These data describe the results of a three year (2011-2013) factorial experiment using plant-soil mesocosms testing the effects of biochar on soil biodiversity and soil carbon fluxes. The experimental design comprised three treatments: (1) biochar (absence or presence at 2% w/w); (2) plant type (barley, perennial ryegrass, or unvegetated); and (3) soil texture (sandy clay, sandy silt loam, clay loam). Ecosystem responses measured were net ecosystem exchange of carbon (NEE) & ecosystem respiration (both g CO2 m-2 h-1) and plant biomass (g aboveground and root). Soil biological responses measured were estimates of microbial community structure (fungal-to-bacterial ratio, total phospho-lipid fatty acid (PFLA) nmol g-1 soil) and densities (g-1 soil) of nematode worms and soil microarthropods (Collembola, Acari). The experiment was done at the Centre for Ecology & Hydrology in Penicuik, near Edinburgh in Scotland (UK). Soils used in the experiment were taken from the top 20 cm of the soil profile, from the James Hutton Institute’s Balruderry Farm near Dundee, Scotland, UK (56° 27’ N, 3° 4’ W). This research was funded by a Natural Environment Research Council Open CASE PhD studentship grant (NE/HO18085/1).