US Corn Belt Research Articles

Sensitivity changes of US maize yields to extreme heat through timely precipitation patterns

Warm temperatures due to increases of greenhouse gas emissions have changed temperature distribution patterns especially for their extremes, which negatively affect crop yields. However, the assessment of these negative impacts remains unclear when surface precipitation patterns are shifted. Using a statistical model along with 23,944 county-year maize-yield data during 1981–2020 in the US Corn Belt, we found that the occurrence of timely precipitation reduced the sensitivity of maize yields to extreme heat by an average of 20% during the growing season with variations across phenological periods. Spatially across the US corn belt, maize in the northern region exhibited more significant benefits from timely precipitation compared to the southern region, despite the pronounced negative effects of extreme heat on yields in cooler regions. This study underscores the necessity of incorporating timely precipitation as a pivotal factor in estimating heat effects under evolving climates, offering valuable insights into complex climate-related challenges.

Using gross primary production data and deep transfer learning for crop yield prediction in the US Corn Belt

Estimating crop yield prior to harvest plays a crucial role in agricultural decision making. Machine learning models that use remote sensing (RS) data can provide quick and early estimates of crop yield. Previously, timeseries gross primary production (GPP) synthesized from RS data has been employed in agricultural applications. However, relevant research on using GPP data and deep learning models in crop yield prediction is scarce. The aim of this study was to assess the capability of MODIS-derived GPP data and deep transfer learning to predict crop yield. We developed and trained deep neural network (DNN), one-dimensional convolutional neural network (1D-CNN), and gated recurrent unit (GRU) models to predict county-level corn and soybean yields. Then we evaluated the performance of crop-type and location transfer learning strategies in crop yield prediction. The results of this study can be summarized as follows: (1) GPP data and deep transfer learning have the potential to predict crop yield with reasonable prediction accuracy; (2) in the context of transfer learning, crop-type transfer learning produced a coefficient of determination (R2) ranging from 0.521 to 0.784 for soybean and from 0.644 to 0.903 for corn. Meanwhile, location transfer learning yielded R2 values ranging from 0.690 to 0.931 for the Great Plains (GP) climatic zone and 0.480 to 0.801 for the Eastern Temperate Forest (ETF) climatic zone; and (3) For corn and soybean yield predictions, the 1D-CNN model (R2 = 0.864 for corn and 0.750 for soybean) outperformed the DNN and GRU models. Our results reveal that MODIS-derived GPP data and deep transfer learning can be used to effectively predict county-level crop yield.

Interviews with farmers from the US corn belt highlight opportunity for improved decision support systems and continued structural barriers to farmland diversification

Diversifying high-input, monocropped landscapes like the US Corn Belt would provide both economic and ecosystem service benefits to the agricultural landscape. Decision support systems (DSS) and digital agriculture could help farmers decide if diversification is suitable for their operation. However, adoption of DSS by farmers remains low, likely due to lack of farmer engagement before and during the DSS development process. This study aimed to better understand the tasks, tools, and people involved in implementing farmland diversification with the goal to inform design of agricultural DSS. Semi-structured interviews were conducted with 11 farmers who had diversified their corn/soybean cropland with government-supported conservation programs (e.g., CRP, wetlands) and alternative crops (e.g., small grains, pasture) in the past four years. Interview data was transcribed and then analyzed using affinity diagramming. Results show farmers needed DSS to layer multiple sources of data and observations over several years to identify field productivity trends and drivers; spatial orientation of practices to fit management and field constraints; matching operation goals to alternative practices; financial planning and market exploration; and information on promising emerging practices like subsidized pollinator habitat. However, the interviews also highlighted structural barriers to diversification that DSS cannot or can only partially address. These included social pressures; market access; crop insurance policy; and quality of relationships with governmental agencies. Results indicate better DSS design can empower individual farmers to diversify cropland, but structural interventions will be needed to successfully diversify the agricultural landscape and support economic and ecosystem health.

