Search for: All records

Nitrous oxide (N₂O) is a potent and persistent greenhouse gas, with rising atmospheric concentrations driven in part by inefficient use of synthetic nitrogen (N) fertilizers in agriculture. Predicting soil N₂O emissions is challenging due to high spatial and temporal variability arising from complex soil biogeochemical processes. Process-based ecosystem models and standalone machine learning (ML) approaches without extensive site-specific calibration often miss high emission episodes. Here, we show how an Ensemble Modeling System (EMS) based on outputs from an ensemble of ecosystem models coupled to an ensemble of ML models can improve predictions and understanding of N2O fluxes from US cropland. Trained and validated on approximately 12,000 N2O chamber measurements at 17 U.S. Midwest sites (six crops, 35 management practices), the EMS accurately predicted daily fluxes of N2O at both training (R² = 0.84, RMSE = 16.4 g N ha⁻¹ d⁻¹) and held-out testing sites (R² = 0.84, RMSE = 6.2 g N ha⁻¹ d⁻¹). Analyses identified six dominant N₂O drivers: soil organic carbon (SOC), NH₄⁺, NO₃⁻, water-filled pore space (WFPS), soil temperature, and biomass production. Wet, warm soils produced large N₂O peaks only with sufficient SOC and mineral N; in low-SOC soils, fluxes remained low. Incorporating these drivers into process-based models might significantly improve their predictive capacity. The EMS demonstrates a strong potential to predict N₂O fluxes at unseen sites, enabling more reliable regional inventories, improved gap-filling where measurements are sparse, and enhanced understanding of mechanisms to advance targeted mitigation strategies in food, feed, and bioenergy crops. # Data from: Coupled machine learning-ecosystem ensemble models substantially improve predictions of nitrous oxide (N~2~O) fluxes from US croplands Dataset DOI: [10.5061/dryad.pvmcvdnzx](10.5061/dryad.pvmcvdnzx) ## Description of the data and file structure We present here the data that were used for the analysis presented in: Coupled machine learning-ecosystem ensemble models substantially improve predictions of nitrous oxide (N~2~O) fluxes from US croplands. ### Files and variables Files: Dataset_S1.xlsx, Dataset_S2.csv, Dataset_S3.csv, #### **Description:**  **Description of data sheets** **Dataset S1A columns** * **Site_ID:** Numeric identifier for the experimental site. * **Treatment_ID:** Numeric code for the management treatment applied at that site * **DataUse:** To assign data to be used for model training (development) and testing (held-out evaluation) * **State/Province:** State acronym * **Latitude decimal deg:** Site location latitude * **Longitude decimal deg:** Site location longitude * **Start Data Year:** Starting year of data used * **End Data Year:** Ending year of data used * **Cover crop:** Type of cover crop used within the treatment * **Rotation Descriptor:** Describe the rotation of crops within the treatment * **Tillage Descriptor:** Describe tillage type within the treatment * **Residual Removal:** Describe residual management within the treatment * **Irrigation:** Describe if irrigation was applied or not within the treatment * **N Treatment Descriptor:** Describe nitrogen amendments within the treatment * **Reference:** Reference for the data **Dataset S1B**: This sheet contains the reference list for the data used **Dataset S2 columns** * **Date:** Gas sampling days * **Site_ID:** Numeric identifier for the experimental site. * **Treatment_ID:** Numeric code for the management treatment applied at that site * **DataUse:** To assign data to be used for model training (development) and testing (held-out evaluation) * **Observed N2O:** Daily average N2O flux measured (g N2O-N ha-1d-1) * **Predicted N2O:** Daily average N2O flux predicted by multimodel hybrid framework (g N2O-N ha-1d-1) * **NH4:** Process-based models simulated daily NH4-N content in the top 30-cm soil layer (kg ha-1) * **SOC:** Process-based models simulated daily soil organic carbon in the top 30-cm soil layer (kg ha-1) * **NO3:** Process-based models simulated daily NO3-N content in the top 30-cm soil layer (kg ha-1) * **ST:** Process-based models simulated daily average soil temperature in the top 30-cm soil layer (°C) * **WFPS:** Process-based models simulated daily water-filled pore space in the top 30-cm soil layer (fraction) * **ABG:** Process-based models simulated daily above-ground biomass (kg ha-1) * **BG:** Process-based models simulated daily below-ground biomass (kg ha-1) * **SRAD:** Average solar radiation for the last five days before gas sampling (Watt m-2) * **Tmax:** Average maximum air temperature for the last three days before gas sampling  (°C) * **APrecip:** Average precipitation in the last fifteen days before gas sampling (mm) * **Wspd:** Average wind in the last fifteen days before gas sampling (m s-1) * **LAI:** Process-based models simulated daily leaf area index (m2 m2) * **Nstress:** Process-based models simulated the daily nitrogen stress factor (fraction) * **Wstress:** Process-based models simulated the daily water stress factor (fraction) * **PET:** Process-based models simulated daily potential evapotranspiration (mm) * **SE:** Process-based models simulated daily soil evaporation (mm) * **SPrecip:** Cumulative precipitation in the last two days before gas sampling (mm) * **SH:** Average specific humidity in the last three days before gas sampling (g kg-1) * **RH:** Average relative humidity in the last fifteen days before gas sampling (%) **Dataset S3 columns** * **Date:** Gas sampling days * **Site_ID:** Numeric identifier for the experimental site. * **Treatment_ID:** Numeric code for the management treatment applied at that site * **DataUse:** To assign data to be used for model training (development) and testing (held-out evaluation) * **SD:** Monte Carlo standard deviation of the simulated daily N₂O flux distribution (g N2O-N ha-1d-1) * **CV**: Monte Carlo coefficient of variation of the simulated daily N₂O flux distribution (%) * **CI05:** 5th percentile (lower 90 % confidence bound) of the Monte Carlo flux distribution(g N2O-N ha-1d-1) * **CI95:** 95th percentile (upper 90 % confidence bound) of the Monte Carlo flux distribution(g N2O-N ha-1d-1) 

