The home-field advantage (HFA) is a well-known phenomenon in decomposition processes, in which plant litter has the tendency to decompose faster in its ‘home’ environment than in its ‘away’ environment even when in a similar abiotic environment (Hunt et al., 1988; Gholz et al., 2000; Ayres et al., 2009; Zheng et al., 2023). The phenomenon shows that the decomposition rate of litter is significantly different under different habitat conditions. Various studies have directly targeted the existence (or absence) of difference in mass loss between different habitat conditions in different types of ecosystems, such as forests, grasslands or wetlands (Gholz et al., 2000; Palozzi and Lindo, 2018; Yeung et al., 2019; Yuan et al., 2019), using either field or greenhouse experiments with different litter types (Purahong et al., 2019; Veen et al., 2019; Wu et al., 2019). However, the difference in decomposition rate is still inconsistently supported by empirical data (Chapman and Koch, 2007; Gießelmann et al., 2011; Freschet et al., 2012; Bachega et al., 2016; Ma et al., 2023), and very few studies have tested difference in mass loss between different habitat conditions across large spatial scales in the field (but see Veen et al., 2019 in a greenhouse experiment). It is still unknown whether difference in mass loss between different habitat conditions in the decomposition process of certain plant species still exists in the field across a large spatial scale.

The existence of difference in mass loss might depend on the dissimilarities in multiple factors between different habitat conditions, such as climatic conditions, aboveground plant species and belowground soil microclimatic conditions or decomposer organisms (Zhu et al., 2024) (Figure 1). One important hypothesis predicts that the difference may result from the local adaptation of the soil organisms to decompose litter from the plant species above them (Bardgett and Walker, 2004; Wardle, 2013). This adaptation might alter physiological activities, shifts in community composition, evolution, or a combination of these processes (Milcu and Manning, 2011). Evidence showed that all size classes of soil fauna might contribute to difference in mass loss (Li X et al., 2021), and large dissimilarities in soil microbial communities between different habitat conditions might result in stronger difference. However, for certain plant species at a large spatial scale, whether different groups of soil fauna or microbes, as well as other environmental factors, might affect the difference in mass loss in decomposition has remained unknown. Especially for soil microbes, it is virtually unknown which groups of soil microbes really matter in driving the difference across large spatial scales.

Previous studies have shown that difference in mass loss was mostly driven by soil microbial communities, especially by soil fungi (Austin et al., 2014; Lin et al., 2019; Veen et al., 2019; Li et al., 2020; Liu L. et al., 2023). More specifically, fungi might contribute to the difference in mass loss, even though possibly only some specific dominant fungi matter to decomposition rather than the whole fungal community (Lin et al., 2019; Veen et al., 2019; Ma Y et al., 2023). Fungal communities are mainly considered as primary decomposers, because many of them can produce a wide range of extracellular enzymes and degrade recalcitrant organic compounds in plant litter (Boer et al., 2005; van der Wal et al., 2012; Voriskova and Baldrian, 2013; Liu et al., 2022). However, we still lack a deeper understanding about the functional role of fungi in driving the difference in decomposition rate. It remains unknown which functional groups or guilds of soil fungi really matter in driving the difference in mass loss. Two prominent publications could help us to further categorize the soil fungal community into different functional groups or guilds, i.e., Pathotrophs, Saprotrophs, Symbiotrophs or their combinations (Nguyen et al., 2016; Põlme S et al., 2020), and hence help to answer the above question.

In terms of testing the difference in mass loss across large spatial scales, P. australis might be a good candidate species, because P. australis as a cosmopolitan grass is dominant in many wetland ecosystems and widely spread both globally and locally (Brix et al., 2014; Hu et al., 2017; Yin et al., 2022; Liu S. et al., 2023). Moreover, P. australis can expand or invade into new communities, with various consequences to ecosystem functioning (Meyerson et al., 2010; Cui et al., 2019). Such an expansion or invasion of P. australis into other plant communities can lead to negative impacts on the local ecology, most notably through decreased biodiversity (Wails et al., 2021) or it can alter the decomposition of other plant species (Cui et al., 2019), which may indirectly also affect biodiversity. This might have important implications for the testing of difference in mass loss in P. australis litter decomposition across large spatial scale: whether an expansion or invasion by P. australis might have ecological consequences via the difference in mass loss across large spatial scales.

