Integrating megafauna into blue carbon strategies: dugongs could enhance seagrass carbon storage

Abstract

Coastal seagrass ecosystems occupy a small fraction of the global ocean yet make disproportionately large contributions to carbon capture and storage. In addition, they are increasingly promoted as blue carbon solutions within nationally determined contributions (NDCs). Despite this, most seagrass carbon budgets implicitly assume bottom-up control and very often neglect the functional role of large herbivores. Dugongs (Dugong dugon) are specialist seagrass grazers who may strongly influence seagrass productivity and sediment carbon storage through a combination of biomass removal, sediment disturbance, and rapid nutrient recycling, but their net effect on ecosystem carbon balance remains unknown. We apply a general animal-driven carbon-nutrient cycling model to estimate dugong effects in one of the world’s most important dugong hotspots, the seagrass beds of Bahrain. We parameterize the sediment-seagrass-dugong model with literature data and validated the model with estimates of net primary production (NPP), net ecosystem carbon balance (NECB) and sediment carbon stocks against literature-based and field measurements comparing scenarios with dugong presence vs. absence. Our results indicate that a realistic dugong aggregation (~700 individuals) can, on average enhance seagrass NPP and NECB by 2.4 times (with a range of uncertainty between 1.1- 4.2 times) and sediment carbon stocks by 2.63 times with a range of uncertainty of 1.1 – 7.6 times) relative to a dugong-absent conditions. This yields substantial additional carbon uptake and storage across the 145 km² focal conservation area which corresponds to an additional 7.9 x10⁷ kg C y^-1 captured and 9.1 x10⁸ kg C stored in sediments. Our findings demonstrate that conserving dugongs and their habitat can significantly increase the climate mitigation value of seagrass ecosystems and that explicitly accounting for animal functional roles is critical to avoid underestimating blue carbon contributions in NDC accounting.

1 Introduction

Despite covering an estimated 160,387–266,562 km² of the world’s coastlines (McKenzie et al., 2020), seagrass ecosystems account for less than 0.2% of the total global marine area (Fourqurean et al., 2012). Yet owing to their comparatively high productivity, and capacity for long-term carbon storage in their sediments, they are increasingly being recognized as having the potential to make outsized contributions to the marine carbon budget (Fourqurean et al., 2012; Duarte, 2017; Mazarrasa et al., 2018; Ralph et al., 2018; Macreadie et al., 2021; Orth and Heck, 2023). This recognition is generating growing interest to enlist the functional role of seagrass ecosystems as a nature-based climate solution, i.e., blue carbon strategy in support of climate change mitigation and adaptation policies (Macreadie et al., 2021; Johannessen, 2022; Unsworth et al., 2022; Orth and Heck, 2023). Indeed, seagrass contributions to climate change mitigation can be large enough that they should be considered in many countries’ accountings of nationally determined contributions (NDCs) (Macreadie et al., 2021). Since countries have begun incorporating blue carbon ecosystems into climate mitigation commitments, accurately quantifying seagrass carbon budgets has become increasingly important.

The discovery that seagrass ecosystems can make substantial contributions to blue carbon strategies stems from growing efforts to characterize and quantify the biogeochemical processes that control carbon capture and storage in seagrass biomass and sediments (Duarte et al., 2010; Holmer, 2019; Johannessen, 2022; Orth and Heck, 2023), as well as through the refinement of modeling and measurement methods to accurately estimate seagrass carbon budgets (Carruthers et al., 2002; Macreadie et al., 2014; Duarte, 2017). However, the accuracy of modeling and measurement rests on conceptualizations of what constitutes a functionally intact seagrass ecosystem and the key controls over its functioning. Estimates of seagrass carbon budgets typically focus on carbon uptake and storage in plant biomass and sediments (Duarte et al., 2010; Macreadie et al., 2014; Duarte, 2017; Ward et al., 2022). This implies that only the bottom trophic compartments of seagrass ecosystems—sediment and plants—are the salient functional components. In this conception, the limiting supplies of sediment nutrients such as nitrogen or phosphorus are considered an important bottom-up control over the rates of ecosystem processes such as primary production and hence carbon capture and storage (Hughes et al., 2004; Holmer, 2019), in addition to other bottom-up biophysical factors such as the availability of light, acidity, temperature and hydrodynamics that control plant production and fate of seagrass carbon (Mazarrasa et al., 2018).

However, fully intact seagrass ecosystems also contain animals, which may exert top-down feedback effects that also control rates of seagrass ecosystem functioning (Carruthers et al., 2002; Maxwell et al., 2017; Scott et al., 2018; Orth and Heck, 2023). Large marine grazers—aka mega-grazers—especially, may have significant effects because they can remove substantial amounts of seagrass biomass and disturb sediments, which may reduce the capacity to capture and store carbon (Johnson et al., 2017; 2019; Wirsing et al., 2022). However, they can also cause the rapid recycling of nutrients that can support new ecosystem production (Hughes et al., 2004; Wirsing et al., 2022; Orth and Heck, 2023; Putillo-Wehry et al., 2025). Thus, failing to account for such animal feedback effects on ecosystem functioning could lead to inaccurate estimates of the ecosystem’s capacity to capture and store carbon (Schmitz et al., 2018).

It has been proposed that because dugongs (Dugon dugon), and related sirenians, have important impacts on seagrass biomass, these species could play a significant role in mediating seagrass ecosystem carbon capture and storage (Scott et al., 2018; Schmitz et al., 2023). This is especially significant given that dugongs almost exclusively feed on seagrasses (Best, 1981; Marsh et al., 2018; Wirsing et al., 2022). However, the degree to which dugongs impact net ecosystem carbon capture and storage remains altogether unknown (Martin et al., 2021; Wirsing et al., 2022). In this study, we address this unknown through quantitative modeling that explores the potential influence of dugongs on whole seagrass ecosystem carbon capture and storage.

Evidence shows that sirenian grazing generally reduces the standing biomass of seagrass (Wirsing et al., 2022). This in turn can reduce the mass of detrital organic matter inputs (and embodied carbon and nutrients) to the sediment (Aragones & Marsh 1999). Such reductions limit the availability of nutrients to plants that are released following microbial decomposition and mineralization of the organic matter, and hence the amount of carbon inevitably stored in sediments (Aragones & Marsh 1999). Yet sirenians could have countervailing fertilizing effects that enhance carbon capture and storage. Generally, this effect can arise via fast nutrient recycling in which readily available nutrients are released as body wastes (urine, feces, reproductive and carcass tissue) that can be quickly taken up by seagrasses, thereby circumventing the slower recycling pathway involving microbial decomposition (Perry and Dennison, 1999; Marsh et al., 2018; Jiménez-Ramos et al., 2025). Our modeling analyses consider the interplay between the countervailing effects of biomass removal and nutrient recycling to address uncertainty about the magnitude of top-down effects of dugongs on carbon cycling.

The analyses reveal that model estimates of carbon capture and storage are sensitive to the values of certain parameters related to dugong impacts on key biogeochemical processes. The measurement of these parameters has not been a priority of research on the trophic ecology of sirenians, nor for many other taxa of seagrass mega-grazers. Thus, the modeling highlights crucial new research needs to better constrain estimates of the functional impact of dugongs, as well as other marine mega-grazers, on seagrass carbon budgets. Nevertheless, the modeling provides first approximation estimates of carbon capture and storage that compare favorably with empirical measurements. The findings suggest that the functional role of marine grazers in seagrass carbon cycling should be accounted for when assessing the contribution of seagrass ecosystems to blue carbon strategies, to avoid deriving inaccurate estimates of seagrass ecosystem carbon capture and storage.

1.1 Background

The focal area for our analysis is the coastal waters of the Arabian Gulf surrounding the Kingdom of Bahrain. This area was chosen because of interest in strengthening the conservation of seagrass ecosystems and resident dugongs in a globally significant dugong conservation hotspot (Preen, 2004; Khamis et al., 2023; Kawiyani et al., 2024), as well as to consider the conservation of the species functional role in the ecosystem as a nature-based climate solution (Schmitz et al., 2023).

