Introduction
Two water-quality issues with major impact on ecosystem health are nutrient pollution and toxic stress by chemicals.
Nutrient pollution occurs when fertilizers, primarily nitrogen and phosphorus, used in food production, enter soils, groundwater and surface water and are transported towards coastal seas. This results in a range of environmental problems, from groundwater pollution, loss of habitat and biodiversity, creation of coastal dead zones, harmful algal blooms, fish kills and human health impacts.1
Currently, more than 350,000 chemicals and mixtures of them have been registered for production and use.2 And, as a result of their use, many of these chemicals find their way to freshwater systems3 and coastal waters.4 There they may accumulate and negatively affect the aquatic ecosystem.
Nutrient pollution
Impact/State
In freshwater, algal blooms are often dominated by cyanobacteria that may generate toxins, rendering the water unsuitable for drinking, irrigation, bathing or swimming. Also, increased growth of algae may result in oxygen depletion and even hypoxia in the water body after the decay of the algal biomass. This, in turn, may lead to bad smells affecting local tourism, as well as to massive fish kills affecting local fisheries.5
Globally, lakes constitute an important source of water, food and recreation. However, increasing water pollution threatens the ability of lakes to provide these and other ecosystem services.6 Lake eutrophication, where nutrient pollution has caused an overgrowth of plants that depletes oxygen, is a global environmental issue that poses a survival risk to aquatic organisms, affecting fisheries and aquaculture. Alarmingly, eutrophication already is a worldwide phenomenon, with rapidly declining aquatic biodiversity.7 One of the symptoms of eutrophication and biodiversity loss is illustrated in Figure 3.1 showing remote-sensing observations for 450 lakes with high cyanobacteria dominance, now occurring across all continents. Trend analyses indicate increases over time, and in several cases even regime shifts, in many lakes.
The global map of the trophic state index (Figure 3.2, top) shows that satellites are able to identify freshwater systems in different stages of eutrophication. Such information can be used in combination with data on phosphorus accumulation in lake and reservoir sediments (Figure 3.2, middle), as an indicator for the potential occurrence of cyanobacteria blooms, and the level of original freshwater biodiversity (Figure 3.2, bottom), i.e. the species loss due to human interferences in aquatic ecosystems, including nutrient loading differentiated by sources and dam construction in rivers.
Drivers/pressures
Eutrophication is, in part, a natural-driven process, but it is greatly increased by anthropogenic nutrient input from agricultural activities or untreated discharge of wastewater.
In the Earth's system, nutrient cycles have intensified dramatically in the past 50 years, with global nitrogen up 75 per cent and phosphorus up 92 per cent between 1970 and 2020. At the same time, the world's population increased by three billion people. As a result, protein and phosphorus consumption and excretion also increased significantly, which reflects growth in population, meat and dairy consumption, and incomes. However, nutrient flows related to food consumption are minor when compared to those from agriculture. It is also clear that the inputs of nutrients in natural ecosystems is relatively stable or slowly declining, and that anthropogenic inputs into the earth system are now dominating.
Globally, delivery to inland water bodies of nitrogen and phosphorus increased rapidly between 1970 and 2020. Natural sources declined slightly, while anthropogenic sources increased from 58 per cent in 1970 to 74 per cent of the total delivery in 2020, implying that there has been an immense intensification of the societal nutrient usage and discharge. It is clear agriculture is now the most important source, contributing 52 per cent in 2020, primarily due to increasing food production.
Around the world, about 40 per cent of the total population is connected to a sewage system, and wastewater treatment installations remove 26 per cent of the emissions from connected households. The remaining nitrogen and phosphorus in the untreated wastewater plus effluents after treatment contribute 15-17 per cent to total nutrient delivery.
Aquaculture is a minor source at the global scale, but particularly in Southeast Asia it is becoming a locally significant source of nutrients.8 With the rapid increase of anthropogenic sources, the relative contribution of natural sources has been decreasing. Natural sources contributed to 37 per cent (for phosphorus) and 42 per cent (for nitrogen) to total nutrient delivery in 1970, and in 2020 this contribution shrank to 27-28 per cent.
Currently global rivers carry 49 teragrams/year-1 of nitrogen and 5 teragrams/year-1 of phosphorus to coastal waters. This is a 60 per cent increase for nitrogen since 1970, and a 31 per cent increase for phosphorus. Natural sources contributed 40-44 per cent in 1970 and 27-30 per cent in 2020 to river export. Sewage contributed 10 per cent to total phosphorus delivery in 1970, and 17 per cent in 2020, while the global contribution of sewage to river export was 15 per cent and 23 per cent, respectively, for 1970 and 2020. This means that while there has been a rapid increase in total nitrogen and phosphorus export, the mix of sources also changes with anthropogenic sources gradually becoming dominant and point sources increasingly important in river export.
