邁向數(shù)智農(nóng)業(yè)：場景、數(shù)據(jù)與智能

全球農(nóng)業(yè)正加速向數(shù)智驅(qū)動范式演進，核心是通過算法重構(gòu)人類認知范式（如決策模型優(yōu)化）與物理操作流程（如自主農(nóng)機系統(tǒng)），實現(xiàn)數(shù)據(jù)智能與農(nóng)業(yè)場景的深度耦合。這一進程催生了橫跨農(nóng)學(xué)、數(shù)據(jù)科學(xué)、環(huán)境科學(xué)等的交叉領(lǐng)域——\"農(nóng)業(yè)數(shù)據(jù)智能\"?！掇r(nóng)業(yè)大數(shù)據(jù)學(xué)報》設(shè)立\"數(shù)據(jù)智能\"長期專欄，圍繞“場景-數(shù)據(jù)-智能”的創(chuàng)新三角，匯聚該領(lǐng)域前沿研究與高質(zhì)量數(shù)據(jù)，推動農(nóng)業(yè)數(shù)智化知識體系發(fā)展。專欄將聚焦以下方向：

農(nóng)業(yè)數(shù)據(jù)的人工智能適配（AI-readyAgriData）

新一代人工智能（artificial intellgence，AI）展現(xiàn)了高度的數(shù)據(jù)依賴性，凸顯了農(nóng)業(yè)等領(lǐng)域中既有數(shù)據(jù)正在面臨的人工智能適配（AI-ready）挑戰(zhàn)。專欄關(guān)注農(nóng)業(yè)AI-ready 數(shù)據(jù)集的構(gòu)建方法與技術(shù)，聚焦農(nóng)業(yè)數(shù)據(jù)（遙感影像、表型組學(xué)、環(huán)境傳感等）高效標注、多模態(tài)對齊、智能適配度評價與規(guī)范等議題，促進AI-ready數(shù)據(jù)集建設(shè)與出版，推動滿足大模型訓(xùn)練需求等智能適配農(nóng)業(yè)領(lǐng)域知識庫的建設(shè)。

農(nóng)業(yè)數(shù)據(jù)處理與分析的智能進化（AIforAgriData）

農(nóng)業(yè)巨系統(tǒng)的開放性與復(fù)雜性，催生了高維異構(gòu)、多模態(tài)、超大規(guī)模和時空異質(zhì)數(shù)據(jù)的系統(tǒng)性涌現(xiàn)，傳統(tǒng)方法面臨維度災(zāi)難、模態(tài)鴻溝和時空耦合建模等多重挑戰(zhàn)，需通過農(nóng)學(xué)機理（例如，作物生長模型）與計算智能的深度耦合，建立“數(shù)據(jù)-知識-決策”貫通的農(nóng)業(yè)數(shù)據(jù)智能處理范式。專欄關(guān)注農(nóng)業(yè)數(shù)據(jù)的智能處理與分析，聚焦表型數(shù)據(jù)高通量采集與邊緣計算、農(nóng)業(yè)時序數(shù)據(jù)的自適應(yīng)特征提取、多源異構(gòu)數(shù)據(jù)融合與跨模態(tài)知識發(fā)現(xiàn)等議題，促進農(nóng)學(xué)機理驅(qū)動和“數(shù)據(jù)-模型”協(xié)同進化的數(shù)據(jù)處理與分析智能化演進。

場景驅(qū)動的農(nóng)業(yè)數(shù)智融合（Scenario-IntelligenceFusion forAgri-Innovation）

人類認知活動的算法替代根本上改變了農(nóng)業(yè)決策的形成與實施，融合日益廣泛深化的人類物理活動機械替代，越來越多的農(nóng)業(yè)場景正在以“數(shù)據(jù)-知識-行動”的閉環(huán)模式成為通向數(shù)智農(nóng)業(yè)的一級級臺階。專欄關(guān)注農(nóng)業(yè)場景的數(shù)智融合創(chuàng)新，聚焦智能育種中的表型-基因關(guān)聯(lián)挖掘、精準種養(yǎng)中的動態(tài)決策系統(tǒng)、農(nóng)業(yè)產(chǎn)業(yè)鏈的數(shù)字孿生建模等前沿方向，重點征集融合農(nóng)學(xué)機理與數(shù)據(jù)智能的跨學(xué)科研究成果，推動形成可解釋、可復(fù)制的農(nóng)業(yè)數(shù)智化范式。

智能時代的農(nóng)業(yè)數(shù)據(jù)治理變革（AgriDataManagementwithinAI）

農(nóng)業(yè)數(shù)據(jù)治理面臨開放共享與隱私保護、流通效率與權(quán)益歸屬、算法權(quán)力與倫理約束的三重悖論。人工智能在提供隱私計算和可解釋性工具的同時，也增加了治理復(fù)雜度。專欄關(guān)注智能時代農(nóng)業(yè)數(shù)據(jù)的治理挑戰(zhàn)，聚焦農(nóng)業(yè)數(shù)據(jù)隱私計算、可解釋AI、區(qū)塊鏈與數(shù)據(jù)保護、數(shù)據(jù)信托與小農(nóng)戶數(shù)據(jù)資產(chǎn)化、農(nóng)業(yè)數(shù)據(jù)的社會化流通和生態(tài)化協(xié)作等關(guān)鍵議題，推動效率與安全兼顧的農(nóng)業(yè)數(shù)據(jù)治理體系。

