Water deficit and water excess constitute severe stresses that limit crop yield and are likely to intensify as climate becomes more variable. Regional crop production aggregates for the US Midwest ...indicate widespread yield losses in past decades due to both extreme rainfall and water limited conditions, though the degree to which these weather impacts are related to site-specific factors such as landscape position and soils has not been examined in a systematic manner. This study offers observational evidence from a large sample of commercial crop fields to support the hypothesis that landscape position is the primary mediator of crop yield responses to weather within unstable field zones (i.e., zones where yields tend to fluctuate between high and low, depending on the year). Results indicate that yield losses in unstable zones driven by water excess and deficits occur throughout a wide range of seasonal rainfall, even simultaneously under normal weather. Field areas prone to water stress are shown to lag as much as 23-33% below the field average during drought years and 26-33% during deluge years. By combining large-scale spatial datasets, we identify 2.65 million hectares of water-stress prone cropland, and estimate an aggregated economic loss impact of $536M USD yr
, 4.0 million tons yr
of less CO
fixed in crop biomass, and 52.6 Gg yr
of more reactive N in the environment. Yield stability maps can be used to spatially implement adaptation practices to mitigate weather-induced stresses in the most vulnerable cropland.
Pre-growing season prediction of crop production outcomes such as grain yields and nitrogen (N) losses can provide insights to farmers and agronomists to make decisions. Simulation crop models can ...assist in scenario planning, but their use is limited because of data requirements and long runtimes. Thus, there is a need for more computationally expedient approaches to scale up predictions. We evaluated the potential of four machine learning (ML) algorithms (LASSO Regression, Ridge Regression, random forests, Extreme Gradient Boosting, and their ensembles) as meta-models for a cropping systems simulator (APSIM) to inform future decision support tool development. We asked: (1) How well do ML meta-models predict maize yield and N losses using pre-season information? (2) How many data are needed to train ML algorithms to achieve acceptable predictions? (3) Which input data variables are most important for accurate prediction? And (4) do ensembles of ML meta-models improve prediction? The simulated dataset included more than three million data including genotype, environment and management scenarios. XGBoost was the most accurate ML model in predicting yields with a relative mean square error (RRMSE) of 13.5%, and Random forests most accurately predicted N loss at planting time, with a RRMSE of 54%. ML meta-models reasonably reproduced simulated maize yields using the information available at planting, but not N loss. They also differed in their sensitivities to the size of the training dataset. Across all ML models, yield prediction error decreased by 10%-40% as the training dataset increased from 0.5 to 1.8 million data points, whereas N loss prediction error showed no consistent pattern. ML models also differed in their sensitivities to input variables (weather, soil properties, management, initial conditions), thus depending on the data availability researchers may use a different ML model. Modest prediction improvements resulted from ML ensembles. These results can help accelerate progress in coupling simulation models and ML toward developing dynamic decision support tools for pre-season management.
Warming generally leads to increased evaporative demand, altering the amount of water needed for growing crops. For the Midwest, some studies have suggested that reaching yield targets by 2050 will ...not be possible without additional precipitation or large expansion of irrigation. Here, we show that this claim is not supported by the historical summer climate trends, which indicate that the warming of daily average temperatures is largely driven by increases in minimum temperatures, while maximum temperatures have decreased. This has translated into a net decrease in vapor pressure deficit (VPD) and potential evapotranspiration (PET). With the increasing rainfall, this suggests that crop water deficits have likely become less frequent in the region despite the warming climate. By projecting these trends into 2050 and ancillary use of a crop model, we estimate minor changes in PET that would have minimal effects on corn yields (<6%) under persistence of these trends.
•We synthesized measured and modeling data to examine rye effects on maize systems.•Soil water deficit caused by rye transpiration affected maize yield only in drought years.•34% of the precipitation ...ended up in subsurface drainage at this site.•Rye produced 47kgha−1 of shoot biomass per each mm of water used.•APSIM simulated reduced NO3-N losses (26±26%), but not drainage (4±13%) or yield (2±6%).
