The study was conducted at the Dongzhaigang National Nature Reserve, Haikou City, Hainan Province, China (19°51′ to 20°01′N, 110°32′ to 110°37′E). This region features a tropical monsoon maritime climate, with an annual average temperature of 23.3–23.8°C and an annual average rainfall of 1676.4 mm (He et al., 2019). The soil types in the area consist of saline sandy loam or saline marsh soil, with a thickness of approximately 1.0–1.5 m (Cai et al., 2022).

The spread of L. racemosa is accompanied by the input of its leaf litter. To better understand the role of L. racemosa leaf litter during its dispersal, we selected three types of mangrove forests within the Dongzhai Harbor Nature Reserve: a native mangrove area dominated by Rhizophora stylosa (R. stylosa forest), a mixed forest of naturally invaded L. racemosa and R. stylosa (mixed forest), and an experimentally planted mangrove forest dominated by L. racemosa (L. racemosa forest). The stages represented three processes of L. racemosa spread: a native mangrove area where L. racemosa has not invaded, a mixed forest during the spread, and a predominantly L. racemosa forest after replacement of the native mangrove.

By examining the long-term cumulative effects of L. racemosa leaf litter decomposition in the three forests and its short-term effects after a single decomposition event in R. stylosa forests, we explored the effects of L. racemosa leaf litter input on nutrient cycling, which is a key ecosystem function in native mangrove forests. The L. racemosa forests were established in 2013 (20 years of growth). Therefore, this study focused on the cumulative effects of L. racemosa litter decomposition over the 20-year period. Rhizophora stylosa is an evergreen tree in the Rhizophoraceae family, and it reaches heights of up to 8 m and is highly resistant to wind and wave impacts. It is one of the most representative mangrove species in China (Pan et al., 2018). Field surveys have reported its cooccurrence with the exotic species L. racemosa and Sonneratia apetala (Liang et al., 2021). We selected R. stylosa due to its widespread presence in various native mangrove communities, including those invaded by L. racemosa (Liu et al., 2019). Rhizophora stylosa covers a significant mangrove area on Hainan Island and accounts for 40% of the total mangrove area (Fang et al., 2022). Moreover, the probability of mixed growth between L. racemosa and R. stylosa is high, as was observed frequently during field surveys. Thus, the combination was highly representative in our study.

Both the long-term cumulative effects and short-term impacts of L. racemosa leaf litter decomposition on native mangrove ecosystems were studied to explore the influence of this species on nutrient cycling and soil microorganisms.

For the long-term cumulative effects study, we examined the impact of inputting L. racemosa litter on the long-term accumulation of soil nutrients and microorganisms in native mangrove soils by comparing soil nutrient elements and microorganisms across the three representative mangrove forest types in the study area: R. stylosa forest, mixed forest, and L. racemosa forest. In June 2023, three plots measuring 10 m × 10 m each were established in the understory of each forest type, with at least 10 m between plots. Within each plot, a five-point sampling method was employed to collect soil samples for nutrient and microorganism analysis. Soil samples from a depth of 0–15 cm were collected using a soil auger for nutrient analysis, which included measurements of organic C, total N (TN), total P (TP), and ammonium N (NH₄ ⁺-N). Additionally, 5 g soil samples were collected in cryotubes, stored in liquid N, and subsequently analyzed to determine their bacterial and fungal communities.

For the short-term effects study after a single decomposition experiment, we conducted in situ decomposition experiments under the R. stylosa forests using litter decomposition bags. Fresh healthy leaves of L. racemosa and R. stylosa were collected, air-dried, and then placed into bags (1 mm mesh nylon bags at 20 cm × 25 cm) with varying types of litter: R. stylosa litter bags, R. stylosa and L. racemosa mixed litter bags, and L. racemosa litter bags. The bags represented scenarios with no L. racemosa litter input (R1), mix of L. racemosa and R. stylosa litter input (R2), and pure L. racemosa litter (R3), which replaced the native mangrove litter.

