On an intertidal scale, mangroves are commonly distributed in bands parallel to the tidal gradient, known as zonation (Duke et al., 1998; Wang et al., 2019). Zonation patterns of mangrove species have been a main research focus for several decades and have led to numerous hypotheses that attempt to explain mangrove zonation. There are four major explanations for mangrove zonation, including (1) propagule dispersal (Watson, 1928; Rabinowitz, 1978a, b; Bunt et al., 1985; McGuinness, 1997; Sousa et al., 2007; De Ryck et al., 2012; Crase et al., 2013; Van der Stocken et al., 2019a), (2) physiological adaptation (Ball, 1998, 2002), (3) animal predation (Smith, 1987a, b; Farnsworth, 1997) and (4) interspecific competition (McKee, 1995a). Mangrove zonation is the outcome of a number of processes driven by surface elevation and variables such as hydroperiod, soil salinity and soil physical-chemical characteristics (Crase et al., 2013; Tomlinson, 2016) The process of mangrove zonation is complex due to interaction of biotic and abiotic factors on the distribution and survival of mangrove seedlings. Therefore, no conclusion has yet satisfactorily explained the mechanism(s) that govern intertidal distribution of mangroves.

Propagule dispersal has been recognized as a main factor affecting population dynamics and community structure (Sousa et al., 2007; Van der Stocken et al., 2019a). Propagule dispersal is foundational to forest structure and dynamics, genetic diversity, and differentiation (Tonné et al., 2017). Additionally, understanding mangrove propagule dispersal is important for determining the potential for natural recovery of mangrove forest (e.g., natural regeneration of mangroves in abandoned ponds) and is critical to support management decisions for reforestation (Di Nitto et al., 2013; Van der Stocken et al., 2022).

Although the propagules of some mangrove species can be dispersed by wind or birds, most mangrove propagules are water-borne (Tomlinson, 2016). Interspecific differences in propagule dispersal patterns may contribute to mangrove zonation, a phenomenon commonly observed in mature mangrove communities (Rabinowitz, 1978b; McKee, 1995b; McGuinness, 1997). The Tidal Sorting Hypothesis (TSH), introduced by Rabinowitz (1978b), has been widely accepted as an explanation for this zonation. TSH hypothesizes that species with heavier propagules are more likely to be found in seaward areas, while species with lighter propagules are more common in landward areas. However, there are notable exceptions where TSH does not hold. For example, Avicennia marina and Sonneratia alba, both species with smaller propagules, are distributed in the lower intertidal zone, whereas species with larger propagules, such as Xylocarpus granatum, dominate the middle to upper intertidal zones (Smith, 1987a; Delgado et al., 2001). De Ryck et al. (2012) suggest that the number of propagules released, and their buoyancy can influence a species’ dispersal capacity. Despite these variations, Tomlinson (2016) maintains that propagule weight is a significant factor in determining zonation.

Propagule dispersal has been shown to be a main factor characterizing species distributions. While the reasons why TSH has not been supported in mangrove environments are unclear, there are marked species differences in dispersal and establishment strategies among mangrove plants (Robert et al., 2015). Ceriops tagal and Rhizophora mucronata (Robert et al., 2015); Bruguiera gymnorhiza and Rhizophora stylosa, which are in the same family (Kadoya and Inoue, 2015), have different propagule dispersal and establishment strategies. Previous studies investigated fewer species and did not consider propagules of different structure. Propagule structure is used to indicate the type of propagule and whether it is viviparous or not. Moreover, semi-mangrove, which commonly have higher tidal distribution and propagules of some species are also diffused by water (Tomlinson, 2016), were also omitted.

Few researchers have noted the influence of water salinity on dispersal ability of mangrove propagules (Alleman and Hester, 2011; Robert et al., 2015; Van der Stocken et al., 2015; Tonné et al., 2017; Van der Stocken et al., 2022). Water with greater salinity is denser therefore, for mangrove propagules whose densities are close to the sea water, small changes in salinity may determine whether propagules continue to float or settle. For example, a study found that the propagules of Aegiceras corniculatum could remain buoyant in full seawater for up to three months, but in brackish water, they sank within one week (Clarke, 1995). B. gymnorhiza and A. marina propagules also floated in saltwater and sank in freshwater (Clarke et al., 2001). Interspecific differences result in different floating states of propagules of different species, which is expressed in terms of species buoyancy (specific gravity: specific gravity refers to the ratio of the density of a substance to the density of a reference substance, typically water, at a specific temperature and pressure). However, data on the specific gravity of propagules of various mangrove plant species are incomplete.