The 145 km² focal dugong occupancy area (25 320–27 90 N; 50 200–51 70 E: Khamis et al., 2023) encompasses one of three key seagrass habitats within the wider Arabian Gulf that support the largest known dugong population outside of Australia (Preen, 2004). Hence this area, dominated by the seagrasses Halodule stipulacea, H. uninervis and Halophila ovalis (Preen, 2004; Supplemental Information S3), provides important habitat and preferred forage for dugong conservation. Dugong numbers in Bahrain waters have been temporally stable (Preen, 2004; Kawiyani et al., 2024), with a mean of 1164 (95% CI range between 530 and 1798) individuals (Khamis et al., 2023). However, increased human activity within Bahrain seagrass beds and adjacent coastal shoreline development is introducing a new risk of habitat alteration that could force dugongs to shift and reduce their range use (Al-Abdulrazzak and Pauly, 2016; Khamis et al., 2023; Kawiyani et al., 2024). The implications of changes in dugong occupancy of these seagrass beds on carbon dynamics remain unaccounted for in blue carbon assessments. Thus, the goal of our assessment was to quantify the implications of conserving dugongs for blue carbon strategies by estimating their effects on the seagrass carbon budget, and to explore the consequences of potential reductions in dugong abundance on the sustainability of carbon capture and storage. While we focus on a case study in the waters of Bahrain, the modeling nonetheless illustrates how to apply a general analytical animal-driven carbon cycle model (Rizzuto et al., 2024) more broadly for empirical analyses aimed at understanding the role of large marine grazers in controlling seagrass biogeochemical cycling to inform blue carbon strategies.

2 The model

2.1 Overview: the carbon cycle model

We estimated carbon capture and storage by parameterizing a general animal-driven carbon cycle model (Rizzuto et al., 2024) for the focal dugong-seagrass study system. The parametrized model was used to estimate the effects of dugongs on (a) net primary production, (NPP) of seagrass biomass, (b) net ecosystem storage potential, also termed net ecosystem metabolism or net ecosystem carbon sink strength (which we define as Net Ecosystem Carbon Balance [NECB] sensuChapin et al., 2006), and (c) carbon stocks in sediment of the seagrass ecosystem.

The model ecosystem (Figure 1) explores the flow and storage of carbon among sediment, seagrass, and dugong trophic compartments and exchanges between the ecosystem and the atmosphere. The dynamical systems mathematical model describing flow and storage (Table 1) is stoichiometrically explicit, characterizing the coupled flow of carbon (C) and nitrogen (N) among ecosystem compartments. This formulation thus accounts for the rate limiting influence of essential nutrients on carbon cycling (Rastetter, 2011; Rizzuto et al., 2024a; Zaehle et al., 2014). The focus on C and N stems from their known importance in seagrass biogeochemical cycling (Carruthers et al., 2002; Hughes et al., 2004; Holmer, 2019).

Figure 1

The mathematical model is solved for steady state conditions that obey fundamental mass balance requirements (Loreau, 2010) in which ecosystem C and N inputs equal outputs plus storage within ecosystem compartments, i.e., it produces a balanced ecosystem carbon budget.

The modeled inputs (Figure 1) include atmospheric N deposition to the ecosystem (I) and atmospheric C uptake by seagrass (αΦ_P).

The model accounts for storage via the redistribution of C and N among ecosystem compartments (Figure 1) through four processes: (i) seagrass uptake of sediment nitrogen N (Φ_P); (ii) uptake of seagrass C (αΦ_H) and N (Φ_H) by dugong herbivory; and recycling of plant organic matter to sediments (iii) via plant litter (i.e., seagrass C ((1 – δ)αθ_P) and N (θ_P)); and (iv) via dugong waste (i.e., C ((1 – π)βθ_H) and N (θ_H)) which includes feces, urine, and body carcasses.

Outputs include sediment N leaching out of the ecosystems (K), and respiratory C release to the atmosphere from sediment (γ_S), seagrass (δαθ_P), and dugongs (πβθ_H + W_CH[H_N, H_C]). The term W_CH[H_N, H_C] partitions carbon intake into respiration versus recycling, reflecting herbivore stoichiometric homeostasis i.e., the need to maintain a fixed C:N ratio (β) despite consuming plant material with a typically higher C:N (α) (Rizzuto et al., 2024). Generally, plant biomass has high C and low N whereas herbivore biomass has low C high and high N: see case study for dugongs in Yamamuro et al., 2004).

The complete set of equations (Table 1) represent the intact ecosystem in which dugongs are present. To simulate a dugong-absent system (i.e., seagrass and sediment trophic compartments only), we set herbivore consumption and recycling parameters to zero i.e., a_H and r_H = 0 in Φ_H, θ_H thus eliminating herbivore consumption and waste-recycling pathways. This provides a baseline on which to quantify the added functional contribution of dugongs to carbon flows and storage. The net effect of dugongs on the seagrass ecosystem carbon budget is then quantified as the difference between these two model scenarios.

2.2 Estimating carbon capture and storage

The equilibrium solutions (steady state carbon stocks of trophic compartments) for the model equations for the animal absent and animal present scenarios were obtained from Rizzuto et al. (2024). Each modeling scenario (i.e., animal absent; animal present) has one equilibrium where all state variables (sediment = S_i, plant = P_i, herbivore = H_i; where i identifies the stock as either carbon (C) or nitrogen (N)) are positive, and thus biologically meaningful (see Supplementary Material S1). Supplementary Material S1 also includes equations used to estimate NPP and NECB obtained from Rizzuto et al. (2024). The equilibrium solutions were applied to the focal seagrass ecosystem by parameterizing them with empirical data obtained from the literature (Table 2). However, data exclusively for the Bahrain or broader Arabian Gulf area were rarely available in the literature (data were only available to estimate inorganic N inputs and sediment C leaching rate). We therefore obtained empirical data from the broader literature for tropical seagrass ecosystems, especially from studies in the Mediterranean region and western Australia, and from empirical syntheses across multiple study systems (see Table 2 footnote). Parameter values in Table 2 were entered into a structured Excel spreadsheet designed to automatically populate programmed model equilibrium solutions from S1 to estimate carbon capture and sediment storage in the seagrass ecosystem.

2.2.1 Estimates for the dugong absent scenario

The model parameterized for the dugong absent scenario (Table 2) was used to obtain baseline estimates of carbon capture and storage in seagrass and sediment trophic compartments of the ecosystem. We further conducted an uncertainty analysis by estimating mean and standard deviation for NPP (kg C km^-2 d^-1), NECB (kg C km^-2 d^-1) and sediment C (kg C m^-2) by varying the five key parameters driving plant-sediment carbon dynamics: (i) inorganic sediment C loss rate, (ii) sediment N loss rate, (iii) plant N uptake rate, (iv) plant N recycling rate, and (v) plant C respiration rate. We varied parameter values by up to +25% and -25% of the mean, based on insights from a sensitivity analyses of the general model (Rizutto et al., 2024), which indicates that a ± 25% range will produce robust estimates of uncertainty. The parameters were varied one-at-a time for the uncertainty analysis.

The daily NPP and NECB estimates were converted to annual values by multiplying by 365 days, consistent approaches used in empirical syntheses and carbon budget analyses that convert daily estimates to annual values (e.g., Duarte et al., 2010; Ganguly et al., 2017). This assumes that the seagrasses are continually productive year-round, an assumption supported by empirical evidence showing year-round photosynthetic activity and nutrient turnover in tropical seagrasses (Ganguly et al., 2017; Cardini et al., 2018). The model’s performance for this scenario was then evaluated by comparing estimates of NPP, NECB and sediment carbon stocks against empirical measurements from the literature and field measurements in Bahrain and other tropical seagrass ecosystems to assesses the ecological plausibility of the model estimates.

