Abstract A field level input–output balance was calculated for macronutrients (phosphorus (P), potassium (K) and magnesium (Mg)) and trace elements (cadmium (Cd), copper (Cu) and zinc (Zn)) during one cropping season in adjacent conventional and organic dairy farming systems at the Öjebyn experimental farm in northern Sweden. Both systems had a 6-year crop rotation: forage crop (oats and peas) with undersown ley (year 1)—grass/clover ley (years 2–4)—barley (year 5)—potatoes/forage crop (year 6). All aboveground biomass was harvested except for the potato crop, where only the tubers were removed. Inputs via farmyard manure and urine were significant for all element balances except Mg where lime (dolomite) was the most important source. Inputs from atmospheric deposition and outputs via water run-off were important for the trace element balances. There were large differences in the element balances between individual crops, with a tendency for a positive balance for fields growing oats and peas, barley and potatoes and a negative balance for grass/clover ley. This could be attributed to management factors, especially the uneven manure and lime application that occurred between crops. Averaged field balances, calculated for the crops included in the crop rotation, indicated that P was approximately in balance for the organic system and slightly more positive in the conventional one. The K balance was negative in the organic system and positive in the conventional system. Positive balances in both farming systems were apparent for Mg, Cd and Zn. In contrast, the balances for Cu were both negative. The mass of materials involved, rather than differences in their composition, mainly accounted for differences in element flows between the two farming systems. Higher stocking rate, total volume of urine and lime application in the conventional system accounted for the main differences in element balances between systems.