Hand washing sign
Trabajadores: The New Workforce in Virginia Seafood Processing
May 6, 2009
Inundated areas in New Orleans following breaking of the levees surrounding the city as the result of Hurricane Katrina. New Orleans, Louisiana. 2005 September 11. © Lieut. Commander Mark Moran, NOAA Corps, NMAO/AOC.
Fall 2009: Adapting Cities to Floods
November 11, 2009
Show all

Forecasting the Rising Tide

Hurricane Isabel

Hurricane Isabel came ashore in September 2003, bringing devastating floods to tidewater Virginia. Photo © NASA

Hurricane Isabel

Hurricane Isabel came ashore in September 2003, bringing devastating floods to tidewater Virginia. © NASA

Virginia Marine Resource Bulletin
Volume 41, Number 1, Spring 2009
By Margaret Pizer

In 2003, when hurricane Isabel barreled into Virginia, it brought thirty inches of water into Danny Bacot’s Ship Store at York River Yacht Haven in Gloucester Point. Isabel also brought some important lessons that Bacot applies to his storm management plans today.

“I would have never thought to tie down a dumpster, but they float if they’re empty,” says Bacot. “In Isabel we made a wild guess and moved all the inventory to be the height of the check-out counter or more. It turns out it was the right wild guess, but for a storm like Ernesto I only had to move things up an inch or two.”

Sometime between Isabel and the 2006 storms, which included Ernesto and an unnamed Northeaster that barely got the Ship Store carpet wet, Bacot started talking to John Boon, a professor emeritus at VIMS, to see if they could predict when the shop was going to flood. The two were able to determine benchmarks based on past storms for the water levels needed to flood Bacot’s store.

Now Bacot monitors water levels and tides for Gloucester Point on the Tidewatch website (www.vims.edu/tidewatch) that Boon maintains. “Then I make it a point in my storm preparations to find John and ask him what the water levels are going to do.”

While not everyone in coastal Virginia has John Boon’s phone number (luckily for him), Boon and several other researchers at VIMS are hard at work on projects that aim to give everyone the same type of predictive power that Bacot gets from those calls. Boon’s goal is to add storm-surge and water-level predictions to the Tidewatch website, while his colleague Harry Wang is working on computer models that can predict flooding at the street level.

The potential of these projects is huge—to help all of us plan for and protect ourselves from the effects of storms that are expected to buffet our coasts with increasing frequency and more devastating effects due to climate change and sea-level rise.

Climate Change Realities
“The Chesapeake Bay has one of the longest records of sea level at Sewells Point and Baltimore Harbor,” says Wang. These records have allowed NOAA scientists to calculate that sea level in the region has risen by an average of about four millimeters per year relative to the land since 1928. A recent report by the U.S. Climate Change Science Program (www.climatescience.gov) suggests a future scenario for the Mid-Atlantic region that includes an additional sea-level rise of more than three feet by 2100—an alarming possibility in a place where many homes and businesses are already within inches of the high-tide line.

Not every region has such a long record of water-level data, but a comparison of the data that are available indicates that the Mid-Atlantic has one of the highest rates of sea-level rise on the East Coast. Scientists attribute this difference to land subsidence in the region (we’re sinking), which combines with sea-level rise due to global climate change. Rising global temperatures contribute to higher sea levels through melting of the Antarctic ice cap and glaciers, as well as through thermal expansion—as ocean waters become warmer, they expand slightly and thus the same amount of water takes up more space.

Warmer oceans can also lead to the formation of more severe tropical storms and hurricanes, but perhaps equally important for coastal Virginia is the fact that rising base sea levels mean that even less-severe storms can produce more damaging floods. In one example of this effect, Boon, Wang, and VIMS colleague Jian Shen compared the August 1933 “storm of the century” at Hampton Roads to Isabel, which was a category one storm when it hit Virginia. Despite being a weaker storm, Isabel brought maximum water levels that were comparable to those seen in the 1933 storm. Boon, Wang, and Shen attribute this to the fact that the monthly mean sea level during Isabel was about 1.4 feet higher, a difference that was mostly due to sea-level rise during the intervening seventy years.

How High Is the Water?
Scientists like Boon and Wang can sometimes speak in an alphabet soup of what they call “datums,” the statistics they use to describe tides and water levels. But to understand the factors that go into determining where flooding will occur, you need to know just a few of these terms. For example, Boon has determined that flooding at York River Yacht Haven will occur when the extratidal high water (XHW) at the Yorktown Coast Guard Training Center is 2.5 feet. To understand what that means, we can take a look at one of the graphs the Tidewatch website produces.

The system Boon and his collaborators, including fellow VIMS professors John Brubaker and Dave Forrest, and technicians Todd Nelson and Tim Gass, have been working on is a network of water-level gauges located at eight locations around Chesapeake Bay. The Tidewatch website updates every half hour and shows astronomical tides (in blue), observed water levels (in red), and storm surge (calculated as the difference between the two, in green) at each location.

