A three-step procedure in SAS to analyze the time series from automatic dendrometers

Continuous measurements of stem radius variation in trees are obtained with automatic dendrometers that provide time series composed of seasonal tree growth and circadian rhythms of water storage and depletion. Several variables can be extracted from the raw data, such as amplitude and duration of radius increase and contraction, which are useful for understanding intra-annual tree growth, tree physiology and for performing growth–climate relationships. These measurements constitute a large dataset whose manipulation needs numerous algorithms and automatic procedures to efficiently and rapidly extract the information. This paper presents a three-step procedure using two SAS routines to extract the time series describing radius variation and associate them with environmental parameters. The first routine organizes and corrects data and generates outputs in the form of files and plots to visualize the results and improve data correction (first step). The second step consists of a reclassification of the hours of contraction or expansion that have been misclassified by the automatic process. The second routine classifies the daily patterns of stem variation into the three phases of contraction, expansion and radius increment and associates the environmental parameters (third step). An example of the procedure is given, with an explanation of the outputs generated. The advantages and shortcomings of the procedure and its importance for the intra-annual analyses of tree growth are discussed.