Introduction
Water availability drives the distribution and dynamics of plant species, even within relatively wet biomes (Engelbrecht et al.2007, Comita & Engelbrecht 2009, Esquivel-Muelbert et al. 2019, Krishnadas et al. 2021). Quantifying plant response to water stress is imperative given that global environmental change will alter patterns of precipitation and seasonal drought. For closed canopy forests, much attention has focused on the response of mature trees to drought (Suresh et al. 2010, Anderegg et al. 2015, Rowlandet al. 2015, Hartmann et al. 2018), but less is known of seedling responses. Seedlings can be highly vulnerable to drought because their shallow roots do not allow access to deeper water layers during dry periods. Moreover, the light-limited forest understory may impose carbon starvation during drought (Slot & Poorter 2007, Amissahet al. 2015)⁠ such that even short dry spells can result in substantial seedling mortality in tropical wet forests (Engelbrechtet al. 2007). Seedling responses to water availability can therefore regulate regeneration and persistence of species in a changing climate and shape community structure of later life-stages (Comita & Engelbrecht 2009, 2014)⁠.
Interspecific variation in drought response can be mediated by species’ traits—anatomical, morphological and physiological characteristics associated with differences in strategies of resource use (Slot & Poorter 2007, Poorter & Markesteijn 2008)⁠. For example, wood density and leaf mass per area can correlate positively with drought tolerance of mature trees (Muscarella et al. 2016, O’Brien et al.2017), but the extent to which these relationships hold true for seedlings is not known. Notably, the role of root traits in mediating species’ drought responses remains poorly explored, despite increasing calls to include below ground traits in assessing plant functional strategies and whole-plant response to stressors (Freschet et al.2021, Laughlin et al. 2021, Weemstra et al. 2023)⁠. In addition, understanding variation in anatomical features of stomata and xylem, which help regulate water balance and response to dry conditions, may serve as stronger proxies for hydraulic function than commonly used traits such as specific leaf area (Bartlett et al. 2016)⁠.
Species of evergreen habit may respond to seasonal droughts either by avoiding or tolerating desiccation, with functionally distinct strategies arising from different trait combinations. Desiccation avoiders try to maintain water status through a drought. Due to higher metabolic demands, these species may be resource-acquisitive with their higher rates of photosynthesis and growth selecting for thinner, cheaper leaves, large stomata and xylem to move adequate water, and deep roots to maintain access to water (Chitra‐Tarak et al. 2021, Reich 2014). Alternatively, species may tolerate desiccation by decreasing photosynthesis to reduce water loss. This strategy impedes growth but reduces the chance of embolism or hydraulic failure (Anderegg et al. 2015, 2016, Adams et al. 2017, Skelton et al. 2017). To prevent carbon starvation, desiccation tolerant species should be resource-conservative, reflected in traits such as low specific leaf area (SLA), high leaf dry matter content (LDMC), smaller xylem and stomata, and relatively lower investment in below-ground tissue to forage for water compared to desiccation avoiders. Alternatively, traits may not be coordinated and different trait combinations may offer equivalent responses to drought.
Easily measurable so-called ‘soft’ traits may help categorize species as avoiding or tolerating desiccation based on an integrated whole-plant evaluation of photosynthesis, growth and survival, and offer a helpful adjunct to hydraulic traits (Bartlett et al. 2012, Sun et al. 2020). With that in mind, we assessed the relationship between traits and species’ performance in dry vs. well-watered conditions for seedlings of 16 tree species from a wet tropical forest region in a global biodiversity hotspot in southern India, the Western Ghats. Across the region, most seedlings establish during the monsoon, but then have to withstand dry periods varying from 3-9 months to persist. The species were chosen to represent a gradient in their affinity to seasonal climatic water deficit during the dry season (henceforth seasonality) based on a previous analysis of tree species distributions in the Western Ghats (Krishnadas et al. 2021)⁠. We asked:
Do species from less vs. more seasonal forests vary in their drought response?
We expected that species associated with greater seasonality would tolerate drought better than species from less seasonal forests, i.e., show smaller decline in survival and/or growth in drought vs. well-watered conditions.
How well do individual traits explain interspecific variation in drought response?
We expected that species with higher SLA, lower LDMC, higher root mass fraction (RMF) or deeper roots, larger stomata and wider xylem would maintain better function (photosynthesis and growth) early in drought, but survive less than species with lower SLA, higher LDMC, and lower RMF or shallow roots, smaller stomata, and narrower vessels.
Do distinct trait combinations characterize species that avoid or tolerate desiccation?
We expected that traits associated with conservative water-use or safe hydraulic strategy, e.g., thicker leaves, denser stem, smaller stomata and xylem, would correspond with responses characteristic of desiccation tolerance – decreased photosynthesis and growth. Species with thinner leaves and larger stomata or xylem would avoid desiccation by functioning at similar levels in drought compared to watered conditions. We also expected that species with higher SLA, lower LDMC, deeper roots and higher RMF would avoid desiccation. Species with lower SLA, higher LDMC, shallow roots and lower RMF were expected to tolerate desiccation.