In addition to nutrient loading per se, the coastal research community has become increasingly aware that the ratio of nitrogen to phosphorus is essential. Disruption of the Redfield molar ratio of 16:1 nitrogen to phosphorus is one of the main causes of harmful algal bloom proliferation, even in a situation of declining nutrient loads. Globally, this ratio has been gradually increasing in the water drained by rivers to the oceans between 1970 and 2020, from values of around 18 to close to 22. For surface waters, this ratio has also increased but at a lower level of 14 to 17. Under such conditions harmful algal blooms often increase in frequency, area and toxicity.9
Data collected in China's coastal waters10 combined with river phosphorus loading data shown indicate that in those waters there is a threshold of 25 for harmful algal bloom proliferation. Since the ratio has exceeded this value from about 1980 onwards, harmful algal bloom has increased in both frequency and area. This problem is rapidly expanding, with global rivers exporting three times the amount of water where the ratio of nitrogen to phosphorus is greater than 25. This suggests that the changes in the nutrient cocktail in river flows to coastal waters may have been one of the major causes of the proliferation of harmful algal blooms observed in recent decades.11
Response
Eutrophication can be mitigated by controlling the nutrient cycles, by improving the resilience of aquatic ecosystems, or reducing perturbations to them. Since agriculture is the main source of nutrients in water bodies, a good place to begin would be reducing the nitrogen and phosphorus cycles in the food production system, i.e. by decreasing inefficient feed-livestock production processes and by cutting food waste.12 Although this is one of the most effective ways to reduce eutrophication, it depends on human behaviour as it is directly related to our diets.13 Furthermore, the food production system could be more efficient in its use of nutrients. This may mean changing nutrient management systems, and tuning inputs to the needs of plants and animals.14
For example, several directives by the European Commission15 have reduced nutrient emissions from agriculture and sewage. As a result, water quality of the Rhine River improved. Since the 1970s, annual average total nitrogen concentrations on the German-Dutch border has declined to close to the EU standard (2.5 mg/L)16 and total phosphorus concentrations to around 0.1 mg/L. However, large parts of European rivers still have "less than good quality" with respect to phosphorus concentrations, clearly affected by diverse sources and the phosphorus legacy effect.17
Unwanted effects of measures such as in the European Union is the increasing ratio of nitrogen to phosphorus, as observed in the Rhine and many European rivers,18 and rivers and lakes in China.19 This calls for a balanced management of both nitrogen and phosphorus from the diverse nutrient sources in river basins, including agriculture, sewage and industry.
Several on-site solutions are available, aimed at lowering local nutrient availability or increasing the resilience of rivers, lakes and coastal waters. For example, this is illustrated by restoring the nutrient buffering capacity of the natural embankment and wetlands around lakes as demonstrated for Lake Taihu in China.20 Re-oligotrophication (lake restoration) may be a costly strategy, and even where vast reductions in nutrient loading are achieved, lakes do not respond as expected. Reducing nutrient loads may lead to less algal production, organic matter sedimentation, less oxygen depletion and, therefore, less de-nitrification. This may lead to increasing nitrate concentrations and rising nitrogen to phosphorus ratios, as observed in a series of large global lakes.21
In order to develop properly balanced response strategies to reduce nutrient pollution that account for the diversity of nutrient sources and transport pathways, the triangulation approach needs to be further explored. This integrates in-situ monitoring of water quality, monitoring impacts using remote sensing, and modelling. Scenarios can be used to drive these models to assess the level of mitigation that is needed to achieve improvement of water quality in a given timeframe. Analysis of future scenarios can help us to compare the effectiveness of the various actions and strategies in future situations, accounting for land-use change, population growth, climate change and human interference in the earth system.
Toxic stress
Impact/State
Effects from chemicals on aquatic species are commonly estimated by comparing the concentrations of chemicals in surface waters to thresholds derived from indicator organisms in laboratory tests on algae, macro-invertebrates and fish species. Projections for acute effects in the field are made on the basis of short-term laboratory tests for lethal endpoints (e.g. mortality). Projections for chronic effects in the field are made on the basis of longer-term laboratory tests for non-lethal endpoints like reproduction or growth. It is furthermore assumed that effects on individual species are likely to cause effects on aquatic ecosystems, such as losses of biodiversity or species shifts.
An assessment based on measured environmental concentrations in more than 10,000 European water bodies22 concluded that chronic effects on aquatic species were expected at 42 per cent of the studied sites and acute lethal effects at 14 per cent. The authors noted that, although these are already serious numbers, they are likely to underestimate the actual risks because of the limited number of chemicals being measured.
A similar assessment was reported23 based on predicted environmental concentrations, generated by state-of-the-art Europe-wide mathematical modelling.24 The expected effects on aquatic species were estimated for a mixture of 1,785 chemicals in 10,658 water bodies across Europe. Results indicated that 79-85 per cent of European waterbodies are expected to experience chronic effects while 16-43 per cent are expected to experience some degree of species loss.
