STATE OF ESTUARY: Vital signs of the San Francisco Estuary fish community

Dr. Christina Swanson takes a look at the status of the San Francisco Estuary’s fish community

The San Francisco estuary is important habitat for many fish species and environmental conditions, such as the amounts and timing of freshwater inflows, water temperatures, the extent of rich tidal marsh habitats, and pollution can affect the numbers and types of fish that the Bay can support. Thus, the measures of fish abundance, diversity, species composition, and distribution reported by SFEI’s 2019 State of the Estuary Report are useful biological gauges for environmental conditions in the estuary and illustrate the connections between the ocean, the estuary, and the estuary’s watershed, as well as underscore the dependence of the estuary’s health on healthy conditions both upstream and downstream of the Bay.

The estuary itself supports over 100 different kinds of fish species, including commercially-important chinook salmon and Pacific herring; sport fisheries like striped bass; native species that spend their entire lives in the entire estuary like the Delta smelt; others that may move back and forth between the estuary and the ocean; and others that may move back and forth between the estuary and the rivers in the watershed.

At the 2019 State of the Estuary conference, Dr. Christina Swanson with the Natural Resources Defense Council gave a presentation on work she did along with Dr. Jon Rosenfield of the San Francisco Baykeeper for the update to the State of the Estuary report that examined the data on the fishes in the estuary to evaluate the overall ecological health of the estuary itself.  These fish indicators are part of the State of the Estuary report.

Dr. Swanson began by noting that the San Francisco Estuary is a link in the chain that connects the inland watersheds to the Pacific Ocean.  It’s a huge system with the estuary right in the middle, and it has its own characteristics in this highly connected system.

It’s a very diverse community for the fish that live here, and in fact, the fish community itself therefore becomes a really important and potent indicator of the ecological health of the estuary,” said Dr. Swanson. “We’re also very fortunate in this estuary because this is one of the most intensively monitored estuaries in the world.  We’ve been surveying the fish among other things for many, many decades, and the data that we have constitutes an incredibly rich resource for analysis and also evaluation of the overall health of the San Francisco Bay.


The San Francisco Estuary is incredibly well monitored, with surveys being conducted by multiple organizations for long periods of time.  Dr. Swanson and Dr. Rosenfeld used four different surveys as the sources of data: the California Department of Fish and Wildlife’s Bay Study and the Fall Midwater Trawl survey, US Fish and Wildlife Service Beach Seine, and the UC Davis Suisun Marsh Survey.  Each of the surveys covers a different geographic region of the estuary, so they were using data from different surveys in different places in the estuary.

Everything is not perfectly uniform, but collectively these surveys and the kinds of indicators that we developed to measure using the data from the surveys give a pretty robust picture of what’s going on in the fish community in the San Francisco Estuary,” she said.

They partitioned the data geographically, looking at Suisun Bay, San Pablo Bay, Central Bay, and South Bay.

The Bay Study uses both a midwater trawl survey and an otter trawl survey and is essentially surveying fish in the offshore open waters of the estuary, but not close to shore.

The data from the surveys were organized into four attributes:

  • Abundance: How many fish are present is what is most commonly measured and reported.
  • Diversity: How many different species of fish are present in the different areas of the Bay.
  • Species composition: The relative proportion of native and non-native species as well as the relative proportions of native and non-native individual fish.
  • Distribution: Which stations throughout the estuary had native fish present.

Dr. Swanson noted that in all the indicators with the exception of species composition, they were focused on native fish species.  “We were looking at the fish that have evolved in this system and using them to tell us on the basis of how they are doing how this system is doing,” she said.

For each of the metrics, they measured the value and compared it to a reference condition.  One of the objectives of the State of the Estuary Report is to help the San Francisco Estuary Partnership evaluate the progress they are making on implementing the Comprehensive Conservation and Management Plan.  The CCMP has articulated broad narrative goals, so they tried to translate the narrative goals and interpret them relative to the results for the indicators, she said.

They measured the value of the indicator itself, and then established a primary reference condition.  The primary reference condition was interpreted to be the level at which measured values above that reference condition were considered good in meeting the CCMP goals, and if the measured values were below the primary reference condition, it was either fair or poor or very poor.

She gave an example.  For the primary reference condition, they tended to use the 1980-1989 period because that was the first ten years for the Bay Study survey, so they compared the measured values to the average for the first ten years of the survey.  They also used other reasons to establish a reference condition; for example, there’s a lot of scientific literature and ecological literature indicating that a high prevalence of non-native species is an indication of poor and often highly altered ecological conditions.  Sometimes, they were left to use their best professional judgement.

