CHRIS ENRIGHT: What Do We Mean by “Natural Functional Delta Inflow” in a Regulated and Modified System?

South Fork, American River, photo by DWR

The issue here for the day is really what do we mean by natural functional flows,” began Chris Enright, Senior Water Resources Engineer for the Delta Science Program (now retired). “What’s a natural hydrograph in regulated rivers, and the science of natural functional inflows to the Delta? So right off the bat, we have a contradiction; natural and functional. Think about it; those two terms in many ways don’t even belong together, and in some sense, that’s the purpose of this workshop is to crack that open and try to make some sense of it. Prior to the Gold Rush, natural and functional would have been redundant, because what was functional was natural, so we didn’t need to distinguish, but the Central Valley watershed is enormously modified and regulated.”

A natural flow paradigm on the other hand is supported by an enormous literature at this point and is extremely well accepted, noted Mr. Enright. “In the first drafts of the Delta Plan, we called for using the natural flow paradigm as a guide to setting flow criteria, and we received a lot of criticism for that, and I think basically rightly so, but we’re not terribly sophisticated about distinguishing about what is natural and what is functional. So I would ask provocatively, does a natural flow paradigm apply here?

The overall goal of the policy of the state of California is the coequal goals that are trying to rectify the conflicting goals of reliable water supply and protection and enhancement of the ecosystem, he said. “It really means as a practical matter conservation of native species, and even more practical matter, conservation of a few species that are endangered.”

When you hear the word ‘reliable’, you should read ‘not variable’ as that’s essentially the main characteristic of reliable, he said. “We can expect it to be this way or that way – not changing. When we hear ‘native species’, it’s just the opposite: we need variable hydrology, geomorphology, and ecosystem pattern. Diverse morphology is the key to diverse ecosystems and ecosystem health.”

Mr. Enright said the question of the conference might be, What measure of hydrogeomorphological restoration confers resilience on native species to present and future stressors? He broke down what that means. “What I am talking about with ‘measure of’ is that we’re really working at the margins of multiple demands. We’re talking about environmental flows for conservation of native species, but hydropower is a huge player and a competitor for water. Water supplies for agriculture and urban areas, a huge competitor. So the question for scientists is, what is enough? Such that we can conserve native species and to communicate that ultimately to policy people.”

Hydrogeomorphological restoration is a concept that needs emphasis, said Mr. Enright. “We tend to talk about hydrographs and morphology as though they are a different thing – and really they are not. We need to always remember that they hydrograph and the morphology are two sides to the same coin. It’s the same issue. Notions of releasing natural hydrographs into leveed systems really makes very little sense to me.”

I’ll coin the term, the natural ‘hydrogeomorphograph’; it confers resilience on native species,” he said. “We know that native species are adapted to physical process characteristics of this watershed, and even at very small scales, we need to concentrate on smaller scales. Population stability really depends on the physical diversity of the system, and disturbance regimes, really the extremes of the hydrology and the morphology really start to confer the most stability.”

There are present and future stressors, such as the growing demand for water supply and clean energy; the impacts of climate change which effectively reduces water supplies; and flood risks are enormously important.

So what’s the role of science? “Our task as scientists is first of all to say, how do native species adapt to physical processes and the disturbance regimes that are native and natural to this system? We need to know that first, and as we gain a better understanding of that, then we can apply it to answering the question of how we can make changes in a highly modified system and the changes to the hydrogeomorphograph,” he said. “There’s a bunch of components of the hydrograph that we need to think about, not just volume: it’s frequency, duration, timing, rate of change, my new favorite is predictability and how do those things affect disturbance adaptive organisms, which is really characterized by native species in our system. And then as we do all of that, to support skillfully, decision making and policy making.

Mr. Enright said the theme of this presentation would be to disentangle notions of natural and functional. “So I’m going to talk about functions of the natural flow regimes, and I’ll talk about this watershed in particular and how modified and regulated it is. I’m going to dip into unimpaired flows, one of my favorite topics, used and abused, I’m going to say. Then we’ll look at some hydrographs of the Sacramento River.”


