The performance of six separate percolation areas was intensively monitored to ascertain the attenuation effects of unsaturated subsoils with respect to on-site wastewater effluent: three sites receiving septic tank effluent, the other three sites receiving secondary treated effluent. The development of a biomat across the percolation areas receiving secondary treated effluent was restricted on these sites compared to those sites receiving septic tank effluent and this created significant differences in terms of the potential nitrogen loading to groundwater. The average nitrogen loading per capita at 1.0m depth of unsaturated subsoil equated to 3.9 g total-N/d for the sites receiving secondary treated effluent, compared to 2.1 g total-N/d for the sites receiving septic tank effluent. Relatively high nitrogen loading was, however, found on the septic tank sites discharging effluent into highly permeable subsoil that counteracted any significant denitrification. Phosphorus removal was generally very good on all of the sites although a clear relationship to the soil mineralogy was determined.

The application of drones has recently revolutionised the mapping of wetlands due to their high spatial resolution and the flexibility in capturing images. In this study, the drone imagery was used to map key vegetation communities in an Irish wetland, Clara Bog, for the spring season. The mapping, carried out through image segmentation or semantic segmentation, was performed using machine learning (ML) and deep learning (DL) algorithms. With the aim of identifying the most appropriate, cost-efficient, and accurate segmentation method, multiple ML classifiers and DL models were compared. Random forest (RF) was identified as the best pixel-based ML classifier, which provided good accuracy (≈85%) when used in conjunction graph cut algorithm for image segmentation. Amongst the DL networks, a convolutional neural network (CNN) architecture in a transfer learning framework was utilised. A combination of ResNet50 and SegNet architecture gave the best semantic segmentation results (≈90%). The high accuracy of DL networks was accompanied with significantly larger labelled training dataset, computation time and hardware requirements compared to ML classifiers with slightly lower accuracy. For specific applications such as wetland mapping where networks are required to be trained for each different site, topography, season, and other atmospheric conditions, ML classifiers proved to be a more pragmatic choice.

More than half of surface water bodies in Europe are at less than good ecological status according to Water Framework Directive assessments, and diffuse pollution from agriculture remains a major, but not the only, cause of this poor performance. Agri-environmental policy and land management practices have, in many areas, reduced nutrient emissions to water. However, additional measures may be required in Ireland to further decouple the relationship between agricultural productivity and emissions to water, which is of vital importance given on-going agricultural intensification. The Source Load Apportionment Model (SLAM) framework characterises sources of phosphorus (P) and nitrogen (N) emissions to water at a range of scales from sub-catchment to national. The SLAM synthesises land use and physical characteristics to predict emissions from point (wastewater, industry discharges and septic tank systems) and diffuse sources (agriculture, forestry, etc.). The predicted annual nutrient emissions were assessed against monitoring data for 16 major river catchments covering 50% of the area of Ireland. At national scale, results indicate that total average annual emissions to surface water in Ireland are over 2700tyr-1 of P and 82,000tyr-1 of N. The proportional contributions from individual sources show that the main sources of P are from municipal wastewater treatment plants and agriculture, with wide variations across the country related to local anthropogenic pressures and the hydrogeological setting. Agriculture is the main source of N emissions to water across all regions of Ireland. These policy-relevant results synthesised large amounts of information in order to identify the dominant sources of nutrients at regional and local scales, contributing to the national nutrient risk assessment of Irish water bodies.