Climate Change Impacts on Native and Exotic Mangrove Distributions and Niche Overlap Analysis

Abstract

Mangroves are important coastal wetland ecosystems with high ecological service values and strong carbon sequestration capacity, serving as a crucial barrier for coastal ecological security. However, current afforestation efforts often ignore environmental suitability differences among mangrove species, while the applicability value and ecological risks of exotic species (Laguncularia racemosa and Sonneratia apetala) for restoration remain poorly understood. Five native and two exotic mangrove species along China’s coasts were selected in this study. Using the MaxEnt model, we identified key environmental factors governing their distribution, predicted their current and future suitable habitats (under the SSP245 scenario in the 2070s), and quantified niche overlap between native and exotic mangroves. The results showed that temperature-related factors (air and sea temperature) are the core climatic drivers shaping the typical mangrove distribution, followed by sea surface salinity, with precipitation contributing little. Currently, niche overlap between native and the two exotic species is low (D.overlap: 0.129–0.340), indicating certain niche differentiation. Under the SSP245 scenario in the 2070s, except for Rhizophora stylosa, other studied species appear to experience expanded suitable habitat areas and a northward latitudinal distribution shift. Compared with Sonneratia apetala, Laguncularia racemosa exhibits a more pronounced expansion of suitable habitats in the future, with its overall suitable area second only to the native Kandelia obovata, indicating its stronger adaptive potential to climate change. Clarifying niche differentiation and constructing species-specific management frameworks may facilitate biological invasion control, mangrove restoration, and species diversity improvement.

1. Introduction

Mangroves are woody wetland communities of evergreen shrubs or small trees dominated by mangrove plants, primarily distributed in the coastal intertidal zones of tropical and subtropical regions. They play a pivotal role in wave attenuation and bank protection, seawater purification, climate regulation, biodiversity conservation, fishery resource preservation, and the provision of cultural ecosystem services such as tourism, entertainment and education [1,2], thus boasting exceptionally high ecological service value and robust ecological functions. In recent years, mangroves have garnered growing attention due to their strong carbon sequestration capacity. On the one hand, despite covering less than 1% of the world’s coastal marine areas, mangroves sequester over 15% of global carbon in modern marine sediments and supply approximately 10% of the total terrestrial organic carbon exported to coastal waters [3,4,5], thus ranking among the most vital blue carbon sinks. On the other hand, as one of the world’s most threatened ecosystems [6], mangrove degradation has boosted global CO₂ emissions from tropical deforestation by 10%, while global mangrove carbon stocks decline 1% yearly, releasing an extra 133 Tg C into the atmosphere each year [7]. Therefore, the protection and restoration of mangroves are of great significance for enhancing carbon sequestration and mitigating global warming.

Although mangroves play a crucial role in promoting human well-being and maintaining ecosystem stability, they have currently become one of the most threatened ecosystem types globally [6]. Over the past few decades, affected by the combined impacts of rapid industrial and agricultural development, accelerated urbanization, and a sharp increase in population in China, mangrove habitats have been largely occupied, leading to a significant reduction in their area [8]. Statistically, the total area of mangroves in China decreased from 48,801 hectares in 1973 to 20,450 hectares in 2000, with a decline rate of more than 60% [9]. In response to this severe situation, China has gradually increased resource investment in mangrove protection and restoration over the past 20 years [10], and a number of artificial mangrove restoration projects and practices have been successively carried out in coastal areas [8,11]. These targeted protection and restoration measures have effectively curbed the trend of sharp decline in mangrove area, making it show a slow recovery trend. China has thus become one of the few countries in the world to achieve a net increase in mangrove area [12,13].

Over recent decades, mangrove restoration practices in China have yielded positive outcomes. Nevertheless, unscientific site selection and irrational species allocation have resulted in low survival rates of replanted mangroves, leading to inefficient consumption of human and financial resources [8,10]. Relevant investigations have demonstrated that most tidal flats designated in national mangrove afforestation plans present harsh site conditions, which are unsuitable for the direct cultivation of native mangrove species [14]. Targeted habitat remediation to meet afforestation requirements further demands systematic technical support and sustained financial investment. Facing such practical restoration constraints, fast-growing exotic mangrove species including Sonneratia apetala and Laguncularia racemosa have been extensively introduced in previous coastal afforestation projects to rapidly expand regional mangrove coverage [14,15]. In addition, exotic species can improve soil fertility and foster favorable conditions for the colonization and growth of late-successional native mangrove species [16,17]. Nevertheless, the fast-growing characteristics and strong dispersal capacity of these two exotic mangrove species have aroused concerns about ecological invasion risks in some regions [18]. Some scholars have proposed that to ensure the safety of the mangrove ecosystem, native mangrove species should be given priority as transplant materials in the process of mangrove restoration [19]. Practical experience in mangrove restoration has demonstrated that in-depth research into the potential habitats of native and exotic species, as well as a clear clarification of the ecological competitive relationships between them, is necessary for formulating scientific and rational conservation strategies, facilitating population restoration, and protecting biodiversity.

