Dynamics of Methane in Mangrove Forest: Will It Worsen with Decreasing Mangrove Forests?

Mangrove forests sequester a significant amount of organic matter in their sediment and are recognized as an important carbon storage source (i.e., blue carbon, including in seagrass ecosystems and other coastal wetlands). The methane-producing archaea in anaerobic sediments releases methane, a greenhouse gas species. The contribution to total greenhouse gas emissions from mangrove ecosystems remains controversial. However, the intensity CH4 emissions from anaerobic mangrove sediment is known to be sensitive to environmental changes, and the sediment is exposed to oxygen by methanotrophic (CH4-oxidizing) bacteria as well as to anthropogenic impacts and climate change in mangrove forests. This review discusses the major factors decreasing the effect of mangroves on CH4 emissions from sediment, the significance of ecosystem protection regarding forest biomass and the hydrosphere/soil environment, and how to evaluate emission status geospatially. An innovative “digital-twin” system overcoming the difficulty of field observation is required for suggesting sustainable mitigation in mangrove ecosystems, such as a locally/regionally/globally heterogenous environment with various random factors.


Introduction
The carbon (C) sequestered in the biomass and deep sediment of vegetated coastal ecosystems, including mangroves, seagrass beds, and tidal marshes, has been called "blue carbon" [1,2]. Although vegetated coastal habitats cover a relatively small area (<2%) of the coastal ocean, they have C burial rates that are 40 times higher than tropical rainforests and account for more than half of the C burial in marine sediment [3].
Although the global area of vegetated ecosystems is one to two orders of magnitude smaller than that of terrestrial forests, the contribution of vegetated coastal habitats per unit area to long-term C sequestration is much greater, which is in part because of their efficiency in trapping suspended matter and associated organic C during tidal inundation [1]. Among vegetated ecosystems, mangroves have been well highlighted as among the major sources of organic matter in tropical areas because they occupy a large part of the tropical coastal area [4]. Additionally, organic C production is more rapid in these areas than for other estuarine and marine primary producers [3,5].

Soil Conditions
Low CH 4 production and emission in mangrove sediment compared to in interior wetland soils is mainly due to the high presence of sulphate in mangrove sediment, which allows sulphate-reducing bacteria to outcompete CH 4 -producing archea (i.e., methanogens) [40,[55][56][57]. However, soil salinity and sulphate concentration show a low negative relationship with methane-producing activities, which suggests that both forms of methanogenesis are not completely inhibited by sulphate reducers with increasing sulphate concentrations [55]. A significant increase in CH 4 production activity caused by the dilution of seawater was also reported [39]. Therefore, mangrove sediment CH 4 production activity is highly and non-linearly sensitive to its specific soil pH/electrical conductivity by being affected by different freshwater intrusion intensities. Despite few studies on the impact of freshwater intrusion on CH 4 emission, it is still important to evaluate because rice paddies/agricultural fields are often found adjacent to protected mangrove zones ( Figure 1). Regardless of a significant correlation between salinity/sulphate concentration in sediment and CH 4 emissions [39,55,63] CH 4 production activity can be significantly increased by the dilution of seawater concentration [39]. The magnitude of CH4 flux in mangrove forests and its relative contribution to global warming compared to CO2 flux remains controversial. The global scale practice of the mangrove C budget has shown that CH4 emissions from soil are 2 Tg C year −1 [9]. Considering its global warming potential, the contribution of CH4 emissions is comparable to the above-mentioned rate of C burial (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34) Tg C year −1 ) and C emission by soil respiration (34 Tg C year −1 ) [1,9,10,13,14]. Recent studies have reported a significant amount of CH4 flux from mangrove sediment [42][43][44][45][46][47][48][49] and have claimed that the contribution of CH4 flux to global warming was non-negligible in estuarine mangrove forests, which could account for 18% to 22% of blue C burial rates [13] and 9% to 33% of plant CO2 sequestration [50]. However, observed CH4 flux from mangrove soils is mostly negligible compared to CO2 emissions from sediment but is highly variable [38][39][40][41][51][52][53][54][55][56][57][58], particularly for non-polluted mangrove sediment [21,49,[59][60][61][62]. To confirm this observation, the authors compared incubation experiments with mangrove sediment collected from the Vietnamese Mekong delta and the Indian Sundarbans forest [39]. CH4 production was equivalent to only 0.05 to 0.27% of the CO2 production under aerobic incubation or 0.05 to 0.22% under anaerobic incubation conditions, even when considering the potential difference caused by (n = 30 in each incubation experiments).

