Mangrove wetlands represent dynamic coastal interfaces where redox conditions and nutrient cycling shape microbial communities and their biogeochemical functions. However, tidal-driven siltation continuously transports sediment from low- to high-tide zones, altering sediment redox potential and nutrient content. The microbial responses to these changes, particularly the response mechanisms of carbon-fixing microorganisms, remain unclear. We integrated metagenomic and metatranscriptomic sequencing with 13C and 15N isotope labeling to examine how oxidation-reduction potential (ORP) and ammonium (NH4+) availability regulate microbial assembly and metabolism in mangrove sediments. ORP emerged as the primary determinant of microbial composition and diversity, while NH4+ exerted variable effects on microbial traits. Under high ORP, CBBL-microorganisms predominantly utilized the cmmG CO2-concentrating mechanism, and carbon fixation rates decreased with increasing NH4+ concentration. Under low ORP, CBBM-harboring genera dominated, primarily utilizing another mechanism cmmE, while NH4+ had little effect and total organic carbon (TOC) exerted stronger control. ORP thus acts as the dominant environmental filter, with NH4+ selectively affecting nitrifiers and carbon-fixing taxa. This was the first study to simultaneously measure ammonia oxidation and carbon fixation rate in mangrove sediments, revealing their mechanistic coupling. This work provides new mechanistic insights into the regulation of microbial metabolic potential in mangrove ecosystems and contributes to a broader understanding of their resilience and function in coastal biogeochemical cycles under fluctuating climate and environmental conditions.

Mangroves are characterized by high carbon sequestration rates, and future changes in mangrove biomass will be impacted by factors such as climate change, sea level rise, and management strategies. Here, we produce a map of present-day global mangrove aboveground biomass (AGB) based on extensive field observations and satellite data, giving a global mangrove AGB of 1.76 Pg dry matter (DM). After accounting for potential growth to maturity, future changes in climatic and hydrological conditions, possible restoration strategies, and sea-level rise, the global total mangrove AGB is projected to increase by 16.7% under the low-warming scenario (Shared Socioeconomic Pathway 1 and Representative Concentration Pathway 2.6, SSP126) and decrease by 19.3% under the high-warming scenario (SSP585) by 2100. Sea-level rise encroaching upon the growth space of mangroves will cause a biomass loss of 41.6% in the high-warming scenario, whereas restoration strategies would only increase AGB by 2.0%. Our study shows that sea-level rise limits the potential of mangrove restoration to maintain carbon stocks, and regions less constrained by sea-level rise offer greater potential for the long-term permanence.

Abstract Accurate estimation of mangrove ecosystem carbon stocks is essential for effective blue carbon management. Significant interspecific variations in carbon storage capacity and estimation methods arise due to species-specific biophysical characteristics, highlighting the need for precise mangrove species identification and species-level carbon stock assessment. However, limited studies assessed mangrove carbon stocks at species-level. This study, conducted in the Gaoqiao Mangrove Nature Reserve in Zhanjiang, Guangdong Province, applied UAV multispectral technology to simultaneously acquire spectral, structural and textural vegetation feature variables for mangrove species identification and established species-specific carbon stock models, thereby achieving species-level carbon stock estimation. Results showed that (1) by integrating spectral and structural features, the study achieved 89.87% overall accuracy in species identification. (2) Species-level carbon stock estimation models, incorporating spectral, structural and textural feature variables alongside field-measured carbon data, demonstrated strong predictive performance (R2 = 0.48-0.95). (3) The most effective vegetation feature variables for carbon estimation varied significantly across species, emphasizing the necessity of accounting for species heterogeneity in mangrove carbon stock estimations. (4) Carbon stocks exhibited significant interspecific variation, with Rhizophora stylosa demonstrating the highest aboveground (97.06 t hm⁻2) and belowground (37.22 t hm⁻2) stocks, compared to Aegiceras corniculatum’s minimum values of 49.14 and 19.88 t hm⁻2, respectively. This study established a UAV-based multispectral framework for mangrove species-level carbon stock estimation and provided new insights for mangrove carbon assessment and management by demonstrating the importance of considering species-specific influences on carbon stocks and their estimation.

