4 Discussion
Increasing studies have showed the importance of ITV on the ecological process. However, the patterns and sources of ITV in themselves are still unclear (Cope et al., 2022). There are large ITVs in mangroves due to the complex environment but are less concerned on propagule which can determine the recruitment or/and distribution (Feller et al., 2010; Petersan & Bell, 2012). It is urgent to study the ITV of mangrove propagule at the background of global changes. We analyzed the relationship between biotic and abiotic factors and hypocotyl traits on a large scale for a typical viviparous mangrove species. We found that PC1 (FW, FL, TDmax and TDmin) was on the establishment dimension, and ITV of PC1 was mainly structured between populations which was directly shaped by climate. While RTD was mainly structured between hypocotyls, which was constrained by fitness tradeoff between dispersal and retention. Our study provides insights to the ITV mechanism of hypocotyl traits in mangrove plants.
4.1 Structured traits variation of mangrove hypocotyl
Pervious study showed that ITV in propagule size has been observed, especially in the genera Rhizophora , Bruguiera ,Kandeli and Ceriops (Tomlinson, 2016). Our results showed that variation of PC1 was significantly negatively correlated with latitude, similar to the results of a previous study (Yang et al., 2020). But there was an inverse relationship in a black mangrove study. The propagule of Aegiceras corniculatum was smallest in the northern (low latitude) population, while the largest propagule was in the southern population in Australian (Saenger & West, 2018). The opposite pattern of structured ITV was possibly due to the different strategies for improving fitness between true viviparous and crypto viviparous mangrove species. Moreover, the genetic analysis also suggested that differences evident in populations of A. corniculatum at the extremes of its range, are explained as evolutionary adaptations in response to the local physical, climatic or biological characteristics (Maguire et al., 2000). Therefore, ITV of PC1 sources from the local biotic and abiotic factors at different sites.
We found that climate (especially temperature) was the main driver of intra-specific variation in PC1. In general, temperature and precipitation have a negative relation with latitude (Hijmans et al., 2005), while oceanic factors such as salinity and tides have little to do with latitude. Obviously, higher temperature means that more materials can be accumulated in hypocotyl. MAP had little effect on PC1, probably because mangrove plants have adapted to tidal environments and are able to obtain fresh water from seawater (Parida & Jha, 2010).
Although oceanic factors were not found to be the main factors affecting the variation of hypocotyl traits, salinity did have a significant effect on PC1 variation. Salinity has long been considered an important factor limiting the growth and distribution of mangroves (Chen & Ye, 2014; Richards et al., 2021), and study also found that length and weight of propagules of mangrove was negatively correlated with the salinity (Alam et al., 2018). In addition, salt-stress is also a factor in the evolution of vivipary in mangrove plants. The hypocotyl of viviparous seeds contains a large amount of nutrients and water, which are needed for the growth of newly colonized seedlings, thus avoiding the vulnerable life stage (Joshi, 1933; Zhou et al., 2016), that also can keep its vitality even floating in the sea for server months. When salinity is too high, maternal tree will invest more energy to against with salt stress (Alam et al., 2018; Parida et al., 2004), which may reduce the investment on the hypocotyl and eventually lead to variation in its traits. Tide is an important factor to mangrove growth and hypocotyl dispersal process (Clarke et al., 2001; Duke et al., 1998; Van der Stocken et al., 2015; Zhang et al., 2021), however, the effect of tides on the variation of hypocotyl traits was marginal in our study. The possible reason is that previous studies have only considered the influence of tidal effects, rather than the combination of multi biotic and abiotic effects. When count on the climate, the effect of tides may be marginal. In addition, we suggest that PC1 is the variability of traits on the establishment axis (Saatkamp et al., 2019; Moles & Westoby, 2004), which is greatly affected by resource-based factors (such as temperature and nutrients etc. ), while tides may not be an important factor affecting hypocotyl mass.
