Explore the vast range of titles available.

Cover crops are known to promote many aspects of soil and water quality, yet estimates find that in 2012 only 2.3% of the total agricultural lands in the Midwestern USA were using cover crops. Focus groups were conducted across the Corn Belt state of Iowa to better understand how farmers confront barriers to cover crop adoption in highly intensive agricultural production systems. Although much prior research has focused on analyzing factors that help predict cover crop use on farms, there is limited research on how farmers navigate and overcome field-level (e.g. proper planting of a cover crop) and structural barriers (e.g. market forces) associated with the use of cover crops. The results from the analysis of these conversations suggest that there is a complex dialectical relationship between farmers' individual management decisions and the broader agricultural context in the region that constrains their decisions. Farmers in these focus groups shared how they navigate complex management decisions within a generally homogenized agricultural and economic landscape that makes cover crop integration challenging. Many who joined the focus groups have found ways to overcome barriers and successfully integrate cover crops into their cropping systems. This is illustrated through farmers' descriptions of their ‘whole system’ approach to cover crops management, where they described how they prioritize the success of their cover crops by focusing on multiple aspects of management, including changes they have made to nutrient application and modifications to equipment. These producers also engage with farmer networks to gain strategies for overcoming management challenges associated with cover crops. Although many participants had successfully planted cover crops, they tended to believe that greater economic incentives and/or more diverse crop and livestock markets would be needed to spur more widespread adoption of the practice. Our results further illustrate how structural and field-level barriers constrain individual actions, as it is not simply the basic agronomic considerations (such as seeding and terminating cover crops) that pose a challenge to their use, but also the broader economic and market drivers that exist in agriculturally intensive systems. Our study provides evidence that reducing structural barriers to adoption may be necessary to increase the use of this conservation practice to reduce environmental impacts associated with intensive agricultural production.

The emergence of new technologies to synthesize and analyze big data with high-performance computing has increased our capacity to more accurately predict crop yields. Recent research has shown that machine learning (ML) can provide reasonable predictions faster and with higher flexibility compared to simulation crop modeling. However, a single machine learning model can be outperformed by a “committee” of models (machine learning ensembles) that can reduce prediction bias, variance, or both and is able to better capture the underlying distribution of the data. Yet, there are many aspects to be investigated with regard to prediction accuracy, time of the prediction, and scale. The earlier the prediction during the growing season the better, but this has not been thoroughly investigated as previous studies considered all data available to predict yields. This paper provides a machine leaning based framework to forecast corn yields in three US Corn Belt states (Illinois, Indiana, and Iowa) considering complete and partial in-season weather knowledge. Several ensemble models are designed using blocked sequential procedure to generate out-of-bag predictions. The forecasts are made in county-level scale and aggregated for agricultural district and state level scales. Results show that the proposed optimized weighted ensemble and the average ensemble are the most precise models with RRMSE of 9.5%. Stacked LASSO makes the least biased predictions (MBE of 53 kg/ha), while other ensemble models also outperformed the base learners in terms of bias. On the contrary, although random k-fold cross-validation is replaced by blocked sequential procedure, it is shown that stacked ensembles perform not as good as weighted ensemble models for time series data sets as they require the data to be non-IID to perform favorably. Comparing our proposed model forecasts with the literature demonstrates the acceptable performance of forecasts made by our proposed ensemble model. Results from the scenario of having partial in-season weather knowledge reveals that decent yield forecasts with RRMSE of 9.2% can be made as early as June 1st. Moreover, it was shown that the proposed model performed better than individual models and benchmark ensembles at agricultural district and state-level scales as well as county-level scale. To find the marginal effect of each input feature on the forecasts made by the proposed ensemble model, a methodology is suggested that is the basis for finding feature importance for the ensemble model. The findings suggest that weather features corresponding to weather in weeks 18–24 (May 1st to June 1st) are the most important input features.

Sun-induced chlorophyll fluorescence (SIF) measurements have shown unique potential for quantifying plant physiological stress. However, recent investigations found canopy structure and radiation largely control SIF, and physiological relevance of SIF remains yet to be fully understood. This study aims to evaluate whether the SIF-derived physiological signal improves quantification of crop responses to environmental stresses, by analyzing data at three different spatial scales within the U.S. Corn Belt, i.e. experiment plot, field, and regional scales, where ground-based portable, stationary and space-borne hyperspectral sensing systems are used, respectively. We found that, when controlling for variations in incoming radiation and canopy structure, crop SIF signals can be decomposed into non-physiological (i.e. canopy structure and radiation, 60% ∼ 82%) and physiological information (i.e. physiological SIF yield, Φ F , 17% ∼ 31%), which confirms the contribution of physiological variation to SIF. We further evaluated whether Φ F indicated plant responses under high-temperature and high vapor pressure deficit (VPD) stresses. The plot-scale data showed that Φ F responded to the proxy for physiological stress (partial correlation coefficient, r p = 0.40, p < 0.001) while non-physiological signals of SIF did not respond ( p > 0.1). The field-scale Φ F data showed water deficit stress from the comparison between irrigated and rainfed fields, and Φ F was positively correlated with canopy-scale stomatal conductance, a reliable indicator of plant physiological condition (correlation coefficient r = 0.60 and 0.56 for an irrigated and rainfed sites, respectively). The regional-scale data showed Φ F was more strongly correlated spatially with air temperature and VPD ( r = 0.23 and 0.39) than SIF ( r = 0.11 and 0.34) for the U.S. Corn Belt. The lines of evidence suggested that Φ F reflects crop physiological responses to environmental stresses with greater sensitivity to stress factors than SIF, and the stress quantification capability of Φ F is spatially scalable. Utilizing Φ F for physiological investigations will contribute to improve our understanding of vegetation responses to high-temperature and high-VPD stresses.

