Efficient agricultural water management requires estimation of root-zone soil water depletion (SWD) under heterogeneous environmental and management conditions. However, predicting SWD is challenging due to complex soil–plant–atmosphere interactions and variability across sites and seasons. This study develops a hybrid mechanistic–machine learning framework as an artificial intelligence–based approach for predicting SWD. The framework integrates soil water balance (SWB)–derived hydrologic state variables with multi-source environmental and remote sensing data across five field sites spanning a hydroclimatic gradient (330–900 mm yr−1) in Nebraska, USA. Extreme Gradient Boosting (XGB) provided the most consistent performance among the ensemble models evaluated. The framework achieved strong accuracy under in-distribution conditions (R2 ≈ 0.87–0.88; RMSE ≈ 14–16 mm) and maintained robust performance under independent evaluation (R2 ≈ 0.66–0.70; RMSE ≈ 24–26 mm). Cross-site (leave-one-site-out) analysis revealed that model generalization is governed by environmental variability and management-driven hydrologic regimes, with performance degradation associated with shifts in soil water state distribution. Feature attribution analysis showed that mechanistic hydrologic state variables dominate model predictions, while environmental and spatial predictors provide complementary information. Ensemble-based analysis indicated that predictive uncertainty is primarily driven by environmental variability rather than model instability. These findings demonstrate that reliable prediction for agricultural water management depends on both physically meaningful system representation and the diversity of environmental and management conditions captured during training to improve generalization across diverse environmental conditions. The proposed framework provides a scalable and transferable approach for irrigation decision-support systems under variable hydroclimatic conditions.

Piglet preweaning mortality (PWM) in the United States averages ≈14–15%, with sow overlaying causing about one third proportion of these losses. This research aimed to develop and evaluate deep learning models to classify six sow postures using depth images to monitor behaviors linked to overlaying risk. Top-down depth images were captured with Kinect V2® cameras at 10 frames min⁻¹ for five consecutive days (2 days before to 2 days after farrowing), yielding 26,506 training images from 18 sows, 17,901 testing images from 12 sows, and 4,697 additional images from three sows in diagonal stalls for external validation. Three transfer learning architectures (YOLOv11m-cls, ResNet-50, and Inception v3) were trained and evaluated on two different depth data transformed images (grayscale and Jet-colormap). Jet colormap images consistently outperformed grayscale, with YOLOv11m-cls achieving the highest accuracy (precision = 0.98, recall = 0.98, F1 = 0.98). External validation confirmed robust performance in diagonal stalls (F1 = 0.94), and cross-validation across stall designs and heat-lamp configurations demonstrated strong generalizability of the models. These findings show that depth imaging combined with deep learning provides a reliable, non-invasive method for sow posture classification. Such model can generate continuous behavioral data streams to improve understanding of postural transitions, stall design, and reduce piglet mortality.

Flow-proportional variable-rate fertigation (VRF) systems are used to support site-specific nitrogen (N) management under center-pivot irrigation; however, field-scale evaluations of actual N delivery under spatially variable operating conditions remain limited. This research brief presents a practical, field-based method to quantify as-applied N rates for a flow-proportional VRF system integrated with variable-rate irrigation (VRI). Fertilizer drawdown measurements, pivot flow-totalizer records, and irrigation prescriptions were combined with management-unit geospatial processing to quantify as-applied irrigation depths and N rates during an in-season fertigation event. System-level irrigation and fertilizer volume accounting was performed to ensure mass-balance closure. As-applied N rates were derived assuming uniform fertilizer concentration under flow-proportional injection and were compared against prescribed rates at the management-plot scale. Strong spatial agreement was observed between prescribed rates (1–27 kg-N/ha) and as-applied N rates (r = 0.993). Root mean square error (RMSE) and mean absolute error (MAE) were 2.52 and 2.41 kg-N/ ha, respectively. A consistent positive bias was observed, attributable to irrigation hydraulic variability and small mismatches between the irrigation-flow meter displayed on the pivot control panel and the flow sensor supplying input to the fertigation controller. System-level volume closure was achieved (R ≈ 1.01), supporting the validity of the quantification approach. Outcome-based as-applied evaluation provides a practical and reproducible means of verifying VRF performance beyond assumed prescription execution. The proposed workflow enables reliable field-scale quantification of N delivery accuracy and supports improved agronomic record keeping and performance assessment in operational VRF–VRI systems without requiring catch-can measurements or nozzle-scale instrumentation.