In October 2023, in situ decomposition experiments were conducted in three plots in the R. stylosa forest to explore the impact of L. racemosa leaf inputs on soil nutrients and microorganisms in native mangrove soils. In each plot, 27 bags of each type of litter were placed adjacent to each other, with 9 retrievals per plot and 3 replicates per retrieval, for a total of 243 litter decomposition bags across all plots. Each decomposition bag contained 15 g of air-dried leaves. After removing surface debris, the bags were secured on the soil surface with homemade stakes to prevent tidal displacement, thus ensuring that the bags were closely arranged without overlapping. Decomposition bags were retrieved at 0, 7, 14, 21, 28, 42, 70, 103, and 149 days. The bags were washed and dried to a constant weight, and the remaining mass of the litter was measured to investigate the decomposition rates and remaining percentages for the three types of litter. The chemical properties of the litter collected at 0 and 103 days were analyzed to assess the impact of L. racemosa litter input on the quality and nutrient release rate of the native mangrove litter substrate. At 103 days of decomposition, samples from a depth of 0–15 cm beneath the three types of litter bags were collected for nutrient analysis, whereas samples at the surface beneath the bags were collected for microbial analysis. These analyses aimed to investigate the short-term effects of L. racemosa litter input on soil nutrients and microorganisms in native mangrove soils. Since the remaining mass at the end of decomposition (149 days) was insufficient for analysis, samples from the 103-day mark were selected for detailed analysis.

Leaf and soil organic C contents were determined using the potassium dichromate-volumetric method. TN was quantified using the Kjeldahl method, and TP was assessed using the molybdenum antimony anti-colorimetric method. Soil NH₄ ⁺-N was quantified using the indophenol blue colorimetric method (Bao, 2000). Leaf lignin contents and cellulose were quantified using a Multiskan™ GO multifunctional microplate reader (Thermo Fisher Scientific, Waltham, MA, USA) (Foster et al., 2010).

The soil bacterial and fungal community compositions were analyzed using the Illumina MiSeq sequencing platform. DNA was extracted from the soil samples using a MagPure Soil DNA LQ Kit (Magan). DNA concentration and purity were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) and agarose gel electrophoresis, respectively. For the bacterial diversity analysis, genomic DNA served as a template for amplifying the V3–V4 variable region of the 16S rRNA gene (using universal primers 343F and 798R). For the fungal diversity analysis, the ITS1 variable region of the ITS gene was amplified using the primers ITS1F and ITS2. The primer sequences were trimmed using Cutadapt software, and the data were processed using the default parameters of QIIME 2 software (version 2020.11). Representative sequences and amplicon-sequence variant abundance tables were obtained, and the classify-sklearn classifier was employed to annotate representative sequences against the SILVA (version 138; for 16S rRNA) and UNITE (for ITS rRNA) databases. To facilitate downstream diversity and composition analyses, each sample was rarefied to the minimum sequence depth. Species alignment annotations were further analyzed using QIIME 2 software (version 2022.2).

The following formula was used to determine the remaining mass percentage (Hu et al., 2022):

where M₀ is the initial litter mass and M_t is the remaining litter mass after time (t) of decomposition.

The litter decomposition rate was estimated using the exponential decay model proposed by Olson (1963) and defined as follows:

where X is the intercept, k is the litter-decomposition constant, y represents M_t /M₀ , a is the negative natural logarithm of X, and e is the base natural logarithm.

The formula for determining the mean relative decomposition rate (MRD) was derived from Patil et al. (2020):

where t₁ –t₀ is the sampling interval in days.

The formulas used to calculate the time required for 50% and 95% litter decomposition (Bockheim et al., 1991; Zhang et al., 2021; Tao et al., 2023) were as follows:

The collected data were analyzed using IBM SPSS Statistics 22 (IBM Corp., Armonk, NY, USA) and MS Excel (Microsoft Corp., Redmond, WA, USA). Before conducting the statistical analyses, all data were assessed for normality and homogeneity of variance. One-way analysis of variance (ANOVA) or Kruskal–Wallis tests were employed to examine the effects of L. racemosa litter on the litter mass, decomposition rate, nutrient residues, and soil nutrients. A p-value < 0.05 was considered statistically significant. Pearson correlation analysis was used to examine the relationship between the average relative decomposition rate of litter and the quality of the litter substrate and determine the correlation between microbial phyla and soil nutrient content. SigmaPlot v15.0 (Grafiti LLC., Palo Alto, CA, USA) was used to fit the litter decomposition rates and their exponential decay models. The litter mass loss at each sampling time point was calculated as the difference between the initial and remaining weights. Bacterial and fungal α-diversity were computed using the stats package in R v3.3.1 (R Foundation for Statistical Computing, Vienna, Austria), alpha diversity was measured using the Shannon index, which considers both the richness and evenness of a sample, and calculated based on a rarefied feature table (QIIME diversity core-metrics-phylogenetic; > 2000 reads). Graphs were generated using Origin 2024 software (OriginLab Inc., Northampton, MA, USA).