Nitrous oxide (N₂O) has a global warming potential that is 300 times that of carbon dioxide on a 100-y timescale, and is of major importance for stratospheric ozone depletion. The climate sensitivity of N₂O emissions is poorly known, which makes it difficult to project how changing fertilizer use and climate will impact radiative forcing and the ozone layer. Analysis of 6 y of hourly N₂O mixing ratios from a very tall tower within the US Corn Belt—one of the most intensive agricultural regions of the world—combined with inverse modeling, shows large interannual variability in N₂O emissions (316 Gg N₂O-N·y−1 to 585 Gg N₂O-N·y−1). This implies that the regional emission factor is highly sensitive to climate. In the warmest year and spring (2012) of the observational period, the emission factor was 7.5%, nearly double that of previous reports. Indirect emissions associated with runoff and leaching dominated the interannual variability of total emissions. Under current trends in climate and anthropogenic N use, we project a strong positive feedback to warmer and wetter conditions and unabated growth of regional N₂O emissions that will exceed 600 Gg N₂O-N·y−1, on average, by 2050. This increasing emission trend in the US Corn Belt may represent a harbinger of intensifying N₂O emissions from other agricultural regions. Such feedbacks will pose a major challenge to the Paris Agreement, which requires large N₂O emission mitigation efforts to achieve its goals.

Photosynthesis is the process by which plants harvest sunlight to produce sugars from carbon dioxide and water. It is the primary source of energy for all life on Earth; hence it is important to understand how this process responds to climate change and human impact. However, model-based estimates of gross primary production (GPP, output from photosynthesis) are highly uncertain, in particular over heavily managed agricultural areas. Recent advances in spectroscopy enable the space-based monitoring of sun-induced chlorophyll fluorescence (SIF) from terrestrial plants. Here we demonstrate that spaceborne SIF retrievals provide a direct measure of the GPP of cropland and grassland ecosystems. Such a strong link with crop photosynthesis is not evident for traditional remotely sensed vegetation indices, nor for more complex carbon cycle models. We use SIF observations to provide a global perspective on agricultural productivity. Our SIF-based crop GPP estimates are 50-75% higher than results from state-of-the-art carbon cycle models over, for example, the US Corn Belt and the Indo-Gangetic Plain, implying that current models severely underestimate the role of management. Our results indicate that SIF data can help us improve our global models for more accurate projections of agricultural productivity and climate impact on crop yields. Extension of our approach to other ecosystems, along with increased observational capabilities for SIF in the near future, holds the prospect of reducing uncertainties in the modeling of the current and future carbon cycle.

We investigate the predictive performance of two novel CNN-DNN machine learning ensemble models in predicting county-level corn yields across the US Corn Belt (12 states). The developed data set is a combination of management, environment, and historical corn yields from 1980 to 2019. Two scenarios for ensemble creation are considered: homogenous and heterogenous ensembles. In homogenous ensembles, the base CNN-DNN models are all the same, but they are generated with a bagging procedure to ensure they exhibit a certain level of diversity. Heterogenous ensembles are created from different base CNN-DNN models which share the same architecture but have different hyperparameters. Three types of ensemble creation methods were used to create several ensembles for either of the scenarios: Basic Ensemble Method (BEM), Generalized Ensemble Method (GEM), and stacked generalized ensembles. Results indicated that both designed ensemble types (heterogenous and homogenous) outperform the ensembles created from five individual ML models (linear regression, LASSO, random forest, XGBoost, and LightGBM). Furthermore, by introducing improvements over the heterogenous ensembles, the homogenous ensembles provide the most accurate yield predictions across US Corn Belt states. This model could make 2019 yield predictions with a root mean square error of 866 kg/ha, equivalent to 8.5% relative root mean square and could successfully explain about 77% of the spatio-temporal variation in the corn grain yields. The significant predictive power of this model can be leveraged for designing a reliable tool for corn yield prediction which will in turn assist agronomic decision makers.

