InaSAFE was designed to predict the result of a disaster by giving us the potential impact on buildings, populations and roads based on specific scenarios. From previous exercises, we have learned how to use Basic and Intermediate InaSAFE functionality. We also learned about using tools in InaSAFE such as define keywords, configure minimum needs and area analysis, using aggregation as well as many others.
In previous exercises, we used the “flooding in Jakarta” scenario and related data to learn InaSAFE functionality. In addition to floods, InaSAFE impact analyses can be run for many hazard types including earthquakes, tsunamis and volcanos. In this exercise, we will explore and learn to run other hazard scenarios in InaSAFE using InaSAFE dock and InaSAFE Function Centric Wizard. Last but not least, we will also use the Generic Impact Function.
To improve the participant’s understanding of how to use InaSAFE to run impact analyses for hazards other than floods. By the end of this exercise, participants will be able to:
The data used for this exercise is the same as that used in previous exercises. The data can be downloaded from InaSAFE Training Data Packages Once the data has been downloaded, we will use the following QGIS project file and spatial data:
Indonesia’s location on the edges of the Pacific, Eurasian, and Australian tectonic plates makes it not only a site of numerous volcanoes but also frequent earthquakes. The hazard layer we are going to use for this example has been provided by Badan Geologi and AIFDR, Australian Government and describes the shaking or Modified Mercalli Intensity (MMI) Scale.
This particular scenario is a modelled version of the 2009 Padang earthquake. Please open QGIS project file Padang.qgs from the InSAFE Training Data > West Sumatera folder. Once opened, you will see the below :
In the InaSAFE dock, note that the InaSAFE form is still empty. It means that your hazard and exposure data does not yet have keywords assigned. Therefore, you must first define keywords for each dataset in the project.
To define keywords, please click on the Keywords Creation Wizard icon and follow the instructions provided. You can refer back to the Run Intermediate InaSAFE module for step-by-step instructions.
While you define the hazard layer, please select hazard classifications for each exposure type. In this example, you must choose ‘Earthquake MMI Classes’ on Population and ‘Generic Classes’ on Structures Classifications to edit the threshold. Do not forget to save the thresholds before you continue to the next step.
Once you have finished defining keywords for each layer, your InaSAFE form should look like this :
Once your InaSAFE dock appears the same as the above image, you are ready to run an earthquake analysis on buildings. It poses the question “In the event of an Earthquake in Padang 2009”, how many buildings will be affected?. Click Run on the bottom right corner in your InaSAFE Dock. 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 layer is named Number of Buildings.
Let’s take a look at the new data layer generated from InaSAFE.
Note
Default Threshold for Earthquake are 0-6 MMI for Low Threshold, 6.1-8 MMI for Medium and more than 8 MMI for high threshold. You can change the threshold of MMI Value for each affected area before running InaSAFE. Please run keyword wizard again and click Edit on Earthquake on Building. Don’t forget to click Save. This configuration will make your result different with Run InaSAFE using default threshold.
In the InaSAFE panel we now see the impact summary. Details are explained below.
We are now ready to run our second InaSAFE analysis using earthquake data in Padang. We will be working with the same earthquake data again, but this time looking at the number of impacted people in a specific area. If you finished defining keywords, these data should have keywords assigned so you will be ready to run InaSAFE.
In QGIS, turn off the Buildings and estimated buildings affected (the layers generated from InaSAFE analysis and turn ON the population layer). Because we want to look at the number people who might be killed or displaced in a specific area, we also need to turn ON the Village layer in QGIS. This layer will be used as an aggregation layer that can show us the result for each administrative area. If you forget the steps needed to define a layer as an aggregation, please refer to the Run Intermediate InaSAFE.
Confirm that the InaSAFE panel on the right side is set to query how many people will be affected :
Your InaSAFE form should appear like the below screenshoot:
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 an Earthquake in Padang (2009), how many people will be affected?” In this analysis we still use Shakemap data which has values from 6 – 8 MMI. If you want to see the minimum relief needs that should be provided based on the result, you can click Options and select Minimum Needs.
