Earthworms play an essential part in determining the greenhouse gas (GHG) balance of soils worldwide. Their activity affects both biotic and abiotic soil properties, which in turn influence soil GHG emissions, carbon (C) sequestration and plant growth. Yet, the balance of earthworms stimulating C sequestration on the one hand and increasing GHG emissions on the other has not been investigated. Indeed, much is still unclear about how earthworms interact with agricultural land use and soil management practices, making predictions on their effects in agro-ecosystems difficult. This thesis determines whether the extent of GHG mitigation by soil C sequestration as affected by earthworms is offset by earthworm-induced GHG emissions from agro-ecosystems under different types of management. To achieve this aim, mesocosm and field studies are combined, as well as meta-analytic methods to quantitatively synthesize the literature. Using meta-analysis, it is shown that, on average, earthworm activity leads to a 24% increase in aboveground biomass, a 33% increase in carbon dioxide (CO 2) emissions and a 42% increase in nitrous oxide (N 2O) emissions. The magnitude of these effects depends on soil factors (e.g., soil organic matter content), experimental factors (e.g., crop residue addition or fertilizer type and rate) and earthworm factors (e.g., earthworm ecological category and -density). Conducting both a mesocosm and a field study record that earthworm activity results in increased N 2O emissions from fertilized grasslands. Further, field conditions record an increase in earthworm-induced N 2O emissions in autumn but not in spring, suggesting that earthworm effects in the field depend on soil physicochemical parameters influenced by meteorological and seasonal dynamics. The unique two-year experiment with a simulated no-tillage (NT) system and a simulated conventional tillage (CT) system, record that earthworm presence increases GHG emissions in an NT system to the same level as in a CT system. This suggests that the GHG mitigation potential of NT agro-ecosystems is limited. When considering the C budget in the simulated NT system, it is demonstrated that over the course of the experiment earthworms increase cumulative CO 2 emissions by at least 25%, indicating a higher C loss compared to the situation without earthworms. Yet, in the presence of earthworms the incorporation of residue-derived C into all measured soil aggregate fractions also increased, indicating that earthworm activity can simultaneously enhance CO 2 emissions and C incorporation into aggregate fractions. In conclusion, the revealed dominance of GHG emissions over C sequestration as affected by earthworms implies that their presence in agro-ecosystems results in a negative impact on the soil greenhouse gas balance.
