../../_images/inasafe_logo.png

Run Basic InaSAFE

Introduction

In this exercise we will work through an example scenario where we show how the different data elements used by InaSAFE are combined in order to analyse the potential impact of a flood in Jakarta on both the buildings and the population.

After we have run the InaSAFE analysis we will print the map and analysis report as pdf and review the results. We will also learn how to change the flood threshold and take a look at the default settings for minimum needs. We will also learn how to save our work.

Learning objective

To develop the participant’s basic understanding of the InaSAFE workflow and application of InaSAFE in the Disaster Management sector. By the end of this exercise, participants will:

  • Be able to run a flood analysis using InaSAFE - on buildings;
  • Be able to run a flood analysis using InaSAFE - on population;
  • Understand the flood impact default settings;
  • Understand the impact summary report;
  • Be able to change the analysis threshold and run a new scenario;
  • Be able to generate a PDF map from the results of an analysis; and
  • Be able to save their work to share results with others.

Data for this exercise

The data for this exercise are available in Run Basic InaSAFE zip which can be downloaded from InaSAFE Training Data Packages . We will use the following QGIS project file and spatial data:

  1. DKI_Jakarta_Basic.qgs
  2. Jakarta_Flood_HKV_WGS84
  3. Jakarta_Buildings_WGS84
  4. Jakarta_Population_WGS84

Exercise

1. Open QGIS Project

Before we can run an InaSAFE analysis, we must open a QGIS project. Please open the QGIS project file DKI_Jakarta_Basic.qgs from the InSAFE Training Data > DKI Jakarta folder. The project looks something like this:

../../_images/run_basic_00.png

As you can see from the picture above, you will be presented with several data sets for Jakarta such as buildings, population, and raster hazard.

You will see that the project has three layers loaded:

  • Buildings: this is exposure data. We will use these to assess the flood impact on buildings.
  • Flood: this is the hazard model. We will use this to determine the depth of water.
  • Population: this is a population raster. We will use it to determine the number of people exposed to the flood.

More detailed information about the data used in this exercise can be found in dataset section. The table below provides a brief summary about the source of the data.

Data Source Description
Buildings OpenStreetMap Most of the important buildings in Jakarta have been mapped through the collaboration of BPBD DKI Jakarta, OpenStreetMap and the Australian Government. See more at..
Jakarta_population_WGS84 WorldPop High resolution, modelled data for human population distributions. See more at..
a Flood similiar to 2007
Jakarta Event
HKV The flood model was created by scientists/engineers in coordination with DKI Jakarta Public Works based on the 2007 flood conditions. See more at..

Let’s move into the next section where we will run our first InaSAFE analysis using these data. We will be working with the flood hazard model to look at the number of affected buildings. These data already have keywords assigned so we are ready to run the analysis.

2. Run InaSAFE Analysis for Building

Take a look at the InaSAFE dock on the right side of QGIS. The InaSAFE dock should show that you are ready to run a flood analysis on buildings. It poses the question “In the event of a flood similar to the 2007 Jakarta event, how many buildings might be flooded?” In this analysis we will use the default flood depth threshold of 1.0 metre. Later on we will learn how to change the threshold.

a. Run InaSAFE Analysis

../../_images/run_basic_01.png

Click Run in the lower right corner of the InaSAFE panel to start the analysis process. If everything was set up correctly, you should get a result in the dock area after a few seconds, and a new map layer should be added to the map.

../../_images/run_basic_02.png

b. Interpret the results

The new impact layer will be generated and called Estimated buildings affected. Let’s take a look at the new impact layer generated by InaSAFE.

  • Zoom in to some area on the map canvas
  • Here we have zoomed in to a location showing two rivers going through the middle of Jakarta. There will be three new different colours generated from InaSAFE (green, orange, and red).
../../_images/run_basic_03.png

Note

If you don’t see these colours, you might need to turn off the data layer above the Estimated buildings affected layer.