開放復(fù)雜農(nóng)業(yè)巨系統(tǒng)的數(shù)字化表達，超高維多模態(tài)大規(guī)模農(nóng)業(yè)數(shù)據(jù)的智能化處理與分析，以及人機混合智能系統(tǒng)賦能的農(nóng)業(yè)場景，正在塑造數(shù)智農(nóng)業(yè)并激發(fā)一系列跨學(xué)科前沿研究。專欄以促進數(shù)智農(nóng)業(yè)發(fā)展為目標，誠邀全球?qū)W者關(guān)注農(nóng)業(yè)數(shù)據(jù)智能理論前沿與實踐邊界，共同推動農(nóng)業(yè)數(shù)智化知識系統(tǒng)發(fā)展。

主編：周國民

Agriculture is rapidly evolving towards a data amp; intelligence-driven paradigm， centered on algorithmization of both human cognition （e.g.，by optimizing decision-making models） and physical operational processes （e.g.，by autonomous agricultural machinerysystems）.This enables the deep integration of data， inteligence withinreal-world agricultural scenarios. Such advancements have catalyzed the emergence of an interdisciplinary field spaning agronomy， data science， and environmental science — \"Agricultural Data Intelligence （ADI）\".

The \"Dataamp; Inteligent\" section of The Journal ofAgricultural Big Data invites submissions for its long-term dedicated section，established to advance the frontierof agricultural data intelligence within the evolving paradigmof agriculture.The sectionis designed to foster cuting-edgeresearchand high-qualitydatacontributions in the field， focusing onthe innovative \"scenario-data-intelligence\"triangle，and promote the knowledge development of data and intelligent agriculture.

The section invites papers on the next topics.

1. Agricultural Data's Artificial Intelligence Adaptation （Al-ready AgriData）

The advent of next-generationartificial inteligence （AI） underscores the critical need for data to be AI-ready， particularly in agriculture where existing datasets facesignificant adaptationchallnges.This sub-theme solicits research on the construction and technical methodologies for AI-ready agricultural datasets.Topics of interest include eficient labelingof agriculturaldata （e.g.，remote sensing imagery，phenomics，environmental sensors），multimodal dataalignment，inteligent adaptabilityevaluation，andstandardization. Contributions thatadvance the development， publication，andutilizationof AI-readydatasets to meet the training demandsof large modelsand build knowledge bases for agricultural intelligence are especially encouraged.

2. Intelligent Evolution of Agricultural Data Processing and Analysis （Al for AgriData）

The complexityandopennessofagricultural mega-systems generate high-dimensional，heterogeneous，multimodal， andspatiotemporally diverse datasets，posing chalenges such as the curse of dimensionality，modal gaps，and spatiotemporal coupling modeling.This direction cals forresearch that integratesagronomic principles （e.g.，crop growth models）withcomputational intellgence to establish aseamless\"data-knowledge-decision\" paradigm.We invite submissions focusing on high-throughput phenotyping data collction and edge computing，adaptive feature extraction from agricultural time-series data，fusion of multi-source heterogeneous data，and cros-modal knowledge discovery. Emphasis is placed on studies that drive the intellgent evolution ofdata processing and analysis through agronomic mechanism-driven and \"data-model\" co-evolution approaches.

3.Scenario-Driven Agricultural Digital-Intelligent Fusion （Scenario-Intelligence Fusion for AgriInnovation）

The algorithmic substitution of human cognitive activities，combined with the mechanical replacement of physical tasks，istransformingagricultural decision-makingand implementation.Thissub-theme exploreshowagricultural scenariosareevolving intodigital-intelligent frameworks through\"data-knowledge-action\"closed loops.Weseek inovative research on phenome-genome assciation mining in intelligent breeding，dynamic decision systems in precision farming，and digital twin modeling across agricultural value chains.Priority willbe given to interdisciplinary studies that integrate agronomic principles with data inteligence，promoting interpretable and replicable paradigms for agricultural digitalization.

4. Agricultural Data Governance Transformation in the Intelligent Era （AgriData Management within AI）

The governance of agricultural data is confronted with paradoxes involving open sharing versus privacy protection， circulation eficiency versus rights atribution，and algorithmic authority versus ethical constraints.The rise of AI introduces tools like privacy computing and explainable AI， yet it also escalates governance complexity.This direction invites research addressing key governance challenges， including privacy-preserving computations， explainable AI in agriculture，blockchainfor data protection，data trusts，the asetization of data forsmalholder farmers，and the socialized circulation and ecologicalcollaboration ofagricultural data.Submissions that proposebalanced solutions for efficiency and security in agricultural data governance are highly encouraged.

The \"Data amp; Intelligent\" section aims to shape the future of digital agriculture by addressing the digital representation of open complex agricultural mega-systems，the intelligent processing and analysis of ultra-highdimensional multimodal large-scale agricultural data，andthe empowerment of human-machine hybrid inteligent systems inagricultural scenarios.We invite scholars worldwide to contribute to the theoretical frontiers and practical boundares of agricultural dataintellgence，colaboratively advancing the knowledge system of agricultural digitalization and intelligence.Submit your originalresearch to TheJournal ofAgricultural Big Data and join us in this transformative journey.

Editorin chief： ZHOUGuoMin