Techno‐economic and life cycle analysis of renewable natural gas derived from anaerobic digestion of grassy biomass: A US Corn Belt watershed case study

AbstractRestoring native grassland vegetation can substantially improve ecosystem service outcomes from agricultural watersheds, but profitable pathways are needed to incentivize conversion from conventional crops. Given growing demand for renewable energy, using grassy biomass to produce biofuels provides a potential solution. We assessed the techno‐economic feasibility and life cycle outcomes of a “grass‐to‐gas” pathway that includes harvesting grassy (lignocellulosic) biomass for renewable natural gas (RNG) production through anaerobic digestion (AD), expanding on previous research that quantified ecosystem service and landowner financial outcomes of simulated grassland restoration in the Grand River Basin of Iowa and Missouri, United States. We found that the amount of RNG produced through AD of grassy biomass ranged 0.12–45.04 million gigajoules (GJ), and the net present value (NPV) of the RNG ranged −$97 to $422 million, depending on the combination of land use, productivity, and environmental credit scenarios. Positive NPVs are achieved with environmental credits for replacement of synthetic agricultural inputs with digestate and clean fuel production (e.g., USEPA D3 Renewable Identification Number, California Low Carbon Fuel Standard). Producing RNG from grassy biomass emits 15.1 g CO2‐eq/MJ, which compares favorably to the fossil natural gas value of 61.1 g CO2‐eq/MJ and exceeds the US Environmental Protection Agency's requirement for cellulosic biofuel. Overall, this study demonstrates opportunities and limitations to using grassy biomass from restored grasslands for sustainable RNG production.

Plot-level satellite imagery can substitute for UAVs in assessing maize phenotypes across multistate field trials.

Abstract Accurate genotype-specific early yield estimates at fields and plots offer potential benefits to farmers in optimizing their agronomic practices, breeders in screening thousands of varieties, and policymakers in decision-making contributing to the improvement of agriculture and food production systems. Effective approaches to track plant growth and predict yield require large datasets of remote sensing and ground truth data collected across multiple environments. Low-altitude drone flights are increasingly being used to collect data from field evaluations of new crop varieties, while satellite imagery is being explored to track yield and management practices at the regional scales. Despite their lower spatial resolution, satellite platforms exhibit logistical and technical advantages in scalability and accessibility, and could facilitate plot-level predictions, especially with steadily improving spatial resolution. However, genotype-specific, plot-level, high-resolution satellite images from multiple environments with ground truth measurements are not yet publicly available. Here we generated, described, and evaluated over 20,000 plot-level images of over 80 hybrid maize varieties grown across the US corn belt under various management practices collected from (near simultaneous) satellite and drone flights integrated with ground truth yield measurement. Of the six baseline models examined, models employing data collected from satellite images often matched or exceeded the performance of models employing drone images for both within and cross-environment yield prediction. Large, multi-environment, genetically diverse datasets such as those generated in this study, along with more complex models could help unlock the power of satellite imagery as an important addition to the tool of farmers, plant geneticists, breeders, and policymakers.

Further adoption of conservation tillage can increase maize yields in the western US Corn Belt

Conservation tillage can reduce soil erosion, increase soil health, and decrease labor and fuel input costs. Despite these benefits, potential yield impacts remain an important concern for farmers considering adoption. Previous research suggests that conservation tillage is likely to have the largest yield benefits in more arid conditions, but a lack of field-level analyses across climatic, management and soil conditions limits confidence in such predictions. Satellite imagery provides the opportunity to monitor agricultural lands at sub-field resolution across large spatial scales and wide environmental gradients. Here we investigate the maize yield impacts of conservation tillage in the semi-arid western US Corn Belt, using sub-field resolution datasets on tillage practices and crop yields derived from satellite data spanning four states (Nebraska, Kansas, South Dakota, and North Dakota) between 2008 and 2020. On these datasets, we estimate heterogenous yield outcomes for several thousand maize fields across gradients in climate, soil quality and irrigation status by using a causal forests analysis, an adaptation of the random forests machine-learning algorithm for causal inference on observational data. We find that long-term adoption of conservation tillage increased rainfed maize yields by an average of 9.9% in the region. Impacts on irrigated yields were small and not statistically significant. These results, along with an analysis of variables related to greater than average yield benefits, indicate that improved water infiltration and retention are the primary reasons for conservation tillage benefits. Despite yield benefits, many fields estimated to see increased yields under long term low till have not adopted the practice. Therefore, we identify specific counties likely to benefit most from increased levels of adoption. Our results strengthen the understanding of the impacts of conservation agriculture on crop yields and help define environments and counties most likely to benefit from conservation tillage.

