Long-term sub-mesophilic swine manure–carcass co-digestion under ammonia stress: Characterization of operating regimes using data-driven modeling

Anaerobic co-digestion of swine manure-carcasses provides a pathway for mortality management, waste stabilization, and renewable-energy recovery. This study evaluated swine manure–carcass co-digestion at 25 °C using six laboratory-scale sequencing batch reactors (20 L working volume) over 1115 days. Following an initial manure-only period, carcass loading rates (CLRs) ranged from 83.2 to 666 kg carcass/m 3 manure, with organic loading rates of 0.15–0.82 g COD t /L·d where available. Reactor performance was evaluated using methane production, pH, volatile fatty acids (VFAs), total ammonia nitrogen (TAN), and free ammonia nitrogen (FAN). At low-to-moderate CLR, reactors maintained pH of 7.2–8.1, methane concentrations of 60–70%, and an average apparent methane yield of 0.39 L CH 4 /g COD t added, potentially including residual biodegradable COD from previous cycles. At high CLR, propionate accumulated (∼1500–4000 mg/L) and TAN reached 7983 mg/L, with FAN exceeding 300 mg/L, coinciding with methane content below 55%. An artificial neural network (ANN), using physicochemical and operational parameters while excluding gas-derived variables, predicted VS- and COD t -based methane yields. Leave-one-digester-out validation demonstrated substantial predictive performance on held-out reactors (overall testing R 2 = 0.813; set-specific R 2 up to 0.867), including TAN >7500 mg/L and FAN >300 mg/L. SHapley Additive exPlanations (SHAP) identified regime-specific predictive contributions from butyrate, buffering indicators, and hydraulic retention time (HRT). The novelty lies in integrating a 1115-day sub-mesophilic co-digestion dataset spanning a wide CLR range with reactor-independent ANN validation and SHAP interpretation. This approach enabled characterization of empirical ammonia–VFA operating ranges and regime-dependent predictive contributors to methane yield, supporting data-informed evaluation of ammonia-rich swine residues.

Authors

Institutions

Publication Details

Journal
Biomass and Bioenergy
Published
2026-09-21
DOI
https://doi.org/10.1016/j.biombioe.2026.110072
Primary Topic
Anaerobic Digestion and Biogas Production
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

Long-term sub-mesophilic swine manure–carcass co-digestion under ammonia stress: Characterization of operating regimes using data-driven modeling

Rajinikanth Rajagopal, Aimaiti Aikeremu, Suman Adhikary
Biomass and Bioenergy
Anaerobic Digestion and Biogas Production
article

Long-term sub-mesophilic swine manure–carcass co-digestion under ammonia stress: Characterization of operating regimes using data-driven modeling

Rajinikanth Rajagopal, Aimaiti Aikeremu, Suman Adhikary
article en

Abstract

Anaerobic co-digestion of swine manure-carcasses provides a pathway for mortality management, waste stabilization, and renewable-energy recovery. This study evaluated swine manure–carcass co-digestion at 25 °C using six laboratory-scale sequencing batch reactors (20 L working volume) over 1115 days. Following an initial manure-only period, carcass loading rates (CLRs) ranged from 83.2 to 666 kg carcass/m 3 manure, with organic loading rates of 0.15–0.82 g COD t /L·d where available. Reactor performance was evaluated using methane production, pH, volatile fatty acids (VFAs), total ammonia nitrogen (TAN), and free ammonia nitrogen (FAN). At low-to-moderate CLR, reactors maintained pH of 7.2–8.1, methane concentrations of 60–70%, and an average apparent methane yield of 0.39 L CH 4 /g COD t added, potentially including residual biodegradable COD from previous cycles. At high CLR, propionate accumulated (∼1500–4000 mg/L) and TAN reached 7983 mg/L, with FAN exceeding 300 mg/L, coinciding with methane content below 55%. An artificial neural network (ANN), using physicochemical and operational parameters while excluding gas-derived variables, predicted VS- and COD t -based methane yields. Leave-one-digester-out validation demonstrated substantial predictive performance on held-out reactors (overall testing R 2 = 0.813; set-specific R 2 up to 0.867), including TAN >7500 mg/L and FAN >300 mg/L. SHapley Additive exPlanations (SHAP) identified regime-specific predictive contributions from butyrate, buffering indicators, and hydraulic retention time (HRT). The novelty lies in integrating a 1115-day sub-mesophilic co-digestion dataset spanning a wide CLR range with reactor-independent ANN validation and SHAP interpretation. This approach enabled characterization of empirical ammonia–VFA operating ranges and regime-dependent predictive contributors to methane yield, supporting data-informed evaluation of ammonia-rich swine residues.

Biomass and BioenergyVol. 217
Agriculture and Agri-Food Canada (CA)
Zero hunger
Openalex Percentile: Top 15%
Anaerobic Digestion and Biogas Production
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Long-term sub-mesophilic swine manure–carcass co-digestion under ammonia stress: Characterization of operating regimes using data-driven modeling — Rajinikanth Rajagopal, Aimaiti Aikeremu, et al. · Biomass and Bioenergy (2026) | TGRS Research Map | TGRS