Empirical Modeling of Storm Surge Inundation Recurrence over the Southern Chesapeake Bay

Storm-driven inundation events in the lower Chesapeake Bay area threaten lives and property, and may be exacerbated by several factors, such as possible future anthropogenically accelerated eustatic global sea level rise. Records show increasingly high tides and storm surges, due to coastal subsidence from the glacial isostatic adjustment, additional subsidence effects from more regionalized geological processes, and global eustatic sea level rise. This work describes the development of a statistical model that quantifies tidal extremes accounting for a long-term global eustatic sea level rise, local subsidence effects, local highest normal tides, extreme events, and includes error budget estimation to derive confidence bounds. Kriging was used to estimate model parameters between measurement stations as well as to quantify the additional uncertainty due to spatial interpolation. A distance metric following the local equivalent hydraulic radius was posed as a method for establishing the empirical structure function required for Kriging within the complex bathymetric depth and topographic elevation topology of the lower Chesapeake Bay area. Topographical analysis of low-laying terrain determined the levels of storm tide necessary to inundate all locations. Comparison of these levels to the modeled extreme tide recurrences allowed high-precision estimation of inundation by geographic point within the region of interest. Successful execution of the overall methodology was found to provide reasonable historical tidal extremes analysis and future tide extremes projections, all with quantitative uncertainty bounds. An analysis of inundation possible under the IPCC A1B scenario for a 100m resolution spatial grid covering the Hampton Roads area of the lower Chesapeake was produced with confidence limits over the span of time from 1900 to 2100 AD.