Abstract Background Efficient monitoring of sow nursing behavior is relevant to animal welfare monitoring and management of preweaning mortality (PWM) risk, which remains a significant challenge in intensive swine production systems. Traditional observation methods are labor-intensive and inadequate for large-scale farms, requiring automated solutions. Precision livestock farming (PLF) technologies, particularly computer vision and deep learning models, offer opportunities for continuous monitoring of sow posture–nursing-state classes. The YOLO (you only look once) architecture has demonstrated high accuracy and speed for object detection in different environments. This study developed and evaluated a modified YOLO11n model optimized with TensorRT to monitor sow posture–nursing-state classes in farrowing crates. In this study, “nursing” was operationalized from top-view images as visible piglet snout/mouth contact oriented toward the udder/teat line; frames without visually confirmable contact were labeled not nursing. The model’s lightweight architecture and inference acceleration address the computational constraints of real-time applications in farrowing-room monitoring. Results The modified YOLO11n achieved an mAP@50 of 98.90%, while reducing complexity to 207 layers and 5.0 GFLOPs by removing the small-object detection head. This enabled inference times of 4.6 ms on an NVIDIA A100 GPU and 6.1 ms on a T4 GPU with TensorRT optimization. The framework detected and classified ten posture–nursing-state classes (sitting, standing, and three lying postures, each labeled as nursing or not nursing) on representative test images, including frames with partial occlusion and variable lighting. Misclassifications were observed primarily among visually similar classes, such as Sow_Sitting_Nursing and Sow_Sitting_Not_Nursing, consistent with limited visibility of piglet-to-teat contact in some postures from the top-view angle. TensorRT optimization further reduced inference latency by 58.56% (A100) and 44.55% (T4), demonstrating consistent inference-latency reductions on the evaluated GPU platforms. Conclusions This study demonstrates the potential of leveraging lightweight deep learning architectures for real-time behavioral monitoring in PLF. The modified YOLO11n and TensorRT optimization provide an efficient framework for automated monitoring of sow posture–nursing-state classes, supporting welfare-oriented monitoring workflows relevant to PWM risk management. Future work will evaluate lightweight temporal post-processing and embedded deployment to reduce frame-to-frame label flicker and assess performance under on-farm compute and I/O constraints.

Improving nitrogen use efficiency (NUE) in commercial maize production remains a persistent challenge. A major barrier is the lack of simple, remote sensing-based decision-support frameworks that enable broad adoption of in-season, site-specific nitrogen (N) management. This study developed and evaluated a practical framework that contextualizes the Holland-Schepers sensor algorithm using PlanetScope (PS) satellite imagery to guide multiple in-season, variable-rate fertigation delivered through a flow-proportional injection system integrated with a center pivot system equipped with variable-rate irrigation. Field implementation was carried out during the 2023 and 2024 seasons across four N rates (0-N, Low-N, Fertigation, Full-N) and three irrigation treatments: full (BMP), deficit (50 %BMP), and rainfed. Normalized Difference Red Edge (NDRE)-derived sufficiency index (SI) values informed the amount, timing, and spatial distribution of N applications. In 2024, PSguided fertigation achieved yields statistically comparable to Full-N while reducing total N input by 23 %. Significant improvements in NUE were observed, with Fertigation outperforming Full-N by 12 % in agronomic efficiency (AE) and 26 % in partial factor productivity of N (PFPN). Satellite-derived SI values were strongly correlated with UAV benchmarks from the MicaSense Altum and RedEdge-3 sensors (rho = 0.82-0.95). However, PS consistently overestimated NDRE relative to UAV data, particularly under N-deficient conditions, underscoring the need for local calibration and bias correction. To improve diagnostic specificity, a biologically informed, rule-based stress-classification framework was developed to differentiate nitrogen stress from water stress using NDRE and soil water depletion (SWD) as diagnostic variables. Retrospective yield and management data were used to establish physiologically meaningful NDRE-SWD thresholds for stress diagnosis during the critical in-season fertigation window (V10-R2). Full-Yield plots achieved approximately 12,000 kg/ha at NDRE = 0.78 and SWD = 64 mm. The resulting NDRE-SWD-yield patterns highlight the feasibility of disentangling stress types under commercial field conditions. However, further validation across seasons and environments, along with integration of canopy water or temperature indices, is needed to improve water-stress detection and enable real-time decision-making. Overall, these results offer actionable guidance for implementing satelliteguided fertigation at commercial scale. The developed framework delivers scalable, data-driven N recommendations that enhance profitability and support environmentally responsible maize production. It also provides a reference for future research and extension programs aiming to turn satellite remote sensing into practical tools for site-specific N management.

