European Journal of Agronomy最新文献

{"title":"Determination of nitrogen dilution curves and nitrogen diagnosis for oilseed flax under different nitrogen and plant density","authors":"Bin Yan , Zhengjun Cui , Ming Wen , Bing Wu , Haidi Wang , Yifan Wang , Yuhong Gao","doi":"10.1016/j.eja.2025.127949","DOIUrl":"10.1016/j.eja.2025.127949","url":null,"abstract":"<div><div>Accurate assessment of nitrogen status in oilseed flax is crucial for optimizing nitrogen use efficiency (NUE) and promoting sustainable agricultural practices. The nitrogen nutrition index (NNI) serves as a key diagnostic tool; however, its response to the interaction between nitrogen fertilization and planting density remains poorly understood. To address this knowledge gap, a field experiment was conducted using three nitrogen application rates (0 (N<sub>0</sub>), 75 (N<sub>75</sub>), and 150 (N<sub>1050</sub>) kg ha⁻¹) and three planting densities (4.5 (N<sub>450</sub>), 7.5 (N<sub>750</sub>), and 10.5 (N<sub>1050</sub>) × 10⁶ plants ha⁻¹) to reveal the effects of nitrogen and planting density on dry matter and nitrogen accumulation, leaf area index (LAI), critical nitrogen dilution curve parameters, and ecaluating subsequent impacts on nitrogen nutition status and yield. Results showed that the high-density treatment (D<sub>1050</sub>) and high-nitrogen treatment (N<sub>150</sub>) significantly enhanced dry matter (DM) accumulation during the later growth stages of flax. The effect of high nitrogen on DM shifted from inhibitory during vegetative growth to promotive during reproductive growth. Planting density had a significant positive effect on DM of oilseed flax. The combination of high nitrogen and high density (N<sub>150</sub>D<sub>1050</sub>) substantially promoted DM accumulation at Kernel and Maturity stages. Both nitrogen application and planting density exerted significant effects on leaf area index (LAI), with the highest LAI values consistently observed under the combined high nitrogen and high density treatment. High nitrogen application suppressed nitrogen accumulation under low planting density but significantly enhanced it under high density. A significant nitrogen × density interaction was evident for the critical nitrogen dilution curve parameter A1 and A2. A1 was estimated with high precision and exhibited a nonlinear response to nitrogen, peaking at moderate application rates (N<sub>75</sub>) under medium planting density (D<sub>750</sub>). In contrast, parameter A2 showed greater estimation uncertainty but pronounced sensitivity to treatments, with its value sharply declining under high nitrogen, particularly at medium density (N<sub>150</sub>D<sub>750</sub>). Higher planting densities generally reduced A2 across all nitrogen levels. A major contribution of vegetative-stage DM and LAI to parameter A1; in contrast, parameter A2 was overwhelmingly contributed by seedling DM. All nitrogen treatments resulted in nitrogen deficiency (NNI < 1) at maturity. The effectiveness of nitrogen application on NNI was highly dependent on planting density. Under low density, high nitrogen maintained superior nitrogen nutrition status through maturity. At medium density, only moderate nitrogen ensured adequate nitrogen supply at the grain-filling stage, while high nitrogen led to late-stage deficiency. Under high density, high","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127949"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145747145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trait-based weed control decisions compared to economic thresholds for site-specific weed management","authors":"Mona Schatke ,&nbsp;Johanna Bensch ,&nbsp;Lena Ulber ,&nbsp;Bärbel Gerowitt ,&nbsp;Christoph von Redwitz","doi":"10.1016/j.eja.2025.127946","DOIUrl":"10.1016/j.eja.2025.127946","url":null,"abstract":"<div><div>Four decision concepts were tested for ex-ante chemical weed control decisions to be integrated into site-specific weed management (SSWM). Two concepts assess the economic profitability of a weed control treatment: one considers the abundance of weed taxonomic groups, and the other takes into account individual weed species. The third and fourth concepts utilize weed functional traits to quantify a species’ ability to provide ecosystem services (service potential) and its competitive ability (disservice potential), thereby informing management decisions. Based on grid-based manual weed assessments in five winter cereal fields in Germany, species-specific weed distribution maps were created using an interpolation approach. For each square meter of the fields, a weed control recommendation was generated using each of the four decision concepts, followed by the creation of weed control maps. Control recommendations by the two economic decision concepts showed the highest similarity across all fields, recommending weed control for 23 %–100 % (weed groups) and 6 %–100 % (species-specific) of the total field area. The concepts based on functional weed traits recommended weed control for 2 %–50 % of the area. Both economic decision concepts recommended a weed control treatment in areas recommended to be left untreated when functional traits are considered. The analysis revealed strengths and weaknesses in all concepts and recommends combining functional weed traits and economic profitability.