Mangrove wetlands play a vital role in global nitrogen (N) cycling, with microbially-driven nitrate reduction as a key pathway. Although phages are ubiquitous in mangrove ecosystems and may profoundly influence nitrogen reduction, direct experimental evidence for their regulatory mechanisms remains lacking. In this study, we established microcosms with three different phage treatments (with native phage, non-native phage, or no phage) in order to systematically investigate the effects of phages on different nitrate reduction pathways in mangrove soils. The results showed that phages significantly reduced the relative contribution of dissimilatory nitrate reduction to ammonium (DNRA) to total nitrate reduction, whereas the absolute DNRA rate showed only a non-significant decreasing trend. This change in process partitioning shifted the dominant nitrate reduction process from DNRA toward denitrification. Such a transition was likely associated with phage-induced lysis of DNRA bacteria and subsequent nutrient depletion. Moreover, native and non-native phages have a certain promoting effect on denitrification, potentially through distinct mechanisms. Native phages enhanced nirS-type denitrifiers via niche specialization and deterministic selection, but non-native phages indirectly stimulated nirK-type denitrifiers by lysing DNRA bacteria to release resources. These distinct mechanisms underscored, for the first time, how phage source shaped divergent microbial responses in nitrate reduction. Anammox exhibited low rates and contributed minimally to nitrate reduction, with no detectable response to phage addition under our experimental conditions. The integrated approach, combining phage inoculation, 15N isotope tracing and microbial profiling, enabled direct experimental resolution of phage impacts on nitrogen reduction, which was a major knowledge gap in the literature. This study could enrich our understanding of virus-microbe interactions and offering theoretical support for nitrogen regulation in coastal ecosystems.

The mobilization of abundant but immobilized phosphorus into bioavailable forms is conventionally attributed to microbial metabolite production. Yet the persistence of reductive transformations in hydrodynamically active environments with limited microbial metabolism implies a previously overlooked abiotic pathway. Here, we reveal hydrodynamic-driven phosphorus mobilization via a piezoelectric energy-transduction pathway. Hydrodynamic forces enhanced phosphate release 3-fold in natural intertidal sediments, whereas ultrasonic treatment further elevated the release through processes operating independently of microbial activity. Mechanistic investigation confirmed that piezoelectrically generated electrons directly reduce Fe(III) species, releasing mineral-bound phosphate while simultaneously facilitating secondary mineral formation. This pathway fundamentally challenges the established paradigm of biologically dominated phosphorus cycling and establishes hydrodynamic energy as a primary driver of phosphorus transformation. Given the global distribution of hydrodynamic energy and piezoelectric minerals, this mechanism likely operates worldwide, particularly under climate-intensified hydrodynamic conditions. These processes may elevate aquatic nutrient fluxes and eutrophication risks, highlighting the need to quantify their environmental significance and develop management strategies. In addition, our findings could advance the interpretation of historical phosphorus cycling while enabling sustainable phosphorus recovery through piezoelectric processes.

Phages have garnered increasing attention due to their potential roles in biogeochemical cycling. However, their impacts on nitrogen cycling have primarily been inferred from the presence of putative auxiliary metabolic genes (AMGs) and the virus-host linkage, despite of very limited direct experimental evidence. In this study, a series of microcosms were established with the inoculation of either native or non-native phages to simulate coastal wetlands with different phage sources and different levels of copper (Cu) contamination. Metagenomics and metatranscriptomics were combined to reveal phages' regulation on microbially-driven nitrogen cycling and to explore how the effects were mediated by Cu stress. Phages significantly impacted denitrification-related genes, with their effects depending on Cu level. Phages inhibited nirK-type denitrification under Cu stress but led to up-regulation of nirS gene in the treatments without Cu addition. Non-native phages also promoted the transcription of genes related to nitrogen assimilation and organic nitrogen transformation. Detection of viral AMGs involved in glutamate synthesis suggested that horizontal gene transfer may be a crucial pathway for phages to facilitate microbial nitrogen uptake. Overall, these findings enhance the understanding of phages' impact on biogeochemical metabolism in coastal wetland, offering novel insights into the links of phages' regulation on microbial nitrogen cycling with Cu stress.

