The Failed Recovery Plan for the Delta and Delta Smelt

By Peter Moyle

Few native species are as controversial as Delta Smelt. It is a 3-4 inch translucent fish that lives only in the California Delta, where the Sacramento and San Joaquin rivers meet. This place also happens to be the heart of California’s complex water supply system which provides fresh drinking water to 35-million Californian’s and supports a multi-billion dollar agricultural industry. As water demand increased over the years, the smelt declined, approaching extinction. This blog probes into lessons we can learn from the failure of efforts to protect the smelt, laid out in the 1990s.

Delta Smelt

In 1993, the Delta Smelt was listed as a threatened species, under the Federal Endangered Species Act of 1973 (ESA). Once listed, actions were initiated to “recover” the species under the powerful, no-nonsense provisions of the ESA. On December 19, 1994, critical habitat for the smelt was defined as essentially the entire Delta, plus the dynamic extent of the low salinity zone (technically, the location of the 2 ppt isohaline or X2), to make sure responses of different life history stages to variability in freshwater outflow were included. On November 11, 1996, the US Fish and Wildlife Service adopted the recovery plan (the Plan) discussed here. The ultimate goal of this Plan was to improve habitat conditions for Delta Smelt so it could be removed from the ESA list once it was no longer threatened with extinction. Twenty-five smelt generations later, this goal has not been achieved, and Delta Smelt populations have declined further.

Today Delta Smelt exist mainly as hatchery fish that are raised, from egg to adult, at two locations. This year, over 55 thousand hatchery Delta Smelt were released into their historical habitat areas and there is now some early evidence of their survival and spawning. But the question remains: can we really expect a self-sustaining population of smelt to re-develop in habitat which has failed to support them in the past?

Recovery Plan

Delta Smelt. Photo credit: Renee Reyes. USBR.

In this blog, I focus on the recovery plan to obtain insight into why recovery efforts have failed, despite a large body of research findings on what smelt populations need for survival. I have a personal interest in the recovery plan because in 1995 I was appointed Team Leader to prepare (officially, to assist), for the USFWS, The Recovery Plan for the Sacramento/San Joaquin Delta Native Fishes (the Plan). The Team consisted of 10 scientists with knowledge of the Delta and its fishes, representing academia, state and federal agencies, and water agencies[1]. Given the perceived urgency of our task we met frequently with a high level of participation from all members, finishing the Plan in about a year.

Early in process, the Team proposed broadening our efforts to include other native fishes that were in decline from poor habitat conditions in the Delta. Our argument was basically that recovering multiple species together fit well under Section 2 of the ESA which says that a primary purpose of the ESA is to provide a means for the conservation of ecosystems on which endangered and threatened species depend. To my surprise, the proposal was accepted by USFWS. “Accordingly, the purpose and scope of this recovery plan is to outline a strategy for the conservation and restoration of the Sacramento-San Joaquin Delta that currently supports or has the potential to support Delta native fishes (USFWS 1996, p.1)”

To create a multi-species plan, we established five criteria for adding additional species to the Plan and chose seven species: Delta Smelt, Longfin Smelt, Sacramento Splittail, Green Sturgeon, Sacramento Late Fall-run Chinook Salmon, Spring-run Chinook Salmon, and San Joaquin Fall-run Chinook Salmon. We did not include Sacramento Winter-run Chinook salmon because it had been listed previously and had its own recovery plan. Sacramento Perch were added by the USFWS after the initial draft of the Plan was completed, even though it was extirpated from the Delta.

A Naïve Approach

In retrospect, this idealistic approach was naïve. Only the Delta Smelt had legal clout, so other species could be largely ignored by managers, at least until Longfin Smelt, Spring-run Chinook, and Green Sturgeon also became listed under the ESA. Sacramento Splittail were listed for a short period but became delisted as more and better information on their biology developed (Moyle et al. 2004). An ecosystem-based approach to management of the Delta also emerged from the signing of the Bay-Delta Accord on December 15, 1994, which was supposed to restrict exports and set salinity standards that restricted outflow, to an extent. Yet the Accord (as CALFED Bay-Delta program) failed in its mission, at least as far as native fishes are concerned, despite the development of the Strategic Plan for the Ecosystem Restoration Program (ERP, issued in 1998). For Delta Smelt and other listed fishes, the goals of the recovery Plan were incorporated the ERP. The expectations were that other native fishes would have increasing or stable populations in 25 years (i.e. by 2023). The evidence is clear that this has not happened [2]. So, the rest of this blog focuses in on why the Plan also failed to recover the Delta Smelt and Delta fish habitats.

