Agronomy Journal最新文献

{"title":"Simulating maize yield and nitrogen leaching under smart fertilizer strategies in subtropical sandy soils","authors":"Dereje Ademe Birhan,&nbsp;Sukhdeep Singh,&nbsp;Lakesh K. Sharma","doi":"10.1002/agj2.70532","DOIUrl":"https://doi.org/10.1002/agj2.70532","url":null,"abstract":"<p>Efficient nitrogen (N) management in Florida's coarse, N-leaching-prone soils is challenging because high N rates are often used to avoid deficiency. This study evaluated whether controlled-release (CRF) and slow-release (SRF) fertilizers can improve maize (<i>Zea mays</i> L.) yield and reduce NO₃–N leaching compared with urea under varying climate. A previously calibrated crop environment resource synthesis-Maize model for the maize hybrid Pioneer P1870 was re-evaluated with 2 years of field data from Live Oak, FL. The 2023 experiment included an unfertilized control, CRF, and SRF at 235, 314, and 392 kg N ha⁻¹. In 2024, the same CRF and SRF treatments were repeated, and split-applied urea at the same rates was added. The model reproduced above-ground biomass, grain yield, and N uptake well (<i>d</i> = 0.81–0.99) and soil NO₃–N with moderate skill (<i>d</i> = 0.57). The evaluated model was then used to run a 30-year simulation (1995–2024) for CRF, SRF, and urea applied at 0–400 kg N ha⁻¹ in 50 kg increments, and outputs were analyzed by growing-season rainfall and temperature categories. Rainfall, not temperature, significantly affected all variables. Across rainfall regimes, CRF gave the highest agronomic and environmental performance. At its source-specific optimum N rate, CRF increased grain yield by ∼6%–7% and cut NO₃–N leaching by ∼40%–45% relative to urea, whereas SRF reduced leaching (∼25% under normal rainfall) with smaller yield gains. Optimal N rates were highest for CRF, intermediate for SRF, and lowest for urea.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148942396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Seeding rate effects on soybean seed yield components and economic return","authors":"Fernando M. Marcos,&nbsp;Caio L. dos Santos,&nbsp;Natan A. Seraglio,&nbsp;Mark A. Licht","doi":"10.1002/agj2.70534","DOIUrl":"https://doi.org/10.1002/agj2.70534","url":null,"abstract":"<p>Soybean (<i>Glycine max</i> [L.] Merr.) seed prices have increased by more than 190% from 1998 to 2018 in the US, making seeding rate one of the most cost-sensitive management decisions. While soybean can maintain maximum seed yield across a wide range of plant populations due to their morphological and reproductive plasticity, the decreasing commodity prices and increased seed cost necessities re-evaluation of economic optimum seeding rates. This study aimed to evaluate soybean yield responses and economic return across a wide range of seeding rates and environments in Iowa, as well as seed quality. Across site-years, the agronomic optimum seeding rate (AOSR) was 234,300 seeds ha^−1, where seed yield plateaued at 4.7 Mg ha⁻¹. Results showed a higher number of pods and seeds at seeding rates lower than 234,300 seeds ha⁻¹; however, at higher seeding rates, the number of pods and seeds were consistently lower. There was no relationship between individual seed weight, oil and protein content, and seeding rates. Seeding rates above the AOSR resulted in no benefit under favorable conditions, since higher plant populations mostly benefit environments exposed to weather events that reduce population (e.g., hail). The economic optimum seeding rate (EOSR) averaged approximately 90.8% of the AOSR. Seed yields are maximized by targeting the AOSR; however, the additional yield from the difference between AOSR and EOSR did not bring financial gains.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/agj2.70534","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148942443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Supplemental nitrogen nanofertilizer doses in maize following soybean: Physiological, nutritional, biochemical, and agronomic responses","authors":"Alessandro Guerra da Silva,&nbsp;Nathalia Dias Batista Silva,&nbsp;María Belén Ciacci,&nbsp;Marcio Rosa,&nbsp;Ricardo de Castro Dias,&nbsp;Augusto Matias de Oliveira,&nbsp;Julia Borges Vilela Carvalho,&nbsp;Arthur Cunha França","doi":"10.1002/agj2.70529","DOIUrl":"https://doi.org/10.1002/agj2.70529","url":null,"abstract":"<p>In recent years, nanofertilizers