Functional traits and the growth–mortality trade-off in tropical trees

WRIGHT, S.J; KITAJIMA, K.; KRAFT, N.B.; REICH PB; WRIGHT, IJ; BUNKER, D.E.; CONDIT, R.; DALLING; J.W; DAVIES, S.J.; DÍAZ, S. M.; ENGELBRECHT, B.M.J.; HARMS, K.E.; HUBBEL, S.P.; MARKS, C.O.; RUIZ- JAEN, M.C.; SALVADOR, C.M.; ZANNE, A.

ECOLOGY

ECOLOGICAL SOC AMER

Año: 2010 vol. 91 p. 3664 - 3674

0012-9658

Abstract. A trade-off between growth and mortality rates characterizes tree species in closedcanopy forests. This trade-off is maintained by inherent differences among species and spatialvariation in light availability caused by canopy-opening disturbances. We evaluated conditionsunder which the trade-off is expressed and relationships with four key functional traits for 103tree species from Barro Colorado Island, Panama. The trade-off is strongest for saplings forgrowth rates of the fastest growing individuals and mortality rates of the slowest growingindividuals (r2 ¼ 0.69), intermediate for saplings for average growth rates and overall mortalityrates (r2¼0.46), and much weaker for large trees (r2  0.10). This parallels likely levels of spatialvariation in light availability, which is greatest for fast- vs. slow-growing saplings and least forlarge trees with foliage in the forest canopy. Inherent attributes of species contributing to thetrade-off include abilities to disperse, acquire resources, grow rapidly, and tolerate shade andother stresses. There is growing interest in the possibility that functional traits might provideinsight into such ecological differences and a growing consensus that seed mass (SM), leaf massper area (LMA), wood density (WD), and maximum height (Hmax) are key traits among foresttrees. Seedmass, LMA,WD, andHmax are predicted to be small for light-demanding species withrapid growth and mortality and large for shade-tolerant species with slow growth and mortality.Six of these trait–demographic rate predictions were realized for saplings; however, with theexception of WD, the relationships were weak (r2 , 0.1 for three and r2 , 0.2 for five of the sixremaining relationships). The four traits together explained 43–44% of interspecific variation inspecies positions on the growth–mortality trade-off; however,WD alone accounted for .80% ofthe explained variation and, after WD was included, LMA and Hmax made insignificantcontributions. Virtually the full range of values of SM, LMA, and Hmax occurred at all positionson the growth–mortality trade-off. Although WD provides a promising start, a successful traitbasedecology of tropical forest trees will require consideration of additional traits.