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127946"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145614020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating canopy structure into remote sensing inversion strategies: Optimizing plant functional trait retrieval","authors":"Zixi Shi ,&nbsp;Shuo Shi ,&nbsp;Jia Sun ,&nbsp;Wei Gong ,&nbsp;Lu Xu ,&nbsp;Binhui Wang ,&nbsp;Chenxi Liu","doi":"10.1016/j.eja.2025.127931","DOIUrl":"10.1016/j.eja.2025.127931","url":null,"abstract":"<div><div>Accurate retrieval of plant functional traits is critical for monitoring crop growth and improving agronomic management. Canopy structural parameters, such as leaf area index (LAI) and leaf inclination distribution function (LIDFa), strongly influence inversion accuracy. Quantifying canopy structural uncertainties and developing strategies to improve retrieval accuracy are crucial. In this study, we developed an inversion framework based on the Soil Canopy Observation of Photosynthesis and Energy fluxes (SCOPE) model, integrating reflectance and solar-induced fluorescence (SIF) data. Using both simulation modelling and field measurements in NEON STER crop field, we introduced multi-level prior noise and evaluated how uncertainties in LAI and LIDFa propagate into the retrieval of chlorophyll content (Cab), maximum carboxylation rate (Vcmax), and fluorescence quantum efficiency (fqe). To assess the influence of canopy structure and improve retrieval accuracy, three inversion strategies—Prior-Matched (PM), Regularized (RI), and No-Prior (NP)—were designed and tested for their accuracy and robustness. The results showed that second-order Sobol’ indices (S2) captured interactions among canopy structural parameters and functional traits, particularly between Cab-LAI, Cab-LIDFa and fqe-LAI, with sensitive spectral ranges at 680–740 nm (fluorescence) and 600–720 nm (reflectance). Error amplification analysis under six noise levels showed that structural uncertainties significant amplified reflectance and fluorescence variations, with red-edge shifts (<span><math><mrow><mi>Δ</mi><mi>RE</mi></mrow></math></span>) and 740 nm fluorescence changes (<span><math><mrow><mi>Δ</mi><msub><mrow><mi>F</mi></mrow><mrow><mn>740</mn></mrow></msub></mrow></math></span>) being most sensitive. Incorporating prior canopy structure information improved inversion accuracy by up to 7.93 % in R² and reduced RMSE by 21.25 %, although this advantage diminished under high noise levels. LAI uncertainty had a greater impact than LIDFa, and additive noise introduced more uncertainty than multiplicative noise. Comparison of the inversion strategies revealed that the RI strategy achieved higher accuracy (simulated data: R²=0.954; measured data: R²=0.799) and greater robustness to noise than the PM strategy. These findings demonstrate the value of integrating canopy structure into computational inversion models to enhance the reliability of remote sensing trait retrieval, supporting precision agriculture and sustainable crop production.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127931"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145614019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic parameter calibration based deep network for paddy yield prediction","authors":"S. ABARNA ,&nbsp;D. KESAVARAJA","doi":"10.1016/j.eja.2025.127929","DOIUrl":"10.1016/j.eja.2025.127929","url":null,"abstract":"<div><div>Paddy yield prediction plays a crucial role in agriculture, enabling farmers to make informed decisions. This work proposes an innovative approach combining Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) for accurate paddy yield forecasting. The hybrid model harnesses the spatial understanding capabilities of CNNs and the sequential learning ability of RNNs to capture both local and temporal dependencies in agricultural data. A key enhancement introduced in this method is the incorporation of a dynamic parameter calibration technique. Traditional regularization methods often rely on static values, which may not adapt effectively to varying complexities in the dataset. The proposed approach dynamically adjusts the regularization strength parameters during training, allowing the model to better converge to different patterns and fluctuations in paddy growth parameters. The dataset utilized for training and evaluation comprises comprehensive agricultural variables, including soil composition, climate conditions, and historical yield data. Experiments demonstrate the effectiveness of the hybrid CNN-RNN architecture with dynamic parameter calibration which improves prediction accuracy over conventional models. This research contributes to the advancement of precision agriculture by providing a more robust and adaptable framework for paddy yield prediction. The integration of spatial and temporal features, along with dynamic parameter calibration, showcases the potential for optimizing agricultural decision-making processes and mitigating the impact of unpredictable factors on paddy production.