BAY DELTA SCIENCE CONFERENCE: Smelt in Hot Water: Is Thermal Stress the Final Blow for Delta Smelt?

Studies show fall water temperatures may play more of a role in Delta smelt survival than flow augmentation

Managing freshwater outflow in late winter-early spring to maintain the Low-Salinity Zone in Suisun Bay has been one of the primary management strategies for supporting Delta Smelt and their habitat. Meanwhile, fall flows have declined over time due to freshwater exports and the Low-Salinity Zone has been located upstream where physical habitat is thought to be poor quality. As a result, fall flows in wet/above normal years are now mandated by the 2009 BiOP to maintain the Low-Salinity Zone no less than 74km in September and October. Since 2009, this has occurred twice – in 2011 and 2017.  In 2011, the Delta smelt rebounded but not in 2017.  While 2011 and 2017 were both wet years, summer water temperatures in 2017 were near the thermal limit for Delta smelt throughout the estuary, while 2011 was much cooler, suggesting temperature may be an important factor for Delta smelt survival.

Dr. James Hobbs is a Professional Research Scientist and Lecturer in the Department of Wildlife, Fish and Conservation Biology, UC Davis.  At the 2018 Bay Delta Science Conference, he gave this presentation which looked at the role that temperature might play in the decline of the Delta smelt.

He started by noting that 2017 was a hot year, as well as the last few years.  But 2017 wasn’t actually the hottest; 2014 and 2015 were the hottest years on record, and it was the first time that mean water temperatures in the Delta were elevated.  He presented a chart of temperature, noting that from 2012 to 2016, mean average temperatures were averaging about 1.4 degrees above normal.

The lower graph shows the Fall Midwater Trawl numbers for Delta smelt abundance.  He noted the Delta smelt abundance further declined during the 2012-2014 drought, and in 2017 Delta smelt abundance did not respond as it did in 2011.

Dr. Hobbs said the question of how to manage flows for Delta smelt has been the question driving Bay Delta ecosystem research for a long time.  He presented a conceptual model for the Delta smelt, noting that this research focuses predominantly in the late summer-fall period, September to December, and there are a lot of things that could be affecting Delta smelt.

Particularly we’re interested in how Delta smelt are growing and surviving from sub-adult in the late summer through the fall,” he said.  “We think that there are a variety of things that affect growth and survival, and we think that growth rates affect survival through this time period.”


The management question about freshwater flows really comes down to a science question of how do we actually understand how the three dynamic habitat attributes that are important for Delta smelt – salinity, temperature, and turbidity – and how do they influence the growth rate for Delta smelt?

To answer this question, they studied fish otoliths.  A fish otolith is a small bone in the ear made of calcium carbonate.  As the otolith grows, it accretes a ring every day, so by counting the number of rings, a fish’s age can be determined; and the increment widths of the daily rings are a proxy for how well they grew.   So it is possible to get daily information on how fast the fish grew, he said.

The research team used two different models to process the large amount of information which gave them different time scales to look at how salinity, temperature, and turbidity affect growth of the Delta smelt.

The first model uses the marginal increments of the otolith rings and shows the mean growth rate for anywhere between 3 and 30 days.  Researchers then look at the salinity, the temperature, and the turbidity simultaneously at capture and look at how those three parameters influence growth rate.

Dr. Hobbs presented results, showing the 30-day mean of the marginal increment growth of the otolith rings.  He pointed out the smooth response to salinity.

From 0 ppt up to 8 ppt, there’s a relatively non-responsive curve; there’s a slight decline around 2 ppt and maybe a little bit about 6 ppt, but pretty smooth,” he said.  “It makes sense, you’re an estuarine fish, you should be able to deal with whatever salinity is given you.”

He pointed out the strong response to temperature.  “At low temperatures, we see slower growth, and that makes sense, being an ectotherm, your metabolism is strongly driven by the temperatures,” he said.  “We see a ramp up in growth rates from about 20 to 21 degrees but once you get about 22 degrees, growth rates are starting to decline.  When we look at Secchi depths, we see that when it’s relatively turbid, growth rates are pretty good, but as the water clears, we see a decline in growth rates.”

