Data-Driven Optimization of Moss Growth Under Variable Light Conditions: A Computational Framework for Ecological Restoration

This study presents a data-driven computational framework for optimizing moss cultivation under variable light environments to support ecological restoration and intelligent environmental management applications. Three moss species (Didymodon tectorum, Abietinella abietina, and Racomitrium canescens) were cultivated under five light spectra (red, blue, green, yellow, and white) and three photoperiod regimes (8 h, 16 h, and 24 h). A standardized experimental dataset comprising 45 treatment combinations was established to evaluate growth characteristics, photosynthetic pigment accumulation, and antioxidant defense responses. Results demonstrated that red light under a 24 h photoperiod produced the highest biomass performance, with Didymodon tectorum achieving a height of 5.45 mm and coverage of 85%. Blue light significantly enhanced photosynthetic activity, yielding a maximum chlorophyll A content of 0.228 mg g−1 FW and promoting proline accumulation associated with stress tolerance. Green light under an 8 h photoperiod generated the strongest antioxidant response, with superoxide dismutase activity reaching 28.817 U·mg−1. Based on the experimental dataset, a computational decision framework and light optimization algorithm were developed to identify optimal spectral–temporal combinations for growth enhancement and stress-resistance management. The proposed framework is compatible with LED-based spectral regulation systems, optical sensing networks, wireless environmental monitoring platforms, and intelligent ecological restoration infrastructures. By integrating biological response modeling with data-driven decision strategies, this work provides an engineering-oriented methodology for adaptive light control and smart cultivation systems in ecological restoration and sustainable environmental applications.