Blending of satellite and tide gauge sea level observations and its assimilation in a storm surge model of the North Sea and Baltic Sea

Abstract

Coastal storm surge forecasts are typically derived from dedicated hydrodynamic model systems, relying on Numerical Weather Prediction (NWP) inputs. Uncertainty in the NWP wind field affects both the preconditioning and the forecast of sea level. Traditionally, tide gauge data have been used to limit preconditioning errors, providing point information. Here we utilize coastal satellite altimetry sea level observations. Careful processing techniques allow data to be retrieved up to 3 km from the coast, combining 1 Hz and 20 Hz data. The use of satellite altimetry directly is limited to times when the satellite passes over the area of interest. Instead, we use a stationary blending method developed by Madsen et al. (2007) to relate the coastal satellite altimetry with corresponding tide gauge measurements, allowing generation of sea level maps whenever tide gauge data are available. We apply the method in the North Sea and Baltic Sea, including the coastal zone, and test it for operational nowcasting and hindcasting of the sea level. The feasibility to assimilate the blended product into a hydrodynamic model is assessed, using the ensemble optimal interpolation method. A 2 year test simulation shows decreased sea level root mean square error of 7–43% and improved correlation by 1–23% in all modeled areas, when validated against independent tide gauges, indicating the feasibility to limit preconditioning errors for storm surge forecasting, using a relatively cost effective assimilation scheme.

Requirements for a Coastal Hazards Observing System

“…Many coastal regions in the world are vulnerable to storm surges, with approximately 45% of the world's population living within 150 km of the coast. Thus it is important to utilize satellite altimetry to enhance our capabilities of observing storm surges to complement traditional tide-gauge networks as demonstrated in Madsen et al (2015).…”

“…These data sets have been shown to be useful to monitor coastal storm surges (Han et al, 2012Lillibridge et al, 2013;Antony et al, 2014;Chen et al, 2014;Fenoglio-Marc et al, 2015b) and to improve their forecasting (Madsen et al, 2015;De Biasio et al, 2016, 2017Bajo et al, 2017;Li et al, 2018). While it is opportunistic for a single altimeter to capture a storm surge, a constellation of altimeter missions especially with wide-swath altimetry could significantly enhance the chance (Antony et al, 2014;Turki et al, 2015;Han, 2017;Han et al, 2017).…”

A basis function approach for exploring the seasonal and spatial features of storm surge events

“…An advantage of these studies is that by solving the governing equations of fluid motion, they preserve faithfulness to the underlying physics and often allow realistic depiction of real‐world storm surge events, including inundation across normally dry coastal areas. However, such methods are often limited by data requirements [ Madsen et al ., ], including lack of data availability and/or insufficient spatial or temporal resolution [ Colberg and McInnes , ]. Additionally, hydrodynamic models are computationally expensive, which can limit their ability to simulate large ensembles of events [ Nuswantoro et al ., ] as often required for risk calculations [ Ball et al ., ].…”

Progress in satellite remote sensing for studying physical processes at the ocean surface and its borders with the atmosphere and sea ice

“…More recently an impact assessment study was carried out by the Danish Meteorological Institute (DMI) within the European Space Agency eSurge project to demonstrate the impact of assimilating a blended tide gauge and altimetry dataset (Madsen et al, 2015). The observations for three storm surge events were assimilated resulting in a stable hydrodynamic ocean model.…”

“…In addition, the spatial spread of information from the coastal tide gauges is implicit in the field, which makes it easier for hydrodynamic models to assimilate. The original approach used standard altimetry observations (Høyer and Andersen, 2003;Madsen et al, 2007), but improved performance was recently obtained through the use of coastal specific altimetry data (Madsen et al, 2015).…”