Projecting inundation in the San Francisco Bay: Sea level and tides

UC Berkeley’s Dr. Mark Stacey with the latest research on the impacts of sea level rise on the Bay Area and possible responses

With sea level projected to rise at least one meter by 2100, possible even more, much needed attention is being paid on how to prepare the San Francisco Bay’s extensive shoreline to adapt to significant amounts of sea level rise. At the seminar, New Approaches for Responding to Climate Change in the San Francisco Bay-Delta presented by the UC Davis Center for Aquatic Biology and Aquaculture and the Delta Science Program, UC Berkeley’s Dr. Mark Stacey discussed the latest research on how sea level rise will impact the bay area.

It was Superstorm Sandy that brought new coastal inundation to the forefront of our consciousness, Dr. Stacey began, presenting a slide from Phil Horton who publishes the Sea & Sky New York. “What I wanted to show here is that there are a number of processes involved with defining that inundation,” he said. “You have a baseline water level that’s set by mean sea level. You see a slow rise in the day or two preceding the big event – that’s the low pressure system in the Atlantic. Then you see a more rapid rise around the event itself, which is driven by the local wind forcing and creating wind setup. But then superposed on all of that are the tides. And so that high water event that flooded parking garages and the like resulted from the combination of all of these processes, so when we start thinking about future inundation of shorelines around San Francisco Bay, it’s a collection of processes that we really need to be thinking about.”

Factors that determine water levels in San Francisco Bay

There are several factors that determine water levels within San Francisco Bay. He noted that these apply for any coastal embayment, but his talk would focus on the processes that are most important for San Francisco Bay.

Sea level: “At the mouth of the bay, you have sea level that is driven by long term trends and we know the projections for that,” he said. “Climatic variability induced by things like the Pacific Decadal Oscillation, upwelling, downwelling, and so on, and specific low pressure events over the Pacific. Then superposed on that are the tides; I’m going to group all of those together under sea level forcing. They have very different time scales associated with them: tides are 12 and 24 hours and the others are much longer than that. That’s important when we think about the response of the bay because those time scales dictate how the bay responds to those different forcings.”

Wind and waves: In the interior bay, wind setup and wave setup are factors. “In the interior of the bay, you’re going to have forcing from local winds and local wind waves, and you also have potential ocean swell influencing some parts of the bay as well, creating this additional set up.”

Freshwater flows: “Terrestrially sourced water is a huge part of the story, and backwater curves and backwater flooding is an important part of the process, so when we start looking out a century, how are precipitation patterns going to change and what does that mean for local urban runoff into the streams that then define local flooding events,” he said.

What defines new inundation?

Those are the processes that drive water level in the bay; when we start to consider the interaction of water level with shorelines, we want to understand what new set of interactions define new inundation, he said.

“The interaction of water levels with the shoreline infrastructure, defined broadly, really governs that,” he said. “That’s the collection of whatever makes up the mosaic of our shoreline, whether it’s marshes or levees or sea walls or whatever comes down the road. That interaction is going to be dominated initially by the low points in the infrastructure, or weak points and failures.”

“Large events can be a confluence of a number of different processes: high water in the Pacific, combined with high tides, combined with winds of the right orientation, and freshwater flow events of large magnitude could overwhelm large segments of that infrastructure, leading to extensive inundation,” he said.

There is an interaction between that inundation that feeds back to the definition of the water level, he said. “We have this set of physical processes that define water level in the bay; that water level interacts with the infrastructure to define flooding, but that flooding now changes the basin, it changes the geometry, the depth of the basin, changing the response of the basin to the forcing that’s defining the water level, so there’s a feedback there that we need to understand as the bay shoreline evolves,” he said. “That feedback really depends on the time scale of that underlying process of the forcing.”

“When we think about the different forcing mechanisms that define water level in the bay where we have the long time scale trends in ocean level, low pressure systems that are evolving slowly, those low frequency variations are only marginally, or probably not at all, affected by new inundations; there’s plenty of water to provide the water for the inundation and the feedback to those low frequency processes is minimal,” he said. “On the other hand, the tides with a time scale of 12 and 24 hours are really nicely tuned to the size of the basin, such that if the basin itself changes, the way the tides interact with the basin may itself change.”

The response of the basin to wind and waves is also altered by the shoreline, but in an uncertain way. “It increases the fetch for wind setup and waves, but it also creates shallow regions that can dissipate waves, so there are two competing influences there,” Mr. Stacey said. “We do know that the wind and wave forcing does increase the risk of failure, and there’s increased risk of infrastructure failure due to that forcing, but the feedback is a little less clear.”