Inclusion of a rye cover crop into maize-based systems can offer environmental benefits, but adoption of the practice in the US Midwest is still low. This is related to the possible risk of reduced maize yields following rye. We hypothesized that the magnitude of rye effects on maize yields and drainage water and nitrate (NO3)-N losses would be proportionally related to rye biomass. We tested this hypothesis by analyzing data from continuous maize treatments (with and without cover crop) in Iowa, US, that were fertilized following recommendations from late spring nitrate tests. Dataset included measurements (2009–2014) of soil water and temperature, drainage water and NO3-N losses, soil NO3, rye shoot and root biomass and C:N, and maize yields. We supplemented our analysis with a literature review and the use of a cropping systems model (APSIM) to calculate trade-offs in system performance characteristics. Experimentally, rye cover crop reduced drainage by 12% and NO3-N losses by 20% (or 31% per unit of N applied), and maize yields by 6%. We also found minimal effects on soil temperature, water deficits that reduced yields only during drought years (2012 and 2013), and lower NO3-N losses that were related to reduced NO3-N concentrations in drainage. Results also revealed a linear relationship between drainage and precipitation (r2=0.96), and rye transpiration and shoot biomass (r2=0.84). Model scenario analysis (4 termination dates×30years) indicated that rye cover crop decreases NO3-N losses (-25.5±26%) but does not always reduce drainage water (-3.9±13%) or grain yields (-1.84±6%), which is consistent with experimental and literature results. However, analysis of the synthesized measured and simulated dataset do not support a strong relationship between these variables and rye biomass. These results are valuable for decision-making and add new fundamental knowledge on rye water and nitrogen use.
Despite the new equipment capabilities, uneven crop stands are still common occurrences in crop fields, mainly due to spatial heterogeneity in soil conditions, seedling mortality due to herbivore ...predation and disease, or human error. Non-uniform plant stands may reduce grain yield in crops like maize. Thus, detecting signs of variability in crop stand density early in the season provides critical information for management decisions and crop yield forecasts. Processing techniques applied on images captured by unmanned aerial vehicles (UAVs) has been used successfully to identify crop rows and estimate stand density and, most recently, to estimate plant-to-plant interval distance. Here, we further test and apply an image processing algorithm on UAV images collected from yield-stability zones in a commercial crop field. Our objective was to implement the algorithm to compare variation of plant-spacing intervals to test whether yield differences within these zones are related to differences in crop stand characteristics. Our analysis indicates that the algorithm can be reliably used to estimate plant counts (precision >95% and recall >97%) and plant distance interval (R2 ~0.9 and relative error <10%). Analysis of the collected data indicated that plant spacing variability differences were small among plots with large yield differences, suggesting that it was not a major cause of yield variability across zones with distinct yield history. This analysis provides an example of how plant-detection algorithms can be applied to improve the understanding of patterns of spatial and temporal yield variability.
In this paper, we advanced the concept of nitrogen-use efficiency (NUE) of cropping systems by linking current crop- and soil-based approaches into a system NUE. We compared this new index to ...traditional metrics, and show how its application can yield insights about N cycling dynamics and tradeoffs.
Display omitted
•Developed a novel systems NUE (sNUE) by synthesizing crop and soil-based approaches.•The maize-soybean rotation removed 87% of N inputs in grains.•45% of the N losses were due to inefficient N input use; the rest to poor N retention.•Balancing production-environmental tradeoffs resulted on a yield penalty of 5–7%.•sNUE was relatively more stable and less correlated to other metrics.
Increasing nitrogen (N)-use efficiency (NUE) is key to improving crop production while mitigating ecologically-damaging environmental N losses. Traditional approaches to assess NUE are principally focused on evaluating crop responses to N inputs, often consider only what happens during the growing season, and ignore other means to improve system efficiency, such as by tightening the cycling of soil N (e.g. with N scavenging cover crops). As the goals of improving production and environmental quality converge, new metrics that can simultaneously capture multiple aspects of system performance are needed. To fill this gap, we developed a theoretical framework that links both crop- and soil-based approaches to derive a system N-use efficiency (sNUE) index. This easily interpretable metric succinctly characterizes N cycling and facilitates comparison of systems that differ in biophysical controls on N dynamics. We demonstrated the application of this new approach and compared it to traditional NUE metrics using data generated with a process-based model (APSIM), trained and tested with experimental datasets (Iowa, USA). Modeling of maize-soybean rotations indicated that despite their high crop NUE, only 45% of N losses could be attributed to the inefficient use of N inputs, whereas the rest originated from the release of native soil N into the environment, due to the asynchrony between soil mineralization and crop uptake. Additionally, sNUE produced estimates of system efficiency that were more stable across weather years and less correlated to other metrics across distinct crop sequences and N fertilizer input levels. We also showed how sNUE allows for the examination of tradeoffs between N cycling and production performance, and thus has the potential to aid in the design of systems that better balance production and environmental outcomes.