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127929"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145599014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identifying optimum sowing dates for rainfed soybeans in Brazil’s new agricultural frontier","authors":"José Wanderson Silva dos Santos ,&nbsp;George do Nascimento Araújo Junior ,&nbsp;Francisco Edson Paulo Ferreira ,&nbsp;Iêdo Peroba de Oliveira Teodoro ,&nbsp;Elvis Felipe Elli ,&nbsp;Wemerson Saulo da Silva Barbosa ,&nbsp;Ricardo Araújo Ferreira Junior ,&nbsp;Guilherme Bastos Lyra ,&nbsp;Iedo Teodoro ,&nbsp;Ivomberg Dourados Magalhães ,&nbsp;Adolpho Emanuel Quintela Rocha ,&nbsp;Alexsandro Claudio dos Santos Almeida","doi":"10.1016/j.eja.2025.127927","DOIUrl":"10.1016/j.eja.2025.127927","url":null,"abstract":"<div><div>Expanding soybean production into new, challenging environments will require advancing knowledge of agriculture management technologies to close yield gaps and enhance system’s sustainability. Crop modeling can be a powerful tool to accurately simulate crop growth and productivity under various cultivation conditions. In this study, the FAO AquaCrop model was calibrated and validated to simulate canopy cover (CC), biomass (B), grain yield (Y), soil water content (SWC), and crop evapotranspiration (ETc) for five soybean cultivars with maturity groups ranging from 6 to 9 under rainfed conditions in the Coastal Tablelands region of Alagoas, Brazil. Model performance was evaluated by modeling efficiency (EF), mean error (E), root mean square error (RMSE), normalized root mean square error (NRMSE), Willmott's index (d), and Pearson’s correlation coefficient (r). The model’s performance for CC ranged from acceptable to good, with the best results for cultivars BRS9383 and M6410. Cultivars AS3730, BMX-POTÊNCIA, and BRS9383 showed the best fit for B (20 % &gt; NRMSE &lt; 30 %; RMSE &lt; 1.30 tons ha⁻¹; EF above 0.8). The model overestimated SWC (E = 16.6 %) but demonstrated low error in predicting ETc for the cultivars (-18–19 mm). After calibration, the model was applied in long-term simulations to evaluate the impacts of alternative sowing dates on crop productivity. For this purpose, a 49-year series of meteorological data was used. The period from the second half of April to the second half of June is recommended for sowing soybean cultivars in the region. During this period, the average soybean productivity can reach 2.72 (±0.38) tons ha⁻¹. Cultivars M6410 (maturity group 6.4) and BMX-POTÊNCIA (maturity group 6.7) were the most suitable for cultivation in the region. AquaCrop model is powerful to optimize rainfed soybean management for cultivation in Brazil’s new agricultural frontier.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127927"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impacts of atmospheric CO2 enrichment on nitrous oxide emissions in wheat and rice cropping systems at global and local scales","authors":"Shengji Yan ,&nbsp;Guiyao Zhou ,&nbsp;Daniel Revillini ,&nbsp;Yunlong Liu ,&nbsp;Haoyu Qian ,&nbsp;Aixing Deng ,&nbsp;Yanfeng Ding ,&nbsp;Yu Jiang ,&nbsp;Manuel Delgado-Baquerizo ,&nbsp;Xin Zhang ,&nbsp;Weijian Zhang","doi":"10.1016/j.eja.2025.127970","DOIUrl":"10.1016/j.eja.2025.127970","url":null,"abstract":"<div><div>Human activity is boosting atmospheric carbon dioxide (CO₂), which further compounds the contributions to climate change through interaction effects with other greenhouse gases, especially nitrous oxide (N₂O). However, the global-scale role of elevated CO₂ (eCO₂) in shaping N₂O emissions across major cereal systems remains insufficiently studied. This effect can be especially important in cereal crops such as rice and wheat, which are among the most predominant crops on the planet and require different management strategies. Here, we combined meta-analysis of a global database (including rice and wheat cropping systems) with a two-year rice-wheat cropping experiment in eastern China; and found that eCO₂ consistently promotes N₂O emissions both at local and global scales in wheat cropping systems. For meta-analysis, we show that wheat cropping increased eCO₂-induced N₂O emissions by 19.6 %, whereas rice cropping showed no significant changes. Local experiments supported the global results and revealed a potential functional mechanism for the positive relationship between eCO₂ and N₂O emissions, where eCO₂ experimentally increased the ratio of microbial nitrite reductase gene abundances (<em>nirK</em>, <em>nirS</em>) to N₂O reductase gene abundance (<em>nosZ</em>) in soil. Taken together, our study highlights the potential positive feedback among eCO₂ and N₂O, as well as the crucial role of cereal type in governing the eCO₂ effect on N₂O emissions, which is an important consideration for management to both mitigate climate change under global change and promote agricultural sustainability.