Jesper Andersen:
Ecosystem-Based Management of Coastal Eutrophication - Connecting Science, Policy and Society

Date: 19-03-2012    Supervisor: Jens Borum

This thesis concerns eutrophication and the management of human activities resulting in nutrient enrichment and the biological effects on aquatic ecosystems.

The term ‘eutrophication’ (noun) has its root in two Greek words: ‘eu’ which means ‘well’ and ‘trope’ which means ’nourishment’. The modern use of the word eutrophication is related to high inputs and effects of inorganic nutrients in ecosystems, especially over-enrichment of aquatic ecosystems.

Management is basically the process of getting people together to accomplish desired goals and objectives. The verb ‘manage’ comes from the Italian ‘maneggiare’ (to handle), which originally derives from the Latin ‘manus’ (hand). In the context of eutrophication, management is about setting up a strategy for control of human activities resulting in discharges (direct sources), losses (diffuse sources, e.g. from agriculture) and emissions (to the atmosphere) of nitrogen, phosphorus and organic matter to the aquatic environment.

An adaptive nutrient management strategy (NMS) should include the following elements: Problem identification and four phases focusing on planning, acting, checking and evaluation. The papers on which this thesis is based upon addresses all of these five elements.

Understanding the context of eutrophication is important both from a scientific point of view, since both definitions and conceptual understanding are constantly developing, and from an implementation of nutrient management strategies. If decision-makers are not informed or do not understand the concept of eutrophication then management is a difficult task. Despite a widespread common European understanding of causes and effects of eutrophication, there is no mutually agreed definition of coastal eutrophication. However, within the European Union (EU) there has been a sound tradition of focusing the measures on the sources causing eutrophication (Elliot et al. 1999, Elliot & de Jonge 2002). Consequently, eutrophication has been defined in relation to sources and/or sectors. For example, in the EU Urban Waste Water Treatment Directive, eutrophication has been defined as “the enrichment of water by nutrients, especially nitrogen and/or phosphorus, causing an accelerated growth of algae and higher forms of plant life to produce an undesirable disturbance to the balance of organisms present in the water and to the quality of water concerned” (Anon. 1991a). The EU Nitrates Directive has an almost identical definition specifically emphasising losses of nitrates from agriculture (Anon. 1991b). Nixon (1995) defined eutrophication as “an increase in the rate of supply of organic matter to an ecosystem”. This definition is short and emphasizes that eutrophication is a process, not a trophic state. Nixon also noted that various factors may increase the supply of organic matter to coastal systems, but the most common is clearly nutrient enrichment. The supply of organic matter to an ecosystem is not restricted to pelagic primary production, even though such an interpretation makes the definition operational. The supply of organic matter to a system includes primary production of higher plants and benthic microalgae as well as inputs of organic matter from adjacent waters or from land via rivers or point sources. Having such a broad interpretation of the term ‘supply’ makes the definition difficult to use in a monitoring and management context.

Eutrophication and definition(s) of eutrophication are widely discussed (Jørgensen & Richardson 1996). The most common use of the term is related to inputs of mineral nutrients, in particular nitrogen and phosphorus, to specific waters. Consequently, eutrophication deals with both the process and the associated effects of nutrient enrichment and natural versus cultural eutrophication. Despite the definitions in existing European directives, the implementation of the EU Water Framework Directive (WFD) revealed a need for a common understanding and definition of eutrophication as well as stronger co-ordination between directives dealing directly or indirectly with eutrophication. Hence, the European Commission convened a process aiming for a development of a pan-European conceptual framework for eutrophication assessment in the context of all European waters and policies (Anon. 2009a). This process did not lead to a common European definition of eutrophication, but it revealed that if ‘undesirable disturbance’ is understood as ‘unacceptable deviation from reference conditions’, the pan-European definition will be coherent with the normative definitions sensu the WFD (Andersen et al. 2006). Accepting this, a pan-European definition of eutrophication, would be: “the enrichment of water by nutrients, especially nitrogen and/or phosphorus, and organic matter, causing an increased growth of algae and/or higher forms of plant life to produce an unacceptable deviation in structure, function and stability of organisms present in the water and to the quality of water concerned, compared to reference conditions”.

The suggested definition includes causative factors (nutrient enrichment), primary effects (increased growth) and secondary effects (sometime referred to as ‘undesirable disturbance’). However, it also is a matter of interpretation, in particular in regard to what an ‘acceptable deviation’ is.

In addition, the definition enables classification of ‘eutrophication status’. Using the definition as a basic assessment principle, an eutrophication quality objective or target (EutroQO) is defined as an indicator with an acceptable deviation (AcDev) from the reference condition (RefCon), EutroQO = RefCon ± AcDev (Andersen et al. 2004, Andersen et al. 2011). As an additional feature, the definition also acknowledges that eutrophication has both quantitative and qualitative perspectives, an aspect not included in Nixon’s definition.

The setting of science-based eutrophication quality objectives (or targets) is a prerequisite for ecosystem-based management. These target setting principles used in Europe are commonly based of information on RefCon and setting of an AcDec from RefCon. The concept originates from the Water Framework Directives and is described and demonstrated by Andersen et al. (2004). The strength of the concept is that it is operational and that the data used derived by science-based process. The weakness is that is allows for expert judgement, e.g. in regard to the setting of AcDev, and thus potentially a weakening of the scientific basis. A specific problem is related to natural variability and it potential influence of the assessment of eutrophication status (Andersen et al. 2004).

