Name: Grzegorz Różyński
room: A07
phone: (+48)585222907
e-mail: c8193953e7a4beaa12d8468e80164266eebbccad8ec0
ORCID: 0000-0001-9822-9847
Projects:
Conferences:
Różyński G.
Coastal nearshore morphology in terms of large data sets
publisher: Wydawnictwo IBW PAN
ISBN: 8385708510
published in: 2003
language: en
Summary: State-of-the-art techniques of data-driven analysis of large data sets, reflecting morphological variability of coastal nearshore seabed topography and shoreline evolution are presented in the book. They are aimed at extraction of key spatial and temporal patterns, which identify and describe physical processes and phenomena in studied coastal segments. The patterns include effects of all interactions and feedbacks amongst various forcings and seabed response and being their unbiased mapping they allow for in-depth investigations of main components that shape morphological evolution of the studied areas. These methods were applied to large morphological data sets collected at the IBW PAN Coastal Research Station at Lubiatowo over many years. Canonical correlation analysis (CCA) was used to assess impacts of the evolution of outer bars onto the variability of inner bars. Singular spectrum analysis (SSA) was employed to extract forced and chaotic elements of shoreline evolution. Principal oscillation patterns (POP) were utilized to build a data-driven model of nearshore bed topography evolution and a model of evolution of shoreline change trends. The role of synergy, resulting from the analysis of one data set with various methods was highlighted, which points to an optimum strategy of intensive statistical investigations of coastal data. Moreover, several variants of empirical orthogonal functions (EOF), applied previously to many analyses of large data sets, were also described. Hence, the book may serve as a manual on data intensive statistical modelling in coastal engineering.