Early forecasting of corn yield and price variations using satellite vegetation products

Several approaches based on process-based, statistical or economic models have been developed and are currently used by market analysts and governmental agencies to forecast agricultural yields and commodity prices at the national and global scales. However, these models require input data that are often difficult to obtain before crop harvest and their predictions are not always reliable. Satellite data products provide easily accessible sources of information on vegetation characteristics, potentially relevant for predicting crop yield and commodity prices at a large scale, before harvest. This study presents and evaluates several methods for forecasting variations in US corn yield and global corn prices using regional vegetation satellite products. The spatial dimension of maps of vegetation indices derived from satellite products in the US Corn Belt is first reduced either by their spatial means or by principal components derived using the Empirical Orthogonal Function (EOF) method. Mean values and EOF principal components are then used as explanatory variables in generalized LASSO linear models to forecast corn price and yield changes at different time horizons, before harvest. A cross-validation method was implemented to assess the model performances using Area under the ROC curve (AUC), the Brier Skill Score (BSS) and the Matthew Correlation Coefficient (MCC). The results revealed that EOF principal components derived from satellite-based gross primary production (GPP) in July could be used to predict variations in US corn yields, with an AUC of 0.97 (95% CI 0.92-1), a BSS of 0.75 and an MCC of 0.83. The results also showed that the July spatial averages of satellite GPP data obtained in the US Corn Belt could be used to forecast the forthcoming increase or decrease of global corn prices as evidenced by an AUC reaching up to 0.91, a BSS up to 0.5, and an MCC up to 0.67. Finally, we compared our forecasts of yield and price variations to those derived from the WASDE forecasting method currently implemented in the US. This comparison showed that our approach was at least as accurate and often more accurate than WASDE. This study demonstrates how satellite data can be used to forecast variations in crop yields and corn prices several months before harvest.

A scalable crop yield estimation framework based on remote sensing of solar-induced chlorophyll fluorescence (SIF)

Projected increases in food demand driven by population growth coupled with heightened agricultural vulnerability to climate change jointly pose severe threats to global food security in the coming decades, especially for developing nations. By providing real-time and low-cost observations, satellite remote sensing has been widely employed to estimate crop yield across various scales. Most such efforts are based on statistical approaches that require large amounts of ground measurements for model training/calibration, which may be challenging to obtain on a large scale in developing countries that are most food-insecure and climate-vulnerable. In this paper, we develop a generalizable framework that is mechanism-guided and practically parsimonious for crop yield estimation. We then apply this framework to estimate crop yield for two crops (corn and wheat) in two contrasting regions, the US Corn Belt US-CB, and India’s Indo–Gangetic plain Wheat Belt IGP-WB, respectively. This framework is based on the mechanistic light reactions (MLR) model utilizing remotely sensed solar-induced chlorophyll fluorescence (SIF) as a major input. We compared the performance of MLR to two commonly used machine learning (ML) algorithms: artificial neural network and random forest. We found that MLR-SIF has comparable performance to ML algorithms in US-CB, where abundant and high-quality ground measurements of crop yield are routinely available (for model calibration). In IGP-WB, MLR-SIF significantly outperforms ML algorithms. These results demonstrate the potential advantage of MLR-SIF for yield estimation in developing countries where ground truth data is limited in quantity and quality. In addition, high-resolution and crop-specific satellite SIF is crucial for accurate yield estimation. Therefore, harnessing the mechanism-guided MLR-SIF and rapidly growing satellite SIF measurements (with high resolution and crop-specificity) hold promise to enhance food security in developing countries towards more effective responses to food crises, agricultural policies, and more efficient commodity pricing.

Assessing the Potential of Multi-Temporal Conditional Generative Adversarial Networks in SAR-to-Optical Image Translation for Early-Stage Crop Monitoring