You can refer to the Run Basic InaSAFE to learn more about the basis of default minimum needs in InaSAFE or if you want edit an item or add new minimum needs, you can refer to the Minimum Needs Configuration manuals. After everything is setup accordingly, click Run to process the new scenario.
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 People Displaced. Let’s explore the result again to help you better understand its meaning :
In the InaSAFE panel we now see the impact summary. The details are explained below.
The 1992 Flores earthquake occurred on December 12, 1992 on the island of Flores in Indonesia. With a magnitude of 7.8, it was the largest and also the deadliest earthquake in 1992. This earthquake triggered another hazard in that area a tsunami in Maumere, Flores.
Next, we will run another scenario in InaSAFE using Tsunami Hazard Model. It is a modelled version of a Magnitude 8.1 earthquake generating a tsunami which impacts Maumere.
Please open the QGIS project file Maumere.qgs from the InaSAFE Training Data > Maumere folder. Once opened, the project should appear similar to the screenshot below:
You will see in the InaSAFE dock that keywords for each layer have not yet been defined. As before, we use the Keyword Creation Wizard icon to define keyword. For detailed steps, please reference the Run Intermediate InaSAFE. module
Once you have finished defining keywords for each layer, your InaSAFE form should look like this :
In this scenario we will use Buildings as an exposure and Village Boundary as an aggregation layer. Once your InaSAFE form appears the same as the above screenshoot, click Run at the bottom right corner in your InaSAFE dock.
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 Number of buildings. Let’s explore the result again to help you understand more.
Here we clicked on one of the red building and find that there is a value of hazard class ‘high’. This means that the building is located in an affected area which is predicted to be flooded more than 3 metres.
In the InaSAFE panel we now see the impact summary. Details are explained below.
The results show the buildings that will be affected by tsunami starting from 1 metre. What if the disaster manager decides that buildings in 80 cm of water are also considered inundated? You can change the water depth threshold to see the result, refer to the Run Basic InaSAFE. module.
Note
InaSAFE Tsunami Impact Function is very similar with Flood, but due to the force of tsunami waves, the maximum depth of water that would affect people and infrastructure is shallower.
Indonesia has many volcanoes, and most of them are still active today. In fact, one of the most frequent disasters in Indonesia is volcano eruptions. There are 129 active volcanoes across the country, and it is valuable to know how many people and how much infrastructure is within a certain perimeter of the vent.
InaSAFE also has an impact function for volcano eruption scenarios. This function can run some type of hazard data. For detail information about the InaSAFE volcano hazard function, please look at the Hazard Data Section.
In this section we will be using the Sinabung volcano hazard from the National Disaster Management Agency (BNPB) as hazard data and building from OpenStreetMap as exposure data. For this run, we will using Impact Function Centric Wizard (IFCW). For more information about IFCW you can refer to Key concepts in disaster management planning with InaSAFE.
Please open a new QGIS project in order to use IFCW to run this project. You new QGIS project should look like this:
To use the Impact Function Centric Wizard, please click Impact Fuction Centric Wizard icon.
After clicking that icon, you will see the following dialogue box appear:
In the box above, there are some fields that help us select the scenario to use. Green fields mean those scenarios are available and ready to run in InaSAFE. Grey fields means those scenarios are not available in InaSAFE at the moment.
Because we want to run Volcano with building in this session, please click Field Volcano’ and :guilabel: ‘Structure. The resulting dialogue box appears like this:
You can click Next and select the geometry types for the hazard and exposure layers, click Polygon.
You can click Next and follow the instructions in the IFCW box.
Hazard Data that we want to use for this scenario can be found in InaSAFE Training Data > Sinabung > Hazard Data folder and please select Sinabung_Hazard_Map_2015_WGS84.shp.
Building Exposure data can be found in InSAFE Training Data > Sinabung > Exposure Data folder and please select Sinabung_buildings_WGS84.shp .
There is optional to use aggregation for your analysis, either you load aggregation layer from your own local file or you just aggregate results for the entire analysis windows.
Note
The differences between Volcano and Volcanic Ash can be seen in Hazard Data Section, and for detail explanation about type of data you can be found in Key concepts in disaster management planning with InaSAFE..