In this study, the specific gravity and weight of 35 species of mangrove plants were measured; these included 25 true mangroves, 10 semi-mangroves, and mangrove species with different propagule structure. We posed two main research questions: (1) Does tidal sorting determine mangrove species zonation? (2) Which is the primary factor driving tidal sorting: propagule weight or propagule specific gravity? We hypothesized that mangrove zonation is determined by the tidal sorting of propagules. Furthermore, we hypothesized that tidal sorting is not driven by propagule weight but by their specific gravity. Propagules distributed along the intertidal zone were also collected and surface elevations were recorded. Additions to the tidal sorting hypothesis and the mangrove plant specific gravity database are critical for understanding mangrove distribution patterns, which are of great importance for mangrove restoration and predicting the distribution of mangrove plants in response to global changes.

In our study, we highlight the importance of specific gravity for small scale zonation and offer precise measurements for propagules across 35 mangrove species. The results of this study show that the mangrove propagules are all slightly less dense than seawater (1.00-1.02 g/cm³), which facilitates propagule dispersal and ensures sinking in suitable areas. Mangrove propagules can remain buoyant and viable for long periods to expand their effective range of dispersal (Steinke, 1975, 1986; Rabinowitz, 1978b; Duke et al., 1998). Propagule specific gravity determines their buoyancy, and propagule buoyancy is an important factor that influences dispersal in frequently flooded environments (Johansson et al., 1996). The propagules that had lower specific gravity than seawater could keep floating during dispersal and may move away from mature trees. In contrast, the propagules which have higher specific gravity than seawater would sink. This means that there are species-specific differences to water salinity during the propagule dispersal phase. We found that there was no relationship between propagule specific gravity and weight. Additionally, the dispersal of mangrove propagules was not explained by propagule weight but rather by propagule specific gravity ( Figure 4 ). While many studies have used models to predict the future distributions or restoration outcomes of mangroves (Hamilton et al., 2017; Van der Stocken and Menemenlis, 2017; Van der Stocken et al., 2019b), our study through comparisons of more species, shows that mangrove propagules have significant interspecific differences in specific gravity and difference in settlement position, which further supports the conclusion of Tom Van der Stocken et al. (2019a). Almost all of the models ignore interspecific differences in propagules and assume that they can disperse anywhere with ocean currents. Therefore, the inclusion of different species propagule characteristics in the model is necessary in future predictions (Di Nitto et al., 2013; Wang et al., 2019), which is what our results want to emphasize.

In general, the seaward distance was positively correlated with surface elevation in the intertidal zone (Fu et al., 2019; Ma et al., 2020). The results indicate that the distribution pattern of mangrove propagules is perpendicular to the coastline of Bamen Bay, exhibiting a consistent zonation pattern with mangrove plants (Wang et al., 2019; Ma et al., 2020). It was once widely thought that mangrove propagules dispersed over the entire intertidal zone (Smith, 1992). The previous conclusion overlooked the influence of dispersal on species distribution and magnified the role of environmental factors. Meanwhile, our results revealed that both mangrove trees and propagules exhibit zonation patterns at the intertidal scale. The zonation of propagule was similarly obtained with the field mark-recapture experiment (Sousa et al., 2007), as well as inferences gathered from measured propagule specific gravity (Wang et al., 2019).

The closer to the downstream of the estuary, the higher the salinity of the sea water. Previous studies have shown that S. caseolaris seeds have a limited range of adaptation to salinity, with conditions below 10‰ being the most suitable for growth (Liao et al., 1997). The propagules and plants of S. caseolaris only occurred in transect A and were not found in the downstream transects B and C. This suggested that while the propagules of S. caseolaris cannot be carried to downstream of estuary, the seedlings are protected from high salinity. In contrast to transect A, the propagules and S. alba, C. tagal and L. racemosa were found in the downstream transects. The differences between propagule specific gravity and sea water specific gravity (Wang et al., 2019) prevented propagules from reaching the upper reaches of the river. Our results suggest that mangrove propagules can be dispersed to suitable habitats that influenced the distribution of mangroves. The similarity in the distribution of propagules and plants suggests propagule dispersal as a key driver of mangrove distribution patterns, alongside other biotic and abiotic factors (Sousa et al., 2007; Kadoya and Inoue, 2015; Wang et al., 2019; Van der Stocken et al., 2019a). Settlement of propagules in the intertidal is set to limit animal predation and is influenced by environmental factors such as tidal scour, water salinity, and inundation, which all contribute to the early establishment of mangrove plants (Krauss et al., 2008). The differences in settling positions of propagules, adaptability to water salinity, tolerance to inundation, and efficiency in nutrient utilization further contribute to the zonal distribution of mangrove plants (Tomlinson, 2016).

