Effects of treeshelter on seedling performance: a meta-analysis

Abstract

Treeshelters are used in a variety of environments for forestry and forest restoration purposes. Although their use is often effective, in some cases treeshelters are ineffective. To clarify the effects of treeshelters on seedling performance, a systematic literature review and meta-analysis was conducted. Of the 296 studies related to treeshelters, 139 studies were suitable for extraction of data on mortality rate, browsing damage rate, height growth, or diameter growth. The quality of the literature did not affect the results of the meta-analysis. Mortality, browsing damage, and height growth were positively affected by treeshelters in many subgroup analyses, but diameter growth did not conform with the perceived inhibitory effect of treeshelters. Regarding conifer species, treeshelters had no effect on mortality rate and non-ventilated treeshelters promoted diameter growth. However, many subgroup analyses showed publication bias as indicated by the significantly fewer negative-effect studies. Therefore, the effects of treeshelters may be overestimated although the effects revealed by the meta-analysis were reasonable from the viewpoint of plant physiological reactions. Heterogeneity was high in most of the subgroups, suggesting that factors other than the analyzed subgroups may contribute to the variance in treeshelter effects. Studies on treeshelters are concentrated under specific conditions, such as on species of the genus Quercus and a temperate climate. Thus, other under-represented conditions are considered to constitute research gaps. Further studies in these research gap areas are required to perform a meta-analysis with low heterogeneity.