2.2.2 Estimates for the dugong present scenarios

The modeling of dugong effects on carbon dynamics used parameter values for both seagrass and dugongs (Table 2). This scenario required adjustment of four key parameters previously used to represent the dugong absent seagrass-sediment dynamics to account for the dugong effects on biogeochemical processes. The parameters were: (i) sediment C leaching rate, (ii) sediment N leaching rate, (iii) seagrass N uptake rate, and (iv) plant C respiration rate. Direct measurements of dugong effects on these parameters were used whenever possible. However, available data are limited; therefore, values were adjusted using evidence from studies of other grazers and by identifying parameter values that yielded biologically realistic dugong densities under a balanced ecosystem budget

We represented bioturbation associated with dugong feeding by increasing sediment C leaching rate by 1.42×. Dugong foraging disturbs sediments directly and indirectly by reducing seagrass root structure, which can increase sediment carbon export to the water column (Wirsing et al., 2022). We assumed that dugong effects were equivalent to that observed for other seagrass inhabiting animals (i.e., burrowing shrimp, and grazing urchins and sea turtles) and in seagrass removal experiments which can cause 1.30× - 1.44× or more of sediment carbon to be lost (Thomson et al., 2019; Githaiga et al., 2019; Johnson et al., 2019; Carnell et al., 2020; Dahl et al., 2021). Using the same rationale, we increased sediment N leaching by 1.42×.

While grazing can limit the physiological regulation of seagrass N uptake (Valentine et al., 2004) the magnitude of decline in seagrass N uptake remains unknown. Measures for the few studied terrestrial systems indicate that plant N uptake by grazed plants is 0.1 – 0.33 × that of ungrazed plants (Guo et al., 2017; Zhou et al., 2024). We assumed, given arguments that general grazing principles apply to dugongs (Marsh et al., 2018), that dugongs cause similar effects on seagrasses. We thus initially altered grazed seagrass N uptake rate by 0.1 – 0.33 × that of ungrazed seagrass. However, we found that the parameter value needed to be reduced further to 0.046× that of ungrazed seagrass to obtain solutions that reached biologically realistic dugong densities. Finally, we assumed that plant recycling rate with dugongs present declined due to less seagrass biomass available for recycling after dugong consumption. We reduced recycling rate down to 0.01×the original value based on evidence that this value falls within the range of dugong grazing reduced available seagrass biomass (Preen, 1995; D’Souza et al., 2015; Masini et al., 2001).

We estimated the effects of dugongs on seagrass carbon dynamics in two ways. (1) We solved for multiple balanced carbon budgets by adjusting each key parameter singly until reaching a value that closely approximated the maximum dugong aggregate herd size observed in Bahrain waters (i.e., 700 animals: Khamis et al., 2023); (2) We explored potential effects across a range of dugong densities. We solved for balanced carbon budgets by adjusting each key parameter singly or in combination to produce a range of steady state herd sizes that bracketed the commonly observed maximum aggregate herd size (700 individuals) and the range in population size in Bahrain waters (i.e., 530–1798 dugongs: Khamis et al., 2023).

We conducted analysis (1) by estimating net carbon capture (NPP and NECB) and sediment carbon stock with the proviso that the estimated steady state herd size of dugongs needed to closely approximate the aggregate herd size of 700. Steady state aggregation size was estimated in several steps. We first converted the carbon stock estimates of the dugong trophic compartment into whole-animal aggregate density (individuals km^-2) by dividing the predicted dugong carbon stock (kg C km^-2) by the product of average adult dugong body mass (500 kg: Lanyon et al., 2025) and % carbon in dugong body mass. We assumed that dugong body mass contains 47% C, which is the midpoint between measures of percentage C in cattle muscle tissue biomass (51%: Franke and Weniger, 1958) and in the tail fluke biomass of a dugong (43%: Yamamuro et al., 2004). This conversion aligns herbivore carbon stocks with realistic population-level biomass. Dugong density was then multiplied by the study area size (145 km²: Khamis et al., 2023) to estimate aggregate herd size. In estimating NECB, we also accounted for enteric methane (CH₄) emissions from dugongs. As no empirical measurement of dugong CH₄ emission measurements exist yet, we used an allometric relationship (Smith et al., 2015) to estimate annual CH₄ emission per average 500 kg dugong individual and scaled that up to the release by the aggregate herd.

We generated multiple daily estimates of carbon capture and storage through an uncertainty analysis in which the seven parameters were varied individually and in combination: sediment C leaching rate; sediment C leaching rate; plant N uptake rate; dugong N uptake rate; plant C respiration rate; dugong C respiration rate; plant recycling rate and herbivore recycling rate. Parameters were adjusted until modeled dugong abundance approximated an aggregate herd size of ~700 (see Supplementary Material S2). Once a solution was obtained, the parameters were reset to their original values before exploring new combinations.

To further explore uncertainty, we conducted analyses in which some individual parameter values were reverted to those used in the dugong absent scenario or set to the midpoint between the dugong absent and present scenarios. The remaining parameters were then adjusted until modeled herd size again approximated 700 individuals. This approach allowed exploration of parameter space across a range of values. The daily estimates were converted to annual values by multiplying by 365 days, assuming year-round dugong presence in seagrass habitats. This assumption is supported by empirical evidence for the focal area (Khamis et al., 2023).

We conducted analysis (2) by examining how model estimates of NPP and NECB in a balanced carbon budget varied across a range of steady state dugong population sizes. Key model parameters were again varied individually; however, each parameter was systematically adjusted across the full specified range to generate multiple solutions spanning a range of steady state aggregate herd sizes from low (~10’s of individuals) to high (up to 2470 individuals).

3 Model estimates and validation

3.1 Estimates for dugong absent scenario

The model estimates for sediment carbon stocks, NPP and NECB in the absence of dugongs (Table 3) fall within the range of values reported in the literature. This comparison provides an initial test of ecological plausibility for the baseline scenario. The uncertainty (standard deviation) of the model estimates of sediment C stock, NPP and NECB (Table 3) were calculated from the replicate estimates obtained by individually modifying key parameter estimates by ± 25% of each parameter value (Supplementary Material S2A). Sediment carbon stocks were most sensitive (estimate change > 25%) to variation in plant N uptake rate, sediment N leaching rate and plant recycling rate (electronic Supplementary Material S2). NPP and NECB were most sensitive to soil N leaching rate, and plant recycling rate (Supplementary Material S2A).

3.2 Estimates for dugong present scenarios

The modeling indicates that dugong presence could increase average NPP and NECB by 2.4× relative to the dugong-absent scenario (Table 3). The estimated uncertainty range (1.1×- 4.2×) was derived from the range of parameter permutations explored (see S2). The lack of difference between NPP and NECB arises because at ~700 individuals, dugong herd size is not sufficiently abundant to cause CO₂ and methane releases to become a liability on the net carbon balance. The magnitude of dugong effects on seagrass NPP and NECB have not been measured. Hence those model estimates cannot be validated empirically with dugongs. However, the estimated 1.1×- 4.2× range of enhancement of NPP encompasses empirical measures of productivity enhancement of 1.7 – 1.8 × under moderate to high grazing by sea turtles (Christiansen et al., 2012; Gulick et al., 2020).

The model indicates that the presence of dugongs in the seagrass ecosystem could lead to an average sediment carbon storage of 7.1 kg C m^-2 and range between 3.6 – 11.4 kg C m^-2 (Table 3). Thus, the model indicates that on average there could be 2.63× more sediment C in the presence of dugongs than in their absence with an uncertainty range of 1.1×- 7.6×. Field measures of sediment carbon stocks have been obtained for the waters in Bahrain. Data from two different surveys conducted within two different areas estimated that the top 0–15 cm of sediments hold on average 3.5 kg C m^-2 but ranged from 0.5 – 8.5 kg C m^-2 (Alkhuzaei and Brolly, 2025). The average values, when extrapolated to 1 m depth (~23.3 kg C m^-2), is 1.2 × higher than the estimated global average of 19.4 kg C m^-2 (Fourqurean et al., 2012).

