An algal bloom fills the Baltic Sea in this image from 2005
Large sources of pollution to the Baltic Sea have been missed by existing monitoring efforts, according to research from Sweden.
The health of the Baltic Sea has been declining since the 1960s, due to the disposal of untreated human waste and toxic materials such as heavy metals.
Countries bordering the Baltic have put together an action plan to stem the tide of contaminants entering the sea.
The pollution hotspots are deceptively small settlements near the coast.
Professor Gia Destouni, from Stockholm University, said these areas were being left without systematic environmental monitoring.
The Baltic Sea is also being harmed by the "nutrients" from fertiliser used in agriculture.
The lack of water exchange between the Baltic and the Atlantic and Arctic Oceans makes the sea water naturally more prone to dead zone episodes, where the water has very low concentrations of dissolved oxygen.
But the problem is exacerbated by the use of fertiliser - including nitrogen and phosphorus - which runs off fields into streams and rivers. When it reaches the ocean, it provides nutrients for algae, which can form blooms.
This in turn leads to more organic matter reaching the bottom of the sea, where bacteria break it down, using up the oxygen required by marine animals on the sea floor.
"Because of practical difficulties, you cannot monitor everywhere. So people do focus on the big rivers, for a number of different reasons," Professor Destouni told BBC News.
"You have these 'blind spots' in coastal zones all over the world. But the Baltic Sea is one of the most polluted and (nutrient-filled) seas in the world."
She added: "In earlier work, when we tried to quantify these holes, we saw that Sweden had an extra large fraction of these blind spots and an extra large population in them."
Although the areas themselves may be small, they extend along most of Sweden's coastline and are home to a large proportion of the country's population.
Sweden was found to have an unmonitored area where water drains into the Baltic of 20%. This same area is home to 55% of the country's population.
The new research shows that concentrations of nitrogen and phosphorus and organic pollutants in the water flowing from these unmonitored areas may be much larger than in the main rivers that are subject to systematic environmental monitoring.
In the journal Global Biogeochemical Cycles, Professor Destouni and her colleagues found that the reported loads of nitrogen and phosphorous from Sweden to the Baltic Sea are significantly smaller than would be expected.
This expectation is based on strong connections, found by the study, between a country's loads of different contaminants and its population size, area and economic activity.
Bridging the gap
Professor Destouni explained: "If we look at the reported nutrient loads from all around the Baltic Sea drainage basin, and compare the rest of the countries - which have a much more similar fraction of blind spots - with the Swedish reports - which have a higher fraction of blind spots - we should see a significant difference.
"We see a very significant difference between what Sweden is reporting and what the other countries are reporting in relation to the area that is contributing, to the population that is contributing, and the GDP of each country."
This year, Sweden has set up 10 new stations to augment the monitoring network.
And some observers stress that calculations of the pollutant load are not based on water monitoring data only. For example, populations connected to waste water treatment plants are monitored, and leakage from agricultural land is modelled.
The unmonitored areas are taken into account when Swedish nutrient loads to the Baltic Sea are estimated. The data gaps are bridged with the help of computer modelling.
However, the authors of the latest study point out that because these results cannot be checked against real data for the unmonitored areas, they could be significantly wrong.