Skip to main content
Doctoral thesis, 2017

Stand dynamics in undisturbed and selectively logged tropical rainforest over 18 years

Lussetti, Daniel


Rainforests in Southeast Asia are enormously important, both ecologically and socio-economically, but their sustainability is severely threatened by various forms of disturbance that perturb the dynamics of forests here and elsewhere. To help efforts to counter such threats, this thesis focuses on two of the most important forms of disturbance that currently affect tropical rainforests: logging and extreme climatic events. It is based on research conducted in the Gunung Rara Forest Reserve, in Sabah, Malaysia (approx. 4°33´N, 117°02´E). The project started in 1992 in a mixed dipterocarp lowland rainforest, involving establishment of a long-term experiment with a 2 x 2 factorial randomized block design, where slope was the blocking factor. Two selective logging methods, supervised logging (SL) and conventional logging (CL), both with (CC) and without (NCC) pre-harvest climber cutting, and a control treatment (C), were applied to evaluate their effects on key parameters of forest dynamics, including: ingrowth of new trees (≥ 10 cm diameter at breast height, DBH), growth of standing trees (survivor growth), tree mortality, and overall development of the stands. Analysis of data covering 18 years of subsequent stand development shows that SL with CC resulted in consistently higher stand volume recovery and ingrowth of dipterocarps, while suppressing establishment of pioneer species by up to 40%, compared to either CL treatment, especially at high harvest intensities In addition, CC reduced overall mortality, suggesting that it can effectively reduce negative effects of harvest operations. Using linear mixed effect models, I also found that the stand level growth rate of dipterocarps was two times higher following SL with CC than following the CL treatment. Furthermore, diversity patterns of naturally regenerating trees (≥ 10 cm DBH) were studied over 10 years after selective logging, using the gradient of harvest intensity (0-16 trees ha-1). Species richness peaked at intermediate levels of extraction, corroborating the intermediate disturbance hypothesis (IDH), and suggesting that moderate levels (< 8 trees ha-1) of harvesting may maintain tree biodiversity. Finally, I studied the effects of El Niño Southern Oscillation events on uncut forests and found that both the intensity and duration of these events strongly influenced the forest dynamics. However, long periods of moderate-intensity El Niño-associated drought had stronger effects on tree mortality and subsequent growth of survivor trees than peak intensities. The results also indicate that cumulative effects of multiple events, rather single events, should be considered. However, although fluctuations in basal area were observed over 19 years of repeated events, the findings suggest that the uncut tropical rainforest was quite resilient and has fairly good capacity to withstand repeated extreme climatic fluctuations


ENSO events, Forest dynamics, RIL, Sustainable forest management, growth models, logging damage, IDH

Published in

Acta Universitatis Agriculturae Sueciae
2017, number: 2017:45
ISBN: 978-91-576-8863-7, eISBN: 978-91-576-8864-4
Publisher: Deparment of Forest Ecology and Management

Authors' information

Swedish University of Agricultural Sciences, Department of Forest Ecology and Management

UKÄ Subject classification

Agricultural Science
Forest Science

URI (permanent link to this page)