There is, however, strong spatial variability with higher carbon stocks in southern waters near Hawar Islands compared to more northerly sites on the east and west sites of Bahrain (Alkhuzaei and Brolly, 2025). Overlaying dugong occurrence data (Khamis et al., 2023) may explain some of the variation given that most historical dugong occurrences are in the southern waters and that sediment carbon stocks can be strongly influenced by the long-term presence and intensity of megaherbivore grazing (Elschot et al., 2015). To explore this possibility further, we partitioned the sediment carbon stock data from Alkhuzaei and Brolly (2025) by classifying the northern locations with infrequent dugong sightings on the east and west side of Bahrain as dugong “absent” sites and the southern locations as dugong “present” sites. The resulting ranges for carbon stock measurements include model estimates for both dugong absent and dugong present scenarios (Table 3). The field measurements (Alkhuzaei and Brolly, 2025) reveal, conservatively, that dugongs could be enhancing sediment C storage by between 1.7 × and 4 ×. Because the previous sediment C sampling was conducted without regard to the spatial occurrences or persistence of dugongs at sampling sites, we conducted an additional sediment C survey targeted to a known dugong occurrence location near Hawar Islands (Khamis et al., 2025). Our survey (see Supplementary Material S3) produced estimates for sediment C stocks in the top 0–10 cm of 14 ± 2.15 (SD) kg C m^-2. This measurement is 25% higher than the upper end of the range of model estimated sediment C in the presence of dugongs (electronic Supplementary Material S2).

The model estimates of NPP and NECB across different dugong aggregation sizes involved systematically changing the original values of the 5 key parameters (Table 2) to produce a balanced budget across a range from 10’s to 1800 individuals (see Supplementary Material S2C). We used those values to examine variation in the dugong abundance-NPP relationship in terms of additional NPP arising from dugong presence. This was estimated by subtracting the mean estimated values of NPP for the dugong absent (baseline) scenario (Table 3) from each estimate of dugong impact on NPP across the different dugong densities (Supplementary Material S2C). The same analysis was completed for NECB. The analysis revealed that NPP varied positively with dugong herd size (Figure 2). However, the data segregated into three distinct domains of high, medium and low NPP (Figure 2). Most parameters, and parameter changes, resulted in the highest estimated carbon capture across dugong density. However, changes in plant N recycling rate led to high and intermediate levels of NPP, and changes in herbivore recycling rate and plant respiration rate reduced NPP from the high to the low domain. Across a gradient from the highest to lowest modeled aggregate herd sizes NPP declined on average by 25% (range 7% - 53%) depending on which domain of NPP vs dugong herd size relationship is being considered. An identical trend was observed for NECB. However, NECB was 0.7% lower than NPP at the highest modeled population sizes. This implies that within the modeled range, dugong herd sizes were never large enough to cause their methane and CO₂ releases to become a serious liability to the carbon budget. The modeling suggests, however, that methane and CO₂ releases could start to become a liability on the carbon budgets of Bahrain seagrass ecosystems at herd or population sizes greater than ~2000 individuals. This underscores that the predominant impacts of dugongs on the modeled seagrass ecosystem carbon budget (Table 3) are via the recycling feedbacks.

Figure 2

The difference in the magnitudes of modeled NPP and NECB across the gradient from high to low dugong herd sizes required only small changes in the magnitudes of parameter values. Hence, the model estimates were sensitive (i.e., % change in estimate > % change in parameter value) to most of the key parameters (sediment C leaching rate, plant N uptake rate, herbivore N uptake rate, plant N recycling rate, herbivore recycling rate and plant respiration rate: Supplementary Material S2C). Increasing or decreasing parameter values beyond those that produced the data for Figure 2 led either to unrealistically high dugong herd sizes or negative herd sizes (i.e., herds could not be sustained by the higher or lower rates; Supplementary Material S2C). Hence the model estimates of ecosystem C budgets for realistic dugong abundances in the study region are produced by narrow range of parameter values. This suggests that a balanced budget when dugongs are present might be easily disrupted by even modest environmental changes that alter the magnitude of dugong impacts on biogeochemical parameters.

4 Discussion

The conservation of seagrass ecosystems is increasingly seen as being vital to ensure the sustainability of ecosystem services, especially regulating services such as carbon capture and storage, and provisioning services in support of providing habitat for animal biodiversity and food webs (Henjowicz et al., 2015; Arkema et al., 2023). However; the provisioning of seagrass habitat for animals and food webs tends to be treated as a co-benefit of conserving ecosystems for carbon capture and storage (Henjowicz et al., 2015; Arkema et al., 2023), in line with the current custom in broader discussions about animal conservation and nature-based climate mitigation (e.g., Soto-Navarro et al., 2020; Smith et al., 2022; Lamba et al., 2023). The consideration of animal conservation as a co-benefit, however, underappreciates the integral functional role of animals in marine ecosystems where their food web interactions may exert feedback control over carbon capture and storage (Martin et al., 2021; Maxwell et al., 2017; Schmitz and Leroux, 2020). However, the science supporting the inclusion of large marine vertebrates such as dugongs as part of feasible blue carbon solutions is still in its infancy; consequently, there is a general lack of understanding of the magnitude of their impacts (Meynecke et al., 2023; Jiménez-Ramos et al., 2025).

Our modeling aimed to enhance quantitative understanding of dugong effects by exploring how the interplay among elemental (e.g., C and N) uptake via foraging, elemental capture and storage within vegetation and animal biomass, elemental release via respiration and body wastes (urine, feces and carcasses), and disturbance that cause elemental leaching from sediments (Figure 1) controls seagrass ecosystem carbon capture in vegetation and storage in sediment. Dugong grazing has the potential to significantly reduce the capacity of seagrass ecosystems to capture and store carbon by reducing aboveground seagrass biomass by up to 96%, belowground biomass by up to 73%, and overall biomass by up to 86%, and coverage by as much as 94% (Wirsing et al., 2022). Yet seagrasses have been found to be resilient to heavy dugong grazing even over relatively short time periods (3–8 months; Wirsing et al., 2022). Our modelling creates this high resilience through dugong release of nutrients that support enhanced seagrass primary production and net ecosystem carbon balance. This conforms with empirical observations of dugong nutrient recycling that increase microbial activity in the sediment, therefore contributing to increased nitrogen fixation and plant production (Aragones et al., 2006). However, other mechanisms not considered in our modelling, could also explain the resilience. For instance, a high level of biomass loss requires seagrass meadows to completely regrow with NPP derived largely from young shoots, rather than regrow from older shoots that have higher content of non-photosynthetic tissues, higher community respiration rates, and possible self-shading (Egea et al., 2023). Thus, heavy grazing by mega-grazers such as dugongs may revert older seagrass beds to younger-age seagrass ecosystem states thereby stimulating increases in carbon capture and storage (Egea et al., 2023).

Our modeling analysis reveals that a herd size of ~700 individuals may be contributing to blue carbon strategies by enhancing sediment C storage by 2.63 times above that stored in their absence, with a range of uncertainty of 1.1 – 7.6 times higher. The modeling illustrates that this increased sediment storage is a consequence of dugongs enhancing NPP and NECB by an estimated average 2.4 times relative to their absence, with a range of uncertainty between 1.1 – 4.2 times higher. These estimated increases in carbon capture and storage due to dugong presence fall within the range of empirical estimates for animal driven effects in 11other terrestrial and coastal marine ecosystems, including sharks in coral reefs and sea otters in kelp forests (Schmitz and Leroux, 2026).

At a herd size of ~700, it is estimated that dugong impacts on the ecosystem could lead to an additional 5.5 x10⁵ kg C km^-2 y^-1 being captured, with an uncertainty range between 1.9 x10⁵ – 9.5 x10⁵ kg C km^-2 y^-1 (Figure 2). Across the entire 145 km² focal seagrass conservation area in Bahrain, this translates into a potential additional uptake of 7.9 x10⁷ kg C y^-1 (i.e., 79, 700 metric tons C y^-1), with an uncertainty range of 2.7 x10⁷ - 1.4 x10⁸ kg C y^-1 (or 27, 000 -140, 000 metric tons C y^-1). Moreover, sediment carbon storage is estimated to increase by 4.4 kg C m^-2 in the presence of dugongs (with a range of uncertainty of 0.1 – 9.4 kg C m^-2). This could mean that dugong impacts on the focal seagrass ecosystem could eventually increase sediment carbon stocks by 6.38 x10⁸ kg (i.e., 638, 000 metric tons, with an uncertainty range of 14, 500 – 1, 363, 000 metric tons). These values could increase by up to 4 times if dugongs occupied the entire spatial extent of seagrass ecosystem around Bahrain (591 km²: Khamis et al., 2023).

