Predicting Height Growth of Yellow‐poplar from Soils and Topography in Southeastern Ohio

Abstract

AbstractIntensive forest management requires an accurate knowledge of site productivity. Where inadequate stocking prohibits direct measurement of site quality, predictive models based on soils and topography may be used to estimate site potential. A predictive model was developed for estimating height growth of yellow‐poplar (Liriodendron tulipifera L.) on a Muskingum‐Neotoma‐Berks soils complex (Typic Dystrochrepts and Ultic Halpludalfs) in Tar Hollow State Forest, Ross County, Ohio. Measurements were made of selected topographic and soil physical and chemical factors for 203 dominant or codominant yellow‐poplars in 39 plots. Observed variation in site index within the mapping unit was 7.0 m. Multiple regression equations of the form: LOG(tree height) = C0 + B0 (1/AGE) + B1V1 + … + BnVN. were used to construct prediction equations. Significant correlations (p = 0.05 level) were found between tree height (with age accounted for), and the following variables; slope position; A horizon thickness; depth to the B2 horizon; depth to a restriction; corrected aspect (minor angle from the southwest); soil pH and lime test index; kg/ha of available P (p = 0.10 level) and Mn; kg/ha of exchangeable Ca, Mg, and K; percent base saturation; and soil organic matter. Approximately 65% of the observed variability in tree height was accounted for by the regressions.