Highlights The performance of RTK-GNSS-enabled drones without GCPs was assessed in agricultural applications. A centimeter-level positioning accuracy was achieved when drones were functioning with RTK-GNSS-enabled. RTK-GNSS-enabled drones without GCPs performed competitively with GCP-based methods in plant height estimation. Using RTK-enabled drones without GCPs provides a time- and labor-saving solution for agricultural practices. ABSTRACT. Commercial drones equipped with Real-Time Kinematic GNSS (RTK-GNSS) technology have been available for several years and are now a standard feature in modern models. This technology has the potential to eliminate the need for ground control points (GCPs). However, despite its widespread adoption and the positioning accuracy claimed in product specifications, limited literature exists on quantifying and systematically evaluating its performance in agricultural field settings for specific applications, such as measuring canopy height profiles. This study aimed to evaluate the potential of RTK-GNSS-enabled drones to estimate plant height without GCPs by comparing it to a benchmark method that uses GCPs across three crops: triticale, maize, and soybean. Three methods were compared: (A) regular-GNSS with GCPs, (B) RTK-GNSS with GCPs, and (C) RTK-GNSS without GCPs. Prior to evaluating the performance in height estimation, the positioning accuracy of RTK-GNSS-enabled drones was assessed, and results confirmed the centimeter-level positioning accuracy achieved by drones with RTK-GNSS-enabled. For plant height estimation, Method-C (RTK-GNSS without GCPs) performed competitively with GCP-based methods, achieving R 2 values from 0.84 to 0.99 for drone-estimated versus manually measured plant heights for the three crops, with a mean RMSE of 12.15 cm (compared to Method-A R 2 of 0.86 to 0.99 with a mean RMSE of 11.86 cm, and Method-B R 2 of 0.84 to 0.97 with a mean RMSE of 11.49 cm). Results suggest that RTK-GNSS-enabled drones can reliably estimate plant height without GCPs, offering a time- and labor-saving solution for field measurements in breeding research and agricultural production. Keywords: Keywords.,Global positioning system (GPS), Ground control point (GCP), High-throughput plant phenotyping, Plant height, Structure-from-motion, Unmanned aerial vehicle (UAV).

