The Interagency Ecological Program releases the peer-reviewed report, An updated conceptual model of Delta smelt biology: our evolving understanding of an estuarine fish
The Delta smelt is perhaps California’s most famous fish, having found itself placed firmly in the center of courtroom battles and water project operations. Once abundant throughout the estuary, the Delta smelt has been protected under federal and state endangered species regulations since 1993. After a 2008 biological opinion determined that continued operation of these two water projects was likely to jeopardize the continued existence of the delta smelt and adversely modify its critical habitat, restrictions on water exported from the Delta were placed on both the Central Valley Project and the State Water Project in order to protect the species. Those biological opinions have been the subject of long and intense litigation and numerous court rulings and appeals, with recent court decisions upholding them.
However, these regulatory mechanisms have not been able to halt the decline of the species. Population abundance of the Delta smelt decreased in the 1980s, prompting its listing; abundance dropped even lower after the onset of the Pelagic Organism Decline in 2002. Delta smelt abundance briefly rebounded in 2011, a wet year, but subsequent years have continued to show very low numbers. The 2014 Fall Midwater Trawl results found the lowest numbers for Delta smelt ever.
Much science has been underway over the decades to try to find answers to the decline of the Delta smelt and other native species in the estuary, most notably, the Interagency Ecological Program (IEP), which is a collaborative science effort between nine state and federal agencies and one NGO that work to develop a better understanding of the Delta’s ecology and the effects of water project operations on the physical, chemical, and biological conditions of the estuary.
Recently, the IEP released an updated peer-reviewed conceptual model and report for the Delta smelt describing habitat conditions and ecosystem drivers affecting the species throughout its different life stages and across the different seasons. A conceptual model summarizes what is known about a system or species, as well as the hypotheses about ecosystem structure and function; it is used a tool used by scientists to describe linkages and cause-and-effect pathways in an ecosystem. The updated Delta smelt model has been useful for deriving and evaluating testable hypotheses about the Delta smelt, as well as identifying critical data and information gaps. The report also summarizes and synthesizes the substantial research that has been done, as well as identifies significant gaps in data and research.
At the January meeting of the Delta Stewardship Council, Dr. Ted Sommer from the Department of Water Resources gave a presentation to the Council on the new report. He was accompanied by Dr. Larry Brown with the USGS.
Dr. Ted Sommer, Department of Water Resources
Dr. Ted Sommer began by noting that he was here on behalf of the Interagency Ecological Program. “Any week of the year, we have an army of folks out there collecting data on fish and invertebrates and water quality and so forth,” he said. “We’ve been pretty good over the years at getting that information out, but one of the things we are really behind on is synthesizing the information and putting it in a format that managers can use, so this is something that we’re really working hard now to try and correct.”
One of the challenges is that there is a lot of diverse kinds of information, so they’ve been working with conceptual models, Dr. Sommer said. “Conceptual models are a way of organizing the information that we have,” he said. “It allows you to organize diverse information and it also provides some tools for resource management.”
Over the years, there have been conceptual models for the Pelagic Organism Decline and for the fall habitat work, he said. “One of the most intensive things that we tried to do is our effort to build a conceptual model for Delta smelt by putting together all of the available information that we have,” he said. “Obviously Delta smelt are one of the species of management interest. We’ve had biological opinions recently and some high profile court battles. We’ve learned a lot since then, and thought it was timely to try and organize the information in a useful way.”
He then presented a slide of the conceptual model, explaining that conceptual models are formalized versions of mental models that are communicated to others verbally and graphically. He then briefly described how the model was organized: “In the center, we have seasons: winter, spring, summer and fall. That is surrounded by the next layer which is the smelt at different life stages. We include the habitat conditions that affect the species, the environmental drivers that control the habitat conditions, and the habitat features that we have in the estuary. These all fit together in a seasonal approach.”
He then presented a slide of a section of the model pertaining to the spring, pointing out that portions of the model can be broken out and expanded to include more information. “This is a breakout of some of the factors that we identify that are influencing the fish in the spring,” he said. “And so starting at the top, the blue part are the fish; the next layer are the habitat attributes, below that, the environmental drivers, and lastly the landscape attributes. You can see in the figure how all kinds of different things fit together to influence the growth and survival of the species.”
Dr. Sommer noted that the report is much more than just figures. “The document has a lot of really detailed information with whole chapters that are basically devoted to each layer to really explain the basis for the thinking,” he said.
The information was used to generate specific hypotheses. “It allows us to look at some of the linkages of all the different things that work together to influence Delta smelt,” he said. “Here’s an example for how turbidity affects juveniles and sub-adults. We have juveniles on the far left that we’re interested in the growth and survival of; the main way they die is they usually just get eaten. At the top, you see if there are predators nearby, that’s a big issue, but down below and to the right, you see that turbidity also plays a big role whether the fish can hide in turbid water. It’s influenced by environmental conditions, flow, and sediment supply in the watershed.”
