Number of soil collars required to measure soil respiration in a 2 m×0.5 m plot:case studies of common and immature forest soils

Abstract

We examined the spatial variation in soil respiration and the number of required sampling points within 1 m2 plots. Twenty-four collars with 0.0065 m2 surface areas were placed in each plot at two Japanese forest sites: four plots at the Yamashiro forest, a deciduous forest on immature soil and four plots at the Kahoku forest, an evergreen forest on common forest soil. Almost all intra-plot coefficients of variation were 23-32%, with a maximum value of 43%. We concluded that the Yamashiro and Kahoku forest sites would require 40 and 30 sampling collars with 0.0065 m2 surface area when the acceptable error rate was 10%, and 10 and 7 sampling collars when the acceptable error rate was 20%, respectively. These values are similar to those reported by previous studies that focused on larger spatial scales.