Site productivity estimation using height-diameter relationships in Costa Rican secondary forests

Abstract

Site quality was assessed in four secondary forests dominated by the commercially important tree species Vochysia ferruginea in Northern Costa Rica. A variation of the site quality concept known as site form (i.e. the expected mean height at 25 cm dbh) was used. Twenty-nine sample plots (20 × 20 m) were installed covering soil and topographic variation within and between sites. Site form was estimated for each plot using a linear model. The site form estimations show a high variation between plots (from 16.2 m to 32.2 m), which could indicate large variation in site productivity. Site form of V. ferruginea was significantly correlated with total stand basal area of all species, which supports the potential value of site form as an indicator of site productivity. Twenty-six physical and chemical soil variables showed significant correlation with site form. Adjusted multiple regression models of site form on soil variables were fitted by stepwise regression analysis (forward selection method). The best fit was obtained using only the percentage of sand at 12-30 cm as the independent variable (r2 = 59.3%, s.e.= 3.0 m). The results of this study suggest that site form appears to be a potentially useful indicator of site quality in broadleaved neotropical secondary rainforests when age is unknown.