The Stony Brook storm surge models

In order to better understand regional and local meteorology and oceanography, and to advance storm surge science, the Stony Brook Storm Surge Group has been developing modern, integrated weather/storm surge hind-casting and predictive models, typically running with 60-hour time horizons. We analyze both historical and current surge events for coastal New York, Long Island Sound, the NY-NJ Harbor estuary system, and northern New Jersey (Bowman et al, 2005; Colle et al, in review). (See http://stormy.msrc.sunysb.edu for further details.)

Our models utilize surface winds and sea level pressures derived from the Penn State - National Center for Atmospheric Research (PSU-NCAR) MM5 mesoscale model (see www.mmm.ucar.edu/mm5/) running at 12 km resolution to drive the Advanced Circulation Model for Coastal Ocean Hydrodynamics (ADCIRC) (see www.adcirc.org/). ADCIRC is run on an unstructured grid whose resolution ranges from about 75 km far offshore down to around 5 m in inland waterways. Further details are given in Bowman et al (2005) and Colle et al (in review). We avoid using synthetic wind and sea level pressure predictions used in models such as SLOSH, but, rather, depend on predictions derived from modern mesoscale research models like MM5 and WRF (see www.wrf-model.org/index.php).

The skill of our models is presently being evaluated against significant historical weather situations (for example, hurricanes and nor'easters) and associated archived surge data gathered over the last 50 years, as well as current events as they occur. These data also support a real-time Web-based storm surge warning system being developed for the New York metropolitan region and Long Island (see http://stormy.msrc.sunysb.edu).

Using these models, our responses to the four questions regarding the efficacy of storm surge barriers are (Bowman et al, 2005):

• Closing the barriers in sequence at local slack water near low tide before the arrival of the surge and keeping the barriers closed for the duration of the surge would indeed work as expected.

• The two ocean barriers would lead to only a negligible rise (a few centimeters) in sea level outside (to the south of) the barriers, but the third barrier in the upper east River could lead to an increase in surge levels outside (to the east) of the barrier by around 30 cm, depending upon its location (tested for conditions observed during the 25 December 2002 nor'easter).

• Entrapped river water, precipitation and runoff inside the barriers would be within safety limits even during the annual peak of Hudson River discharge in late winter.

• Partial blockage with only one or two barriers operational would not be sufficient.

The models are also being used to:

• improve the quality of coastal surge-related flooding predictions in the metropolitan region (but not of precipitation-induced inland flooding), both for the present era and for a future milieu of rapid climate change and rising sea level;

• further investigate the hydrologic feasibility of establishing a valid scientific basis for deciding whether to further consider the construction of a set of retractable storm surge barriers;

• establish the hydrologic performance requirements (as contrasted with a civil engineering or cost analysis) of such barriers; for example, most effective locations to protect the core city, their efficacy, effects on circulation and water quality during deployment and during normal out-of-service conditions, and so on.

It is not our intent to present candidate barrier designs for discussion, although much has been learned from European and New England experiences, to be discussed below. Our general approach has been to apply our storm surge models to the bathymetry and topography of the metropolitan region, and determine the flooding that would result - inside and outside the barriers - under extreme storm conditions, with/without barriers, singly/in combination, and now/future scenarios.

Was this article helpful?

0 0

Post a comment