Here we conducted a large-scale decomposition experiment (ranging from 26° N to 41° N in latitude) to examine the leaf litter decomposition rates of P. australis, and tested whether it showed difference in mass loss at sites under P. australis communities, compared to the decomposition rates at sites under another respective dominant plant species across Chinese coastal regions. This experimental design ruled out the litter quality effect on testing the difference (Milcu and Manning, 2011), and simplified our study system to better explore the role of the dissimilarity in belowground biota in driving the difference. Therefore, we hypothesized that (1) the difference in mass loss should mainly depend on the dissimilarities in microbial communities and soil properties between different habitat conditions; (2) other factors such as different soil fauna groups and different stages of decomposition might also matter for the strength of the difference in decomposition across large spatial scales. In addition, we were able to test which taxonomic group(s) or functional guild(s) really matter when testing the difference in a large spatial scale along the Chinese coast. This might improve our understanding about the underlying mechanisms that drive the difference in mass loss and provide important implications for linking soil microbial community to ecosystem functioning.

Our 24 sites ranged across 15 degrees of latitude. Soil TOC ranged from 0.081% to 3.144%; soil TN ranged from 0.026% to 0.270%; soil TP ranged from 0.027% to 0.096; soil pH ranged from 6.16 to 8.89 (Supplementary Table S1). Soil fungal diversity ranged from 53 to 374 in terms of Sobs diversity index, 0.230 to 4.109 in terms of Shannon diversity index, 53.1 to 389.0 in terms of Chao diversity index, 1.0 to 24.4 in terms of Inv-simpson diversity index and 27823 to 65572 in terms of number of OTUs; soil bacteria diversity ranged from 1,352 to 3,022 in terms of Sobs diversity index, 3.744 to 7.086 in terms of Shannon diversity index, 1709.6 to 4,178.1 in terms of Chao diversity index, 6.4 to 565.3 in terms of Inv-simpson diversity index and 20199 to 49274 in terms of number of OTUs (Supplementary Table S2).

There were significant differences in mass loss after 9 and 12 months between between different habitat conditions (Figures 2B, C, p = 0.02 & p = 0.03), but only a trend in mass loss difference after 3 months (Figure 2A, p = 0.09). When separating other species sites into different aboveground plant species, there were significant differences in the mass losses after 9 and 12 months between PA and SS, and between PA and CM (Figures 2E, F, p < 0.01), while there was no site effect after 3 months (Figure 2D). Mesh size had significant effects on litter mass loss of all three harvests (Figure 2, p < 0.01), but no significant effects on differences in mass losses between different habitat conditions across all three harvests.

There was an overall positive correlation between differences in mass loss and the Bray-Curtis distances in initial soil fungal communities between different habitat conditions (Figure 3 left panel, p < 0.01), and this correlation remained for differences in mass loss after 9 months (Figure 3 left panel, p < 0.01) and differences in mass loss after 12 months (Figure 3 left panel, p < 0.01). However, there were no significant correlation between differences in mass loss and the Bray-Curtis distances in bacterial communities between different habitat conditions (Figure 3 right panel, p > 0.05). In addition, there was no significant correlation between differences in mass loss and the dissimilarities in climatic factors or soil properties (Supplementary Table S3).

