Storm Surge Reconstruction And Return Water Level Estimation In Southeast Asia For The 20Th Century

Keywords

extreme value analysis; return water level; Southeast Asia; statistical model; storm surge

Abstract

We present a methodology to reconstruct the daily maximum storm surge levels, obtained from tide gauges, based on the surrounding atmospheric conditions from an atmospheric reanalysis (20th Century Reanalysis—20CR). Tide gauge records in Southeast Asia are relatively short, so this area is often underrepresented in studies based on long observational records, and there are just a few studies that have analyzed storm surge trends, variability or return water levels (RWLs) from numerical models in this area. Here we develop, calibrate, and validate a multivariate linear regression model that relates the storm surge with the principal components of the local atmospheric conditions. This allows us to reconstruct storm surges for the 147 year 20CR period (1866–2012) and therefore to calculate more robust RWLs from the entire simulated data set and subsets thereof. RWLs are obtained by fitting the monthly maxima values to the Generalize Extreme Value (GEV) distribution. We find an increase in the 50 year RWL from the second half of the 19th century to the present unrelated to mean sea level; this increase is less noticeable when comparing only recent periods. Therefore, further research is needed since there is evidence that atmospheric reanalyses can include spurious trends in the late 19th and early 20th. RWLs obtained from the statistical reconstruction are validated against the ones obtained from observations and from a numerical model. Agreements are generally higher when using surge levels from the statistical model, even before its calibration.

Publication Date

1-1-2018

Publication Title

Journal of Geophysical Research: Oceans

Volume

123

Issue

1

Number of Pages

437-451

Document Type

Article

Personal Identifier

scopus

DOI Link

https://doi.org/10.1002/2017JC013143

Socpus ID

85040745772 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85040745772

This document is currently not available here.

Share

COinS