Species distribution models (SDMs) serve as a methodological tool for guiding the identification of species-specific suitable habitats, as they quantify relationships between species distribution data and environmental factors via statistical algorithms to predict the distribution probability of target species [20,21]. Among them, the Maximum Entropy (MaxEnt) algorithm, which is based on presence-only data, has shown the optimal performance and has been widely applied in predicting the suitable areas for mangroves in China [22,23]. Nevertheless, there remain certain knowledge gaps in current mangrove distribution research. Most previous studies have focused on community-scale patterns rather than species-level responses, failing to provide precise and species-specific guidance for mangrove restoration and site selection. In particular, few studies have horizontally compared the potential distribution patterns of native versus exotic mangrove species as well as evaluating their spatial variability under climate change. Such key information is essential for clarifying the mechanisms underlying climate change impacts on different mangrove species and developing targeted strategies to enhance the adaptability and resilience of mangrove ecosystems.

To address the aforementioned research gaps and fill the deficiencies in current mangrove conservation and management practices, this study selects five typical native mangrove species widely distributed in coastal areas of China and two exotic mangrove species that have been extensively introduced as research objects. Utilizing the MaxEnt model, we aim to identify the most influential environmental factors governing the growth of mangrove species, predicting the potential habitats of each species. Then, we quantitatively analyze the current niche overlap between native and exotic species to clarify their interspecific competitive relationships and the potential ecological invasion risks posed by exotic species. Furthermore, we will explore the differences in changes in potential suitable habitats and latitudinal patterns of native and exotic species by the 2070s under the SSP245 climate scenario. Beyond providing scientific insights into the response mechanisms of mangrove species to climate change and the interspecific competition between native and exotic species, this research is expected to offer targeted and spatially explicit management strategies for mangrove ecosystem conservation and restoration. Specifically, it provides a theoretical basis for selecting priority transplant species, assessing the ecological risks of exotic mangroves, and guiding mangrove restoration in vulnerable coastal areas.

2. Materials and Methods

2.1. Study Area

Mangroves in China naturally occur from Yulin Harbor, Sanya of Hainan (18°09′ N) to Shacheng Harbor, Fuding of Fujian (27°20′ N) [24], while their northern limit for artificial introduction extends to Yueqing County, Zhejiang Province (28°25′ N) [8]. In China, mangroves are predominantly distributed across Hainan, Guangdong, and Guangxi, which together account for 96% of the nation’s total mangrove area. Smaller mangrove patches also occur in Fujian, Zhejiang, Taiwan, Hong Kong, and Macao. The study area is characterized by a tropical to subtropical monsoon climate, featuring hot, humid summers and cool, rainless winters.

China boasts a rich diversity of mangrove plant resources. Investigations have confirmed the presence of 26 species of true mangroves and 12 species of semi-mangroves [25], accounting for approximately one-third of the world’s total mangrove species. Among these taxa, Kandelia obovata, Avicennia marina, Aegiceras corniculatum, Bruguiera gymnorrhiza and Rhizophora stylosa are native dominant species in the subtropical coastal areas of China. Their natural distribution spans multiple provinces along the southeastern coast, making them the core constructive species of China’s mangrove communities. Among introduced species, Sonneratia apetala ranks as one of the most important tree species applied in mangrove restoration projects along South China’s coasts [25]. In recent years, Laguncularia racemosa has been successively introduced to Hainan, Guangdong, Fujian and other provinces for afforestation and landscape purposes [18]. However, these two exotic mangrove species exhibit rapid growth and prominent competitive advantages after establishment in some regions, showing a tendency to outcompete native indigenous plants. Their potential risks of biological invasion have thus aroused widespread concerns [26,27].

2.2. Mangrove Distribution Data

The mangrove species distribution data in this study were first retrieved and collected from the Global Biodiversity Information Facility (GBIF; https://www.gbif.org/, accessed on 24 June 2025). Samples with missing precise latitude and longitude information, duplicate records, or obviously erroneous geographic coordinates were excluded. To address the limitation of insufficient distribution points for some species in the GBIF database, more data were obtained from several published datasets. Specifically, the distribution data of Kandelia obovata, Avicennia marina, Aegiceras corniculatum, Bruguiera gymnorrhiza, Rhizophora stylosa, and Laguncularia racemosa were supplemented using the dataset from Zhao et al. [28]. This dataset, for the first time based on remote sensing technology, systematically revealed the spatial distribution information of major mangrove species in China, with the distribution information verified through relevant field sample surveys. The distribution data of Sonneratia apetala in China for 2020 were acquired from the Science Data Bank [29] and validated with three datasets. The test and independent evaluation datasets achieved overall accuracies of 98.1% and 96.4%, respectively; field verification with 145 expert-provided samples also showed a 91.7% accuracy, confirming the dataset’s high classification accuracy and suitability for this study’s analytical needs.

