DR. GEOFFREY PETTS: In-stream Flow Science for Sustainable River Management

The plenary lecture was given by Dr. Geoffrey Petts, Vice-Chancellor of the University of Westminster, a position he has held since 2007. His research is at the interface of hydrology, geomorphology and ecology to address the sustainable development of water projects. He has published 20 books and more than 100 scientific papers, and is founder and Editor-in-Chief of the international journal River Research and Applications.

Dr. Petts began his presentation with a thought: “Are we really concerned with rivers flowing to the Delta, or are we talking about river flows for the Delta? The starting point is very important, because I think if you were starting with the question, what are the river flows that you need to sustain the functional development of the Delta and the transition zone, that’s a very different starting point and end point perhaps than talking about the functional flows in the regulated river.”

In Britain, one of the areas they are struggling with is that their estuary, their Delta, has been owned by no one and ignored by everyone for probably the last 30 to 40 years. “There has been a lot of focus on the catchment and a lot of focus on the deep ocean, but nobody has been looking at the estuary,” he said. “We now believe many of our environmental problems associated with rivers in the UK is in fact the legacy pollution and bad management in the past in the estuary. It’s a fundamental change in the way we’re looking at our river systems, so for me, it’s an interesting question. Is it river flows for the Delta or river flows to the Delta? Basically it’s to the Delta; the Delta has to make do with what it gets because we’re optimizing for the river.”

As the opening speaker, Dr. Petts said he wanted to dip into a whole range of questions, just to seed some ideas. “The end point is going to be hopefully a demonstration that smart hydrology with flexible operations managed locally, within a conjunctive use regional framework gives us best protection of river ecology for the future, but also uses less water for ecology, making more water available for supply. So smarter hydrology means a win-win, more for supply and more for ecology. That’s what I hope to convince you of in the next half an hour.”

He acknowledged that there are other parallel programs to improve water management: for instance, metering of water usage and other ways to improve water usage, or the debate around virtual water. “All of that is going on in parallel with our discussions about how we can drive forward river management for the 21st century, and what river management is in a world of uncertain environmental change,” he said. “Another message I want to drive forward is that despite the fact I came from a geological background originally, I don’t believe the past is the key to the future.”

So how we can derive environmental flow management scenarios for heavily developed rivers? Dr. Petts noted that in Britain, they have 4,000-6,000 years of environmental change, and their rivers are very heavily developed. “We’re not talking about restoration to anything ever approaching pristine,” he said.

In Britain, they have some real problems. There is real environmental damage caused by hydropower peaking. “This is mainstem; this is not tucked away in some headwater. We still have major problems with hydropower and certainly in the UK now, there were 200 new applications for hydropower schemes in the UK last year. Hydropower is on the up in Europe and we have to deal with that. It’s relatively new; hydropower has been relatively small beer for us over the past decades; now it’s becoming a real problem.”

Dr. Petts described the rivers in the UK as ‘naturalized’. “We’re not trying to put them back to something pristine; it’s how do we work with the contemporary catchment ecosystem to deliver the best ecological outcomes under those conditions?,” he said. “It’s more like landscape architecture than it is around natural science, but that is the reality. The key message for me is for making sure that we use soft solutions that can adapt as the world around us changes, so in the future, we are not locking the system in to a management solution that might be best for today, but has little benefit in whatever future scenario we look at. It’s looking at scenarios where we have natural evolution of our management systems in line with the changing catchment and climatic systems.”

He presented a slide with two pictures. Dr. Petts described the slide on the left as the quintessential English landscape with the garden, fields of green, coming down to a beautiful chalk stream. “It’s idyllic, but it is totally unnatural; it is safe, it is comfortable, and it is what people are used to, but let’s not kid ourselves that this is what nature intended,” he said. “The picture on the right is what nature intended – a wooded stream that’s inaccessible, you have to fight your way through the undergrowth. It’s dark, it’s damp, it’s dangerous, and the general public hate it. The fishermen hate it because they can’t get along the bank.”

The European Framework Directive requires all countries in Europe to determine for every water course, what is a good ecological status or the potential in severe cases by 2015. “What does this word ‘good’ mean in a modern environment with a highly modified landscape with generations of people who have been brought up to believe what they see as natural is indeed natural?,” he said. “There’s a real big issue here about the societal influence as opposed to the natural science influence on the word ‘good’ ecological status.”