These findings suggest a need to reconsider the prevailing view that conserving seagrass beds for carbon storage merely provides a conservation co-benefit of protecting habitat for animals (Henjowicz et al., 2015; Arkema et al., 2023), to one that sees animals such as dugongs undergirding conservation by engineering seagrass ecosystem structure and productivity via their integral functional role in the ecosystem (Scott et al., 2018; Schmitz et al., 2023). This perspective may help explain why dugong spatial occurrences are often regarded as indicators of productive seagrass habitats beds (Khamis et al., 2022). Comparison of the model outputs with available empirical data provides some preliminary support for the ecological plausibility of the modelled carbon dynamics in the seagrass ecosystem. The results, therefore, highlight the importance of considering dugong presence when estimating seagrass carbon budgets. However, further empirical research is needed to explicitly test model predictions by quantifying seagrass carbon budgets in the presence and absence of dugongs.

Quantifying dugong effects will avoid producing incomplete or biased country-level accounts of seagrass ecosystems to NDC’s. Indeed, when placed in the context of Bahrain’s national emissions reporting (https://unfccc.int/documents/655129), this estimated magnitude of dugong-associated carbon capture and storage, when expressed in CO₂e, is comparable in scale to the individual annual emissions of several major sectors (i.e., components of fuel combustion, industry, agriculture and wastewater treatment in 2022). It is also equivalent to roughly 10-30% of 2022 emissions from large sectors such as transport, oil and gas production and aviation (https://unfccc.int/documents/655129). The modelling thus offers an approach to assist conservation decision-making aimed at jointly considering the conservation of dugongs and their functional roles in seagrass ecosystems and improving nature’s contribution to carbon capture and storage as reflected in NDC’s.

The modeling further highlights the importance of defining target dugong population sizes in efforts to conserve or restore their populations, given that the amount of additional carbon captured by the seagrass ecosystem is estimated to vary with dugong herd size (Figure 2). Across the entire range of realistic dugong numbers for the Bahrain waters, from lowest (~ 10’s individuals) to highest (1800 individuals), the modeling suggests that the exact amount of carbon captured could increase by an average of 25% with a range of uncertainty between 7% - 53%. These percentage changes are estimated to be identical for both NPP and NECB, with NECB accounting for storage of C in animal biomass (secondary production) and release of C via CO₂ respiration and enteric methane emissions. Differences between NPP and NECB only begin to arise at much higher dugong aggregation (population) sizes. Thus, the modeling suggests that despite their large sizes (500 kg), dugongs at abundances observed in waters surrounding the Kingdom of Bahrain are unlikely to have an appreciable additive effect on greenhouse gas emissions from their bodies.

The trophic structure and mass balance principles used to model the ecosystem dynamics in this study are similar to those used in previous modeling to describe functional relationships among seagrass animals, including sirenians (manatee [Trichecus manatus]), and plants (Freire et al., 2008; Castelblanco-Martínez et al., 2012; Lee et al., 2021). However, unlike the modeling presented here which explicitly predicts productivity and biomass as emergent properties of animal driven biogeochemical cycling, the previous modeling starts with data on species biomass, species production, and biomass to production ratios as input variables. The general aim of previous modeling is to use these input data to quantify the amount of biomass and energy flowing from different plant trophic groups to different higher trophic compartments (i.e., assumes the system is bottom-up controlled). There was the further aim to estimate the amount of primary productivity needed to support an intact functional ecosystem in support fisheries exploitation (Freire et al., 2008). That modeling accounts for the proportion of total ecosystem productivity flowing through manatees and thus isolates their functional role as a consumer of biomass (e.g., Freire et al., 2008; Castelblanco-Martínez et al., 2012). However, it does not account for top-down recycling feedback and thus does not account for the role that sirenians may play in controlling the level of ecosystem production and associated carbon capture via feedback mechanisms. Consistent with the previous modeling, however, our study shows that under mass-balance an intact seagrass ecosystem that includes animals requires a higher level of productivity to sustain it. But that higher level of seagrass productivity in our modeling is engineered by the recycling feedback by the very animals that consume it (i.e., a sustained interplay between bottom-up and top-down control). This highlights that conventional approaches to study sirenian effects on seagrass ecosystems, focused predominantly on seagrass foraging, will provide an incomplete understanding of dugong effects.

Conventional studies of dugong ecology align with previous modeling (Freire et al., 2008; Castelblanco-Martínez et al., 2012; Lee et al., 2021) by assuming that seagrass ecosystems are bottom-up controlled. These studies thus focus on quantifying biomass removal during feeding (Preen, 1995; Masini et al., 2001; Aragones et al., 2006; Wirsing et al., 2022) motivated by a classic bioenergetics paradigm in animal ecology that considers intake of available biomass and energy in relation to metabolic demands (Best, 1981). There is recognition that dugong foraging may stimulate plant productivity as a compensatory response, and that N recycling is key a key driver of that response (Aragones et al., 2006; Wirsing et al., 2022). However, this recognition is used merely to explain how dugongs may improve the nutritional quality of their plant resources. It does not consider further the whole ecosystem dynamical consequences of recycling.

Our modeling shows that reducing uncertainty in estimated dugong effects on whole ecosystem carbon cycling will require complementing the classic energetics approach in animal foraging ecology with a stoichiometric approach that explicitly quantifies elemental (N and C) intake in relation to metabolic elemental demand while foraging, and elemental release via metabolic body wastes, to obtain more accurate consumption and recycling estimates. As well, it requires expanding the field measurement of C and N fluxes and storage between sediment and plant compartments (Macreadie et al., 2014) to include measurement of fluxes and storage due to dugong presence. However, accounting of dugong effects will require systematic comparison between two ecosystem states in which dugongs are present vs experimentally excluded or naturally absent, as was emulated in the modeling scenario analyses and conducted for other mega-grazers (e.g., Gulick et al., 2020). Accounting for animal driven effects on stocks and flows of carbon also requires better data on how dugong bioturbation affects the amount of sediment C and N that is released into the water column and exported out of the ecosystem relative to conditions where dugongs are absent. Furthermore, there is a need for more detailed measurement on how dugong grazing affects seagrass C respiration and N uptake and C and N allocation in seagrass tissue, which can be facilitated using isotope tracer experiments (e.g., Strickland et al., 2013).

The modeling presented here embodies principles of elemental recycling to sediments by animals as an important feedback mechanism in seagrass ecosystems (Maxwell et al., 2017). However, it treats the sediment as a “black box” and thereby does not explicitly consider the role of microbes in regulating elemental cycling. Our modeling merely approximates the effects of elemental recycling inputs and respiratory and leaching outputs of C by adjusting parameter estimates that account for dugong release and disturbance (Figure 1). Alternative principles suggest that much of dugong body waste that is released to the water column is unlikely to settle in the sediment bed, and thereby unlikely to have a measurable effect on seagrass ecosystem functioning (Aragones et al., 2006). Instead, it may be dugong feedback via disturbance of sediments while foraging, and cascading impacts on microbial functioning that may be the more important control on sediment recycling processes and C storage (Aragones et al., 2006). Further empirical work is needed to resolve how the mechanisms operate and their relative magnitude of effect.

In conclusion, we illustrated how to apply a general model of animal-driven carbon cycling to derive first approximation estimates of the feasibility of using specific conservation projects to enhance ecosystem carbon capture and storage. Our analysis for dugongs shows that they may strongly influence seagrass productivity and sediment carbon storage through a combination of biomass removal, sediment disturbance, and rapid nutrient recycling. Comparison of our model estimates with empirical data provides some support for its predictive reliability, even though there is a margin of uncertainty that needs to be better constrained with more concerted field measures that deliberately quantify dugong foraging and recycling effects on net primary production (NPP), net ecosystem carbon balance (NECB) and sediment carbon stocks. Nevertheless, our findings underscore the need to consider the link between conserving dugongs (and perhaps other mega-grazers) and their habitat with the climate mitigation value of seagrass ecosystems because explicitly accounting for their functional roles may be critical to avoid underestimating blue carbon contributions in NDC accounting.