So turbidity has an effect, temperature has an effect, and salinity somewhat of an effect, but its pretty weak,” he said.  “The plot is a 3D plot looking at the relationship between salinity and temperature, and when you combine the intersection of salinity and temperature, there is a downturn in growth rates when you get above about 4 ppt and you get above about 20 degrees.  So that’s pretty interesting.”

The second model looks at the entire otolith chronology – the growth increment widths from the hatch all the way to the edge.  Using this information, the salinity that the fish has experienced during its entire life can be reconstructed using strontium isotope ratios.

Isotopes are present everywhere in the world, but the balance (or ratios) in which different isotopes occur varies between different substances and ecosystems.  As an organism grows, the isotopes that are ingested are incorporated into all the tissues including the skeleton.  By measuring the ratios of different isotopes and using scientific knowledge about how they occur in nature to trace them back to the sources that they came from, scientists can determine many things about the organism, such as the its diet or the environment it grew up in.  (More about isotopes by clicking here and clicking here.)

We developed this model that can predict salinity based on the strontium isotope ratio,” Dr. Hobbs said.  “We can reconstruct the strontium isotope ratio profile, merge that with the increment data, and reconstruct when the fish transitioned from fresh water to brackish water.  That gives us the salinity history for each individual fish.  Then based on the salinity history, we used continuous water quality data sondes throughout the estuary to figure out what the temperature was in freshwater and in brackish water, and then we basically infer what the temperature was.  We also used the sondes data to infer what the turbidity was.”

The second model has better data because there are a lot more data points, Dr. Hobbs noted.  “The salinity affect is sort of nil from fresh to about 1 and then there’s a steady decline of growth with salinity up until about 6 ppt.  That really makes sense.  Delta smelt typically like it around 1 to 2 ppt.  Up to about 6 ppt, they are growing okay, but then after that, their growth rates start to decline.”

Looking at the results for temperature, the smelt grow slow early in the season when it’s cool.  As springtime ramps up, they grow faster up until about 22 degrees, at which point growth rate declines, he said.  The results are similar for turbidity; once the water gets pretty clear, growth rate drops off, he said.

The interaction of salinity and temperature gives us a very similar result, maybe slightly downtown growth at about 3 ppt and again at about 20 degrees,” he said.

The two different modeling approaches both tell us that when Delta smelt are in saltier water along with high temperatures, which is generally the conditions we see during drought, growth rates decline,” he said.  “These are scaled as orders of magnitude change, so we’re actually seeing quite a large change in growth – orders of magnitudes of 2 to 3 times, so it’s a pretty drastic effect.”


The question is how will flow augmentation actually influence or significantly affect the vital rates of Delta smelt, in terms of growth and survival?  “The answer generally is that it will have an effect if the flows will actually reduce salinity, increase turbidity or reduce temperature,” Dr. Hobbs said.  “It could be any one of these things, and ideally we’d like to have all three of them.”

The chart on the left shows the mean salinity for the Fall Midwater Trawls from 1999 to present when Delta smelt have been present in the survey.  “What we see is that when it is typically saltier throughout Suisun Bay and the Delta, fish are caught at higher salinities,” he said.  “We know freshwater flows will influence the salinity distribution through the estuary, at least to some scale.  Freshwater flows in the fall do influence the salinity they experience.”

However, flow doesn’t really affect turbidity and temperature.  “The average temperature from 1999 to present shows that 2014 and 2015 were exceptionally warm and the water has been getting clearer throughout the estuary since the early 2000s.  How are we going to manage freshwater flows to affect these other two important variables?

