Shedding light on tree growth : ring analysis of juvenile tropical trees

In the understory of tropical forests light is believed to be the main limiting growth factor for the newly established trees. Trees growing in shade of the understory may experience periods of slow radial growth. It is expected that gaps created by tree or branch fall will provoke tree growth releases due to improved light levels. Such growth releases can be observed in the annual tree-rings of some tropical species. This PhD study aims to provide insights into the response of juvenile tropical trees to changes in light conditions, and to propose an alternative way to evaluate changes in growth rates due to variations in the light levels. As a first step, the rings of three Bolivian rainforest tree species (two shade-tolerant: Peltogyne cf. heterophylla and Clarisisa racemosa and one long-lived pioneer: Cedrelinga catenaeformis) were characterized anatomically. Then, the annual nature of juvenile tree-rings was proved: a chronology for each species was developed of which the date was verified by 14C bomb peak dating; also, a correlation between ring width and rainfall was determined for all species. Secondly, using core and disc samples, the diameter growth responses of juvenile individuals of the three species to gap formation were assessed. These, by the analysis of change in the 7 year-median diameter growth before and after logging gaps were formed. The distance to the gap was also considered in these analyses. Finally, based on information on growth responses of juvenile individuals of Peltogyne cf. heterophylla to gap formation, criteria for release identification were established. Improved light conditions after gap formation were observed for juvenile trees of the two shade-tolerant species, P.heterophylla and C.racemosa, while no effect was observed for the long-lived pioneer C.catenaeformis. A response of significant increase in diameter growth rates after gap formation was observed for P.heterophylla and C.catenaeformis, but not for C.racemosa. For P.heterophylla this response was detected closer to logging gaps: it was strong for large juvenile trees (20-25 cm dbh) and almost absent in small juveniles. For the long-lived pioneer species, differences in growth responses were not related to tree size or distance to gaps. Based on Peltogyne cf. heterophylla growth responses I established criteria for release identification. For the first time, empirical evidence of growth response to increased light levels was used to set a threshold of growth release for a tropical tree species. This empirical threshold takes into consideration the ontogenetic growth trend and size-dependent growth variations. Alternatively, I present a method of release detection that uses a double detrending approach i.e. it removes variation due to the ontogenetic trend and size-dependent growth. The results suggest that interpretation of growth releases in tree-ring series should be cautiously done as gap formation does not always lead to increased light levels, does not always invoke growth responses, and if growth responses are found, their strength depends on species and tree size