Seed sourcing recommendations for forest restoration: impacts of tree isolation on progeny performance of cedar and mahogany in the neutropics
Date
2011
Authors
Navarro, C.
Cavers, S.
Lowe, A.
Editors
Montagnini, F.
Finney, C.
Finney, C.
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Book chapter
Citation
Restoring Degraded Landscapes with Native Species in Latin America, 2011 / Montagnini, F., Finney, C. (ed./s), pp.51-61
Statement of Responsibility
Carlos Navarro, Stephen Cavers, and Andrew Lowe
Conference Name
Abstract
Swietenia macrophylla King and Cedrela odorata Linnaeus are globally and locally important timber species; both have been overexploited, driving many populations to extinction. Extensive deforestation in Central America has led to widespread interest in efforts to restore forest and increase connectivity between remnant fragments. A series of progeny trials were evaluated for variation in performance due to 1) region of seed collection; 2) the forest context of seed source trees (forest, grassland or swamp); and 3) the level of isolation of the parent tree. Region of collection accounts for a large component of the variation in seedling growth, indicating the importance of local adaptation to soils, climate and pests; forest context was not significant in this study. Seed producers isolated by more than 500 meters from the nearest conspecific tree produce progeny that perform less well than those with closer pollen donors. Isolated seed sources, such as trees in pastures or agricultural fields, should be avoided. Where possible, progeny trials should also be reevaluated considering the isolation level of source trees. This will also provide support for efforts to prevent fragmentation and tree isolation, increase survival capacity of restored forests and avoid problems resulting from a dependence on isolated trees for restoration. © 2011 by Nova Science Publishers, Inc. All rights reserved.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright status unknown