They also have to have plans in place by 2015 to assure that there’s no deterioration. “Deterioration against what benchmark?,” Dr. Petts said. “How do we define whether the change is good or bad, given that actually our rivers our very different to what nature intended anyway? There are some really big issues that are driving some useful discussions about what is the responsibility to future generations in providing a landscape and a river system or river environment that is good, and making sure we have management processes in place to ensure no deterioration against that level of good. There have been interesting discussions around what we mean, given we don’t have the pristine environment as a benchmark.”

2012 was a really, really exciting year, Dr. Petts said. Winter 2011-2012 was the driest winter on record, and they have climatic records going back several hundred years. In May, 2012, the government declared a drought; there were restrictions on water use. “The next day, the cricket season started and it started raining, and it didn’t stop raining until Christmas. We had the wettest summer on record. We had the wettest autumn on record. We had the most disrupted Christmas holiday period on record, with thousands of people unable to get home because the railway lines were down and the road network was flooded. That was all in one year.”

After that, the government declared that Britain now acknowledges they are in a phase of climate change. “Now that is very important, because we now have a management framework for discussing uncertainty of the future,” Dr. Petts said. “We’re no longer looking back at our 20, 30 years of hydrological time series, and selecting some historic hydrological index to use as a benchmark for the future. We’re saying the future is uncertain, and that was a fundamental and very important statement that I think gives us real opportunity now to actually deal with some of the big issues that we have to face in the UK that we were unable to face in the past because of this belief that the past is the behavior of the future in terms of how the environment was going to change.”

The reality is the location of Britain means that we are going to see significant climate change,” he said. “We know theoretically that a warmer world means more intensive hydrological cycle with more moisture in the atmosphere. We’re on the West Coast of the continent with a big ocean next door and more importantly, halfway between the Azures where there’s no temperature change and the Arctic where everybody agrees there’s 1 degree C per decade – we are on that temperature gradient, between north and south and east and west and we are going to experience very severe changes.”

Following on from the driest winter to the wettest summer to the wettest autumn we now have the heaviest snow ever, and looking at the forecast of the behavior of the Jetstream, over the next few months, it is coming further south, more intense, tracking Arctic air through and I think we’re in for a very interesting period,” Dr. Petts said. “The reality is Britain has woken up to the fact that the future is uncertain, and that there are really no models adequately predicting how the future will change and we need to bring that to bear on our management of our river systems.”

Dr. Pett’s work has been on the question, how can we maintain hydrographs as natural as possible given water supply commitments? “I believe the approach should be to allow change in our system, change as nature intended, as opposed to change as humans want, but we need to recognize that we have water supply commitments as well, and of course, flood control is an integrated part of the management system.”

In Britain, the only funding for ecological work and conservation work on river system is through flood control projects; as part of a major flood control project, the government will require considerable investment in the environment. “Ironically, without a major flood control scheme, there’s very little money available to do environmental work, so the big floods give us an opportunity here,” he said.

Dr. Petts stated again that the past is the key to the future. In many cases, he said they have relied on space-for-time substitution models to try to predict what could happen; there has also been a progressive drive towards regional models that incorporate regional hydrology and ecology, identifying regions within which every river is meant to behave the same. “Clearly one of the challenges that I’ve taken on is to try to convince government and managers that actually we need to look locally, because actually every river is different, and that can be a benefit in a world of change, not a constraint,” he said.

In Britain, they have been developing biological response models to predict how the flora and fauna of the riverine landscapes will change with hydrological change or climate change when their environment is so severely degraded already. “Our biological populations are severely degraded already; our communities are artificial, fish stocking happens everywhere … The reality is to what extent can these models, based upon artificial systems, provide any basis for predicting how natural systems will respond?

Dr. Petts said that they have worked will models of various types, looking at how they can measure, monitor, and then use data from streams to predict changes of biological populations. “They are fine up to habitat level, they all fall apart when we try to look at population dynamics in the medium to long-term, but we all benefit from dealing with our managers because they produce pretty pictures. We’ve been hugely successful in convincing people that these pictures mean something. We can model hydraulics, we can model hydraulics and we can come up with habitat dynamics, and we should look very, very impressive. But do they actually tell us anything fundamental that we really believe is real? I have to say, I don’t think they do.”