have gained popularity as a means of enhancing maize (<i>Zea mays</i> L.) yield. However, there is limited information on their effectiveness when maize is grown in succession to soybean (<i>Glycine max</i> (L.) Merr.). Therefore, this study aimed to evaluate the effects of supplemental nitrogen nanofertilizer on the physiological, nutritional, biochemical, and agronomic variables of maize grown in succession to soybean cultivation. Field experiments were conducted in the Midwest region of Brazil from 2022 to 2024 in a randomized complete block design with six replications and five treatments corresponding to MIST-N nanofertilizer doses (0-control, 1.0, 2.0, 3.0, and 4.0 L ha⁻¹) on maize following soybean. The supplemental nitrogen nanofertilizer doses did not consistently and significantly affect most agronomic, physiological, or biochemical variables. The crop exhibited a major leaf area and grain yield in the crop year with favorable precipitation conditions (2024), without the effect of the nanofertilizer. In the crop year with less rainfall (2022), increasing the supplemental nanofertilizer doses up to 1.6 L ha⁻¹ of MIST-N promoted lower plants, higher number of ears, grain protein content, and grain yield. The results of this study allow us to conclude that generalized recommendations for nitrogen nanofertilizer use should not be made when maize is grown after soybean crop.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/agj2.70529","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148942448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating camelina genotypes for improved nitrogen use efficiency and yield with reduced nitrogen","authors":"Chengci Chen,&nbsp;Maral Etesami,&nbsp;William Franck,&nbsp;Chrisanne Kuester,&nbsp;Sooyoung Franck,&nbsp;Shreya Gautam,&nbsp;Tehmina Arshad,&nbsp;Chaofu Lu","doi":"10.1002/agj2.70527","DOIUrl":"https://doi.org/10.1002/agj2.70527","url":null,"abstract":"<p>Camelina (<i>Camelina sativa</i> L.) is a promising bioenergy crop with wide environmental adaptation, but nitrogen (N) requirements affect the sustainable production. Sixteen camelina genotypes were planted in 3 years in Sideny, MT at high and low N levels to evaluate nitrogen use efficiency (NUE) and yield in response to soil N. Results showed extensive variations among genotypes for plant height, seed yield, and seed oil and N concentrations at both N levels. The top-yielding genotypes were MT143 (2122 kg ha⁻¹) and MT229 (2119 kg ha⁻¹) across years and N levels. Genotypes MT144 and MT229 had the highest oil concentrations independent of N levels with an average value of 407 g kg⁻¹ for both genotypes. The nitrogen use efficiency based on oil yield (NUEOY) determined by (1) N uptake efficiency (NUpE) and (2) N transfer efficiency based on oil yield (NUtEOY) showed large genotypic variations. MT143 had the highest NUpE (2.38 kg kg⁻¹), especially at low N, and MT229 and MT144 had the highest NUtEOY (8.67 and 8.58 kg kg⁻¹, respectively). Overall, especially in low N, MT229 had the highest NUEOY (19.47 kg oil kg N⁻¹) followed by MT143 (18.12 kg oil kg N⁻¹). High broad-sense heritability (<i>H</i>²b) values were observed for seed yield and oil concentration, but not for seed N concentration at low N. The genotype + genotype-by-environment interaction biplots demonstrated which genotype won low and high N environments, and the analysis of means showed genotypes that had traits significantly higher or lower than the group means. The information and genetic materials may be used for further plant breeding.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/agj2.70527","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148942407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Subsoil water nitrate and phosphate dynamics in an organic integrated crop–livestock rotation system","authors":"Binaya Baral,&nbsp;Anuj Chiluwal,&nbsp;Nathaniel J. Colten,&nbsp;Arjun Chhetri,&nbsp;Ellyn Anthony,&nbsp;Kenneth Andries,&nbsp;Shawn T. Lucas","doi":"10.1002/agj2.70516","DOIUrl":"https://doi.org/10.1002/agj2.70516","url":null,"abstract":"<p>Organic integrated crop–livestock systems (ICLSs) can improve nutrient cycling and soil cover, but subsurface nutrient dynamics in organic rotations that include small ruminants remain poorly characterized. This study evaluated nitrate–N (NO₃–N) and phosphate–phosphorus (PO₄–P) concentrations