The Influence of Sampling Method, Sample Size, and Frequency of Observations on Plant Phenological Patterns and Interpretation in Tropical Forest Trees

• Autor(a) principal: Morellato, L. Patricia C. [UNESP]
• Data de Publicação: 2010
• Outros Autores: Camargo, Maria Gabriela G. [UNESP], D'Eca Neves, Fernanda F. [UNESP], Luize, Bruno G. [UNESP], Mantovani, Adelar [UNESP], Hudson, Irene L., Hudson, I. L., Keatley, M. R.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1007/978-90-481-3335-2_5; http://hdl.handle.net/11449/245320
• topic: Frequency of observations; Phenological methods; Sample size; Tree phenology; Tropical trees

The research field of plant phenology, which often involves the monitoring of several to hundreds of species of different life forms and/or different vegetation types, has increased exponentially over the last three decades. This has occurred in general, without consideration of the comparability of data and patterns across areas, and its influence on the interpretation of resultant patterns. In this chapter we address the influence of sampling method, sample size and the frequency of observations on the analysis of tropical tree phenology. Our approach is to compare the results of direct observations on transects with those obtained from litter traps. Transects and litter traps are the two most common methods used to sample and monitor plant phenology. Data from 3 locations were used to simulate different sample sizes and frequencies, and results were then compared with the original data. We conclude that sample size influences the patterns observed and there is a clear trade off between sample size and the frequency of observations. We show that direct observations were more accurate in defining both the beginning and the peak of phenological phases, and there was a significant difference between the peaks and seasonal patterns detected by both sampling methods. For tropical tree forest applications we recommend a minimum sample size of 15 trees and that a fortnightly frequency of observation be used especially if the sample size is small. We advocate the combination of presence/absence data and a quantification method to estimate plant phenology, a careful application of indices and a cautious generalization of pattern.