The optimum samples size to estimate throughfall for individual Brant`s oaks (Quercus brantii) in Zagros forests

Abstract

Throughfall (TF) has a large spatial variability due to the heterogeneous canopy structure and variable rainfall patterns. In this study, the aim was to estimate the optimum number of collectors needed to obtain a mean cumulative TF value within certain error limits for five individual Brant’s oak trees (Quercus brantii) in the Zagros forests of Ilam in western Iran. Sixteen TF manual gauges were placed beneath the five selected tree canopies in eight geographic directions and the gross rainfall (GR) was measured by the mean of six homemade gauges for a period of 14 months in leaf-on condition. During this period, 23 rainfall events with cumulative depth of 257.4 mm were collected, of which an average TF rate of 68.9% reached the forest floor. An average of 70 (range between 41-138), 18 (range between 10-35), and 8 (range between 5-15) gauges would be required to estimate mean cumulative TF to within ±5, ±10, and ±15% percent errors at the 95% confidence level, respectively. Based on the results of this study, 16 gauges are sufficient to estimate mean cumulative TF with an error of 15% and a confidence interval of 95%. However, the number of gauges should be increased if a lower error rate for the mean cumulative TF estimations is required.