To improve data sufficiency and spatial representativeness for species distribution modeling, we integrated GBIF occurrence records with high-precision remote sensing-derived mangrove maps. This combination effectively compensates for the sparsity and uneven survey effort in GBIF data, while incorporating systematic, full-coverage distribution information from remotely sensed products. Nevertheless, integrating multi-source data may introduce potential sampling biases, such as opportunistic recording bias in GBIF and minor classification uncertainties in remote sensing products. To alleviate these effects and reduce spatial autocorrelation among occurrence points, the Fishnet tool in ArcGIS 10.8.1 was used to resample points at 1 km × 1 km resolution, retaining only one record per grid cell. The final numbers of valid occurrence points for each species were as follows: native species including Kandelia obovata (506), Aegiceras corniculatum (399), Avicennia marina (1096), Bruguiera gymnorrhiza (129) and Rhizophora stylosa (292); exotic species comprising Laguncularia racemosa (92) and Sonneratia apetala (1400). The distribution patterns of all mangrove species are presented in Figure 1.

2.3. Environmental Dataset and Climate Change Scenarios

Mangroves are distributed in coastal intertidal zones, and their distribution is jointly influenced by both terrestrial and marine environments. Temperature and rainfall regimes govern the global distribution, structure, and function of mangrove forests [30]. Topography and bathymetry of the coast result in three dominant features like rivers, tides and waves, which play a role in shaping the geomorphic settings of mangroves [31]. In this study, a total of 28 variables were selected as key environmental factors, including bioclimatic, marine environmental, and topographic variables. Bioclimatic data were retrieved from WorldClim V2.1 (https://www.worldclim.org/, accessed on 3 March 2025), encompassing 19 standard temperature- and precipitation-related bioclimatic variables at a 30-arcsecond spatial resolution. Elevation data were sourced from ETOPO2022 (www.ngdc.noaa.gov, accessed on 23 July 2025), and the slope and aspect data were extracted using spatial analysis tools in ArcGIS 10.8. Marine variables were retrieved from the Bio-ORACLE V3.0 database (https://www.bio-oracle.org/, accessed on 16 March 2025), a global environmental database for marine biodiversity distribution modeling. Among the oceanic variables, the average sea surface temperature, long-term average minimum and maximum sea surface temperature, sea surface temperature range, mean sea surface salinity, and mean sea surface current velocity were selected as key variables in this study.

The future bioclimatic data for China used in this study were derived from the dataset generated by the Beijing Climate Center Climate System Model (medium-resolution version, BCC-CSM2-MR) under the framework of the Coupled Model Intercomparison Project Phase 6 (CMIP6). This dataset includes a suite of scenario data based on the Shared Socioeconomic Pathways (SSPs) [32]. Various SSP scenarios exhibit distinct future emission and socioeconomic trajectories. The SSP126 pathway entails stringent mitigation prerequisites and overidealized developmental assumptions, with limited practical feasibility under current global policy contexts. In contrast, SSP370 and SSP585 represent extreme high-emission scenarios with overly pessimistic projections, which diverge markedly from the ongoing global low-carbon transition and climate governance progress. As a moderate intermediate pathway, SSP245 reasonably balances socioeconomic demand and conventional climate mitigation efforts and conforms well to existing national climate action commitments and realistic development trends [33]. Accordingly, simulating the suitable habitats of mangroves based on the SSP245 scenario can yield reliable results consistent with current trends [34].

The time period of the 2070s (average for 2061–2080) was selected as the simulation window for future climate scenarios in this study, based on two core considerations. First, the impacts of climate change on mangroves and the corresponding responses of mangroves exhibit a long-term cumulative effect, making it difficult to identify distinct patterns over short timescales. Second, 2060 marks the target year for carbon neutrality under China’s Dual Carbon Strategy; prior to this year, the climate system is prone to perturbations from unpeaked carbon emissions, whereas the climate system will enter a relatively stable phase after 2060, which can effectively improve the reliability of habitat prediction. This time window not only aligns with the ecological conservation scenarios following the achievement of the carbon neutrality target but also provides forward-looking references for the formulation of relevant policies after the target is met, thus ensuring the temporal alignment of this study with national strategies.

Future marine environmental variables were also represented by the 2060–2080 averages under the SSP245 scenario, consistent with the bioclimatic variables, while topographic variables remained unchanged relative to the present. All the aforementioned environmental variables were processed at a spatial resolution of 30 arcseconds (~1 km), and a 10 km buffer zone was created on both sides of the coastline to trim these datasets. To avoid model overfitting caused by multicollinearity among variables, a correlation analysis was conducted on the 28 environmental variables. For any pair of variables with an absolute correlation coefficient |r| > 0.8, the variable with stronger ecological relevance to mangrove species distribution was retained, while the less ecologically important one was excluded. A total of 17 environmental variables (

Table 1. Environmental factors used for MaxEnt models.

Data Types	Variables	Variable Description	Unit
Bioclimate	Bio2	Mean diurnal range	°C
	Bio3	Isothermality	%
	Bio5	Max temperature of warmest month	°C
	Bio6	Min temperature of coldest month	°C
	Bio8	Mean temperature of wettest quarter	°C
	Bio9	Mean temperature of driest quarter	°C
	Bio10	Mean temperature of warmest quarter	°C
	Bio13	Precipitation of wettest month	mm
	Bio15	Precipitation seasonality	%
	Bio19	Precipitation of coldest quarter	mm
Topography	Elevation	Surface elevation	m
	Slope	Slope	°
	Aspect	Aspect	°
Marine	SSS	Mean sea surface salinity	‰
	SSV	Mean sea surface water velocity	m·s−1
	SSTmean	Mean sea surface temperature	°C
	SSTmax	Mean long-term maximum sea surface temperature	°C

) were ultimately selected for model simulation. Prior to input into MaxEnt, all environmental variable data were converted to ASCII format in the ArcGIS 10.8 software.