He then presented a slide of some coastal flooding from a king tidal period in January a few years ago. “When we think about this kind of flooding, right now we worry about the king tides in January; usually whenever there’s a freshwater flow event, but when we look a century out, it’s more like the daily high tides that would create these types of events,” he said. “As I go through this talk, I might be talking about change of 5 centimeters or 10 centimeters and that’s not a big difference, but that can be the difference between a few times annual inundation and 300x annual inundation because of the threshold nature of those crossings.”

The importance of tides

Next, he presented a slide showing sea level at the mouth of San Francisco Bay, noting that the white line on the graph on the left is sea level rise, about 10 centimeters over the last century and on the right is a tidal time series from the mouth of San Francisco Bay. “The tidal range in San Francisco Bay is about an order of magnitude larger than the long-term trend in sea level rise, so small changes in the tidal response of the bay could be comparable to long term sea level rise when we think about changes in water level,” he said.

Mr. Stacey then presented a graph of 150 years of hourly water level at the mouth of San Francisco Bay, noting that it has been compiled by Stefan Talke, who has created this data set by translating handwritten archives into digital form. “In the end, it’s this phenomenal data set,” he said. “1983 stands out – that’s when the 150 year high water events were … 1983 is just a higher water year in general because of things like El Nino, freshwater flows, and other low frequency variations.”

“But when we focus on the particular event that defines this 150-year high water mark, it’s really about the tides,” he said. “The tides dominate the local signature of that event, but that’s within this backdrop of longer term variations and other processes that layer on top of long term sea level rise and underlie the tides. So the tides are a huge part of the story in San Francisco Bay when we think about water level. The tides interact with the basin to define their own dynamics.”

Tides in the San Francisco Bay

The typical tidal range of the San Francisco Bay is about seven and a half feet. When the tide propagates into the Bay, it splits between north and south bay. As the tidal wave propagates through the north bay, at Suisun Bay the tidal range is reduced to about five feet. “It’s a progressive tidal signal,” he said. “The tides progress through the northern bay and dissipate as they go. San Pablo Bay functions a little differently, but overall the story is that in north bay, as you move along the embayment, the tides dissipate.”

This is not the case in the south bay, he said. “At the southern tip of south bay, the tidal range is amplified to about eleven feet,” he said. “It’s about a 50-60% amplification of the tides at the mouth at the bay to what’s experienced at the head. That’s specifically because of the way the tides interact with the shoreline and the bathymetry of south bay, so as the shoreline and bathymetry change, that interaction changes, and that’s why we need to pay attention to it.”

It’s really a competition between tidal amplification and tidal dissipation, he said. “Amplification occurs because of two processes. One is the tidal wave reflecting off of a hard boundary, and then interacting with the incoming wave. That superposition, if it’s the right wave length or frequency, creates an amplification of that tide,” Mr. Stacey explained. “The second is due to an estuary shape or an embayment shape that provides a funneling mechanism for the tidal energy. So even in the absence of that reflection mechanism, if an embayment is shaped like a funnel, the energy is intensified as you move towards the head of the embayment. Those two processes create tidal amplification.”

Competing with amplification is dissipation by friction, he said. “So in any embayment, you’ve got a competition between these processes that are acting to amplify the tides and dissipate the tides, and the question is what wins and where does each win?”

There are examples of both in San Francisco Bay, Mr. Stacey said. “The north bay is dominated by dissipation and the south bay is dominated by amplification,” he said. “The shape of south Bay really feeds both of those amplification mechanisms: The shape of south bay provides a funneling mechanism even for a progressing tide to amplify it, but also the nature of the south bay shoreline being hardened to a great extent provides a lot of reflection of that tidal energy, creating a near resident process that amplifies the tidal wave.”

Rough estimates suggest the amplification in the south bay is due about equally to those two processes, he said. “If we inundate or change this shoreline in some way, through restoration, through accidental flooding, or through deliberate management actions of the shoreline, we may change these interactions, altering the tidal dynamics in the embayment.”

Modeling results of sea level rise simulations

Mr. Stacey then presented some modeling results that looked at how tidal dynamics might be altered, noting that this study was pursued under Coastal Conservancy funding using a model capable of resolving the detailed bathymetry, the topography of newly inundated regions, and the time scales of the tides effectively. “In order to get the analysis right, because the tides themselves are responding to shoreline, to the basin, the analysis needs to explicitly include that feedback,” he said. “As you change the shorelines, you need to feed it back into the tidal dynamics to see how the tides respond which then defines the shorelines, so it’s a completely coupled system.”

He then presented a slide showing the topography surrounding the shoreline, noting that this map does not include tides. “This is what we call our bathtub simulation,” he said. “Set the ocean level and let the water fill wherever it goes; then raise the ocean a bit and let it fill. This really just a map of the elevation of the connections of the water, but you can see that at sea level rise of about one meter, we get inundation of these broad regions north of San Pablo Bay and around the southern end of south bay.” He noted they did not really consider Suisun Bay as the focus of the modeling effort was on San Pablo and the south bay.”