I reread the Plan after a long absence and the first thing that struck me was how little we knew about Delta Smelt at the time. We relied heavily on a few older published studies, agency reports where smelt were incidental catches, anecdotal and unpublished reports, and the life history study by Moyle et al. (1992). For quantitative assessment of distribution and abundance, such as it was, the Team relied on data from the Fall Midwater Trawl Survey of CDFW, which we thought had the best representation of both adult smelt numbers and distribution. This sampling program was established to monitor the abundance of Striped Bass, not smelt, but was regarded as generally useful for monitoring all pelagic fishes in the upper estuary and Delta.

Causes of Smelt Decline

The Team generally agreed Delta Smelt had declined to a small fraction of their historical population size, which was assumed to be represented by base-lined data from 1967-1981. The reduced populations in 1982-1992 were used for comparison, although 1987-92 were drought years. After the initial decline, most smelt seemed to be confined to the Sacramento River channel between Collinsville and Rio Vista. The Team outlined several plausible explanations for the decline. In the Plan, the causes of decline were stated to be “multiple and synergistic but seemed to be in the following order of importance (p. 19):”

1. Reduction in outflow

2. Entrainment losses to water diversions

3. High outflows

4. Changes in food

5. Toxic substances

6. Disease, competition, and predation.

Reduction in inflow to the Delta, the result of decades of dam construction and diversions, was considered to be a logical cause, because subsequent outflow from the Delta was seemingly insufficient to carry juvenile smelt to their most productive rearing area, Suisun Bay, as outflow. However, Delta Smelt numbers showed a positive relationship with outflow (X2) before 1981-82 and negative relationship after that (Kimmerer 2002). In contrast, other pelagic fishes such as Striped Bass, Longfin Smelt and American Shad showed a fairly strong positive relationship between numbers and outflow in all years.

Estimates of fish entrainment in water diversions, especially in the big state and federal pumping plants in the south Delta, were largely based on analyses of CDFG (now CDFW) that entrainment of juvenile Striped Bass was a major cause of bass decline. This justified a hatchery program for rearing Striped Bass and a program for taking young-of-year bass from the salvage operations and rearing them in cages, to increase survivorship. For Delta Smelt, Kimmerer (2008) was the first to demonstrate potential large effects of the pumping plants on their survival.

High outflows were included as a possible cause of decline because the period of decline included both years with low outflow (1987-91) and years with exceptionally high outflows (1982, 1986). It was speculated that high outflow years could “flush” Delta Smelt, and their planktonic food supply, to unfavorable habitats downstream. The Team speculated that the combination of both exceptionally high and low outflows during the evaluation period created unfavorable conditions for smelt.

Changes in food reflected a general recognition that zooplankton seemed less abundant in Delta Smelt rearing habitat than they once were. Three lines of evidence were presented: the shift in copepod species from native to non-native species, blooms of the diatom Melosira, which was regarded as low quality food for zooplankton, and the invasion of the overbite clam, Corbula amurensis, which by 1986-87, was demonstrably depleting zooplankton populations in Suisun Bay. The first two causes of change were largely dismissed, but it was quickly accepted that the overbite clam invasion was likely depleting plankton populations (Kimmerer et al. 1994). The problem from a Delta Smelt perspective is that the invasion occurred after the major smelt decline had taken place. So, the clam might be preventing recovery but was regarded as an unlikely cause for the initial decline.

Toxic substances (contaminants), especially agricultural pesticides, were listed as a potential threat but unstudied. Subsequent laboratory studies indicated contaminants of various sorts were potentially a problem, but impacts of individual toxins were hard to pin down. Delta waters at times seemed a soup of contaminants, unhealthy for fish.

The Plan found “no evidence that disease, competition, or predation has caused Delta Smelt populations to decline, despite the abundance of introduced species in the estuary (p.22-23).” It was implied, however, that further study might show that species such as Striped Bass and Mississippi Silversides were having a negative impact. It was noted that CDFW stopped planting Striped Bass in the estuary in 1992 because of potential predation on endangered species.

The loss of genetic integrity due to hybridization of Delta Smelt with non-native Wakasagi had no published support. Later studies indicated this was not a problem.


The Plan and scientific thinking in the 1980s and 90s revealed no ‘smoking gun’ causes of decline of Delta Smelt. But the Team agreed the key to recovery of smelt was habitat restoration for each major life stage: spawning, larval and juvenile transport, rearing, and adult migration. This meant having adequate outflows for each life stage, each living in a food-rich environment that was relatively free of toxins and had low entrainment in diversions. The Plan indicated the smelt population would be considered restored “when its population dynamics and distribution pattern within the estuary are similar to those that existed in the 1967-1981 period.” The distribution requirement was particularly important because recovery of endangered fish species is usually focused on population size, not distribution. The Team developed very specific criteria for distribution, because before the decline Delta Smelt were widely distributed in the Delta, including the south and central Delta. The temporary resurgence of a smelt population in the 1990s almost met the abundance and distributional requirements for delisting. Rapid changes after that made recovery no longer possible.

