Recapping the basics

Once you have been through the introductory tutorials (Getting started modelling in RiskScape, How to build RiskScape models, and Creating a RiskScape project) you will have learnt how to build models that identify elements-at-risk that were exposed to a hazard.

These tutorials cover a lot of material on things you might need to know about RiskScape modelling. However, you don’t have to understand everything covered in order to do simple modelling in RiskScape. This page is a recap on the basics required to put a simple model together.

If you are comfortable with the concepts on this page, you could try Going beyond simple RiskScape models.

General

When you install RiskScape, make sure that the riskscape executable is added to your PATH. Windows users might want to use a Desktop shortcut to do this.

You run RiskScape from a terminal (on Windows this will be the Command Prompt or PowerShell).
Use riskscape --version to sanity-check you can run RiskScape successfully.
It pays to always run the most recent version of RiskScape.

If you forget the exact RiskScape command to run, you can use riskscape --help, or add --help to the end of any riskscape command, to find out more.

Always run RiskScape commands from the directory containing your project.ini file.

Use the cd command to change to the correct directory.
Use the dir (Windows) or ls (Linux or Mac) commands to check what files are present (you should see a project.ini file in the output).

To run a model

To run a saved model, use the riskscape model run MODEL_NAME command.

To see the names of your saved models, as well as what model parameters they use, enter the riskscape model list command.

To change a model parameter on the fly, use the --parameter or -p CLI option on the end of the riskscape model run command. This is in the format -p "PARAMETER_NAME=NEW_VALUE".

Typically you might want to run the same model against different hazard-layers. To change the hazard-layer input used by your model, use:

riskscape model run MODEL_NAME -p "input-hazards.layer=MY_HAZARD.tif"

To create a project

You need to create a project.ini file before you can build a model. Try using the following template:

[project]
auto-import = true

[bookmark exposure_layer]
location = MY_BUILDINGS.shp

[bookmark hazard_layer]
location = MY_HAZARD.tif

[bookmark area_layer]
location = MY_REGIONS.shp

# function is optional - you could just use the built-in is_exposed function
[function exposure_status]
location = exposure_status.py
argument-types = [ building: anything, hazard: nullable(floating) ]
return-type = text

A simple example of a Python function (i.e. exposure_status.py) that you could modify to suit your model.

def function(building, hazard_depth):
    if hazard_depth is None or hazard_depth <= 0:
        return 'Not exposed'

    if hazard_depth > 2.0:
        status = 'Exposure >2.0m'
    elif hazard_depth > 1.0:
        status = 'Exposure >1.0m to <=2.0m'
    else:
        status = 'Exposure >0.0m to <=1.0m'

    return status

Replace the file locations with the input data you want to use. Make sure you save the file as project.ini and not project.ini.txt.

To build a model

To build a model, use the riskscape wizard command. At the end of the wizard, save the model so you can run it again later. To start with, we recommend choosing ‘closest’ sampling - it’s the simplest spatial sampling option.

The RiskScape wizard requires that bookmarks are present. But everywhere else you can usually use files directly instead of a bookmark. This works for shapefiles, GeoTIFF, GeoJSON, and ESRI grid (.asc).

Remember that you can build projects and models and then share these projects with other people. A user doesn’t necessarily need to know all the ins and outs of model building in order to run a RiskScape model.