The incomplete construction of optical image time series caused by cloud contamination is one of the major limitations facing the application of optical satellite images in crop monitoring. Thus, the construction of a complete optical image time series via image reconstruction of cloud-contaminated regions is essential for thematic mapping in croplands. This study investigates the potential of multi-temporal conditional generative adversarial networks (MTcGANs) that use a single synthetic aperture radar (SAR) image acquired on a prediction date and a pair of SAR and optical images acquired on a reference date in the context of early-stage crop monitoring. MTcGAN has an advantage over conventional SAR-to-optical image translation methods as it allows input data of various compositions. As the prediction performance of MTcGAN depends on the input data composition, the variations in the prediction performance should be assessed for different input data combination cases. Such an assessment was performed through experiments using Sentinel-1 and -2 images acquired in the US Corn Belt. MTcGAN outperformed existing SAR-to-optical image translation methods, including Pix2Pix and supervised CycleGAN (S-CycleGAN), in cases representing various input compositions. In particular, MTcGAN was substantially superior when there was little change in crop vitality between the reference and prediction dates. For the SWIR1 band, the root mean square error of MTcGAN (0.021) for corn was significantly improved by 54.4% and 50.0% compared to Pix2Pix (0.046) and S-CycleGAN (0.042), respectively. Even when there were large changes in crop vitality, the prediction accuracy of MTcGAN was more than twice that of Pix2Pix and S-CycleGAN. Without considering the temporal intervals between input image acquisition dates, MTcGAN was found to be beneficial when crops were visually distinct in both SAR and optical images. These experimental results demonstrate the potential of MTcGAN in SAR-to-optical image translation for crop monitoring during the early growth stage and can serve as a guideline for selecting appropriate input images for MTcGAN.

Do newer maize hybrids grow roots faster and deeper?

AbstractUnderstanding historical changes in root depth attributes is needed for crop productivity and sustainability assessments, but such information is rare. We explored whether newer maize (Zea mays L.) hybrids grow roots faster and deeper than older hybrids and quantified the role of management and environment on root trait expression. We measured root front velocity (RFV) and maximum root depth in 11 Bayer Crop Science legacy hybrids released from 1983 to 2017 across five environments in the US Corn Belt during 2021 and 2022. Root depth was measured weekly during vegetative stages with manual probes and the maximum root depth at crop harvest with a Giddings probe. Results indicated that the RFV and maximum root depth slightly increased with the year of hybrid release (0.13% per year, p = 0.1) at 8.7 plants m−2. Historical increases in plant density from 4.7 to 8.7 plants m−2 lowered RFV and maximum root depth, but the new hybrids compensated for this loss, resulting in 4% higher RFV and 3% higher maximum depth when comparing systems from 1983 to 2017. The environment strongly influenced root trait expression (>41%). Rain anomaly and soil bulk density explained a portion of this variation. We found a linear relationship between root depth and leaf number (R2 = 0.95) and a nonlinear relationship between RFV and maximum root depth (R2 = 0.77), which can stimulate crop model improvements. Faster and deeper roots were not correlated with maize yields in our environments. This study enhances our understanding of maize breeding impacts on root traits.

Comparing impacts of corn residue removal and amelioration practices on soil properties after 3, 6, and 10 years

AbstractInformation on corn (Zea mays L.) residue removal and organic amendment additions to ameliorate removal effects on soil properties is limited. Thus, we compared impacts of corn residue removal at 59% and organic amendments (winter rye [Secale cereale L.]) cover crop and manure every other year at 24 Mg ha−1 on soil organic carbon (SOC) and physical properties after 10 years under irrigated no‐till continuous corn on a silt loam in the western US Corn Belt. We collected soil property data after 10 years and then compared it with earlier results from this experiment (3 and 6 years). Cover crop and manure amendments maintained SOC concentration, but no amelioration practice reduced SOC concentration (−15%) from 3 to 10 years when residues were annually removed in 0‐ to 5‐cm depth. Under residue retention, SOC concentration increased with time with largest increase (+41%) occurring under manure amelioration practice in the 0‐ to 5‐cm depth. Residue removal reduced soil wet aggregate stability by 34%, soil dry aggregate stability by 73%, bulk density by 6%, water content at −33 kPa matric potential by 14%, and plant available water by 17% after 10 years in the 0‐ to 5‐cm soil depth but did not affect infiltration rate. Cover crops increased wet aggregate stability (32%), but manure and cover crops had limited effects on other soil properties. Residue removal increased the susceptibility of the soil to erosion and reduced water retention and SOC after 3 years, but such effects were not fully offset by amendments.