in subsoil water from a certified-organic corn–soybean–goat pasture rotation in central Kentucky. Subsoil water was monitored at 1-m depth for 5 years under seven treatments arranged in a randomized complete block design: corn–soybean no-pasture treatment (CNP), ungrazed perennial pasture (PP), two crop phases (rotation corn phase [CS1] and rotation soybean phase [CS2]), and three goat-grazed pasture phases (P1 [first-year goat-grazed pasture], P2 [second-year goat-grazed pasture], and P3 [third-year goat-grazed pasture]). NO₃–N showed significant treatment, year, and treatment × year effects. Five-year mean NO₃–N ranged from 6.8 mg L⁻¹ in PP to 21.1 mg L⁻¹ in P3, with significantly high concentrations in CNP (19.4 mg L⁻¹). Peak seasonal NO₃–N concentrations occurred mainly during annual crop phases and pasture-to-crop transition periods, whereas PP and early pasture phases generally maintained lower concentrations under continuous vegetative cover. Managed goat grazing did not consistently increase NO₃–N relative to the ungrazed pasture. PO₄–P responses differed from NO₃–N, with a strong year effect but limited treatment separation. Five-year mean PO₄–P ranged from 0.6 mg L⁻¹ in CNP to 1.1 mg L⁻¹ in PP. Overall, maintaining continuous cover, minimizing soil disturbance during rotation transitions, and managing grazing to preserve vegetation may reduce nutrient movement into subsoil water in an organic ICLS.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148942454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A mobile discrimination system for improper cured tobacco leaves based on machine vision and lightweight machine–deep learning methods","authors":"Jun Bin,&nbsp;Zheng Zhou,&nbsp;Jinliang Liu,&nbsp;Mengjiao Ding,&nbsp;Yi Chen","doi":"10.1002/agj2.70525","DOIUrl":"https://doi.org/10.1002/agj2.70525","url":null,"abstract":"<p>During the tobacco (<i>Nicotiana tabacum</i> L.) leaf flue-curing process, various types of nonconforming tobacco leaves are often produced due to the influence of multiple factors, significantly reducing the usability value of cured tobacco leaves. Based on different formation mechanisms, defective tobacco leaves can be broadly categorized into eight types, including steamed slice, black-rotten leaf, waterlogged vein/slice, greenish-yellow leaf, mold-damaged leaf, over-reddened cured leaf, gray-spotted leaf, and toxicant-damaged leaf. Traditional manual sensory classification methods are highly subjective and error-prone, hindering early detection and intervention of curing defects. To improve the accuracy, convenience, and automation of defective tobacco leaves sorting, a mobile intelligent recognition application integrating machine vision with machine and deep learning technologies was developed in this study. The application adopted an innovative dual-mode architecture: an offline mode utilizing the random forest algorithm for operation without network connectivity with the classification accuracy of 85.8%, and an online mode employing the visual self-supervised pretraining (VIMER-CAE) model through cloud collaboration with 90.4% recognition accuracy. In addition to precise identification, the application combined cause analysis with preventive recommendations, requiring no specialized equipment or manual intervention beyond a smartphone. Compared to traditional empirical methods, this application replaces subjective judgment with machine vision analysis and provides process optimization guidance to reduce tobacco leaf quality losses. Its portable, high-precision, and intelligent design offers an innovative solution for curing process control, helping tobacco farmers improve production efficiency and economic benefits while promoting the standardization and intelligent development of tobacco cultivation.