Canyon section of the Shasta River in the late fall when spawning Chinook return and the focus of a water transfer program to address water quality conditions. Photo: Carson Jeffres

by Ann Willis

Unimpaired streamflow has long been the benchmark against which current stream flows are evaluated for environmental purposes. The underlying assumption is that if there is water in a stream, the stream must be healthy.

A closer look shows why unimpaired flows is often a flawed basis for environmental management, particularly when water quality is the primary problem.

Environmental flow studies seem ubiquitous. In California’s Shasta River watershed, a tributary to the Klamath River, unimpaired flows have been the basis of recent Instream Flow Needs studies. Recently, another study uses unimpaired flows for a larger regulatory effort to address California’s Water Action Plan.

The approach in the draft Shasta River study plan follows a familiar pattern: first, develop a flow model to better understand current water supply, water demand, and instream flow patterns. Then, remove all human activity (e.g., diversions, pumping, storage) to estimate unimpaired flows. Ultimately, use the model to evaluate water management scenarios that address ecosystem objectives – in California, these ecosystem objectives frequently focus on anadromous fish such as salmon.

At first, such an approach seems reasonable and systematic. However, a regulatory strategy built on unimpaired flows makes critical assumptions on how streams work; assumptions that seem increasingly invalid.

Four questionable assumptions form the basis of many unimpaired flow strategies:

1. Today’s unimpaired flows would support anadromy and native fish.

Because salmon thrived in rivers before European settlers diverted flows for mining/agriculture/etc, many assume that they would thrive again if diversions stopped today. Many streams have had too much water diverted for too long, resulting in a persistent environmental drought. But unimpaired flows do not guarantee that other vital conditions exist for fish to thrive (like desirable stream temperature, dissolved oxygen, or nutrient levels). Further, climate change has already influenced a variety of physical and ecological conditions, including the timing of flows as snowmelt shifts earlier to rainfall runoff.  Thus, even if we could restore unimpaired flow patterns to our streams, historical conditions may no longer be a realistic benchmark for current or long-term management.

2. Water temperature and water quality problems are changed by instream flow management.

Stream temperature is a major characteristic of aquatic habitat. It drives water quality processes, food webs, animal health, reproductive success, and much more. Stream temperature is naturally influenced in many ways.  Developing a management strategy for stream temperatures requires understanding the processes that drive those temperatures.

Aquatic plants form a natural canopy in Big Springs Creek, and control water temperatures during mid- to late-summer.

In the Shasta River, the things that drive temperature change as the river flows downstream. In some areas, aquatic plants control water temperature. In other places, stream temperatures are controlled by large groundwater springs far upstream that overwhelm other local influences, including diversions (similar phenomenon have been observed downstream of large dams, too). Eventually, water flowing on the surface reaches a balance with surrounding air temperature. In these situations, more water does little to influence temperature (or other water qualities), so unimpaired flows often fail to address undesirable temperature.

Generally, stream flow and water temperature are inextricably linked. But for stream temperature problems, flow-based management strategies is not always the right approach.

3. Watershed-scale models provide detailed insights for managing localized water temperature and water quality.

Fish management is as much a question of water quality as it is water quantity. Often, the data needed to understand and model these processes is highly detailed and covers many features.

Watershed models look at large areas, when the true areas of interest may only be a few key places. The Shasta Basin (outlined) is part of the much larger Klamath Basin (inset). Source: UC Davis Center for Watershed Sciences

Watershed models are the 30,000-ft view of these processes. Modelling an entire watershed to develop a management strategy specific to anadromous fish and water quality issues is like using a wide-angle lens when you really need a magnifying glass.

Local actions have local results that are often lost in a watershed-scale model. Cumulatively, many local actions may be needed to mitigate basin-scale problems. But the right combination of specific, effective local actions can only be identified on a more detailed scale than can be shown by a watershed-scale model. Figuring out where the priority areas are depends on a dedicated investigation to understand the underlying processes, which leads to myth #4:

4. “Best available data” will be enough to develop effective, watershed-wide regulatory management.

Regulators are in the unfortunate position where they must implement regulatory policies, but can only implement effective strategies on the “best available data.” The “best available data” may be insufficient to define the underlying stream processes that prompt regulatory actions. Nevertheless, agencies are mandated to regulate, whether or not data is available to develop effective strategies.

Generally, no one benefits from a poor understanding of stream processes. Science brings vital insights that are necessary to manage our resources, including what data we can afford to forgo and what data we cannot. The scientific studies funded as part of Proposition 1 are good examples of how public investment can realize public benefit through well-informed scientific findings. But Prop 1 science funding is limited to the Delta.  Proposition 68 focuses on projects in California’s rivers and streams, including the Klamath, but has no comparable funding category for scientific studies of those areas.

The Klamath watershed, though geographically remote, has already seen regulatory and litigation decisions that have profound implications for the rest of California, including the Delta. Though the short-term result of robust public funding for watersheds like the Shasta and Klamath may be an uncomfortable admission of how poorly we have understood our rivers and streams, in the long-term we provide more stability to all water users and more effective guidance to regulatory policies. A step away from unimpaired flows as a default framework for managing our streams would be a step in the right direction.

Ann Willis is an engineer at the Center for Watershed Sciences and PhD student in Civil and Environmental Engineering at U.C. Davis. Her work is currently supported by fellowships with the National Science Foundation (Graduate Research Fellowship Program) and the John Muir Institute for the Environment.

Further readings

Paradigm Environmental. 2018. Draft Shasta River Watershed Characterization and Model Study Plan.

Willis et al. 2017. Seasonal aquatic macrophytes reduce water temperatures via a riverine canopy in a spring-fed stream.

Nichols et al. 2014. Water temperature patterns below large groundwater springs: management implications for coho salmon in the Shasta River, California.

Willis et al. 2015. Instream flows: new tools to quantify water quality conditions for returning adult Chinook salmon.