Assessment protocol

Data sources

The HELCOM Contracting Parties annually report waterborne inputs of nitrogen and phosphorus from rivers and direct point sources to Baltic Sea sub-basins. Data on atmospheric emissions and monitored atmospheric deposition are submitted by countries to the Co-operative programme for monitoring and evaluation of the long-range transmission of air pollutants in Europe (EMEP), which subsequently compiles and reports this information to HELCOM. In accordance with Recommendation 37-38-1

"Waterborne pollution input assessment (PLC-Water)", sources of nutrients input are assessed every six year.

Nutrient input data can be viewed in HELCOM PLC reports (e.g. HELCOM 2012, HELCOM 2013d and HELCOM 2015).


Trend analysis and statistical processing

Annually reported data on riverine and directs inputs are quality assured and approved by national data reporters and data assurers. Assessment dataset for 1995-2015 based on the reported data was established after expert reviewing and filling data gaps in. Riverine data are flow normalized for individual rivers. Input from unmonitored areas aggregated and normalized for sub-basin. EMEP delivered actual and climate normalize annual nitrogen deposition data. For information about normalization of airborne and flow normalization of waterborne input data, see Annexes 9.3 and 9.4 of the PLC-5.5 report (HELCOM 2015) and Larsen and Svendsen (2013).

As part of the HELCOM PLC-6, PLC-7 and MAI CART OPER projects the trend analysis was carried out by DCE, Aarhus University (Denmark), with linear regression Mann-Kendall methodology (Hirsch et al. 1982) on:

  • flow normalized waterborne inputs (sum of flow normalized riverine data and direct point sources)
  • normalized airborne inputs)
  • total normalized inputs of nitrogen and phosphorus

for all relevant combinations of Contracting Parties and sub-basins of the Baltic Sea. Where there is a significant trend, the annual changes were determined with a Theil-Sen slope estimator (Hirsch et al., 1982). The change since 1995 and the reference period (average of 1997-2003) was calculated based on the normalized inputs. The methodology has been agreed on by HELCOM LOAD and HELCOM PRESSURE (more information on trend analysis and determining the changes in input can be found in Larsen & Svendsen 2013, and in document "Comparison of methods applied to evaluate progress towards CARTS and fulfilment of national inputs ceilings" for CART WS 1-2017). Compared to the first evaluation of MAI fulfilment also a test for break points has been performed for all sub-basins of the Baltic Sea. The breakpoints were identified using an iterative statistical process which determines the most significant break point (see document 6-1, HELCOM PLC-7 1-2017. If a break point is identified, the time series is divided into at least 2 segments, and trends are tested for each segment of the series. It is tested if the trend in the segment are significant, and if the trend are significant in the latest sections a linear regression model are used to estimate the annual input values in 2015. If there is no significant trend the estimated values in 2015 is the average of the normalized annual inputs in the time series after the break point if any.

The evaluation of MAI fulfilment is based on comparing MAI for each basin and the Baltic Sea with the trend estimated normalized annual total nitrogen and phosphorus inputs in 2015 including uncertainty on these inputs (Table 2) (described in document "Comparison of methods applied to evaluate progress towards CARTS and fulfilment of national inputs ceilings" for CART WS 1-2017 . In the first and second evaluation of MAI fulfilment uncertainty of average of respectively 2010-2012 and 2912-2014 normalized inputs was estimated from the variation of the 3year inputs around the average.

Student t-test is used testing for significant difference in the trend estimated input and inputs in the reference period (average of normalized annual inputs 1997-2003).


Assessment units

Nutrient input data have been compiled in accordance with PLC guidelines for the following nine sub-basins: Bothnian Bay, Bothnian Sea, Archipelago Sea, Gulf of Finland, Gulf of Riga, Baltic Proper, Western Baltic, The Sound and Kattegat. The boundaries of the sub-basins coincide with the main terrestrial river basin catchments.

The BALTSEM model has divided the Baltic Sea into seven sub-basins in accordance with natural marine boundaries and hence the MAIs have been calculated for the following seven sub-basins: Kattegat, Danish Straits, Baltic Proper, Bothnian Sea, Bothnian Bay, Gulf of Riga and Gulf of Finland. In the BALTSEM sub-division, the Bothnian Sea includes the Archipelago Sea area and the Danish Straits combine Western Baltic and The Sound.

The entire Baltic Sea is covered by the assessment.