2.4. Potential Suitable Habitat and Niche Overlap Analysis

By integrating species distribution data and the selected environmental variables, we predicted the potentially suitable habitats of each species under current and future scenarios with the MaxEnt software (Version 3.4.1). For each species, 75% of occurrence records were randomly selected for model training, and the remaining 25% for testing. To ensure the stability of the simulation, a “subsample” mode with 10 replicates was adopted. The Jackknife method was applied to evaluate variable importance and quantify the effects of environmental variables on mangrove distribution. Response curves were further generated to illustrate the response patterns of habitat suitability to individual environmental variables. The “logistic” format was also conducted to produce the suitability maps. Other settings were kept at their default values within the software. We transformed the continuous suitability values (0–1) from the MaxEnt model into a spatial habitat distribution map using ArcGIS 10.8.1. Binary habitat suitability maps (suitable = 1/unsuitable = 0) were produced using the training sensitivity plus specificity threshold [35,36].

The predictive performance of MaxEnt model was evaluated using the area under the Receiver Operating Characteristic (ROC) curve (AUC) and the true skill statistic (TSS). The threshold-independent AUC has been widely used to assess the accuracy of species distribution models Model performance was graded as failing (0.5–0.6), poor (0.6–0.7), general (0.7–0.8), good (0.8–0.9), and excellent (0.9–1.0) based on AUC values [37]. TSS accounts for both omission and commission errors and is not influenced by species prevalence [38]. The value of TSS ranges from −1 to +1, where a value of +1 represents perfect agreement, while values of zero or lower suggest performance no better than random [39].

To better understand the environmental resource competition between native and exotic mangrove species and accurately quantify their niche overlap, we used the “ecospat” package in R 4.4.0 [40]. Niche overlap was quantified using Schoener’s D index, which ranges from 0 (indicating no overlap) to 1 (indicating complete overlap). Schoener’s D was selected because it is widely applied in niche studies and comprehensively accounts for differences in species occurrences and habitat suitability, providing a robust metric for quantifying niche overlap [41].

3. Results

3.1. Assessment of Model Performance and Identification of Key Environmental Variables

Model predictive accuracy was evaluated using the AUC and TSS metrics. As shown in

Table 2. The values of AUC and TSS for mangrove species.

Species	AUC	TSS
Kandelia obovata	0.895	0.632
Aegiceras corniculatum	0.950	0.761
Avicennia marina	0.933	0.765
Bruguiera gymnorhiza	0.924	0.669
Rhizophora stylosa	0.957	0.777
Laguncularia racemosa	0.940	0.704
Sonneratia apetala	0.936	0.796

, the AUC values for all target species ranged from 0.895 to 0.957, while the TSS values varied between 0.632 and 0.796, indicating overall good to high performance. These results confirm that the models have high predictive reliability and can accurately reflect species distribution patterns, providing a solid basis for subsequent spatial and scenario analyses.

The percentage contribution method was employed to quantify the influence of different environmental variables on the potential distribution of mangrove species. This metric is calculated during model operation by sequentially adjusting the coefficients of individual variables and allocating the gain increments to the corresponding environmental variables. The results (Figure 2) indicated that, overall, temperature-related factors (air temperature and sea surface temperature) were the core drivers of habitat suitability for mangrove species, among which the mean long-term maximum sea surface temperature (SSTmax), mean temperature of the wettest quarter (Bio8), and mean sea surface temperature (SSTmean) accounted for the highest total contribution; sea surface salinity (SSS) was the secondary driver. Among topographic factors, slope exerted a relatively prominent influence on species distribution, whereas precipitation-related factors (Bio13, Bio15, Bio19) had an unremarkable contribution proportion among all environmental factors.

However, the percentage contributions of these environmental factors to determining habitat suitability varied among species (Figure 2). Slope had a significantly greater influence on Kandelia obovata (13%) and Bruguiera gymnorrhiza (10.6%) than on other species. Among all species, Sonneratia apetala was most strongly affected by precipitation of the coldest quarter (Bio19). Temperature seasonality (Bio3) and mean temperature of the warmest quarter (Bio10) contributed the most to predicting the distribution of Laguncularia racemosa. These differences indicate the presence of species-specific ecological adaptations.

To clarify the species-specific responses to changing environmental conditions, we selected the four environmental variables with the highest percent contributions to construct response curves. These curves depict how the occurrence probability of each species varies with gradients of key environmental factors. We found that different mangrove species differ significantly in both habitat suitability thresholds and the shapes of their response curves. As shown in Figure 3, Kandelia obovata exhibited a considerably wider suitable range for SSTmax, Bio8, and SSTmean than other species, with a higher occurrence probability under lower temperature conditions. Sonneratia apetala and Laguncularia racemosa were less tolerant to low temperatures than K. obovata, similar to Avicennia marina, but more tolerant than the other three native species. Figure 3d shows that K. obovata grows better in moderate and low salinity but is less tolerant to high salinity than other native species; S. apetala prefers environments with lower seawater salinity.