Nitrous oxide (N₂O) is a potent and persistent greenhouse gas, with rising atmospheric concentrations driven in part by inefficient use of synthetic nitrogen (N) fertilizers in agriculture. Predicting soil N₂O emissions is challenging due to high spatial and temporal variability arising from complex soil biogeochemical processes. Process-based ecosystem models and standalone machine learning (ML) approaches without extensive site-specific calibration often miss high-emission episodes. Here, we show how an Ensemble Modeling System (EMS) based on outputs from an ensemble of ecosystem models coupled to an ensemble of ML models can improve predictions and understanding of N₂O fluxes from US cropland. Trained and validated on ~12,000 N₂O chamber measurements at 17 US Midwest sites (six crops, 35 management practices), the EMS accurately predicted daily fluxes of N₂O at both training (R²= 0.84, RMSE = 16.4 g N ha⁻¹d⁻¹) and held-out testing sites (R²= 0.84, RMSE = 6.2 g N ha⁻¹d⁻¹). Analyses identified six dominant N₂O drivers: soil organic carbon (SOC), NH₄⁺, NO₃^-, water-filled pore space, temperature, and aboveground biomass production. Wet, warm soils produced large N₂O peaks only with sufficient SOC and mineral N; in low-SOC soils, fluxes remained low. Incorporating these drivers into process-based models might significantly improve their predictive capacity. The EMS demonstrates a strong potential to predict N₂O fluxes at unseen sites, enabling more reliable regional inventories, improved gap-filling where measurements are sparse, and enhanced understanding of mechanisms to advance targeted mitigation strategies in food, feed, and bioenergy crops. 