Enhanced weathering in the US Corn Belt delivers carbon removal with agronomic benefits

Terrestrial enhanced weathering (EW) of silicate rocks, such as crushed basalt, on farmlands is a promising scalable atmospheric carbon dioxide removal (CDR) strategy that urgently requires performance assessment with commercial farming practices. We report findings from a large-scale replicated EW field trial across a typical maize-soybean rotation on an experimental farm in the heart of the United Sates Corn Belt over 4 y (2016 to 2020). We show an average combined loss of major cations (Ca2+ and Mg2+) from crushed basalt applied each fall over 4 y (50 t ha-1 y-1) gave a conservative time-integrated cumulative CDR potential of 10.5 ± 3.8 t CO2 ha-1. Maize and soybean yields increased significantly (P < 0.05) by 12 to 16% with EW following improved soil fertility, decreased soil acidification, and upregulation of root nutrient transport genes. Yield enhancements with EW were achieved with significantly (P < 0.05) increased key micro- and macronutrient concentrations (including potassium, magnesium, manganese, phosphorus, and zinc), thus improving or maintaining crop nutritional status. We observed no significant increase in the content of trace metals in grains of maize or soybean or soil exchangeable pools relative to controls. Our findings suggest that widespread adoption of EW across farming sectors has the potential to contribute significantly to net-zero greenhouse gas emissions goals while simultaneously improving food and soil security.

Genetic gains in short‐season corn hybrids: Grain yield, yield components, and grain quality traits

AbstractEra studies are important to understand historical changes in maize (Zea mays) germplasm and estimate genetic gains, yet information for short‐season maize hybrids is limited. Here, we determine grain yield genetic gain in Bayer short‐season hybrids (100–105 days) and investigate indirect changes made on 17 secondary traits, including yield components (kernel number, weight, and shelling efficiency), and grain quality traits (oil, protein, starch, ethanol, moisture, and test weight). We evaluated 40 maize hybrids released from 1980 to 2020 across 18 environments in the US Corn Belt. Plant density and N‐fertilizer were held constant within each environment. Results indicated a linear increase in grain yield (from 11.1 to 15.3 Mg ha−1, 105 kg ha−1 year−1, or 0.8% year−1) with no sign of a plateau. The increase in grain yield was attributed more to increased kernels per m2 (0.57% year−1) than kernel weight (0.23% year−1). Grain protein concentration decreased until the late 2000s and plateaued thereafter, while starch and ethanol concentration increased until the early 2000s and plateaued thereafter. However, the total amount of protein, starch, and ethanol increased linearly from 1980 to 2020. We concluded that maize breeding for increased grain yield has indirectly affected many traits at different rates and directions. Our results are encouraging for future progress in grain yield increase, update genetic gain information to 2020 for short‐season hybrids, and can inform plant breeders, crop physiologists, agronomists, and crop modelers.

What would it look like? Visualizing a future US Corn Belt landscape with more table food production

Abstract Most farmland in the US Corn Belt is used to grow row crops at large scales (e.g., corn, soybean) that are highly processed before entering the human food stream rather than specialty crops grown in smaller areas and meant for direct human consumption (table food). Bolstering local table food production close to urban populations in this region through peri-urban agriculture (PUA) could enhance sustainability and resilience. Understanding factors influencing PUA producers' preferences and willingness to produce table food would enable supportive planning and policy efforts. This study combined land use visualization and survey data to examine the potential for increased local table food production for the US Corn Belt. We developed a spatial visualization of current agricultural land use and a future scenario with increased table food production designed to meet 50% of dietary requirements for a metropolitan population in 2050. A survey was administered to row crop (1360) and specialty crop (55) producers near Des Moines, Iowa, US to understand current and intended agricultural land use and factors influencing production. Responses from 316 row crop and 25 specialty crop producers were eligible for this analysis. A future scenario with increased table food production would require less than 3% of available agricultural land and some additional producers (approximately 130, primarily for grain production). Survey responses indicated PUA producers planned small increases in table food production in the next three to five years. Producer plans, including land rental for table food production, could provide approximately 25% of residents' fruit, vegetables, and grains, an increase from the baseline of 2%. Row crop producers ranked food safety regulations, and specialty producers ranked labor concerns as strong influences on their decision-making. Both groups indicated that crop insurance and processing facilities were also important. Increasing table food production by clustering mid-scale operations to increase economies of scale and strengthening supply chains and production infrastructure could provide new profitable opportunities for farmers and more resilient food systems for growing urban regions in the US Corn Belt. Continuing to address producer factors and landscape-scale environmental impacts will be critical in considering food system sustainability challenges holistically.