An aerobic phosphorus-recovering strain, designated WS-203T, was isolated from mangrove sediment. Strain WS-203T showed the highest 16S rRNA gene sequence similarity to the type strain of Pelagibacterium lacus (98.1

Mangrove ecosystems, renowned for their carbon sequestration capabilities, also contribute to the emission of greenhouse gases from their sediments. They are under stress from human activities and global changes, such as heavy metal pollution and rising sea levels, which can affect carbon stability and greenhouse gas emissions. However, the combined impact of these factors is not fully understood to date. This study assessed the impact of cadmium contamination at varying levels of inundation depths on the emission of CO2, N2O, and CH4 from mangrove sediments. The findings indicated that increased inundation depth intensifies the effect of low cadmium concentrations on CO2 emissions, with sediment CO2 flux doubling from 3430 mu mol m- 2 h- 1 to 6720 mu mol m- 2 h- 1. In contrast, deeper water levels reduced N2O emissions in the presence of high cadmium concentrations, lowering sediment N2O flux from 14.10 mu mol m- 2 h- 1 to 10.13 mu mol m- 2 h- 1. These results highlight that the combined effects of rising inundation depth and cadmium pollution can enhance greenhouse gas emissions from mangrove sediments. It is crucial to consider the synergistic effects of pollution and rising sea levels on the stability of sediment carbon and gas emission rates in the conservation and management of mangroves. Such assessments will bolster efforts to protect blue carbon and achieve carbon neutrality in the future.

Plant functional traits, the key to driving carbon cycling and the energy balance of ecosystems, are widely demonstrated associated with environments across a wide range of biomes. However, it is unknown whether global trait-environmental relationships extend to climatic extremes. We tested if these relationships hold in the environmental extremes of forest on Earth using a compiled database of mangrove plant leaf traits from 338 individuals across 58 species, 71 sites worldwide. We show that mangroves were clustered at the conservative end along the leaf economic spectrum axis of global vascular plants. Mangrove leaf traits were significantly (p < 0.05) correlated with salinity, tidal current velocity and mean temperature diurnal range (Bio2), and were seen to undergo changes in response to future environmental changes. Trait-environmental relationships are thus generalizable to the tidal-dominated forest, informing prediction of mangrove ecosystem function shifts in a changing world.

Mangrove wetlands are crucial for carbon sequestration, however, the contributions of bacterial carbon fixation in these ecosystems are often overlooked, and the predominant pathways remains unknown. This gap seriously hinders the understanding and precise assessment of carbon sequestration. This study systematically investigates the pathways, rates, and influential factors of bacterial carbon fixation in mangrove wetlands, utilizing soils from various tidal zones and depths. Through an integrated approach that combines in situ metagenome sequencing, 13CO2 tagging experiment, functional gene abundance measurement, and 16S rRNA sequencing, we provide the first evidence that the reverse tricarboxylic acid cycle is the predominant pathway for carbon (C) fixation in mangrove soils. The mangrove ecosystem was identified as a significant hotspot for bacterial carbon fixation, with rates in topsoil ranging from 15 to 63 mmol C/(m²·day), significantly influenced by environmental variables such as oxidation-reduction potential, and ammonium and nitrate concentrations. In deep soils, high carbon fixation rates were detected in low tidal zones but not in middle and high tidal zones, which did not align with the abundance of carbon fixation functional genes. Notably, we found a strong correlation between carbon fixation rates and nitrogen metabolism processes, underscoring the ecological interactions between these biogeochemical cycles. These findings greatly enhance our understanding of microbial contributions to carbon cycling in mangrove ecosystems and offer novel insights into blue carbon sequestration and the management of coastal wetlands under varying environmental conditions.

Mangrove wetlands in China are naturally expanding seaward due to rapid sediment accretion, yet the dynamics of species competition and stand structure during this expansion remain unclear. Here, we developed an integrated remote sensing framework combining satellite imagery, unmanned aerial vehicle (UAV) data, and deep learning to investigate species composition and stand structure in China's largest contiguous mangrove forest. The algorithm achieved high accuracy in identifying two dominant species (Aegiceras corniculatum and Avicennia marina, overall accuracy: 87.7 %) and extracting individual crown parameter (precision > 70 %). Results show that A. corniculatum initially colonizes new mudflats but is later outcompeted by A. marina, forming monospecific stands over time. Interspecific competition intensity peaks during early succession (1-7 years) and declines with stand age, accompanied by a shift in spatial distribution from random to uniform. These findings reveal a distinct pattern of sequential species replacement and community succession during mangrove seaward expansion. The proposed framework and ecological insights provide valuable guidance for near-natural mangrove afforestation and the restoration of degraded coastal ecosystems, contributing to sustainable wetland conservation strategies.

