Measuring Stem Growth Rates for Determining Age and Cohort Analysis of a Tropical Evergreen Tree

Abstract

Metrosideros polymorpha (Myrtaceae) is the dominant canopy tree in many Hawaiian lowland and montane rain forests. It is a shade-intolerant species that persists throughout forest succession. Stands usually regenerate following synchronized dieback of the canopy cohort. Like most tropical evergreen trees, Metrosideros does not form growth rings, making determination of tree age and stand turnover rates difficult. This study measured the annual stem diameter growth rate for 3 yr in cohorts of six different stem size classes on young volcanic substrates at 1100 m above sea level on the island of Hawai'i. These side-by-side cohorts were assumed to represent a chronosequence of stand development in early primary succession. Growth rates were used to predict mean cohort tree age based on mean tree diameter, adjusting for variation in growth rate during the life cycle of the trees. Mean annual growth rate was about 2 mm yc1 for all the cohorts except the largest, which was significantly lower. This cohort was undergoing stand dieback, with regeneration of a new cohort and is assumed to represent the terminal stage of the cohort life cycle. The predicted age of this cohort was about 200 years; this appears to be a reasonable estimate of the turnover rate for cohorts in this environment. Individual growth rates within cohorts were highly variable. Other parameters, such as crown area and nearest neighbor distances, could not account for the variation. Analysis indicates that the growth rate of each individual tree probably fluctuates about the mean growth rate throughout its life. Year-to-year variation in mean cohort growth rates was significant only for the two largest cohorts. For these large trees, mean growth rate was negatively associated with rainfall. It is suggested that these trees may be light limited, because solar radiation itself is known to be negatively correlated with annual rainfall in the study area.