Multiple herbicide‐resistant (MHR) Palmer amaranth has been ranked as the most problematic weed in row crop fields in Nebraska. Integration of narrow row spacing with herbicide might augment control of MHR Palmer amaranth. The objectives of this study were to evaluate an integrated effect of row spacing and herbicide programs for MHR Palmer amaranth control, density, seed production, corn injury, photosynthetically active radiation (PAR) interception, and grain yield in glyphosate/glufosinate‐resistant corn. Field experiments were conducted in 2020 and 2021 in a grower's field infested with acetolactate synthase‐inhibitor/atrazine/glyphosate‐resistant Palmer amaranth near Carleton, NE. Herbicide‐by‐row spacing interactions were significant for most variables. Flufenacet/isoxaflutole/thiencarbazone‐methyl followed by (fb) glufosinate, acetochlor/mesotrione applied pre‐emergence (PRE) or fb glufosinate, acetochlor/clopyralid/flumetsulam fb glufosinate, and glufosinate fb dicamba/tembotrione controlled Palmer amaranth ≥90% at 90 days after late post‐emergence (DALPOST) herbicide application. Glufosinate fb dicamba/tembotrione and acetochlor/mesotrione fb glufosinate with 38‐ and 76‐cm row spacing, flufenacet/isoxaflutole/thiencarbazone‐methyl fb glufosinate, and acetochlor/clopyralid/flumetsulam fb glufosinate with 38‐cm row spacing recorded no Palmer amaranth 30 DALPOST. Herbicide programs with narrow row spacing had numerically higher PAR interception compared with 76‐cm row spacing 30 days after early post‐emergence (DAEPOST) and 15 DALPOST. Higher corn yield (13,222–13,596 kg ha −1 ) was obtained with acetochlor/clopyralid/flumetsulam fb glufosinate with 38‐cm row spacing. Palmer amaranth seed production was zero with acetochlor/mesotrione fb glufosinate, and glufosinate fb dicamba/tembotrione; flufenacet/isoxaflutole/thiencarbazone‐methyl fb glufosinate in 76‐cm row spacing; and acetochlor/clopyralid/flumetsulam fb glufosinate in 38‐cm row spacing.

With the resurgence of on-the-go site-specific weed control technologies, research has increasingly focused on real-time weed identification using computer vision and machine learning. While Convolutional Neural Networks (CNNs) have been successfully used for automatic weed detection, many studies prioritize accuracy over inference speed and computational resource demand. The objective of this work was to develop, deploy, and evaluate CNN-based computer vision models on computationally resource-restrained edge devices for real-time weed classification and segmentation. RGB images were collected by UAVs over corn and soybean fields infested with Palmer Amaranth (Amaranthus palmeri). A few object detection (OD) and semantic segmentation (SS) models were modified, if necessary, to be lightweight, then trained and deployed on edge devices for real-time weed detection with resized input images in dimensions of 256 x 256 pixels. Specifically, the original backbones of Single-Shot Detector (SSD) and DeepLabv3+ models were replaced by a pre-trained MobileNetV3 backbone and deployed on a Jetson Nano. A YOLOv8n model and a proposed MobileNetV4-Seg model were trained and deployed on a Jetson Orin Nano. All models achieved real-time performance with acceptable accuracies with the data collected in corn and soybean fields. Among them, the MobilenetV4-Seg achieved the best performance, with IoUs of 69.9 % and 76.8 %, F1 scores of 82.3 % and 86.9 %, for the corn and soybean datasets, respectively, and 44 FPS on the Jetson Orin Nano. Overall, this study demonstrated a use case and feasible workflow for developing and deploying lightweight CNN models on resource-constrained edge devices for on-the-go real-time site-specific applications.

Aboveground plant biomass production can exhibit varied responses to within- and across-season weather variability. Using a long-term data set (2007-2021) collected annually from grazing exclosures in a Sandhills grassland in Nebraska, USA, our part-1 paper reported 15 yr of changes in grassland plant production. In this paper, we modeled seasonal weather impacts on total plant biomass and biomass of three plant functional groups using stepwise (forward) linear regression. Biomass data were measured for the following three periods each year: early season (April to midJune), late season (midJune to midAugust), and full season (April to midAugust). Weather variables, derived from precipitation and temperature, were categorized into amount, index, and pattern variables. The temporal variability of each weather variable was quantified across four time periods: three within-season conditions (early season, late-season, and full-season periods of the current year) and an across-season condition (the full-season period of the previous year). The results indicated that plant production responses varied among functional groups and across time periods. Late-season C-4-grass production significantly increased under wetter summer conditions (P < 0.05). A dry growing season in the previous year tended to decrease subsequent-year early season C-3-grass production (P < 0.05), and a warmer growing season in the previous year was likely to enhance subsequent-year forb production (P < 0.001). Total plant production exhibited more complex seasonal patterns, primarily driven by the differences in individual plant functional groups during specific growing periods. This complexity reflects the collective responses of mixed plant functional groups to weather variability. Understanding these complex relationships is fundamental to predicting grassland plant production as well as developing and implementing appropriate grazing strategies that adapt to changing climate variability. This study will ultimately support the enhancement of ecological sustainability and resilience of the Sandhills semiarid grassland ecosystem. (c) 2025 The Authors. Published by Elsevier Inc. on behalf of The Society for Range Management. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ )

Conventional soil water balance (SWB) irrigation scheduling tools, such as FAO-56-based Spreadsheets and the Spatial Evapotranspiration Modeling Interface (SETMI), rely heavily on manual inputs and periodic field measurements, leading to delayed recommendations and missed opportunities to prevent crop stress. More critically, these tools lack the computational scalability and adaptability to leverage the high-frequency, high-volume datasets now available through modern sensing technologies. As precision irrigation increasingly depends on integrating spatially and temporally nuanced field information, there is a pressing need for decision-support systems that can process Big Data efficiently and respond in real-time. To overcome these limitations, we developed and validated a machine learning (ML) framework for near real-time prediction of soil water depletion (SWD) and site-specific irrigation recommendations in maize and soybean production systems. Unlike prior models, our approach integrates multi-source, multi-year (2020 and 2023) datasets, including remote sensing data, weather variables, soil properties, management practices, yield records, time-related features, and geospatial information, to train Decision Tree, Random Forest, Gradient Boosting, and Extreme Gradient Boosting (XGB) models. Feature selection combined agronomic domain knowledge, correlation analysis, and Random Forest feature importance to retain relevant predictors while minimizing model complexity. The SWD was typically between 0 mm (field capacity) and 110 mm (management allowable depletion with a dynamic root zone increasing up to 1000 mm for the second half of the season) for irrigated plots, and up to 180 mm in rainfed conditions. Among the models, XGB performed best in the 2024 independent validation, predicting SWD with high accuracy (maize: R2 = 0.72, RMSE = 22 mm; soybean: R2 = 0.78, RMSE = 24 mm). The average predicted SWD values were 53 mm (maize) and 54 mm (soybean), closely matching SWB Spreadsheet estimates (47 mm and 54 mm, respectively). Field deployment in 2024 demonstrated the model's operational potential, with ML-generated irrigation recommendations (62-70 mm for maize; 73-83 mm for soybean) closely aligning with FAO-56 Spreadsheet (61 mm maize; 79 mm soybean) and SETMI (64-96 mm soybean) benchmarks. However, testing on independent 2021 data revealed reduced generalization performance, highlighting the need for more diverse training datasets. Overall, this study advances a practical, scalable, ML-driven decision support framework for real-time precision irrigation in commercial cropping systems.

Weed management is always a challenge in crop production, exacerbated by the issue of herbicide resistance. Excessive herbicide application not only leads to the development of herbicide resistance weeds but also causes environmental problems. In precision agriculture, innovative weed management methods, especially advanced remote sensing and computer vision technologies for targeted herbicide applications, i.e., site-specific weed management (SSWM), have recently drawn a lot of attention. Challenges exist in accurately and reliably detecting diverse weed species under varying field conditions. Significant efforts have been made to advance computer vision technologies for weed detection. This comprehensive review provides an in-depth examination of various methodologies used in developing weed detection systems. These methodologies encompass a spectrum ranging from traditional image processing techniques to state-of-the-art machine and deep learning models. The review further discusses the potential of these methods for real-time applications, highlighting recent innovations, and identifying future research hotspots in SSWM. These advancements hold great promise for further enhancing and innovating weed management practices in precision agriculture.