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127970"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Response of maize productivity and protein quality to straw incorporation combined with nitrogen fertilizer in Northeast China: A 10-year field experiment","authors":"Anran Long ,&nbsp;Xinjie Ji ,&nbsp;Xiangyu Li ,&nbsp;Xuelian Wang ,&nbsp;Lv Yang ,&nbsp;Liyun Chang ,&nbsp;Jingwen Yang ,&nbsp;Ying Jiang ,&nbsp;Hua Qi ,&nbsp;Xiangwei Gong","doi":"10.1016/j.eja.2025.127965","DOIUrl":"10.1016/j.eja.2025.127965","url":null,"abstract":"<div><div>The combination of straw and nitrogen (N) fertilizer enhances ecological sustainability and food security; however, few studies have examined how maize (<em>Zea mays</em> L.) productivity, protein synthesis, and protein structure–function relationships respond to these practices. In 2015, a study used a two-factor split design with eight treatments: two straw incorporation methods, rotary (RTS) and plow tillage (PTS), and four N fertilizer levels (N0, N1, N2, N3; 0, 112, 187, and 262 kg ha⁻¹). Compared with RTS, PTS significantly increased the maize biomass accumulation by 9.7 % and grain yield by 3.5 % over three years. In contrast, higher N use efficiency, activities of N metabolism enzymes, and increased protein components during the grain-filling stage were observed under RTS conditions than under PTS conditions, thereby increasing the protein and amino acid contents of mature grains. N2 treatment resulted in a smoother protein surface and a more stable secondary and tertiary structure than the other N fertilizer treatments, which was conducive to optimizing the physicochemical properties of maize proteins in RTS combined with N2 practices. Partial least squares path modeling and random forest analyses revealed that high yield and superior protein quality could not be enhanced simultaneously under the current management practices. Overall, PTS increased maize yield, whereas RTS optimized protein quality in Northeast China in a 10-year field experiment. Our results provide important references for improving productivity and protein quality from the perspective of straw incorporation and N fertilizer management at the field scale when considering different requirements.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127965"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Monitoring crop leaf area index using improved global structure-from-motion and multi-feature data fusion on a phenotyping robot","authors":"Miao Su ,&nbsp;Weixing Cao ,&nbsp;Yaze Yun ,&nbsp;Peng Xia ,&nbsp;Yue Guo ,&nbsp;Xiangming Zhou ,&nbsp;Chongya Jiang ,&nbsp;Jie Jiang ,&nbsp;Yan Zhu ,&nbsp;Xia Yao ,&nbsp;Dong Zhou","doi":"10.1016/j.eja.2025.127974","DOIUrl":"10.1016/j.eja.2025.127974","url":null,"abstract":"<div><div>Reconstruction of crop three-dimensional (3D) point clouds is essential for monitoring phenotypic parameters, like plant height and leaf area index (LAI), which is a critical phenotype predictor for smart crop breeding. The main 3D reconstruction technologies include image-based approaches, laser scanning, and depth camera methods. Among these methods, image-based structure-from-motion (SfM) is widely used due to its low cost and high accuracy. However, field crop canopy image data for high-resolution point cloud construction are often large-scale, unordered, and uncalibrated. Conventional SfM methods struggle with 3D reconstruction due to high computational costs and long processing times, delaying phenotypic analysis. To address this issue, we developed an improved global SfM algorithm, which increases the point cloud reconstruction speed by an average of 1.39 times compared to traditional incremental SfM methods and by more than 10 % on average compared to two mainstream global SfM algorithms. In addition, we integrated three types of predictors, point cloud features, color indices and texture features, through multi-feature data fusion and machine learning. A random forest algorithm for the prediction of LAI for a combined data set of four different crops, and using all three categories of predictors, achieved higher monitoring accuracy compared to using a single feature category (R²=0.78 vs R²=0.71–0.74). This new method, which includes an improved global SfM algorithm and a three-predictor fusion-based LAI monitoring approach, offers an efficient and reliable solution for precise crop phenotyping and continuous growth monitoring in complex field environments, enabling accurate assessment of crop morphology and developmental dynamics.