002Fpst.15821",{"doi":504,"openalex_id":506,"authors":507,"venue":447,"cited_by_count":106,"oa_url":526,"card":527,"direction":111,"ingested_from":114},"W7214623646",[508,511,514,517,520,523],{"name":509,"orcid":510},"R M Dhinu","https:\u002F\u002Forcid.org\u002F0009-0008-3471-1093",{"name":512,"orcid":513},"M. Vijayabhama","https:\u002F\u002Forcid.org\u002F0000-0001-9437-9477",{"name":515,"orcid":516},"Balaji Kannan","https:\u002F\u002Forcid.org\u002F0000-0003-2330-0893",{"name":518,"orcid":519},"K Baranidharan","https:\u002F\u002Forcid.org\u002F0000-0001-9371-8588",{"name":521,"orcid":522},"M Kalpana","https:\u002F\u002Forcid.org\u002F0000-0002-0201-6968",{"name":524,"orcid":525},"M Sankavi","https:\u002F\u002Forcid.org\u002F0009-0006-1493-1123","https:\u002F\u002Fhorizonepublishing.com\u002Fjournals\u002Findex.php\u002FPST\u002Farticle\u002Fdownload\u002F15821\u002F16343",{"tldr":528,"method":529,"finding":530,"direction":365,"opportunity":531},"综述橡胶林生物量与碳建模研究进展，指出机器学习与深度学习精度最高但标准化和区域数据仍不足。","PRISMA 2020 文献计量分析 104 篇 Scopus 文献，比较异速生","机器学习\u002F深度学习精度优于传统模型，但标准协议、印度南部数据和全生命周期碳核算仍存空白。","可面向印度小农户构建多年代遥感与机器学习融合的橡胶林碳核算与REDD+监测体系。","2026-09-30T23:30:12.585164Z",{"id":534,"title":535,"url":536,"summary":537,"summary_zh":538,"content":21,"source_name":539,"source_url":536,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":540,"score_detail":541,"sources":544,"tags":546,"primary_topic":551,"secondary_topics":554,"classification":561,"search_phrases":584,"slug":587,"view_count":156,"doi":588,"paper":589,"created_at":606},3845,"Problems of soil degradation assessment using spectral indices from space images","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-71433-1","This article presents an integrated method of spectral indices for assessing soil degradation using Sentinel-2 Earth Remote Sensing (ERS) data. Unlike approaches based on individual indices, the method combines indicators of vegetation, humidity, thermal state, and salinity, increasing the accuracy of mapping saline and degraded lands while minimizing errors caused by variations in natural conditions. The application of the method in the Tashkent region of Uzbekistan (41.1465°N, 69.3009°E) showed that the category of “extremely high” and “high” degradation accounts for 3.29% and 8.28% of pixels, “medium” and “low” for 9.4% and 9.52%, while the remaining 69.51% belong to other category (class) of land cover, confirming the concentration of degradation foci in lowland areas. Comparison with 2020 showed a notable increase in the share of critical degradation (growth from 10.46% to 11.57%), while the share of the “extremely high” category increased from 2.80% to 3.29%, indicating a shift toward more severe conditions that aligns with official regional salinization statistics. Analysis of meteorological data (2013–2025) revealed an inverse relationship between temperature and precipitation, confirming the climatic nature of salinization; four stages of salinization were identified—latent, progressive, severe, and complete degradation and the main components method (PCA) was applied to reduce the size of the data. These results correspond to the expansion of irrigated lands in Uzbekistan from 3.5 to 8.5 million hectares and the associated increase in groundwater levels. Consequently, this method is not the final means of delimiting degraded lands, but represents a quick, inexpensive preliminary check before field surveys: it marks potentially degraded areas for priority ground surveys, supplementing official surveys and government statistics rather than replacing them. The method is applicable in agriculture, land cadastre, and environmental monitoring as an auxiliary tool for climate adaptation and water resource management.","本文提出了一种利用Sentinel-2地球遥感（ERS）数据评估土壤退化的光谱指数集成方法。与基于单一指数的方法不同，该方法综合了植被、湿度、热状态和盐分指标，提高了盐碱化和退化土地制图的精度，同时最大限度地减少了自然条件变化引起的误差。该方法在乌兹别克斯坦塔什干地区（北纬41.1465°，东经69.3009°）的应用表明，“极高”和“高”退化类别分别占像元的3.29%和8.28%，“中等”和“低”分别占9.4%和9.52%，其余69.51%属于其他土地覆盖类别（等级），证实了退化热点集中在低地地区。与2020年相比，临界退化比例显著增加（从10.46%增至11.57%），而“极高”类别比例从2.80%增至3.29%，表明退化状况趋于严重，这与官方区域盐渍化统计数据一致。气象数据（2013—2025年）分析揭示了气温与降水之间的反比关系，证实了盐渍化的气候成因；识别出盐渍化的四个阶段——潜伏期、进展期、严重期和完全退化期，并应用主成分分析方法（PCA）对数据进行了降维处理。这些结果与乌兹别克斯坦灌溉面积从350万公顷扩展至850万公顷以及相关的地下水位上升相吻合。因此，该方法并非划定退化土地的最终手段，而是实地调查前一种快速、低成本的初步筛查：它标记出潜在退化区域以供优先开展地面调查，是对官方调查和政府统计的补充而非替代。该方法适用于农业、土地地籍和环境监测领域，可作为气候适应和水资源管理的辅助工具。","Scientific Reports",79,{"impact":542,"substance":29,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":543},16,"基于Sentinel-2多光谱指数的土壤退化综合评估方法，方法新颖、数据翔实，对盐渍化遥感监测与耕地保护有参考价值。",[545],{"name":539,"url":536},[547,455,548,549,550],"Sentinel-2","土壤退化","盐渍化","土地退化评估",{"code":485,"slug":489,"name":552,"path":553,"parent_code":142},"土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil",[555,560],{"code":556,"slug":557,"name":558,"path":559,"parent_code":142},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation",{"code":318,"slug":462,"name":463,"path":464,"parent_code":49},{"status":52,"locked":53,"version":54,"method":55,"evidence":562},{"candidates":563,"agricultural_evidence":26},[564,568,575,579],{"code":485,"score":308,"supported":26,"title":565,"abstract":566,"tags":567},[489],[488,489],[549],{"code":556,"score":127,"supported":26,"title":569,"abstract":570,"tags":574},[],[571,572,573],"灌溉","水资源","water