Fish populations are severely depressed compared with what the pristine environment was like. “When you have so little competition in rivers for habitat, does fish behavior mean anything? Is it giving us a good model for understanding how fish behavior works in natural systems? These fish are heavily stocked any way. Stocked fish are trained to surface feed are behaving very differently to the native fish that are feeding more on the bottom. All sorts of interesting of problems there of using this sort of data.”

Over the last 50 years, they have moved through a whole series of different approaches: they have tried simple approaches and biological response models; they’ve moved into holistic models that link hydrology and ecology; a building block approach – a whole range of different types of approaches that are more conceptual than they are deterministic. “At the end of all this, having tried all these models, having applied all these models, and having made enormous mistakes … I got on the bandwagon of excitement of these news quasi-deterministic models that tried to give us answers but actually in reality have been really not very helpful.”

Dr. Petts said they were under pressure from the government to accomplish something within an election term. “We’ve got five years to come up with an answer,” he said. “We’re looking unashamedly and without any apology to developing models that focus on the hydrology. Get the hydrology right and then hope that there are positive benefits for the biota of the system. We are not going to be in the position of not doing anything because we cannot predict convincingly how the biological populations will respond over the longer time scales that they need. Action is what has been required for.”

There are certain principles they kept in mind as they developed the plan:

To derive environmental flow management scenarios for heavily developed rivers:River ecosystems are adapted to the range of flows,” Dr. Petts said. “But that requires you to accept that we need to look at the interannual variability of flows; and that dry years as important as wet years, and that very dry years as important as very wet years. It’s that inter-annual dynamic that we need to allow in the future, as well as looking at seasonal dynamics.

To provide for the ‘best’ ecological configurations under current and future naturalized river and catchment conditions:We know that there are certain key ecological periods when major flow impacts can have greater effect, and that means we need to reduce our time window of looking at riverine systems. Looking at them as a year is not good enough. I would argue looking at them on a season timescale is not good enough. I can’t see any reason technically why we can’t use real-time for the management of these systems.”

To advance ‘smarter’ flow management to optimize resource use and ecological protection. This is a contentious point, he said. “There’s a presumption in many quarters in Europe that if we can maintain natural flow and thermal regimes, this will discourage alien species. I don’t know if any actual data knowledge to prove that or support that.

Dr. Petts said there’s also an assumption that the resilience of riverine ecosystems declines with the degree of modification. “I’m not sure that’s right either. If what we’re trying to sustain is a modified system, what does resilience mean in the context of a heavily modified environment? I think that’s a real question for scientists as well.”

One of the things we can quantify with confidence is a whole range of factors around the physical landscape,” Dr. Petts said. “There’s that wonderful saying that you’ve probably heard, ‘more physics and less biology makes life easier.’ We can model the physical bit; it’s the biological bit that causes the problem, so spatial variability of habitat flows across the drainage network is real and is important; interannual flow variability is important; and morphological complexity is important.

In terms of the basic understanding of the natural history of their river basins, they are looking at resilience, linking connectivity through the drainage network, hydrological variability and access to floodplains, access to high priority interfaces, and morphological diversity. The picture on the right is of the Rhine River as nature intended it to be; it now looks like the main stem of the Mississippi. It’s dead straight, it’s channelized, it’s impounded by locks and dams, Dr. Petts said. The picture below it is of the Tagliamento River in Italy, which is the only remaining active dynamic braided river system in Europe, which has been designated as a conservation site.

What this does show us is little things,” said Dr. Petts. “There is a correlation between biological diversity and shoreline length. If you want to improve your diversity, morphological diversity is fundamental. Absolutely fundamental. It’s not just the morphological diversity in the planned form; we know there are biological hotspots. In many cases, the most important hotspot in the UK context are the tributary confluences. We’re very lucky that many of our tributary confluences have not been developed because they were dynamic and they always flooded. That means we actually have some relic habitats available to enable us to build our conservation strategies. It’s knowing where are these morphological hotspots and understanding how morphological dynamics can do a lot of help us preserve our rivers in the future.”

