The kayaker and the frog

Or a brief study of recreational dam flow releases, the natural flow regime, and yellow-legged frogs

Most paddlers are aware of the annual Gauley Festival in September with its predictable flow releases and raucous partying.  At my first Gauley Festival, I remember sitting in my boat just below the dam waiting with great anticipation for the water to be turned on.  At the prescribed time, there was a series of loud beeps and then water shot out of pipes below the dam, filling the river.  I loved it.  It was water when I wanted it and at the level I wanted and right on time.  It wasn’t until years later during my career as a scientist that I realized, filling rivers when I wanted and how I wanted might be a false victory for the rivers we love.  While I later learned that the Gauley releases are done to drawdown the reservoir for flood control purposes and paddlers just happen to benefit, further the festival celebrates the protection of this run from two proposed projects that would have inundated or dried it up, it was my introduction to dam release river paddling.

As whitewater paddlers, we celebrate recreational flow releases as victories.  A year into my graduate studies in river restoration, the victory of recreational flow releases began to hollow.  At first, the pejorative terms leading aquatic ecologists gave recreational flow releases, like “w”recreation, put a sour taste in my mouth and left me defensive.  After all, dams stopped water and recreational releases kept some water in the river during the end of the paddling season, water that would have been diverted out of the river or used for hydropower which certainly wasn’t ideal.  Compared to the dewatering common with dams and all of the impacts of hydropower on flow timing, I assumed recreational releases were an improvement.  But my colleagues in freshwater ecology were sewing doubts in my assumptions.  As a scientist, my proclivity for evidence and example lead me to research on the foothill yellow legged frogs on the Feather River in California.  Here ecological needs for an endangered species, whitewater paddlers’ interests in flow releases, and hydroelectric interests collided in a complex game of winners and losers.  This story begins with an awareness of the natural flow regime, something paddlers are often keenly aware of, and ends with over a decade of negotiations between paddlers, hydropower interests, and advocates for the endangered frog.


The Natural Flow Regime

On the most general annual cycle, in the Western U.S. streams are dominated by a flush of snowmelt in the spring and in the Eastern U.S. flows peak in the late fall when deciduous trees drop their leaves and less water evapotranspires, and, therefore, more reaches the streams.  Scientists refer to these patterns of annual flow as ‘natural flow regimes.’  The concept of the natural flow regime was clearly articulated and illustrated by LeRoy Poff, a scientist at Colorado State University, who also coined the term, natural flow regime.  The natural flow regime is important to understanding aquatic species life histories and adaptations.  Many species including fish, macroinvertebrates, and riparian plants synchronize their reproduction and other life history traits in relation to the seasonal flow regime.

When a dam is built, the impact on the natural flow regime is drastic.  The high flood flows are reduced and the low flow during summer periods are increased and made more consistent (Figure 1).   Conservationist’s early attempts to mitigate the impact of the dam on the natural flow regime saw them push for the establishment of minimal flow requirements.  To find this minimal flow conservationists and scientists worked to determine how much water is needed to support a species.  With the benefit of years of data, scientists and conservationist are learning that the issue may be more nuanced than simply letting a certain minimum of water out of a dam.  In the case of some aquatic species the timing of the flows is as important as the amount of water.  This is the case for the yellow-legged frog in California.  The timing of their reproduction and the vulnerability of the eggs to scouring by sudden high flows- such as recreational flow releases- made them an ideal case study for my shaken faith in dam releases for paddlers.

Figure 1. Hydrograph of the Green River, Utah with pre-dam flows before 1963 and post-dam flows. The high flood flows (blue) are fewer and less pronounced after the dam, while summer flows are higher and more consistent. Data from US Geological Survey, published in Trends in Ecology (Feb. 2004, Vol. 19, No. 2).