resource",[],{"code":318,"score":33,"supported":26,"title":576,"abstract":577,"tags":578},[],[290,321],[547,455],{"code":138,"score":156,"supported":26,"title":580,"abstract":581,"tags":583},[],[582],"气候适应",[],[585,586],"乌兹别克斯坦 塔什干 土壤盐渍化","Sentinel-2 光谱指数 土壤退化","乌兹别克斯坦塔什干土壤盐渍化-3845","10.1038\u002Fs41598-026-71433-1",{"doi":588,"openalex_id":590,"authors":591,"venue":539,"cited_by_count":106,"oa_url":600,"card":601,"direction":365,"ingested_from":114},"W7214801255",[592,595,598],{"name":593,"orcid":594},"Z. Z. Shamsiev","https:\u002F\u002Forcid.org\u002F0000-0002-0323-9741",{"name":596,"orcid":597},"L. F. Sulyukova","https:\u002F\u002Forcid.org\u002F0000-0002-7150-4359",{"name":599,"orcid":21},"Rasul Shamsiev","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-71433-1_reference.pdf",{"tldr":602,"method":603,"finding":604,"direction":365,"opportunity":605},"提出融合多光谱指数的Sentinel-2土壤退化评估方法，在乌兹别克斯坦验证并识别盐渍化阶段。","Sentinel-2影像多指数融合（植被、湿度、热、盐分）结合PCA与气象数据分","退化集中于低地，2020年以来极重度退化占比由2.80%升至3.29%，与官方盐渍化统计一致。","可将该快速预筛方法与地面调查、机器学习结合，构建多区域可迁移的退化早期预警与制图框架。","2026-09-30T23:30:29.386707Z",{"id":608,"title":609,"url":610,"summary":611,"summary_zh":612,"content":21,"source_name":613,"source_url":610,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":540,"score_detail":614,"sources":616,"tags":618,"primary_topic":622,"secondary_topics":623,"classification":625,"search_phrases":655,"slug":658,"view_count":156,"doi":659,"paper":660,"created_at":690},3843,"A multi-scale remote sensing approach for assessing drought impacts on rainfed and irrigated cereal yields in Lleida, Spain","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.agwat.2026.110826","This study evaluates the performance of agricultural drought indices at provincial and field scales, as well as their ability to explain the variability of wheat and barley yields under rainfed and irrigated conditions. The analysis was conducted in the Lleida region (northeastern Spain) from 2010 to 2024 using high-resolution (30 m) multi-source remote sensing data and yield information from 5030 cereal fields. Five drought indices representing vegetative stress, thermal stress, soil moisture conditions, and their combination were evaluated: the Vegetative Condition Index (VCI), the Thermal Condition Index (TCI), the Soil Moisture Condition Index (SMCI), the Vegetative Health Index (VHI), and the Combined Drought Anomaly Index (CDAI). The analysis of drought index–yield relationships reveal a strong dependence on crop type, phenological stage, spatial scale, and water management practices. Barley exhibits greater sensitivity to drought than wheat and responds earlier in the growing season. Correlations are generally weak during early growth stages but become much stronger during the reproductive and maturation phases. Using high spatial resolution data (30 m) significantly enhances the performance of drought indices. Under rainfed conditions, most indices perform well except for the TCI, with the CDAI standing out as the most effective, achieving correlation values of r = 0.85 (p \u003C 0.01) at the provincial scale and a median of r = 0.52, with a maximum of r = 0.95 (p \u003C 0.01) at the field scale. In contrast, drought index–yield correlations in irrigated areas are consistently lower, highlighting the buffering role of irrigation in mitigating drought impacts. In irrigated fields across the study area, yield variability is primarily driven by heat stress rather than water or vegetation stress. Overall, composite drought indices such as the CDAI are particularly well suited for monitoring agricultural drought in rainfed conditions, whereas indices specifically designed for irrigated agriculture remain essential.","本研究在省域和田块尺度上评估了农业干旱指数的表现，以及其在雨养和灌溉条件下解释小麦和大麦产量变异的能力。分析于2010—2024年在西班牙东北部莱里达地区开展，使用了高分辨率（30 m）多源遥感数据和5030块谷类田块的产量信息。研究评估了五个代表植被胁迫、热胁迫、土壤水分状况及其组合的干旱指数：植被状态指数（VCI）、热状态指数（TCI）、土壤水分状态指数（SMCI）、植被健康指数（VHI）和综合干旱异常指数（CDAI）。对干旱指数与产量关系的分析表明，其强烈依赖于作物类型、物候阶段、空间尺度及水分管理方式。大麦对干旱的敏感性高于小麦，且在生长季中响应更早。在生长早期阶段相关性普遍较弱，但在生殖和成熟阶段显著增强。使用高空间分辨率数据（30 m）可显著提升干旱指数的表现。在雨养条件下，除TCI外，大多数指数表现良好，其中CDAI最为有效，在省域尺度上相关系数达到r = 0.85（p \u003C 0.01），在田块尺度上中位数为r = 0.52，最大值为r = 0.95（p \u003C 0.01）。相比之下，灌溉区域的干旱指数与产量相关性 consistently 较低，凸显了灌溉在缓解干旱影响中的缓冲作用。在研究区的灌溉田块中，产量变异主要由热胁迫驱动，而非水分或植被胁迫。