Derek C. Manheim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Derek C. Manheim has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Nazli Yesiller
California Polytechnic State Univ
5 shared publications - James L. Hanson
California Polytechnic State University
4 shared publications Paul Imhoff
Civil, Construction and Environmental Engineering, University of Delaware, 344A DuPont Hall Newark, Delaware 19716, USA
3 shared publications- Ramin Yazdani
Air Quality Research Center, University of California at Davis, One Shields Avenue, Davis, CA 95616, USA
3 shared publications - Sajjad Karimi
Air Quality Research Center, University of California at Davis, One Shields Avenue, Davis, CA 95616, USA; School of Civil Engineering, Iran University of Science and Technology, Narmak, Tehran 16846-13114, Iran
3 shared publications Fatemeh Nayebloie
2 shared publications
Latest Publications
- Open Access
- 10 July 2026
- Published by Elsevier in SSRN Electronic Journal
Machine learning predictions of methane generation from dry thermophilic anaerobic digestion of organic co-substrates
Dry thermophilic anaerobic digestion (DTAD) of organic solid wastes plays a critical role in climate-friendly waste management. Conventional biochemical methane potential (BMP) tests, often conducted to support digestion system design and operation, are often time-consuming and laborious, thereby limiting screening to a few candidate organic feedstocks. Machine learning (ML) models, when trained and validated on experimental BMP data, can overcome many technical challenges by establishing an automated, reliable approach to prediction of methane generation. This investigation combines standardized laboratory BMP data collected using samples from a full-scale DTAD facility with ML models to predict methane generation from a diverse range in co-substrates, including: cooking oil solids, grease trap waste, distillery mash, glycerin, and strawberry syrup. A combined dataset of 452 samples was used to train and test nine regression models (CatBoost, Gradient Boosting, LightGBM, Random Forest, XGBoost, K-Nearest Neighbors, Support Vector Regression, Neural Network and Linear Regression) with substrate type, digestion time, and organic matter content as key predictor variables. Model performance was evaluated using 3-fold cross-validation and the metrics mean absolute error (MAE), root mean squared error (RMSE) and coefficient of determination (R²). Tree-based ensemble models substantially outperformed linear and kernel-based approaches, with CatBoost achieving the lowest prediction error for cumulative methane generation. The results demonstrate that reliable ML models can be derived from relatively small and noisy laboratory BMP datasets using practical input features, consistent with recent reviews of ML in anaerobic digestion. These ML models provide a rapid tool to assess methane generation potential of alternative co-substrates.
- 10 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMachine learning predictions of methane generation from dry thermophilic anaerobic digestion of organic co-substrates
Dry thermophilic anaerobic digestion (DTAD) of organic solid wastes plays a critical role in climate-friendly waste management. Conventional biochemical methane potential (BMP) tests, often conducted to support digestion system design and operation, are often time-consuming and laborious, thereby limiting screening to a few candidate organic feedstocks. Machine learning (ML) models, when trained and validated on experimental BMP data, can overcome many technical challenges by establishing an automated, reliable approach to prediction of methane generation. This investigation combines standardized laboratory BMP data collected using samples from a full-scale DTAD facility with ML models to predict methane generation from a diverse range in co-substrates, including: cooking oil solids, grease trap waste, distillery mash, glycerin, and strawberry syrup. A combined dataset of 452 samples was used to train and test nine regression models (CatBoost, Gradient Boosting, LightGBM, Random Forest, XGBoost, K-Nearest Neighbors, Support Vector Regression, Neural Network and Linear Regression) with substrate type, digestion time, and organic matter content as key predictor variables. Model performance was evaluated using 3-fold cross-validation and the metrics mean absolute error (MAE), root mean squared error (RMSE) and coefficient of determination (R²). Tree-based ensemble models substantially outperformed linear and kernel-based approaches, with CatBoost achieving the lowest prediction error for cumulative methane generation. The results demonstrate that reliable ML models can be derived from relatively small and noisy laboratory BMP datasets using practical input features, consistent with recent reviews of ML in anaerobic digestion. These ML models provide a rapid tool to assess methane generation potential of alternative co-substrates.
- 9 June 2026
- Published by Elsevier in Waste Management
- Vol. 222, 115618
Beyond the digester: facility-wide greenhouse gas and pollutant emissions and energy performance of hybrid anaerobic–aerobic organic waste composting
- 9 June 2026
- Published by Elsevier in Waste Management
- Vol. 222, 115618
Beyond the digester: facility-wide greenhouse gas and pollutant emissions and energy performance of hybrid anaerobic–aerobic organic waste composting
- Open Access
- 15 April 2026
- Published by Elsevier in Water Research
- Vol. 300, 125950
Multi-scale UAV, flux-chamber, and point-source measurement framework for facility-wide greenhouse gas and air pollutant emissions assessment from wastewater treatment operations
- 15 April 2026
- Published by Elsevier in Water Research
- Vol. 300, 125950
Open AccessMulti-scale UAV, flux-chamber, and point-source measurement framework for facility-wide greenhouse gas and air pollutant emissions assessment from wastewater treatment operations
- 17 March 2026
- Published by Elsevier in Waste Management
- Vol. 216, 115468
Facility-wide greenhouse gas and pollutant emissions from covered aerated static pile composting
- 17 March 2026
- Published by Elsevier in Waste Management
- Vol. 216, 115468
Facility-wide greenhouse gas and pollutant emissions from covered aerated static pile composting
- Open Access
- 1 July 2024
- Published by Elsevier in Waste Management
- Vol. 186, 318-330
Climate impacts of landfill gas emissions: Analysis for 20-year and 100-year time horizons
- 1 July 2024
- Published by Elsevier in Waste Management
- Vol. 186, 318-330
Open AccessClimate impacts of landfill gas emissions: Analysis for 20-year and 100-year time horizons
- 9 April 2024
- Published by Springer Nature
- p. 521-531
Management of Flood Debris: A Case History from Michigan, USA
- 9 April 2024
- Published by Springer Nature
- p. 521-531
Management of Flood Debris: A Case History from Michigan, USA
- Open Access
- 4 August 2023
- Published by Elsevier in Waste Management
- Vol. 169, 392-398
Disaster reconnaissance framework for sustainable post-disaster materials management
- 4 August 2023
- Published by Elsevier in Waste Management
- Vol. 169, 392-398
Open AccessDisaster reconnaissance framework for sustainable post-disaster materials management
- Open Access
- 8 February 2023
- Published by Elsevier in Soils and Foundations
- Vol. 63 (2)
Geoenvironmental assessment of climate impacts from landfill gas emissions
- 8 February 2023
- Published by Elsevier in Soils and Foundations
- Vol. 63 (2)
Open AccessGeoenvironmental assessment of climate impacts from landfill gas emissions
- Open Access
- 12 October 2022
- Published by Elsevier in Waste Management
- Vol. 154, 146-159
Assessment of methane emissions from a California landfill using concurrent experimental, inventory, and modeling approaches
- 12 October 2022
- Published by Elsevier in Waste Management
- Vol. 154, 146-159
Open AccessAssessment of methane emissions from a California landfill using concurrent experimental, inventory, and modeling approaches