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127974"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating peat pasture performance under elevated groundwater conditions","authors":"Hilde M. van Dijk ,&nbsp;Nyncke J. Hoekstra ,&nbsp;Raimon Ripoll-Bosch ,&nbsp;Idse Hoving ,&nbsp;Jeroen Pijlman ,&nbsp;Nick van Eekeren","doi":"10.1016/j.eja.2025.127961","DOIUrl":"10.1016/j.eja.2025.127961","url":null,"abstract":"<div><div>Historically, Dutch peatlands have been drained for agriculture, particularly as pastures for dairy production. While drainage increases productivity, it degrades the peat layer, leading to high greenhouse gas emissions and soil subsidence. Raising the depth of the groundwater table (WT) to 20 cm below field level, contrasted to commonly drained peat at approximately 50 cm below field level, is considered an effective solution to limit peat degradation. However, it is unclear how raising the WT with an active water infiltration system (AWIS) would affect grass yield and quality, and whether dairy farming can be maintained. We studied the effect of raising WT on grass production, quality, and nitrogen utilization, using experimental plots with different WT and fertilization levels in 2020–2024. Raising WT from 50 to 20 cm below field level resulted in an average decrease of 9 % in herbage dry matter yield (DMY), with significant variation between years due to varying weather conditions. The reduction in DMY could partly be explained by the observed decrease in soil N supply in wetter conditions, due to a lower mineralization rate. Increased N fertilization could mediate the lower DMY but increases nutrient losses. Herbage crude protein concentration was moderately affected by WT, and more strongly by yearly variation. Although results are on plot level, which excludes losses originating from trafficking by machinery and animals, they indicate that raising WT is compatible with relatively high grass production and similar grass quality. Consequences on farm level need to be further explored.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127961"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145731697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and validation of a crop yield prediction framework accounting for regional heterogeneity: A case study of spring maize in Jilin Province","authors":"Jun Gao ,&nbsp;Daming Dong ,&nbsp;Wenjiang Huang ,&nbsp;Yingying Dong ,&nbsp;Tongren Xu ,&nbsp;Chongya Jiang ,&nbsp;Kun Wang","doi":"10.1016/j.eja.2025.127977","DOIUrl":"10.1016/j.eja.2025.127977","url":null,"abstract":"<div><div>Accurate and timely crop-yield prediction is essential for ensuring food security, managing agricultural risk, and supporting policy formulation. To address the respective limitations of traditional crop growth models and deep learning methods under complex environmental conditions, a hybrid modeling framework is proposed that integrates remote-sensing data assimilation, a process-based crop growth model, and deep learning techniques. Using spring maize in Jilin Province, China (2015–2020) as the case study, leaf area index (LAI) retrieval accuracy is first improved by coupling the PROSAIL model with machine-learning algorithms. A complete meteorological sequence for the target year is then constructed using a dynamic time warping (DTW) algorithm to overcome early-season prediction challenges caused by missing real-time weather data. Retrieved LAI is assimilated into the WOFOST crop growth model through an ensemble Kalman filter (ENKF) to calibrate state variables and enable dynamic yield prediction across growth stages. Finally, a deep learning model (Convolutional Neural Network–Attention Long Short-Term Memory with Multi-Task Learning, CNN-ALSTM-MTL) is developed to fuse assimilation outputs with multi-source heterogeneous data, leveraging multi-task learning to enhance adaptability to regional heterogeneity and improve yield prediction performance at the regional scale. Assimilation is found to substantially improve maize-yield estimation, increasing R² by 0.2 and reducing RMSE by 276 kg ha⁻¹. Compared with the assimilated crop growth model alone, the hybrid framework further increases R² by 35 % and decreases RMSE by 23 % by hierarchically capturing feature information relevant to maize-yield estimation. The best performance is achieved during the key growth stage (jointing to tasseling stage), with an R² of 0.75 and an RMSE of 592 kg ha⁻¹, enabling reliable yield prediction approximately two months before harvest. This framework demonstrates potential for cross-crop and cross-regional applications and provides robust methodological support for regional-scale yield forecasting and food-security early warning.</div></div>","PeriodicalId":51045,"journal":{"name":"European Journal of Agronomy","volume":"174 ","pages":"Article 127977"},"PeriodicalIF":5.5,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}