总体而言，诸如CDAI之类的复合干旱指数特别适用于监测雨养条件下的农业干旱，而专门为灌溉农业设计的指数仍然不可或缺。","Agricultural Water Management",{"impact":542,"substance":29,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":615},"基于30米高分辨率遥感与5030块田块数据，量化五种干旱指数对雨养与灌溉谷物产量的解释力，方法扎实、结论对农业干旱监测与灌溉管理有参考价值。",[617],{"name":613,"url":610},[39,619,620,455,621],"作物产量","精准灌溉","农业干旱",{"code":318,"slug":462,"name":463,"path":464,"parent_code":49},[624],{"code":138,"slug":139,"name":140,"path":141,"parent_code":142},{"status":52,"locked":53,"version":54,"method":55,"evidence":626},{"candidates":627,"agricultural_evidence":26},[628,632,636,640,647,651],{"code":318,"score":542,"supported":26,"title":629,"abstract":630,"tags":631},[321],[290,321],[619,455],{"code":138,"score":126,"supported":26,"title":633,"abstract":634,"tags":635},[152],[151,152],[],{"code":556,"score":66,"supported":26,"title":637,"abstract":638,"tags":639},[],[571,557],[620],{"code":641,"score":148,"supported":26,"title":642,"abstract":643,"tags":646},"A2",[],[644,645],"小麦","wheat",[],{"code":485,"score":148,"supported":26,"title":648,"abstract":649,"tags":650},[],[488,489],[],{"code":65,"score":34,"supported":53,"title":652,"abstract":653,"tags":654},[],[],[39],[656,657],"Lleida 干旱 遥感 谷物产量","CDAI 干旱指数 灌溉","Lleida干旱遥感谷物产量-3843","10.1016\u002Fj.agwat.2026.110826",{"doi":659,"openalex_id":661,"authors":662,"venue":613,"cited_by_count":106,"oa_url":610,"card":685,"direction":365,"ingested_from":114},"W7214788171",[663,666,669,672,674,677,680,682],{"name":664,"orcid":665},"Youness Ablila","https:\u002F\u002Forcid.org\u002F0009-0006-3081-5990",{"name":667,"orcid":668},"Maria‐José Escorihuela","https:\u002F\u002Forcid.org\u002F0000-0002-7780-7334",{"name":670,"orcid":671},"Abdelhakim Amazirh","https:\u002F\u002Forcid.org\u002F0000-0002-6665-3843",{"name":673,"orcid":21},"Jose Antonio Martinez Casasnovas",{"name":675,"orcid":676},"El Houssaine Bouras","https:\u002F\u002Forcid.org\u002F0000-0002-6973-6644",{"name":678,"orcid":679},"Saïd Khabba","https:\u002F\u002Forcid.org\u002F0000-0003-3309-9935",{"name":681,"orcid":21},"Zaineb Bouswir",{"name":683,"orcid":684},"Salah Er‐Raki","https:\u002F\u002Forcid.org\u002F0000-0002-8595-7949",{"tldr":686,"method":687,"finding":688,"direction":365,"opportunity":689},"评估多尺度遥感干旱指数对西班牙Lleida地区雨养与灌溉谷物产量的解释能力。","2010-2024年30米多源遥感数据与5030块田产量，比较VCI、TCI、S","CDAI在雨养区表现最佳（省域r=0.85），灌溉区相关性低且产量主要受热胁迫驱动。","可开发针对灌溉农业的专用干旱指数，并融合热胁迫与灌溉管理数据提升产量预测。","2026-09-30T23:30:29.241563Z",{"id":692,"title":693,"url":694,"summary":695,"summary_zh":696,"content":21,"source_name":697,"source_url":694,"published_at":448,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":540,"score_detail":698,"sources":700,"tags":702,"primary_topic":705,"secondary_topics":710,"classification":712,"search_phrases":738,"slug":741,"view_count":332,"doi":742,"paper":743,"created_at":761},3829,"Towards accessible smallholder crop classification in the groundnut basin of Senegal","https:\u002F\u002Fdoi.org\u002F10.1088\u002F2976-601x\u002Faea0b8","Abstract Crop type mapping is a critical lever for decision making in smallholder food systems, yet high cost and expertise requirements of developing remote sensing-driven maps as well as need for recurrent data are persistent barriers to incorporating geospatial tools into decision making. To address this gap, we establish a three-part framework to evaluate crop type mapping: (1) performance, (2) temporal transferability, and (3) accessibility. We evaluate geospatial foundation model embeddings (TESSERA and AlphaEarth) against traditional baselines in Senegal’s groundnut basin. We find that the embedding-based approaches, in particular TESSERA, best fulfill these criteria. The TESSERA-based method shows a 78% reduction in both compute compared to a timeseries approach, a reduction in need for custom feature generation, and 28% higher relative accuracy (an absolute improvement of approximately 0.10 weighted F 1) in one crop type classification temporal transfer test. By reducing the reliance on training data, computational resource use, and expertise for data processing, this approach shows potential to contribute to improved smallholder crop type classification.","摘要 作物类型制图是小农粮食系统决策的关键手段，但开发遥感驱动地图的高成本与专业知识要求，以及 recurring 数据的需求，仍是将地理空间工具纳入决策的持续障碍。为弥补这一差距，我们建立了一个三部分框架来评估作物类型制图：（1）性能，（2）时间可迁移性，（3）可及性。我们在塞内加尔花生盆地评估了地理空间基础模型嵌入（TESSERA 和 AlphaEarth）与传统基线方法。我们发现，基于嵌入的方法，尤其是 TESSERA，最能满足这些标准。基于 TESSERA 的方法显示，与时间序列方法相比，计算量减少 78%，对自定义特征生成的需求降低，并且在一项作物类型分类时间迁移测试中相对精度提高 28%（加权 F1 绝对提升约 0.10）。通过减少对训练数据、计算资源使用和数据处理专业知识的依赖，该方法显示出改善小农作物类型分类的潜力。","Environmental Research Food Systems",{"impact":31,"substance":29,"depth":31,"authority":32,"freshness":33,"relevant":34,"comment":699},"该研究提出兼顾性能、时序迁移性与可及性的作物分类评估框架，用地理空间基础模型显著降低算力与训练数据依赖，对小农户农业遥感应用具有实质参考价值。",[701],{"name":697,"url":694},[39,40,703,494,704],"农业遥感","作物分类",{"code":241,"slug":706,"name":707,"path":708,"parent_code":709},"farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[711],{"code":318,"slug":462,"name":463,"path":464,"parent_code":49},{"status":52,"locked":53,"version":54,"method":55,"evidence":713},{"candidates":714,"agricultural_evidence":26},[715,720,724,730,734],{"code":241,"score":716,"supported":26,"title":717,"abstract":718,"tags":719},12,[244],[244],[494],{"code":318,"score":33,"supported":26,"title":721,"abstract":722,"tags":723},[],[290,321],[704,703],{"code":725,"score":156,"supported":26,"title":726,"abstract":727,"tags":729},"A4",[],[728],"花生",[],{"code":65,"score":34,"supported":53,"title":731,"abstract":732,"tags":733},[],[],[39],{"code":45,"score":34,"supported":53,"title":735,"abstract":736,"tags":737},[],[],[40],[739,740],"塞内加尔 花生盆地 作物分类","TESSERA 遥感 作物制图","塞内加尔花生盆地作物分类-3829","10.1088\u002F2976-601x\u002Faea0b8",{"doi":742,"openalex_id":744,"authors":745,"venue":697,"cited_by_count":106,"oa_url":755,"card":756,"direction":111,"ingested_from":114},"W7214087594",[746,748,750,752],{"name":747,"orcid":21},"Madeline