But we always started by comparing the measured values to the primary reference conditions,” said Dr. Swanson.  “In most of the graphs that I’ll show you, the primary reference condition is indicated by the red line; the incremental lines often were determined on the basis of statistical variability of the measured values.  We tried to make sure we were making meaningful distinctions between measured values that corresponded to good versus fair versus poor.  We used a five point scale, and these evaluations were being made for each of the individual indicators.

Those abundance metrics were then combined together into a single index for that particular attribute.  “We did it by mathematically assigning a score to each one of these evaluation categories, sort of like a grade point.  After we had done that for each of the four attribute categories, then we combined the attributes together, equally weighting each of them into ultimately a fish index.  And the index is essentially compiling the results from all of the indicators mathematically together to give an indication of the collective measured value of the condition of the fish community in one particular area of the estuary.”


Dr. Swanson then went over the findings for the update of the State of the Estuary report fish indicators, discussing the status and trends of the fish community in the different regions of the estuary, the possible drivers influencing the results, and how the information relates to potential management opportunities for ongoing efforts to improve the ecological condition of the estuary.

The first finding is that the fish community condition varies geographically, she said.  Based on all of those fish community attributes, the condition of the fish community is good in San Pablo Bay, Central Bay, and South Bay, and it’s poor or very poor in Suisun Bay and the Delta.  One exception is that the Suisun Marsh is the one place in the estuary where conditions are not quite as bad as in Suisun Bay or the Delta.

For most of us who work on the system, this isn’t a real surprise, but it’s being consistently shown across all of these different surveys,” Dr. Swanson said.

She presented a slide (lower, left) showing the results from the Bay Delta Study data for a diversity measure of the percentage of native estuary-dependent fish were present in the four regions of the bay.  “The Suisun Bay consistently has a low proportion of the estuary-dependent fish assemblage whereas San Pablo Bay has a somewhat intermediate percentage of that assemblage, sort of jumping up and down between good and fair.  Central Bay has the largest percentage of the estuary-dependent fish assemblage, and South Bay is on a par with that.”

The next finding (above, right) was that the overall fish community is declining.   When the results of all the indicators are aggregated collectively into the quantitative fish index, and looking at the long-term trends in the measured index over time, which in the case of the Bay Study is a 38 year record of data, the overall condition of the fish community is declining long-term, Dr. Swanson said.

This is a new result for some of the regions in the lower estuary,” she said.  “In the earlier State of the Estuary report released in 2015, some of the regions in the lower estuary were stable.  Now, each region here is showing a statistically significant decline in the overall condition of the fish index.  It is much worse in Suisun Bay, it is somewhat intermediate in San Pablo Bay, and the decline in Central Bay and South Bay is really very slight.  But statistically it’s there, and it’s the first time that we have seen and been able to detect from these data using the metrics that were using, this decline over time.”

The declines in all of the regions are being driven by declines in different numbers of attributes.  There are four attributes: abundance, species composition, distribution, and diversity.  Diversity is stable across the entire estuary; it may be worse in Suisun, but it isn’t changing over the long-time period, she said.  However, in Suisun, abundance is going down, species composition is going down meaning that non-native species are becoming increasingly more prevalent, and distribution is going down meaning that an increasing number of survey stations are not finding any native species.

Native fish are disappearing from certain regions within Suisun Bay.  In San Pablo Bay and South Bay, both abundance is going down and species composition is going down, meaning that non-native species and non-native fish are becoming more prevalent in those two regions of the Bay, whereas in the Central Bay, the only attribute that is showing a significant decline is abundance.”

In addition, for many of the indicators which translate into the calculated attribute measurements which then aggregate together into the index, the measured results for the most recent years and particularly the last two (2015 and 2017), those measured values were for many of the indicators at or near record low levels.

So in fact, the other thing that is driving this long-term decline as measured by the fish index in all these different regions of the Bay is the fact that in these most recent years, we’re seeing lower measured values then we’ve seen in the past, and those are the data points which are driving this into a statistically significant decline,” Dr. Swanson said.

The Suisun Marsh was again an exception to those trends, she noted.  “In Suisun Marsh, overall conditions remain for the most part poor, but over the long-term with the exception of abundance where we saw a decline in the early 1980s, things sort of stabilized, but what we’re seeing since the early to mid-90s is a very slow incremental increase or improvement in conditions or at least stable conditions.  This is in stark contrast to what we’re seeing in the Delta and also in Suisun Bay and the open waters of the upstream region of the estuary, so Suisun Marsh is again standing out as a little bit different.