Afforestation of mangroves in abandoned ponds is considered an effective approach for global mangrove restoration. Although several studies have assessed recovery outcomes of reforested mangroves, the studies about the synergistic recovery of mangrove vegetation and ecosystem carbon stocks in abandoned ponds are still limited. The present study compared the vegetation structure and carbon stocks of pond-to-mangrove sites with different restored methods in Dongzhaigang Bay, Hainan Province, China. We found that the vegetation structure (tree density and Shannon diversity) had no significant differences between afforestation sites and natural mangroves. The total ecosystem carbon stocks of all the sample sites ranged from 64.07 to 224.91 (mean +/- SD: 114.82 +/- 49.30) Mg C ha(-1), while there were no significant variations between restoration sites in 2014 (p > 0.05). In addition, 85.44 % of the total carbon stocks were stored in the soil among the study sites. The soil physicochemical properties such as soil pH, total organic carbon, total nitrogen had effects on tree height, diameter at breast height and carbon stocks composition. Meanwhile, in-situ soil redox potentials, Shannon diversity, tree height, density and diameter at breast height affected carbon stock. Moreover, the soil physicochemical properties, absolute elevation and vegetation structural variables could explain 76.83 % of the carbon stock variance. This study demonstrated that artificial planting had no superiority in vegetation restoration and carbon sequestration over natural regeneration. Under appropriate environmental conditions, natural regeneration is an economically effective way to restore mangroves from abandoned ponds to realize the co-benefits of biodiversity and carbon storage restoration.

The mangrove ecosystem is characterized by high carbon sequestration rates and plays a crucial role for the exchange of carbon between land and ocean. Understanding the carbon dynamics of mangroves under climate change and human disturbances is therefore essential for quantifying their contributions to global carbon cycle. However, most land surface models do not have a specific module for mangroves, leading to potential biases in simulating the uptake of atmospheric carbon by terrestrial ecosystems and the carbon budget of tropical countries. In this study, we introduced a new plant functional type for mangroves into the land surface model ORCHIDEE-MICT-PEAT-LEAK. We added the mangrove-specific carbon pools of aboveground roots and a new carbon allocation scheme. We also added the effects of salinity and tidal inundation on mangrove productivity. We then calibrated and optimized the parameters related to photosynthesis, autotrophic respiration, carbon allocation and mortality using data collected from literature reviews. Compared to the original model version, the modified version shows greatly improved performance when evaluated against the observational data sets from field measurements. Specifically, ORCHIDEE-MAN can generally reproduce the spatial distribution of aboveground biomass density from satellite-based map, as well as the seasonal cycles of gross primary productivity observed at three eddy covariance flux towers for mangroves. This model version with the mangrove module provides a useful tool for understanding the carbon cycle processes and estimating carbon budgets in mangrove ecosystems. The processes and parameters described here may support the development of mangrove module in other land surface models.

Phosphorus affects microbial metabolic activity, nitrogen and carbon cycling in mangrove sediment, but its influence on carbon stability and greenhouse gases emission remains unclear. This study compared greenhouse gases (CO2, N2O, and CH4) emissions from mangrove sediment receiving wastewater containing various phosphorus concentrations, and evaluated its long term effect on sediment carbon flux when phosphorus pollution is eliminated. Significant increases in greenhouse gases flux and decrease of total organic carbon and readily oxidizable organic carbon in the sediment were observed after phosphorus discharge. Specifically, the N2O flux was reduced significantly at high phosphorus levels while the CO2 flux and the microbial biomass organic carbon was increased. The copy numbers of ammonia oxidation (AOA-amoA, AOB-amoA) gene, denitrification (narG, nirK) gene and methanogenesis (mcrA) gene increased with the increasing phosphorus concentration. During the wastewater discharge period for 70 days, the global warming potential of sediment flux at high phosphorus discharge condition was more than 4 times that of the control group, and the loss of total organic carbon and readily oxidizable organic carbon was 4.66 % and 7.1 %, respectively. During the remediation period (71-101 days), the greenhouse gases flux decreased rapidly, ends up with a similar level of the control group. Our results indicate that using mangrove wetland for pollution minimization in the coastal aquaculture industry could increase greenhouse gases emisison significantly, it is therefore essential to reduce phosphorus discharges from various anthropogenic activities, and local authorities must set up more stringent discharge standards in the future.