A delayed harvest of maize and soybean crops is associated with yield or revenue losses, whereas a premature harvest requires additional costs for artificial grain drying. Accurately predicting the ...ideal harvest date can increase profitability of US Midwest farms, but today's predictive capacity is low. To fill this gap, we collected and analyzed time-series grain moisture datasets from field experiments in Iowa, Minnesota and North Dakota, US with various maize (n = 102) and soybean (n = 36) genotype-by-environment treatments. Our goal was to examine factors driving the post-maturity grain drying process, and develop scalable algorithms for decision-making. The algorithms evaluated are driven by changes in the grain equilibrium moisture content (function of air relative humidity and temperature) and require three input parameters: moisture content at physiological maturity, a drying coefficient and a power constant. Across independent genotypes and environments, the calibrated algorithms accurately predicted grain dry-down of maize (r
= 0.79; root mean square error, RMSE = 1.8% grain moisture) and soybean field crops (r
= 0.72; RMSE = 6.7% grain moisture). Evaluation of variance components and treatment effects revealed that genotypes, weather-years, and planting dates had little influence on the post-maturity drying coefficient, but significantly influenced grain moisture content at physiological maturity. Therefore, accurate implementation of the algorithms across environments would require estimating the initial grain moisture content, via modeling approaches or in-field measurements. Our work contributes new insights to understand the post-maturity grain dry-down and provides a robust and scalable predictive algorithm to forecast grain dry-down and ideal harvest dates across environments in the US Corn Belt.
•Integrated landscape and supply chain optimization leads to environmental benefit.•High-resolution stochastic model accounts for biomass yield uncertainty.•Framework for generating and integrating ...high-resolution yield and soil carbon data.
This paper proposes an integrated stochastic mixed-integer linear programming model for biofuel supply chain and landscape design optimization that considers the interactions between uncertainty in biomass yield, spatially explicit feedstock availability, supply chain configuration, operational decisions, and the system's environmental impact. By modeling crop establishment and fertilization as strategic decisions made before the realization of uncertainty, model solutions identify integrated supply chain configurations better suited to mitigate uncertain biomass yields. Importantly, the paper presents an approach for gathering and processing the large amount of necessary real-world data, including an accurate accounting of soil carbon sequestration. A case study located in Michigan, USA, demonstrating the capabilities of the integrated model with realistic data, is presented. Results at a variety of harvesting site resolutions and number of uncertainty scenarios, show that large-scale instances, at fine spatial resolutions, identify attractive environmental solutions and that solving the stochastic problem leads to an economic benefit.
Water deficit and water excess constitute severe stresses that limit crop yield and are likely to intensify as climate becomes more variable. Regional crop production aggregates for the US Midwest ...indicate widespread yield losses in past decades due to both extreme rainfall and water limited conditions, though the degree to which these weather impacts are related to site-specific factors such as landscape position and soils has not been examined in a systematic manner. This study offers observational evidence from a large sample of commercial crop fields to support the hypothesis that landscape position is the primary mediator of crop yield responses to weather within unstable field zones (i.e., zones where yields tend to fluctuate between high and low, depending on the year). Results indicate that yield losses in unstable zones driven by water excess and deficits occur throughout a wide range of seasonal rainfall, even simultaneously under normal weather. Field areas prone to water stress are shown to lag as much as 23–33% below the field average during drought years and 26–33% during deluge years. By combining large-scale spatial datasets, we identify 2.65 million hectares of water-stress prone cropland, and estimate an aggregated economic loss impact of $536M USD yr-1, 4.0 million tons yr-1 of less CO2 fixed in crop biomass, and 52.6 Gg yr-1 of more reactive N in the environment. Yield stability maps can be used to spatially implement adaptation practices to mitigate weather-induced stresses in the most vulnerable cropland
There is an urgent need to transform unsustainable “linear” grain production systems in the United States (U.S.) and other countries like China, Brazil, Argentina, Canada, Russia, Australia and ...Europe, into more circular and sustainable systems to address the simultaneous challenges of resource depletion, environmental degradation, and the growing global demand for food under the threat of climate change.
In this perspective, we survey the current state of circularity of U.S. grain production, and discuss how we can transform the systems into more circular systems.
Specifically, we lay out a vision of circular grain production enabled by novel digital, mechanical, and biological technologies that allow closing loops of nutrient and energy flows within the farm, through the optimization of land-use choices and crop management. We also examine market- and policy-based mechanisms that could incentivize the widespread adoption of these key technologies.
Display omitted
•There is an urgent need to transform “linear” grain production systems into circular and sustainable systems•Digital agriculture technologies can increase circularity in the grain production systems•Regenerative practices, new genetics, robotics and electrification of fertilizer production will increase circularity•Widespread adoption of circular systems will depend critically on their profitability•Profitability will in turn depend on public policies needed to encourage farmers to adopt sustainable practices