The approach taken will have non-scientific and scientific components.

First step is to identify and deal with pollutant sources. The sources could be non-point (farm runoff or urban storm water) or point (e.g. municipal sewage or industrial).

The next step is to identify all stakeholders (which will include polluters, funders of the project, interested parties) and form a group to deal with the issue. If there is no interest from the stakeholders there may be regulation introduced by the government, which could trigger sufficient interest.

Apart from devising ways to minimise pollutant inputs (it may not be possible to totally avoid pollutant inputs) there may be mitigation methods to deal with the remaining pollutants in the waterway. As an example, tree/shrub shades can be provided to reduce water temperature which should minimise algal blooms even if the nutrient levels are high in water.

Such restoration projects are already active in New Zealand (e.g. Lakes Taupo and Rotorua, Waikato River). As a first step all point source of pollution have been either dealt with (please refer my publication on consent process to deal with wastewater discharges in the Otago Region) or are being deal with. Recently, central government has introduced a national policy to deal with freshwater quality. The government has also stepped in to providing restoration funding (e.g. NZ$80 for Lake Taupo) which is being used in part to retire agricultural land into native bush land to reduce nutrient input to the lake. Local government in the lake catchment has been regulating nutrient input to the lake.

In short, while there is sufficient scientific information exist on waterway restoration there should be a community will to achieve the desire environmental outcome.

Environmental laws should be very stringent.  The industrial and domestic discharges without treatment should be stopped first. The biological organisms should be introduced again for restoration. The biological filtration by gastropods will be more efficient for the aquatic system for long term maintenance of the ecosystem. Many aquatic plants can also be planted for proper functioning of the system. Artificial reefs are excellent structures for the restoration of fouling organisms. 

Good point raised by colleague Kenneth. The term natural state is subjective, i.e. subject to community's realistic expectation of the state of the waterway. During the value-setting-consultation process stakeholders have to agree on the type of values they can collectively achieve and maintain. Values (e.g. freshwater fish, swimming, aesthetics) chosen are a product of a balance struck between a range of waterway users such as recreational users (including people who aspire aesthetics), people who discharge into it and funding bodies (e.g. government). When work together the group will realise that setting high values will stifle the discharges activities or rendering the project practically unachievable and unaffordable. Equally, aiming for low denominations will lose interest with wider waterway users resulting in immobilisation of any action.

In New Zealand we are lucky in that we have several alpine lakes and rivers that already have excellent water quality and are virtually in pristine status. In such cases, the values are set at very high levels to maintain in the current status. For example E.Coli levels are expected to be below 10 cfu (colony forming units) in the so called pristine waterways. On the hand, in lowland areas (where farming is predominant) the values are set to sustain farming and basic values (e.g. E.Coli levels are set between 250 and 550 cfu, which will still meet the health guideline for swimming).

Thank you so much for all of your active conversation in this regard; hopefully back sooner after go through some of the materials you provided.

Very interesting answers.

Restoration is a human-nature issue, so stakeholders need to be the main actors to improve the state of the lake. It is important to get agreements on the goals of restoration to achieve and the tasks for every social actor. There are evidence about the water bodies recovering some physicochemical features in a few years but not the biological communities. There are different rythms of recovery for the different aspects involved in the process. Then, it can be useful to set a group of indicators or variables to follow (biophysical and social ones) and keep tracking them along a period of time.

Always consider upstream influences and sources of pollution, watershed scale is best.  Also consider what is within the potential of the stream. 

Some sources:

http://www.amazon.com/Restoration-Aquatic-Ecosystems-Science-Technology/dp/0309045347/ref=sr_1_1?s=books&ie=UTF8&qid=1443307469&sr=1-1&keywords=national+research+council+aquatic+restoration

Beechie, T. J., et al. (2010). "Process-Based Principles for Restoring River Ecosystems." BioScience 60(3): 209-222.

Bernhardt, E. S. and M. A. Palmer (2011). "River restoration: the fuzzy logic of repairing reaches to reverse catchment scale degradation." Ecological Applications 21(6): 1926-1931.

Palmer, M. A., et al. (2005). "Standards for ecologically successful river restoration." Journal of Applied Ecology 42: 208-217.

Roni, P. and T. Beechie (2012). Stream and Watershed Restoration : A Guide to Restoring Riverine Processes and Habitats. Hoboken, Wiley.

The following methodologies are effective for aquatic body restoration:

• Preserve and protect aquatic resources: Existing, relatively intact ecosystems are the keystone for conserving biodiversity, and provide the biota and other natural materials needed for the recovery of impaired systems.
• Restore ecological integrity: Ecological integrity refers to the condition   of an ecosystem - particularly the structure, composition, and natural processes of its biotic communities and physical environment.
• Restore natural structure: Many aquatic resources in need of restoration have problems that originated with harmful alteration of channel form or other physical characteristics, which in turn may have led to problems such as habitat degradation, changes in flow regimes, and siltation.
• Restore natural function: Structure and function are closely linked in river corridors, lakes, wetlands, estuaries and other aquatic resources. Reestablishing the appropriate natural structure can bring back beneficial functions.
• Work within the watershed and broader landscape context: Restoration requires a design based on the entire watershed, not just the part of the water body that may be the most degraded site. Activities throughout the watershed can have adverse effects on the aquatic resource that is being restored. By considering the watershed context in this case, restoration planners may be able to design a project for the desired benefits of restoration, while also withstanding or even helping to remediate the effects of adjacent land uses on runoff and non-point source pollution.
• Understand the natural potential of the watershed: Restoration planning should take into account any irreversible changes in the watershed that may affect the system being restored, and focus on restoring its retpaining natural potential.
• Address ongoing causes of degradation: Identify the causes of degradation and eliminate or remediate ongoing stresses wherever possible.
• Develop clear, achievable, and measurable goals: Goals direct implementation and provide the standards for measuring success. The chosen goals should be achievable ecologically, given the natural potential of the area, and socio-economically, given the available resources and the extent of community support for the project.
• Focus on feasibility taking into account scientific, financial, social and other considerations.
• Anticipate future changes: As the environment and our communities are both dynamic, many foreseeable ecological and societal changes can and should be factored into restoration design.