“Again, the idea here is that we have a conceptual model that gives folks an understanding of how things really fit together, instead of the usual process of picking one favorite factor and trying to look at its individual importance,” Dr. Sommer said.
He then gave his ‘take home’ messages:
It takes a whole year to make a smelt. “Everyone has their favorite season, so fall, winter, spring, whatever, what we’ve learned is that every season for this species really matters,” he said. “That’s somewhat intuitive, but we try and provide the basis for that.”
Model used to evaluate hypotheses: The model was used to try and answer the question, why did Delta smelt suddenly do so well in 2011? “They’d been doing miserably for the past decade, why in 2011, when it was a wet year, did they finally do well? One of the things that we found is compared to another recent year, 2006, they both produced lots of babies in each of those years, but what happened differently in 2006 is we had a very hot summer and we think that undermined a lot of the fish. The idea here is that you can have really good conditions early in the year, but it can easily change if something happens later in the year.”
The report includes a number of recommendations for how the information can be used. “We’ve tried to use this information to address specific questions that managers are interested in,” he said. “We’ve also tried to use it to help identify some of the science priorities and gaps so the document has a whole section about some of the key things that we’re missing for Delta smelt. We’re hoping that helps inform the Science Action Agenda as well.”
The document can provide guidance for adaptive management, he said. “As part of the adaptive management, a key initial step is to have a conceptual model, so we’re hoping we’re giving people a head start here by providing a document,” Dr. Sommer said. “And lastly, we’re hoping some of the descriptive information we’ve provided here provides a springboard for the development of quantitative models.”
Putting the model to use in the drought
Dr. Sommer then gave an example of how they are applying this information in the drought. “There’s a lot of information pouring in, and we thought it was a good opportunity to show how it could be used to help organize the information,” he said. “So many of the same folks that worked on the MAST effort took up the task of trying to use some of the information that’s been coming in from our long-term monitoring and also some additional special studies to answer three questions relative to the drought: what sort of changes have there been relative to the last decade, how different are conditions that we’re seeing now compared to the last drought, and are we seeing any big ecosystem changes.”
First, using the same tiers as in the model (fish, habitat attributes, environmental drivers, and landscape attributes), and then for each part of the model, they summarized the different features of it, including only those variables they thought would be influenced by the drought. “After that, we looked at how each of those variables might apply to different life stages, and then lastly, for each of the seasons of interest, we made predictions about what sort of change we thought would occur during the drought. We thought abundance, distribution and growth would all decline, and we have much more detailed predictions for a number of other things.”
“We start with predictions, we start compiling all of the information and then look to see if its different than what we expected and what sort of changes might be necessary,” he said. “As of last week, we have a nice progress report summarizing some of the results of that analysis. A lot of the managers, our bosses, were interested in what sort of changes are going on. There’s a big move to do environmental analysis relative to some of the proposals, things like barriers in the Delta, so people need this sort of information now.”
Other observations and applications
Dr. Sommer said that the decline of the Delta smelt has made headlines. “We’re seeing record low abundance and that’s definitely something that we predicted here,” he said. “But there have been a number of other interesting changes as well.” He then gave a few examples:
Increasing water clarity: He presented a graph of four different salinity zones of the estuary. “You see the Delta has been getting clearer and clearer and during the drought,” he said. “We’re seeing a dramatic increase in water clarity, and this is a big issue for the habitat for a number of these species that really need turbid water.”
Increasing water temperatures: “There’s a major increase in water temperatures during the drought, and that’s devastating for Delta smelt – it gets hot the Delta smelt get cooked. It’s also a disaster for some of the other fish – salmon, longfin smelt, and others.”
More harmful algal blooms: “Our surveys also record information on harmful algal blooms. This is a graph showing for all the different stations that we do observations on in fall. Which of the stations do we see harmful algal blooms? Basically during the drought, almost everywhere we look we see this blue green algae called Microsystis, known toxin, producing alga, it’s throughout the system, so a real degradation in habitat conditions.”
Increasing invasion rate: “One of the things that we predicted as part of the drought would be an increasing invasion rate,” he said. “Indeed as a hint that this might be going on, we actually got two new invaders this year. One is a striped mullet, familiar worldwide as an estuarine fish. It appeared this year at the salvage facilities. The other is a weather loach that seems to have become established in the lower San Joaquin River.”
Dr. Sommer concluded by noting that the MAST approach useful for other science groups:
Collaborative Adaptive Management Team (CAMT), an effort of the key agencies and organizations that have been involved with the federal lawsuits for Delta smelt and salmon to work together to collaboratively resolve some of the sources of disagreement and uncertainty, has adopted this model as a way to understand the role of entrainment within the life cycle of Delta smelt.
Tidal wetlands restoration: The team has been applying the Delta smelt MAST and identifying the features of the model they expect might change as part of habitat restoration activities, and then picking methods, metrics, and hypotheses that can then be tested as part of habitat restoration.