When we talk about the hydrograph for purposes of conservation, we need to recognize that there are components to the hydrograph that are important to the way native species are adapted to the hydrogeomorphology of this system, he pointed out. It’s not just volumes, he said, but duration, and that leads into notions of hydroperiod and depth and duration of the magnitudes of especially extreme events and the timing of those events, the rates of change, and whether or not those rates of change are predictable or not.

There are also physical processes associated with the hydrograph, especially the extreme events that reset the geomorphology of river systems. “When we have extreme events, that’s when we inundate our floodplains,” Mr. Enright said. “There’s lots of retention on those floodplains which recharges groundwater and historically there would have been important interaction between groundwater and the surface water runoff.”

There are also chemical processes: Floodplains are importing organic matter and nutrients, transforming them over time, and then releasing them back into the river and estuarine environment. Base flows are extremely important to protect.

We can think about base flows in different ways,” said Mr. Enright. “For example, you can think about base flows that we would release into the rivers because we want to maintain a base flow, so we’d spill that out of a dam. The historical base flows which would have been coming out of the ground because we had higher water tables and we had influent streams would lead maybe to the same base flow but with a very different chemical signature, temperature, and everything else.”

Biological adaptation modes

Native species have evolved in relation to the hydrograph and hydrogeomorphograph in different ways: By their life history, by their morphology, and by their behavior, and those different modes are related to different components of the hydrograph, he said.

For example, the life history behaviors. Native species that use life history to adapt the hydrographs try to synchronize their life cycle growth and reproduction to the timing and predictability components of the hydrograph, such as the snowmelt recession: yellow-legged frogs are adapted particularly to the snowmelt recession portion of the hydrograph, which is highly predictable. Cottonwood trees tend to release their seeds when their wet feet in the riparian areas start to sense that the water table is dropping a little bit, so they are adapted to the predictable parts of the hydrograph.

Species that have behavioral adaptations are really responding to event magnitudes and the associated environmental cues that come with those,” said Mr. Enright. “The timing of high flow events are generally not predictable at all. They are predictable within seasons, but not within weeks. So for example, Delta smelt are staging in the low salinity zone up at the upper edge of the turbidity maximum, but it’s not until that first high magnitude event comes with the suspended sediment that comes along with it, and the optical quality or turbidity of the water changes, and that gives them the cover they need to actually migrate upstream relatively safely. Aquatic insects do it in a slightly different way; they’ve adapted to get out of the stream when in the larval stage before the flood comes, so they respond to rain.”

There are also morphological responses, which is mostly about energy allocation. They respond primarily to the frequency and magnitude components, especially those of the extreme events.

So for example, there are several floodplain plants that have figured out how to shed some of their above ground biomass during floods to protect their root systems, which is adaptive. They don’t’ want to exert a lot of drag with high flows and risk losing the whole plant,” Mr. Enright said. “Cottonwoods are ragged, beautiful trees, largely because they’ll drop big limbs during droughts so they don’t have to support that biomass during a stressful period, and that cascades into a variety of ecosystem processes, such as wood ducks up in the hole that’s left behind.”

Managed flow regimes and conservation

There are a number of different ways to think about managed flow regimes and conservation. “We can do things like impose a percentage of the natural hydrograph. So here would be a natural hydrograph and say 50% of the natural hydrograph, so we might imagine that the native species that are adapted particularly to the timing and the predictability may do quite well, because that signal is still there. The snowmelt recession will still be there in the 50% hydrograph. However, it could be that the magnitude adaptors, the ones on the floodplain that are now not inundated, they may be a bit disadvantaged.”

An alternative might be to consider adapting the hydrograph slightly differently and impose what might be called functional flows. “We would retain some portions of the flood magnitudes – we want to be sure we have at least high magnitudes but perhaps the durations aren’t as long, in order to set the environmental cues. The Delta smelt for example can respond to a big turbidity event. To vary the moderate base flows, maintain those base flows, recognizing the baseflows have their own characteristic related to geomorphology and modified landscapes, and probably in our system we would want to maintain that first fall freshet because Delta smelt are pretty wired into that.”


The Delta watershed is 22,000 square miles. “Historically the watershed was highly connected, there were four races of salmon up and down the Central Valley in these streams,” Mr. Enright said. “Today we have about 23 or so major rim dams which capture much of the runoff from the Sierra and have blocked off salmon habitat over most of the Central Valley.”

