Search

In a unique phenomenon restricted to the ever wet forests of Southeast Asia, hundreds of species from dozens of plant families reproduce synchronously at irregular, multi-year intervals. The proximate environmental cues that synchronize these general flowering events have not been evaluated systematically because there have been no long-term, high temporal resolution, species-level records from the region. We present 13 years of weekly flowering records for five Shorea species as well as daily temperature and rainfall records from the Pasoh Forest Reserve, Peninsular Malaysia. We constructed models to evaluate hypothesized relationships between flowering and cool temperature, drought, and additive and synergistic effects of cool temperature and drought for each species. Model parameters include periods of time for floral cue accumulation and flower development and temperature and/or rainfall thresholds required for floral initiation. Parameters estimated using flowering observations from 2001 to 2011 were used to forecast flowering for 2011–2014. We show that drought and cool temperatures acting synergistically best explain the timing of flowering events for all Shorea species in the section Mutica and forecast the largest general flowering event accurately. Periods estimated for signal accumulation ranged from 54 to 90 days among species. Periods estimated for flowers to develop ranged between 43 and 96 days and closely followed the interspecific sequence of flowering in the Shorea species. Drought and temperature thresholds also varied among species, with Shorea maxwelliana requiring the most severe drought and Shorea leprosula the lowest temperatures. Synthesis. Our results indicate that cool temperatures and low rainfall occurring on seasonal time-scales of about 2–3 months rather than brief cold snaps or brief droughts best explain general flowering in Shorea species at the Pasoh Forest Reserve. Low rainfall is equally likely in winter (December–February) and summer (July–August) and cool temperatures are most likely in winter at Pasoh, which explains why general flowering events are restricted to spring and fall, with more frequent and stronger flowering in spring. In addition, species-specific sensitivity to environmental cues suggests that future climate change will have differential impacts on the frequency of reproduction, with potential consequences for regeneration of these dominant species of lowland tropical forests.

We examined relationships between mortality rate, relative growth rate (RGR), and spatial patterns of three growth stages (small, medium, and large trees) for 11 dipterocarp species in the Pasoh 50-ha plot. Mortality rates for these species tended to be positively correlated with RGRs, although the correlation was significant only at the small-tree stage. Seven species with high growth and mortality rates exhibited peaks in spatial aggregation at small distances (100 m) in small trees, but this aggregation disappeared in medium and large trees. In contrast, the other four species with low growth and mortality rates aggregated at large distances (200 m) throughout the three growth stages in all but one species. Negative associations between different growth stages were observed only for the high-mortality species, suggesting density-dependent mortality. The high-mortality species showed habitat associations with topography, soil type, and the forest regeneration phase after gap formation, whereas the three low-mortality species only had associations with the forest regeneration phase. A randomization procedure revealed that these habitat associations explained little of their spatial aggregation. Our results suggest that the growth strategy has a large effect on the structuring of the spatial distribution of tree species through mortality processes.