Efficient water management is vital for sustainable agriculture, yet integrating real-time data for precise irrigation remains a challenge. This study designed the Crop2Cloud (C2C) platform, a system that leverages advanced sensors using Internet of Things (IoT), edge and cloud computing techniques, and computed Water Stress Indices (WSIs) and machine learning models (i.e., fuzzy logic), to provide scalable and real-time irrigation decisions. The C2C platform aggregates several data including Volumetric Water Content (VWC) from TDR sensors (Acclima Inc., US) installed at four multiple depths, canopy temperatures (Tc) measured by Infrared Radiometers (IRTs) (Apogee Instruments, US), as well as weather information and estimated Crop Evapotranspiration (ETc) from FAO56 approach. Computed WSIs included the theoretical Crop Water Stress Index (CWSI) and Soil Water Stress Index (SWSI) as a ratio of Volumetric Water Content (VWC), measured and that at Field Capacity (FC) and Maximum Allowable Depletion (MAD). Additionally, fuzzy-logic irrigation schedule was developed using different fuzzy rules and three available water use indicators – CWSI, SWSI, and ETc. A designed dashboard can display collected data, computed WSIs, and irrigation recommendations from selected methods: only CWSI, only SWSI, combining SWSI + CWSI, and fuzzy logic. The C2C platform can provide quick and real-time crop performance insights and data-driven decisions for timely water application. However, there are logistical challenges such as sensor damage and power management which impact the platform’s performance and efficiency. Future work will involve refining the system to avoid data gaps and improving scheduling methods to optimize irrigation applications to increase water and energy savings.

Accurate interpretation of soil moisture patterns is critical for irrigation scheduling and crop management, yet existing approaches for soil moisture time-series analysis either rely on threshold-based rules or data-hungry machine learning or deep learning models that are limited in adaptability and interpretability. In this study, we introduce SPADE (Soil moisture Pattern and Anomaly DEtection), an integrated framework that leverages large language models (LLMs) to jointly detect irrigation patterns and anomalies in soil moisture time-series data. SPADE utilizes ChatGPT-4.1 for its advanced reasoning and instruction-following capabilities, enabling zero-shot analysis without requiring task-specific annotation or fine-tuning. By converting time-series data into a textual representation and designing domain-informed prompt templates, SPADE identifies irrigation events, estimates net irrigation gains, detects, classifies anomalies, and produces structured, interpretable reports. Experiments were conducted on real-world soil moisture sensor data from commercial and experimental farms cultivating multiple crops across the United States. Results demonstrate that SPADE outperforms the existing method in anomaly detection, achieving higher recall and F1 scores and accurately classifying anomaly types. Furthermore, SPADE achieved high precision and recall in detecting irrigation events, indicating its strong capability to capture irrigation patterns accurately. SPADE's reports provide interpretability and usability of soil moisture analytics. This study highlights the potential of LLMs as scalable, adaptable tools for precision agriculture, which is capable of integrating qualitative knowledge and data-driven reasoning to produce actionable insights for accurate soil moisture monitoring and improved irrigation scheduling from soil moisture time-series data.

Commercial drones equipped with Real-Time Kinematic GNSS (RTK-GNSS) technology have been available for several years and are now a standard feature in modern models. This technology has the potential to eliminate the need for ground control points (GCPs). However, despite its widespread adoption and the positioning accuracy claimed in product specifications, limited literature exists on quantifying and systematically evaluating its performance in agricultural field settings for specific applications, such as measuring canopy height profiles. This study aimed to evaluate the potential of RTK-GNSS-enabled drones to estimate plant height without GCPs by comparing it to a benchmark method that uses GCPs across three crops: triticale, maize, and soybean. Three methods were compared: (A) regular-GNSS with GCPs, (B) RTK-GNSS with GCPs, and (C) RTK-GNSS without GCPs. Prior to evaluating the performance in height estimation, the positioning accuracy ofRTK-GNSS-enabled drones was assessed, and results confirmed the centimeter-level positioning accuracy achieved by drones with RTK-GNSS-enabled. For plant height estimation, Method-C (RTK-GNSS without GCPs) performed competitively with GCP-based methods, achieving R2 values from 0.84 to 0.99 for drone-estimated versus manually measured plant heights for the three crops, with a mean RMSE of 12.15 cm (compared to Method-A R2 of 0.86 to 0.99 with a mean RMSE of 11.86 cm, and Method-B R2 of 0.84 to 0.97 with a mean RMSE of 11.49 cm). Results suggest that RTK-GNSS-enabled drones can reliably estimate plant height without GCPs, offering a time-and labor-saving solution for field measurements in breeding research and agricultural production.