References

• 1

  Al-AbdulrazzakD.PaulyD. (2016). Reconstructing historical baselines for the Persian/Arabian Gulf Dugong, Dugong dugon (Mammalia: Sirena). Zool. Middle East63, 95–102. doi: 10.1080/09397140.2017.1315853

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  AlexandreA.GeorgiouD.SantosR. (2014). Inorganic nitrogen acquisition by the tropical seagrass Halophila stipulacea. Mar. Ecol.35, 387–394. doi: 10.1111/maec.12128

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  AlkhuzaeiM.BrollyM. (2025). Remote sensing-based estimation of below-ground sediment carbon in seagrass habitats of Bahrain waters. Remote Sens. Earth Syst. Sci.8, 1084–1101. doi: 10.1007/s41976-025-00239-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  AragonesL.MarshH. (1999). Impact of dugong grazing and turtle cropping on tropical seagrass communities. Pac. Conserv. Biol.5, 277–288. doi: 10.1071/PC000277

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  AragonesL. V.LawlerI. R.FoleyW. J.MarshH. (2006). Dugong grazing and turtle cropping: grazing optimization in tropical seagrass systems? Oecologia149, 635–647. doi: 10.1007/s00442-006-0477-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  ArkemaK. K.DelevauxJ. M.SilverJ. M.WinderS. G.SChile-BeersL. M.BoodN.et al. (2023). Evidence-based target setting informs blue carbon strategies for nationally determined contributions. Nat. Ecol. Evol.7, 1045–1059. doi: 10.1038/s41559-023-02081-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  BestR. C. (1981). Foods and feeding habits of wild and captive Sirenia. Mammal Rev.11, 3–29. doi: 10.1111/j.1365-2907.1981.tb00243.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  BurkholzC.Garcias-BonetN.DuarteC. M. (2020). Warming enhances carbon dioxide and methane fluxes from Red Sea seagrass (Halophila stipulacea) sediments. Biogeosci.17, 1717–1730. doi: 10.5194/bg-17-1717-2020

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  CardiniU.van HoytemaN.BednarzV. N.Al-RshaidatM. M. D.WildC. (2018). N2 fixation and primary productivity in a red sea Halophila stipulacea meadow exposed to seasonality. Limnol. Oceaongr.63, 786–796. doi: 10.1002/lno.10669

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  CarnellP. E.IerodiaconoD.AtwoodT. B.MacreadieP. I. (2020). Overgrazing of seagrass by sea urchins diminishes blue carbon stocks. Ecosystems23, 1437–1448. doi: 10.1007/s10021-020-00479-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  CarruthersT. J. B.DennisonW. C.LongstaffB. J.WaycottM.AbalE. G.McKenzieL. J.et al. (2002). Seagrass habitats of northeast Australia: models of key processes and controls. Bull. Mar. Sci.71, 1153–1169. Available online at: https://api.semanticscholar.org/CorpusID:37947519 (Accessed May 21, 2025).

  • Google Scholar
  • View reference in article

• 12

  Castelblanco-MartínezD. N.BarbaE.Schmitter-SotoJ. J.Hernández-AranaH. A.Morales-VelaB. (2012). The trophic role of the endangered Caribbean manatee Trichechus manatus in an estuary with low abundance of seagrass. Estuaries Coasts35, 60–77. doi: 10.1007/s12237-011-9420-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  ChapinF. S.WoodwellG. M.RandersonJ. T.RastetterE. B.LovettG. M.BaldocchiD. D.et al. (2006). Reconciling carbon‐cycle concepts, terminology, and methods. Ecosystems9, 1041–1050. doi: 10.1007/s10021‐005‐0105‐7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  ChristianenM. J.GoversL. L.BoumaT. J.KiswaraW.RoelofsJ. G.LamersL. P.et al. (2012). Marine megaherbivore grazing may increase seagrass tolerance to high nutrient loads. J. Ecol.100, 546–560. doi: 10.1111/j.1365-2745.2011.01900.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  D’SouzaE.PatankarV.ArthurR.MarbàN.AlcoverroT. (2015). Seagrass herbivory levels sustain site-fidelity in a remnant dugong population. PloS One10, e0141224. doi: 10.1371/journal.pone.0141224

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  DahlM.BjörkM.GullströmM. (2021). Effects of seagrass overgrazing on sediment erosion and carbon sink capacity: Current understanding and future priorities. Limnol. Oceanogr. Lett.6, 309–319. doi: 10.1002/lol2.10211

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  DuarteC. M. (1990). Seagrass nutrient content. Mar. Ecol. Prog. Ser.67, 201–207. Available online at: https://www.jstor.org/stable/24816762 (Accessed May 15, 2025).

  • Google Scholar
  • View reference in article

• 18

  DuarteC. M. (2017). Reviews and syntheses: Hidden forests, the role of vegetated coastal habitats in the ocean carbon budget. Biogeosci.14, 301–310. doi: 10.5194/bg-14-301-2017

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  DuarteC. M.MarbàN.GaciaN. E.FourqureanJ. W.BegginsJ.BarrónC.et al. (2010). Seagrass community metabolism: Assessing the carbon sink capacity of seagrass meadows. Global Biogeochem. Cycles24, GB4032. doi: 10.1029/2010GB003793

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  EgeaL. G.Pérez-EstradaC. J.Jiménez-RamosR.HernándezI.López-LópezS.BrunF. G. (2023). Changes in carbon metabolism and dissolved organic carbon fluxes on seagrass patches (Halodule wrightii) with different ages in Southern Gulf of California. Mar. Env.l Res.191, 106136. doi: 10.1016/j.marenvres.2023.106136

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  ElschotK.BakkerJ. P.TemmermanS.van de KoppelJ.BoumaT. J. (2015). Ecosystem engineering by large grazers enhances carbon stocks in a tidal salt marsh. Mar. Ecol. Prog. Ser.537, 9–21. doi: 10.3354/meps11447

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  FourqureanJ.DuarteC.KennedyH.MarbàN.HolmerM.MateoM. A.et al. (2012). Seagrass ecosystems as a globally significant carbon stock. Nat. Geosci.5, 505–509. doi: 10.1038/ngeo1477

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  FrankeE.-R.WenigerJ. H. (1958). Der stickstoff-, kohlenstoff- und energiegehalt des fleisches und der wärmewert des fettes verschiedener nutztierarten. Archiv Fr. Tierernährung.8, 81–94. doi: 10.1080/17450395809424770

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  FreireK. M.ChristensenV.PaulyD. (2008). Description of the east Brazil large marine ecosystem using a trophic model. Sci. Mar.72, 477–491. doi: 10.3989/scimar.2008.72n3477

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  GangulyD.SinghG.RamachandranP.SelvamA. P.BanerjeeK.RamachandranR. (2017). Seagrass metabolism and carbon dynamics in a tropical coastal embayment. AMBIO46, 667–679. doi: 10.1007/s13280-017-0916-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  GithaigaM. N.FrouwsA. M.KairoJ. G.HuxhamM. (2019). Seagrass removal leads to rapid changes in fauna and loss of carbon. Front. Ecol. Evol.7, 62. doi: 10.3389/fevo.2019.00062

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  GotoM.WatanabeA.KaritaS.TokitaN.YamamotoY.WakakiY.et al. (2008). Nutrient and energy consumption of captive mature dugong (Dugong dugon) consuming eelgrass at the Toba Aquarium. Mar. Freshw. Behav. Physiol.41, 169–177. doi: 10.1080/10236240802284445

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  GulickA. G.JohnsonR. A.PollockC. G.Hillis‐StarrZ.BoltenA. B.BjorndalK. A. (2020). Recovery of a large herbivore changes regulation of seagrass productivity in a naturally grazed Caribbean ecosystem. Ecology101, e03180. doi: 10.1002/ecy.3180

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  GuoY.LiuL. P.ZhengL. L.YuF. H.SongM. H.ZhangX. Z. (2017). Long-term grazing affects relationships between nitrogen form uptake and biomass of alpine meadow plants. Plant Ecolog.218, 1035–1046. doi: 10.1007/s11258-017-0746-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  HanY.FengG.SwaneyD. P.DentenerF.KoebleR.OuyangY.et al. (2020). Global and regional estimation of net anthropogenic nitrogen inputs (NANI). Geoderma361, 114066. doi: 10.1016/j.geoderma.2019.114066

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  HejnowiczA. P.KennedyH.RuddM. A.HuxhamM. R. (2015). Harnessing the climate mitigation, conservation and poverty alleviation potential of seagrasses: prospects for developing blue carbon initiatives and payment for ecosystem service programmes. Front. Mar. Sci.2, 32. doi: 10.3389/fmars.2015.00032

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  HelberS. B.WintersG.StuhrM.BelsheE. F.BröhlS.SchmidM.et al. (2021). Nutrient history affects the response and resilience of the tropical seagrass Halophila stipulacea to further enrichment in its native habitat. Front. Plant Sci.12, 678341. doi: 10.3389/fpls.2021.678341

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  HolmerM. (2019). “ Productivity and biogeochemical cycling in seagrass ecosystems,” in Coastal wetlands; and integrated ecosystem approach, second edition. Eds. PerilloG. M. E.WolanskiE.CahoonD. R.HopkinsonC. S. (Amsterdam, NL: Elsevier), 433–477.