Many studies have shown that the input of exotic plant litter can accelerate the decomposition rate of litter in invaded areas (Liao et al., 2008; Poulette and Arthur, 2012; Chen et al., 2013). In our study, although the litter quality of L. racemosa was lower than that of R. stylosa, its litter input reduced the litter quality in the invaded sites, which led to accelerated decomposition rates ( Table 1 ) and increased nutrient release rates ( Figure 2D ). This finding is consistent with the results of previous reports and supports our first hypothesis that the input of L. racemosa litter will accelerate the decomposition rate and nutrient cycling of invaded areas.

In terrestrial ecosystems, the litter quality of exotic plants is generally higher than that of native plants and thus enhances the local litter quality (Woodworth et al., 2020; Patil et al., 2020; Meisner et al., 2012; Zhang et al., 2021). Litter quality is a crucial factor affecting decomposition (Gao et al., 2016; Cheng et al., 2024), and higher-quality litter has been reported to decompose more easily than lower-quality litter (Kennedy and El-Sabaawi, 2017; Carrasco-Barea et al., 2022). However, whether this dynamic occurs in wetland ecosystems, particularly mangrove ecosystems, has not been previously clarified. Our findings on the exotic mangrove L. racemosa were inconsistent with those of previous studies. The litter of L. racemosa has higher C:N and C:P ratios than the litter of the native mangrove species R. stylosa, indicating that it has lower-quality litter. However, the low litter quality of L. racemosa accelerated its decomposition in the invaded area ( Table 1 ), which is contrary to previous findings. Li et al. (2020) also highlighted the low-quality characteristics of L. racemosa litter. Under similar climatic conditions, litter decomposition is regulated by litter quality and decomposer communities (Patil et al., 2020; Zhang et al., 2021; Cheng et al., 2024). Thus, we sought to determine the factors underlying the faster decomposition rate of the lower-quality L. racemosa litter. An analysis of the soil microbial community structure revealed that the input of L. racemosa litter significantly increased the relative abundance of microbes related to litter decomposition, such as Proteobacteria and Bdellovibrionota, and enhanced the alpha diversity of soil fungi. Thus, a microbial environment conducive to the decomposition of L. racemosa leaf litter was established. This issue will be further discussed in the following section.

We found that the input of L. racemosa leaf litter altered the soil microbial community in native mangroves and stimulated the growth of certain dominant bacteria and fungi associated with litter decomposition, such as Proteobacteria, Bdellovibrionota, and Ascomycota. These microbial groups play a crucial role in accelerating leaf decomposition (Gladkov et al., 2022; Andronov et al., 2022; Kimeklis et al., 2023) and presented a higher relative abundance after the input of L. racemosa litter; thus, they accelerated the decomposition of the L. racemosa litter. The long-term cumulative effects indicated that L. racemosa litter decomposition also increased the soil fungal α-diversity. Higher microbial diversity supports elevated levels of ecosystem functions, which further facilitate the decomposition of L. racemosa litter. Correlation analyses further revealed that certain soil bacteria and fungi were significantly related to the MRD and nutrient content of the soil after litter decomposition ( Supplementary Tables S2 , S3 ), indicating a correlation between the faster decomposition rates of L. racemosa leaf litter and the structure of the soil decomposer community. This supports our hypothesis that the input of L. racemosa litter alters the microbial community structure in the invaded area. Additionally, these results suggest that a key reason for the faster decomposition rate of L. racemosa litter compared to R. stylosa litter may be changes in the soil microbial community structure caused by the input of L. racemosa litter. One of the ecological impacts of exotic plant dispersal is the alteration of soil microbial communities, which affects the diversity, abundance, and functionality of microbes, thereby leading to changes in litter decomposition rates and soil nutrient availability (Torres et al., 2021). Specific types of litter have distinct morphological and chemical characteristics that can act as ecological filters that select for or exclude certain microbial groups from the shared soil pool. Variations in the chemical exudates from litter can also directly influence soil microbial communities (Waldrop and Firestone, 2004). Therefore, the input of litter from exotic plants with different morphological and chemical traits leads to alterations in the local soil microbial community composition. Microbial communities with higher abundance and diversity often support higher levels of ecosystem functions, such as biomass decomposition, by facilitating interactions and resource allocation among microbial species (Zhang et al., 2023), thereby accelerating the decomposition process.

