Evaluation of total tree height subsampling strategies for estimating volume in loblolly pine plantations

Abstract

Assessing current conditions and projecting future forest productivity are primary objectives in designing forest inventory for forest management planning. Stand net volume, expressed as sum of all standing trees’ volume on a per-unit area basis, has been widely applied to quantify forest productivity. Total tree height is a principal predictor commonly used to estimate tree volume, weight and biomass. However, measuring tree heights in the field is time consuming and labor intensive. Measuring a small portion of trees on the sample plots, denoted as subsampling heights, is a relatively efficient alternative in many assessments of forest cover. In this study, sampling plans with different levels of subsampling intensity were evaluated. Four 37.63-ha loblolly pine plantations with varying degrees of spatial heterogeneity at four inventory entry points (ages 10, 15, 20 and 25) were simulated. The impact of tree height measurement error was assessed using data from 210 standing and felled trees measurements across southeastern US and incorporated in the simulations.Results indicate that sampling intensity, stand age, and spatial heterogeneity have greater influence on the reliability for total volume estimation than height subsampling intensity and measurement error. Precision of the total volume estimates decreased with increasing degrees of spatial heterogeneity and stand age. Adding sample plots is a more effective way to reduce error for total volume estimation than adding measurement trees on plots. The plantations and repeated samples simulated in this study provide important information about the behavior of total volume and related estimates under various scenarios, which provides guidance for ecologists and resource managers to design efficient subsampling strategies for height measurements.