Lisaius",{"name":749,"orcid":21},"Srinivasan Keshav",{"name":751,"orcid":21},"Andrew Blake",{"name":753,"orcid":754},"Clement Atzberger","https:\u002F\u002Forcid.org\u002F0000-0003-2169-8009","https:\u002F\u002Fiopscience.iop.org\u002Farticle\u002F10.1088\u002F2976-601X\u002Faea0b8\u002Fpdf",{"tldr":757,"method":758,"finding":759,"direction":365,"opportunity":760},"评估地理空间基础模型嵌入在塞内加尔花生盆地小农作物分类中的性能、时序迁移性和可及性。","对比TESSERA和AlphaEarth嵌入与传统基线，用加权F1和计算量评估。","TESSERA嵌入精度相对提高28%，计算量减少78%，且无需自定义特征生成。","可探索基础模型嵌入在更多小农区域和作物类型上的泛化能力及低成本部署方案。","2026-09-30T23:30:11.490990Z",{"id":763,"title":764,"url":765,"summary":766,"summary_zh":767,"content":21,"source_name":122,"source_url":765,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":540,"score_detail":768,"sources":770,"tags":772,"primary_topic":776,"secondary_topics":777,"classification":778,"search_phrases":797,"slug":800,"view_count":76,"doi":801,"paper":802,"created_at":816},3817,"Exploring the boundary conditions of AI-driven environmental value creation: evidence from China’s livestock sector","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1934939","The transition toward sustainable food systems demands effective interventions that can reduce the environmental footprint of livestock production without compromising its economic functions. Artificial intelligence (AI) offers a promising technological lever, yet evidence on its environmental effectiveness and the conditions under which its benefits materialize remains scarce. This study investigates whether and when AI adoption improves corporate environmental management performance, drawing on a panel data-set of Chinese A-share listed livestock firms from 2011 to 2024. We identify two critical boundary conditions-financing constraints and the separation of ownership and control rights-that shape the environmental outcomes of AI deployment. Our findings demonstrate that AI adoption significantly enhances environmental management performance, confirming its viability as a sustainability intervention in the livestock sector. However, this positive effect is substantially weakened by financing constraints, indicating that capital shortages impede firms’ ability to realize AI-driven environmental gains. Conversely, a higher degree of separation between ownership and control rights amplifies the positive impact of AI, suggesting that governance structures insulating managerial decisions from short-term ownership pressures facilitate long-term investments in digital and green technologies. These results highlight that the environmental value of AI is not automatic but conditional upon adequate financial resources and appropriate governance arrangements. Our study contributes empirical evidence on the economics of sustainable food system transitions, offering actionable policy implications for promoting digital-green synergies in resource-intensive agricultural sectors.","向可持续食物系统的转型，需要能够在不损害畜牧业经济功能的前提下降低其环境足迹的有效干预措施。人工智能（AI）为此提供了一种颇具前景的技术杠杆，但关于其环境有效性以及其效益在何种条件下才能实现的证据仍然稀缺。本研究基于2011年至2024年中国A股上市畜牧企业的面板数据集，考察AI应用是否以及何时能够改善企业环境管理绩效。我们识别出两个塑造AI部署环境成效的关键边界条件——融资约束与所有权和控制权的分离。研究发现，AI应用显著提升了环境管理绩效，证实了其作为畜牧业可持续性干预措施的可行性。然而，这一正向效应受到融资约束的显著削弱，表明资本短缺阻碍了企业实现AI驱动的环境收益。相反，所有权与控制权分离程度的提高则放大了AI的正向影响，表明使管理决策免受短期所有权压力的治理结构有助于对数字与绿色技术进行长期投资。这些结果表明，AI的环境价值并非自动实现，而是取决于充足的财务资源和适当的治理安排。本研究为可持续食物系统转型的经济学提供了经验证据，并为在资源密集型农业部门促进数字—绿色协同提供了可操作的政策启示。",{"impact":31,"substance":29,"depth":31,"authority":32,"freshness":33,"relevant":34,"comment":769},"基于2011—2024年中国A股畜牧企业面板数据，揭示融资约束与两权分离对AI环境价值实现的边界条件，方法规范、结论有政策含义，属细分领域实证进展。",[771],{"name":122,"url":765},[39,40,773,774,775],"畜牧业","农业上市公司","数字绿色协同",{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},[],{"status":52,"locked":53,"version":54,"method":55,"evidence":779},{"candidates":780,"agricultural_evidence":26},[781,788,793],{"code":782,"score":308,"supported":26,"title":783,"abstract":785,"tags":787},"C",[784],"livestock",[786,784],"畜牧",[773],{"code":45,"score":76,"supported":26,"title":789,"abstract":790,"tags":792},[],[791],"artificial intelligence",[40],{"code":65,"score":34,"supported":53,"title":794,"abstract":795,"tags":796},[],[],[39],[798,799],"中国畜牧业 AI 环境绩效","A股畜牧企业 人工智能","中国畜牧业AI环境绩效-3817","10.3389\u002Ffsufs.2026.1934939",{"doi":801,"openalex_id":803,"authors":804,"venue":122,"cited_by_count":106,"oa_url":765,"card":811,"direction":182,"ingested_from":114},"W7214777826",[805,808],{"name":806,"orcid":807},"Wang Xiaojie","https:\u002F\u002Forcid.org\u002F0009-0002-0470-1878",{"name":809,"orcid":810},"Yimin Yang","https:\u002F\u002Forcid.org\u002F0000-0002-5714-0886",{"tldr":812,"method":813,"finding":814,"direction":182,"opportunity":815},"基于中国畜牧业上市公司数据，检验AI应用能否提升环境管理绩效及其边界条件。","2011-2024年中国A股畜牧业上市公司面板数据，实证回归分析。","AI显著改善环境绩效，但融资约束削弱该效应，两权分离则放大其正向作用。","可探究不同畜种、规模及区域下AI绿色效应的异质性，并识别其他治理与政策调节变量。","2026-09-30T23:30:07.380143Z",{"id":818,"title":819,"url":820,"summary":821,"summary_zh":21,"content":21,"source_name":822,"source_url":820,"published_at":823,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":540,"score_detail":824,"sources":827,"tags":829,"primary_topic":833,"secondary_topics":839,"classification":840,"search_phrases":856,"slug":859,"view_count":156,"doi":860,"paper":861,"created_at":877},3808,"Orchard worker detection using spatial-context attention on 2D LiDAR data","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.biosystemseng.2026.104604","Orchard worker detection using spatial-context attention on 2D LiDAR data。