Figuring out how to make recovery possible required a major research program, which was quickly instituted. Generous funding for research was important for producing the insights into smelt biology that were gained. Meanwhile, the decline continued, to the point that the natural population has largely disappeared, and hatchery smelt have been planted in ‘vacant’ Delta habitat. But the total amount of suitable habitat for Delta Smelt has further diminished by invasions of aquatic plants, especially Egeria densa. These plants now line Delta channels, slowing outflow and tidal flows, filtering out sediment and organic matter and making the water clearer and warmer. Habitat in the south and central Delta is now largely free of Delta Smelt and most other open-water fishes. Instead, it is lake-like habitat that supports common non-native species such as Largemouth Bass, Bluegill, and Black Bullhead. Elsewhere, Asian and overbite clams suppress zooplankton populations while fishes such as Mississippi Silverside prey on larval smelt and other fishes. Clearly, habitat for Delta Smelt has not recovered, nor has it for other species in the Delta Native Fishes Recovery Plan, with the possible exception of Sacramento Splittail (but see )

Failure of the Plan

The causes of decline and suppression of Delta Smelt and other native fishes continue to be multiple and are covered elsewhere (e.g., Moyle et al. 2018. Hobbs et al. 2017) so are beyond the scope of this blog. But it is clear that habitat for native fishes has worsened dramatically since the 1996 Plan was adopted. This trend continues with sea level rise, climate change, new invaders, and other factors. The questions then become: “How much habitat remains available for native fishes? Is it possible to create sufficient habitat through restoration?” Can we work with rapid environmental change in ways that favor native fishes or do we just ‘roll with the punches’? The discussion of alternative responses to climate-related changes in Suisun Marsh (Moyle et al. 2014) provides some ideas of how to respond.

In short, developing the Recovery Plan for the Sacramento/San Joaquin Delta Native Fishes was basically a good idea because it amounted to a Delta native fish habitat recovery plan. It failed in part because it was never instituted on a large enough scale under a coherent plan of shared governance among state and federal agencies. Providing water to agriculture has almost always trumped providing significant water for the Delta ecosystem and its fishes. This inequity is increasingly being recognized as a violation of the Public Trust, which has a particularly long history in relation to sharing water and has been undergoing a revival in recent years in California litigation. (e.g., D. Des Jardins. 2022,

But the Plan also failed because the governance structure for Delta water and fishes has not been prepared for unanticipated changes to Delta habitats, especially those related to climate change and invasive species. Even the scientific community has been surprised by the rate of change, if not the trajectory. A grand experiment is underway to see if current Delta habitat can support a re-established, self-sustaining population of Delta Smelt. Regardless, the severe decline of Delta Smelt is a major indicator of the Plan’s failure and of the failure of more recent management efforts (Börk et al. 2020). Unless large-scale action is taken to make Delta habitats more favorable to native freshwater fishes, the other native fishes in the Plan (and outside the Plan) will likely follow the trajectory of Delta Smelt, if on longer time scales.


An ecosystem-based approach is needed to allow native aquatic species and their ecosystems to persist (Mount et al. 2019). As the recovery Plan and the CalFed strategic plan show, using this approach has attractive features, but so far the approach has failed to protect native fishes. One option is to develop biological goals for the Delta and then use multiple linked species models to define habitat conditions needed to reach those goals (Dahm et al. 2019). Then, all it takes is strong cooperative leadership, especially from state and federal fish and water agencies, to develop and use an ecosystem-based approach and to acquire the resources to implement it. Can these agencies actually work together, along with NGOs and diverse groups with interests in the Delta, to create conditions favorable for recovery of desirable species? In particular, can sufficient water be provided for the environment under the Public Trust or other laws? Or is it too late to ‘restore’ appropriate habitats? If the latter is the case, we need to decide what the Delta ecosystem of the future should look like and what species we want living there. Otherwise, nature will decide for us.

Peter B. Moyle is a Distinguished Professor Emeritus at the University of California, Davis and is Associate Director of the Center for Watershed Sciences. This blog is originally posting on Dr. Moyle’s birthday. Happy birthday Dr. Moyle!!! 🙂

Further Reading

Börk, K.S., P. Moyle, J. Durand, T. Hung, and A. L. Rypel. 2020. Small populations in jeopardy: a Delta Smelt case study. Environmental Law Reporter 50 ELR 10714 -10722 92020.