There is enormous capability for capturing and storing that water. “Overall in the Sacramento Valley, the storage capability is 80% of the mean inflow, mean runoff from the Sacramento Valley, and storage in the San Joaquin Valley is quite a bit higher,” said Mr. Enright.

On top of that, there’s an enormous infrastructure on all of the major branches of the rivers,” Mr. Enright said. “There are the State Water Project facilities and the Central Valley Project facilities; there are the rim dams that have responsibility both for flood control and water supply, obviously conflicting beneficial uses. Then there are a large number of hydropower reservoirs up in the watersheds. And then an extensive flood control and levee system which starts up at some point along the main tributaries and gathers in all of that Sierra Nevada flow, funneling it into the river system, or worse, the bypasses.”

The flood basins were historically much larger in the past with an enormous storage capacity, maybe around 4 MAF in the Sacramento Valley among the Butte and the Sutter, American, Yolo, and Colusa basins. Today there is a flood bypass system which can convey about half a million cfs through flood bypasses.

We’re also enormously highly regulated by various biological opinions, agreements, and regulations and standards up and down the watersheds, regulating flows and releases from the major reservoirs at important points along river systems and into the Delta,” said Mr. Enright. “And then we have an extremely modified system over the last 150 years.”

Water users, municipal utilities, ag water districts, and urban water suppliers who are analyzing the board’s approach of using a percentage of natural or unimpaired flows, and they are suggesting that the use of natural flow regime is a one dimensional approach that is going to cause a great deal of problems, said Mr. Enright. “They say, with a degree of truth, that if we rely solely on the natural hydrograph, we will have impacts on hydropower, and as a result of that impacts on our goals for curbing greenhouse gases, it would reduce water supply reliability, cause loss of cold water habitat protection, which would be a functional flow, and recreation impacts.”

We are highly modified and regulated, that’s the truth,” said Mr. Enright. “It’s not natural really much at all. The climate system of course is natural, the signal coming out of the sky is natural, but how it’s routed through our state is not. It’s not likely to change a whole lot, so the question for us is, what are the marginal opportunities that we have for manipulating that hydrology and hydrogeomorphology to benefit natives and what is enough? We can do that at the margin by better connecting floodplain bypasses for multiple purposes; ramping releases from reservoirs; dam removal is starting to happen; Fish bypasses are being contemplated at enormous costs around our largest reservoirs; and riparian and tidal marsh restoration.”


‘Unimpaired flows’ is an attempt to estimate what was the natural flows on all the tributaries and in the Delta; what it attempts to do is remove the effects of upstream reservoirs, imports to those basins, and exports to those basins above the reservoirs, said Mr. Enright.

But as most people know, it doesn’t include a lot of important things, and we tend to say ‘ok, it’s close enough.’ But it’s very important that estimates of unimpaired flow don’t include the infrastructure value, including levees, all of the channelization improvements that have been made in the Valley, no notion of storage in wetlands and on the floodplains and evaporation from that retention. Forest practices in the Sierra Nevada has changed the ecosystem in the Sierra Nevada enormously in some way that I don’t know. Lot’s of hardened surfaces and groundwater interaction is not included in our estimates.”

Mr. Enright said there are two kinds of unimpaired flow. “One is what I’m going to call the rim dam unimpaired flow; the other is the one most people think of, I’m calling it the Central Valley unimpaired flow. They have the same name, they are both called unimpaired flow, but they are distinctly different beasts.”

The rim dam unimpaired flows are calculated at a site below each of the rim dams. We do get a daily average from those. It is site-specific and it depends on many data sources: there are electric utilities upstream, there are water districts upstream, so different users are supplying data for that daily calculation.

A lot of the inflows to the reservoirs are measured as change in the reservoir elevation, so if the reservoir elevation changes a centimeter over the day, then that’s multiplied by the area and that’s a volume that can be changed into a rate for the flow,” he said. “Imagine if it’s windy, which it always is, so the reservoirs are building up on one side, there’s a long fetch, and then the wind relaxes … so whenever you happen to make that measurement in the change in storage, you’re going to get those effects. But it should average out over several days.”

There’s no upstream groundwater interaction considered; it’s a mass balance,” he said. “The unimpaired flow is what’s coming out of the reservoir minus the imports to the system upstream plus diversions from that plus the change in storage which is the estimation of the inflow, plus an estimate of evaporation. And it’s used in a variety of really useful and reasonable ways. We use it for water year classifications, and that’s a reasonable use.”