3.2. Suitability of the Mangrove Forest Habitat

3.2.1. Suitable Habitats for Mangrove Forests Under Current Climate Scenarios

The continuous habitat suitability predicted for each species was converted into binary maps using the maximum training sensitivity plus specificity threshold rule. Only suitable areas (grid cells with suitability probability greater than the threshold) were mapped and used to calculate suitable habitat area. The suitability distributions of native and exotic mangrove species under current climate scenarios are presented in Figure 4a–g. The MaxEnt simulation results indicated that the simulated area of mangrove suitable habitat under current conditions followed the order: Kandelia obovata (21,117 km²) > Sonneratia apetala (15,906 km²) > Avicennia marina (13,687 km²) > Bruguiera gymnorrhiza (11,375 km²) > Aegiceras corniculatum (10,629 km²) > Rhizophora stylosa (8737 km²) > Laguncularia racemosa (6175 km²). Based on the predicted suitable distribution of all species, a species richness map along the coastal areas of China was generated through spatial superposition (Figure 4h). The results showed that among the species simulated in this study, the regions with the highest species richness were mainly distributed along the Leizhou Peninsula and the coastal zone of the Beibu Gulf in Guangxi, while the regions with moderate species richness included the Pearl River Estuary, the coasts of Yangjiang and Maoming, Dongzhai Harbor in Hainan, and Xiamen in Fujian.

By comparing the occurrence points of mangrove species with the simulated current potential suitable habitats, the model predictions show an overall spatial consistency with the realistic distribution of mangroves, and the highly suitable areas are well matched with the concentrated existing populations. Nevertheless, local underestimation of habitat suitability and spatial prediction biases remain across certain coastal regions. Specifically, the simulated northern distribution limit of Kandelia obovata is consistent with its actual latitudinal boundary, while its habitat suitability along the coast of Guangxi is underestimated, and the natural populations in northern Hainan are not fully captured by the model. Considerable prediction deviations are observed for Aegiceras corniculatum in central Guangdong and Hainan coasts. Omission errors occur in the localized distribution of Avicennia marina, Bruguiera gymnorrhiza and Rhizophora stylosa in specific coastal zones of Fujian, Hainan and scattered habitats of central Guangdong. In contrast, the distribution simulations of exotic mangroves (Laguncularia racemosa and Sonneratia apetala) achieve higher overall accuracy, with satisfactory consistency between predicted results and field distribution records. Regionally, northern Hainan exhibits the most severe omission bias, which further leads to the underestimation of mangrove species richness across Hainan Province.

3.2.2. Distribution of Mangrove Habitats Under Future Climate Scenarios

We also simulated suitable habitats for each mangrove species under the 2070s SSP245 scenario, and the results indicated that their potential distribution patterns had changed significantly (Figure 5). The predicted suitable habitat of most species in this study showed a significant increasing trend except Rhizophora stylosa (Figure 6a): the suitable habitat extent of four native mangrove species, namely Kandelia obovata, Aegiceras corniculatum, Avicennia marina, and Bruguiera gymnorrhiza, increased by 44.90%–202.35%, whereas that of the native Rhizophora stylosa drastically declined by 98.94%. Among exotic species, Laguncularia racemosa exhibited the largest expansion in suitable habitat (555.79%), expanding from 6175 km² to 40,495 km², with its coverage second only to Kandelia obovata; the suitable habitat of Sonneratia apetala also increased by 64.73%, yet its overall suitable area remained smaller than that of the native mangrove species Kandelia obovata, Aegiceras corniculatum and Avicennia marina.

Under future climate scenarios, as carbon emissions continue to increase, the climatic suitability of all mangrove species except Rhizophora stylosa will show an upward trend by the 2070s, which is specifically reflected in the significant increase in the suitability probability in coastal areas. Meanwhile, the suitable areas of all mangrove species will exhibit a tendency of northward expansion in terms of latitudinal distribution (Figure 6b), among which the northernmost boundary of the suitable habitats for Kandelia obovata, Aegiceras corniculatum, Bruguiera gymnorrhiza, and Laguncularia racemosa will extend north of 30° N. The increase in the area of suitable areas for Sonneratia apetala is mainly derived from the improved suitability in the coastal areas of western Guangdong, the Beibu Gulf region of Guangxi, and the northeastern part of Hainan Island; therefore, there will be no significant change in the main distribution range (25%~75%) of its suitable regions. Affected by the sharp decrease in the climatic suitability of Rhizophora stylosa, the maximum mangrove species richness will decrease from 7 to 6 under future climate scenarios; specifically, the species richness in the Leizhou Peninsula will decrease, while that in other coastal cities of Guangdong, Fujian, Zhejiang, Hainan, and Taiwan will all show an increasing trend.

3.3. Current Ecological Niche Overlap Between Native and Invasive Mangroves

As shown in Figure 7 and

Table 3. Niche overlap index (Schoener’s D) between native and exotic mangrove species.

