Sporadic temporal patterns of seed production are a challenge for the regeneration and restoration of longleaf pine, which is a keystone component of an endangered ecosystem in the southeastern United States. In this study, long-term data for longleaf pine cone production, collected at six sites across the southeastern region, was examined from the perspective of information entropy. Our results indicate that long-term monitoring data usefully reflects the information entropy and trajectory in cone production. The entropy of cone production for longleaf pine forests at all sites increased slowly through time. However, a slight decrease in entropy was also noted. Entropy across all sites remained within 1.28-1.77 until 2016. High linear correlation existed between entropy and log (time length). The maximum entropy model (MaxEnt) overestimated the information entropy at each site, although the dynamics were similar. Joint entropy among all sites might reflect an emergent pattern for entropy across the region. Our study provides an important approach for characterizing ecosystem dynamics by information flow in adaptive ecosystem management.
Chen, Xiongwen; Brockway, Dale G.; Guo, Qinfeng. 2016. Entropy dynamics in cone production of longleaf pine forests in the southeastern United States. Mathematical and Computational Forestry and Natural-Resource Sciences. 8(2): 11-15. 5 p.