The second one is Central Valley unimpaired flow. “This one is dangerous, and we use it all the time,” said Mr. Enright. “It’s a monthly average; we have it from 1921 to the present and people love it because it’s this great long dataset. It’s calculated on the basis of 24 individual unimpaired flow subbasins, and it assumes many of the same things: no groundwater interactions, surface retention, channel reconfigurations are not considered.”

But it largely also depends on Bulletin 1 from DWR in 1951, where the data in that bulletin was lots of short records and stage-discharge relationships,” he said. “A lot of them are based on volume proportions and correlations to nearby basins, so a lot of the basins weren’t gauged at all. They are estimated based on the size of the basin, compared to the ones that were gauged. Even worse, there’s lots of precipitation flow correlations where they do magic with polygons and try to relate the amount of precipitation around the entire basin to what’s running off. I’m actually pretty amazed and I admire what they did, but we need to know this so we can use this dataset with caution.”

Mr. Enright then presented a chart showing the monthly average Delta unimpaired outflow. “So we gather up all those basins, add them up, call it unimpaired outflow from the Delta, and people have used this data in the literature to do things like drive models. What was the natural salinity regime in the Delta, and they drove it with these unimpaired inflows or outflows from the Delta, and then say that the salinity in the regime in the Delta historically was different than today by this amount. That is an inappropriate use I would say, and we need to at least do an error analysis of all that. It does not exist. And we can probably do much better.”


Mr. Enright then presented a hydrograph of the Sacramento River from 2007 to 2012, pointing out the shape. “Now as a dis to unimpaired flow, I’m going to use it and compare it to the measured flow in a couple of key places. The top graph is the rim dam unimpaired flow, which was far more reasonably used. You can see that we have essentially an unimpaired flow hydrograph in the wetter years, largely because where the gauges are there are some unimpaired streams.”

He noted there are also several other patterns. “We have this big hole, usually in March or April where the floods have gone by and now we’re trying to fill up the reservoir storage and so you tend to see that truncated snowmelt hydrograph which is an important pattern for lots of species. And essentially a hydrograph inversion happens just about every year.”

Then these measured flow releases are done purposely. These are essentially functional flows for cold water habitat, for ag demands, for hydropower as this is summer time when hydropower peaking demands are valuable and high, and for Delta demands. There are lots of Delta standards, so this inversion of a hydrograph of essentially the base flows is a functional flow. It doesn’t look much like a natural base flow.”

The Yuba River has many of the same things, he said. “The Yuba River is highly modified. Salmon and steelhead, which are the focus of the Yuba River accord; it’s held up as an example for very good reason. Salmon and steelhead are relegated to this portion of the river and kept out of their historic range, and so we have to do heroic things such as releasing cold water through the summer so spring run salmon can actually hold over the summer in the lower river, which is essentially the valley. The river is wide, much of it is the Goldfields, so that water heats up fast, and we have to release water to protect that species. We’re not going to get a natural hydrograph under those conditions.”


Until it’s natural, it’s necessarily functional,” said Mr. Enright. “The degree of landscape alteration is very high today. The heroic things we do to the hydrograph are quite large today in response to the change in the landscape, but as we move through time, hopefully there is a convergence. As we restore the connectivity between aquatic and terrestrial environments in the future, then perhaps the flow regimes should start to look more natural, but they have to happen together, and hopefully it would converge over time.

So natural flow has a geomorphic context,” he said. “We have to keep that in mind. Native species are adapted to components of the hydrogeomorphograph. But the watershed is highly modified, so function flow prescriptions are required at the margins. Restoration of diverse and accessible aquatic and terrestrial habitats are needed.

I think we should establish consistent terms and methods for hydrograph characterizations,” he said. “Unimpaired flow calculations need a geomorphic and a vegetation context. Central Valley unimpaired flow shouldn’t be used I think for anything; it’s an index at best. Stop using it, or let’s improve it if we think it’s important. I’m not sure why that would be; maybe that’s a discussion we can have. In any case, we should characterize the error in that both the rim dam estimates and the Central Valley unimpaired flow.”

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