	Laguncularia racemosa	Sonneratia apetala
Kandelia obovata	0.129	0.136
Aegiceras corniculatum	0.285	0.340
Avicennia marina	0.246	0.282
Bruguiera gymnorrhiza	0.233	0.165
Rhizophora stylosa	0.323	0.210

, the degree of ecological niche overlap between five native mangrove species and two exotic species is illustrated. The results demonstrate that both Laguncularia racemosa and Sonneratia apetala exhibit ecological niche overlap with native mangrove species (D.overlap > 0). The Schoener’s D values for niche overlap between Laguncularia racemosa and native species range from 0.129 to 0.323, in the order: Rhizophora stylosa > Aegiceras corniculatum > Avicennia marina > Bruguiera gymnorrhiza > Kandelia obovata. For Sonneratia apetala, its Schoener’s D values with native species fall within the range of 0.136 to 0.340, with the sequence: Aegiceras corniculatum > Avicennia marina > Rhizophora stylosa > Bruguiera gymnorrhiza > Kandelia obovata. Higher niche overlap values typically reflect high similarity in resource utilization and habitat requirements among co-occurring species under limited environmental resources. In this study, Kandelia obovata, Aegiceras corniculatum and Avicennia marina showed greater niche similarity with S. apetala than with L. racemosa, while Bruguiera gymnorrhiza and Rhizophora stylosa shared more analogous resource niches with L. racemosa. Collectively, the native species Kandelia obovata had the lowest niche similarity with the two exotic mangroves.

4. Discussion

4.1. Key Environmental Factors Influencing Suitability

Changes in temperature and precipitation patterns influence ecosystem processes, alter the suitability for growth and community composition of mangrove species [42] and govern the global distribution, structure, and function of mangrove ecosystems [30,43].

Temperature can affect plant physiological activities and developmental processes [44]. Laboratory experiments and empirical studies have demonstrated that temperature regimes can constrain species’ geographic ranges by suppressing metabolic activities and/or damaging cell or tissues [45]. Global mangrove distribution is restricted to areas within the 16 °C isotherm of the coldest month or the 20 °C isotherm of winter sea surface temperature [46]. Precipitation indirectly influences mangrove distribution patterns by regulating coastal salinity, freshwater supply, and soil inundation. In arid regions, where evaporation exceeds precipitation, elevated soil salinity inhibits mangrove growth and may even lead to widespread die-back [47].

Although ecologists have long recognized that temperature and precipitation regimes determine the global distribution, structure, and function of mangroves, their relative importance often varies at regional scales [30]. Existing research indicates that in humid coastal zones, temperature exerts a stronger driving effect on mangrove spatial distribution than precipitation [30]. Furthermore, in high-latitude coastal areas, constrained by frost stress, the influence of temperature on mangrove growth may be most pronounced at latitudinal range limits [48]. Our study area is located along the southeast coast of China, characterized by abundant precipitation and positioned at the northern edge of the world’s natural mangrove distribution. This biogeographical pattern reasonably explains why mangrove species distributions in this study are primarily governed by temperature factors (air and sea surface temperature), with precipitation contributing relatively weakly. These results are consistent with those of several previous potential distribution simulations of mangroves in China [49,50,51,52]. Notably, while winter low temperature is commonly regarded as a key limiting factor for mangrove biogeography, our results reveal that sea surface high temperature also contributes substantially to mangrove distribution. A previous study on long-term mangrove cover changes in China also suggested that metrics based on high temperature are more suitable for explaining and predicting mangrove area dynamics in warm regions of southern China [53].

Beyond climatic factors, slope shapes the spatial distribution of mangroves within the intertidal zone mainly by regulating tidal inundation duration, sediment stability, and microtopographic characteristics. Relevant studies show that mangroves exhibit higher suitability and occurrence probability in gentle-slope areas with low wave energy, and gentle intertidal topography can effectively improve the wave attenuation capacity of mangrove forests [54]. Slope and tidal processes jointly regulate the vertical zonation of mangroves in the intertidal zone; in China, these forests commonly form a sequential distribution gradient from the sea to the land, consisting of Avicennia marina, Aegiceras corniculatum, Kandelia obovata, Rhizophora stylosa, and Bruguiera gymnorhiza [55]. However, the influences of intertidal topography and tidal dynamics on mangrove species distribution appear to be better identified only at finer spatial scales [55,56]. Constrained by the 1 km spatial resolution of this study, the constraining effect of topographic factors on mangrove distribution may have been somewhat underestimated in the simulations. In addition, other environmental factors important to mangrove distribution, such as soil properties and water quality parameters [57], were not incorporated in this study, which may also affect the accuracy of the model results to some extent.

