Focus on the impact of climate change on wetland ecosystems and carbon dynamics

The renewed growth in atmospheric methane (CH4) since 2007 after a decade of stabilization has drawn much attention to its causes and future trends. Wetlands are the single largest source of atmospheric CH4. Understanding wetland ecosystems and carbon dynamics is critical to the estimation of global CH4 and carbon budgets. After approximately 7 years of CH4 related research following the renewed growth in atmospheric CH4, Environmental Research Letters launched a special issue of research letters on wetland ecosystems and carbon dynamics in 2014. This special issue highlights recent developments in terrestrial ecosystem models and field measurements of carbon fluxes across different types of wetland ecosystems. The 14 research letters emphasize the importance of wetland ecosystems in the global CO2 and CH4 budget.


Introduction
Wetlands cover approximately 6% of the Earth's land surface and contain a large portion of the world's biodiversity (Junk et al 2013). Wetland ecosystems are an important component in the global carbon cycle and play an important role in terrestrial ecosystem and atmosphere interactions. On the one hand, wetland ecosystems are an ideal natural environment for the carbon sequestration and long-term storage of atmospheric CO 2 (Frolking et al 2011); on the other hand, wetlands are the largest single source of atmospheric methane (Meng et al 2012). Therefore, wetlands serve as both a carbon sink and source from global warming perspectives (Junk et al 2013). It is highly uncertain that whether wetlands act as a positive or negative feedback to the climate system.
Currently, carbon stored in wetlands is close to that stored in the atmosphere (Lenhart 2009). Understanding the impact of climate changes and human activities on carbon dynamics of wetland ecosystems is critical to climate predictability. The 14 letters included in this special issue 'focus on the impact of climate change on wetland ecosystem carbon dynamics' deal with carbon dynamics in different types of wetlands in high latitudes, altitudes (e.g., Tibetan plateau), and tropical regions through observational studies and modeling simulations. Thus, this special issue provides a set of valuable findings and data related to wetlands carbon cycling and its environmental controls.

Northern wetlands (45°N north)
Northern wetlands store over 50% of the global soil organic carbon due to the slow organic carbon decomposition rate as a result of wet surface conditions and cold temperatures (Hugelius et al 2013 temperatures. A few letters from this special issue address the impact of environmental changes on wetland ecosystem carbon balance and methane emissions. Peichl et al (2014) used 12 years of Eddy covariance (EC) measurements to investigate the inter-annual variation in the long-term net ecosystem exchange (NEE) of carbon dioxide (CO 2 ) and its relationship to environmental factors such as water table level and pre-growing season air temperatures in an oligotrophic mire in northern Sweden. It was found that this boreal peatland serves as a persistent longterm sink of atmospheric CO 2 that is not subject to moderate inter-annual climate variations except under extreme low water table level conditions. Another study found that dry conditions have no persistent effects on the CO 2 exchange dynamics in the Andoya blanket bog in northern Norway on seasonal to annual time scales, and that growing season onset and amount of incoming light are more important controls of inter-annual variation of the CO 2 exchange in this blanket bog (Lund et al 2015). Similarly, Pelletier et al (2015) found boreal peatlands with pools in Quebec, Canada are net sinks of CO 2 at the ecosystem level during the growing season using oneseason of EC measurement of NEE and CO 2 flux despite the pools themselves being persistent sources of CO 2 .
In addition to the natural causes of carbon dynamics and changes on high latitude wetland ecosystems, agricultural activities such as grazing also affect vegetation structure and greenhouse gas emissions in a high arctic mire (Falk et al 2015). Falk et al (2015) conducted controlled grazed and ungrazed experiments to investigate the impact of muskox grazing on the vegetation composition and the carbon balance in a high arctic mire located in the Zachenberg valley, NE Greenland. It was found that net ecosystem uptake of CO 2 and CH 4 emissions have decreased by 47% and 44% in the ungrazed experiment plots respectively, suggesting the importance of grazing mammals in ecosystem greenhouse gas dynamics. Another study evaluated 5 years of EC measurement of CO 2 and 4 years of chamber measurements of CH 4 from a freshwater Marsh in Ottawa, Ontario, and found that the Marsh C budget as a sink was significantly reduced when combining CO 2 and CH 4 emissions (Strachan et al 2015) due to the substantial Marsh CH 4 emissions. Strachan et al (2015) also demonstrated that cold season ecosystem respiration, driven by temperature and snowfall, was an important component of the Marsh annual C budget and Marshes should be included in national and global estimates of wetland greenhouse gas contributions to the global carbon cycle.
Johnston et al (2014) investigated the effects of lowland permafrost thaw over millennial timescales using a measurement of CO 2 and CH 4 from thermokarst collapse bogs and adjacent fen locations that contain a ∼1000 yr thaw chronosequence in western Alaska. They found that CH 4 efflux has been enhanced during the growing season following lowland permafrost thaw over decadal time scales, suggesting that thaw features (such as thaw stages and cycles) should be included when evaluating high latitude CH 4 dynamics (Johnston et al 2014).
This special issue also includes a perspective by Stark and Ylanne (2015) on the importance of grazing in affecting ecosystem processes and C budgets in Arctic peatlands. They develop a conceptual model of possible pathways by which grazing and climate change may affect peatland C storage, CO 2 balance, and CH 4 release in the Arctic.

