A hybrid downscaling approach for the analysis of coastline response to changing climate forcing in the period 1900-2010
Jose A.A. Antolinez1, A. Brad Murray2, Fernando J. Mendez1, María Yolanda Medrano Gutiérrez3 and Antonio S Cofino4, (1)University of Cantabria, Ciencias y Tecnicas del Agua y del Medio Ambiente, Santander, Spain, (2)Duke University, Nicholas School of the Environment, Durham, NC, United States, (3)University of Cantabria, Departamento de Ciencias y Tecnicas del Agua y del Medio Ambiente, Santander, Spain, (4)University of Cantabria, Departamento de Matematica Aplicada y Ciencias de la Computación, Santander, Spain
Abstract:
Long-term coastal evolution models can be forced with long-term time series of sea state parameters or with representative angular distribution of wave influences on alongshore sediment transport. The use of observational data –buoys– is usually restricted to 15-20 years, requiring transformation of waves, and in most of the cases, there are gaps in the time series. The development of atmospheric reanalysis can dramatically extend the period to more than 100 years, depending on the resolution of the wind forcing (1979-2015 for the hourly 0.3º spatial resolution CFSR atmospheric reanalysis, 1948-2015 for the 6-hourly 2º spatial res. NCEP/NCAR reanalysis, 1871-2010 for the 6-hourly 20th century reanalysis, 20CRv2). The spatio-temporal wind fields associated with these reanalysis are the source terms of global or regional 3rd generation wave model forcing that leads to different available state-of-the-art wave hindcast models at a spatial scale varying between 0.5-1º. The need to model wave transformations from deep water to at the base of the shoreface (~ 20 meters depth), at a spatial resolution of ~1 Km, requiring many hours of CPU time, presents an additional challenge.
To address these challenges, we propose a hybrid approach combining: (a) the use of long-term deep water buoy time series; (b) efficient hybrid wave transformation to the shoreface with SWAN model and data mining techniques (Camus et al, 2011); (c) a statistical downscaling model based on daily sea level pressure-based (SLP) weather types that allows us to downscale multivariate sea state parameters (significant wave height, SWH, peak period, Tp, mean wave direction, MWD) (Camus et al, 2014); (d) reconstruction of daily values of SWH, Tp and MWD in the period 1871-2010 using the 20CRv2 reanalysis; (e) use of the Coastline Evolution Model, CEM (Ashton and Murray, 2006) forced with the resulting wave climate.
Comparing the average wave climate for the whole century with different durations and time spans, we will look for the likely interannual and interdecadal shifts in the climate forcing. We will then run model experiments to see what sort of coastline responses we would expect to have occurred over the century, comparing with the results of Moore et al. (2013).