Agriculture serves as both a source and a sink of global greenhouse gases (GHGs), with agricultural intensification continuing to contribute to GHG emissions. Climate-smart agriculture, encompassing both nature- and technology-based actions, offers promising solutions to mitigate GHG emissions. We synthesized global data, between 1990 and 2021, from the Food and Agriculture Organization (FAO) of the United Nations to analyze the impacts of agricultural activities on global GHG emissions from agricultural land, using structural equation modeling. We then obtained predictive estimates of agricultural GHG emissions for the future period of 2022-2050 using deep-learning models. The FAO data show that, from 1990 to 2021, global livestock numbers, inorganic nitrogen (N) fertilizer use, crop residue, and irrigation area increased by 27%, 47%, 49%, and 37%, respectively. The increased livestock numbers contributed to the increases in CH4 and N2O emissions, while inorganic N fertilizer, crop residue, and irrigation mainly contributed to the increases in N2O emissions. Emissions of CO2 decreased because of a 29% reduction in net forest loss. As a result of the reduced deforestation emissions, the overall agricultural GHG emissions declined from 11.50 to 10.89 GtCO2eq from 1990 to 2021 despite the increases in livestock numbers, inorganic N fertilizer, crop residue, and irrigation. Looking ahead, our model predicts that if current agricultural trends persist, GHG emissions will rise to 11.82 ± 0.07 GtCO2eq in 2050. However, maintaining agricultural GHG emissions at the 2021 level through 2050 is possible if the rate of reduction in net forest loss is doubled. Furthermore, if the rate is tripled, agricultural GHG emissions can be limited to 9.85 ± 0.07 GtCO2eq in 2050. Our findings suggest that reductions in agricultural GHG emissions, alongside sustainable agricultural intensification and climate-smart agricultural practices, can be achieved through parallel efforts emphasizing accelerated forest conservation.

Weed management is always a challenge in crop production, exacerbated by the issue of herbicide resistance. Excessive herbicide application not only leads to the development of herbicide resistance weeds but also causes environmental problems. In precision agriculture, innovative weed management methods, especially advanced remote sensing and computer vision technologies for targeted herbicide applications, i.e., site-specific weed management (SSWM), have recently drawn a lot of attention. Challenges exist in accurately and reliably detecting diverse weed species under varying field conditions. Significant efforts have been made to advance computer vision technologies for weed detection. This comprehensive review provides an in-depth examination of various methodologies used in developing weed detection systems. These methodologies encompass a spectrum ranging from traditional image processing techniques to state-of-the-art machine and deep learning models. The review further discusses the potential of these methods for real-time applications, highlighting recent innovations, and identifying future research hotspots in SSWM. These advancements hold great promise for further enhancing and innovating weed management practices in precision agriculture.

Forage biomass exhibits temporal and spatial variability driven by microenvironmental conditions and diverse management practices in pastures. Common forage sampling strategies, such as random or transect-based sampling, often overlook this variability, potentially resulting in unrepresentative samples and inaccurate estimates. This study investigated the potential of a forage biomass estimation tool based on unmanned aerial vehicles (UAV) remote sensing for assessing spatial and temporal effects of various pasture management practices. Using UAV-derived forage dry matter (DM) maps collected over two years from for managed Bromus inermis dominated pastures in eastern Nebraska, we evaluated the impacts of grazing and fertilization management practices on biomass. Forage DM model was developed using vegetation indices derived from UAV multispectral data, cumulative precipitation and growing degree days, ground-sampled biomass data, and Random Forest machine learning (ML) algorithm. Result showed that total DM was estimated with R-2 of 0.76 and RMSE of 1068 kg.ha(-1) (rRMSE = 33.1 %). Significant DM differences were observed for three management treatments-control, feed supplement with Dry Distillers Grains Solubles (DDGS), and pasture nitrogen fertilization-with and without grazing across 2021 and 2022 growing seasons using the UAV estimated DMs (p < 0.001). In addition, we also suggest conducting the UAV flight and data processing prior to ground sampling to guide the sampling locations, if both ground biomass samples and UAV remote sensing data are intended to be collected, to better capture spatial variability across the pasture.