Mangroves are crucial plant communities in the coastal intertidal zones, providing various ecological services and supporting biodiversity. To identify the primary factors affecting the poleward shift of subtropical mangroves and the northernmost extent of subtropical mangroves under various carbon emission scenarios, the MaxEnt model was utilized with presence data of subtropical mangrove and environmental data under different carbon emission scenarios. Kandelia obovata , Avicennia marina , and Aegiceras corniculatum in China were included for modeling, utilizing 22 bioclimate variables to predict mangrove distributions, and kernel density analysis was further employed to identify conservation and afforestation hotspots for mangroves . The results indicated that the number of days below 10 degrees C per year is the major climatic factor shaping the distribution of subtropical native mangroves, while precipitation exerts a relatively lower contribution. Under the low and medium emissions scenarios (SSP126 and SSP245), the northern boundary of the distribution of K. obovata remains stable, while A. marina and A. corniculatum are projected to shift northward to Fuzhou, Fujian Province (26.1 degrees N) from 24.9 degrees N by 2100. Under higher emissions scenarios (SSP370 and SSP585), the suitable habitat of K. obovata is expected to shift north to the mouth of the Yangtze River (31.9 degrees N) from 26.8 degrees N, and A. marina and A. corniculatum are projected to move northward to the coastal area of Ningde, Fujian Province (26.8 degrees N) by 2100. In conclusion, with the increase in carbon emissions, there is a significant northward migration observed in subtropical native mangroves ( p < 0.01), and suitable areas for restoration in China have been identified, with the northernmost region located in Zhejiang Province. The spatial and temporal patterns of range change predicted in this study provide valuable information for conservation and afforestation strategies for these ecologically important species.

Mangrove wetlands, as one of the natural ecosystems with the most ecological services, have garnered widespread attention about their microbial driven biogeochemical cycling. Urbanization have led to different spatial patterns of environmental conditions and microbial communities in mangroves. However, viruses, as the pivotal drivers of biogeochemical cycling in mangroves, remain inadequately explored in terms of how their ecological potential and complex interactions with host respond to functional zonings. To address this knowledge gap, we conducted a comprehensive investigation on the structural and functional properties of temperate and lytic viruses in mangrove wetlands from different functional zonings by jointly using high-throughput sequencing, prokaryotic and viral metagenomics. Multiple environmental factors were found to significantly influence the taxonomic and functional composition, as well as lysogen-lysis decision-making of mangrove viruses. Furthermore, enriched auxiliary metabolic genes (AMGs) involved in methane, nitrogen and sulfur metabolism, and heavy metal resistance were unveiled in mangrove viruses, whose community composition was closely related to lifestyle and host. The virus-host pairs with different lifestyles were also discovered to react to environmental changes in different ways, which provided an empirical evidence for how virus and bacteria dynamics were specific to viral lifestyles in nature. This study expands our comprehension of the intricate interactions among virus, prokaryotic host and the environment in mangrove wetlands from multiple perspectives, including viral lifestyles, virus-host interactions, and habitat dependence. Importantly, it provides a new ecological perspective on how mangrove viruses are adapted to the stress posed by urbanization.

Organophosphate flame retardants (OPFRs) have become a ubiquitous environmental contaminant due to their extensive usage in various applications such as electronics, electrical equipment, building materials, automobiles, textiles, and cables. It have been detected in water bodies and sediment worldwide, indicating their global presence and potential environmental impact. Mangrove forests in the marine-terrestrial ecotone serve as the primary repository for terrestrial pollutants. However, the distribution of OPFRs and the underlying factors driving their presence are not yet well understood. In this study, three urban mangroves in Shenzhen were selected based on urban functional zoning: Shajing (SJ) mangrove, located in an industrial district; Futian (FT) mangrove and Baguang (BG) mangrove, located in the central business district and ecological preserve, respectively. This study revealed that the total concentrations of the seven OPFRs ranged from 30.0 to 418.3 ng L-1 in surface seawater and from 53.1 to 151.6 ng g-1 dry weight in sediment. Triethyl phosphate-TEP and tributyl phosphate-TnBP were detected in surface seawater from SJ and in all sediment samples. The dominant OPFRs varied between surface seawater (tris(1-chloro-2-propyl) phosphate-TCPP and triphenylphosphine oxideTPPO) and sediments (TCPP and triphenyl phosphate-TPhP). Further analysis showed transfer of OPFRs was relatively weak in SJ but improved in BG and FT, suggesting varying degrees of contamination and accumulation across the different functional zones. Ecological risk assessment indicated that only TPHP exhibited moderate risk in surface seawater and sediment. This study provides a comprehensive understanding of the spatial distribution (surface seawater and sediment) of OPFRs in urban mangroves characterized by diverse functional zoning. It emphasizes the necessity of reducing OPFRs emissions into the environment and effectively managing OPFRs contamination in urban mangroves.

