Effects of land use, moisture, temperature and substrate on production of CO2 and consumption of CH4 were measured in a series of laboratory incubation experiments on bulk soil samples from 0–10, 10–20, 20–40, 40–60, 60–80 and 80–100 cm soil depths under four dominant land uses [forest, grazing land, irrigated rice on level terraces (Khet), and upland maize–millet on sloping terraces (Bari)] of Mardi watershed (area=144 km2), Nepal. In addition, baseline physical and chemical properties of these soils were measured. The production of CO2-C day−1 per unit soil organic carbon (SOC) content in topsoil was lowest in grazing land, indicating a possibility of higher C sequestration with this land use than with other land uses. There was a decreasing trend of CO2 emission with soil depth in all land uses, as was also the case with the SOC content. The CO2 emission was increased by 90, 58, 27 and 23% for Bari, Khet, grazing and forest soil, respectively, with increase in moisture level from 40 to 60% (w/w). The CO2 release from forest soil went up from 15 to 50 μg CO2 g−1 dry soil with increase in temperature from 5 to 15 °C and it further increased to 67 μg CO2 g−1 at 20 °C with estimated Q10 values of 3.4 and 1.8, respectively. Significantly higher amounts of CO2 were emitted from all the land use types upon addition of glucose to the soil, indicating high potential of microbial activity. Consumption of CH4 was more rapid in the soil from 10 to 20 cm depth for all the land use types. There was a 89% increase in CH4 consumption in forest soil with increase in moisture level from 40 to 60%, while it was decreased by 38, 73, and 40% for Khet, Bari and grazing soil, respectively. Addition of (NH4)2SO4 inhibited CH4 oxidation in soils of all land uses, indicating a negative effect of N fertiliser input on CH4 uptake in soil.