Impact of agronomic practices of an intensive dairy farm on nitrogen concentrations in a karst aquifer in Ireland

•Statistical analysis of N occurrence in groundwater underneath a dairy farm.•Long-term study in a karst region with thin overburden.•Assessment of climatic, agronomic and (hydro-)geologic influences.•A vertical time lag from source to receptor of 1–2 years was estimated.•Water Framework Directive targets can be achieved on similar sites. Exploring the relationship between agricultural nitrogen loading on a dairy farm and groundwater reactive nitrogen concentration such as nitrate is particularly challenging in areas underlain by thin soils and karstified limestone aquifers. The objective of this study is to relate changes in detailed agronomic N-loading, local weather conditions, hydrogeological and geological site characteristics with groundwater N occurrence over an 11-year period on an intensive dairy farm with free draining soils and a vulnerable limestone aquifer. In addition, the concept of vertical time lag from source to receptor is considered. Statistical analysis used regression with automatic variable selection. Four scenarios were proposed to describe the relationships between paddock and groundwater wells using topographic and hydrogeological assumptions. Monitored nitrate concentrations in the studied limestone aquifer showed a general decrease in the observed time period (2002–2011). Statistical results showed that a combination of improved agronomic practices and site specific characteristics such as thicknesses of the soil and unsaturated zone together with hydrogeological connections of wells and local weather conditions such as rainfall, sunshine and soil moisture deficit were important explanatory variables for nitrate concentrations. Statistical results suggested that the following agronomic changes improved groundwater quality over the 11-year period: reductions in inorganic fertiliser usage, improvements in timing of slurry application, the movement of a dairy soiled water irrigator to less karstified areas of the farm and the usage of minimum cultivation reseeding on the farm. In many cases the explanatory variables of farm management practices tended to become more important after a 1- or 2-year time lag. Results indicated that the present approach can be used to elucidate the effect of farm management changes to groundwater quality and therefore the assessment of present and future legislation implementations.