Methods of Study

Due to the complexity of the problem of 'pollution of coastal waters, seas and oceans with nutrients (nitrogen N and phosphorus P)', there are no standardized scientific methods for studies and research of this issue. Consequently various different methods have been used, which has lead to a great variety of results.

Though there is vast data available for some seas, e.g. the North Sea or the Baltic Sea, nonetheless there is still a considerable need for the development of standardized methods for the estimation of nutrient inputs, and for the preparation of guidelines for assessing comparable nutrient input data (153), which may include appropriate sampling programmes (210). The lack of basic data may in some cases lead to an overestimation, especially in the case of phosphorus (P) loadings to surface waters (184).

In spite of extensive monitoring in some research studies, the greater number of such studies is based on general assessment (46, 87, 280, 344; SCOPE Nitrogen Project 229, 230), on risk assessment of land-derived nitrogen (N) loads, on general estimates and/or uses a 'model' as it's base.

Therefore, no recommendation can be given on what method would be the best to study nutrient (nitrogen N and phosphorus P) pollution of coastal or marines waters in any specific case.

In addition to the studies based on general monitoring (17, 41, 129, 162, 164, 193, 198), and such based on a 'model' (7, 53, 69, 109, 110, 124, 126, 199, 214, 291), the following details of methods of studies used have been reported:

• Developing a scoring system for quantitative assessment of eutrophication (161); monitoring acc. to the PRISMA Project (216); monitoring water pollution using conductivity values (264); computer modelling (366); simulation model (228); physical model simulation (214); dynamic N loading model (NLM) (223); agricultural non-point source model (AGNPS) (226); box model examining internal vertical N fluxes (249); Waquoit Bay Nitrogen Loading Model (251); methods used in the Chesapeake Bay Pollution Abatement Program (42); HBN - N model (Sweden) (272); HBN -N / SCOBI / MATCH model (Sweden) (301); Princeton Ocean Model (273); DELPHI - model (tracer experiments) (308); simulation / optimization (S/O) model (81); effect-dose-sensitivity models for aquatic ecosystems (148); and catchment modeling (314);

• Multinominal logistic regression (58); statistical evaluation (45, 89); statistical process analysis (196); multivariate statistical analysis (114, 182); multiple regression analysis (101, 311, 379); linear regression equation (326); time series regression models (220); logistic regression model (237); model calculation (109); two-dimensional numerical model (110); non parametric method for calculating changes (116); using macro- and micro-gradients (119); time series analysis (132); mathematical simulation (213); simulation modelling using four systems of ordinary differential equations (317); model based on the emission approach (120); method comparing anthropological and 'natural' inputs (186); materials accounting techniques (165); evaluation using ANOVA (328);

• Mass balance approach (38, 43, 108); mass budget equation (14); budget approach (208); using mass fraction (355);

• Isotopic analysis (delta 15N, delta13C, delta 14S) (221); delta 15N (234); measuring de-nitrification with delta 15N (232); using 137Cs dating techniques (P) (44); based on sequential extraction scheme (339); liquid-solid phase extraction followed by gas chromatography - electron capture detection (GC - ECD) (47);

• Water sample analysing acc. to Murphey and Riley (32); base-flow sampling (219); method based on intensive and strictly controlled sampling regime (118); analysis based on water sampling by cruises (179); based on monthly sampling (304);

• Establishing a Geographic Information System (GIS) for site modelling (57); GIS watershed approach (255); method based on digitized maps of land use and other factors (120); using GIS and RS (remote sensing) (240);

• Method based on reconstituted sediment-water cores under laboratory conditions (206);

• Method based on the relationship between pore water NO3(-) N concentration and
  NO3-N flux rates for estimating nitrogen loading from ground water (239);

• Method based on size distribution of particulates (262); based on measurement of N : P atomic ratios in phytoplankton (330);

• Simulating nutrient runoff (48);

• Methodology estimating nitrous oxide (N2O) emissions acc. to Seitzinger and Kroeze (20).