Dahm, C., W. Kimmerer, J. Korman, P. B. Moyle, G. T. Ruggerone, and C.A. Simenstad. 2019. Developing Biological Goals for the Bay-Delta Plan: Concepts and Ideas from an Independent Scientific Advisory Panel. A final report to the Delta Science Program. Sacramento: Delta Stewardship Council.

Hobbs, J.A, P.B. Moyle, N. Fangue and R. E. Connon. 2017. Is extinction inevitable for Delta Smelt and Longfin Smelt? An opinion and recommendations for recovery. San Francisco Estuary and Watershed Science 15 (2): San Francisco Estuary and Watershed Science 15(2).

Kimmerer, W.J. 2002. Effects of freshwater flow on abundance of estuarine organisms: physical effects or trophic linkages? Marine Ecology Progress Series 243:39-55.

Kimmerer WJ. 2008. Losses of Sacramento River Chinook salmon and Delta smelt to entrainment in water diversions in the Sacramento-San Joaquin Delta. San Francisco Estuary Watershed Science 6(2).

Mount, J., B. Gray, K. Bork, J. E. Cloern, F. W. Davis, T. Grantham, L. Grenier, J. Harder, Y. Kuwayama, P. Moyle, M. W. Schwartz, A. Whipple, and S. Yarnell. 2019. A Path Forward for California’s Freshwater Ecosystems. San Francisco: Public Policy Institute of California. 32 pp.

Moyle, P.B., R. D. Baxter, T. Sommer, T. C. Foin, and S. A. Matern. 2004. Biology and population dynamics of Sacramento splittail (Pogonichthys macrolepidotus) in the San Francisco Estuary: a review. San Francisco Estuary and Watershed Science [online serial] 2(2):1-47.

Moyle, P.B., J. Durand, and C. Jeffres. 2018. Making the Delta a Better Place for Native Fishes. White Paper, Orange County Coastkeeper and Center for Watershed Sciences, University of California. 63 pp.[available from author]

Moyle, P.B., J. A. Hobbs, and J. R. Durand. 2018. Delta smelt and the politics of water in California. Fisheries 43:42-51.

Moyle, P.B., A. D. Manfree, and P. L. Fiedler. 2014. Suisun Marsh: Ecological History and Possible Futures. Berkeley: University of California Press.

Moyle, P.B., R. M. Quiñones, J.V.E. Katz, and J. Weaver. 2015. Fish Species of Special Concern in California. 3rd edition. 842 pp. Sacramento: California Department of Fish and Wildlife.

[1] I apologize for giving the ERP Strategic Plan such short shrift. It has many good features including taking an adaptive management approach and basing management recommendations on conceptual models (full disclosure: I was a member of the Core Team that produced the document).

[2] Team members were Peter Moyle, Robert Pine (Executive Secretary, USFWS), Larry R. Brown (USGS), C.H. Hanson (Hanson Environmental), Bruce Herbold (USEPA), Kenneth M. Lenz (USBR), Lesa Meng (USFWS), Jerry J. Smith (San Jose State University), Dale Sweetnam (CDFG), and Leo Winternitz (CDWR).

About Andrew Rypel

Andrew L. Rypel is a Professor and the Peter B. Moyle and California Trout Chair of coldwater fish ecology at the University of California, Davis. He is a faculty member in the Department of Wildlife, Fish & Conservation Biology and Director of the Center for Watershed Sciences.
This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to The Failed Recovery Plan for the Delta and Delta Smelt

  1. The largest regret of my career was our failure to recover Delta Smelt and other native species. I believe that goal was achievable for some period of time. Now, with strong external influences including warmer temperatures and prolonged droughts I fear that goal is tenuous. However I still have warm memories of the team led by Peter Moyle who developed the recovery plan. We worked very well together and strongly believed in our shared goal.

    • Peter Moyle says:

      Leo, I agree that the team working on the recovery plan was a good one (including you, of course). We all felt we were doing something positive and put aside differences to work together, which did not preclude some heated discussion!

  2. Dan Ray says:

    “All it takes is strong cooperative leadership from state & federal fish & water agencies ….” Sounds simple enough, if only….

  3. The only hope I see is contained in the last sentence of the penultimate section of the blog:
    “Providing water to agriculture has almost always trumped providing significant water for the Delta ecosystem and its fishes. This inequity is increasingly being recognized as a violation of the Public Trust, which has been undergoing a revival in recent years in California litigation.” Let us hope that that revival continues and is ultimately successful in protecting the Delta and its native fishes.

Leave a Reply