So what are the possible drivers to explain these patterns of response?  Dr. Swanson said that in the upstream estuary, one of the most influential and key environmental drivers for the upper portion of the estuary are the amounts and timing of the freshwater inflow into the system; in contrast, freshwater inflow has less influence down in the lower estuary.  However, there have been elevated temperatures in the lower estuary in recent years.

In the upper estuary, freshwater flows have been chronically low and have continued to decline, she said.  Another element in the State of the Estuary report is a series of indicators on freshwater inflow conditions; the graph on the slide on the upper right shows the amount of freshwater flow into the estuary on an annual basis.  The bars are each color coded with red indicating very, very low inflow conditions comparable to a critically dry year, and blue indicates high freshwater inflow conditions.  The graph essentially integrates both variations in hydrologic conditions in the watershed, whether it’s a wet or dry year, as well as the amounts of fresh water that are withdrawn from the system before it flows into the estuary.

This is a combination of hydrological drought and manmade drought as a result of water management operations,” Dr. Swanson said.  “As you can see in the last 20 years, in fact in the last 18 years, almost every single year is showing up as red, meaning that the amount of fresh water that flowed into the bay in that year, regardless of what kind of hydrologic water year it was, was comparable to what would flow in in a critically dry year.  This is an estuary that is in chronic manmade drought and in fact, the conditions over the last 10 and 20 years are worse than ever.

In contrast, down in the lower estuary, temperatures have been unusually warm in the last few recent years, which may be one of the things driving the recent lower measured levels for some of the fish indicators in that portion of the estuary, she said.

However, things are different for Suisun Marsh, and while it is not necessarily an intact marsh, it is a marsh that has been monitored for a long period of time.

The fact that conditions there as measured by the fish community appear to be stable and possibly improving I think is an indicator of the value of marshes and it something that we should be looking forward to as we continue to restore marshes across the estuary,” she said.

Dr. Swanson noted that all of these conditions translate into other ecological processes, including creating conditions that foster the high prevalence and high abundance of non-native species like corbula as well as overall productivity, whether it be native species like mysis or changes in productivity as a result of non-natives.

All of these things are also connected, and the fish are essentially the ultimate indicators of all of these drivers collectively,” she said.


In terms of management opportunities, Dr. Swanson said there are only certain things that can be managed in the system.  “We can manage fresh water inflows, we do that already, and it’s pretty clear that is a critical driver which is, in particular in the upper estuary, having very large influences on the fish community,” she said.  “In contrast, we can’t manage temperature down in the lower estuary which is really being driven by a combination of the Pacific Decadal Oscillations and those other multi-year trends, as well as climate change.  We can restore tidal marshes, and while the Suisun data is interesting and mostly because it’s a long-term record, I think this underscores the importance of monitoring the tidal marsh restoration that is going on because these kinds of results may be an indicator suggesting that these are in fact really important habitats.

She concluded by noting that this research is part of the updated State of the Estuary report, which has not only the general results she presented, but also all the detailed technical appendices which describe the methods, the calculations, and the indicators to support the results.


Question:  A lot of the non-natives and even some of our native long-term fish are not well sampled by the open water sampling.  For example, the largemouth bass that supports a huge sport fishery in the Delta now but isn’t monitored in any meaningful way.  There’s a lot of other stuff that isn’t getting well monitored to my knowledge.  Do you have any thoughts on what other monitoring we should be doing to better assess the health of the fish populations and what that might be looking like?

Even the native fish aren’t necessarily well monitored by some of these surveys,” said Dr. Swanson.  “For example, we really don’t do a very good job of surveying with any of these nearshore and in the marsh habitats, so those represent opportunities for additional surveys to get a better understanding of how the fish are utilizing these systems.  I think the same thing goes for the non-natives like the largemouth bass in the Delta.  The value that these surveys do bring to us is they have been conducted using identical methods and sampling in the identical ways and the same places over a very long time period, so we were able to look at the trends of time and also compare across.”

There are always opportunities for better and more monitoring, but I really do think we need to decide whether or not that monitoring is going to be answering a question that you really want to know about,” continued Dr. Swanson. “The first thing you should be asking about when developing a new monitoring program is what is the question that you’re trying to address?  Are you uncertain as to what’s going on in this area, or are you just doing it because you know you’re not monitoring the fish right now, but maybe you really don’t care about it.  That’s a bit of an opinion there, but surveys need to be designed based on what you want to find out.

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