The Swedish Environmental Protection Agency has assigned to SMED the task of calculating waterborne nitrogen and phosphorus loads to the sea basins around Sweden for its HELCOM PLC5 (Fifth Pollution Load Compilation) report. The aim is for these calculations also to be usable by the Swedish water authorities, to monitor fulfilment of environmental objectives etc. Similar calculations were performed by the Swedish Meteorological and Hydrological Institute (SMHI) and the Swedish University of Agricultural Sciences (SLU) for the PLC4 report, in the ‘TRK’ (‘Transport — Retention — Source Apportionment: Pollution Loads to the Sea’) project (Brandt and Ejhed, 2002). The calculation methods used for PLC5 are the same, but collection of input data has been developed and improved. For phosphorus, agricultural leaching was estimated by means of a new method. The calculations relate to 2006 but are flow-corrected for the period 1985–2004 to prevent marked variations in climate situation for individual years, such as wet years with heavy runoff, skewing the figures. Flow correction means that the longterm averages for runoff are used. This method involves calculation of ‘gross load’, i.e. nitrogen and phosphorus loads from various sources, such as diffuse load and point sources, to the study area. Modelling (simulation) of retention, transformation and erosion along streams and rivers is then carried out by means of the estimated gross load and measured data. Finally, the ‘net load’, i.e. what reaches the sea, is calculated. With this method, both gross and net loads can be apportioned to their various sources. In the TRK project, the gross and net loads for nitrogen were reported, but only the gross load for phosphorus. In this calculation, retention calculations have been performed for phosphorus so that the net load for phosphorus, too, can be stated. Land-use leaching is calculated by multiplying the area (in km2) devoted to a particular land use by concentration figures or leaching coefficient (mg/l) for the same land use and by runoff (l/s per km2). The figure for total gross land-use leaching is then obtained by summarising the inputs from all the various land uses. In this project, calculations were based on subcatchments on land and islands located in marine areas (numbering some 13,500). These were subsequently summarised as ‘PLC5 catchments’ (numbering some 1,100) and then as main river catchments (119) and areas draining into HELCOM sea basins. The EU list of water bodies in Sweden had not been digitised by the time the calculations started, and subcatchments were therefore selected as the most detailed geographical level in the calculations. Land-use data were compiled using the Geographic Information System (GIS) from the National Land Survey of Sweden; the Swedish Board of Agriculture’s block maps; the Integrated Administration and Control System (IACS), an EU support database; and the Swedish Forest Agency’s data on cleared forest areas. Runoff on a daily basis for 1985–2004 was calculated for the PLC5 catchments using the HBV hydrological model, and then summarised to obtain annual and monthly averages. Leaching coefficients for nitrogen leaching from agriculture were estimated by means of the SOILNDB simulation tool, and phosphorus leaching from agricultural land was estimated using the ICECREAMDB simulation tool. Other land-use concentrations are based on summaries of measured data in small, more or less homogeneous catchments. Figures for atmospheric deposition of nitrogen on water (in lakes) were taken from the MATCH model, and those for phosphorus deposition were obtained from measurements. Calculation of loads from hard, i.e. impermeable surfaces (‘storm water’ in this report) in urban areas was performed using improved methods compared with those used in the TRK project. Figures for point-source pollution from large wastewater treatment plants and industrial facilities were taken from the county administrative boards’ EMIR database. Load from smaller plants were estimated on a standard basis, according to population equivalents, removal efficiency and emission factors. Estimation of loads from local on-site wastewater treatment was based on the number of person days per year, municipal information on the degree of treatment, and standardised emission data. Total gross nutrient load was calculated by summarising land-use leaching, atmospheric deposition on water areas and point sources per subcatchment. Using the HBV-NP model and measuring data from recipients, the biogeochemical changes in (retention of) nitrogen and phosphorus were simulated during transport from the subcatchment in question to the sea. Net load was calculated by subtracting the retention from the gross load. The load figures include inputs from catchments located in Norway and Finland but drained via Sweden (mainly the rivers Klarälven and Torneälven. With this method, concentrations and leaching coefficients from a number of land-use classes are used to represent land-use leaching. Relatively small deviations in these have a major bearing on load if they apply to large areas and may also affect the source apportionment. The table below shows the estimated nitrogen and phosphorus loads to water from Swedish sources in 2006, flow-corrected for the period 1985–2004. ‘Gross’ load refers to the amount entering water in the subcatchments (at source). Direct discharges from wastewater treatment plants and industries to the sea do not involve retention, and are included in both ‘gross’ and ‘net’ loads.