%A Stephen McAuliffe, James D. Ackerman and Raymond L. Tremblay %T Land use legacy for a tropical myco-heterotroph: how spatial patterns of abundance, reproductive effort and success vary %0 Journal Article %D 2019 %J J Plant Ecol %R 10.1093/jpe/rty029 %P 367-375 %V 12 %N 2 %U {https://www.jpe.ac.cn/CN/abstract/article_59701.shtml} %8 2019-04-01 %X
Aims

Human land use such as agriculture and logging can have cascading effects on the environment and severely influence forest ecosystems by altering structure, species composition and community processes. These activities may have long-term consequences, which impact forest recovery. We investigated the legacy of historical anthropogenic land use on the current reproductive effort (RE) and success of the understory, myco-heterotrophic orchid, Wullschlaegelia calcarata in Puerto Rico’s tropical rain forest after 80 years of forest recovery.

Methods

Our study site was the 16-ha Luquillo Forest Dynamics Plot in the Luquillo Experimental Forest. We used six 10 m × 500 m transect lines that spanned areas with differing levels of historic canopy coverage which are correlated with land use history. We recorded the abundance of W. calcarata plants and measured shoot height, number of flowers, fruit set for all plants and seed set from the most mature, undehisced fruit on a random subset of plants measured. We sought to determine whether or not there is a legacy of land use history on the RE and success of W. calcarata. Of the varying degrees of historic disturbance, we predicted that RE and success would be highest in minimally disturbed old-growth forest, and that soil type differences would be insufficient to affect RE or success.

Important Findings

We found 1607 plants of W. calcarata, and only one was detected in the most historically disturbed area of the forest. The orchids were most abundant in the two least historically disturbed sites. However, the prevailing trend in all measures of RE is in the opposite direction with greater RE in the forest plots with intermediate levels of historical disturbance. Furthermore, the best model (as a function of AICc and weights) to predict RE is a combination of soil type and cover class. Nevertheless, our measures of reproductive success (fruit and seed set) were best in the least historically disturbed sites and were not associated with soil type. Thus, the best sites for growth are not always the same as those for abundance and reproduction, and after >80 years of recovery, components of the rainforest community have not fully recovered.