AgroFlux: A Spatial-Temporal Benchmark for Carbon and Nitrogen Flux Prediction in Agricultural Ecosystems
AgroFlux: 農業生態系における炭素・窒素フラックス予測のための時空間ベンチマーク (AI 翻訳)
Qi Cheng, Licheng Liu, Yao Zhang, M. Hong, Yiqun Xie, Xiaowei Jia
🤖 gxceed AI 要約
日本語
農業生態系からの温室効果ガス排出の正確な定量化は重要だが難しい。この研究では、物理モデルシミュレーションと実観測を統合した初の時空間ベンチマークデータセット「AgroFlux」を構築し、LSTM、CNN、Transformerなどの深層学習モデルで炭素・窒素フラックス予測を評価した。転移学習によりシミュレーションデータの有効性も確認。AI駆動の農業生態系モデル開発に貢献。
English
This work introduces AgroFlux, a first-of-its-kind spatiotemporal benchmark dataset for carbon and nitrogen flux prediction in agricultural ecosystems, integrating simulations from Ecosys and DayCent with observations from flux towers. It evaluates various deep learning models (LSTM, TCN, Transformers) and explores transfer learning from simulated to real data, advancing AI-driven agroecosystem modeling for GHG mitigation.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本の農業分野では、農林水産省が「みどりの食料システム戦略」で温室効果ガス削減目標を掲げるが、排出量の高精度推定が課題。AgroFluxはAIによるフラックス予測の基盤を提供し、日本の水田や畑作地への応用が期待される。SSBJやGHG排出量算定への間接的貢献も考えられる。
In the global GX context
Globally, the agricultural sector accounts for ~25% of GHG emissions, yet accurate measurement remains challenging. AgroFlux addresses this by providing a benchmark dataset and evaluation framework for AI-based flux prediction, which is critical for climate-smart agriculture policies and carbon credit markets. The transfer learning approach is particularly relevant for scaling models across regions with limited ground data.
👥 読者別の含意
🔬研究者:This benchmark dataset and model evaluation framework provides a standardized testbed for advancing deep learning methods in agroecosystem GHG prediction.
🏢実務担当者:AgroFlux offers a pathway to deploy AI-driven flux estimation for farm-level carbon accounting and sustainability reporting.
🏛政策担当者:The dataset supports evidence-based policy on agricultural GHG mitigation by enabling more accurate and scalable emission monitoring.
📄 Abstract（原文）
Agroecosystem, which heavily influenced by human actions and accounts for a quarter of global greenhouse gas emissions (GHGs), plays a crucial role in mitigating global climate change and securing environmental sustainability. However, we can't manage what we can't measure. Accurately quantifying the pools and fluxes in the carbon, nutrient, and water nexus of the agroecosystem is therefore essential for understanding the underlying drivers of GHG and developing effective mitigation strategies. Conventional approaches like soil sampling, process-based models, and black-box machine learning models are facing challenges such as data sparsity, high spatiotemporal heterogeneity, and complex subsurface biogeochemical and physical processes. Developing new trustworthy approaches such as AI-empowered models, will require the AI-ready benchmark dataset and outlined protocols, which unfortunately do not exist. In this work, we introduce a first-of-its-kind spatial-temporal agroecosystem GHG benchmark dataset that integrates physics-based model simulations from Ecosys and DayCent with real-world observations from eddy covariance flux towers and controlled-environment facilities. We evaluate the performance of various sequential deep learning models on carbon and nitrogen flux prediction, including LSTM-based models, temporal CNN-based model, and Transformer-based models. Furthermore, we explored transfer learning to leverage simulated data to improve the generalization of deep learning models on real-world observations. Our benchmark dataset and evaluation framework contribute to the development of more accurate and scalable AI-driven agroecosystem models, advancing our understanding of ecosystem-climate interactions.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.48550/arxiv.2602.01614first seen 2026-06-29 08:38:14
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。