4.2. Changes in the Suitable Habitats and Species Richness of Mangroves Under Climate Change

Predicting the changes in species’ suitable habitats under climate change scenarios is crucial, as it serves as a key prerequisite for assessing species survival risks and formulating scientific conservation strategies. This study investigated the responses of the potential suitable distributions of typical mangrove species to climate change under the SSP245 pathway in the 2070s. The results showed that the overall suitable areas of the studied species exhibited an expanding trend, reflecting that future global warming will be conducive to the growth of mangrove plants [58]. In terms of latitudinal patterns, our study predicts that the suitable ranges of most mangrove species will expand northward in the future. This trend is consistent with global findings: driven by winter warming and a reduction in extreme cold events, mangroves worldwide are generally shifting toward higher latitudes with continuously widening latitudinal distributions [59]. Studies focusing on China have suggested that a 2 °C rise in temperature could push the northern limit of Chinese mangroves approximately 2.5 degrees northward, from the current Leqing County in Zhejiang Province to Hangzhou Bay [58,60], which is basically consistent with the projected future northern boundaries of the species distributions in our study. However, long-term investigations of historical mangrove distributions indicate that the northernmost latitudinal limits of mangroves in some regions have not shown regular poleward expansion [61], suggesting that the mechanisms driving the northward expansion of mangroves are complex and may be constrained by dispersal limitations and other factors [61,62]. For instance, a study on mangroves in southern Africa reported that dispersal ability is a key factor limiting the southern distribution limit of mangroves and recommended incorporating dispersal processes into species distribution models [63]. Furthermore, the northward expansion of mangroves may encroach into communities dominated by salt marsh herbs [62,64]. Highly complex ecological interactions exist between herbaceous salt marsh plants and mangroves [64]. To fully reveal the impacts of climate change on mangrove communities, biotic interactions should be integrated into the construction of species range models [65]. Due to the challenges in data acquisition, several complex limiting factors that are critical to mangrove distribution were not incorporated into our analysis. Consequently, our projections of potential suitable distribution areas may be somewhat optimistic.

For specific species, our findings revealed that under future climate scenarios, the predicted suitable habitat area of Laguncularia racemosa ranks second only to Kandelia obovata, with the highest growth rate; its main latitudinal distribution range (25th–75th percentiles) is comparable to that of Avicennia marina. This may be attributed to the physiological factor that the low-temperature tolerance of Laguncularia racemosa is lower than that of Kandelia obovata, similar to that of Aegiceras corniculatum and Avicennia marina, but stronger than that of Sonneratia apetala and other native mangrove species [66]. Previous physiological studies have demonstrated that Laguncularia racemosa and Sonneratia apetala can tolerate low temperatures to a certain extent, thus enabling their survival at higher latitudes [67,68]. In northern Fujian Province, China, the growth rate and expansion capacity of Laguncularia racemosa are higher than those of Sonneratia apetala of the same age [26]. Notably, the projected potential distribution areas of these exotic mangrove species exceed those of several native species under future climates. Such advantages in habitat suitability and range expansion may imply potential competitive pressures and ecological impacts on native mangrove species. From the perspective of climatic suitability, our study confirms that exotic mangrove species possess certain application potential in climate change adaptation and high-latitude mangrove restoration, yet their introduction and large-scale planting should be cautiously evaluated with full consideration of invasion risks and long-term ecological consequences.

The impacts of climatic conditions on mangroves are also reflected in species diversity. China lies at the northern limit of the global natural distribution of mangroves, and the species diversity of naturally distributed mangrove communities across different regions decreases with increasing latitude [69]. Our study predicts that under the current climate scenario, the studied species reach the highest richness in the Leizhou Peninsula and the Beibu Gulf of Guangxi, which aligns with the current status of mangroves in China. The predicted low species richness in Hainan stems from the exclusion of many endemic species [70] of the province from our study scope. Under the SSP245 scenario, we predict that future climate change will expand the range of suitable mangrove habitats along the southeast coast of China, with a marked increase in the areas experiencing rising species richness along the coasts of Guangdong and Fujian Provinces.

4.3. Assessment of Invasion Risk from the Perspective of Niche Overlap

Since the 20th century, mangrove areas have been shrinking continuously due to the impacts of intensive human activities such as aquaculture and urbanization, which has become a global concern. To accelerate the ecological restoration of mangroves, Sonneratia apetala and Laguncularia racemosa were introduced into China as pioneer species, and they have now become the most widely applied exotic mangrove species in mangrove afforestation practices in South China [15,71]. Pioneer species are characterized by rapid growth and strong environmental adaptability [26], able to form a closed canopy quickly and establish mangrove communities on bare tidal flats [72]. The formation of mangrove communities can improve soil fertility and habitat conditions, creating more favorable habitats for the colonization and growth of late-successional native mangrove plants [16,17]. However, exotic fast-growing species often have advantages over native species in growth-related traits, such as photosynthetic rate and resource use efficiency [73,74,75]. Consequently, they not only pose a risk of displacing native mangroves but also may evolve into invasive species [26,76].