Biosystems Engineering","Biosystems Engineering","2026-09-30T00:00:00Z",{"impact":542,"substance":825,"depth":31,"authority":126,"freshness":25,"relevant":34,"comment":826},21,"提出基于二维激光雷达与空间上下文注意力的果园工人检测方法，属农业人工智能与智能农机细分领域的方法创新，具备一定专业参考价值。",[828],{"name":822,"url":820},[39,40,830,831,832],"目标检测","果园管理","激光雷达",{"code":834,"slug":835,"name":836,"path":837,"parent_code":838},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[],{"status":52,"locked":53,"version":54,"method":55,"evidence":841},{"candidates":842,"agricultural_evidence":26},[843,848,852],{"code":834,"score":224,"supported":26,"title":844,"abstract":846,"tags":847},[845],"orchard",[845],[],{"code":45,"score":170,"supported":53,"title":849,"abstract":850,"tags":851},[],[],[40,830],{"code":65,"score":34,"supported":53,"title":853,"abstract":854,"tags":855},[],[],[39],[857,858],"2D LiDAR 果园 工人检测","农业人工智能 智慧农业 果园管理 激光雷达","2DLiDAR果园工人检测-3808","10.1016\u002Fj.biosystemseng.2026.104604",{"doi":860,"openalex_id":862,"authors":863,"venue":822,"cited_by_count":106,"oa_url":820,"card":21,"direction":21,"ingested_from":114},"W7214801786",[864,866,868,871,874],{"name":865,"orcid":21},"Jintack Jeon",{"name":867,"orcid":21},"Huijong Chung",{"name":869,"orcid":870},"Gookhwan Kim","https:\u002F\u002Forcid.org\u002F0000-0002-7278-3476",{"name":872,"orcid":873},"Yong–Joo Kim","https:\u002F\u002Forcid.org\u002F0000-0002-1212-9018",{"name":875,"orcid":876},"Dae-Hyun Lee","https:\u002F\u002Forcid.org\u002F0000-0001-9544-5974","2026-09-30T23:30:05.310491Z",{"id":879,"title":880,"url":881,"summary":882,"summary_zh":883,"content":21,"source_name":884,"source_url":881,"published_at":23,"category":24,"cover_url":21,"hotness":885,"is_selected":53,"score":886,"score_detail":887,"sources":889,"tags":893,"primary_topic":896,"secondary_topics":897,"classification":898,"search_phrases":921,"slug":924,"view_count":148,"doi":925,"paper":926,"created_at":936},3822,"Climate Value Bifurcation (CVB) Model: Methodology, Classification Protocol, and Quantification Framework","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23031418","The Climate Value Bifurcation (CVB) Model is an open methodology for valuing the climate outcomes of agroecological and climate-smart agriculture programmes led by smallholder farmers. Such programmes produce mitigation, adaptation and resilience outcomes at the same time, and reporting them together invites \"triple-win\" claims, inconsistent classification and double counting. The CVB Model separates these outcomes into three independent tracks: Mitigation (carbon value), Adaptation (farm margin expansion) and Resilience (probability-weighted income protected against defined climate shocks). A five-step protocol assigns each intervention a primary track based on its climate mechanism, not its financial magnitude, anchored where relevant to IPCC and Verra pathways. The document sets out valuation formulas for each track, a three-tier data hierarchy with status codes, a structured literature review method, emission factor and carbon price conventions, a cash-flow separability test that governs when values may be aggregated, confidence ratings, sensitivity ranges and reporting conventions. An Engagement Localisation Annex adapts the method to any Indian landscape, with a worked example from Patratu Block, Jharkhand. Outputs are pre-feasibility estimates, not verified carbon credits. Field validation sharpens them at diligence.","气候价值分叉（Climate Value Bifurcation, CVB）模型是一种开放方法论，用于评估小农户主导的农业生态项目和气候智慧型农业项目的气候成果。此类项目同时产生减缓、适应和韧性成果，将其合并报告容易引发“三赢”主张、分类不一致和重复计算。CVB模型将这些成果分为三个独立轨道：减缓（碳价值）、适应（农场利润扩张）和韧性（针对既定气候冲击的概率加权保护收入）。五步协议根据每项干预措施的气候机制而非其财务规模，将其归入主要轨道，并在相关处锚定IPCC和Verra路径。文件阐述了各轨道的估值公式、带状态码的三级数据层级、结构化文献综述方法、排放因子与碳价格惯例、控制价值何时可聚合的现金流可分离性检验、置信度评级、敏感性区间和报告惯例。参与本地化附录将该方法适配于印度任意景观，并附有贾坎德邦帕特拉图区块的实例。输出结果为预可行性估算，而非经核证的碳信用。实地验证可在尽职调查阶段对其加以完善。","Zenodo (CERN European Organization for Nuclear Research)",25,78,{"impact":542,"substance":29,"depth":31,"authority":32,"freshness":127,"relevant":34,"comment":888},"提出气候价值三分轨核算方法学并附印度案例，对农业碳汇与气候智慧型农业评估有参考价值，但属预可行性方法框架，非政策或产业级突破。",[890,891],{"name":884,"url":881},{"name":884,"url":892},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23031417",[494,894,232,197,895],"气候智慧型农业","方法学",{"code":241,"slug":706,"name":707,"path":708,"parent_code":709},[],{"status":52,"locked":53,"version":54,"method":55,"evidence":899},{"candidates":900,"agricultural_evidence":26},[901,905,909,913,917],{"code":241,"score":66,"supported":26,"title":902,"abstract":903,"tags":904},[],[494,244],[494],{"code":216,"score":76,"supported":26,"title":906,"abstract":907,"tags":908},[],[232],[232],{"code":65,"score":156,"supported":26,"title":910,"abstract":911,"tags":912},[],[392],[],{"code":211,"score":34,"supported":53,"title":914,"abstract":915,"tags":916},[],[],[197],{"code":138,"score":34,"supported":53,"title":918,"abstract":919,"tags":920},[],[],[894],[922,923],"气候智慧型农业 