For the study, at first they focused on two scenarios, which were end points for the analysis. The first scenario, the ‘hardened’ scenario, considered what would happen if sea walls were built as high as needed to key the bay waters where they are today, and the other scenario was if they did nothing and allowed the water to overtop levees and inundate the lands.

They then created two hybrid scenarios, one where the south bay was hardened and nothing was done in the north bay, and the other was if they hardened the north bay but did nothing in the south bay. Late in the project, on additional scenario, a restoration scenario, was added which considered what would happen if all planned restoration projects on the table were to proceed.

He then presented a slide of current conditions, noting that this map shows high water level minus oceanic high water level. He explained that if everywhere in the bay, the water rose exactly to the same level of the ocean, everything would be blue; the purples indicate that high water is below that of the ocean because of the dissipation of the tides; the hot colors indicate where the tides are amplified. “Under current conditions with the current shorelines and current sea level, the water at the south end of south bay is about 75 centimeters higher than ocean water level. That’s because of the amplification process.”

If we keep the shorelines where they are and build sea walls as high as needed, the picture looks very much the same, he said. “You can see about 70 cm of super elevation, maybe a slight reduction here but pretty much the same high water mark,” said Mr. Stacey. “So with existing shorelines, sea level rise doesn’t change the tidal dynamics which means you could add sea level rise linearly. A meter of sea level rise means water goes up by a meter.”

If we don’t harden levees but instead allow water to just go over the top, we get a very different picture, he said. “First of all, the high water is much more dispersed in south bay, it’s not focused at the southern tip, so we’ve lost that funneling effect, and we’ve also lost that reflection,” he said. “Perhaps more importantly, it’s reduced by 25 to 30 centimeters, so the high water here is only about 40 or 45 centimeters above ocean water level. … the response of the bay is muted a little bit … the response is reduced. There’s a protective, a mitigating effect of that inundation, because it eliminates some of those amplification processes.”

The intermediate scenarios behave a lot like the end points, he said. “If we harden north bay, but allow south bay to inundate, it looks a lot like the free inundation case, because south bay is where this amplification occurs,” he said. “If we harden south bay and allow north bay to inundate, it looks a lot like the hardened case because again it’s south bay that is where that action is. This provides some protection in north bay, but it doesn’t change the overall dynamics.”

Mr. Stacey explained that if you cut along the axis of the south bay through central bay into north bay, the lateral variation of water level is much smaller than the variation along the axis of the bay. “We’re going to aggregate in sections across the bay, just to look at how water level changes from the southern tip of south bay, through central bay to north bay, and that’s what is shown here.”

He then presented a graph of the scenarios, noting that this is extracting high water level, relative to a current condition case in the ocean. He explained the graph: “The dotted line on the top is if we hardened everything and added one meter of sea level rise, so there’s a little bit of increase, about 5 centimeters over current conditions. We compare that to the solid black line, that’s the allow free inundation case, so the difference between those two lines provide us with a quantification of the extent of protection provided by inundation, and it ranges from about 20 centimeters in south bay to something like 5 to 10 centimeters in north bay.”

“Again, 5 centimeters may not be very much, but when we start thinking about the frequency of inundation due to high water events, this 5 centimeters is going to cross different thresholds at different locations at different times, and so you’re going to increase the frequency of coastal flooding, even with a small change in the water level,” he said.

He then presented a slide showing the effects of hardening the south bay shoreline. “With only that change, it’s about a 15 to 20 centimeter level of protection in south bay,” he said. “What’s interesting is it even provides a few centimeters of protection in north bay, and that’s because south bay is creating that amplification both within south bay but also at the mouth of south bay, which is central bay, which feeds into the northern reach as well. So actions in south bay have this much larger footprint associated with them. It’s a small, a few centimeters, but this signature is robust.”

“On the other hand, the north bay is really a local process,” he said. “Inundation in north bay provides something like 5 centimeters of protection in north bay, but in south bay, a really minimal effect. And that’s because in north bay, the tides propagate up through north bay and are dissipated and there’s very little feedback back to the central part of the bay, so we have this different response to actions in south bay versus north bay.”

“There’s a link between local actions and the regional response of water level and inundation, which is governed by how the tides are interacting with the shorelines and the shoreline infrastructure,” Mr. Stacey said. “So decisions we make about levees, sea wall, marsh restoration, have a local effect on water level and potentially protection, but may also have a regional footprint associated with them, such that we really need to be thinking about spatial interactions in our analyses of these strategies.”