Origin and fate of nitrates in groundwater from the central Po plain : insights from isotopic investigations

Original data and information from recent studies are presented with the goal of identifying the N sources, sinks, major transformations and sites of accumulation within groundwater of the Lombardy plain. Particular emphasis is placed on the use of stable isotopes (δ15NNO3, δ18ONO3 and δ11B) in fulfilling these objectives. The area covered by this study accounts for about 18% of the Po catchment and 30% of the Po plain area. Nitrates in groundwater are not homogeneously distributed: concentrations >50 mg L−1 are often observed at the Alpine foothills, while values below 25 mg L−1 characterise other areas, especially the lower plain. An inventory of the potential N sources is conducted, revealing a complex pattern, as the distribution of nitrates in groundwater does not fully match the distribution of any of the identified N sources. In the higher plain, groundwater contamination is favoured by (i) the high permeability of the unsaturated zone and of the aquifer; (ii) the great depth of the water table; (iii) intensive cattle raising and the consequent manure effluents; (iv) the large amount of water used for irrigation. Stable isotopes of dissolved nitrates indicate the absence of denitrification, while the coupled use of B isotopes evidences, even in rural areas, a contribution from septic effluents. At the transition between the higher and the lower plain, in the so called “spring belt”, substantial recycling to the surface of the contaminated groundwater occurs, affecting the N content of surface water. Denitrification is evidenced in the lower plain both by hydrochemical (Fe, Mn) and isotopic tools, and is favoured by a shallow water table, within 5 m of the surface. Denitrification is estimated to remove about 40–60% of the initial nitrates from groundwater in maize fields, and up to 80% in rice fields. The findings suggest that, in order to limit the N export to the Po river and ultimately to the Adriatic Sea, groundwater contamination first needs to be remediated. The amount of N estimated to be stored in the higher plain is about 370,000 t as N–View the MathML sourceNO3- or about 1 t ha−1. If only agricultural N is regarded as a source, this stock would result from approximately 15 a of surplus N accumulation. If mitigation only targets the agricultural N surplus, recovery times in the order of decade(s) are to be expected for the aquifers.