Soil Conditions
Low CH4 production and emission in mangrove sediment compared to in interior wetland soils is mainly due to the high presence of sulphate in mangrove sediment, which allows sulphate-reducing bacteria to outcompete CH4-producing archea (i.e., methanogens) [40,[55][56][57]. However, soil salinity and sulphate concentration show a low negative relationship with methane-producing activities, which suggests that both forms of methanogenesis are not completely inhibited by sulphate reducers with increasing sulphate concentrations [55]. A significant increase in CH4 production activity caused by the dilution of seawater was also reported [39]. Therefore, mangrove sediment CH4 production activity is highly and non-linearly sensitive to its specific soil pH/electrical conductivity by being affected by different freshwater intrusion intensities. Despite few studies on the impact of freshwater intrusion on CH4 emission, it is still important to evaluate because rice paddies/agricultural fields are often found adjacent to protected mangrove zones ( Figure 1). Regardless of a significant correlation between salinity/sulphate concentration in sediment and CH4 emissions [39,55,63] CH4 production activity can be significantly increased by the dilution of seawater concentration [39]. Another reason for decreased CH4 production is that compared to herbaceous organic matter, woody organic matter derived from mangrove trees is relatively recalcitrant to methanogens using it as a substrate [58]. Additionally, mangrove ecosystems are Another reason for decreased CH 4 production is that compared to herbaceous organic matter, woody organic matter derived from mangrove trees is relatively recalcitrant to methanogens using it as a substrate [58]. Additionally, mangrove ecosystems are inundated by irregular periodic tides affected by the tidally mediated exchange of porewater between sediment and surface water via the ebb and flow of tides [10,59,60]. The tidally mediated exchange of porewater between sediment and surface water occurs via the ebb and flow of tides (i.e., tidal pumping [64][65][66][67]). Tidal pumping is a potential source of solutes to mangrove water and causes ebullition, but the process has only recently been quantified and directly linked to the export of C and nutrients [67,68]. In addition to the spatiotemporal heterogeneity caused by irregular tidal pumping, CH 4 flux is spatio-temporally heterogeneous and is highly variable because of the heterogenic spatial distribution of aerial mangrove tree roots [69] and burrows created by crabs/goby fish. Such activity enhances hydraulic connectivity and increases the surface area of the sediment-water interface [70] where the exchange of the by-products of subterranean respiration can occur during tidal inundation [71][72][73][74][75][76]. Owing to the difficulty of observing CH 4 flux precisely, CH 4 data for mangrove forests are limited compared to data observing interior wetlands and underestimate the global emission [13,60].

Methanogenic and Methanotrophic Communities
Although CH 4 flux micrometeorological observation data are limited [76], recent studies on the community structures of methanogens and methanotrophs have revealed the biological processes common to interior wetlands and as unique characteristics in coastal wetlands. Previous studies on CH 4 metabolism have indicated that CH 4 emission in natural ecosystems is largely driven by microorganisms, especially methanogens and methanotrophs [34,[77][78][79][80]. Highly diverse methanogenic and methanotrophic communities can promote CH 4 production and oxidation [81,82]. However, different types of methanogens and methanotrophs have preferable growing conditions, which further affect CH 4 emissions in natural ecosystems [77,83,84]. Methanogens include hydrogenotrophic, acetoclastic, and methylotrophic methanogens [84,85]. Methanotrophs exist under both aerobic and anaerobic conditions. Aerobic methanotrophs are phylogenetically divided into two main groups: type I (Gammaproteobacteria, e.g., Methylococcaceae) and type II (Alphaproteobacteria, e.g., Methylocystaceae) [86][87][88][89], nitrate-or nitrite-dependent [90,91] and metal-dependent [92] CH 4 oxidizers, respectively. Type I methanotrophs tend to be dominant in natural environments with sufficient nutrients and substrates (i.e., relatively high O 2 concentration, low CH 4 concentration) [39,58,93], whereas type II methanotrophs tend to be abundant in resource-limited environments with a high affinity for their nutrients and substrates (i.e., relatively low O 2 concentration, high CH 4 concentration) [84,87,94,95]. Of note, methanogens and methanotrophs in coastal wetland soils have unique characteristics that are rarely found in interior wetland soils. Hydrogenotrophic and acetoclastic methanogens are considered dominant in natural freshwater wetland soils. However, methylotrophic methanogens are dominant in hypersaline and sulphate-rich environments including coastal wetlands, and they make different contributions to CH 4 production [82,96,97]. In coastal wetlands, anaerobic methanotrophs include sulphate-dependent methanotrophs [39,98], which might have an important role in controlling low coastal CH 4 fluxes. Furthermore, several reports have described the possibility of active CH 4 production under aerobic conditions in mangrove forests based on laboratory incubation experiments and field observations [39,99].