Impacts of Recreational Flows on Yellow-Legged Frogs

Foothill yellow-legged frogs, Rana boylii, are listed as a California Species of Special Concern (Figure 2) and have disappeared from 66% of their historic range in the Sierra Nevada Mountains.  Males and females live most of the time in smaller tributaries where there is abundant riparian cover keeping the stream cool, but frogs can’t complete their life cycle in dark and shady tributaries.  Tadpoles, unlike adult frogs, are herbivores. They require warm sun-lit patches of slow moving water to scrape algae and diatoms off rocks. Male frogs are the first to migrate from tributaries to the mainstream of the river were they congregate at historic breeding areas (bars and pool tail-outs) that are used year after year and what biologists’ call “leks”. The frogs broadcast their territory with a low-pitched raspy series of notes, grunts, oinks and  rattling  that can only be heard with an underwater microphone (To hear the frog calls check out this link).  Females travel to the mainstem a little later when the water temperatures are between 12-15 degrees usually from late May to early April. Females travel farther than males, the record is over 7 km based on a female with a radio transmitter in Tehama Co (work done by Ryan Bourque). In the North Fork Feather, there is also a radio-telemetry record of a frog crossing the channel at high flows, probably crawling on the bottom to get from the tributary on one bank to the cobble bar where the breeding site is on the opposite bank.  The spawning occurs in synchrony with the declining spring runoff from May to June when the flows are declining in what hydrologists call the spring snowmelt recession.  Depending on her age and size, a female frog lays between 300 to 3000 eggs in a clump along the sides and undersides of cobbles and boulders.  After 7-30 days, depending on temperature, the eggs hatch into tadpoles.  Tadpoles then develop over 2-3 months into frogs over the summer.  When the rain comes in the fall the young frogs move into the tributary streams for the winter, completing the cycle from tributary to mainstem river and back to tributary.

Figure 2. A yellow-legged old female frog full with eggs takes a break by a pool in a small tributary of the South Fork Eel River. (Alessandro Catenazzi, Flickr Creative Commons License).

In the Sierra Nevada, the hydropower utility Pacific Gas and Electric (PG&E) paid Garcia and Associates, a research firm, as part of the relicensing requirement to study yellow-legged frogs in the North Fork of the Feather River in two reaches of river, one with recreational flow releases (Cresta) and another without (Poe)(Figure 3).  They found that the timing and frequency of recreational flows do not match the life cycle of yellow-legged frogs and result in negative impacts to their reproductive success (Kupferberg et al. 2011).  Even though the recreational flow releases are not as big in magnitude as a spring flood flow, the timing later in the season causes higher mortality because the egg jelly adhesion is lower and the eggs wash off the rock and tadpoles are not strong swimmers.  The impact on the population is detected three years after the egg and tadpoles scouring because of the lag time required for a young frog to reach reproductive maturity (Figure 4).  In the case of the Cresta reach with the recreational flows beginning in 2002, the impact and decline of the population was detected three years out (Figure 4).  The Poe reach, with no recreational flow releases, showed an increase in the population over the same period.

Figure 3. Annual hydrographs from 2001 through 2002 and 2005 through 2006 for the Cresta reach of the North Fork Feather compared to the Poe Reach (light grey is the time of frog breeding and dark gray is the earlier breeding in the Poe reach. Graph from Kupferberg et al. 2011.


Figure 4. Decline in egg masses on the Cresta reach of the North Fork Feather from 2002 until 2010 compared to the increase on the Poe reach. There is a three-year lag in response to the 2002 recreational releases due to the reproductive timing of the yellow-legged frog. Graph modified from Kupferberg 2009.


Through field observations, scientists found that males select breeding sites at wide and shallow channel cross-sections.  Females lay their eggs in areas with low water velocity, behind or under rocks.  These behavioral adaptations help protect their egg and tadpoles from changes in flow. When I asked one of the main researchers, Sarah Kupferberg, about recreational flow schedules, she recommended the timing and duration of recreational flows avoid the breeding and rearing season. She suggested a minimum flow to prevent the eggs from stranding and drying out and a maximum flow to prevent scouring the eggs off the rocks and tadpoles being swept into deeper water where fish predators lurk.  Further, the ramping rate or the speed that the flow is released from a dam would not be too fast or quick to scour the frog eggs or to strand tadpoles in pools when the flows were dropped.