Before implementing a nutrient management strategy, having a complete overview of the human activities and pressures is necessary to focus on the activities resulting in impaired conditions. An example can be found in Korpinen et al. (2012), where cumulative pressures and impacts in the Baltic Sea region have been estimated. The estimate is based on the mapping of human activities, maps of key ecosystem components and expert judgement of the impact of a specific pressure upon a specific ecosystem component. A matrix is established and from it, the dominant pressures in the Baltic Sea were estimated to be: (1) nutrient enrichment, (2) fishing activity, (3) input of contaminants, and (4) physical modification. This study is the first ever assessment of cumulative pressures and impacts for a regional sea, and is a useful tool for documenting the causes of impaired conditions as well as targeting of measures, regionally and sub-regionally.

The targets of nutrient management can in principle be established in two ways: (1) the traditional way where load reduction targets are agreed upon, and (2) a more modern way where Eutrophication Quality Objectives (EutroQO’s) are established and the critical loads matching the EutroQO’s are calculated. Two different case studies are analysed in this thesis. The Danish Action Plans on the Aquatic Environment have a strong focus on load reduction targets for agricultural discharges and losses as well as discharges from urban water treatment plants and industries with separate discharge (Conley et al. 2003, Carstensen et al. 2009, Andersen & Conley 2009). The HELCOM Baltic Sea Action Plan, which is based on an ecological target and subsequent calculation of a critical load, represent a more evidence-based way to estimate the load reductions (Andersen et al. 2011).

A key step in any nutrient management strategy is monitoring for expected improvements in ecological quality, specifically eutrophication status, in the marine environment. The Danish Action Plans on the Aquatic Environment included a well-designed monitoring programme for Danish marine waters (Conley et al. 2002, Ærtebjerg et al. 2003). The data and information originating from monitoring activities have not only resulted in annually national reports, which have been used for regular evaluations of the effectiveness of the Danish Action Plan, but also in papers of eutrophication trends (Carstensen et al. 2006). Both the reductions in loads and the effects of the loads reductions in Danish coastal waters are well documented: (1) inputs have decreased significantly, both for nitrogen and phosphorus, (2) nutrient concentrations have decreased significantly, (3) primary productivity and phytoplankton biomass have decreased as well, and (4) benthic communities have in some areas improved their ecological status.

The work on assessing eutrophication in Danish marine waters and the Baltic Sea has lead to important advances in our understanding. For decades, eutrophication assessments have focused on state for a given indicator supplemented by temporal trend assessment for individual indicators. Recently, multi-metric indicator-based assessment tools are emerging (Andersen 2010, Andersen et al. 2011). With the development of the HELCOM Eutrophication Assessment Tool (HEAT) (Andersen et al. 2011), status assessments can now be supplemented with a simple estimate of confidence (Andersen et al. 2010). The approach to solve a statistical challenge in a non-statistical way is based on expert judgement of the confidence of information in regard to RefCon, AcDec and observations of the state. This information is combined for each indicator and integrated into an overall estimate of confidence. Information in regard to confidence estimates is useful for setting up evidence-based nutrient management strategies, but also essential when redesigning monitoring programmes.

Based on the lessons from the Danish Action Plans on the Aquatic Environment and the HELCOM Baltic Sea Action Plan, the following DO’s and DON’T’s of evidence-based nutrient management strategies can be made:

- DO understand that ecosystem-based management is adaptive and science-based.
- DON’T assume that decisions can not be taken because of incomplete knowledge and uncertainty.

- DO evidence-based target setting and exhaustive planning, the latter involving decisionmakers, authorities and all stakeholders.
- DON’T wait for perfection and all-inclusive ecosystem understanding.

- DO a full execution of the plan.
- DON’T rely on voluntary agreements or guidelines.

- DO monitoring with ecologically relevant resolution in time and space.
- DON’T underestimate resources needed for sampling, quality assurance, analysing data and reporting.

- DO regular evaluations in regard to the progress of the nutrient management strategy.
- DON’T disregard the advantages of a dual monitoring strategy focusing on both nutrient inputs as well as ecological responses to lowered nutrient inputs.

An important lesson learned from the Danish Action Plans on the Aquatic Environment and the HELCOM Baltic Sea Action Plan is that decisions are often made in short windows of opportunity. It is critical to prepare for those brief moments where decisions and actions can be taken. Preparation of decision support systems and determining the best possible scientific basis for decision-making can provide the scientific basis for actions to be implemented.

Perhaps the most important lesson is that time is needed before the effects of changes in human behaviour can be observed in nutrient inputs and, eventually, in the ecological quality of the marine environment. It would, therefore, be prudent to ask if we, within a decade or two, can expect to have a marine environment not affected by eutrophication as required by national and international processes, e.g. the Danish Action Plan on the Aquatic Environment, the EU Water Framework Directive, the EU Marine Strategy Directive and the HELCOM Baltic Sea Action Plan. There are a number of factors underlying the slow response of ecosystems, e.g. delays in nutrient inputs from fields to streams and rivers caused by retention in the soil. There is a growing recognition that the recovery trajectories differ from the well known degradation trajectories (Duarte et al. 2009; Laamanen et al. submitted). Another challenge is a shifting baseline caused by increasing temperatures, resulting in a situation where loads of nutrients probably have to be reduced more than estimated in a situation with stable temperatures (Laamanen et al. submitted). Apparently, we face two counteracting process, one where nutrient loads are progressively reduced, and one where sea temperatures are rising. The prospects in regard to the long-lasting eutrophication crisis are not good due to the lack of political will to act, a fact being sustained by the current financial crisis.