Can Long-Term Precipitation Trends Explain Net Annual Carbon Loss From High Elevation Alpine Tundra?

Five continuous years of eddy covariance measurements over predominantly snow-free alpine tundra on Niwot Ridge, Colorado show that ecosystem respiration dominates over gross primary productivity on an annual basis, and that this ecosystem is a significant source of carbon to the atmosphere over long periods of time. Long-term data also show that precipitation has increased since the 1960s, in contrast to modeled forecasts that generally predict decreasing precipitation through the 21st century across the Rocky Mountain region. To constrain the specific relationship between precipitation and the alpine tundra carbon cycle, we tested the degree to which precipitation and soil moisture determined respiration fluxes over the course of three years, and across a range of 17 sites, including xeric, mesic, and hydric alpine tundra soils, within the footprint of ongoing eddy covariance measurements. Overall, we found that respiration from this ecosystem was principally moisture-limited. Cumulatively, the highest respiration rates were measured from hydric soils associated with seasonal ice lenses and perched water tables, however, growing season respiration rates peaked in mesic areas when hydric soils were saturated. Respiration from xeric soils increased with soil moisture, but fluxes from these areas were small in magnitude relative to mesic and hydric soils. Changes in precipitation and resultant soil moisture thus invoked a bidirectional response from alpine tundra soils, as moisture and respiration were positively correlated in some areas, but negatively correlated in others, depending on landscape position and prevailing soil moisture regime. Interannually, however, respiration fluxes were highest in wet years, indicating that moisture stimulated respiration from xeric and mesic soils more than it was suppressed from hydric soils. In sum, increased precipitation over the last half-century may be augmenting respiratory fluxes from alpine tundra, but changes in precipitation alone are insufficient to explain the magnitude of net annual carbon loss, in part due to the opposing responses of xeric, mesic, and hydric alpine tundra soils to changes in soil moisture.