A meta-analysis of the effects of invasive plants on soil microbial communities revealed that exotic plants can enhance soil nutrient cycling by altering soil microbes, thus promoting invasion success (Hou et al., 2022). Our study revealed that after the input of L. racemosa leaf litter, the α-diversity of soil fungi increased. Changes in the relative abundance of dominant bacteria and fungi associated with leaf litter decomposition led to increased decomposition rates and nutrient release from native mangrove leaf litter. These findings are consistent with those of previous studies on invasive plants (Arthur et al., 2012; Yan et al., 2018; Zhang et al., 2023) and suggest that L. racemosa shares similar characteristics with invasive species in terms of modifying soil microbes to enhance soil nutrient cycling. This finding indicates the potential invasiveness of L. racemosa.

Research on the impact of exotic plants on soil nutrient cycling is important for evaluating the effects of these plants (Ehrenfeld, 2003). Exotic plants often have higher leaf litter quality and decomposition rates, leading to increased nutrient content in the soil through leaf decomposition, which alters ecosystem nutrient cycling and promotes exotic plant adaptation and growth (Meisner et al., 2012; Patil et al., 2020; Rothstein et al., 2004). However, our experiments showed that the short-term and long-term inputs of L. racemosa leaf litter resulted in decreased levels of carbon, nitrogen, and ammonium nitrogen in the soil of the invaded native mangrove forests, which is inconsistent with the increase in nutrients typically observed in studies of exotic plant litter (Sanon et al., 2012; Bray et al., 2017).

An area of research is why L. racemosa exhibits rapid litter decomposition but causes lower soil nutrient contents. Although the leaf litter of L. racemosa decomposes quickly, it has low quality and nutrient contents. However, L. racemosa is a tall fast-growing species, suggesting that its rapid growth requires substantial nutrients. Thus, the high amount of nutrients released from the decomposing leaf litter may be absorbed by L. racemosa plants to support their growth and development. This hypothesis was validated by Zhu et al. (2023), who performed genomic analyses of the leaves of L. racemosa from Dongzhaigang (Hainan Province) and discovered convergent evolution in genes responsible for nutrient absorption, which enhanced N and P uptake and transport capacity. The results of Liu et al. (2020) further supported the hypothesis that the introduced mangrove L. racemosa predominantly absorbs and enriches nutrients in plant tissues. They observed significantly higher richness in rhizospheric and endophytic diazotrophic communities as well as rhizospheric nutrient components (such as TN) in L. racemosa mangroves than in native mangroves, thus providing evidence based on the rhizospheric microbiota and nutrition. In our study, we also discovered that Alphaproteobacteria, which facilitates N fixation in plants, was significantly more abundant in L. racemosa forests than in mixed and R. stylosa forests, thus providing further support for our hypothesis. Therefore, the rapid decomposition of L. racemosa litter and lower soil nutrient contents might be attributed to the retention of nutrient elements in L. racemosa plants to facilitate their own rapid growth and development. In summary, inputting exotic mangrove L. racemosa litter accelerated nutrient cycling within mangrove ecosystems. The rapid decomposition of L. racemosa litter and the lower soil nutrient contents could reflect the absorption and retention of released nutrients to promote their rapid growth and development. Ecosystem nutrient cycling is a continuous process involving leaf litter decomposition, soil nutrient increase, plant growth, and subsequent leaf litter decomposition. The soil nutrient contents depend not only on the amount of nutrients released during leaf litter decomposition but also on the amount of nutrients absorbed and fixed within the plant biomass, thus reflecting a balance between nutrient inputs and outputs. Due to the rapid growth and high biomass of exotic plants, more nutrients are sequestered in their larger structures, leading to reduced nutrient levels in the soil. Compared to native plants, these larger exotic plants can gain a competitive advantage based on their enhanced absorption of sunlight and uptake of soil nutrients, which further facilitate their spread into new areas.

