FloodTool

FloodTool is a very lightweight plug-in for QGis geographical information system. QGis is a very complex piece of free software with a lot of frameworks that makes the platform very extensible and, most importantly, we can reuse a lot of available python code. Extensibility can be achieved by means of algorithms, models and plug-ins. FloodTool makes extensive use of the three, algorithms to preprocess layer fields and geometry, models to define workflows of different algorithms and, finally, the real plug-in to perform the actual computation of flood level.

This plug-in has been developed under MyCOAST project.

Figure 1: A screenshot of FloodTool in QGIS

Requierements

The tool have several prerequisites:

• QGIS: QGIS 3.10 LTR is recommended but the plug-in works from 3.8 and up. In Windows operating system, it is necessary to install it from OSGeo4W.
• Python modules: The plug-in makes extensive use of several libraries to perform data access tasks. A recent netCDF4 and cftime.

Installation

FloodTool is available for Linux and Windows operating systems. We recommend having a version of QGis 3.8 or higher installed. In Windows operating system, it is necessary to install a copy of QGis from OSGeo4W. Installation on Linux operating system is straightforward as far as all Linux distributions have latest versions of both, software and libraries.

To install the program under Windows, visit this link and choose the option to download it from OSGeo4W. When executing the “setup” of the OSGeo4W we must follow these steps:

Figure 2: QGis setup from OSGeo4W.

It is important that you mark, download and install all software packages that shown in 2 (c). Once the packages to be installed have been selected, they will be downloaded and installed automatically on our computer. This task will take several minutes (depending on the network download rate). Once finished, and without obtaining any warning message, the QGis will be installed on our system. Next, several dependencies must be satisfied in order Qgis to have access to OpenDAP data servers. To fulfill the requirements we recommend the following steps:

• Open OSGeo4W Shell

• Path update, using “py3_env.”

• Install “pip” command:

  • Download “get_pip.py” from https://bootstrap.pypa.io
  • Put he file in OSGeo4W folder and execute from Shell (python get_pip.py). This program will download and install pip, setuptools and wheel.
  • Update the paths, again, with “py3_env.”

• From https://www.lfd.uci.edu/~gohlke/pythonlibs/ download the following libraries:

  • cftime-1.0.4-cp37-cp37m-win_amd64.whl or later.
  • netCDF4-1.5.3-cp37-cp37m-win_amd64.whl or later

• Install these packages with pip install package.whl

Once NetCDF and cftime are installed we can install the plug-in itself. Clone the code tree or download the zip file from this Github repository. Look for the QGis plug-in folder (in Windows systems usually C:\OSGeo4W64\apps\qgis\python\plugins) and copy the plug-in code to this location. In QGis program, in tab “Complements/Manage and install plugins” looking for “FloodTool” and activate it. If you are able to see a connector icon at the toolbar, congratulations, you have the plug-in installed.

Usage

FloodTool plug-in operates over an specific GIS type of vector layer with several fields and metadata. Several tools (algorithms and models) are provided to accomplish the task of layer preparation but these tools are not covered in this document. So, in order to use the plug-in, a project with input layer is needed. Additional layer, raster or vector, can be added to the project and you can add symbols as needed as in a normal QGIS project.

• To activate the plug-in, find the plug button or select the plug-in at the plug-in menu (3). Please, make sure you have your input layer is selected. So far, no control over input layer is performed and an exception will be raised if selected layer is not the right one. The plug-in docked window appears with three different tabs:

  • Layers where kind of output layer is selected
  • Hydrodynamic where hydrodynamic model is selected
  • Wave where wave model is selected.

Figure 3: FloodTool main window.

• The first step is filling in the Hydrodynamic tab selecting the hydrodynamic model (Model grid combo box). After model selection, calendar is updated accordingly with available files on THREDDS server. The attribute of active layer that contains the nearest neighbour assignation and netCDF standard_name attribute of variable you want to use must also be filled (5). So far, model grids are hard coded into the plug-in.

Some hydrodynamic models have a tidal solution that can be used to synthesize time series of astronomical tide level only (4). These models contains a subset of principal tidal armonics, so knowing the list of armonics included in tidal signal is posible to adjust the model output timeseris to a sum of tidal armonics of the form:

by means of least squares method, where A_i, ω_i and ϕ_i are amplitudes, frequencies and phases of different tidal armonics present in the tidal signal. If the selected model grid has a tidal solution, the check box will be enabled. Make sure you check it if you want to use it.

Figure 4: Synthetic astronomical tide calculated from ROMS tidal solution.

Figure 5: Hydrodynamic model selection.

• Fill in the Wave tab in the same way you filled Hydrodynamic tab. Data sources for Hydrodynamic and Wave model can have different time spans. The plug-in will select the overlapping time span to perform actual calculations. Once hydrodynamic and wave models are configured, we need to specify the kind of output we want. So far, several options are available (6). Time control options indicate if you want to compute Max level or a Timeseries of expected flood level. Additionally, you can compute the real level or use a “traffic light”-like color codes. For this last option to work you will need an input layer with flood level return periods.

Figure 6: Wave model selection tab.

• Once configured, Run button will be enabled. Click on it to start with data download, perform calculation of flood level and categorization of output. Progress bar will evolve accordingly (7).

Figure 7: Running the configuration and output layer.

• After running, and Output vector layer will be created with several attributes that can be used to define a QGIS symbols. If you choose Timeseries as Time control option, your output vector layer has an additional Datetime field called time that can be used to animate the output with an optional plug-in called Time Manager. Installation of this plug-in is very straightforward (8).

Figure 8: TimeManager plug-in output animation.

Credits

• author:
  • Pablo E. Carracedo García (DXCASCC, Consellería Medio Ambiente Territorio e Vivenda, Xunta de Galicia, pablo.enrique.carracedo.garcia@xunta.es)
• license: Copyright (c) Xunta de Galicia 2020. Licensed under MIT
• funding: MYCOAST Interreg Atlantic Programme, Project nr. EAPA 285/2016 http://www.mycoast-project.org