Please use this identifier to cite or link to this item: http://repositorio.ufla.br/jspui/handle/1/58319
Title: Multi-objective forest harvesting under sustainable and economic principles
Keywords: Amazon rainforest management
Computational intelligence
Multi-objective functions
Evolutionary computing
Issue Date: 2023
Publisher: Springer
Citation: LACERDA, T. H. S. et al. Multi-objective forest harvesting under sustainable and economic principles. Journal of Forestry Research, [S.l.], v. 34, p. 1379-1394, 2023.
Abstract: Selective logging is well-recognized as an effective practice in sustainable forest management. However, the ecological efficiency or resilience of the residual stand is often in doubt. Recovery time depends on operational variables, diversity, and forest structure. Selective logging is excellent but is open to changes. This may be resolved by mathematical programming and this study integrates the economic-ecological aspects in multi-objective function by applying two evolutionary algorithms. The function maximizes remaining stand diversity, merchantable logs, and the inverse of distance between trees for harvesting and log landings points. The Brazilian rainforest database (566 trees) was used to simulate our 216-ha model. The log landing design has a maximum volume limit of 500 m3. The nondominated sorting genetic algorithm was applied to solve the main optimization problem. In parallel, a sub-problem (p-facility allocation) was solved for landing allocation by a genetic algorithm. Pareto frontier analysis was applied to distinguish the gradients α-economic, β-ecological, and γ-equilibrium. As expected, the solutions have high diameter changes in the residual stand (average removal of approximately 16 m3 ha−1). All solutions showed a grouping of trees selected for harvesting, although there was no formation of large clearings (percentage of canopy removal < 7%, with an average of 2.5 ind ha−1). There were no differences in floristic composition by preferentially selecting species with greater frequency in the initial stand for harvesting. This implies a lower impact on the demographic rates of the remaining stand. The methodology should support projects of reduced impact logging by using spatial-diversity information to guide better practices in tropical forests.
URI: https://link.springer.com/article/10.1007/s11676-023-01614-5
http://repositorio.ufla.br/jspui/handle/1/58319
Appears in Collections:DEG - Artigos publicados em periódicos

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.