Widespread expansion of agriculture and forestry has altered the surface of the Earth, the composition of the atmosphere, and as a result, the climate. Here we quantify the radiative forcing caused by the deforestation of an ecoregion of the U.S. Upper Midwest and the adoption of eight nature-based climate solutions. We combined forest inventory data with over three decades of remote sensing and in situ data from a replicated land use change experiment. Deforestation of the region caused net global warming (1626 ± 44 µW m-2), mainly from the 76 % reduction of ecosystem carbon stocks, but also from the 84 % reduction of the soil methane sink and the 115 % increase in soil nitrous oxide emissions. The associated albedo increase offset 24 % of the greenhouse gas induced warming. For the adoption of nature-based climate solutions, we found that conservation agriculture provided a modest -39 to -76 ± 31 µW m-2 of climate mitigation, short/medium length forestry rotations provided more at -296 to -881 ± 44 µW m-2, and natural forest regeneration provided the most at -1555 ± 44 µW m-2. As the impacts of climate change on nature and society intensify, consideration should be given to the climate mitigation, habitat, and ecosystem services that nature-based climate solutions can provide. 

ABSTRACT Switchgrass (Panicum virgatumL.) is a native North American grass currently considered a high‐potential bioenergy feedstock crop. However, previous reports questioned its effectiveness in generating soil organic carbon (SOC) gains, with resultant uncertainty regarding the monoculture switchgrass's impact on the environmental sustainability of bioenergy agriculture. We hypothesize that the inconsistencies in past SOC accrual results might be due, in part, to differences in prior land management among the systems subsequently planted to switchgrass. To test this hypothesis, we measured SOC and other soil properties, root biomass, and switchgrass growth in an experimental site with a 30‐year history of contrasting tillage and N‐fertilization treatments, 7 years after switchgrass establishment. We determined switchgrass' monthly gross primary production (GPP) for six consecutive years and conducted deep soil sampling. Nitrogen fertilization expectedly stimulated switchgrass growth; however, a tendency for better plant growth was also observed under unfertilized settings in the former no‐till soil. In topsoil, SOC significantly increased from 2007 to 2023 in fertilized treatments of both tillage histories, with the greatest increase observed in fertilized no‐till. Fertilized no‐till also had the highest particulate organic matter content in the topsoil, with no differences among the treatments observed in deeper soil layers. However, regardless of fertilization, the tillage history had a strong effect on stratification with depth of SOC, total N, and microbial biomass C. Results suggested that historic and ongoing N fertilization had a substantial impact on switchgrass growth and soil characteristics, while tillage legacy had a much weaker, but still discernible, effect. 

Abstract The transition from conventional to more regenerative cropping systems can be economically risky due to variable transition period yields and unforeseen costs. We compared yields and economic returns for the first 3 years of the transition from a business as usual (BAU) conventional corn (Zea mays)–soybean (Glycine max) rotation to an aspirational (ASP) five‐crop (corn‐soybean‐winter wheat [Triticum aestivum]–winter canola [Brassica napus]‐forage) rotation in the Upper Midwest United States. Regenerative ASP cropping practices included the more diverse crop rotation, continuous no‐till, cover crops, precision inputs, and livestock (compost) integration. For the first two transition years, BAU corn yields were 8%–12% higher than ASP while in the third transition year, BAU corn yields were 5% lower. Soybean yields were similar for the first 2 years but higher in BAU in the third year due to an ASP pest outbreak. Equivalent yields for other ASP crops were lower than BAU in the first 2 years but similar in the third year except for canola, which suffered from slug damage. Whole‐system economic returns narrowed across years; by year three, whole system comparisons for the ASP corn and soybean entry points (corn‐soybean‐wheat and soybean‐wheat‐canola, respectively) showed equivalent economic returns for BAU and ASP, despite yield differences, owing largely to the ASP system's reduced operational costs. Overall findings suggest that early regenerative systems can be as profitable as conventional systems with careful attention to rotation entry points and inputs. 