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148943668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Insights from a two-decade study of crop rotation and biocover impact on yield in no-till corn","authors":"Isaac Mirahki,&nbsp;Virginia R. Sykes","doi":"10.1002/agj2.70526","DOIUrl":"https://doi.org/10.1002/agj2.70526","url":null,"abstract":"<p>Crop rotation and cover crop (biocover) management can improve the productivity and yield stability of no-till corn, yet long-term evidence for humid subtropical silt loams remains limited. This study evaluated corn (<i>Zea mays</i> L.) grain yield across a 20-year period (2002–2021) under varying crop sequences and biocovers at two Tennessee locations: the Research and Education Center at Milan (RECM; Loring silt loam) and the Middle Tennessee Research and Education Center (MTREC; Maury silt loam). Treatments were applied as a split-block (strip-plot) design. The main plot consisted of repeated 4-year phases of varying crop sequences containing corn, soybean (<i>Glycine max</i> L.), and/or cotton (<i>Gossypium hirsutum</i> L.) (10 sequences at RECM and seven at MTREC). Sub-plots contained biocover treatments: Austrian winter pea (<i>Pisum sativum</i> L. <i>sativum</i> var. <i>avense</i>), hairy vetch (<i>Vicia villosa</i> L.), wheat (<i>Triticum aestivum</i> L.), poultry litter, or winter fallow. Data were analyzed using generalized linear mixed models to test management effects and their interactions with phase and location. Across phases and sites, rotation increased yield relative to continuous corn, with corn following soybean showing the most consistent performance across phases. Legume biocovers and poultry litter improved yield relative to fallow in several phases and site-specific comparisons. In contrast, wheat often reduced yield in early phases, although responses became less negative in later phases. For southeastern United States producers, prioritizing soybean-based rotations and pairing them with legume biocovers or poultry litter offers a practical pathway to improve corn yield in long-term no-till systems in the southeastern United States.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/agj2.70526","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148943637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The influence of sowing date and biostimulants foliar spray on yield and yield components of chickpea (Cicer arietinum L.)","authors":"Bita Abbasi,&nbsp;Alireza Bagheri,&nbsp;Shahab Khoshkhooy,&nbsp;Sadegh Jalilian,&nbsp;Mahmood Khoramivafa","doi":"10.1002/agj2.70528","DOIUrl":"https://doi.org/10.1002/agj2.70528","url":null,"abstract":"<p>Chickpea (<i>Cicer arietinum</i> L.) is a key rainfed crop in western Iran, whose productivity depends heavily on optimal sowing dates and biostimulant use. This study evaluated the effects of sowing date and foliar application of biostimulants on seed yield and yield components of desi chickpea. The study employed a factorial randomized complete block design with three replications across two locations in Kermanshah Province, Iran, in 2024–2025. Treatments consisted of three sowing dates (February 19, March 5, and March 21) and four foliar biostimulants applications (control, seaweed extract, amino acid, and a growth-promoting biofertilizer). The highest number of pods per plant (33.41) and the greatest 100-seed weight (17.08 g) were obtained from the March 5 sowing date at the Kermanshah site. Foliar application of seaweed extracts increased pods per plant and 100-seed weight by 87.36% and 61.41%, respectively, compared with the control at Salas-e Babajani. The maximum seed yield (1216.67 kg ha⁻¹) and aboveground biomass (2892.33 kg ha⁻¹) were achieved with the combination of the March 5 sowing date and seaweed extract application at Kermanshah, representing 3.28- and 3.64-fold increases, respectively, relative to the late sowing (March 21) control treatment at the Salas-e Babajani site. The highest seed protein content (25.70%) resulted from amino acid application at Salas-e Babajani on February 19. Seaweed extract applied on March 5 at Kermanshah also resulted in high protein content (23.76%). Optimizing sowing dates combined with these foliar biostimulants can substantially enhance chickpea yield and seed quality under western Iran's rainfed conditions.