XHW is defined as an observed water level that exceeds the highest predicted high tide for that location. Storms are the most obvious factor that can cause water levels to exceed the predicted high-tide level, and since most people in coastal areas are prepared for the typical range of predicted tides, flooding should only become a problem when water levels rise into extratidal range. On the Tidewatch graphs, XHW is measured as the height above the highest tide predicted using astronomical models (called highest astronomical tide or HAT).

“One of the things that we’ve been trying to do is put something together that has all of the information that emergency managers might need but is not so overly technical that people can’t get their minds around it,” says Boon. Because the storm surge is not always in phase with the tide, that next extreme water level may not occur at predicted high tide, Boon explains. “That phasing can alter dramatically when you’re going to actually see your highest high water (XHW).”

“We’ve asked ourselves this question,” says Boon, “if we knew something about what that surge was going to do in the next twelve hours, could we then add that on to our prediction of the astronomic tide and come up with a forecast of the next high water?” Storm surge predictions are available from the National Weather Service, and Boon is working to incorporate them into his graphs (dotted green line). “We think that would be very important to emergency managers,” he adds. “We could tell them how high that next extreme is going to be and when it is going to occur at places where we have active water level monitoring going on.”

Tide Graph

Tides and water levels at Sewells Point, VA, during Hurricane Isabel. Observed water levels (red) are the sum of astronomical tide and storm surge. Boon has added predicted storm surge (dotted green) and used it to predict observed water level beyond the last observed data point on the graph (yellow). Key to abbreviations: HAT=Highest Astronomical Tide, LAT=Lowest Astronomical Tide, MHHW=Mean High High Water, MLLW=Mean Low Low Water, MSL=Mean Sea Level (averaged from 1983–2001), m30=Mean water level (averaged over last thirty days).

Testing the Models
Another benefit of the network of water gauges that Boon and his colleagues maintain is that they provide ground truthing for large-scale models that show great potential for predicting how flooding will happen in the event of the next big storm. Wang is working with a team of other researchers on one such model that would provide city-block-level predictions of inundation. The model is the product of the NOAA-funded Chesapeake Bay Inundation Prediction System (CIPS).

The challenge is to model flooding over a large area, but with very fine resolution, explains Wang. “A hurricane is on the order of 300 to 500 km wide, so you don’t want to wait until it reaches your doorstep. You want to track it when it’s far away. But the real impact—the scale of the damage—is on the order of 100 meters, like a home or a street.”

To meet this challenge, the model breaks up a large area into a grid that is finest over land, and least fine over deep water. Within each section of the grid, water levels are predicted based on wind velocities and directions, tides, topography, and water depth. “Once we simulate with the model, we need to see how well we’ve done,” says Wang, and this is where data from the water level gauges come in. Wang can compare what his model would have predicted during Isabel or Ernesto with the actual data collected during the storms at Sewells Point or other gauges. “Right now our predictions are accurate to within about fifteen centimeters (about six inches). That’s where the uncertainty begins to kick in.”

One of the limiting factors for these models at the moment is the availability of good topographic data. The output of the model at the street level is only as accurate as the data fed into it about the contours of the land.

Wang’s ultimate goal is similar to Boon’s—to help coastal residents and emergency managers prepare for storms. “Eventually we’d like to generate flooding maps in real time, but a lot of work and investment will be needed,” says Wang.

Putting Science to Work
The modeling and predictions that Wang and Boon are working on have real, practical effects for coastal residents. When a storm approaches, Danny Bacot has to decide how to manage a variety of things, from how high he moves his merchandise to whether to haul out boats from his marina. “Do I go ahead and reserve the carpet vacuums or not?” Bacot asks with a smile. “Those are the types of decisions that I make using John’s information.”

All joking aside, storms like Katrina have driven home the message that being able to plan and prepare for storms has life-and-death implications—from evacuating nursing home residents to protecting critical infrastructure like power grids. With this in mind, the VIMS researchers are seeking a variety of community collaborations.

Boon is hoping to work with the National Weather Service and local meteorologists to get accurate water-level predictions onto the nightly newscast. He and Wang and their VIMS colleagues have conducted workshops in communities like Poquoson and Jamestown that are at risk from flooding, and the scientists are working with city engineers, planners, and emergency managers to help them take the effects of sea-level rise into account in their development and emergency management plans. Brubaker, Forrest, and Boon have also recently garnered Virginia Sea Grant (VASG) funding for an outreach project in Middlesex County that will include workshops and demonstrations of the Tidewatch system to emergency managers and other local officials.

VASG marine extension leader Tom Murray says facilitating these outreach efforts is a natural fit. “Virginia Sea Grant supports this outreach aimed at providing the most contemporary, local, science-based information to the Commonwealth’s coastal communities. A better understanding of the local risks arising from the increasing intensity of coastal storms and flooding will help residents and leaders develop practices to reduce vulnerability and allow for a quicker and more effective response when coastal flooding occurs.”