In talking with Sarah about her research, she asked me why whitewater paddlers do not run the rivers in the spring during the snowmelt peak.  After all my years of paddling, I still had to think about this for a bit, but it quickly became clear that recreational releases allow paddlers to run rivers when they want and paddlers usually want to run rivers on the weekend in the summer when it was warm outside.  She and another professor suggested hydroelectric utilities like PG&E pay for drysuits for paddlers so that we would be warm and satisfied with paddling in the colder spring months when releases would fit with the natural flow regime.  The comedy of their suggestion aside, their underlying question was intriguing: why don’t paddlers- a usually, anti-dam, pro-river-conservation group, want flows that are more natural?

Paddlers’ voices are heard most markedly through their advocacy groups of which the most prominent when negotiating with dam authorities is American Whitewater.  Their mission statement is “to conserve and restore America’s whitewater resources and to enhance opportunities to enjoy them safely.”  American Whitewater(AW) is at the nexus of recreational flow releases and conservation, and has a strong history of negotiating releases that are ecological in nature.

American Whitewater and the Feather River

In 2000, the Rock Creek-Cresta relicensing settlement determined the flow requirements and operations of PG&E’s hydropower dams on the North Fork Feather River.  Simultaneously, the settlement set the stage and required funding for research on the impacts of flow releases on yellow-legged frogs and established the first recreational flow releases on the Feather.  In 2000, American Whitewater conducted flow studies to determine the optimum flow levels for paddling.  As required, PG&E agreed to release recreational flows for whitewater paddlers once a month in the summer, beginning in 2002.  By 2004, an agreement between PG&E and AW was signed to set recreational flow release levels.  During this same time, frog researchers found frogs were declining due to the recreational releases.  Although the decline was not definitive in 2004, the Rock Creek-Cresta Ecological Resources Committee and US Forest Service suspended the recreational releases in 2006 because of the Special Concern status of the frog and the potential adverse effects.  In 2007, an interim three-year recreational flow plan for the Rock Creek reach and a one-year cancellation of the Cresta recreational flows was the consensus.

Paddlers were unhappy with the lack of water and conservationists were deeply concerned about the plummeting frog population.  David Steindorf, the California Stewardship Director for American Whitewater, expressed concern with the research on the frogs. He felt the sample sizes were small, with only 2-4 egg masses in some years, and small population numbers leading to potentially skewed results.  Further, he argued the tadpole data may not represent reality.  When scientists released tadpoles to detect where they moved in the stream and what habitats they preferred, they were only able to find half of the initial tadpoles. Despite his doubts about the science and initial conflicts between AW and the science community, Dave reassured me AW strives to put conservation on an equal footing with recreation and takes an integrated approach to modifications to river hydrographs.  To that end, American Whitewater has taken the position of  pushing for more natural flow recession limbs (down-ramping rates) on the Poe reach to prevent the kind of mass egg loss that occurred in 2011 when dam operators dropping flows too quickly.

When the dam required reliscensing, a new type of flow regime began to take form.  Parties, including American Whitewater and conservation groups, worked to determine a flow regime that would provide both whitewater recreational releases and ecological conditions supportive of yellow-legged frogs.  The collaborative project identified a solution: release water allocated for summertime recreational flows during the spring to benefit channel processes such as sediment transport and frogs.  A simple solution, but one that required entrenched interests to compromise under pressure and eliminated summertime recreational releases on the Cresta reach.

