How mangroves quietly move carbon — and why it matters for climate accounting

Our recent article highlights the importance of lateral carbon export from mangroves, averaging 101 mmol m⁻² d⁻¹. Under warming scenarios, this could increase by up to 305% before declining past a 2.5 °C threshold, emphasizing its role in carbon sequestration.

## The team behind the science

This work was a truly international collaboration. Our team brought together expertise in coastal carbon biogeochemistry, mangrove ecology, and environmental modelling, with field campaigns in China and data synthesis spanning the globe. We are grateful to the students who helped with field sampling, and to the many researchers whose local-scale studies made our global synthesis possible.

## The question that started it all

Mangroves are among the most carbon-dense ecosystems on Earth. They pull carbon dioxide out of the atmosphere, store it in their biomass and soils, and bury it in sediments over millennial timescales. This is the foundation of “blue carbon” — a natural climate solution worth an estimated US$190 billion per year.

But there is a piece of the mangrove carbon budget that has remained stubbornly elusive: lateral export. Mangroves don’t just store carbon — they also leak it. Tides, groundwater, and tidal creeks carry dissolved and particulate carbon out of the forest and into the coastal ocean. This lateral transfer is thought to sustain nearshore fisheries, drive coastal nutrient dynamics, and potentially contribute to long-term carbon sequestration. Yet while global estimates exist for carbon exported via submarine groundwater discharge, the surface-water pathway — water flowing through tidal creeks — had only ever been quantified at individual local sites.

We wanted to know: how much carbon do mangroves actually export through their surface waters, what controls it, and how will global warming change it?

## From local creeks to a global picture

Our study spans an enormous range of scales — from 24-hour time series in individual tidal creeks to a global synthesis across 61 mangrove sites in 14 nations, and from diurnal to decadal timescales. We combined a global literature dataset with primary field data from three contrasting mangrove types in Guangdong Province, China: an estuarine freshwater mangrove, an estuarine saline mangrove, and a coastal mangrove.

The first surprise was just how large the surface-water lateral carbon exchange (SWLCE) really is. At the global scale, we found that mangroves export an average of 101 mmol m⁻² d⁻¹ of total carbon, dominated (62.6–93.6%) by dissolved inorganic carbon (DIC). When scaled to the current global mangrove area, this amounts to roughly 64.1 TgC yr⁻¹— about 65% of the total carbon sequestered by global mangrove vegetation, and nearly three times the global mangrove carbon burial rate.

In other words, lateral export may be the largest single component of mangrove carbon sequestration, yet it has been largely overlooked in carbon accounting frameworks.

## What’s driving the export?

To understand what controls this export, we turned to interpretable machine learning. Random forest models, combined with Kernel SHAP analysis, allowed us to quantify both the importance and the direction of influence of each environmental factor. Water temperature, dissolved oxygen, chlorophyll a, and salinity emerged as the key drivers for DIC and DOC export, while season and dissolved oxygen were paramount for POC export.

We also used stable carbon isotopes (δ¹³C-DIC) to trace where the exported DIC comes from. By comparing observed values against conservative mixing lines between freshwater and seawater endmembers, we identified the dominant biogeochemical processes: organic carbon degradation (41% of samples) and carbonate dissolution (33%). In some cases, these processes occurred sequentially — for example, CO₂ outgassing followed by carbonate dissolution — revealing the complex, coupled nature of carbon cycling in mangrove tidal creeks.

## Looking ahead: warming scenarios

Using the random forest models, we projected SWLCE through to 2100 under low (SSP1-2.6), intermediate (SSP2-4.5), and high (SSP5-8.5) emissions scenarios. The results were striking:

- Under intermediate-to-high emissions, total carbon export is projected to increase by up to 305%.

- However, this increase reaches a plateau at a warming threshold of approximately 2.5°C, beyond which export declines.

- For China’s mangroves, total carbon export more than doubles by 2100 under SSP2-4.5 and increases by about 92% under SSP5-8.5.

The temperature threshold is particularly significant. It suggests that while warming may initially stimulate carbon export, beyond a certain point the system shifts — a disruption of carbon sequestration that aligns with other findings on warming-stimulated carbon emissions.

## Why this matters for policy and people

Our findings have direct implications for how we value and manage mangrove ecosystems.

First, lateral carbon export should be incorporated into mangrove carbon accounting. Current frameworks recognize only sediment carbon burial, but our results show that surface-water export is a significant — and previously underquantified — carbon sink. This is particularly relevant for the development of marketable blue carbon credits, where uncertainty about lateral fluxes has been flagged as a major obstacle.

Second, the exported carbon has co-benefits beyond climate mitigation. It supports nearshore fisheries — an estimated 700 billion juvenile fish and invertebrates depend on mangrove-derived carbon — and sustains livelihoods for approximately 4.1 million fishers. This connects directly to Sustainable Development Goals 13 (Climate Action) and 14 (Life Below Water).

Third, our projections suggest that China’s mangrove carbon export is resilient to warming under intermediate and high emissions scenarios, with increased export through 2100. But this resilience is not unlimited — the temperature threshold we identified highlights the importance of mitigation. 

We are also mindful of the gaps that remain. There are still few studies on SWLCE from Central American mangroves and limited data from African mangroves. Future research should aim to fill these geographical gaps, refine estimates of turnover times for different carbon species, and fully value lateral carbon export for its role in enabling multiple ecosystem services.

## What unites us

What unites our team is a shared interest in coastal carbon cycling and a commitment to improving how we account for nature’s contributions to climate mitigation. We welcome networking with anyone who shares this commitment — because collaboration is the only way we can truly make an impact.

*Read the full paper: Ouyang, X., Maher, D.T., Lee, S.Y. & Yang, Z. Surface-water carbon exchange from mangroves and responses to global warming. Nature Communications (2026). https://doi.org/10.1038/s41467-026-77058-2*

*Data and code are available via Figshare: https://doi.org/10.6084/m9.figshare.31476808*