Q&A:

The Braided Side of the Earth: modelling the long-term morphological impact of dams on the gravel-bed braided rivers of New Zealand to support restoration of the heavily-impacted European rivers. (For short: "BraidSideEarth")

... what's that?

BraidSideEarth is a three-year project that aims at improving our capability of predicting and quantifying the impact of dam construction and operation on highly dynamic, multi-thread river systems through the development and application of a novel physics-based numerical model.

What is a highly-dynamic, multi-thread river system?

In alpine and piedmont regions, gravel-bed rivers naturally tend to assume morphological configurations characterised by a wide planform, where the flow splits into a multiple-channel pattern - hence the definition "braided river", as opposed to single-thread or "meandering". Braided morphologies are generally subject to rapid evolution and reworking of their channel pattern due to a multiplicity of processes. Planform evolution mainly happens under flood condition, when abundant sediment transport drives morphological change which reshapes the channel network.

Switch to this page to have a look at two healthy braided rivers, or have a look at the two pictures below, which show the braided planforms of (left) the Waimakariri River (New Zealand) and (right) of the Tagliamento River (Italy) .

Are all gravel-bed rivers on the Earth naturally braided?

No. Gravel bed rivers assume a braided configuration where the conditions needed to sustain braiding exist. These conditions are essentially availability of high-energy flows and high sediment supply. For this reason gravel-bed braided rivers are generally found in alpine and piedmont areas, where high slopes determine the required flow conditions and sediment production and delivery to the river.

Notably, rivers can assume different planform configurations in different portions of their course, depending on the changing conditions which they encounter. For example, a river can be braided in its high and middle course, and then switch to meandering in the lowlands, when the gradient is not sufficient to sustain braiding .

OK. But ... if a river reach is braided, will it stay braided forever?

Not necessarily! As the conditions and controls which determine the river's planform shape can change in time, both due to natural and anthropogenic changes, then the river shape in a single reach can change in time quite significantly.

In the last 150 years, most of (formerly) braided reaches in Europe have experienced, at least partially, a transition to single-thread configurations. Overall, adjustments were very significant and determined loss of most of braided reaches in Europe. The general pattern of changes identified for Italian rivers comprises four phases: i) a first period (19th century) characterized by very small changes; ii) an phase (1870-1950) of river narrowing without significant incision essentially driven by landuse change, iii) a phase of dramatic adjustments (more rapid narrowing accompanied by incision) driven by sediment depletion induced by gravel mining, and iv) a final phase (1990-present) of widening after abandonment of in-stream mining.

So... what are the management causes which drove observed changes in configurations in Europe?

Many!

Sediment delivery to the main course of rivers was reduced by landuse change and catchment reforestation (for instance following abandonment of mountain agriculture) and by torrent control works on tributaries (check dams).

Rivers were strenghtened and meanders were cut off to reclaim land and favour road construction. Bank protection measures isolated the flow from sources of gravel on the floodplain. In-stream gravel mining reduced the sediment available for transport and forced rivers to scour the main channels. Spread of vegetation on the floodplain was a consequence of flow concentration in a few main channels, and induced further flow concentration, bank strenghtening, and in-channel degradation.

Dams have a prominent role in promoting these alterations. The river's bedload gets trapped in the dam reservoir and cannot be delivered to the downstream reach. Dams also alter the flow regime (reducing the frequency and magnitude of floods), which further favours vegetation spread across the river, because floods are no longer competent to clear vegetation. The construction of a large number of hydropower dams and reservoirs in the European alpine rivers has been a key driver of overall observed transition of river configurations from braiding to transitional and single-thread.

What are the morphological changes induced by dams?

Bed degradation and channel incision; narrowing of the active planform; reduction of braiding complexity and transition towards single-thread morphology. Jump to this page for an example.

Wait. Why does all of this matter? Isn't a river "just a river", regardless of the shape it takes?

Well, the shape of the river does matter. Braided rivers possess a unique fluvial landscape, able to fulfil valuable morpho-ecological functions and provide ecological services. Impacted rivers which have experienced severe morphological changes are, at least in some degree, incapable of providing the same services.

Notice that the European Water Framework Directive (WFD, 2000/60/EEC), along with the Flood Risk Directive (2007/60/EEC) and the Habitats Directive (1992/43/EEC), mandate the restoration and maintenance of healthy aquatic and riparian ecosystems, where “health” is assessed on the basis of hydro-morphological, chemical and biological attributes.

So, what can we do about that?

We cannot obviously think of restoring natural and healthy rivers by restoring the state of the beginning of the 19th century.

BUT

We can at least try and understand in advance what would happen if we could modify some controls and constraints which are currently in place on a river. For instance, we can investigate how the river would respond if more an environmentally-targeted operational rule for dam releases was adopted. Such an understanding could inform targeted restoration process when and where these are allowed by the local conditions.

What is our understanding of the complete picture?

Still pretty limited. Rivers are complex systems. Their evolution is driven by multiple (hydraulic, morphodynamic, biological) processes which interact with each other. Disentangling all of this is not straighforward. We do have some conceptual understanding of "what happens if ...", but forecasting "how much will a river change in one century's time if one control parameter value is modified" is beyond the capability of presently available science.

So? How can we make forecasts?

Well, that's where the present project comes into play. We can use physics-based numerical models. These models are schematised renderings of the physics of rivers through equations, which are solved through computer programs. Models can be used to try and work out how the river shape could change if we alter the controls.

Sounds good. Does a model capable of representing all the relevant processes exist?

Aha, you wish! Almost.

In recent years lots of work has been done on modelling 2D hydro-morphodynamics of braided rivers and river morphodynamics driven by vegetation processes. Designing a model which can account for all the relevant processes is one of the objectives of the present project, and is being done within this project. Just go to the Modelling page of this website for a practical example of our model's capabilities.

Is that all?

Not really. Model development must be informed by observation and applications to real rivers and laboratory prototypes. Getting feedbacks from real rivers is essential when designing such a complex model. In detail, the work can benefit from the observation of weakly-impacted braided rivers. These are mainly located outside Europe, because, as said previously, anthropogenic pressure on European rivers has been very high in the last two centuries.

The South Island of New Zealand, with its alluvial planes close to steep alpine mountain ranges, frequent and flashy floods, high sediment delivery to rivers due to poorly consolidated rock type, earthquakes and landslides, and very low anthropic pressure due to very low population density, is the ideal setting to observe minimally impacted braided rivers. Then, the model developed and calibrated using these observations can be used to forecast the behaviour of heavily-impacted braided rivers in Europe under different and hypothetical management scenarios.

Sounds like a fun project. Just one more question. Who pays the bill?

The Research Executive Agency of the European Commission, through a funding scheme called "Marie Curie Outgoing Fellowship for Career Development". I wish to acknowledge financial support of Marie Curie funding, which made this project come true.

The project is managed and administrated at the University of Trento, Italy, which serves as Return Host, where the local Research Group in River Morphodynamics and Environmental Hydraulics has a wealth of experience on braided rivers and numerical models. During the first two years, the project is carried out at the National Institute of Water and Atmospheric Research (NIWA), Christchurch, New Zealand, which serves as Outgoing Host. The Sediment Processes Group has a unique knowledge of the New Zealand river environment and great experience in numerical modelling of braided rivers. This represents for me (the Fellow) a once-in-a-lifetime chance for learning.