In the US Corn Belt, a recent doubling in commodity prices has created incentives for landowners to convert grassland to corn and soybean cropping. Here, we use land cover data from the National Agricultural Statistics Service Cropland Data Layer to assess grassland conversion from 2006 to 2011 in the Western Corn Belt (WCB): five states including North Dakota, South Dakota, Nebraska, Minnesota, and Iowa. Our analysis identifies areas with elevated rates of grass-to-corn/soy conversion (1.0–5.4% annually). Across the WCB, we found a net decline in grass-dominated land cover totaling nearly 530,000 ha. With respect to agronomic attributes of lands undergoing grassland conversion, corn/soy production is expanding onto marginal lands characterized by high erosion risk and vulnerability to drought. Grassland conversion is also concentrated in close proximity to wetlands, posing a threat to waterfowl breeding in the Prairie Pothole Region. Longer-term land cover trends from North Dakota and Iowa indicate that recent grassland conversion represents a persistent shift in land use rather than short-term variability in crop rotation patterns. Our results show that the WCB is rapidly moving down a pathway of increased corn and soybean cultivation. As a result, the window of opportunity for realizing the benefits of a biofuel industry based on perennial bioenergy crops, rather than corn ethanol and soy biodiesel, may be closing in the WCB.

The impacts of climate change on agricultural production are a global concern and have already begun to occur (Kawasaki 2018 Am. J. Agric. Econ . 101 172–92; Ortiz-Bobea et al 2021 Nat. Clim. Change 11 306–12), with major drivers including warmer temperatures and the occurrence of extreme weather events (Lobell and Field 2007 Environ. Res. Lett. 2 014002; Challinor et al 2014 Nat. Clim. Chang e 4 287; Rosenzweig et al 2001 Glob. Change Hum. Health 2 90–104; Schlenker and Roberts 2009 Proc. Natl Acad. Sci. USA 106 15594–8; Lobell et al 2014 Science 344 516–9; Ortiz-Bobea et al 2019 Environ. Res. Lett. 14 064003). An important dimension of the climate change-crop yield relationship that has often been overlooked in the empirical literature is the influence that warming temperatures can have on plant damage arriving through biotic channels, such as pest infestation or fungal infection (Rosenzweig et al 2001 Glob. Change Hum. Health 2 90–104). Aflatoxins are carcinogenic chemicals produced by the fungi Aspergillus flavus and A. parasiticus, which commonly infect food crops. Currently, in the United States, aflatoxin is a perennial contaminant in corn grown in the South, but rare in the Corn Belt and northern states. Climate change may expand aflatoxin’s geographical prevalence, however; because hot, dry summers promote aflatoxin accumulation. Here we model aflatoxin risk as a function of corn plant growth stages and weather to predict US regions with high aflatoxin risk in 2031–2040, based on 16 climate change models. Our results suggest that over 89.5% of corn-growing counties in 15 states, including the Corn Belt, will experience increased aflatoxin contamination in 2031–2040 compared to 2011–2020. Interestingly, the results are spatially heterogeneous and include several southern counties expected to have lower aflatoxin risk, because the causative fungi become inactivated at very high temperatures.

Conservation tillage is a primary tenet of conservation agriculture aimed at restoring and maintaining soil health for long-term crop productivity. Because soil degradation typically operates on century timescales, farmer adoption is influenced by near-term yield impacts and profitability. Although numerous localized field trials have examined the yield impacts of conservation tillage, their results are mixed and often unrepresentative of real-world conditions. Here, we applied a machine-learning causal inference approach to satellite-derived datasets of tillage practices and crop yields spanning the US Corn Belt from 2005 to 2017 to assess on-the-ground yield impacts at field-level resolution across thousands of fields. We found an average 3.3% and 0.74% yield increase for maize and soybeans, respectively, for fields with long-term conservation tillage. This effect was diminished in fields that only recently converted to conservation tillage. We also found significant variability in these effects, and we identified soil and weather characteristics that mediate the direction and magnitude of yield responses. This work supports soil conservation practices by demonstrating they can be used with minimal and typically positive yield impacts.

Crop yield prediction before the harvest is crucial for food security, grain trade, and policy making. Previously, several machine learning methods have been applied to predict crop yield using different types of variables. In this study, we propose using the Geographically Weighted Random Forest Regression (GWRFR) approach to improve crop yield prediction at the county level in the US Corn Belt. We trained the GWRFR and five other popular machine learning algorithms (Multiple Linear Regression (MLR), Partial Least Square Regression (PLSR), Support Vector Regression (SVR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)) with the following different sets of features: (1) full length features; (2) vegetation indices; (3) gross primary production (GPP); (4) climate data; and (5) soil data. We compared the results of the GWRFR with those of the other five models. The results show that the GWRFR with full length features (R2 = 0.90 and RMSE = 0.764 MT/ha) outperforms other machine learning algorithms. For individual categories of features such as GPP, vegetation indices, climate, and soil features, the GWRFR also outperforms other models. The Moran’s I value of the residuals generated by GWRFR is smaller than that of other models, which shows that GWRFR can better address the spatial non-stationarity issue. The proposed method in this article can also be potentially used to improve yield prediction for other types of crops in other regions.