Dr. Petts then presented a slide of a flow regime. “It doesn’t matter if you look at the median flow regime, the most probable flow regime, the most frequent flow regime – this is your typical warm temperate deciduous environment, bioclimatic region flow regime. Peak flows January; low flows in July. These, all three of these, are a human artifact. The real world does not look like this. Therefore, to manage the flows like this would be an unacceptable artifact in my view.”

In reality, it looks like this,” he said. “If you analyze the flow regime in relation to the weather patterns that drive those flow regimes, there are four basic types in the UK. The most common one which does occur 53% of the time, so once every two years is the one you’d expect with the high flows in January, the lows flows in July. But for 26% of the time, we get a much earlier and higher peak in December – 13% of the time we get a dual peak very early December and a bigger peak in March. Another 9% of years, we get another twin peak with a very, very early flood coming through in November and a second flood peak in January.”

Dr. Petts noted that the last pattern is directly linked to climate change and did not occur in the historic record before 1990. “So we are seeing a change in the detail of the flow regime, and in my view, if we are managing a flow regime for the future, we must allow for that diversity of flow regimes. I would argue that the diversity of flow regimes benefits different species, and the reason why we have such a diverse flora and fauna is because different species benefit from having the different timings and magnitudes of flows at different times of the year. If that’s changing with different types of flow regimes coming in, as the weather patterns change, which they are, then we need to allow our ecosystems to adapt to those hydrological changes.”

Dr. Petts then presented a picture of his home river, the River Trent, noting that it’s big in terms of the UK, although not necessarily in terms of the US. “It’s 10,000 square kilometers, 275 kilometers of main river. It’s been affected by abstractions (or diversions), groundwater pumping, damns, flow augmentation from regulation works, and cooling of industrial water and mine drainage. It is totally artificial in its characteristics. There are 23 individual river systems in the basin; we have 31 gauging stations with at least 20 years of data, and we have a huge diversity of characteristics. This is a heterogeneous catchment, but it also typical of a western European catchment in terms of the diversity of sub-catchments that persist within it. And it’s that diversity that I think gives us options.”

He said the River Trent has very diverse flow regimes, from surface water dominated to groundwater dominated, and to discharge-rich rivers with import dominated flow regimes. “Interestingly if we ask, do we want to impose restrictions on abstractions (diversions), and to impose constraints on the water resources end of the spectrum, how often would we need to do that? The answer is about once every five years. It’s not every year; most years, we can allow the river to have the water, the river wants people to have the water they want, it’s not a problem. The tensions arise once every five years. That’s important from a management point of view, because it helps us evaluate the options that we have available.”

Dr. Petts said that 20% of average daily flow is a basic indicator of when rivers become degraded. So how many days does flow actually fall below 20% of average daily flow? “50% of our gauging stations never go below 20% ADF, even in a severe drought,” he said.

The message is to understand the diversity we’re dealing with,” he said. “Quite often the problem is not as severe, the tension between the water developer and the conservationist is not as strong, not as intense as we tend to think it is. We can work with the natural environment to make our life easier.”

If we look locally, we make our lives easier,” he said. “No single measure of low flow is appropriate for the different types of hydrological regime we see in different catchments. Flow protection for what supply is needed, no more frequently than every five years. And at the end of the day, exceptional rainfall events, either high or low, never affect every tributary in same way. There’s a natural resilience, a physical resilience at least, built into the system.”

In terms of smart flow regulation, Dr. Petts said they are playing with two ideas to develop operational rules for distributing flows to mimic nature and both are very simple.

First approach

The first one is to identify a seasonably-variable base flow. In Britain we would use the term ‘hands-off’ flow; this is when the flow declines to a particular low where the abstractions (diversions) have to stop. Below that level, the environment gets 100% of the low flow.”

Secondly, we know what the maximum capacity of abstraction outtakes is, so therefore anything above that is the river, that’s the flood flow component, so what we’re talking about is how we distribute the bit in the middle,” he continued. “It’s actually the intermediate flows that we’re trying to distribute. The example is based on using 20% of the flow, whatever the natural flow is. So if the natural flow is 1 Qbk, then we allocate .2 Qbks to the river. If it’s 10 Qbks, we allocate 2 Qbks to the river. We do this by using an innovative weir design, so it’s a fixed structure that automatically allocates 20% of whatever the flow is down the river and 80% down the outtake.”

