Post provided by Rebecca Sanders-DeMott and Pamela Templer

Processes that occur in winter are a significant component of annual carbon and nutrient cycles. ©Travel Stock Photos
Processes that occur in winter are a significant component of annual carbon and nutrient cycles. ©Travel Stock Photos

The climate is changing throughout the globe with consequences for the biogeochemical processes and ecological relationships that drive ecosystems. Scientists have been conducting manipulative experiments to determine the effect of climate warming on ecosystems for several decades. These experiments allow us to observe ecosystem responses before the climate changes occur and have yielded invaluable insight that has expanded our understanding of the natural world.

There is a wide range of creative approaches to mimicking climate warming that have been used, for example open-topped chambers which passively heat small areas of soil and small stature plants (like the ITEX global network), burying heating cables in the soil to directly increase soil temperatures (e.g. Harvard Forest experiments), infrared heating lamps (like Jasper Ridge), or even large scale chambers that can encompass taller stature plants like trees and actively warm the air (like the SPRUCE experiment). The focus of much of these inquiries has been on changes that occur during the growing season, when biological activity is at its peak.

Although traditionally considered the “dormant” period, we’re learning that in many seasonal ecosystems winter is much more important than previously understood. The processes that occur in winter are a significant component of annual carbon and nutrient cycles and can have effects that play out during the growing season. In our recent Methods in Ecology and Evolution paper – ‘What about winter? Integrating the missing season into climate change experiments in seasonally snow covered ecosystems’ – we reviewed climate warming experiments from the last three decades to determine how much these experiments can teach us about winter and what’s left to learn.

What does Snow have to do with Warming?

Thermal image of heated forest plots.  © Jon Chapell, Science Metrics LLC 2015
Thermal image of heated forest plots. © Jon Chapell, Science Metrics LLC 2015

Climate warming is not evenly distributed throughout the globe and continued warming is expected to be most pronounced at high latitudes and during winter, particularly in the northern hemisphere (Pithan & Mauritsen 2014; Xia et al. 2014).  So, accelerated warming will affect ecosystems that experience seasonal snow and ice cover, which make up 45% of land area in the northern hemisphere (Zhang et al. 2003).

The fact that warming is occurring rapidly in ecosystems at high latitudes, high elevations, and during winter is important because the effects of climate change on soil temperatures are mediated by interactions with snowpack. Snow is an excellent insulator, so underneath a deep, persistent snowpack, heat is trapped underground and soils remain right around 0 °C, even though air temperatures can get much, much colder.

In some regions, like many areas of tundra, where climate warming is expected to lead to greater winter precipitation as snow in the future, a deeper snowpack may better insulate soils and lead to warmer soils throughout the winter (Sturm et al. 2005).  In other areas, like many mid-high latitude forests and grasslands, reduced snowfall and accumulation are projected to occur.  This change will result in soils being exposed to cold and inconsistent winter air temperatures and is expected to lead to colder soils, often with increased frequency of freeze-thaw events (Henry 2008; Brown & DeGaetano 2011).  Also important are shifts in the timing of snowmelt in spring and/or the onset of snowpack in early winter, which have implications for growing season length and patterns of soil temperature in the shoulder seasons (spring and autumn).

Snow insulates soils from cold air (left). Deeper snow leads to warmer soil (centre). Less snow exposes soils to winter air (right).
Snow insulates soils from cold air (left). Deeper snow leads to warmer soil (centre). Less snow exposes soils to winter air (right).

Warming Experiments and Winter

In our review of 64 experiments designed to simulate future climate in seasonally snow-covered ecosystems, we found five distinct approaches to applying warming treatment across seasons. The most common form of ecosystem warming (60% of studies) included increasing soil temperatures by a consistent temperature in the growing season only (no treatment in winter, “A” below). The second most common warming method used in 20% of studies included increasing mean soil temperatures by a consistent temperature throughout the entire year (e.g. 5 °C above ambient soil temperatures in the growing season and in winter, “B” below).

The remaining three approaches employed warming methods that allow for varying degrees of distinct winter and growing season temperature manipulation. Together make up only 20% of warming experiments. Of these, experiments that applied a different type of warming in the winter versus the snow-free season accounted for 8% of studies (“C” below). This approach included experiments with a different magnitude of warming in the winter compared to the growing season to reflect projections of varying amounts of temperature change in different seasons, as well as those that employed alternate methods of warming in the winter, such as snow addition.

An additional 6% of experiments increased soil temperature variability in winter (usually by reducing winter snowpack depth and/or duration) and combined this treatment with increased temperatures in the snow-free season (“D” below). The remaining 6% of studies experimentally accelerated snowmelt and/or delayed the onset of snow pack coupled with warmed temperatures in the growing season (“E” below), allowing researchers to examine of the effects of an extended growing season on ecosystem processes.

What Might We be Missing?

Since there are few examples of experiments that have incorporated distinct winter change, we don’t really know the answer to this question.  But results from a few selected studies indicate that we might be missing a lot.  For example, Turner and Henry (2010) conducted a warming study that increased winter soil temperature variability in a cold temperate grassland and found that winter warming increased rates of nitrogen mineralization in soils  but this signal was completely offset by warming year-round. By warming in the tundra with snow addition, Natali et al. (2014) found that increased winter soil carbon losses from increased respiration completely offset the carbon uptake gains by accelerated photosynthesis during the growing season. And in a montane meadow in the Rocky Mountains, Hart et al. (2012) found accelerated snowmelt from warming was the strongest driver of shifts in plant community composition observed in a 23-year warming study. These studies all suggest that including winter manipulations can have important effects on the conclusions that we draw from ecosystem experiments.

Simulating Future Winters

Increasing recognition of the importance of winter climate change is evidenced by the numerous experiments that have been done to isolate the effects of snow and temperature in winter.  Methods include:

  • Manually removing snow from forests, grasslands, and agricultural lands via shoveling or by constructing shelters
  • Warming soils during winter by enhancing snowpack depth through snow fences that trap blowing snow
  • Actively warming soil with heat lamps
  • Increasing winter soil temperature variability passively through snow removal or actively through warming pulses
  • Shifting the timing of snowpack onset or melt through snow removal or addition.

(To learn more about how isolated changes in snow and soil temperatures influence ecosystems see reviews by Wipf & Rixen 2010, Blankinship & Hart 2012, and Makoto et al. 2014.)

Experimental snow removal plot. © Annie Socci 2012
Experimental snow removal plot. © Annie Socci 2012

At the Hubbard Brook Experimental Forest in New Hampshire, USA, a new experiment called the Climate Change Across Seasons Experiment has been explicitly designed to simulate the distinct effects of growing season and winter climate change in northern forests. It combines well-established methods for both ecosystem warming (via soil heating cables) and winter climate change simulation (via snow removal). Early results show that by including the projected changes in winter climate (that are distinct from mean annual warming in the snow-free season), ecosystem responses are observed that are not captured by warming alone (Sanders-DeMott in press; Sorensen et al. 2018).

There’s a great opportunity for more researchers to integrate established winter climate change methods with ecosystem-scale warming approaches in regions with seasonal snow cover. By designing experiments that examine the effects of both winter and growing season climate change together, we can develop a better understanding of ecosystem responses to climate change.

To find out more, read our Methods in Ecology and Evolution review article ‘What about winter? Integrating the missing season into climate change experiments in seasonally snow covered ecosystems’.