Microclimatic responses to different thinning intensities in a Japanese cedar plantation of northern Taiwan

To assess the effects of thinning intensities on, and to reveal the temporal trends of, the short-term microclimates in a Japanese cedar (Cryptomeriajaponica D. Don) plantation in a montane region of northern Taiwan, we monitored air and soil temperatures, air and soil temperature ranges, soil water potential, and ambient air water pressure deficit within the plantation over a 2-year period. The thinning experiment consisted of three thinning intensities (weak, moderate, and strong) and a control (unthinned). We used a set of semiparametric smoothing spline and linear mixed-effects models to analyze the monthly averages of the six variables. The results showed that the temporal trends of the six variables mainly reflected the regional climatic patterns during the study period and the topographic effects. Averaged over the 2-year period, the strongly thinned stand had significantly higher air and soil temperatures than that for the other three treatments. All the thinned stands had higher average air temperature ranges than that for the control, and the average of the strongly thinned stand was significantly higher than that for the other two thinning treatments. The air temperature ranges of the thinned stands, however, returned almost to the pre-thinning stage at the end of the monitoring period. For soil temperature range, the strongly thinned stand had a significantly higher average than that for the other three stands. Structural heterogeneity created by thinning also led to greater month-to-month variations in air and soil temperature ranges for the thinned stands. As a combination of thinning and its aspect, the moderately thinned stand had the highest soil water potential, followed by the strongly thinned stand. The thinned stands all had higher average soil water potential than for the control in a severe summer drought that occurred during the study period. For water vapor pressure deficit, no significant thinning effect was detected, probably because the study site was located within the cloud belt of northern Taiwan. This study has demonstrated that we can understand better how thinning truly affects microclimates by separating the influences that are due to higher-level climatic factors from those due to thinning. Implications of the effects of different thinning intensities were also addressed. (c) 2007 Elsevier B.V. All rights reserved.