As to different functional groups of fungi, there were significant positive correlations between differences in mass loss and the dissimilarities in Symbiotroph composition (Figure 4, R = 0.64, p < 0.01). There were no significant correlations between differences in mass loss and the other two functional guilds, i.e., Pathotrophs and Saprotrophs (Figure 4, p > 0.05). Moreover, there were significant correlations between differences in mass loss and the dissimilarities in the following taxonomic groups: Dothideomycetes (Supplementary Table S4, p < 0.01), Agaricomycetes (Supplementary Table S4, p < 0.01) and Eurotiomycetes (Supplementary Table S4, p = 0.03).

Using a litter transplantation experiment, we tested the existence of difference in mass loss of P. australis litter decomposition across a large spatial scale within coastal wetlands. Our results showed that there were clear and consistent differences in mass loss and that this difference mostly depended on the dissimilarities in initial soil fungal communities between different habitat conditions, but not on those in soil bacterial communities. Moreover, we found that difference tended to be more evident at later stages of decomposition, and could be better explained by the dissimilarities in soil symbiotrophic fungi between different habitat conditions, rather than by those in soil pathotrophic or saprotrophic fungi. Below we discuss our findings and highlight the implications of the observed large-scale difference in coastal wetlands as well as the possible underlying microbial-driven mechanisms.

The widespread distribution of P. australis in coastal wetlands has allowed us to test the generality of differences in the mass loss across a spatial scale. Overall, we did find a difference in the decomposition of P. australis across Chinese coastal areas and the magnitude of difference increased with the dissimilarities in initial soil fungal communities between different habitat conditions. This may be due to the fact that there is a home-field advantage (HFA) in the process of litter decomposition. These results were consistent with multiple recent studies, indicating that soil fungi play a vital role in driving the HFA (Miura et al., 2015; Purahong et al., 2019; Ma et al., 2023). Moreover, our results also showed that increasing dissimilarity in the dominant fungal taxa, i.e., Dothideomycetes, Agaricomycetes, and Eurotiomycetes between different habitat conditions soils corresponded with stronger differences in mass loss. Previous studies have also shown that the existence or absence of HFA in decomposition might be related to some specific taxa of fungi but not all fungi (Lin et al., 2019; Veen et al., 2019), or depended on different body sizes of soil microbes (Li et al., 2020). However, no previous studies have tried to link different functional groups of soil fungi, i.e., fungal guilds, to the differences in mass loss. Our results have shown that the dissimilarity in composition within the Symbiotroph group between different habitats was positively correlated with the observed difference in mass loss, while the differences in mass loss was not significantly correlated with compositional dissimilarities within either the Pathotroph or Saprotroph group (Figure 4; Supplementary Figure S2). Symbiotrophic and saprotrophic fungi might affect plant litter decomposition differently, and symbiotrophic fungi might indirectly regulate litter decomposition rates by restraining activities of more efficient litter saprotrophs (Bödeker et al., 2016). Moreover, some of the symbiotrophic fungi associated with plant roots, are capable of oxidizing organic matter (Lindahl and Tunlid, 2015; Nguyen et al., 2016; Choreño‐Parra and Treseder, 2024). Therefore, we hypothesize that symbiotrophic fungi associated with the P. australis roots versus those of other dominants might be a predominant driver for the observed difference in mass loss. To take this further, in-depth study should specifically test the extent to which differences is linked to dissimilarities in the extent and genetic/functional composition of symbiotrophic networks in sites dominated by P. australis versus other dominants. Glomeromycota (Glomeromycetes), the taxon representing symbiotrophic fungi, were not among the dominant fungi in our experiment, but our in-depth exploration revealed that they might still have an important role to play in differences in mass loss in our study: the dissimilarities in composition within the Glomeromycetes between different habitat conditions were positively correlated with the observed difference in mass loss (Supplementary Figure S3). This could give symbiotrophic fungi either a direct role in differences in mass loss or an indirect role via influencing the abundance of the saprotrophic taxa that were involved in our study (e.g., Agaricomycetes).