He showed a graph from a river in Italy (not included in the archived power point) that shows how this was implemented: a fixed base flow all year around and then pulses of floodwater coming down. “The timing of flows is linked to whatever nature intended,” Dr. Petts said. “The magnitudes are lower, but the pattern reflects what nature intended. High flows are more or less the same, base flows are protected.”

As for the ecological response, it’s wait and see, he said. “There’s a program of research ongoing now over the next few years to see how the environment responds to it, but from a hydrological point of view, my message is simple. We can do it. We can vary this, make it as complicated or as simple as you like, but it can be done. In the past, certainly in Britain, the response was, well we can’t do that, it’s too complicated. It isn’t.”

Second approach

The second approach was to look at a reservoir system with the same objective of determining operational rules for distributing flows to mimic nature. “Here the core steps were determine seasonally variable base flow, optimize flow allocations to water supply and the environment based on flow roles, and then to deliver the environmental flows according to the timing that nature intended, linked to either a gauging station either at the reservoir inflow which triggers the timing, or a gauging station on a downstream tributary to trigger the timing. So we’re trying to map both the flood magnitude and duration, rate of rise and rate of fall, and duration to what nature intended.”

So we defined the reservoir storage zones or the amount of water available for the river and at different times, define the flow levels with the timing of releases, optimize the flow rules to balance abstractions for water supply and environmental performance using a standard optimization program, and then testing it with that very simple approach to actually test the degree of flow alteration, the aim being to minimize the degree of flow alteration,” he said. “That is the only environmental indicator performance indicator that we are concerned with – to minimize the degree of flow alteration.”

Dr. Petts said that they played with a whole range of indicators: high flows, low flows, various water supply scenarios, limitations on the valves on the dam to control the amount of water, and how flexible we can be to let water out to the downstream regulated river. “We came up with a response that for this particular system with a reliable yield of five Qbks, we would have a degree of flow alteration of flow alteration of 0.4. The interesting point here is no matter what we did in this particular system, we could not get better than 0.4. It gives us a benchmark of achievement. What is theoretically possible; what is the best we can hope for? The best we can hope for is not 100%, we can’t get back to absolutely what nature intended. It gives a scale though of what is achievable potentially is all else falls in our favor. And we believe that was a very important step forward.”

Dr. Petts said it’s about using smart hydrology with flexible operations to develop locally informed conjunctive use systems. “It gives us better ecological protections using less water,” he said.

In conclusion …

So in conclusion: there are four key elements to the regulated river hydrograph,” said Dr. Petts. One is that we have to have a hands-off base flow regime. Not least because our rivers are modified and we just don’t know what that means in terms of the loss of natural resilience or the ability of the system to recover after natural drought. That’s our safeguard. We have to allow high flows at the right time with the right frequency and the right duration and the right rise and fall. We have to optimize the distribution and allocation of water through intermediate flows and we have to allow the model to develop in an evolutionary way so that as climate changes in the future, allocation of water follows that climate change.”

I leave you with this,” said Dr. Petts. “In 1853, a cartoon in the newspaper portrayed the River Thames as dead. 1853 was a very hot summer. Parliament was so appalled by the stench of the river that Parliament shut down and the politicians went home to their country seats in the clean air of the rural England. But they were so disgusted by what they saw that it led to the very rapid and very quick development of base water control systems, sewage transport systems, and the world for river management changed and changed for the better.”

“I believe that the drought of 2011-12, the floods of 2012, and the realization by our government that the world has an uncertain future is an equivalent landmark,” he said. “We’re going to see some really significant and positive change over the next 5 to 10 years and I think it’s up to scientists both in universities and in agencies to respond to that challenge and to come up with answers that are practicable and not get embedded developing the more monastic views of how science should be developed.”


Daily emailsSign up for daily email service and you’ll never miss a post!

Sign up for daily emails and get all the Notebook’s aggregated and original water news content delivered to your email box by 9AM. Breaking news alerts, too. Sign me up!

Print Friendly, PDF & Email