Article: Parmentier, Frans‐Jan W., Wenxin Zhang, Yanjiao Mi, Xudong Zhu, Jacobus Huissteden, Daniel J. Hayes, Qianlai Zhuang, Torben R. Christensen, and A. David McGuire. “Rising methane emissions from northern wetlands associated with sea ice decline.” Geophysical Research Letters (2015). doi:10.1002/2015GL065013
Feedback is not always positive
One of the greatest examples of climate change is the stark decrease in Arctic sea ice. When sea ice melts, it causes a climate feedback. Let me explain. That more-than-normal sea ice melt is a result of warming temperatures. However, when that sea ice disappears, it changes the regional albedo, or reflectivity of the land type. Sea ice is white, which generally causes a lot of solar irradiance to be reflected, while the underlying ocean is darker and typically absorbs more solar energy. In other words, as sea ice melts, more heat energy is absorbed by the Arctic region, ultimately making the Arctic water even warmer (Figure 1).
Unfortunately, sea ice melting may be a double whammy to warming in the Arctic. Parmentier et al. (2015) set out to test a hypothesis that would link disappearing sea ice to increased methane emissions from nearby terrestrial wetlands. Microbes in wetland soils produce methane through an anaerobic respiration process called methanogenesis. These microbes live in soils with no oxygen, so when they breathe, they produce methane instead of carbon dioxide. Methane is a powerful greenhouse gas and methane production increases when temperatures are warmer. Think about your own breathing patterns; after a light jog, you are likely to breathe heavier when it’s the dead of summer and lighter after lovely fall day!
Thus, is it possible that the warming associated with losing sea ice also causes more methane to be emitted from Arctic wetlands by intensifying microbial “breathing”?
The approach: Satellites and Models allow us to be “everywhere”
Arctic sea ice coverage and declines were measured using satellite observations from 1991-2010 and methane emissions were estimated using three different models (LPJ-GUESS, Peatland-UV, and TEM6) from 1981-2010 (Figure 2). The use of satellites and multiple models allowed for a large region to be investigated. Scientists cannot be physically present everywhere, so we use helpful tools like models to analyze more data than humanly possible!
The investigators in this study wanted to understand if A) the melting sea ice in the Arctic lead to increased methane emissions in nearby wetlands, or B) if that same warming that initially caused the ice to melt is also causing more methane to be produced. The best way to separate the two: distance! If enhanced methane was only being emitted near melting sea ice, then their hypothesis would be wrong. But if the northern wetland region were emitting more methane, despite distance from melting ice, then their hypothesis may hold true.
Parmentier et al. (2015), used grid cells (small units of area, 0.5° latitude by 0.5° longitude) to correlate sea ice content with modeled methane emissions. These Arctic grid cells were fitted with a linear correlation by removing the distance between sea ice and wetlands, which sets the mean to 0. This prevents “false” or coincidental trends from being observed. This detrending set the correlations on a scale from -1 and +1. A more negative correlation (near -1) would support the notion that melting sea ice results in increased emissions. A 0 would indicate no trend and a more positive correlation (+1) would suggest methane emissions are not correlated with sea ice loss.
What did they find?
From May to October, there was a clear negative correlation suggesting that decreasing sea ice increased methane emissions from the northern wetlands. It is reasonable to think that as climate change progresses, and sea ice decreases further, this feedback could promote even more methane emissions. This feedback (an endless spiral!) could in fact increase atmospheric temperatures and further decrease sea ice content.
This analysis predicted that an additional 1.7 Tg of methane per year from 2005 to 2010 was emitted as a result of this feedback loop, with an uncertainty of ±0.4 to 4.1 Tg of methane per year. This study was only able to assess the magnitude of methane production, and not the resulted warming.
Surprisingly, a few terrestrial regions, like the Canadian Archipelago, saw a decrease in methane. This region has very few wetlands, suggesting that the soils actually uptake this greenhouse gas. Clearly there is an important feedback between wetlands and sea ice.
The researchers suggest investigating this connection between sea ice and wetlands further. For example, most methane measurements used to validate the models are from the summer months while some of the strongest correlations where observed in the spring and fall.
Overall, methane has been increasing in the Arctic. While thawing permafrost is one potential cause, this study offers another contributing factor. The ocean and land are more connected than you may think!
Climate change comes in many different forms. This study used satellite measurements of sea ice decline and modeled methane emissions from land to demonstrate just how interconnected the land and sea are. Decreases in sea ice not only lower the albedo (making the regional temperature warmer), but also potentially increases methane production in Arctic wetlands. These processes ultimately make the Arctic region even warmer. Broadly, this climate feedback loop decreases sea ice even more, which reduces vital arctic habitat (the poor polar bears!) and creates more open water all year long (open shipping lanes)!