Semiarid grasslands of the Nebraska Sandhills provide critical ecosystem services and are an important forage resource for the local cattle industry. Over the past decades, warming and climate-related extremes have affected grassland production worldwide, which promotes the initiation of numerous grassland monitoring projects. Despite this, production trends for plant functional groups in the Sandhills regions in recent years have remained unknown. In this study, we analyzed plant biomass production of the Sandhills grasslands with a dataset collected over 15 yr from 2007 to 2021. Ungrazed total biomass and biomass of individual plant functional groups were assessed in grazing exclosures twice a year, in mid-June (for early season) and mid-August (for late season). This first paper reports our findings on total biomass and compositional changes of the three major plant functional groups, as well as trends in precipitation and temperature during the study period. A significant increasing trend ( P < 0.05) was observed in temperature over time during the early season (April to mid-June), with a weak monotonic increasing trend ( P = 0.07) during the full season (April to mid-August), whereas no significant pattern was reported for precipitation during the study, although it displayed complex within- and across-season patterns. The proportion of C3 -grass biomass in total biomass increased ( P < 0.05), while the proportion of C4 -grass biomass decreased ( P < 0.01). We did not observe any significant trends for forbs; however, the drought of 2012 resulted in up to a fivefold increase in the proportion of forb biomass the following year. These findings enhance our understanding of current patterns in grassland production and contribute to regional evidence on the response of plant functional groups to variability and extremes in intra-annual weather variables, which can improve our capability to perform adaptive grazing management in a similar semiarid grassland ecosystem. (c) 2024 The Authors. Published by Elsevier Inc.

Multiple herbicide-resistant (MHR) Palmer amaranth is among the most problematic summer annual broadleaf weeds in Nebraska and several other states. A new MHR corn cultivar (resistant to 2,4-D/glufosinate/glyphosate, also known as Enlist corn) has been commercially available in the United States since 2018. Growers are searching for herbicide programs for control and reduce seed production of MHR Palmer amaranth among Enlist corn crops. The objectives of this study were to evaluate herbicides applied preemergence, early postemergence, or preemergence followed by (fb) late postemergence for the management of MHR Palmer amaranth in Enlist corn fields and to assess their effect on Palmer amaranth biomass, density, seed production, and corn yield. Field experiments were conducted near Carleton, NE, in 2020 and 2021, in a grower's field of Enlist corn infested with acetolactate synthase-inhibitor/atrazine/glyphosate-resistant Palmer amaranth. Herbicides applied preemergence, such as flufenacet/isoxaflutole/thiencarbazone-methyl, acetochlor/clopyralid/flumetsulam, or acetochlor/clopyralid/mesotrione, provided 75% to 99% control of Palmer amaranth 30 d after preemergence. Preemergence fb late postemergence herbicides resulted in 94% Palmer amaranth control 90 d after late postemergence, reduced weed density to 0 to 8 plants m(-2) 30 d after late postemergence, and reduced biomass to 2 to 14 g m(-2) 15 d after late postemergence compared to preemergence-only (59% control, 0 to 15 plants m(-2), and 4 to 123 g m(-2)) and early postemergence-only herbicides (78% control, 6 to 30 plants m(-2), and 8 to 25 g m(-2)). Based on contrast analysis, Palmer amaranth seed production was reduced to 14,050 seeds m(-2) in preemergence fb late postemergence herbicide programs compared with 325,490 seed m(-2 )in preemergence-only and 376,750 seed m(-2) in early postemergence-only programs. Based on orthogonal contrast, higher corn yield of 12,340 and 11,730 kg ha(-1) was obtained with preemergence fb late postemergence herbicide programs compared with preemergence-only (10,840 and 11,510 kg ha(-1)) and early postemergence-only programs (10,850 and 10,030 kg ha(-1)) in 2020 and 2021, respectively.