Cold Climate Factors in Nitrogen Management for Maize

Among essential crop nutrients, nitrogen is the greatest management challenge in maize (Zea mays L.) production due to high requisite rates as well as dynamic transformations and losses. Climate plays a role in N management through changes in crop calendars, soil properties, agronomic practices, and yield effects. This study focuses on climate influences on maize N management and the objectives are to (i) review cold climate factors impacting economic optimum N rates (EONR), (ii) discuss approaches and climate considerations in estimating optimum N rates, and (iii) illustrate unexplored climate aspects related to optimum N rate assessment. Cold climate effects are expressed through inherent soil properties, agronomic management, and N fertilizer management. Most current N rate calculators do not explicitly account for climate factors, but implicitly integrate them through regional calibrations. Yield and EONR data from the US Corn Belt region indicate a positive correlation where lower means are associated with colder climates. High variability within climate regions is explained by differences in annual production environments, notably seasonal weather. Soil health models show that colder climates in the US are associated with higher stocks of soil organic matter, especially labile fractions. Adapt-N model simulations of a colder (North Central Wisconsin; 45.50, −89.70) and warmer (South Central Illinois; 38.50, −89.70) Corn Belt location show that higher soil organic N stocks do not increase crop N availability, presumably due to temperature-constrained N mineralization rates. The EONR for the colder site is 58 kg N ha−1 lower than the warmer site, which is well explained by differences in yield potential. Overall, abductive inferences suggest that colder climates are generally associated with higher levels of organic N stocks, but lower yields and crop N demands lessen EONRs. Seasonal weather and interactions with soil and agronomic factors also critically impact EONR, which can be assessed with model-based decision tools.

Growing season convective systems in the US Corn Belt in relation to land use-land cover and synoptic patterns

Growing season convective systems in the US Corn Belt in relation to land use-land cover and synoptic patterns

Long-term croplands water productivity in response to management and climate in the Western US Corn Belt

Global population growth and water scarcity are raising concerns about agricultural systems' ability to meet future food, fuel, feed, and fiber demands. Water productivity (WP), the ratio of crop production to water use, is crucial for assessing agricultural resilience and sustainability. Yet, research on WP often lacks long-term observational data to understand the impact of management and climate variability. Long-term monitoring of crop yield and water use, using Eddy Covariance technique, allows for accurate assessment of crop performance and their response to climate change in major cropping systems. In this study, we used data collected over a 20-year period (2001–2020) to investigate the interannual variability in yield (Y), crop evapotranspiration (ET), and water productivity (WP, ratio of Y and ET), and their response to management and climate in three major cropping systems located in Eastern Nebraska: irrigated continuous maize, irrigated maize-soybean rotation, and rainfed maize-soybean rotation. Our results showed significant differences (p <0.05) in WP between irrigated and rainfed sites, mainly attributed to variations in Y rather than ET, while there was no significant effect of crop rotation on measured responses. WP was 18.4% higher in irrigated maize in rotation relative to the rainfed site. Water input (mm, sum of precipitation and irrigation) was the main management factor in rainfed maize WP (R=0.67, p = 0.05) and Y (R=0.79, p <0.05). Vapor pressure deficit was negatively correlated with Y in rainfed maize (R=−0.72, p <0.05) and therefore was considered a determinant in WP (R=−0.7, p <0.05). For soybean, soil water content had the highest correlations with Y and WP (irrigated: R=−0.77; rainfed: R=0.49, only significant in irrigated sites). These findings can aid in formulating strategies to enhance water productivity and resilience in the US Corn Belt.

Corn Rootworm (CRW) Diabrotica SPP.: A Review of its Impact to Maize Production and Current and Future Control Options

Western corn rootworm (WCRW), Diabrotica virgifera virgifera (Le Conte), and Northern corn rootworm (NCRW), Diabrotica barberi, are economically significant pests of maize production in the US corn belt. Maize plants are primarily damaged via CRW beetle larvae feeding on root tissue, which impairs nutrient uptake, vertical stability, and ultimately agronomic potential. CRW species are remarkably adaptable and able to overcome many of the agricultural practices and controls developed by farmers and the crop protection industry. CRW has also developed resistance to multiple insecticide classes, both chemical and protein in nature. Modern integrated approaches utilizing agricultural best practices, chemical insecticides, and genetically modified maize varieties with pyramided trait proteins are needed to control resistant populations of CRW. Next-generation technologies that target CRW will provide farmers with effective options to combat this challenging pest.