Fiddler crabs, as coastal ecosystem engineers, play a crucial role in enhancing biodiversity and accelerating the flow of material and energy. Here we show how widespread crab burrows modify the carbon sequestration capacity of different habitats across a large climatic gradient. The process of crab burrowing results in the reallocation of sediment organic carbon and humus. Crab burrows can increase more greenhouse gases emissions compared to the sediment matrix (CO2: by 17–30%; CH4: by 49–141%). Straightforward calculations indicate that these increased emissions could offset 35–134% of sediment carbon burial in these two ecosystems. This research highlights the complex interactions between crab burrows, habitat type, and climate which reveal a potential lower carbon sink function of blue carbon ecosystems than previously expected without considering crab burrows. Bioturbation in wetlands can increase carbon dioxide and methane emissions, partially offsetting their sediment carbon burial capacity, according to a large-scale data set from sediment samples collected along the Chinese coastline and laboratory incubations.

Heavy metal pollution leads to severe soil contamination and raises environmental concerns. Phosphorus-bearing minerals have been identified as effective and environmentally friendly agents for remediating contaminated soil. However, using recycled phosphorus for metal remediation remains limited. In this study, we investigated the potential use of recovered magnesium ammonium phosphate (MAP), potassium magnesium phosphate hydrate (KMP), and hydroxyapatite (HAP) to immobilize lead (Pb) and cadmium (Cd) in contaminated soil. The effectiveness of MAP, KMP, and HAP in modifying the chemical composition of Pb and Cd in the soil and reducing their mobility was demonstrated by the results. When added at dosages of 5:1 and 10:1, the extractable fraction of Pb decreased to 0.43-13.43 % of the control group. Similarly, the extractable fraction of Cd decreased to 49.18-76.44 % of the control group at the same dosages. Moreover, the application of these products resulted in an increase in soil pH value, urease activity, and bacterial Shannon index, thereby enhancing soil properties. The immobilization effects were more significant in Pb-contaminated soil compared to Cd-contaminated soil, and MAP and KMP exhibited a stronger passivation effect on Pb in comparison to HAP. For soil fertility improvement, a molar ratio of 10:1 was recommended, whereas a ratio of 5:1 effectively passivated Pb in contaminated soil. These findings provide evidence for the feasibility of utilizing MAP, KMP, and HAP in the immobilization of heavy metals in polluted soil.

The effective management of the mangrove forests is especially important, which requires precise and timely monitor of the stand structure of mangrove forest. Meanwhile, stand structure is a key feature that indicates the health status of mangrove forests. Therefore, we developed an effective method to interpret the stand structure of urban mangrove forests using visible light remote sensing images from Unmanned Aerial Vehicles (UAV). The results showed that: (1) Pixel-based deep natural networks under the constraint of physical parameters (tree height) have been explored for mangrove species classification, and the overall classification accuracy of the whole study site was 88.29%. (2) Based on watershed algorithm and the relationship between tree height and tree crown, single tree crown segmentation in non-closed mangrove area was realized, with recognition accuracy of 78.57%~ 88.89%. (3) Based on the obtained stand structure parameters, including the size ratio, angle scale, mixing degree, etc., a stand structure health evaluation model was established to determine the health level of a mangrove forest. Stand structure evaluation score (A) in Futian Mangrove National Nature Reserve was derived that the core area (A: 0.9315~0.9391) was better than the non-core area (A: 0.8622~0.9127). The proposed method will reduce the time and effort required for future mangrove plant observations and contribute to future assessments of the health of mangrove forests.