  • The red buildings are situated in water greater than one metre, the orange buildings are situated in water between zero and one metre, while the green buildings are considered as unaffected as they are situated in dry areas according to the flood hazard model.
  • Click Estimated buildings affected in the layer list to select it and click Identify Feature tool and then click on a building to view the attributes of that building.
../../_images/run_basic_04.png

In the InaSAFE panel we now see the impact summary. The details of are explained below.

../../_images/run_basic_05.png
  • Hazard Category: divides the results into several categories based on the threshold set in the hazard analysis. In this impact summary, InaSAFE divides the impact buildings into three categories: number of buildings inundated (buildings affected by water deeper than the analysis threshold), number of wet buildings (buildings affected by flood water but not as deep as the analysis threshold), and number of dry buildings (buildings that are not affected by any flood water)
  • Building type: divides the exposed buildings into several categories based on the building type attribute for each building. In this impact summary, InaSAFE breaks down the results into a more detailed report by looking at each type of the building, for example the number of inundated hospitals and the total number of hospitals in analysis area.
  • Action checklist: designed to make disaster managers think about what they need to do/discuss when planning for a similar event in the future.
  • Notes & assumptions: provides details about the input data and any limitations or assumptions in the analysis or report summary. In this example, it explains why buildings are said to be inundated, wet and dry.
  • Detailed building type report: statistical breakdown of the building types affected by the simulated event. When you choose to use an aggregation layer with your analysis (we will do this later) this table will show the number of buildings by aggregation boundary.
  • Hazard details: explains where the hazard data come from
  • Exposure detail: explains where the exposure come from

The results show the buildings that will be affected by flood water 1m deep. But what if the disaster manager decides that buildings in 80cm of water are also flooded? In order to assess this new scenario, we need to change the water depth threshold at which buildings are considered to be inundated. With InaSAFE it is easy to run a new scenario, all you need to do is change the Thresholds [m] in the Options tab to 0.8 and run the scenario again. We will do this next.

c. Changing threshold

In the Jakarta flood scenario we are running; the threshold refers to the depth of water that a disaster manager decides is the boundary between buildings being flooded (affected) and buildings not being affected.

Note

You can only change the threshold for raster hazard data.The default threshold for this hazard is 1m or 100cm.

If you want to open the InaSAFE question panel again, click on Show question form at the top of the InaSAFE panel. You will see the InaSAFE question panel again and you can click the Options button next to be flooded.

../../_images/run_basic_06.png

It will open the InaSAFE impact function configuration.

../../_images/run_basic_07.png

Here you can change the threshold of the flood according to your needs. In this example we change it to 0.8m. After you change the threshold to 0.8, click OK to close the dialog and then run the analysis again to see the change in the results.

When the function completes, take a look at the impact summary in the InaSAFE panel. How do the results compare to the first analysis results? The result should be different to the first analysis because in the first analysis InaSAFE buildings are said to be inundated if the flood level exceeds 1.0m and now we have changed the flood level to 0.8m. This means that buildings are said to be inundated when the flood level exceeds 0.8m. By reducing the flood threshold value from 1.0 to 0.8, more buildings will be considered inundated because a greater area of Jakarta is flooded at this depth.

Note

ask your tutor to explain if you do not understand this.

This completes our first InaSAFE analysis using the flood hazard model to look at the number of affected buildings.

3. Run InaSAFE for population

We are now ready to run our second InaSAFE analysis using the flood hazard data for Jakarta. We will be working with the flood hazard model again, but this time to look at the number of impacted people. These data already have keywords assigned so we will be ready to run the analysis as soon as we have turned on the relevant data layers.

In the QGIS, turn OFF the Buildings and estimated building affected (the layers generated from InaSAFE analysis and turn ON Jakarta_Population_WGS84 layer.

Confirm that the InaSAFE panel on the right side is set to query how many people might need evacuation:

  • A flood similar to the 2007 Jakarta event
  • People
  • Need evacuation
../../_images/run_basic_08.png

a. Run InaSAFE Analysis

If everything is setup correctly, the InaSAFE dock should show that you are ready to run a flood analysis on population. It poses the question “In the event of a flood similar to the 2007 Jakarta event, how many people might need evacuation?” In this analysis we will use the default flood depth threshold of 1.0 metre to find out how many people are in 1 metre of water. After everything is setup accordingly click Run to process the new scenario.

