Attention:The NSF Public Access Repository (PAR) system and access will be unavailable from 5:00 PM ET until 8:00 PM ET on Friday, September 11 due to maintenance. We apologize for the inconvenience.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


This content will become publicly available on May 1, 2027

Title: Predicting Biochar‐Induced Changes in Soil Organic Carbon With Ensemble Machine Learning
ABSTRACT Biochar is a promising soil amendment for enhancing soil organic carbon (SOC), but accurately predicting its effect under diverse environmental conditions remains challenging due to complex, nonlinear interactions among biochar properties, soil characteristics, climate, and management practices. To address this research gap, we developed an ensemble machine learning (ML) framework, combining Extremely Randomized Trees (ExtraTrees), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost) regressors, to model SOC responses to biochar application using a globally curated dataset of 800 field observations. The ensemble model showed strong predictive performance (R2 = 0.86, RMSE = 0.11) and generalized well across a wide range of conditions. Shapley Additive exPlanations (SHAP) analysis identified biochar addition rates, crop types, soil type, and soil pH were the most influential predictors of SOC changes. The most effective biochar application rate was about 40 t/ha, and the saturation point was 121.7 t/ha. Partial dependence plots revealed nonlinear and threshold effects of pyrolysis temperature, initial SOC levels, and nitrogen content. To facilitate practical application, we also developed a user‐friendly graphical interface for SOC prediction under various biochar‐soil‐climate scenarios. This work highlights the predictive power and interpretability of ML tools in digital soil carbon modeling and supports data‐driven strategies for optimizing biochar use in climate‐smart agriculture.  more » « less
Award ID(s):
2000058
PAR ID:
10699482
Author(s) / Creator(s):
 ;  ;  ;  ;  ;  
Publisher / Repository:
Wiley
Date Published:
Journal Name:
GCB Bioenergy
Volume:
18
Issue:
5
ISSN:
1757-1693
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, Biochar)
    Abstract Biochar is a promising climate-smart agriculture (CSA) solution to sustainably support food security while mitigating the adverse impacts of climate change on agroecosystems. Yet, its effectiveness across diverse environments is not well quantified. We developed a process-based biochar model and used it to evaluate biochar’s impacts on agroecosystem production and the dynamics of soil biogeochemical cycles (e.g., key CSA indicators such as crop yield, soil organic carbon (SOC), and greenhouse gas (GHG) emissions) across 48 globally distributed field experiment sites. The biochar model was calibrated and evaluated in maize, wheat, and soybean cropping systems, with an average root mean square error of 1878.9 kg ha−1(R2 = 0.78) for crop yield, 4129.3 kg C ha−1(R2 = 0.72) for SOC, and 1995.7 kg CO2 ha−1(R2 = 0.91) for GHG emissions. The model accuracy varied across environments, with yield predictions performing better in tropical (R2 = 0.90) and temperate (R2 = 0.81) zones and on medium-textured soils (R2 = 0.87), but declining in arid regions (R2 = 0.55) and on coarse soils (R2 = 0.65). Simulation accuracy of SOC and CO2was higher in maize than in soybean systems. Biochar application rates also influenced model performance, with medium rates best for crop yield and high rates optimal for SOC and CO2 emissions. These results highlight the need for robust modeling tools to optimize biochar application across diverse soil and climate conditions. These tools can be important for stakeholders, from farmers to policymakers, and can help refine biochar management strategies and advance global goals of sustainable intensification and net-zero agricultural systems. HighlightsA process-based biochar model was developed and calibrated using observational data from 48 global field studies.Model’s performance was evaluated across a range of climate conditions, soil types, cropping systems, and biochar application rates.The model serves as a robust tool for optimizing site-specific biochar application and advancing climate-smart agriculture. Graphical Abstract 
    more » « less
  2. ; ; ; ; (, Biochar)
    Abstract Biochar is well-accepted as a viable climate mitigation strategy to promote agricultural and environmental benefits such as soil carbon sequestration and crop productivity while reducing greenhouse gas emissions. However, its effects on soil microbial biomass carbon (SMBC) in field experiments have not yet been thoroughly explored. In this study, we collected 539 paired globally published observations to study the impacts of biochar on SMBC under field experiments. Our results suggested an overall positive impact of biochar (21.31%) on SMBC, varying widely with different climate conditions, soil types, biochar properties, and management practices. Biochar application exhibits significant impacts under climates with mean annual temperature (MAT) < 15 °C and mean annual precipitation (MAP) between 500 and 1000 mm. Soils of coarse and fine texture, alkaline pH (SPH), soil total organic carbon (STC) content up to 10 g/kg, soil total nitrogen (STN) content up to 1.5 g/kg, and low soil cation exchange capacity (SCEC) content of < 5 cmol/kg received higher positive effects of biochar application on SMBC. Biochar produced from crop residue, specifically from cotton and maize residue, at pyrolysis temperature (BTM) of < 400 °C, with a pH (BPH) between 8 and 9, low application rate (BAP) of < 10 t/ha, and high ash content (BASH) > 400 g/kg resulted in an increase in SMBC. Low biochar total carbon (BTC) and high total nitrogen (BTN) positively affect the SMBC. Repeated application significantly increased the SMBC by 50.11%, and fresh biochar in the soil (≤ 6 months) enhanced SMBC compared to the single application and aged biochar. Biochar applied with nitrogen fertilizer (up to 300 kg/ha) and manure/compost showed significant improvements in SMBC, but co-application with straw resulted in a slight negative impact on the SMBC. The best-fit gradient boosting machines model, which had the lowest root mean square error, demonstrated the relative importance of various factors on biochar effectiveness: biochar, soil, climate, and nitrogen applications at 46.2%, 38.1%, 8.3%, and 7.4%, respectively. Soil clay proportion, BAP, nitrogen application, and MAT were the most critical variables for biochar impacts on SMBC. The results showed that biochar efficiency varies significantly in different climatic conditions, soil environments, field management practices, biochar properties, and feedstock types. Our meta-analysis of field experiments provides the first quantitative review of biochar impacts on SMBC, demonstrating its potential for rehabilitating nutrient-deprived soils and promoting sustainable land management. To improve the efficiency of biochar amendment, we call for long-term field experiments to measure SMBC across diverse agroecosystems. Graphical Abstract 
    more » « less
  3. ; (, Agronomy Journal)
    Abstract Greenhouse gas (GHG) emissions reduction efforts are underway to mitigate climate change worldwide. Climate‐smart agriculture (CSA) practices have been shown to both increase soil organic carbon (SOC) inputs and reduce net greenhouse gas emissions (GHGnet). We evaluated the GHGnet of several management practices with three biogeochemical models (APSIM, Daycent, and RothC) at two sites with contrasting soils, climates, and cropping systems. Additionally, two future climate scenarios (baseline and high‐emissions) provided alternative outcomes of SOC, N2O, and CH4by 2050. In Michigan, most biochar and residue retention with no‐till treatments increased SOC stocks; leguminous cover crops, no‐till, and reducing fertilizer input lowered N2O emissions. The lowest biochar treatment lowered GHGnet in the baseline climate scenario, but all other management treatments increased GHGnet under both baseline and high emissions, and all management scenarios increased a mean of 8.0 Mg CO2‐equivalent GHG (CO2e) ha−1from baseline to high emissions. Conversely, in Texas, most treatments increased SOC, and N2O was relatively constant. Every no‐till treatment reversed GHGnet in both the baseline and high‐emissions climate scenarios but all management scenarios increased a mean of 0.6 Mg CO2e ha−1under high emissions. At both sites under high‐emissions climate change, cover crops and no‐till resulted in the lowest GHGnet overall. Overall, the study showed that no‐till, especially with residue retention, and cover crops are important CSA practices to lower the GHGnet of agriculture, but there remains much room to find even more effective solutions to adapt to climate change. 
    more » « less
  4. ; ; ; ; ; ; ; ; (, Proceedings of the National Academy of Sciences)
    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. 
    more » « less
  5. ; ; ; ; ; ; (, PLOS ONE)
    Peña-Fernández, Antonio (Ed.)
    Application of crop residues and biochar have been demonstrated to improve soil biological and chemical properties in agroecosystems. However, the integrated effect of organic amendments and hydrological cycles on soil health indicators are not well understood. In this study, we quantified the impact of hemp residue (HR), hemp biochar (HB), and hardwood biochar (HA) on five hydrolytic enzymes, soil microbial phospholipid (PLFA) community structure, pH, permanganate oxidizable carbon (POXC) soil organic carbon (SOC), and total nitrogen (TN). We compared two soil types, Piedmont and Coastal Plain soils of North Carolina, under (i) a 30-d moisture cycle maintained at 60% water-filled pore space (WFPS) (D-W1), followed by (ii) a 7-day alternate dry-wet cycle for 42 days (D-W2), or (iii) maintained at 60% WFPS for 42 days (D-W3) during an aerobic laboratory incubation. Results showed that HR and HB significantly increased the geometric mean enzyme activity by 1-2-fold in the Piedmont soil under the three moisture cycles and about 1.5-fold under D-W in the Coastal soil. In the presence of HA, the measured soil enzyme activities were significantly lower than control under the moisture cycles in both soil types. The shift in microbial community structure was distinct in the Coastal soil but not in the Piedmont soil. Under D-W2, HR and HB significantly increased POXC (600–700 mg POXC kg -1 soil) in the Coastal soil but not in the Piedmont soil while HA increased nitrate (8 mg kg -1 ) retention in the Coastal soil. The differences in amendment effect on pH SOC, TN, POXC, and nitrate were less distinct in the fine-textured Piedmont soil than the coarse-textured Coastal soil. Overall, the results indicate that, unlike HA, HR and HB will have beneficial effects on soil health and productivity, therefore potentially improving soil’s resilience to changing climate. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Text Encoded Conversion
  • XML
  • Save / Share this Record
  • Send to Email