Many studies have assumed that the impact of exotic plants on soil nutrient cycling enhances the competitive advantage of these plants within ecosystems (Yelenik and Levine, 2011) because their high-quality litter increases soil nutrients, thereby promoting their own biomass accumulation and reproductive growth. Such increased nutrient availability may bolster their competitive ability and contribute to their successful dispersal (Patil et al., 2020; Hou et al., 2022). However, contrary to reports on other exotic plants (Meisner et al., 2012; Rothstein et al., 2004), our experiments revealed that litter inputs from L. racemosa actually led to a decrease in soil nutrient levels. Similar outcomes were observed for the invasive plant Agropyron cristatum (Christian and Wilson, 1999). This finding raises the questions of how the reduction in soil nutrients caused by L. racemosa litter aids the spread of this plant and what advantages are conferred by this strategy. In low-resource environments, the competitive ability of invasive plants is often influenced by their ability to enhance resource availability and tolerate low-resource conditions (Funk, 2013). Adaptations to low-resource ecosystems are frequently associated with traits related to resource acquisition or conservation (Chapin, 1980; Craine, 2009). Previous research indicates that the genetic and rhizosphere microbial characteristics of L. racemosa enhance its ability to acquire nutrients and tolerate low-nutrient conditions (Zhu et al., 2023; Liu et al., 2020), potentially strengthening its competitive ability in nutrient-poor environments. Therefore, we hypothesized that L. racemosa may employ a unique invasion strategy in which it reduces soil nutrient levels in native mangroves through low-quality litter inputs. This reduction in quality reduces the competitive advantage of native mangrove species intolerant to low-nutrient conditions but enhances the competitive advantage of L. racemosa, which is tolerant of low-nutrient environments (Liu et al., 2020). Similarly, Leary et al. (2006) found that litter of the invasive plant Ulex europaeus causes lower soil nutrient contents than that of native plants; thus, U. europaeus exhibits a competitive advantage over Panicum virgatum under nutrient-poor conditions.

The carbon content of mangroves is among the highest relative to other ecosystems (Friess et al., 2021; Macreadie et al., 2021); thus, mangroves play a crucial role in global carbon capture and storage (Lu et al., 2024). Nutrients from leaf litter decomposition are a primary source of organic matter in mangrove soils (Kristensen et al., 2008). Our single decomposition experiment and cumulative effect experiments demonstrated that the input of L. racemosa leaf litter decreased the soil organic carbon content. This reduction in soil organic carbon may be due to changes in microbial community structure caused by L. racemosa leaf litter, which enhanced microbial respiration and accelerated the conversion of soil carbon to carbon dioxide (CO₂), leading to a decrease in organic carbon sequestration in the soil. Exotic plants, especially invasive ones, typically have high biomass and rapid growth rates, and L. racemosa is no exception (Li et al., 2020; Feng et al., 2022). These high-biomass, fast-growing plants return more organic carbon to the soil each year, potentially leading to increased carbon sequestration within the soil ecosystem (Cheng et al., 2006). Traditionally, such plants have been considered high carbon-sequestration species. However, in addition to plant biomass, the decomposition rates and patterns of leaf litter are also crucial factors in determining soil carbon storage. If high-biomass plants experience elevated levels of mineralization during litter decomposition, then additional carbon will be released back into the atmosphere, resulting in less organic carbon retained in the soil (Yan et al., 2018). Studies on terrestrial and wetland ecosystems indicate that the input of exotic plant litter can alter the soil microbial community structure by increasing the proportion of microbial communities responsible for mineralization (e.g., Proteobacteria), which can affect organic carbon decomposition patterns. This change can lead to increased soil respiration and CO₂ emissions, thereby reducing soil organic carbon storage (Meisner et al., 2012; Spirito et al., 2014; Zhang et al., 2014; Yan et al., 2018). Our results indirectly support this mechanism.