Approach to Integrate Biophysical Components for a Hydrodynamic Model in a Tropical Wetland

In tropical wetlands, Intertropical Convergence Zone (ITCZ) modified the hydrological regime and pattern of grown vegetation, and this in turn impact in a hydrological budget and friction force in the hydrodynamics. A complex wetland in lowlands has a particularity that has multiple connections around rivers or swamp through natural channels, who depends the season could have a superficial connection. The Ayapel Swamp Complex (ASC) is a freshwater wetland located in the Caribbean Region of Colombia, in the southwestern part of the Momposina Depression, its basin has an extension of 1504 km².

The aim to model the ASC is integrated a hydrological, vegetation and hydrodynamic components for identifying the impact of the spatial-temporal dynamics in a delineation of the wetland. The implementation of the hydrodynamic model was developed in TELEMAC-2D that solves the two-dimensional Saint Venant equations with finite element scheme, supported by Blue Kenue software for the pre and post-processing stages. A pilot model was proposed that includes the water mirror of the swamp and a perimeter corridor in which are the channels that feed and drain the system. In this area it is observed that in large areas the terrain does not have altitudinal variations, slopes are lower (0-10 %), which makes it difficult to define the computational domain, the delineation of the channels and the construction of the mesh. The first results show numerical problems possibly related to the digital elevation model and to the refinement of the elements of the mesh. On the other hand, it has been evidenced that due to the low water velocities and the hydrological regime of the zone, simulations must be performed in windows of time superior to the daily, maintaining the stability of the model (CFL), which implies high computational time and the generation of large volumes of information, so that parallel computing in TELEMAC-2D has been shown to reduce computing time by up to 40%. It is concluded that for improving the performance of the model is necessary to implement the following procedures: 1) refinement of the topography, in this way to obtain better representations of the bed of the channels and floodplains, 2) identification of spatial patterns (geomorphological, vegetation and soil) that allow increasing the size of the mesh in areas where no detail is required, and 3) evaluated the uncertainties associated with the parameters.