Habitat Adaptation Mediates the Influence of Leaf Traits on Canopy Productivity: Evidence from a Tropical Freshwater Swamp Forest

Abstract

Functional traits offer generalizability to the prediction of ecosystem processes such as production, and community-weighted mean trait values are increasingly used for such predictions. However, the underlying causal direction between traits and ecosystem processes are often indirect and sometimes even tenuous. In this study, we aimed to uncover underlying causal mechanisms between traits, habitat adaptation and canopy productivity. We used canopy production data estimated from leaf litter traps, and trait and habitat association data obtained from 40 permanent vegetation plots in the Nee Soon catchment in Singapore, which contains a heterogeneous mix of freshwater swamp and dry-land tropical forests. Mean canopy production across the catchment was estimated to be 768 g m −2 year −1 , which is similar to other tropical dry-land forests in the region. Fortnightly per-basal-area canopy production was found to be consistently lower in swamp than non-swamp plots, and positively correlated with monthly mean temperature. Structural equation models fitted to data of canopy production, leaf traits, plot type (swamp versus non-swamp), basal areas and habitat adaptations of 69 tree speciesplot combinations suggested that tree species possessing leaf traits associated with more conservative resource acquisition strategies, viz., low specific leaf area, high leaf C:N ratio, and thicker leaves, are better adapted to stressful, waterlogged swamp conditions, but that this adaptation also reduces canopy-and likely total-net primary productivity. These observations suggest that the stressful conditions of waterlogged, anoxic swamp habitats significantly reduce the rate at which nutrients are cycled by communities found in such environments.

Leaf and twig traits predict habitat adaptation and demographic strategies in tropical freshwater swamp forest trees

“…We did not find this, but instead found that swamp‐associated species had higher H max and lower D H . More importantly, our findings reinforce a recent study (Lam et al ., 2022) that showed that leaf traits may be as important as twig traits in conferring waterlogging stress tolerance to tropical forest tree species. This is likely because leaves and roots act as the water exchange surfaces of trees (root traits are likely to be important predictors of waterlogging stress tolerance as well; Kozlowski, 1997; Yamauchi et al ., 2018).…”

“…low SLA, low leaf N, etc. ; Moor et al ., 2017; Mori et al ., 2019; Lam et al ., 2022) and xeromorphic traits (thick and sclerophyllous leaves with low transpiration rates; Lugo et al ., 1990). Many studies also find that tree species associated with wetland habitats have lower wood densities (Ordoñez et al ., 2010; Savage & Cavender‐Bares, 2012; McCoy‐Sulentic et al ., 2017; Moor et al ., 2017).…”

Leaf decomposition and flammability are largely decoupled across species in a tropical swamp forest despite sharing some predictive leaf functional traits

“…Several studies have identified biochemical traits, such as nutrient concentrations and stoichiometric ratios, as being the strongest predictors of leaf litter decomposition, with physical traits being relatively less important (reviewed in Krishna & Mohan, 2017). By contrast, we found that few biochemical traits were important predictors of decomposability in our tropical freshwater swamp forest, except leaves with a higher CNR that had a slightly lower decomposability, in line with a previous bulk litter decomposition study at the same site (Lam et al ., 2022). Taken together, our findings suggest that decomposition in swamp forests differs from nonwetland forests in that oxygen availability to microbes, rather than nutrients, is likely to be more important in limiting litter decomposition rates.…”

“…Leaf litter for decomposition experiments, flammability experiments and litter trait measurements was collected between July 2020 and January 2021 from 10 replicate 400 m 2 long-term vegetation plots (from a total of 40 plots) established in 2013 (described in Chong et al, 2021;Lam et al, 2021Lam et al, , 2022. Freshly fallen leaf litter was collected from 1 m long × 1 m wide litter traps suspended at least 1 m above the ground on a fortnightly basis (Lam et al, 2021).…”

“…As an example of the importance of tropical swamp forests, in Southeast Asia, 25 million hectares of fresh and peat swamps store half of the region's ecosystem carbon (Page et al ., 2011; Miettinen et al ., 2016) and have a vascular plant diversity that exceeds 1000 species (Posa et al ., 2011; Giesen et al ., 2018). In undisturbed tropical swamp forests, anoxic waterlogged conditions slow down decomposition rates (Sulistiyanto et al ., 2005; Yule & Gomez, 2009; Lam et al ., 2022) and water‐saturated soils enhance resistance to fire (Turetsky et al ., 2015; Page & Hooijer, 2016). Degradation of swamp forests following logging, drainage or cultivation, as seen extensively across Southeast Asia (Miettinen et al ., 2016), results in well‐drained, aerobic soil conditions that are more conductive to microbial and faunal decomposition and that are more likely to burn (Hooijer et al ., 2010; Cattau et al ., 2016; Mishra et al ., 2021).…”