The science on the frogs occurred after the recreational flow releases and lead to learning on both the scientific and stakeholder sides.  AW responded responsibly on behalf of paddlers and frogs and continues to use sound science when evaluating flow modifications.  In many dam relicensing agreements, the model used to evaluate the ecology and flow release impacts is called PHABSIM.  Without going into the details of this model, the bottom line is that it does not incorporate all of the dynamism inherent in flow regimes and AW is often advocating for improvements to the hydrograph beyond what the model indicates.  AW relies on science, such as Poff’s natural flow regime, to guide flow recommendations, but as science advances as was the case with the frog, so does the approach to more sophisticated flow releases.

However, the elimination of recreational flow releases during the breeding season may have come too late.  Even though releases no longer exist on the Cresta reach, the frog population has yet to rebound.  There were additional factors affecting the frogs after the releases ended: in 2011 dam operations dropped the flow too quickly stranded eggs, in another year a car accident lead to channel dewatering to recover the bodies.  These extra impacts aside, the frog population may have missed a boom year for population growth during the years of the recreational flow releases, and it remains to be seen if the population on Cresta will sustain itself in the future.

The new flow schedule was implemented by PG&E in 2009 for the Cresta reach.  Whitewater paddlers lost recreational releases during the warm summer weekends since 2006, but gained more reliable springtime releases for geomorphic purposes (sediment transport and channel forming flows) not recreation.  The geomorphic release schedule is between May 1-7 with 800 cubic feet per second (cfs) and 1200 cfs the first weekend from noon on Saturday to noon on Sunday.  From May 8 until the end of May the flow is 600 cfs, June is 500 cfs, and July is 400 cfs.

In 2011, a follow on study on the impacts of flow releases on macroinvertebrates in the Feather River was completed.  This million-dollar research study was a continuation of the research on biological response to flow releases, and it detected changes in macroinvertebrate communities related to flow releases.  As was the case with the yellow-legged frog, this data will influence flow release timing and rates in another nod to the natural flow regime.  As before, there is a delay between the research and changes in river management.  AW’s Dave Steindorf summarized his feelings about the macroinvertebrate research as  “some bugs were happy, some didn’t care, and some were sad.”

In the next decade, dam relicensing projects will continue to provide an opportunity to reexamine flow regime impacts and set new flow regimes for the future.  River ecology research influenced the reliscensing releases on the Poe reach of the North Fork Feather, and is garnering attention in conversations about releases on numerous other rivers in the Sierra Nevada from the Pitt to the Kern.

As I grappled with the consequences of recreational releases, I skipped paddling on the Feather in 2009.   I knew the Cresta Reach had shown the impacts of releases on frogs, but I was not sure about other sections upstream and downstream.  I was starting to feel disingenuous as a paddler who might be harming the rivers I loved.  Often recreation and ecology do not align, but looking into the Feather River’s example buoyed my hopes as a model of ecological problem solving and collaboration.  Academics worked with utility biologists and forest service biologist to figure out what was going on, and stakeholders like PG&E and AW negotiated flows to protect the survival of yellow-legged frogs.




Kupferberg, S.J. Palen, W.J., Lind, A.J., Bobzien, S., Catenazzi, A., Drennan, J., and M.E. Power. 2011. Effects of flow regimes altered by dams on survival, population declines, and range-wide losses of California river-breeding frogs. Conservation Biology, 26(3): 513-524.


Kupferberg, S.J. 1996. Hydrologic and geomorphic factors affecting conservation of a river breeding frog (Rana boylii). Ecological Applications 6(4): 1332-1344.


By Kristen Podolak

Kristen works on watershed conservation programs in the Sierra Nevada for The Nature Conservancy. She is an avid paddler and believes in adopting whichever paddling discipline is closest to home.

7 replies on “The kayaker and the frog”

Bravo, Kristen. I’ve used the Conservancy’s Indicators of Hydrologic Alteration software in work on macroinvertebrates and it supports your conclusions. Whitewater paddlers have not come to terms with this dilemma yet.

viyavi misuli. gtdsaiminve Air flow-s metodit, magram me pasdaklebas vegar mivustsari da 60 lari girs. saertod araa mtkivneuli, ai saertod. kargad gavudzeli. marinastan viyavi, mometsona dzalian. akhlo momavalshi dabjenvas vapireb.