A flexible and efficient knowledge-guided machine learning data assimilation (KGML-DA) framework for agroecosystem prediction in the US Midwest

Process-based models are widely used to predict the agroecosystem dynamics, but such modeled results often contain considerable uncertainty due to the imperfect model structure, biased model parameters, and inaccurate or inaccessible model inputs. Data assimilation (DA) techniques are widely adopted to reduce prediction uncertainty by calibrating model parameters or dynamically updating the model state variables using observations. However, high computational cost, difficulties in mitigating model structural error, and low flexibility in framework development hinder its applications in large-scale agroecosystem predictions. In this study, we addressed these challenges by proposing a novel DA framework that integrates a Knowledge-Guided Machine Learning (KGML)-based surrogate with tensorized ensemble Kalman filter (EnKF) and parallelized particle swarm optimization (PSO) to effectively assimilate historical and in-season multi-source remote sensing data. Specifically, we incorporate knowledge from a process-based model, ecosys, into a Gated Recurrent Unit (GRU)-based hierarchical neural network. The hierarchical architecture of KGML-DA mimics key processes of ecosys and builds a causal relationship between target variables. Using carbon budget quantification in the US Corn-Belt as a context, we evaluated KGML-DA's performance in predicting key processes of the carbon cycle at three agricultural sites (US-Ne1, US-Ne2, US-Ne3), along with county-level (627 counties) and 30-m pixel-level (Champaign County, IL) grain yield. The site experiments show that updating the upstream variable, e.g., gross primary production (GPP), improved the prediction of downstream variables such as ecosystem respiration, net ecosystem exchange, biomass, and leaf area index (LAI), with RMSE reductions ranging from 9.2% to 30.5% for corn and 4.8% to 24.6% for soybean. Uncertainty in downstream variables was automatically constrained after correcting the upstream variables, demonstrating the effectiveness of the causality linkages in the hierarchical surrogate. We found joint use of in-season GPP and evapotranspiration (ET) products along with historical GPP and surveyed yields achieved the best prediction for county-level yields, while assimilating in-season LAI observations benefitted the prediction in extreme years. Uncertainty and error analysis of regional yield estimation demonstrated that KGML-DA could reduce prediction error by 26.5% for corn and 36.2% for soybean. Remarkably, the GPU-based tensor operation design makes this DA framework more than 7000 times faster than the PB model with a High-Performance Computing system, indicating the high potential of the proposed framework for in-season, high-resolution agroecosystem predictions.

Improving nitrogen cycling in a land surface model (CLM5) to quantify soil N2O, NO, and NH3 emissions from enhanced rock weathering with croplands

Abstract. Surficial enhanced rock weathering (ERW) is a land-based carbon dioxide removal (CDR) strategy that involves applying crushed silicate rock (e.g., basalt) to agricultural soils. However, unintended biogeochemical interactions with the nitrogen cycle may arise through ERW increasing soil pH as basalt grains undergo dissolution that may reinforce, counteract, or even offset the climate benefits from carbon sequestration. Increases in soil pH could drive changes in the soil emissions of key non-CO2 greenhouse gases, e.g., nitrous oxide (N2O), and trace gases, e.g., nitric oxide (NO) and ammonia (NH3), that affect air quality and crop and human health. We present the development and implementation of a new improved nitrogen cycling scheme for the Community Land Model v5 (CLM5), the land component of the Community Earth System Model, allowing evaluation of ERW effects on soil gas emissions. We base the new parameterizations on datasets derived from soil pH responses of N2O, NO, and NH3 in ERW field trial and mesocosm experiments with crushed basalt. These new capabilities involve the direct implementation of routines within the CLM5 N cycle framework, along with asynchronous coupling of soil pH changes estimated through an ERW model. We successfully validated simulated “control” (i.e., no ERW) seasonal cycles of soil N2O, NO, and NH3 emissions against a wide range of global emission inventories. We benchmark simulated mitigation of soil N2O fluxes in response to ERW against a subset of data from ERW field trials in the US Corn Belt. Using the new scheme, we provide a specific example of the effect of large-scale ERW deployment with croplands on soil nitrogen fluxes across five key regions with high potential for CDR with ERW (North America, Brazil, Europe, India, and China). Across these regions, ERW implementation led to marked reductions in N2O and NO (both 18 %), with moderate increases in NH3 (2 %). While further developments are still required in our implementations when additional ERW data become available, our improved N cycle scheme within CLM5 has utility for investigating the potential of ERW point-source and regional effects of soil N2O, NO, and NH3 fluxes in response to current and future climates. This framework also provides the basis for assessing the implications of ERW for air quality given the role of NO in tropospheric ozone formation, as well as both NO and NH3 in inorganic aerosol formation.