Date of Award


Degree Type


Degree Name

Master of Science (MS)


Environmental Sciences

First Advisor

José A. Amador


Methane, a potent greenhouse gas, is produced or consumed by microorganisms in soil, depending on soil conditions. In general, anaerobic conditions promote activity of methanogenic Archaea, whereas aerobic conditions promote methane consumption by methanotrophic bacteria. Earthworms may have an effect on methane flux in soil by altering soil pore space, aeration, moisture content, and organic carbon availability, all of which affect the activities of methanogens and methanotrophs. As the range of European and Asian earthworm species expands throughout the United States, their potential effects on methane flux may have global implications.

My research focused on the effects of earthworms on methane flux in temperate pastures and on the potential mechanisms involved. In November, 2009, I established 24 experimental plots each at Peckham Farm (PF), Kingston, and at the W. Alton Jones (WAJ) campus of the University of Rhode Island, West Greenwich, both in southern Rhode Island. Three treatments (n=8) were implemented: (i) earthworms removed (D−), (ii) earthworms removed, counted and returned to the plot (D+), and (iii) undisturbed (U). Weekly methane flux measurements were made from April to November, 2010. Soil moisture and temperature were continually recorded. Methane flux measurements were also made three times over a 24-h period at each site to quantify diurnal patterns. I also conducted a second experiment to examine the relationship between earthworm population size and soil properties, and methane flux that did not disturb the soil. I determined methane flux in these supplemental plots at seven randomly selected locations in an area of the field adjacent to the manipulated plot experiment. Once gas sampling was completed, I sampled the soil under each chamber for pH, organic matter content, NH4 and NO3 concentrations, soil moisture, and soil temperature. Earthworms were counted, weighed and identified.

I also examined the potential effects of earthworms on acetoclastic and hydrogenotrophic methanogenesis, and on methane oxidation, using cast, burrow, and bulk soil from soils inoculated with Anecic and epi-endogeic earthworms, the two most common earthworm ecological groups found at the research sites, and a control soil. Soil from bulk, burrow or casts was transferred to microcosms with conditions manipulated to favor only one of the three processes—methane oxidation, hydrogenotrophic methanogenesis, or acetoclastic methanogenesis. Microcosms were periodically sampled for methane concentration in the headspace to determine process rates.

Mean (± s.d.) methane flux (ng CH4-C/m2/sec) in the U treatment plots ranged from -98 ± 6 to -20 ± 9 at PF and -42 ± 7 to 2 ± 5 at WAJ. Methane flux in the D- treatment plots ranged from -95 ± 13 to -14 ± 11 at PF and -46 ± 11 to -4 ± 4 at WAJ. Mean methane flux in the D+ treatment plots ranged from -73 ± 12 to -14 ± 12 at PF and -38 ± 7 to -8 ± 5 at WAJ. Analysis of the data using a two-way repeated measures ANOVA showed no overall difference in methane flux among the three treatments at PF. On two out of a total 27 sampling dates, there was significantly higher methane consumption (more negative flux) in D- than in D+ treatments. Furthermore, on three out of 27 sampling dates the D- treatment had significantly lower methane consumption than the D+ treatment. However, no there were no significant differences among treatments in earthworm population density or biomass at the end of the experiment.

By contrast, there was a significant overall difference in methane flux among treatments at the WAJ site, although there were no significant difference in earthworm population densities or biomass among treatments at the end of the experiment. Net methane consumption was significantly higher in D- than in D+ treatments on seven out of a total of 26 sampling dates. The results from WAJ support the hypothesis that plots with earthworms removed should have higher net methane consumption than plots with earthworms, whereas results from PF suggest no effect from earthworm removal on methane flux. This suggests that the effect of earthworms on methane flux may be site-specific, contingent upon factors not quantified in this experiment.

For the supplemental plots, mean (± s.d.) methane flux (ng CH4-C/m2/sec) ranged from -80 ± 15 to -17 ± 4 at PF and -40 ± 18 to -14 ± 3 at WAJ. There was no apparent relation between methane flux and earthworm population density or mass for either site. Population densities of earthworms ranged from 10 ind./m2 on 19 August to 103 ind./m2 on 7 October at PF, and from 36 ind./m2 on 14 October to 132 ind./m2 on 17 June at WAJ. Using a backward-elimination stepwise multiple regression analysis I examined methane flux as a function of earthworm population density and soil properties - moisture, temperature, pH, O.M. content, NH4, NO3, and sampling date - as well as other earthworm parameters besides earthworm population density, including mass, ecological group, and life stage (juvenile or adult). Results showed no relation between methane flux and any earthworm parameter for either site.

The microcosm experiments yielded evidence for an effect of earthworms on methane dynamics. A one-way ANOVA and Holm-Sidak all-pairwise comparison showed that the anecic cast treatment had a significantly lower methane oxidation rate than all other treatments in soil from both sites for 80% of the sampling events. All other treatments were statistically similar to each other, with the exception of the epi-endogeic burrow soil at WAJ, which had a significantly higher methane oxidation rate than all other treatments on most sampling times. The anecic cast soil had significantly higher levels of nitrate and ammonium than all other treatments, suggesting an inhibitory effect of ammonium and/or nitrate on methane oxidation.

Results of a two-way ANOVA showed that there were no significant differences (p>0.05) in rates of hydrogenotrophic methanogenesis among treatments at any time in soil from either site (0.6 ± 0.4 to 2.3 ± 1.3 (µg C-CH4/sec/g soil) in the PF soil and 0.6 ± 0.4 to 3.1 ± 1.5 (µg C-CH4/sec/g soil) in the WAJ soil), except at t=71 h, when the epi-endogeic bulk soil had significantly higher rates of methanogenesis than the epi-endogeic burrow soil, the anecic burrow soil, and the anecic cast soil in the soil from PF. A two-way ANOVA showed no significant differences between rates of acetoclastic methanogenesis among treatments at any time in the PF soil. Acetoclastic methanogenesis rates ranged from 0 to 16.0 ±10.7 µg CH4-C/sec/g soil. There was no clear trend in the rate of methanogenesis over time. By contrast, there were significant differences in the rates of acetoclastic methanogenesis in WAJ soil at 115 h and 144 h. At 115 h, the epi-endogeic burrow treatment had significantly higher rates of methanogenesis than the control, the epi-endogeic bulk, the anecic burrow, and the anecic cast treatment. At 144 h, the epi-endogeic burrow treatment had significantly higher rates of methanogenesis than the control, the epi-endogeic bulk, and the anecic burrow treatment. These results show that the ability of soil to consume methane was lowered by anecic earthworm activity in soil from both sites (anecic cast) as compared to soil with no earthworms (control). In addition, one earthworm treatment (epi-endogeic burrow) at times (115h and 144h) had an effect of increasing acetoclastic methanogenesis in soil from WAJ as compared with soil from WAJ with no earthworms (control).

In the manipulated plots field experiment the majority of the differences were found at the W. Alton Jones site. This is in agreement with the results of the microcosm experiments, which suggest that earthworms may affect methane flux specifically by increasing acetoclastic methanogenesis and decreasing oxidation, and that effects may be site-specific. At PF, the decrease in methane oxidation as found in the anecic casts treatment in the microcosms may have occurred in the field but to a minimal degree, that which was not enough to cause a measurable difference. However, at WAJ, the combination of lower methane oxidation in anecic cast soil, paired with an increased acetoclastic methanogenesis in the epi-endogeic burrow soil, may drive differences in flux.