As one of the participants in the Rock Creek – Cresta settlement negotations and signatory to the settlement agreement, I find this article inaccurate, misleading and, quite frankly, insulting. For example, to draw the conclusion that several years of whitewater releases in the Cresta Reach has seriously impacted frogs while turning a blind eye to the fact that PG&E and their predecessors have been responsible for massive, frog killing flow fluctuations on the N.F. Feather for over 75 years – yet they get a free pass by calling it “project operations.”

I would point out that once the concerns were made known about the issues with Yellow Legged Frogs in the Cresta reach, the paddling community agreed to cease the recreational flow releases. By contrast, just last year, PG&E wiped out the entire Poe reach with a massive flow release (project operations) after the frogs had laid their eggs.

I would suggest the author spend a bit more time researching all of the issues connected with hydro licensing on the N.F. Feather and talk with the people involved so that she can provide a more account of this issue.

Kevin Lewis

Kristen, You are right to struggle with these topics – it is complicated and can at times pose a dilemma for paddlers. The vast majority of the time though it does not. Natural rivers support both recreation and ecology, and regulated rivers often do not. If you cant paddle because of flow manipulations, chances are some native species cant do something critical in its life history. I work for AW, mostly in the East and Northern Rockies. In many of my flow negotiations I am the first and only person advocating against flatlining the river’s flows at some very low anemic level. I always put the river first, and strive to meet ecological and recreational objectives with very limited amounts of water / clout. Every river is different but there are some themes.

Many of our projects are very ecologically oriented. The Cheoah in NC mimics natural rain storms following a natural pattern. The Bear River (ID) releases use tiny amounts of water to move massive amounts of sediment out of the river during the natural high flow window which creates great habitat with negligible impacts. Releases on the Sultan (WA) serve as fish migration and channel maintenance flows. Many of our CA river projects have slowly receding spring flows that mimic snowmelt – where before there was a flashy Hell or High Water scenario. If it were not for our advocacy for high flows these rivers and others would suffer. Collectively my colleagues and I have restored ecologically valuable flows on many rivers, often using the Nature Conservancy’s ground breaking science and tools to advocate for ecologically meaningful flow restoration.

In many cases like the Deerfield (MA), Nantahala (NC), Tuckasegee (NC), and Yough (MD), regular out of season releases are primarily for hydropower production, and recreation is an ancillary benefit. In other cases we benefit from mandated fall drawdown of flood control dams like the Gauley and Russell Fork. (Incidentally, I believe the proposed dam on the Gauley would have inundated the Upper Gauley under a reservoir – not merely changed the flow – and defeating that dam was indeed a huge victory). Back east, and especially in the Northeast, many rivers naturally have frequent fall flow pulses resulting from rains, making some fall drawdown releases within the range of natural variability. The Beaver and Moose come to mind, and the first ever upper Nantahala release this fall was followed the very next day by an identical natural pulse flow.

Natural is best, but natural is long gone on many regulated rivers. We do the best we can to restore ecological and recreational functions with tiny amounts of water, knowing the natural structure and volume of the flow regime is lost for this generation – and the lions share of water will flow through a pipe for the foreseeable future. There is no one-size-fits all answer to these challenges, and working as river advocates at the cutting edge of science, we are always learning. Humility is mandatory when dealing with complex dynamic systems – as is good scientific feedback. I look forward to the future when all dams have smart, modern, studied flow regimes that support recreation as well as the full suite of ecological values. We are working on it.

Kevin, Thank you for the insightful comments on the article. I am making a few edits based on your feedback (Gauley information and highlighting AWs ecologically oriented flow releases that you mentioned) and will post it soon. I have no doubt that AW will continue to use science to inform their work. I think the Feather River is a lesson learned, and I wanted to highlight it as a model of ecological problem solving and collaboration.

Leave a Reply

Your email address will not be published. Required fields are marked *