Dendrocalamus hamiltonii, a large, clump-forming bamboo, has great potential to contribute towards poverty alleviation efforts across its distributional range. Harvesting methods that maximize yield while they fulfill local objectives and ensure sustainability are a research priority. Documenting local ecological knowledge on the species and identifying local users' goals for its production, we defined three harvesting treatments (selective cut, horseshoe cut, clear cut) and experimentally compared them with a no-intervention control treatment in an action research framework. We implemented harvesting over three seasons and monitored annually and two years post-treatment. Even though the total number of culms positively influenced the number of shoots regenerated, a much stronger relationship was detected between the number of culms harvested and the number of shoots regenerated, indicating compensatory growth mechanisms to guide shoot regeneration. Shoot recruitment declined over time in all treatments as well as the control; however, there was no difference among harvest treatments. Culm recruitment declined with an increase in harvesting intensity. When univariately assessing the number of harvested culms and shoots, there were no differences among treatments. However, multivariate analyses simultaneously considering both variables showed that harvested output of shoots and culms was higher with clear cut and horseshoe cut as compared to selective cut. Given the ease of implementation and issues of work safety, users preferred the horseshoe cut, but the lack of sustainability of shoot production calls for investigating longer cutting cycles.
Keywords: Bamboo silviculture; Dendrocalamus hamiltonii; Local ecological knowledge; NTFP harvesting; Participatory action research.