Mangrove Species
The above-mentioned studies have mainly focused on CH 4 emissions from water and sediment in mangrove forests, and CH 4 emission through stems, leaves, and pneumatophores from mangroves is poorly understood [41]. A positive correlation between the density of pneumatophores and CH4 flux is often found in various mangrove forest sediment [39,49,51,100]. However, some specific mangrove forest tree species such as Sonneratia apetala decrease CH 4 flux [41]. The presence of pneumatophores increased soil redox potential (Eh) and sulphate concentration in sediment in a Tanzanian mangrove forest, which suggests the inhibition effect of peumatophores on CH 4 emission [101]. Pneumatophores of different mangrove species might have different roles in CH 4 flux from sediment [41]. The number of pneumatophores can vary seasonally in some mangrove forests, which could be an important factor in the long-term seasonal dynamics of methanogen activity [49].
The contribution of CH 4 emissions from mangrove tree stems without passing through sediment has been reported. Jeffrey et al. [102] reported that the direct CH 4 emission from living tree stems was 2.3% to 6.7% of the total net ecosystem flux and a significant amount came from dead trees, with emissions contributing to 6.6% to 46.1% of the total net ecosystem flux. These results indicate that the health status of the mangrove tree and the canopy structure of forests also have a substantial effect on CH 4 flux.

Invasive Plant Species
Vegetation invasion is often found to increase regional CH 4 emissions in coastal wetlands because most invasive plant species have higher net primary productivity than native plants and supply a greater amount of fresh organic matter to sediment [41,96]. Furthermore, the vegetation change in coastal wetlands occasionally shifts the major methane-producing process in the sediment from a hydrogenotrophic process to a methylotrophic or acetoclastic process [97]. Soil salinity or sulphate concentration shows a low positive relationship with methane-producing activities, which suggests that both forms of methanogenesis are not completely inhibited by sulphate reducers with increasing sulphate concentrations [39,55,97]. Apart from sulphate supply, fresh organic matter supply was often found to be strongly correlated with methane-procuring activities [39]. Consistently, several reports have described that surface layer sediment has greater methaneproducing activity and contains higher soil organic matter content compared to other sediment [101,103]. However, the substrate supply in sediment is spatio-temporally heterogenous because it is affected by tidal pumping, crab/goby burrowing, and the structural complexity of aerial mangrove tree roots [60]. Recent studies are starting to report a significant enhancement in CH 4 emission caused by invading alien species and mangrove trees planted for reforestation, which are more prone to providing greater amounts of fresh organic matter as substrates for methanogens in sediment than native plants and enhance the amount of emission [41,96,97]. Because the invasion/reforestation status of plant species is sensitive, particularly in lawn zones [104,105], and the accumulation zone of substrate for methane production is relatively dense in hollow zones, microscale land surface structure needs to be carefully considered. Although CH 4 flux from the entire of mangrove ecosystem is highly correlated with micro-meteorological properties such as sensible heat flux [76,106], finer spatial resolution evaluation (e.g., chamber-based method) would also be important for considering the interaction between the pollution/invasion status and such geophysical characteristics in local/regional scales.