农业生态 农业碳汇 小农户","气候智慧型农业 农业生态","气候智慧型农业农业生态农业碳汇小农户-3822","10.5281\u002Fzenodo.23031418",{"doi":925,"openalex_id":927,"authors":928,"venue":884,"cited_by_count":106,"oa_url":881,"card":931,"direction":111,"ingested_from":114},"W7214714568",[929],{"name":930,"orcid":21},"Srajesh Gupta",{"tldr":932,"method":933,"finding":934,"direction":182,"opportunity":935},"提出气候价值分叉模型，将小农农业项目的气候成果分为减缓、适应、韧性三条独立轨道进行估值。","五步分类协议、三轨估值公式、三层数据层级与现金流可分性检验，以印度贾坎德邦为例。","按气候机制而非财务规模分配主轨道，可避免三重赢家声明、分类不一致和重复计算。","可探索该模型在不同区域和作物系统的本地化验证，以及将其与数字平台结合实现自动化估值与监测。","2026-09-30T23:30:09.550801Z",{"id":938,"title":939,"url":940,"summary":941,"summary_zh":942,"content":21,"source_name":122,"source_url":940,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":886,"score_detail":943,"sources":945,"tags":947,"primary_topic":953,"secondary_topics":958,"classification":965,"search_phrases":999,"slug":1002,"view_count":34,"doi":1003,"paper":1004,"created_at":1027},3814,"Optimizing biogas slurry and biochar co-application to enhance soil microenvironment, mitigate greenhouse gas emissions, and increase tomato yield in greenhouse systems","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1866425","To support agricultural carbon peaking and carbon neutrality goals, this study investigated the uncertainties surrounding greenhouse gas emissions and nutrient leaching associated with biogas slurry utilization. It systematically explored the mechanisms and mitigation pathways of co-applying biogas slurry, biochar, and chemical fertilizers for reducing greenhouse gas emissions and nutrient losses under intensive agricultural management characterized by excessive irrigation and fertilizer application. Under equivalent N, P, and K inputs and consistent irrigation levels, different substitution ratios of biogas slurry for chemical fertilizers were evaluated in a point-source underground irrigation system. Treatments included CF, FR, BS25, BS50, BS75, and BS100, with corresponding biochar treatments CF + C, FR + C, BS25 + C, BS50 + C, BS75 + C, and BS100 + C. Tomato yield, soil physical properties, microbial biomass, microbial gene copy numbers, and soil greenhouse gas emissions were measured. Key influencing factors and optimal management strategies were identified using multivariate statistical analysis and a multi-criteria decision approach. The results indicated that biogas slurry combined with biochar significantly improved the soil microenvironment in greenhouse tomato systems by enhancing soil temperature (ST), soil moisture content (SMC), and water-filled pore space (WFPS), especially during the flowering stage in both spring-summer and autumn-winter periods. Biochar also significantly increased microbial biomass carbon (MBC), while microbial gene copy numbers, including AOA, AOB, and mcrA, showed clear seasonal variation. Regarding greenhouse gas emissions, co-application of biogas slurry and biochar significantly affected CO 2 fluxes, particularly from days 8–119, promoting CO 2 emissions. Biochar effectively suppressed N 2 O emissions, promoted CH 4 emissions during days 1–7, and reduced CH 4 fluxes from days 8–119. The BS75 + C treatment significantly increased tomato yield to 151.34 t ha −1 , reduced global warming potential to 9491.43 kg ha −1 , and lowered greenhouse gas emission intensity. Compared with other treatments, BS75 + C consistently showed superior agronomic and environmental performance. Comprehensive analysis showed that ST, SMC, and MBC mainly influenced CO 2 emissions; ST, AOA, and AOB influenced N 2 O emissions; and mcrA, WFPS, and ST influenced CH 4 emissions. Multi-criteria decision analysis identified BS75 + C as the optimal strategy for maximizing yield while minimizing greenhouse gas emissions and improving biogas slurry utilization efficiency in sustainable greenhouse agriculture.","为支撑农业碳达峰与碳中和目标，本研究探讨了沼液利用过程中温室气体排放与养分淋失的不确定性，系统揭示了在过量灌溉与施肥的集约化农业管理条件下，沼液、生物炭与化肥配施减少温室气体排放和养分损失的机理与调控路径。在等氮、磷、钾投入及一致灌溉水平下，采用点源地下灌溉系统，评估了不同沼液替代化肥比例的效果。处理包括CF、FR、BS25、BS50、BS75和BS100，以及相应的生物炭处理CF + C、FR + C、BS25 + C、BS50 + C、BS75 + C和BS100 + C。测定了番茄产量、土壤物理性质、微生物生物量、微生物基因拷贝数及土壤温室气体排放。通过多元统计分析和多准则决策方法，确定了关键影响因素和最优管理策略。结果表明，沼液配施生物炭通过提高土壤温度(ST)、土壤含水量(SMC)和充水孔隙度(WFPS)，显著改善了温室番茄系统的土壤微环境，尤其在春夏季和秋冬季的开花期。生物炭还显著提高了微生物生物量碳(MBC)，而微生物基因拷贝数(包括AOA、AOB和mcrA)则表现出明显的季节性变化。在温室气体排放方面，沼液与生物炭配施显著影响CO₂通量，尤其在8—119天期间促进CO₂排放。生物炭有效抑制了N₂O排放，在1—7天促进CH₄排放，并在8—119天降低CH₄通量。BS75 + C处理显著提高了番茄产量至151.34 t ha⁻¹，将全球增温潜势降至9491.43 kg ha⁻¹，并降低了温室气体排放强度。与其他处理相比，BS75 + C在农艺和环境性能方面始终表现优越。综合分析表明，ST、SMC和MBC主要影响CO₂排放；ST、AOA和AOB影响N₂O排放；mcrA、WFPS和ST影响CH₄排放。