Note

Notice that if you click on the drop-down list on “How Many People, the building option is not available. This is because building is not checked in the Layers panel.

b. Interpret the results

If everything was set up correctly, you should get a result in the dock area after a few seconds, and a new map layer should be added to the map. The new impact layer will be called population which need evacuation. Let’s explore the result again to make you understand more about the InaSAFE result.

  1. Turn off Estimated buildings affected layer and drag the population which need evacuation above a flood similar to 2007 Jakarta event
  2. Zoom in to an the area you choose
  3. Select population which need evacuation in the layer list and use Identify Feature tool again to select a pixel (square) in the map canvas.
  4. Here we clicked on one of the light green pixels and find that there is a value of 80.75106, which means there are approximately 80 people in this one pixel (square) who need to evacuate because of the flood.
../../_images/run_basic_09.png

In the InaSAFE panel we now see the impact summary. The details of this summary are explained below.

../../_images/run_basic_10.png
  • Population needing evacuation: InaSAFE estimates the number of affected and unaffected people from the total number of people in the analysis area. It is assumed that all of these affected people will need to be evacuated.
  • Evacuated population minimum needs: these are the calculated amounts of food, water and other products that are needed by evacuated people. These needs should be provided weekly.
  • Action checklist: designed to make disaster managers think about what they need to do/discuss when planning for a similar event in the future.
  • Notes and assumptions: provides details about the input data and any limitations or assumptions in the analysis or report summary. In this example, it shows the total number of population in the analysis area and the source of minimum needs.
  • Detailed gender report (affected people): provides a breakdown of the number of affected people by age (youth, adults and elderly) and gender based on the default world population demographics and calculates the minimum needs for women’s hygiene and pregnant women.
  • Detailed minimum needs report: provides a breakdown of the minimum needs for evacuated people based on PERKA No 7/2008. These minimum needs consist of rice, drinking water, clean water, family kits and toilets to be provided weekly.
../../_images/run_basic_11.png

c. Understand defaults minimum needs

The InaSAFE impact summary for flood impact on people includes details for the amount of drinking water, rice, clean water, and family kits and for the number of toilets that should be provided for displaced persons each week. The minimum needs in the Jakarta flood impact assessment are based on the Head of Indonesia National Disaster Management Authority, BNPB, regulation, PERKA No 7/2008 guideline procedure for fulfillment of basic needs in Disaster Response. The default minimum needs formula is:

  • 400g rice per person per day (2.8kg per week)
  • 2.5l drinking water per person per day (17.5l per week)
  • 15l clean water per person per day (105l per week)
  • one family kit per family per week (assumes five people per family which is not specified in perka)
  • 20 people per toilet

As described above, the impact summary and minimum needs calculation is based on the default world population demographics (which assumes a ratio of 26.3% youth, 65.9% adult and 7.9% elderly).

You may like to refer to local population statics (for example - Population of DKI Jakarta) to change these defaults for your analysis area, similarly if you have other regulation for minimum needs, you can change in the Impact Function Configuration in Minimum Needs Tab or if you want to create your own minimum needs, you can use minimum needs configuration (see more at Minimum Needs Configuration manuals).

../../_images/run_basic_14.png

Summary

In this exercise you have learned how to run a basic InaSAFE analysis using an existing QGIS project file and what the minimums component that must be there to run InaSAFE properly are. Those components are hazard and exposure data. In this exercise, you have run an InaSAFE impact assessment for a flood scenario in Jakarta using two types of exposure data. The hazard data you used was a modelled flood raster and the exposure data were buildings and population. These analyses produced impact layers and impact summaries for affected buildings and impacted people.

You have also learned how to modify the analysis options through the Impact Function configuration, how to print InaSAFE results in PDF format, understand what minimum needs is and how to save both your impact layers and your QGIS project file.

In the next section you will learn more about how to run InaSAFE in more detail. In that module you will learn how to use more InaSAFE tools such as Agreggation options, OSM Downloader, Minimum Needs Configuration, etc.