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148943670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does replacing fallow with winter crops affect corn yield and soil carbon retention in a Florida cropping system?","authors":"Julia Barra Netto-Ferreira,&nbsp;Gerrit Hoogenboom,&nbsp;Henrique Boriolo Dias,&nbsp;Thiago Berton Ferreira,&nbsp;Chris H. Wilson,&nbsp;Gabriel Maltais-Landry","doi":"10.1002/agj2.70524","DOIUrl":"https://doi.org/10.1002/agj2.70524","url":null,"abstract":"<p>Cropping systems in the southeastern United States established on sandy soils and subject to a subtropical climate with heavy rainfall often lose nutrients and carbon during winter fallow periods. While cover crops can mitigate these losses, their adoption is limited because they are not harvested for profit. Replacing winter fallows with a cash crop may offer economic and environmental benefits, but benefits may take years to emerge. Crop simulation models, such as those within the Decision Support System for Agrotechnology Transfer (DSSAT), can complement short-term field trials to assess longer term responses. This study assessed the impacts of replacing winter fallows with chickpea (<i>Cicer arietinum</i> L.) or rye (<i>Secale cereale</i> L.) on soil carbon and nitrogen in a corn (<i>Zea mays</i> L.) rotation system, using a combination of experiments and crop modeling. Two years of field data were used to calibrate and evaluate DSSAT-cropping system model (CSM) models, which were then used to simulate 10-year responses under North Florida conditions. Chickpea and rye had limited short-term effects on corn responses and soil conditions compared to fallow. Model calibration achieved acceptable accuracy across crops, despite challenges due to disease and extreme weather during the second experimental year. Soil organic carbon (SOC) simulations were moderately accurate due to low variability, while predictions of soil nitrogen dynamics were less accurate. Despite these limitations, DSSAT captured biomass and soil trends, supporting its use for long-term simulations. Ten-year simulations showed that chickpea and rye helped retain SOC but did not improve corn yields, highlighting how long-term soil benefits might be the primary advantage of integrating a winter crop in these systems.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148823187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-resolution satellite imagery and machine learning for predicting yield and nitrogen uptake in potatoes","authors":"Ravinder Singh,&nbsp;Sehijpreet Kaur,&nbsp;Lincoln Zotarelli,&nbsp;Hardeep Singh,&nbsp;Chang Zhao,&nbsp;Joel B. Harley,&nbsp;Lakesh K. Sharma","doi":"10.1002/agj2.70463","DOIUrl":"https://doi.org/10.1002/agj2.70463","url":null,"abstract":"<p>Optimizing nitrogen (N) management is crucial for maximizing potato (<i>Solanum tuberosum</i> L.) yield while minimizing environmental impacts. Achieving this requires timely, in-season estimation of crop N status and the ability to predict final tuber yield early in the season. However, commonly used approaches, including destructive sampling or UAV-based sensing, are often labor-intensive, costly, and limited in scalability. This study evaluated the utility of high-resolution SkySat satellite imagery combined with machine learning (ML) to predict final tuber yield and in-season N uptake within five nutrient management experiments in northeast Florida. Vegetation indices (VIs) derived from satellite imagery were used to train three ML algorithms: random forest (RF), support vector regression (SVR), and partial least squares regression. Under nested cross-validation, models achieved high predictive accuracy for yield (<i>R</i>² = 0.74–0.79), with RF performing well up to 77 days after planting (<i>R</i>² = 0.78, normalized root mean square error [nRMSE]% = 16.1%), and SVR enabling early-season predictions before the final N application window (<i>R</i>² = 0.76, nRMSE% = 16.5%). SkySat-derived VIs also effectively estimated mid-season N uptake (<i>R</i>² up to 0.82). SHapley Additive exPlanations analysis identified modified soil-adjusted vegetation index, soil-adjusted vegetation index, and Thiam's transformed vegetation index as key predictors. However, when evaluated under a leave-one-experiment-out framework, model performance declined substantially, with negative <i>R</i>² values, indicating limited generalization across experiments. These findings demonstrate that while high-resolution satellite imagery combined with ML can provide accurate in-season predictions within a given experimental setting, model transferability across environments remains a critical limitation. The results highlight the need for multi-environment datasets and robust modeling approaches for precision agriculture applications.</p>","PeriodicalId":7522,"journal":{"name":"Agronomy Journal","volume":"118 4","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148466893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}