多准则决策分析确定BS75 + C为最优策",{"impact":542,"substance":29,"depth":31,"authority":32,"freshness":127,"relevant":34,"comment":944},"该研究通过沼液与生物炭配施实现温室番茄增产与温室气体减排，方法系统、数据扎实，对农业绿色低碳发展具有参考价值。",[946],{"name":122,"url":940},[948,949,950,951,952],"生物炭","温室番茄","农业碳中和","沼液利用","减排增效",{"code":954,"slug":955,"name":956,"path":957,"parent_code":838},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture",[959,964],{"code":960,"slug":961,"name":962,"path":963,"parent_code":838},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",{"code":485,"slug":489,"name":552,"path":553,"parent_code":142},{"status":52,"locked":53,"version":54,"method":55,"evidence":966},{"candidates":967,"agricultural_evidence":26},[968,974,978,984,990,994],{"code":954,"score":308,"supported":26,"title":969,"abstract":971,"tags":973},[970],"greenhouse",[972,970],"温室",[949],{"code":485,"score":308,"supported":26,"title":975,"abstract":976,"tags":977},[489],[488,489],[948],{"code":960,"score":126,"supported":26,"title":979,"abstract":981,"tags":983},[980],"tomato",[982,980],"番茄",[],{"code":155,"score":148,"supported":26,"title":985,"abstract":986,"tags":989},[],[987,988],"施肥","fertilizer",[],{"code":556,"score":148,"supported":26,"title":991,"abstract":992,"tags":993},[],[571,557],[],{"code":211,"score":156,"supported":26,"title":995,"abstract":996,"tags":998},[],[997],"农业碳",[],[1000,1001],"沼液 生物炭 温室番茄","BS75 C 温室气体减排","沼液生物炭温室番茄-3814","10.3389\u002Ffsufs.2026.1866425",{"doi":1003,"openalex_id":1005,"authors":1006,"venue":122,"cited_by_count":106,"oa_url":940,"card":1022,"direction":182,"ingested_from":114},"W7214739741",[1007,1010,1013,1015,1018,1020],{"name":1008,"orcid":1009},"Qinglin Sa","https:\u002F\u002Forcid.org\u002F0009-0007-5248-566X",{"name":1011,"orcid":1012},"Jian Zheng","https:\u002F\u002Forcid.org\u002F0009-0003-5723-1950",{"name":1014,"orcid":21},"Ke Zhang",{"name":1016,"orcid":1017},"Yan Wang","https:\u002F\u002Forcid.org\u002F0000-0002-5950-8904",{"name":1019,"orcid":21},"Xuqin FU",{"name":1021,"orcid":21},"Shikun Sun",{"tldr":1023,"method":1024,"finding":1025,"direction":182,"opportunity":1026},"研究沼液与生物炭配施对温室番茄土壤微环境、温室气体排放及产量的影响。","等氮磷钾滴灌试验，设沼液替代化肥比例与生物炭处理，测产量、土壤及气体。","BS75+C处理番茄产量最高达151.34 t\u002Fha，全球增温潜势最低，减排增产效果最优。","可探究不同气候区或土壤类型下沼液生物炭配施的长期固碳减排效应与微生物机制。","2026-09-30T23:30:07.179554Z",{"id":1029,"title":1030,"url":1031,"summary":1032,"summary_zh":1033,"content":21,"source_name":822,"source_url":1031,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":1034,"score_detail":1035,"sources":1037,"tags":1039,"primary_topic":1043,"secondary_topics":1044,"classification":1046,"search_phrases":1064,"slug":1067,"view_count":34,"doi":1068,"paper":1069,"created_at":1089},3811,"Automated extraction of architectural traits from apple tree branches via 3D instance segmentation and skeletonisation","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.biosystemseng.2026.104607","Accurate and efficient extraction of 3D phenotypic traits of apple tree branches is crucial for apple tree breeding and growth monitoring. Current methods, however, rely heavily on manual measurement, which are time-consuming, costly, and difficult to scale. Moreover, existing approaches based on deep learning and skeleton extraction still face critical challenges when processing high-density, large-scale point cloud data, including insufficient branch segmentation accuracy, suboptimal skeleton extraction quality, and a lack of category-specific fine-grained phenotypic quantification capabilities. To overcome these limitations, a dedicated framework for automated phenotypic trait extraction was presented. This approach consists of three key steps: (1) instance segmentation of branches from 3D data using the Relation3D model to distinguish individual branches; (2) skeletonisation of each segmented branch based on geometric characteristics and the L1-median algorithm to obtain fine-grained skeletal structures; and (3) quantitative trait extraction, where trunk height, branch count by category, and branch length are automatically computed. Experimental results demonstrate that Relation3D achieves high-precision instance segmentation, and our skeletonisation method reliably recovers branch geometry. Consequently, the framework accurately extracts key phenotypic traits, including branch number and length. This study provides an efficient and accurate solution for branch-level phenotyping, with significant potential to enhance apple tree breeding and precision horticulture.","准确高效地提取苹果树枝条的三维表型性状对于苹果树育种和生长监测至关重要。然而，现有方法严重依赖人工测量，耗时、成本高且难以规模化。此外，基于深度学习和骨架提取的现有方法在处理高密度、大规模点云数据时仍面临关键挑战，包括枝条分割精度不足、骨架提取质量欠佳以及缺乏类别特定的细粒度表型量化能力。为克服这些局限，本文提出了一种用于自动化表型性状提取的专用框架。该方法包括三个关键步骤：（1）利用Relation3D模型从三维数据中对枝条进行实例分割，以区分单个枝条；（2）基于几何特征和L1中位数算法对每个分割后的枝条进行骨架化，以获得细粒度骨架结构；（3）定量性状提取，自动计算主干高度、按类别统计的枝条数量和枝条长度。实验结果表明，Relation3D实现了高精度的实例分割，且本文的骨架化方法能够可靠地恢复枝条几何形态。因此，该框架能够准确提取关键表型性状，包括枝条数量和长度。