Multidimensional modelling of compost quality integrating nutrient dynamics, energetics and pathogen inactivation

Abstract

• C/N ratio strongly inversely correlated with NRP and Available N indicates compost maturity • TES and decomposition rate showed perfect correlation linking energy storage to breakdown rate • Regression models showed a significant combined effect on C/N ratio • Higher Plastic Contamination Index reduced plastic contamination hinders nutrient release • CO₂ evolution rate decreased to 12 mg CO₂-C/g VS/day, confirming compost stabilization Accurate compost quality assessment requires robust tools that capture the complex biological, chemical, and energetic transformations that occur during decomposition. Unlike other studies that apply conventional single-indicator or purely descriptive approaches, this study applied a model-driven and chemometric framework to evaluate the quality of compost produced at 20 randomly selected sites in Ghana. Traditional single-parameter assessments often fail to reflect compost maturity and safety comprehensively; hence, predictive models were used to quantify nutrient availability, energy dynamics, microbial stability, and decomposition behaviour. Composts with lower carbon-to-nitrogen (C/N) ratios (mean: 10.32) exhibited significantly higher available nitrogen (up to 2.48%) and nutrient release potential (NRP), with a strong inverse correlation (r = –0.92). Total energy stored (TES) ranged from 12.5 to 20.8 MJ/kg, while microbial energy use (MEU) declined below 5 MJ/kg in mature samples, indicating biological stabilization. Decomposition rate constants (K) varied from 0.05/day to 0.16/day, with higher values linked to active microbial breakdown. Pathogen inactivation was substantial: faecal coliforms declined by 94.5% (117 to 6.5 cfu/g), and Escherichia coli and Salmonella spp. were completely eliminated, reducing the pathogenic index (PI) from 0.427 to 0.013 (p < 0.0001). Principal Component Regression (PCR) extracted five components explaining 81.97% of total variance, with strong predictive performance (R² = 0.82; CV R² = 0.75 ± 0.08). These findings affirm that model-based compost assessment provides a comprehensive and scalable approach for quality assurance, capturing subtle but critical dynamics often overlooked by conventional methods, and supporting certification, agronomic optimization, quality control, and integration into climate-smart agricultural systems.