Simultaneous Assimilation of Water Levels from Stage Gauges and Satellite Derived Flood Extents for Improving the Performances of a 2d Flood Routing Model
Abstract
Flood wave routing algorithms are effectively used within hydro-climatic extreme event for Early Warning Systems (EWS), allowing to simulate the inundation flow dynamics and forecast flooding scenarios. However, flood models are affected by several uncertainties due to the complexity of the represented physical phenomena, the simplifications of the model itself and the inaccuracy and scarcity of input data used for calibration and validation. To reduce the uncertainty of these models in near-real time, Data Assimilation (DA) frameworks are implemented supporting the update of the model state variables and/or their parameters using the available observations of the flood event. In this work, an Ensemble Kalman Filter (EnKF) is implemented in a DA framework to simultaneously assimilate the water levels both directly gathered from stage gauges and determined from the multispectral and SAR satellite images. The adopted DA framework includes a hydrologic model - simulating rainfall-runoff processes using a geomorphic approach (WFIUH) - producing the flood wave to be given as input to a Quasi-2D hydraulic model (FLO-2D) for the simulation of the flow wave propagation. The presented approach presents some novelties in the methodology of the water levels updating of the hydraulic model and in the methodology of the water levels extraction from the flood extent derived by the satellite image. The case study of the Tiber River basin is selected and preliminary results are presented showing the improvements on the performances in terms of predicted water levels and flood extension. This work aims to investigate the potential use of distributed observational information, coming from different sources, to overcome the limitations related to the exclusive use of stage gauges or satellite images as observations for updating the hydrologic-hydraulic models within DA-based EWS flood risk mitigation platforms.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.H31J1841A
- Keywords:
-
- 1847 Modeling;
- HYDROLOGY;
- 1855 Remote sensing;
- HYDROLOGY;
- 1910 Data assimilation;
- integration and fusion;
- INFORMATICS