Anthropogenic Effects on Decreasing Mangrove Forests with Increasing Methane Emission
Causes of mangrove habitat conversion vary globally and include mechanical destruction (mariculture/aquaculture ponds), forest over-exploitation, land-use changes for urbanization, industrial use, chemical spills, upstream dams, dredging, eutrophication of overlying waters, urban development, and conversion to open water due to accelerated sea-level rise and subsidence [1,18,[23][24][25][26][27][28][29]. As mentioned above, these mangrove forest-decreasing factors with high PAI (26% for forestry use, 14% for fish culture, 11% for freshwater diversion, 5% for land reclamation [18]) not only decrease mangrove forests but also enhance CH 4 emission/production. Particularly, with sea-level rise/increasing temperatures, CO 2 emission caused by anthropogenic global warming is suggested to be the biggest threat to mangroves in future decades [13,19,30]. Consistently, these globalscale factors are also recognized as CH 4 emission-enhancing factors, as they accelerate C cycles in wetland soils [107,108]. Even though the contribution of CH 4 emissions from mangrove ecosystems to the total global warming potential is still not clear, various reports have suggested significant impacts on CH 4 emission from mangrove forests [40][41][42]47]. Mangrove ecosystems cleared for aquaculture or that are polluted by organic waste show significantly more methanodynamic activities (i.e., CH 4 production and oxidation activities) and emissions [47,48,109]. A recent estimate of global CH 4 emissions per area of coastal aquaculture farms is equivalent to 7-430 times higher than emissions from general coastal habitats, including from mangrove forests, salt marshes, or seagrasses [110]. Enhancing CH 4 emissions through the aquacultural land-use adjacent to the mangrove ecosystem and external species invasion also needs continuous monitoring with high spatial resolution.
Fortunately, satellite remote-sensing technologies to observe these factors have been advancing. Most major studies targeting mangrove forests aim to quantify biomass or canopy height based on synthetic aperture RADAR (SAR) observation (e.g., polarimetric interferometry [111] and SAR-tomography [112]). However, various technologies to detect changes in these factors include aquacultural pond detection in mangrove forest ecosystems by optical sensor data classification with object-based image analysis [113], SAR shadowing-effect analysis for deforestation detection [114], and hyperspectral-sensor data analysis for mangrove tree species identification and for salinity and vegetation stress [115,116]. Particularly, technologies that can detect soil inundation even with soil covered by cloud/vegetation have been well developed. They involve analysing the contribution of double-bounce intensity to the total back-scattering intensity of the L/P band SAR data used for regionalizing CH 4 emissions in various wetland studies [117][118][119][120][121][122].
For the next phase, we now need to integrate these remote-sensing technologies with biogeochemistry models that consider the above complex factors controlling CH 4 emissions from mangrove soils. Anthropogenic factors monitored by remote-sensing technology are expected to help support decision-making for efficient mangrove forest protection. High CH 4 emission risk is accompanied by a reduction in mangrove forest area, so sophisticated regional evaluation with high transparency and an inter-disciplinary approach is desirable for further development. To realize this, we need to start creating operational applications by developing new processes based on C cycle models that explicitly simulate CH 4 emissions considering the unique behavior of methanogens/methanotrophs in mangrove soils; these applications can realistically represent anthropogenic impacts by using satellite data as input data (i.e., forcing data) or "observation data" for data assimilation [121][122][123].
Although reports on CH 4 emission simulation studies in coastal catchment areas have been increasing recently, most of the model still does not consider the unique behavior of methanogens in mangrove ecosystems to explicitly solve the above-mentioned distinctive CH 4 production/emissions from the sediment. Among emission studies, most of the reports and validation data for the simulation are derived from tidal marshes, and contributions from mangrove ecosystems are still limited [124,125]. Such an innovative "digitaltwin" system overcoming the difficulty of field observation is required in order to suggest sustainable mitigation in mangrove ecosystems, such as a locally/regionally/globally heterogenous environment with various random factors. Funding: This work was partly supported by the Ministry of Environment, Japan.

Conflicts of Interest:
The authors declare no conflict of interest.