Dr. Hobbs pointed out that the effective temperature on growth rates of Delta smelt is cumulative; it is a metabolic process.  The graph on the left shows the continuous water quality data from 1999 to present; it shows the number of days of the year that temperatures are above 20 degrees, 22 degrees, and 24 degrees.  “Every year, there’s at least 60 to 80 days where its warmer than 20 but it’s not always warmer than 22, and during 2009-2012 period, it didn’t actually get above 22 degrees that often through the estuary,” he said.  “In 2006, it was a really hot summer in July.  We had a short period of time when it got above 24 degrees.  But from 2015-2017, we had an excessive period of time when it was above 22 degrees throughout the estuary.”

The plot on the right side of the slide shows the cumulative days above 20 degrees and the mean growth rate of the fish for the years they studied from 1999 to present.  “We’re seeing a pretty strong effect on the number of days that temperatures are above this moderate threshold,” he said.  “It’s pretty much a similar result for 22 degrees; there’s just fewer data points, but the amount of time that temperatures are above a certain threshold has a strong effect on growth rate.”

Dr. Hobbs then used recent data from Leo Polansky’s abundance estimates from the Fall Midwater Trawl and calculated daily mortality rates (lower, left).  “It looks like the number of days the estuary is above 22 degrees is having some influence on mortality rates,” he said.  “You can see here in 2006, we had one of the warmest summers where it even got above 24 degrees for a period of time, and we had the greatest daily mortality rate, so the amount of time that temperatures are above these thresholds are obviously having clear impacts on the growth and survival of Delta smelt.”

He presented a slide (upper, right) showing a time series of freshwater outflows, noting that there were several years in the time series where it was wet: 2017, 2011, 2006.  Even 1999 and 2000 were rather wet years, and in 1999 and 2000, Delta smelt were abundant, but through this time series, Delta smelt abundance has drastically declined.  In 2011 they recovered, and that happened to coincide with a couple of years of cool conditions, he noted.


So how can we manage freshwater flows to improve Delta smelt?  “Looking at the flows during this time period and growth rate, there’s a slight effect that’s really probably bound to a temperature,” he said.  “We looked at mortality rates and they go in the opposite direction. 2017 and 2006 are the warmest years.  When we have the cumulative amount of time above temperature thresholds, we had the highest mortality rates.  So it’s not all flow; it’s temperature.”

We’ve been thinking about how to manage freshwater flows for Delta smelt for the better part of 20 years, and what we need to be thinking about now is how do we manage temperature for Delta smelt?  How do we manage temperature at all? Can we even manage temperature?

Several studies have shown that air temperature is a major driver of water temperatures, and that by the time the water released from reservoirs gets to the Delta, it has very little influence on the temperatures in the Delta itself, he said.  “Maybe we should be thinking about what habitat features are actually promoting heat.  With restoration coming online and the creation of more shallow water habitat and more open water habitat, we should be thinking clearly about what we’re doing in terms of restoration and how that can influence thermal management in the estuary.  It’s definitely something that needs to be integrated in our management.”

Dr. Hobbs then gave some examples of factors other than flow that can affect the water temperature.  Agricultural drainage is discharged into the Delta from multiple places, and we have no idea of what it’s actually doing to temperatures in the estuary.  There are rip-rapped levees with large blue stones that suck up the heat in the daytime and remain hot well into the evening.

All these kinds of habitat features are having some influence on the temperatures in the estuary,” he said.  “They may not be what made it hot in the last few years, but you probably now need to think about those because with climate change coming, it’s only going to get hotter.

Future work being considered is looking at thermal refuge in the estuary.  There is a consistent temperature gradient geographically from the Golden Gate to the Delta, a definite west to east temperature refuge, he said.  What about riparian habitats or deepwater habitats?  There are places in the North Delta that aren’t being sampled, but we know from surveys that it might actually be cooler.  What kind of features can we create that will keep temperatures cooler?

Dr. Hobbs concluded by recommending a recent paper which looked at vegetation and how marsh vegetation cover on Twitchell Island can actually improve thermal conditions and decrease temperatures four to five degrees.


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