Abstract Model projections predict increasing temperatures and precipitation change in many locations in the Central United States. To provide perspective on what these trends might bring relative to what has already happened, we compared historical temperature and precipitation change with what models from the Coupled Model Intercomparison Project (CMIP6) predict. The analysis focuses on regions represented by five long‐term agroecosystem research sites along a latitudinal transect from Michigan to Iowa, Missouri, Oklahoma, and Mississippi. We analyzed trends in long‐term records (≥50 years) of precipitation and temperature data at annual and monthly scales using indicators that characterize extreme and average temperature and rainfall amounts. Results show that temperatures have changed from 1900 to 2020, more for minimum (0.1°C–0.3°C decade⁻¹) than maximum (−0.1°C–0.2°C decade⁻¹), more for winter (−0.1°C–0.3°C decade⁻¹) than summer (−0.1°C–0.1°C decade⁻¹), and more often in the north than in the south. Except in Mississippi, annual precipitation has increased at rates of 25 mm decade⁻¹or greater over 1950–2020, but monthly trends were inconsistent. Projected trends suggest continued temperature increases, highlighting the urgent need for research on management systems that are resilient to such increases. 

Abstract Nitrogen (N) supply from cover crops to subsequent crop primarily depends on cover crop biomass production. Questions remain on how cover crop biomass interacts with abiotic factors to affect soil inorganic N and when N availability is highest following cover crop termination. This study identified key variables influencing N mineralization dynamics in cover crop systems, including air temperature, precipitation, gravimetric water content, cover crop biomass, weed biomass, and days after cover crop termination (DAT). Using random forest modeling with leave‐one‐year‐out cross‐validation, we analyzed 30 years (1990–2020) of bi‐weekly soil N measurements in two corn (Zea maysL.)‐soybean (Glycine max)‐wheat (Triticum aestivumL.) rotations with a legume cover crop (Trifolium pratenseL.) to identify key drivers of soil inorganic N release. Models explained 35% of variability in soil NO₃⁻‐N and 15%–32% variability in soil NH₄⁺‐N in the two systems. Variable importance analysis revealed that DAT was the most important driver affecting soil inorganic N availability, with air temperature as a close second. Partial dependence plots showed that soil NO₃⁻‐N increased rapidly following cover crop termination and peaked at approximately 50 DAT. Two‐dimensional partial dependence plots revealed interactions among DAT, temperature, and cover crop biomass in affecting soil NO₃⁻‐N. Temperature >12°C and cover crop biomass above 4000 kg ha⁻¹were associated with high soil NO₃⁻‐N levels. There were productivity differences between the management systems studied, yet both systems showed similar N dynamics, suggesting this approach was robust for understanding underlying drivers concerning N mineralization. 

Abstract Phosphorus (P) budgets for cropping systems provide insights for keeping soil P at optimal levels for crops while avoiding excess inputs. We quantified 12 years of P inputs (fertilizer and atmospheric deposition) and outputs (harvest and leaching losses) for replicated maize (Zea maysL.)—soybean (Glycine maxL.)—wheat (Triticum aestivum) crop rotations under conventional, no‐till, reduced input, and biologically based (organic without compost or manure) management systems at the Kellogg Biological Station LTAR site in southwest Michigan. Conventional, no‐till, and reduced input systems were fertilized between 13 and 50 kg P ha⁻¹depending on year. Soil test phosphorus (STP) was measured at 0‐ to 25‐cm depth every autumn. Leached P was measured as dissolved P in the soil solution sampled beneath the rooting depth and combined with modeled percolation. Fertilization and harvest were the predominant P fluxes in the fertilized systems, whereas only harvest dominated P flux in the unfertilized organic system. Leaching losses were minor terms in the budgets, but leachate concentrations were nevertheless close to the range of concern for downstream eutrophication. Over the 12‐year study period, the organic system exhibited a negative P balance (−82.0 kg P ha⁻¹), coinciding with suboptimal STP levels, suggesting a need for P supplementation. In contrast, the fertilized systems showed positive P balances (mean: 70.1 kg P ha⁻¹) with STP levels well above agronomic optima. Results underscore the importance of tailored P management strategies to sustain crop productivity while mitigating environmental impacts.