Date of Award

Spring 2020

Project Type


Program or Major

Earth Sciences

Degree Name

Doctor of Philosophy

First Advisor

Ruth K Varner

Second Advisor

Michael W Palace

Third Advisor

Wilfred M Wollheim


Arctic regions are experiencing more rapid warming than other parts of the world, leading to destabilization of carbon (C) that has been sequestered in permafrost, especially in peatlands where the C content of the peat is very high. More frequent incidence of thaw in permafrost peatlands is leading to the development of small thaw ponds that are known to be sources of methane (CH4) to the atmosphere, yet there is a lack in long-term studies of CH4 emission from these formations. This is of concern because CH4 has thirty-two times the global warming potential of carbon dioxide over a one-hundred-year timescale (Holmes et al., 2013). At a site in northern Sweden, we have collected over 3000 measurements of CH4 ebullition, or bubbling, from eight small thaw ponds (<0.001 km2) differing in physical and hydrological characteristics over seven growing seasons (2012-2018).

We found ebullitive emission to be highly variable over space and time, with an average emission rate of 21.9 mg CH4 m-2 d-1. Between 2012 and 2015, ebullitive emission was weakly correlated with environmental conditions like atmospheric pressure and temperature and potentially more influenced by the physical characteristics of the ponds themselves. Based on their rates of daily ebullitive emission, the ponds fell into four statistically significant groups which appeared to differ from each other based on physical characteristics among the ponds within each group. This grouping, further called pond types, distinguishes ponds from one another based on vegetation presence, pond depth, and hydrologic connectivity to neighboring fen areas (or lack there-of). Type 1, with the lowest daily ebullitive emissions measured, are the shallowest ponds, they are hydrologically isolated have low instances of sedge vegetation (Carex spp. and Eriophorum spp.) and have Sphagnum spp. mosses present within them. Type 2 ponds, which emit more ebullitive CH4 than type 1, are deeper, have more sedge vegetation present and are hydrologically isolated. Type 3 ponds are this highest emitting on a daily scale and are the deepest, with more sedge vegetation present than type 3 yet remain hydrologically isolated. Type 4, are shallower than type 3, have no Sphagnum spp. present, are surrounded by sedge vegetation and connected to a neighboring fen area allowing water to flow. Based on our findings, and the available literature, we estimate that small ponds (< 0.001 km2) emit between 0.2 and 1.0 Tg of CH4 through ebullition over an estimated 149 ice-free days. Using acoustic techniques, we determined that on a sub-daily timescale CH4 emission rates varied significantly over space and time within a single pond with diel variability in bubbling rate following that of air temperature, shortwave radiation and wind speed. Using remotely collected imagery from an unmanned aerial system (UAS) platform of seven ponds collected over five sampling seasons (2014 — 2018) we found pond edge and water area varied significantly between ponds as well as over time, with water area varying significantly between pond types. Annual ebullitive flux was highest in ponds that ranged in pond edge area of 50 – 150 m2 with smaller and larger ponds emitting less, however this relationship is likely more related to physical differences between the ponds, rather than differences in overall size.

This work supports the importance of long-term studies that take advantage of a range of spatial and temporal scale sampling techniques in order to adequately capture the variability in CH4 ebullition from these highly dynamic formations. Not only are high resolution measurements of CH4 ebullition important, but the tandem monitoring of pond size and other physical characteristics that distinguish ponds from one another are also important to better understand the observed CH4 emissions. With an increase in the number of long-term studies such as this, we will be better able to model CH4 emissions from thawing permafrost ecosystems in the future.