Both C. tagal and R. mucronata, belong to the same family and reproduce viviparously, however, R. mucronata was found to strand and establish in the lower intertidal zone (Robert et al., 2015). In the present study, although S. caseolaris and S. alba both belong to Sonneratiaceae, their propagules and plants were distributed in different locations along the river. R. apiculata, B. sexangular, and C. tagal belong to the family Rhizophoraceae, however, B. sexangula, R. apiculate, and C. tagal were distributed in order from sea edge to landedge in the intertidal zone. Differences in the intertidal positioning of propagules resulted from varying retention periods for species in the intertidal zone. The retention duration of mangrove propagules is in fact linked to the duration of the window for establishment (Balke et al., 2013; Van der Stocken et al., 2019a; van Hespen et al., 2022). This window denotes the timeframe during the retention period, when mangrove propagules have the opportunity to encounter favorable growth conditions within the intertidal zone, facilitating their establishment into plants (Balke et al., 2011; 2013).

The appearance of zonation with propagule dispersal and plant establishment also supported the tidal zonation hypothesis (Rabinowitz, 1978b), which results from tide sorting of propagules based on their traits or differences in propagule colonization ability. However, unlike Rabinowitz (1978b), we found that it was not weight but specific gravity that affects the distribution in the intertidal zone. Van der Stocken et al. (2022) proposed that the interaction between specific gravity and seawater can influence the distribution of mangrove plants, particularly in the context of human activities and climate change induced changes in sea water salinity (Cinco-Castro and Herrera-Silverira, 2020; Chen et al., 2021), which lead to changes in hydrological characteristics.

The influence of the propagule dispersal stage must not be overlooked in future discussions of the cause(s) or predictions of mangrove distribution patterns (Di Nitto et al., 2013; Van der Stocken et al., 2019a; Chen et al., 2020). Our study measured the specific gravity and weight of 35 species of mangrove plants, including 25 true mangroves, 10 semi-mangroves, and mangrove species with different propagule structures. This comprehensive data collection and analysis contribute to a deeper understanding of the ecological characteristics of different propagule structures of mangrove plants. We conducted a reassessment of the tidal sorting hypothesis, which has led to new insights into the distribution mechanisms of mangrove plants and proposed alternative perspectives diverging from traditional views. Our results indicate that dispersal has an impact on mangrove zonation, and environmental adaptation and interspecific competition may further intensify mangrove zonation. If interspecific differences are not properly accounted for, the effects of environmental filtering or interspecific competition on patterns may be magnified. In the case of mangrove conservation and restoration, it is desirable to obtain a community structure similar to that of natural mangrove forests. The complexity of elevation in the restored area should be set artificially in restoration to cope with species differences in propagule retention. Meanwhile, the setting of critical elevations is essential to ensure that propagules can sink and seedlings survive in all future mangrove restoration projects. We suggest that more consideration should be given to interactions of hydrodynamics with propagule dispersal processes in different species in future studies/investigations of mangrove plants. Additionally, the relationship between the survival and dispersal abilities of propagules should also be taken into account for more accurate estimations of mangrove distribution.

Publicly available datasets were analyzed in this study. This data can be found here: https://zenodo.org/records/10476887.

CL: Writing – original draft, Visualization, Software, Project administration, Methodology, Formal analysis, Data curation, Conceptualization. LZ: Writing – review & editing, Software, Methodology, Conceptualization. XS: Writing – review & editing, Methodology, Investigation, Conceptualization. XL: Writing – review & editing, Methodology, Investigation, Conceptualization. YD: Writing – review & editing, Methodology, Investigation, Conceptualization. MW: Writing – review & editing, Validation, Supervision, Funding acquisition, Conceptualization. WW: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.