Tropical and subtropical wetlands
Tropical and subtropical wetlands contribute at least 50% of total wetland methane emissions, and over 80% of the natural sources, to the atmosphere partially due to large inundated areas and high temperatures Understanding the carbon dynamics and their influential factors will be critical to the estimation of global carbon budgets and its spatial distribution. Jauhiainen et al (2014) evaluated the heterotrophic CO 2 , N 2 O, and CH 4 fluxes in a drained tropical peatland under four shading conditions with different treatments in Central Kalimantan, Indonesia. They found that soil shading by vegetation decreases greater greenhouse gas (GHG) emissions to the atmosphere through reducing peat temperatures. Land management differences affect how greenhouse gas fluxes respond to peat temperature and fertilization, but increases in temperatures do not necessary translate into GHG emissions. The internal characteristics of peatlands (such as open degraded peatlands, intact peatland forests) may decrease the sensitivity of GHG emissions to temperature change (Jauhiainen et al 2014). A study by Oliveras et al (2014) assessed net primary productivity and carbon cycling in Andean tropical alpine grasslands based on direct measurements and found that Andean grasslands had similar NPP and soil carbon stocks to tropical montane cloud forests. Therefore, high elevation tropical grasslands are an important reservoir for soil carbon stocks and production (Oliveras et al 2014). A study of CO 2 budget dynamics in a subtropical estuarial Marsh wetland ecosystem in Taiwan shows that temperature and radiation have stronger influence on gross primary production of CO 2 than soil moisture content and vapor pressure deficit. Further, these environmental variables have strong but different impacts on the CO 2 budget in the two different low-latitude ecosystems (para grass and reed) (Lee et al 2015).

Wetland ecosystem modeling
Wetland ecosystem models have been used to estimate CH 4 and CO 2 fluxes at regional and global scales (Zhuang et al 2004, Wania et al 2010. Uncertainties associated with the environmental controls of methane biogeochemical processes prevent the accurate prediction of global and regional methane budgets  (2014), estimated the annual methane emission to be 67.8±6.2 Tg CH 4 yr −1 and regional CO 2 sink of −1.28±0.03 Pg C yr −1 . Zhuang et al (2015) found northern wetlands to be a consistent greenhouse gas sink per global warming potential and that variability in wetland inundation plays a very important role in affecting regional greenhouse gas budgets. Jin et al (2015) applied the TEM to estimate net exchanges of methane and CO 2 fluxes on the Qinghai-Tibetan Plateau from 1979 to 2100 under different climate change scenarios. Their simulations, by considering both CO 2 and CH 4 fluxes, demonstrate that the Qinghai-Tibetan Plateau will serve as a carbon source from 1979-2030 and turn to a carbon sink from 2030 to 2100.

Coastal wetlands and methane measurement techniques
The flux chamber technique has been one of the widely used methods to measure methane fluxes. Luan and Wu (2014) assessed how chamber characteristics affect artificial biases in methane fluxes measured using static chambers and found that opaque chambers tended to measure a significantly higher methane emission from hummocks than transparent chambers, but not at the hollows in a boreal peatland. It was concluded that both microtopography (hummocks versus hollows) and chamber characteristics can have large influence on the accuracy of methane flux measurements (Luan and Wu 2014). This information will be very valuable to field measurements to get more accurate measurement of CH 4 fluxes.
Caplan et al (2015) studied the primary production of Phragmites in response to increased CO 2 level and N enrichments in a North American mid-Atlantic Coast wetland using observations and model simulations. It was found that N enrichment increased carbon assimilation in the early and late growing season and CO 2 elevation enhanced assimilation more moderately in the early and mid-growing season. This suggests that carbon sequestration could be enhanced through the increase in Phragmites productivity if Phragmites were allowed to invade into the coastal wetland. Further research is needed to address the impact of Phragmites on coastal wetland ecosystems and carbon budgets.

Concluding remarks and future research
Wetland ecosystems are one primary component of the global carbon budget. The collection of 14 letters and one perspective in this special issue provides an excellent starting point for researchers who are interested in wetland ecosystem and carbon dynamics. Both field measurements and modeling studies are included to address different aspects of carbon cycling at local and regional scales. Environmental variables that affect CO 2 and CH 4 fluxes at local and regional scales are investigated, including temperature, radiation, agricultural treatments and land management, soil moisture content, and vapor pressure deficit. With the current global warming trend, it is expected that both tropical and northern wetland ecosystems will experience large changes in temperature and wetting and drying cycles. Future research should continue focusing on the influential factors that affect carbon dynamics across different wetland ecosystems. A few variables that were not addressed in this special issue are soil pH and redox potentials, soil moisture dynamics, and vegetation types and their impacts on CH 4 and CO 2 emissions. More field measurements of carbon budget and fluxes are required to further understand the dominant environmental variables that impact wetland carbon dynamics.