Laurance, William F., Andrade, Ana S., Magrach, Ainhoa, Camargo, José L.C., Valsko, Jefferson J., Campbell, Mason, Fearnside, Philip M., Edwards, Will, Lovejoy, Thomas E., and Laurance, Susan G. (2014) Long-term changes in liana abundance and forest dynamics in undisturbed Amazonian forests. Ecology, 95 (6). pp. 1604-1611.

PDF (Published Version) - Published Version Restricted to Repository staff only

 

83

2

Abstract

Lianas (climbing woody vines) are important structural parasites of tropical trees and may be increasing in abundance in response to global-change drivers. We assessed long-term (∼14-year) changes in liana abundance and forest dynamics within 36 1-ha permanent plots spanning ∼600 km2 of undisturbed rainforest in central Amazonia. Within each plot, we counted each liana stem (≥2 cm diameter) and measured its diameter at 1.3 m height, and then used these data to estimate liana aboveground biomass. An initial liana survey was completed in 1997–1999 and then repeated in 2012, using identical methods.

Liana abundance in the plots increased by an average of 1.00% ± 0.88% per year, leading to a highly significant (t = 6.58, df = 35, P < 0.00001) increase in liana stem numbers. Liana biomass rose more slowly over time (0.32% ± 1.37% per year) and the mean difference between the two sampling intervals was nonsignificant (t = 1.46, df = 35, P = 0.15; paired t tests). Liana size distributions shifted significantly (χ2 = 191, df = 8, P < 0.0001; Chi-square test for independence) between censuses, mainly as a result of a nearly 40% increase in the number of smaller (2–3 cm diameter) lianas, suggesting that lianas recruited rapidly during the study.

We used long-term data on rainfall and forest dynamics from our study site to test hypotheses about potential drivers of change in liana communities. Lianas generally increase with rainfall seasonality, but we found no significant trends over time (1997–2012) in five rainfall parameters (total annual rainfall, dry-season rainfall, wet-season rainfall, number of very dry months, CV of monthly rainfall). However, rates of tree mortality and recruitment have increased significantly over time in our plots, and general linear mixed-effect models suggested that lianas were more abundant at sites with higher tree mortality and flatter topography. Rising concentrations of atmospheric CO2, which may stimulate liana growth, might also have promoted liana increases.

Our findings clearly support the view that lianas are increasing in abundance in old-growth tropical forests, possibly in response to accelerating forest dynamics and rising CO2 concentrations. The aboveground biomass of trees was lowest in plots with abundant lianas, suggesting that lianas could reduce forest carbon storage and potentially alter forest dynamics if they continue to proliferate.

Actions (Repository Staff Only)

Item Control Page