4.2 Unstructured traits variation of mangrove hypocotyl
We noted that the shape index RTD did not vary with latitude, and it was also less affected by abiotic and biotic factors. There are three possible reasons for ITV of RTD was structured between individuals (hypocotyls): 1) those traits may be constrained by limited genetic variation (Bradshaw 1991). Studies have shown that the genetic diversity of mangrove populations is low (Richards et al., 2021; Ruan et al., 2013) and the complex matrix also limits gene flow between mangrove populations (De Ryck et al., 2012; Van der Stocken et al., 2015); 2) traits are related to plant fitness tradeoffs. Such as costs induce tradeoff, they give rise to covariation between dispersal and other life-history traits at different scales of organismal organization (Bonte et al., 2012). Dispersal event through investments in special morphologies, are facing the risk and opportunity for transfer and settlement; 3) low ITV may also be caused by very sharp contrasts in conditions in space and time (Holt & Gomulkiewicz, 2004), which induce the plant stress. This is exactly what happens in coastal ecosystems where mangroves grow. Plants evolving in stressful environments have less plasticity in making use of resources upon their increased availability (Quadros et al., 2021). However, even the low genetic diversity and stressful environments, there were still large ITV of mangroves (Tomlinson, 2016), or such as the ITV of PC1. Consequently, RTD was related to plant fitness tradeoffs.
Specifically, we inferred that RTD was correlated with dispersal/retention tradeoff, because the center of gravity was important to determine the position relative to the water surface while floating (Van der Stocken et al., 2019; Figs. 1B, 1C). Although dispersal can avoid competition with conspecific species, it also faces the risk of colonization or/and establishment in an unfavorable environment. On the contrary, retention means colonizing in a similar environment to the parent tree, but cannot avoid the between conspecific species and natural enemies (Stump & Comita, 2020). Not all mangrove species recolonize proportionately with the previous generation at the same site (Kamruzzaman et al., 2017). Therefore, mangroves may regulate the RTD to determine more dispersal or retention.K. obovata is the most widely distributed species of mangrove (Sheue et al., 2003; Wang et al., 2011), the RTD of K. obovata may be relatively small which is facilitated to dispersal. While other Rhizophoraceae spp., likeBruguiera sexangular , B. gymnorrhiza , Rhizophora stylosa etc. are thermophilic species (Wu et al., 2018), the RTD of those species may relatively large which can ensure the settlement/establishment of offspring (propagule) in a suitable place. Based on that strategy of tradeoff, mangroves can maintain a high-level fitness.
Many studies have emphasized the importance of maternal effects in the variation of offspring traits (Alam et al., 2018; Cochrane et al., 2015; Galloway, 2005), but the strength of maternal effect on ITV was minor in this study. As we only recorded the height and DBH of maternal plants, which reflecting the size and age, that may have less to do with morphology than with yield (Hangelbroek & Santamaria, 2004). Rather, heredity may be the major contributor to maternal effects. In general, genetic diversity is low in the natural mangrove population (Hodel et al., 2018; Richards et al., 2021), but researchers discovered that high epigenetic diversity can maintain the high plasticity of traits in mangrove (Mounger et al., 2021). That was possibly why the ITV of RTD was mainly contributed from within individuals. Due to the lack of dormant seed periods, true mangrove species do not have seed banks and are difficult to regenerate after anthropogenic disturbances or natural disasters (Nettel & Dodd, 2007). Therefore, the collection of germplasm resources and artificial cultivation of true-viviparous mangrove species are particularly important, especially under global climate change. Study showed that K. obovata had higher regenerative capacity than other viviparous species (Zhang et al., 2021), because K. obovata had higher dispersal capacity. We recommend that the RTD of hypocotyl should be considered in germplasm collection and restoration, thus conducive to the regeneration of the population. More studies should be combined those biotic and abiotic factors to help us to have a comprehensive understanding of the trait variation of mangroves and to better protect the mangroves.