  • Google Scholar
  • View reference in article

• 34

  HolmerM.OlsenA. B. (2002). Role of decomposition of mangrove and seagrass detritus in sediment carbon and nitrogen cycling in a tropical mangrove forest. Mar. Ecol. Prog. Ser.230, 87–101. doi: 10.3354/meps230087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  HughesA. R.BandoK. J.RodriguezL. F.WilliamsS. L. (2004). Relative effects of grazers and nutrients on seagrasses: a meta-analysis approach. Mar. Ecol. Prog. Ser.282, 87–99. doi: 10.3354/meps282087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  Jiménez-RamosR.EgeaL. G.D’AgostinoV. C.DegratiM.LoizagaR. (2025). Carbon and nitrogen stocks in sediment at Península Valdés Biosphere Reserve: novel insights into the potential contribution of large marine vertebrates to carbon sequestration. Front. Mar. Sci.12, 1500594. doi: 10.3389/fmars.2025.1500594

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  JohannessenS. C. (2022). How can blue carbon burial in seagrass meadows increase long-term, net sequestration of carbon? A critical review. Environ. Res. Lett.17, 093004. doi: 10.1088/1748-9326/ac8ab4

  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  JohnsonR. A.GulickA. G.BoltenA. B.BjorndalK. A. (2017). Blue carbon stores in tropical seagrass meadows maintained under green turtle grazing. Sci. Rep.7, 13545. doi: 10.1038/s41598-017-13142-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  JohnsonR. A.GulickA. G.BoltenA. B.BjorndalK. A. (2019). Rates of sediment resuspension and erosion following green turtle grazing in a shallow Caribbean Thalassia testudinum meadow. Ecosystems22, 1787–1802. doi: 10.1007/s10021-019-00372-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  KawiyaniR.Ben-HasanA.MohsenK.AlmojilD. (2024). Status, threats, and conservation considerations of selected marine habitats and organisms in the Arabian/Persian Gulf. Mar. Environ. Res.198, 106556. doi: 10.1016/j.marenvres.2024.106556

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  KhamisA.AbdullaA.D’SouzaE.KelkarN.ArthurR.Al KhalifaB. H.et al. (2023). Long-term persistence of large dugong groups in a conservation hotspot around Hawar Island, Kingdom of Bahrain. Aquat. Conserv.33, 592–605. doi: 10.1002/aqc.3936

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 42

  KhamisA.AlcoverroT.D'SouzaE.ArthurR.PagèsJ. F.ShahJ.et al. (2022). Identifying conservation priorities for a widespread dugong population in the Red Sea: Megaherbivore grazing patterns inform management planning. Mar. Environ. Res.181, 105762. 10.1016/j.marenvres.2022.105762.

  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  KhamisA.PagèsJ.ArthurR.D'SouzaE.HameidN.Al Khalifaet al. (2025). Why do large dugong aggregations persist in continuous seagrass meadows despite anthropogenic disturbances around Hawar Island, Bahrain. Aquat. Conserv. Mar. Freshw. Ecosyst.35, e70157. doi: 10.1002/aqc.70157

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 44

  LambaA.TeoH. C.SreekarR.ZengY.CarrascoL. R.KohL. P. (2023). Climate co-benefits of tiger conservation. Nat. Ecol. Evol.7, 1104–1113. doi: 10.1038/s41559-023-02069-x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 45

  LanyonJ. M.NewgrainK.AlliT. S. (2006). Estimation of water turnover rate in captive dugongs (Dugong dugon). Aquat. Mamm.32, 103. doi: 10.1578/AM.32.1.2006.103

  • CrossRef
  • Google Scholar
  • View reference in article

• 46

  LanyonJ. M.SneathH. L.LongT.BlanshardW. H.WorthyG. A. J.BoothD. T. (2025). How much seagrass does a dugong need? Metabolic rate of live wild dugongs, Dugong dugon, determined through indirect calorimetry (oxygen consumption). Mar. Mamm. Sci.41, e13190. doi: 10.1111/mms.13190

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 47

  LeeC.-L.LinW.-J.LiuP.-J.ShaoK.-T.LinH.-J. (2021). Highly productive tropical seagrass beds support dverse consumers and a large organic carbon pool in the sediments. Diversity.13, 544. doi: 10.3390/d13110544

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 48

  LepointG.MilletS.DaubyP.GobertS.BouquegneauJ. M. (2002). Annual nitrogen budget of the seagrass Posidonia oceanica as determined by in situ uptake experiments. Mar. Ecol. Prog. Ser.237, 87–96. doi: 10.3354/meps237087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 49

  LoreauM. (2010). From populations to ecosystems: theoretical foundations for a new ecological synthesis (Princeton, NJ: Princeton University Press).

  • Google Scholar
  • View reference in article

• 50

  MacreadieP. I.BairdM. E.Trevathan-TackettS. M.LarkumA. W. D.RalphP. J. (2014). Quantifying and modelling the carbon sequestration capacity of seagrass meadows–a critical assessment. Mar. pollut. Bull.83, 430–439. doi: 10.1016/j.marpolbul.2013.07.038

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 51

  MacreadieP. I.CostaM. D.AtwoodT. B.FriessD. A.KellewayJ. J.KennedyH.et al. (2021). Blue carbon as a natural climate solution. Nat. Rev. Earth Environ.2, 826–839. doi: 10.1038/s43017-021-00224-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 52

  MarshH.GrechA.McMahonK. (2018). “ Dugongs: seagrass community specialists,” in Seagrasses of Australia: structure, ecology and conservation. Eds. LarkumA. D. W.KendrickG. A.RalphP. J. ( Springer International Publishing, Cham), 743–758. doi: 10.1007/978-3-319-71354-0_19

  • CrossRef
  • Google Scholar
  • View reference in article

• 53

  MartinA. H.PearsonH. C.SabaG. K.OlsenE. M. (2021). Integral functions of marine vertebrates in the ocean carbon cycle and climate change mitigation. One Earth4, 680–693. doi: 10.1016/j.oneear.2021.04.019

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 54

  MasiniR. J.AndersonP. K.McCombA. J. (2001). A Halodule-dominated community in a subtropical embayment: physical environment, productivity, biomass, and impact of dugong grazing. Aquat. Bot.71, 179–197. doi: 10.1016/S0304-3770(01)00181-4

  • CrossRef
  • Google Scholar
  • View reference in article

• 55

  MateoM. A.RomeroJ. (1997). Detritus dynamics in the seagrass Posidonia oceanica: elements for an ecosystem carbon and nutrient budget. Mar. Ecol. Prog. Ser.151, 43–53. doi: 10.3354/meps151043

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 56

  MaxwellP. S.EklöfJ. S.van KatwijkM. M.O'BrienK. R.de la Torre‐CastroM.BoströmC.et al. (2017). The fundamental role of ecological feedback mechanisms for the adaptive management of seagrass ecosystems–a review. Biol. Rev.92, 1521–1538. doi: 10.1111/brv.12294

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 57

  MazarrasaI.Samper-VillarrealJ.SerranoO.LaveryP. S.LovelockC. E.MarbàN.et al. (2018). Habitat characteristics provide insights of carbon storage in seagrass meadows. Mar. pollut. Bull.134, 106–117. doi: 10.1016/j.marpolbul.2018.01.059