This study reveals the complex relationships between exotic species and typical native mangrove species in terms of resource use from the perspective of niche overlap analysis and provides new insights into their coexistence mechanisms. The niche overlap indices of Sonneratia apetala and Laguncularia racemosa with each of the five native mangrove species range from 0.129 to 0.340, indicating a certain degree of similarity in their ecological requirements and habitat preferences, yet the similarity is relatively low. This may suggest that the species occupy distinct ecological spaces for key resources and have developed resource use differentiation [41]. Previous studies have demonstrated the differences between these two exotic species and native species in their preferences for light [75,77], temperature [66], salinity [78], tidal level [79], nutrient utilization and other such factors. By occupying the resource niches that are underutilized by native species, exotic species may to a certain extent enhance the overall resource use efficiency of the community. Therefore, clarifying the differences in species’ resource preferences and thereby altering the competitive balance between exotic and native species through the regulation of resource availability constitutes an effective strategy for preventing potential biological invasions and restoring wetland habitats [17].

4.4. Uncertain Niche Assessment and Potential Invasion Risks

Although our quantitative results reveal a low degree of niche overlap between native and exotic mangrove species, the invasion risks of Laguncularia racemosa and Sonneratia apetala still require cautious assessment. The niche overlap analysis in this study only considered abiotic environmental factors, while the neglected biotic factors may affect the colonization and growth of native mangroves. Our study adopted a relatively coarse spatial resolution and limited environmental variables, which may underestimate the actual niche overlap to a certain extent.

Compared with native species, exotic fast-growing mangrove species generally exhibit superior growth-related traits, such as higher photosynthetic rate and resource use efficiency [73,74,75], which enables them to establish populations more rapidly and occupy vacant intertidal habitats at an early stage. Moreover, these exotic mangrove species possess stronger environmental stress tolerance, such as high salinity and poor soil nutrient, enabling them to better adapt to harsh habitats such as bare tidal flats [80], which further enhances their competitive dominance. Exotic species may also encroach on the living space of native mangroves. Previous studies have confirmed that exotic mangroves have invaded natural mangrove forests in some regions [15,81,82], profoundly altering the structure and function of local mangrove ecosystems. Notably, whether L. racemosa exerts competitive exclusion on native mangrove plants during natural dispersal and drives the succession of native communities into L. racemosa-dominated communities remains to be further verified by community dynamic studies [81].

In addition, allelopathy is also a critical mechanism underlying biological invasion. Plants can release allelochemicals into the surrounding environment to regulate the growth, development and metabolism of neighboring plants. Existing evidence has demonstrated that both Laguncularia racemosa and Sonneratia apetala exert allelopathic effects on native mangrove species [17,83,84], thereby inhibiting seed germination and seedling establishment of native mangroves. Furthermore, climate change and human activities act synergistically to accelerate the invasion process of exotic species [75]. Rising atmospheric CO₂ concentration and enhanced nitrogen deposition can jointly improve the resource utilization capacity of exotic mangroves, providing favorable conditions for their range expansion [85,86].

Therefore, careful consideration is indispensable when introducing Laguncularia racemosa and Sonneratia apetala into pristine mangrove habitats for afforestation [75]. Long-term monitoring of their growth dynamics and seedling dispersal is urgently needed [26]. Under effective risk control, targeted and hierarchical management strategies can be implemented for mangrove ecological restoration. For regions with severely degraded and difficult-to-restore mangrove habitats, rational introduction and appropriate popularization of these two exotic mangrove species are feasible to meet the national demands for coastal wetland ecological restoration [18].

5. Conclusions

At the species scale, this study systematically analyzed the habitat characteristics of typical native and exotic mangrove species, clarified their current and future potential habitat patterns, and quantified niche overlap between native and exotic species. Contribution analysis indicated that temperature-related factors (air and sea temperature) are the core drivers of mangrove habitat patterns, followed by salinity, while precipitation and topography have relatively weak effects; notably, different species respond specifically to these key environmental factors.

Under the SSP245 scenario, all studied mangrove species are projected to show increased climatic suitability by the 2070s, except Rhizophora stylosa. Among them, the exotic Laguncularia racemosa has the highest growth rate of potential suitable habitats, with an area second only to the native Kandelia obovata, indicating strong adaptability to future climate change. Additionally, driven by global warming, mangrove potential suitable habitats exhibit a clear northward latitudinal expansion. These findings imply that introduced exotic mangroves could have potential value in climate change adaptation and high-latitude restoration, but their application should be cautious and based on rigorous invasion risk assessment.

This study further quantified pairwise niche overlap between native and exotic mangroves. Schoener’s D values between Laguncularia racemosa, Sonneratia apetala and five native species ranged from 0.129–0.323 and 0.136–0.340, respectively, indicating low current niche overlap and a certain degree of niche differentiation. Clarifying interspecific resource preference differences and establishing species-specific management frameworks in conservation planning can support potential biological invasion control and wetland restoration. Optimizing resource use and community composition can also enhance species diversity, providing a theoretical basis for prioritizing multi-species communities in future mangrove restoration to improve ecosystem productivity and resilience.

Notably, although exotic mangroves exhibit strong climatic adaptability and certain restoration potential under future climate scenarios, their artificial introduction and large-scale cultivation should be treated with great caution. Their competitive advantages may aggravate their potential invasion risks and threaten native mangrove communities. Therefore, long-term dynamic monitoring, risk assessment and targeted management for exotic mangroves are urgently required to constrain their uncontrolled expansion, maintain the stability of indigenous mangrove ecosystems, and guarantee the long-term ecological security of coastal wetlands.