IC3_I

Introduction
Bedform dynamics is ruled by the mutual adjustment of hydrodynamics, topography and sediment transport. A number of morphological irregularities, i.e. bed defects (cf. Huntley et al., 2008), occur over bedform fields, and have been detected and characterized both in subaqueous environments and in aeolian systems (Kocurek et al., 2010), as well as in flume experiments (Venditti et al., 2005). Among them, bedform crest bifurcations can be commonly found in bedform fields. The knowledge of the mechanisms behind bifurcations can lead to a better understanding of bedform pattern organization and bedform dynamics.
Some attempts of explaining the occurrence and development of crest bifurcations have been made in the last decades (e.g. Aliotta and Perillo, 1987; Baas, 1994; Flemming, 2000; Dohmen-Jannsen et al., 2008), but none of them has led to a complete explanation of such a phenomenon. In fact, taking single process-response mechanisms into consideration for explaining crest bifurcations automatically excludes the generally accepted idea of a strict interaction and mutual influence between morphology, hydrodynamics and sediment dynamics. In addition, the extensive use of models and laboratory experiments, not always or properly supported by field data, do not allow to completely relate the results to natural features, as they only give an approximation. Furthermore, the lack of studies in tidal environments represents a gap in the general understanding of crest bifurcations.

Research questions and objectives
The general aim of the project is to contribute to the understanding bedform crest bifurcations, with a focus on bedform fields in tidal channels, in order to answer fundamental questions such as a) Where do crest bifurcations occur? Why do they seem to appear only at definite locations? b) How do bifurcations develop? c) Why do bifurcations occur at all?
Two main objectives have been delineated: 1) to understand how bifurcations evolve in space and time; 2) to identify the processes in charge for the generation of bifurcations. They have been pursued through a holistic and integrated approach, combining field observations and numerical simulations.
The working hypothesis is that bifurcations occur in response to variations in sediment composition and transport, hydrodynamics or morphology, or a combination of two or more factors. In particular, they might be related to threshold values of these parameters over, or above, which larger bedforms are not stable anymore and need to split into smaller ones (and vice versa). Different case studies have been selected for the scope, two tidal inlets in Denmark, where repeated surveys have been conducted over the past decade, and the Tauranga Harbour in New Zealand.

Results
The description and characterization of bedforms and their dynamics are based on the application of different methods in order to gain as much objective information as possible. Thus, continuous wavelet transform of bed elevation profiles was tested and performed along a 1600 m long transect across natural compound bedforms, using high resolution bathymetric data from the Grådyb tidal inlet channel (Danish Wadden Sea) from 2002 to 2009 (Fraccascia et al., 2011). The analysis revealed the consistency of the spectral patterns in the long time, with generally low coefficients of variation, despite morphological changes due to bedform evolution and migration had been documented in the area (cf. Ernstsen et al., 2005).
For a first characterization of crest bifurcations, the Knudedyb tidal inlet channel (Danish Wadden Sea) has been selected, having preserved its natural form and being not affected by human interventions. Bathymetric data collected in the Knudedyb tidal inlet channel show that the bottom of the channel is extensively covered by large to very large primary bedforms, several hundred meter long and several meter high, generally asymmetric and ebb-oriented, that bifurcate into smaller bedforms on the northern side of the channel. Geometrical descriptors of bedforms have been derived applying continuous wavelet analysis (Fraccascia et al., 2011) and zero-crossing techniques (cf. Ernstsen et al., 2010), as well as morphological indices described in the literature, to bed elevation profiles extracted from the multibeam echo sounding data. The morphodynamic processes of the Knudedyb tidal inlet are being investigated with the Delft3D-FLOW numerical model. Model results are being explored and compared with bifurcations.