However, our results only partially supported our first hypothesis, because we found that the magnitude of difference did not change with the dissimilarity of soil bacterial communities between different habitat conditions. These results might be due to bacteria having much faster turnover in composition as related to their short lifespans. Thus soil bacteria at t₀ may not represent the community composition well through decomposition time, while for fungi t₀ sampling may have been more representative of the entire decomposition period. Note that our study only focused on whether the microbial communities in the initial soil, rather than those on the remaining litter, can explain the observed difference. Both sources of microbial community measures are important indicators for understanding the microbe-driven decomposition process: the former focused on examining the role of soil origin in driving HFA (Elias et al., 2020), while the latter focused on quantifying the contribution of microbe colonization on the remaining litter to HFA (Lin et al., 2019) and hence might be more sensitive to litter qualities. Examining the dissimilarity in soil fungal communities might provide more direct evidence to test the soil microbial adaptation hypotheses in HFA studies (Bardgett and Walker, 2004; Ayres et al., 2009; Wardle, 2013). Our experimental design ruled out the effects of litter quality on the existence of difference in mass loss; instead, it zoomed in on the belowground soil biota or incubation environmental drivers on difference.

The existence or absence of HFA in decomposition might depend on different stages of decomposition (Ayres et al., 2009; Güsewell and Gessner, 2009; Purahong et al., 2019) and on the identity of the species growing at other habitat conditions, but not on the dissimilarity of soil properties or different soil fauna groups in coastal habitats. As for the latter, the presence of mesofauna appeared not to interfere with the relationship between HFA and fungal community composition, as evidenced indirectly by a lack of interaction with the mesh size treatment. Our results for decomposition stage showed that there were significantly higher differences in mass loss after 9 and 12 months of decomposition (Figure 1), and the strength (regression coefficient) of the relationship between difference in mass loss between different habitat conditions and dissimilarity of initial soil fungal communities only increased at later stages of decomposition (Figure 3). These results might be explained by most fungi being considered to have low nutritional needs and especially tending to decompose recalcitrant compounds, and therefore might be better adapted to later decomposition stages with poor-quality litter substances. Different decomposing fungi in the soil might have strong inertia and only attack the most recalcitrant compounds such as lignin at later stages of decomposition (Perez et al., 2013). On the other hand, there might be a large amount of variation in HFA among plant species, as was shown previously through reciprocal transplants between different plant species (Ayres et al., 2009). If we investigate different plant species at other dominant species sites in our study, there was only a significant HFA effect when comparing PA sites with SS and CM sites. This might be due to the microbial community between PA and SS or CM being more distinct than those between PA and other dominants (Supplementary Figure S1). More distinct decomposer communities in ‘home’ sites may generate stronger HFA effects than when different habitat conditions soils have a similar composition (Veen et al., 2015a). In addition, our results showed that dissimilarities in soil properties could not explain the difference in mass loss either, and different groups of soil fauna (as deduced from differences in differences in mass loss between mesh sizes) did not affect the relationship between initial soil fungi dissimilarity and the difference in the mass loss (Figure 3). These results re-emphasize that, among soil biota, it is mainly soil fungi that play a predominant role in driving the HFA in decomposition.

Our results have provided empirical evidence on the existence of a large-scale difference in the mass loss of P. australis litter across coastal wetlands, and has linked the difference in the mass loss with different initial soil fungal guilds, notably symbiotrophic fungi and, to a lesser extent, saprotrophic fungi. These results might improve our understanding about nutrient cycling within coastal wetlands, and provide a potential mechanism via different functional groups of soil fungi to explain the inconsistent results about the difference in mass loss across different types of ecosystems. To further reveal the microbe-driven difference in mass loss, we suggest that similar large-scale decomposition should be conducted in other types of ecosystems. Such future studies should further investigate the microbial community in more detail and try to quantify the relative contribution of different soil fungi to difference in mass loss, with particular emphasis on the separate and interactive contributions of symbiotrophic and saprotrophic fungi.