If you have followed the instruction in IFCW box, before running InaSAFE you should see the final form below:
If your IFCW box looks like the screenshot above, click Run and wait for analysis processing until a result box appears.
Once you have finished running the analysis, you will see the result has new layer named “Number of buildings” . This layer will show you which buildings are affected for each hazard zone. There will be four new colours of building generated from InaSAFE (light yellow, orange, red, and grey).
The light yellow buildings mean these buildings are located in a Low Risk Zone (Risk Zone 1), the orange buildings mean these buildings are located in an Intermediate Risk Zone (Risk Zone 2), red buildings are considered situated in a High Risk Zone (Risk Zone 3) of Sinabung Volcano, and grey buildings considered not exposed by the hazard.
In the InaSAFE result box, we now see the impact summary. It is very similar to the InaSAFE Dock. Details are explained below.
You have now run InaSAFE for a volcano scenario using the Impact Function Centric Wizard (IFCW). Quite different than the InaSAFE Dock, this tool is designed to help a user run InaSAFE more easily without needing to open all the required data one by one in QGIS. IFCW provides precise guidance and detail about what actions should be taken step by step until the scenario is ready to run. This tool is very useful, especially for InaSAFE user who are less familiar with QGIS and Spatial Data.
InaSAFE can run analyses for multiple hazards, using scenarios that we set up based on data availability. These scenarios include seven types of hazards : floods, earthquakes, volcanos, volcanic ash, tsunamis, dam break, and, cyclone. What if our hazard scenario is not included in this list (for example, a land slide or drought). To solve this problem, InaSAFE provides a tool called the Generic Hazard Impact Function that can run analyses for any hazard not available via a specific Scenario Impact Function.
Next, we will explore this tool using a landslide hazard in Nagekeo, East Nusa Tenggara with building and population data for each scenario. Please open the QGIS project file Nagekeo.qgs from the InaSAFE Training Data > Nagekeo folder. Once opened, the project should look like the screenshot below:
In the InaSAFE dock, the InaSAFE form still appears empty. This means that your hazard and exposure data lacks keywords. Before proceeding, you should define keywords for each dataset in this project.
To define a keyword, please click on the Keywords Creation Wizard icon
and follow the instructions. You can refer to the Run Intermediate InaSAFE. if you need additional assistance. Once you finish defining keyword for each layer, your InaSAFE form should appear like the below screenshot:
Your InaSAFE Dock now poses the question “In the event of Landslide Hazard Zone, how many buildings will be affected?” Click Run on the bottom right corner in your InaSAFE Dock. 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 layer will be named “Number of buildings”.
Let’s take a look at the new data layer generated from InaSAFE:
In the InaSAFE panel we now see the impact summary. Details are explained below.
Turn off the Buildings Affected by each hazard zone (the layers generated from InaSAFE analysis and turn ON the People layer. Since we want to look at the number of people who might die or be displaced in a specific area, we also need to turn ON the Village layer in QGIS. This layer will be used as an aggregation layer that can show us results grouped by administrative boundary.
If you forget the steps to define an aggregation layer, refer to the Run Intermediate InaSAFE. section. Edit the question form in the InaSAFE Dock so that it appears similar to the below screenshot:
Next, click Run to start the analysis.
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 Number of People. Let’s explore the result again to help you understand more.
Here we clicked on the red zone and find that there is a value of hazard class ‘high’. This means that the people is located in affected area which is predicted as high hazard zone.
In the InaSAFE panel we now see the impact summary. The details are explained below.
Note
In the result of InaSAFE, Action Checklist and Notes might be unrelated with hazard that we run. For instance, if we run drought hazard the action checklist might have some topics such as how many building closed, or people die or displaced. Those topics are not really related with drought.
Congratulation! You have now learned to use most of InaSAFE‘s functionality. You can run analyses for specific hazard using tools such as the Impact Function Centric Wizard (IFCW) and the Generic Impact Function which will make using InaSAFE easier.
Now, to become an expert user of InaSAFE, try utilising all those tools that you explored in this module using your own scenarios and data, and practice interpreting the results.