Use of satellite altimeter data for validating large

  • Slides: 24
Download presentation
Use of satellite altimeter data for validating large scale hydraulic models Matt Wilson, University

Use of satellite altimeter data for validating large scale hydraulic models Matt Wilson, University of Exeter Doug Alsdorf, Ohio State University Paul Bates, University of Bristol Matt Horritt, University of Bristol

Hydraulic modelling of floodplains • Flow on floodplains is controlled by topography and friction

Hydraulic modelling of floodplains • Flow on floodplains is controlled by topography and friction • Leads to complex spatial patterns of water depth and velocity that are 2 D in space and dynamic in time • Until recently modelling of such flows has only been possible for small river reaches (10 -50 km)

New opportunities • Large scale modelling has now been made possible by: – Simplified

New opportunities • Large scale modelling has now been made possible by: – Simplified 2 D hydraulic models – Faster computers – New satellite data sources e. g. SRTM, satellite radars

LISFLOOD-FP • Hybrid 1 D/2 D model – Based on raster DEM – 1

LISFLOOD-FP • Hybrid 1 D/2 D model – Based on raster DEM – 1 D Kinematic or diffusion wave routing in channel – Once bankfull depth is exceeded calculates a flux to floodplain cells using Manning’s equation or 2 D diffusive wave to route water over complex floodplain topography Model discretization of floodplain and channel topography In-channel flow routed using a 1 D wave equation Once bankful depth is exceeded water can flow laterally over adjacent low lying floodplains according to topography and free surface gradient

LISFLOOD-FP • Inundation is based on a simple continuity equation: where: V = cell

LISFLOOD-FP • Inundation is based on a simple continuity equation: where: V = cell volume t = time Qup, Qdown, Qleft and Qright = flow rates in each direction into (positive Q) and out of (negative Q) the cell

LISFLOOD-FP i j • Flux between cells calculated using the Manning equation where: Qi,

LISFLOOD-FP i j • Flux between cells calculated using the Manning equation where: Qi, j = Ai, j = Ri, j = Si, j = n = flux between two cells i and j, cross sectional area at the cell interface, hydraulic radius at cell interface, water surface slope between cells, Manning friction coefficient.

Simplified 2 D models • Advantages – Floodplain flow is solved analytically rather than

Simplified 2 D models • Advantages – Floodplain flow is solved analytically rather than numerically so very efficient • 150 -500 k cells for full dynamic events should run in less than 1 day on a pc – Can use large elements (e. g. 250 m – 1000 m grids) – Intrinsically mass conservative treatment of floodplain flow • Disadvantages – Simplified floodplain flow representation – Wetting front propagation may be grid and time step dependent

LISFLOOD-FP • Successfully applied to reaches 10 -50 km in UK and continental Europe

LISFLOOD-FP • Successfully applied to reaches 10 -50 km in UK and continental Europe • We are now applying LISFLOOD-FP to 100 -500 km reaches

Solimões inflow Purus inflow 100 km

Solimões inflow Purus inflow 100 km

Amazon model implementation • 200 x 280 km model domain, Purus and Solimões as

Amazon model implementation • 200 x 280 km model domain, Purus and Solimões as 1 D channels – 90 m SRTM DEM averaged to 270 m, processing for vegetation – Around ~800, 000 cells – Simplified channel information • 4 year simulation, each annual hydrograph is the average of 20 years of gauged data

Comparison with JERS-1 imagery

Comparison with JERS-1 imagery

Topex/POSEIDON altimetry in Amazon region

Topex/POSEIDON altimetry in Amazon region

Altimetry data for study site

Altimetry data for study site

Model vs. altimetry

Model vs. altimetry

Model vs. altimetry

Model vs. altimetry

Model vs. altimetry

Model vs. altimetry

Floodplain flow, mid-rising

Floodplain flow, mid-rising

Some conclusions • Comparison of inundation extent using JERS image is useful but uncertain.

Some conclusions • Comparison of inundation extent using JERS image is useful but uncertain. • Altimetry data are a welcome addition, but miss the spatial complexity floodplain flows. • To capture these complexities, a twodimensional approach is required.