本研究为枝条级表型分析提供了一种高效且准确的解决方案，在提升苹果树育种和精准园艺方面具有显著潜力。",77,{"impact":308,"substance":825,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":1036},"提出基于Relation3D实例分割与L1中值骨架化的苹果树枝条三维表型自动提取框架，方法新颖、结论可靠，对果树育种与精准园艺有实用价值。",[1038],{"name":822,"url":1031},[39,40,1040,1041,1042],"三维点云","果树表型","苹果育种",{"code":834,"slug":835,"name":836,"path":837,"parent_code":838},[1045],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"status":52,"locked":53,"version":54,"method":55,"evidence":1047},{"candidates":1048,"agricultural_evidence":26},[1049,1056,1060],{"code":834,"score":1050,"supported":26,"title":1051,"abstract":1053,"tags":1055},17,[1052],"apple",[836,1054,1052],"苹果",[],{"code":45,"score":33,"supported":26,"title":1057,"abstract":1058,"tags":1059},[],[304,302],[1040,40],{"code":65,"score":34,"supported":53,"title":1061,"abstract":1062,"tags":1063},[],[],[39],[1065,1066],"苹果树 三维点云 表型","苹果树 分枝 骨架提取","苹果树三维点云表型-3811","10.1016\u002Fj.biosystemseng.2026.104607",{"doi":1068,"openalex_id":1070,"authors":1071,"venue":822,"cited_by_count":106,"oa_url":1031,"card":1084,"direction":365,"ingested_from":114},"W7214750336",[1072,1075,1078,1080,1082],{"name":1073,"orcid":1074},"Anjian Fei","https:\u002F\u002Forcid.org\u002F0009-0000-9349-9223",{"name":1076,"orcid":1077},"Jintao Wu","https:\u002F\u002Forcid.org\u002F0000-0003-4257-2288",{"name":1079,"orcid":21},"Jian Zhuang",{"name":1081,"orcid":21},"Leyou Yang",{"name":1083,"orcid":21},"Hao Yang",{"tldr":1085,"method":1086,"finding":1087,"direction":365,"opportunity":1088},"提出3D实例分割与骨架化框架，自动提取苹果树枝条表型性状。","Relation3D实例分割、L1中值骨架化、高密度点云处理。","框架能高精度分割枝条并可靠提取枝长、枝数等表型性状。","可扩展至多树种、多尺度动态监测，并与基因组关联推动智能育种。","2026-09-30T23:30:05.626390Z",{"id":1091,"title":1092,"url":1093,"summary":1094,"summary_zh":1095,"content":21,"source_name":822,"source_url":1093,"published_at":23,"category":24,"cover_url":21,"hotness":25,"is_selected":53,"score":1096,"score_detail":1097,"sources":1099,"tags":1101,"primary_topic":21,"secondary_topics":1106,"classification":1107,"search_phrases":1120,"slug":1123,"view_count":170,"doi":1124,"paper":1125,"created_at":1150},3810,"Assessment of pressure drop across discs in irrigation filters: New experimental and calculation methodology","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.biosystemseng.2026.104606","The pressure drop across the discs in disc filters for irrigation systems is a key parameter for assessing hydraulic performance and suggesting design improvements to allow energy savings. However, to date, no clear procedure or method has been reported for obtaining this information. Therefore, an experimental apparatus was developed to test four types of discs from two different manufacturers. This device allowed the measurement of the pressure drop across the discs alone, without any additional element that could mask it. The effects of the disc pack height, compression, and filtration grade were investigated. Furthermore, the discs were imaged using an ultra-high resolution field emission scanning electron microscope to characterise the groove dimensions. These data were employed to calculate the pressure drop using a proposed analytical model based on the Darcy-Weisbach and minor loss equations, assuming two limiting cases for the grooves: open and closed conduits, alongside an additional bound-interpolated weighted model (BIW model) that combined both cases. The results showed an inverse relationship between the groove cross-sectional area and the pressure drop for the tested discs, while regression curves fitted to experimental data for each disc revealed a slightly quadratic behaviour. The analytical model successfully predicted a region containing experimental data. The information obtained may allow the development of improved disc filter designs.","灌溉系统叠片过滤器中叠片间的压降是评估水力性能、提出设计改进以实现节能的关键参数。然而，迄今为止，尚未有明确的程序或方法被报道用于获取该信息。因此，本研究开发了一套实验装置，对来自两个不同制造商的四种叠片进行测试。该装置能够单独测量叠片间的压降，而不受任何可能掩盖该压降的附加元件的影响。研究了叠片组高度、压缩量和过滤等级的影响。此外，利用超高分辨率场发射扫描电子显微镜对叠片进行成像，以表征沟槽尺寸。这些数据被用于通过所提出的解析模型计算压降，该模型基于达西-魏斯巴赫（Darcy-Weisbach）方程和局部损失方程，假设沟槽的两种极限情况：明渠和封闭管道，同时结合了一个额外的边界插值加权模型（BIW模型），该模型综合了上述两种情况。结果表明，对于所测试的叠片，沟槽横截面积与压降之间呈反比关系，而各叠片实验数据的拟合回归曲线则呈现出轻微的二次方特征。该解析模型成功预测了包含实验数据的区域。所获得的信息可为改进叠片过滤器设计提供依据。",75,{"impact":716,"substance":29,"depth":31,"authority":126,"freshness":127,"relevant":34,"comment":1098},"提出叠片过滤器压降测量装置与解析模型，方法新颖、数据扎实，对灌溉系统节能设计有参考价值，但属细分领域技术论文，产业影响有限。",[1100],{"name":822,"url":1093},[39,1102,1103,1104,1105],"节水灌溉","灌水器","水力性能","过滤设备",[],{"status":1108,"locked":53,"version":54,"method":55,"evidence":1109},"needs_review",{"candidates":1110,"agricultural_evidence":53,"reason":1119},[1111,1115],{"code":556,"score":308,"supported":26,"title":1112,"abstract":1113,"tags":1114},[557],[571,557],[1102],{"code":65,"score":34,"supported":53,"title":1116,"abstract":1117,"tags":1118},[],[],[39],"农业相关性需要复核",[1121,1122],"叠片过滤器 压降 试验","智慧农业 水力性能 节水灌溉 过滤设备","叠片过滤器压降试验-3810","10.1016\u002Fj.biosystemseng.2026.104606",{"doi":1124,"openalex_id":1126,"authors":1127,"venue":822,"cited_by_count":106,"oa_url":1093,"card":1145,"direction":111,"ingested_from":114},"W7214745100",[1128,1130,1133,1136,1139,1142],{"name":1129,"orcid":21},"Aniol Castells",{"name":1131,"orcid":1132},"Miquel Duran–Ros","https:\u002F\u002Forcid.org\u002F0000-0002-3286-8811",{"name":1134,"orcid":1135},"Gerard Arbat","https:\u002F\u002Forcid.org\u002F0000-0003-2192-0781",{"name":1137,"orcid":1138},"F. Ramírez de Cartagena","https:\u002F\u002Forcid.org\u002F0000-0001-8844-7461",{"name":1140,"orcid":1141},"Jaume Puig–Bargués","https:\u002F\u002Forcid.org\u002F0000-0002-6712-269X",{"name":1143,"orcid":1144},"Toni Pujol","https:\u002F\u002Forcid.org\u002F0000-0001-7013-2867",{"tldr":1146,"method":1147,"finding":1148,"direction":111,"opportunity":1149},"提出测量与计算灌溉叠片过滤器叠片压降的新实验装置与解析模型。","搭建实验台测四类叠片压降，用SEM表征沟槽，基于达西-魏斯巴赫与局部损失建模。","沟槽截面积与压降呈反比，解析模型能预测实验数据所在区域。","可结合物联网实时监测压降，开发低能耗自适应叠片过滤器设计与堵塞预警模型。","2026-09-30T23:30:05.559936Z"]