But these kinds of assessments cannot be made on a global scale, because insufficient data is available. A tentative assessment for other continents has been proposed that extrapolates the results obtained for Europe on the basis of a simple proxy called Human Impact and Water Availability Indicator (HIWAI).
This proxy combines information about population density, economic activity, river dilution capacity and downstream transport and can be easily calculated using a hydrological model. Currently this work is being extended to the global scale.
Drivers/pressures
Chemicals are used for a reason. Obvious drivers are the increasing demand for food, leading to increased use of pesticides and veterinary drugs, as well as the demand for improved health and public well-being, leading to increased use of pharmaceuticals. In general, increasing economic development leads to an increasing use of a wide range of chemicals in industry, building and construction, and in the outdoor and indoor environments. Different types of uses in combination with the properties of the chemicals in question lead to variable fractions of the volume of chemicals used leaking to the environment (air, wastewater, surface waters, soils). On average, this leakage fraction may amount to 10-20 per cent.25
In 1965, the CAS Chemical Registry System was introduced, which issued unique CAS Registry Number® to identify chemical substances without ambiguity. In 2009, this system made its 50 millionth substance registration, while that number had doubled to 100 million in 2015.26
Global chemical sales (excluding pharmaceuticals) are projected to grow from €3.47 trillion in 2017 to €6.6 trillion by 2030.27 Asia is expected to account for almost 70 per cent of sales by then.
It is challenging to obtain a quantitative and complete overview of the resulting local pressures. Model-based approaches may partly help out.28 The quantification of contaminant emissions may be provided by various methods, e.g. by Wastewater Treatment Plant pathway modelling, by modelling of the terrestrial run-off and erosion pathway, or by inverse modelling. However, none of these approaches can easily address hundreds or thousands of chemicals on a continental or global scale. This limits the possibilities to manage chemicals effectively in aquatic ecosystems.
Response
Responses to the presence of hazardous chemicals can be arranged at two levels. The most effective is prevention of the use of chemicals presenting unacceptable risks to humans and the environment. This is particularly relevant for persistent and mobile toxic chemicals as these are not removed by natural processes and may spread over large distances. This requires an effective chemicals 'admission to market' legislative system to be in place and to be enforced. A challenge is to determine if a chemical is presenting unacceptable risks to humans and the environment, in a way that can withstand the economic pressure of bringing new chemicals to the market. Reactive responses by mitigation or remediation measures are also possible, but not always effective. Relatively cheap is the (additional) treatment of already intercepted wastewater flows in treatment plants. Chemical leakages that reach the environment in a more diffuse way are much harder to abate. In such cases, only good application practices (e.g. for pesticides, paints) may offer some degree of reduction of leakages to the environment.
Data and knowledge gaps
This assessment shows that there still is a strong need for better and more regularly monitored data on nutrient pollution:
• Long-term monitoring data on nutrient concentrations at various locations within river basins is scant and is limiting our ability to validate models at different scales. Data for some of the largest rivers in the world is not available at all.
• The contributions of sewage, agriculture, aquaculture and natural sources to nutrient loading of global river basins is uncertain. This is related to data limitations but also knowledge gaps on the importance of the various transport pathways, the biogeochemical processing in groundwater systems, riparian zones and in streams, lakes and reservoirs.
• Nutrient legacies from past management has been shown to have an important contribution to current river nitrate and phosphate loading; however, the magnitude of these nutrient fluxes is uncertain. Reducing this uncertainty requires specialized experiments in combination with models.
Concerning toxic stress, effective management of chemical pollution is hampered by both knowledge and data gaps:
• Data to quantify chemicals use volumes and use types are sparse at best (e.g. Europe, US) and often non-existent.
• The knowledge to set up effective legislative frameworks to deal with chemicals is lacking. In some parts of the world, substance authorization procedures are in place that aim at eliminating non-desirable chemicals even before they are used. Such procedures have proven extremely useful. It is difficult, however, to incorporate chemical mixtures in such procedures.
• Regulations to protect the environment and human health also have difficulties with mixtures, i.e. cumulative effects. As there are simply too many substances and degradation products, too many protection targets, and too many modes of action for chemicals to affect humans and the environment, approaches that single out any of the above will never provide holistic solutions. There is at present no ready-to-use solution.
• In Europe, a recent "Green Deal" call for research projects aims at providing a sound knowledge base for innovative regulatory solutions. Elsewhere, there are definitely no-regret actions to be taken. This includes awareness raising among policy makers worldwide, investments in sanitation while making sure that collected wastewater is not discharged untreated, and initiating substances authorization procedures world-wide in order to avoid a situation that hazardous substances banned in one place find their way to other parts of the world.
• Adapted from World Water Quality Alliance (2021). World Water Quality Assessment:
First Global Display of a Water Quality Baseline.