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  McKenzieL. J.NordlundL. M.JonesB. L.Cullen-UnsworthL. C.RoelfsemaC.UnsworthR. K. F. (2020). The global distribution of seagrass meadows. Environ. Res. Lett.15, 074041. doi: 10.1088/1748-9326/ab7d06

  • CrossRef
  • Google Scholar

• 59

  MeyneckeJ.-O.SamantaS.de BieJ.SeybothE.Prakash DeyS.FearonG.et al. (2023). Do whales really increase the oceanic removal of atmospheric carbon? Front. Mar. Sci.10, 1117409. doi: 10.3389/fmars.2023.1117409

  • CrossRef
  • Google Scholar

• 60

  OrthR. J.HeckK. L. (2023). The dynamics of seagrass ecosystems: History, past accomplishments, and future prospects. Estuaries Coasts46, 1653–1676. doi: 10.1007/s12237-023-01252-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61

  OrtizR. M. (2001). Osmoregulation in marine mammals. J. Exp. Biol.204, 1831–1844. doi: 10.1242/jeb.204.11.1831

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  PerryC. J.DennisonW. C. (1999). Microbial nutrient cycling in seagrass sediments. AGSO J. Aust. Geol. Geophys.17, 227–231.

  • Google Scholar

• 63

  PreenA. (1995). Impacts of dugong foraging on seagrass habitats: observational and experimental evidence for cultivation grazing. Mar. Ecol. Prog. Ser.124, 201–213. doi: 10.3354/meps124201

  • CrossRef
  • Google Scholar

• 64

  PreenA. (2004). Distribution, abundance and conservation status of dugongs and dolphins in the southern and western Arabian Gulf. Biol. Cons.118, 205–218. doi: 10.1016/j.biocon.2003.08.014

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 65

  Putillo-WehryA.LefcheckJ.FuentesM. (2025). Effects of green turtle herbivory on seagrass ecosystems: an experimental test in the context of a global meta-analysis. Mar. Ecol. Prog. Ser.766, 57–71. doi: 10.3354/meps14902

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66

  RalphP. J.CrosswellJ. R.CannardT.StevenA. D. (2018). “ Estimating seagrass blue carbon and policy implications: The Australian perspective,” in Seagrasses of Australia: structure, ecology and conservation. Eds. LarkumA. D. W.KendrickG. A.RalphP. J. ( Springer International Publishing, Cham), 743–758. doi: 10.1007/978-3-319-71354-0_22

  • CrossRef
  • Google Scholar

• 67

  RastetterE. B. (2011). Modeling coupled biogeochemical cycles. Front. Ecol. Environ.9, 68–73. doi: 10.1890/090223

  • CrossRef
  • Google Scholar

• 68

  RizzutoM.LerouxS. J.SchmitzO. J. (2024). Rewiring the carbon cycle: A theoretical framework for animal‐driven ecosystem carbon sequestration. J. Geophys. Res. Biogeosci.129, e2024JG008026. doi: 10.1029/2024JG008026

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 69

  SchmitzO. J.LerouxS. J. (2020). Food webs and ecosystems: Linking species interactions to the carbon cycle. Ann. Rev. Ecol. Evol. Syst.51, 271–295. doi: 10.1146/annurev-ecolsys-011720-104730

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 70

  SchmitzO. J.LerouxS. J. (2026). “ Trophic rewilding to animate the carbon cycle and mitigate climate change,” in Trophic rewilding: research, policy and practice. Eds. GordonI.ManningA. (Amsterdam, NL: Elsevier).

  • Google Scholar

• 71

  SchmitzO. J.SylvénM.AtwoodT. B.BakkerE. S.BerzaghiF.BrodieJ. F.et al. (2023). Trophic rewilding can expand natural climate solutions. Nat. Clim. Change13, 324–333. doi: 10.1038/s41558-023-01631-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 72

  SchmitzO. J.WilmersC. C.LerouxS. J.DoughtyC. E.AtwoodT. B.GalettiM.et al. (2018). Animals and the zoogeochemistry of the carbon cycle. Science362, eaar3213. doi: 10.1126/science.aar3213

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 73

  ScottA. L.YorkP. H.DuncanC.MacreadieP. I.ConnollyR. M.EllisM. T.et al. (2018). The role of herbivory in structuring tropical seagrass ecosystem service delivery. Front. Plant Sci.9, 127. doi: 10.3389/fpls.2018.00127

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 74

  SmithF. A.LyonsS. K.WagnerP. J.ElliottS. M. (2015). The importance of considering animal body mass in IPCC greenhouse inventories and the underappreciated role of wild herbivores. Glob Change Biol.21, 3880–3888. doi: 10.1111/gcb.12973

  • CrossRef
  • Google Scholar

• 75

  SmithP.ArnethA.BarnesD. K. A.IchiiK.MarquetP. A.PoppA.et al. (2022). How do we best synergize climate mitigation actions to co-benefit biodiversity? Glob. Change Biol.28, 2555–2557. doi: 10.1111/gcb.16056

  • CrossRef
  • Google Scholar

• 76

  Soto-NavarroC.RaviliousC.ArnellA.De LamoX.HarfootM.HillS. L.et al. (2020). Mapping co-benefits for carbon storage and biodiversity to inform conservation policy and action. Phil. Trans. R. Soc B.375, 20190128. doi: 10.1098/rstb.2019.0128

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 77

  StricklandM. S.HawlenaD.ReeseA.BradfordM. A.SchmitzO. J. (2013). Trophic cascade alters ecosystem carbon exchange. Proc. Nat. Acad. Sci. U.S.A.110, 11035–11038. doi: 10.1073/pnas.1305191110

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 78

  ThomsonA. C.Trevathan‐TackettS. M.MaherD. T.RalphP. J.MacreadieP. I. (2019). Bioturbator‐stimulated loss of seagrass sediment carbon stocks. Limnol. Oceanogr.64, 342–356. doi: 10.1002/lno.11044

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 79

  TouchetteB. W.BurkholderJ. M. (2000). Review of nitrogen and phosphorus metabolism in seagrasses. J. Exp. Mar. Biol. Ecol.250, 133–167. doi: 10.1016/S0022-0981(00)00195-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 80

  UnsworthR. K.Cullen-UnsworthL. C.JonesB. L.LilleyR. J. (2022). The planetary role of seagrass conservation. Science.377, 609–613. doi: 10.1126/science.abq6923

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 81

  ValentineJ. F.BlytheE. F.MadhavanS.ShermanT. D. (2004). Effects of simulated herbivory on nitrogen enzyme levels, assimilation and allocation in Thalassia testudinum. Aquat. Bot.79, 235–255. doi: 10.1016/j.aquabot.2004.04.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 82

  WardM.KindingerT. L.HirshH. K.HillT. M.JellisonB. M.LummisS.et al. (2022). Reviews and syntheses: Spatial and temporal patterns in seagrass metabolic fluxes. Biogeosciences19, 689–699. doi: 10.5194/bg-19-689-2022

  • CrossRef
  • Google Scholar

• 83

  WirsingA. J.KiszkaJ. J.AllenA. C.HeithausM. R. (2022). Ecological roles and importance of sea cows (Order: Sirenia): a review and prospectus. Mar. Ecol. Prog. Ser.689, 191–215. doi: 10.3354/meps14031

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 84

  YamamuroM.AketaK.UchidaS. (2004). Carbon and nitrogen stable isotope ratios of the tissues and gut contents of a dugong from the temperate coast of Japan. Mammal Study29, 179–183. doi: 10.3106/mammalstudy.29.179

  • CrossRef
  • Google Scholar

• 85

  ZaehleS.MedlynB. E.De KauweM. G.WalkerA. P.DietzeM. C.HicklerT.et al. (2014). Evaluation of 11 terrestrial carbon–nitrogen cycle models against observations from two temperate Free-air CO2 Enrichment studies. New Phytol.202, 803–822. doi: 10.1111/nph.12697

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 86

  ZhouJ.DingY.TianY.XuX.GuninaA. (2024). Exploring grazing intensity effects: nitrogen uptake in grassland species and soil carbon allocation. Plant Soil504, 833–846. doi: 10.1007/s11104-024-06662-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar