QGIS Trigger Workflow for Madagascar

French Translation

The french version of this page can be found here: La version française de cet article se trouve ici:

Version française

Download the data and project files

The QGIS project file, the model, and the datasets required for the model can be downloaded from HeiGITs Gitlab here:

The QGIS workflow presented in this article was developed in the framework of the Anticipatory-Action (AA) Project of the Croix-Rouge Malagasy (CRM), the German Red Cross (GRC) and the Heidelberg Institute for Geoinformation Technology (HeiGIT).

The workflow is almost fully automated through a QGIS model, requiring no manual intervention. The chapterFunctionality of the workflow outlines the process and its practical implementation. Each step included in the model is explained in detail to provide a complete understanding of the workflow and how the analysis was carried out.

Background

Setting triggers is one of the cornerstones of the Forecast-based Financing system. For a National Society to have access to automatically released funding for their early actions, their Early Action Protocol needs to clearly define where and when funds will be allocated, and assistance will be provided. In AA, this is decided according to specific threshold values, so-called triggers, based on weather and climate forecasts, which are defined for each region (see FbF Manual).

Trigger Statement

Pre-Activation Trigger: at least one of the meteorological forecasts from Météo Madagascar, RMSC La Reunion, or ECMWF projects a greater than 50% likelihood of landfall by a tropical cyclone of tropical storm strength or higher within the next 7 days.

Activation Trigger: if the Meteo Madagascar (DGM) forecast indicates landfall of a tropical cyclone with wind speeds in excess of 118 km/h within the next 48-72 hours.

Functionality of the Workflow

The Trigger Process concept is displayed in the figure below.

The provided QGIS project contains the necessary layers and a QGIS model file to perform an assessment of the potential impact of the predicted cyclone event. The analysis workflow will be run in the QGIS model, which automates the steps for assessing the impact of a tropical cyclone event. It integrates cyclone storm track data with administrative boundaries, population data, infrastructure, and service locations to identify and quantify exposed areas and resources. Based on the cyclone forecast by Météo Madagascar, the model calculates the area likely to be exposed to the cyclone, the potentially exposed population, number of exposed buildings, exposed agricultural land, and potentially exposed health and education facilities.

Additionally, the QGIS file includes layers with the CRM warehouses and the areas they can service, allowing for a quick accessibilty assessement. The provided folder also contains map templates and style files to generate map reports based on the model calculations.

The documentation is separated into two parts. The first part covers the spatial analysis using the automated QGIS model. The second part documents how to create the map reports using the map templates and style files.

Attention

The QGIS project and QGIS model have been created using QGIS version 3.40.9 (LTR) Bratislava. To ensure that the model is working correctly, do not use older QGIS versions.

Available Data

For the trigger mechanism to work properly we currently use different datasets: data that we assume to be static in the near term, and variable data which describe the datasets that will be checked for triggering on a regular basis depending on the occurrence of anticipated cyclone events.

Download the data and project files

The QGIS project file, the model, and the datasets required for the model can be downloaded from HeiGITs Gitlab here:

Download the project.

Downlaoding the project.

To download the project click on the

Fixed Data

By fixed data we mean datasets that are needed to create the map reports, that will most probably not change in the near term. In the long term these datasets can be adapted easily.

Dataset	Source	Descriptions
Administrative Boundaries	HDX	The administrative boundaries on level 0-4 for Madagascar can be accessed via HDX provided by OCHA. For this trigger mechanism we provide the administrative boundaries on level 1 (regional level) and 2 (district level) as a shapefile.
POI counts	HOT Export Tool	The POI data (education facilities and health sites) is downloaded using the HOT Export Tool based on OpenStreetMap data.
CRM Warehouses	Croix-Rouge Malagasy	This layer contains points representing the locations of the CRM warehouses
CRM Warehouse Isochrones	HeiGIT	Using the Global Friction Surface, we calculated the area which can be reached within a specific amount of time from a given warehouse by car.
Population Counts	WorldPop	The worldpop dataset in raster format provides the estimated total number of people per grid-cell for the year 2020. We will be working with the Constrained Individual countries 2020 dataset at a resolution of 100m.
Buildings Counts	Global ML Building Footprints	The building counts dataset in raster format counts the number of buildings per 100m grid cell. The workflow on how this dataset was created can be found on GitLab
Land Cover	Copernicus Land Cover	The land cover dataset in raster format provides an overview over the dominant land cover type at a resolution of 100m. The workflow on how this dataset was downloaded can be found on GitLab

Master Raster

The three raster datasets are combined into a Master Raster — a multi-band raster layer with a spatial resolution of 100 meters. This composite layer includes the following information across three channels:

1. Population counts per grid cell from Worldpop constrained (2020)

2. Building counts per grid cell derived from ML Building Footprints (2021)

3. Land Cover type per grid cell derived from Copernicus Land Cover (2019)

Monitoring Data

Attention

The forecast information will be sourced from DGM (Météo Madagascar), which will provide tropical cyclone track data for the trigger workflow.

For an analysis of past events, data provided by NOAA (National Centers for Environmental Information) can be used. The cyclone storm tracks are provided within the International Best Track Archive for Climate Stewardship (IBTrACS) project. It is the most complete global collection of tropical cyclones available and merges recent and historical tropical cyclone data from multiple agencies to create a unified, publicly available, best-track dataset. IBTrACS was developed collaboratively with all the World Meteorological Organization (WMO) Regional Specialized Meteorological Centres, as well as other organizations and individuals from around the world.

Cyclone tracks

Tropical cyclone track data is available in various subsets, depending on the temporal scale of interest. Regional subsets can also be generated, with data for the South Indian Ocean being particularly relevant for this trigger mechanism.

Estimating the impact of the cyclone using the QGIS model

As explained at the start of this chapter the developed trigger workflow is done automatically by a QGIS model. In this chapter we will explain its functionality and in a subsequent step it is explained how to run the automated model.

Functionality of the model

The following key processing steps are run inside the model:

1. Cyclone Buffering & Impact Area Extraction

   • The input cyclone track is buffered to create an estimated zone of impact. The buffer is dissolved to generate a single polygon representing the exposed cyclone area. This layer serves as the base for subsequent exposure calculations.

2. Exposed Administrative Units

   • The buffered cyclone area is intersected with district (Admin 2) boundaries to extract the exposed districts. These are further linked with regions (Admin 1) using the region (Admin 1) names attribute to structure exposed districts by region. This hierarchy is used for reporting and map layout purposes.

3. Population Impact

   • The model uses the population raster to calculate zonal statistics over the exposed districts. This determines the total population per district and the exposed population, which is then exported to a table.

4. Infrastructure Impact

   • The cyclone buffer is intersected with:

     • Buildings to extract exposed buildings.

     • Health sites and education facilities layers to extract and summarize exposed points of interest. These datasets are combined into a table, summarizing exposed infrastructure.

How to run the model

The QGIS Model Designer is a visual tool that allows users to create and edit a workflow with all tools available in QGIS that can be used repeatedly in a simple and time-efficient manner, while ensuring reproducibility. It provides a graphical interface to build workflows by connecting geoprocessing tools and algorithms. The user can define inputs, outputs, and the flow of data between different processing steps.

Step 1: Explanation of the folder structure

Purpose: This step outlines the recommended folder structure to simplify the analysis and ensure consistent, reproducible results.

Tool: No special tools or programs are needed

Instruction

Folder Structure

Open the folder “AA_Cyclone_Monitoring_Trigger_MAD”. Input data is located in the folder “fixed_input_data”. The QGIS model can be found in the “trigger_model” folder. Resources for styling and map creation are located in: layer_styles – predefined layer symbology logos_pictures – logos and visual elements map_templates – templates for final map layouts example_map_results – example outputs for orientation Save your own results in the “map_outputs” folder. To start the process, open the QGIS project file “AA_Cyclone_Monitoring_Trigger_MAD.qgz” by double-clicking it. This will launch the full analysis workflow.

Step 2: Open the project in QGIS and load the model in the QGIS Model Designer

In this step we will open our Trigger project in QGIS and load the QGIS model which will automatically run the analysis for us.

1. Open the file AA_Cyclone_Monitoring_Trigger_MAD.qgz by double clicking on it.

2. The QGIS project will open with lots of data pre-loaded. This data is required for running the QGIS model and create some output maps.

The data will be structured into five groups:

Group 1: Fixed_Input_data

This group contains all the fixed input data required to successfully run the model. These datasets remain constant and do not change between events. The only additional input needed is the storm track for the event under investigation, which should be added before this data group in the layer panel.

Attention

Always ensure you are using the most up-to-date storm track for the event being analyzed. To add the layer, simply drag and drop it into the Layers panel, placing it directly above the Fixed_Input_data group for clarity.

For better data management, give the storm track a descriptive name, such as storm_track_eventname_year (e.g. storm_track_freddy_2023). This naming convention helps keep your workspace organized and ensures the correct data is used during the analysis.

Group 2: Model_outputs

This group is used to organize all output layers generated by the model after it has been run. You can review the outputs here and identify which layers are relevant for specific maps. Once identified, move them to the appropriate map group for visualization and layout.

Group 3: Map_Cyclone_Impact_Overview

This group includes all the layers required to create the Cyclone Impact Overview Map (shown below). The storm track and region (Admin1) boundaries (Admin1_Impact_Overview_Map) are pre-loaded to help you get started quickly. Make sure you’re working with the correct and updated storm track for the event under investigation.

Fig. 56 This map will be created using the layers in group 3.

Group 4: Map_Cyclone_Impact_Assessment

This group contains all the necessary layers to generate detailed impact assessment maps. As with the overview map, both the storm track and region (Admin1) boundaries are pre-loaded. Ensure you’re using the correct event data to maintain consistency and accuracy in the assessment. In this section we can create 5 different maps for different impacts:

• exposed population

• exposed buildings

• exposed education facilities

• exposed health sites

• exposed agricultural land cover

The final map outputs will look like the following.

Fig. 57 This map will be created using the layers in group 4

Group 5: CRM_Warehouse_Isochrones

This group includes isochrones for all warehouses, calculated for time intervals up to 24 hours. These layers are useful for assessing accessibility of locations in emergency response planning. This group is used to create the CRM warehouse accessibility matrix map. It is also possible to add a specific warehouse isochrone to one of the previous maps. We will cover this further below.

---

Opening the model in QGIS

Let’s open the QGIS model:

1. In the tob bar of your QGIS window, navigate to Processing -> Model Designer. A new window will open. This is the model designer.

2. In the upper panel click Model -> Open Model and navigate to your folder “AA_Cyclone_Monitoring_Trigger_MAD/trigger_model”.

3. Select the “Cyclones_EAP_MAD_Trigger.model3” file and click on Open. The model will open and you will see yellow, white, green and grey boxes.

Fig. 58 Opening the graphical modeler in QGIS 3.44

Box	Significance	Description
Yellow	Model Input	Definition of the input data for the model the model will perform on.
White	Algorithms	Algorithms or Tools are specific geoprocessing steps that perform specific tasks, such as clipping, reprojecting or buffering.
Green	Model Output	The results created by the model (Output layers) are automatically added to your layers panel in your QGIS project interface.
Grey	Comments	The boxes are used to further explain the specific processes.

Step 3: Run the model

Model Inputs & Outputs

1. A QGIS model can be run by navigating to the top bar > Model (Modèle) > Run Model (Exécuter le modèle) or by clicking on the icon.

2. A new window will open. Here you need to define the model’s inputs and outputs. For each of these mandatory inputs, you click on the dropdown arrow and choose the respective file.

Fig. 59 Selecting the inputs and defining the outputs before running the model.

Note

In the dropdown list, only layers that are currently loaded in your QGIS Project will be displayed.

3. To run the model, select the following 5 input layers:

   1. ADM1 = mdg_admbnda_adm1_BNGRC_OCHA_20181031

   2. ADM2 & Risk = mdg_adm2_risk - mad_adm2_risk

   3. Cyclone_monitoring_data = cyclone track of the current event

   4. Madagascar_Health_and_Education_Facilities = Madagascar_Health_and_Education_Facilities

   5. Master Raster = MAD_pop_constrained_buildings_landcover

Attention

If you don’t specify the location to save the output files, the outputs will be loaded as temporary layers after running the model and disappear even after saving the project file.

3. Further down, you have to specify where to save the outputs. For each output, click on the three points > Save to Geopackage (Enregistrer dans un Geopackage...). A File explorer window will open. Navigate to the folder .../AA_Cyclone_Monitoring_Trigger_MDG/model_outputs/ and give it the name of the output layer and the date (YYYYMMDD).

   1. Exposed_Cyclone_Area_YYYYMMDD, for example, Exposed_Cyclone_Area_YYYYMMDD_20250805

   2. One output is called Spreadsheet_Exposed_District for which the model will ouput a .csv-file. For this layer, choose Save to file (Enregistrer vers un fichier...), navigate to the folder .../AA_Cyclone_Monitoring_Trigger_MDG/model_outputs/ and give it the name Spreadsheet_Exposed_Districts_YYYYMMDD

   3. Exposed_Education_Facilities_points_YYYYMMDD

   4. Exposed_Health_Facilities_points_YYYYMMDD

   5. Exposed_Regions_YYYYMMDD

   6. Exposed_Districts_YYYYMMDD

   7. Exposed_Population_YYYYMMDD

   8. Exposed_Buildings_YYYYMMDD

   9. Exposed_Agricultural_Landcover_YYYYMMDD

   10. Exposed_Education_Facilities_YYYYMMDD

   11. Exposed_Health_Facilities_YYYYMMDD

4. Once you have set the names and saving locations for the output layers, click Run to execute the model. The output result layers will be automatically added to the main QGIS window upon completion. Once the process has finished, you can close the Model Designer window.

5. You will see new layers added to the map canvas and the layers panel (on the bottom left). Move the new layers to the group “Model_Outputs”.

Video: Input and output Model

Results

We have all the necessary layers to create the individual maps. The next section will cover how to use the predetermined and calculated layers to create the maps using the map templates and layer style files.

Creating the map reports using the map templates

Visualization and Styling of the Model Outputs and creating the Print Map

Output maps

We will generate two different types of output maps to support the analysis:

• Map 1 will provide an cyclone impact overview of the affected districts, the extent of the cyclone event, and the locations of warehouses.

• Map 2 will focus on the impact to infrastructure and population. We will create 5 different impact maps displaying the following information:

  • exposed population

  • exposed buildings

  • exposed health sites

  • exposed education facilities

  • exposed agricultural landcover

We will create the maps in two steps: First, we will use the layer styling panel and the layer style files (.qml) to adjust the visualisation of the layers on the map canvas.

In a second step, we will use the print layout composer to create printable maps with additional datatables.

Map 1: Cyclone Impact Overview: Affected Districts, Event Extent, and Warehouse Locations

Layers needed for this map:

• CRM_Warehouses

• cyclone_track

• Exposed_Cyclone_Area

• Admin1_Impact_Overview_Map already loaded and styled in QGIS

• Exposed_Districts

Right-click on each of the layers and select Duplicate this layer. Move the copy to the group “Map_Cyclone_Impact_Overview”.

The layers should be arranged as shown in the figure below.

Styling of the layers

1. Right click on the exposed_districts layer -> Properties -> Symbology

2. In the down left corner click on Style -> Load Style

3. In the new window click on the three points . Navigate to the “AA_Cyclone_Monitoring_Trigger_MAD/layer_styles” folder and select the file “exposed_districts_style.qml”.

4. Click Open. Then click on Load Style

5. Back in the “Layer Properties” window click Apply and OK

Repeat this process for the following output layers, along with their corresponding style sheets:

Layer name	Style	Comment
Admin1_Impact_Overview_Map	adm1_style.qml	pre-loaded
CRM_warehouses	CRM_warehouses_style.qml	model output
exposed_cyclone_area	exposed_cyclone_area_style.qml	model output
cyclone_track	storm_track_cyclone_style.qml	pre-loaded

6. The styling for the layer CRM_warehouses is not fixed yet. Right-click on the layer CRM_warehouses > Properties and navigate to the symbology tab.

• Select Raster Image Marker and then click on the three points .

• In the project folder, navigate to the subfolder called logo_pictures. Here you will find a png-file called “ngo-office”. Select it and click Open.

Attention

Ensure that all relevant output layers are properly added to the QGIS project. If any layers are missing, try re-running the model or check your Model Outputs folder to see if the files were created successfully.

To maintain a clear and organized workspace, group the output layers in the Layers panel under the appropriate group (e.g., Map_Cyclone_Impact_Overview). This helps keep your project structured and makes navigation easier during the map creation process.

Making the Print Layout

For easier visualization, we have created these map templates for presenting the results of the trigger analysis. These templates serve as a base for your own visualizations and are available in the following directory: AA_Cyclone_Monitoring_Trigger_MAD/map_templates. You can customize the templates to suit your needs and preferences. You can find help here.

1. Deactivate all Layer Groups except the group Map_Cyclone_Impact_Overview and the OpenStreetMap basemap.

2. Open a new print layout by clicking on Project -> Layout Manager. A small new window will appear. Here you can select an existing layout or create a new layout from a template.

3. We want to create a new layout from a template. Click on the Empty Layout dropdown menu and select Specific.

4. Below, click on the three dots and navigate to the folder ../AA_Cyclone_Monitoring_Trigger_MAD/map_templates/ and select the file with the name cyclone_impact_overview_map_template. Click Open, then Create.

5. QGIS will ask you to name the new layout. Give it a name such as “Cyclone_Overview_Map_Freddy_2023”. Click OK. A new window will open. This is the print layout composer. It should look similar to the figure below.

Fig. 60 The print layout composer after opening the template file.

The print layout will automatically load the map canvas. However, to finish the report, we need to adjust and update some of the elements on the print layout. For example, on the right side of the map, the attribute table is not displayed correctly, the legend seems to be wrong, and the logos of the CRM and HeiGIT are displayed as red crosses.

6. Update the Map Title

   • Click on the title text element at the top of the map.

   • In the Item Properties panel, edit the Main Label text to match your event, e.g. Cyclone Harald – 2025.

   • Adjust font size or alignment as needed.

7. Update the Attribute Table on the Right-Hand Side of the Map

• On the right side, there is a attribute table that did not fully load. We want to update the attribute table to display the exposed districts.

• In the Item Properties panel, select the Exposed_Districts layer and click Refresh Table Data

• Click on Attributes...

• In the Columns section:

  • Click Clear

  • ➕ Add the columns: ADM1_EN, ADM2_EN, ADM2_PCODE

• In the Sorting section:

  • ➕ Add ADM1_EN and set the sort order to Ascending

• Click OK to apply

Note

💡 If too many districts are affected, the attribute table might not fit the page. Reduce the font size in the table’s item properties to make everything visible — but be aware that this may reduce readability.

8. Adjust the Legend

   • Select the legend item, navigate to the Item Properties tab and scroll down until you see the Legend items field. If it is not there, check if you have to open the dropdown. Make sure Auto update is not checked.

   • Remove all items in the legend by clicking on each item and then the red minus icon

     • In the pop-up, check Only show visible layers to help you find the correct ones

     • To rename a legend item, double-click on the layer name in the legend item list and enter the new name

   • ➕ Add the following layers by clicking the green plus:

     • Admin1_Impact_Overview_Map

     • exposed_districts

     • Cyclone Track

     • Exposed_Cyclone_Area

     • CRM_warehouses

     • OpenStreetMap

     • Now, let’s rename the layers in the legend. In the Item properties, below the list of the legend layers, there is a Edit selected item properties-button. By clicking on it, you can edit the label of the icon in the legend. Rename the layers as follows:

       • Admin1_Impact_Overview_Map → rename to

       Regions
       

       • exposed_districts → rename to

       Exposed Districts
       

       • Cyclone Track → rename to

       Projected Cyclone Track
       

       • Exposed_Cyclone_Area → rename to

       Exposed Cyclone Area
       

       • CRM_warehouses → rename to

       CRM Warehouses
       

       • Background Map: OpenStreetMap → rename to

       Background Map:
       OpenStreetMap
       

9. Adjust the icons by clicking on the field in the items list or on the red cross in the map template.

• In the Item Properties, correct the path to the CRM logo by clicking on the three dots and navigate to \aa_madagascar\AA_Cyclone_Monitoring_Trigger_MAD\logos_pictures and selecting the CRM logo file.

• Repeat the process for the second missing image. This time, select the HeiGIT Logo

10. Below the logos, adjust the information in the text box by selecting the text box and navigating to the Item properties.

11. Finally, let’s lock the layers and layer styles so that changes in the main QGIS window do not affect our print layout:

• In the item list, select Map 1.

• In the item properties, check the boxes for lock layers and lock styles for layers. This will prevent the map to automatically when we make changes to the map canvas

Attention

Checklist for final map output:

• Map Information: Review and update all text elements as needed.

• Legend: Remove unnecessary items and rename layers with clear, meaningful descriptions.

• Exposed Districts: Include only districts that are actually impacted in your “List of Exposed Districts”. Update them according to the event.

Your final output should look like this after styling the layer

You will now see the exposed districts and the locations of relevant warehouses clearly displayed on the map. Additionally, the original storm track line — used as input data — is highlighted, along with the buffered impact area, which serves as a proxy for identifying exposed districts.

Exporting the Map

When you have finished the design of your map, you can export it as pdf or image file in different data formats.

Export as Image

1. In the print layout click on Layer -> Export as Image

2. Choose the map_outputs folder. Give the file the name of the event e.g MDG_Trigger_Impact_Overview_Map_Freddy_2023.

3. Click on Save

4. The window Image Export Options will appear. Click Save. Now the image can be found in the result folder.

Export as PDF

1. In the print layout click on Layer -> Export as PDF

2. Choose the map_outputs folder. Give the file the name of the event e.g MDG_Trigger_Impact_Overview_Map_Freddy_2023.

3. Click on Save.

4. The window PDF Export Options will appear. For the best results, select the lossless image compression.

5. Click Save. Now the image can be found in the result folder.

Map 2: Impact Assessment: Exposed Population and Critical Infrastructure

Layers needed for this map:

• CRM_Warehouses

• cyclone_track

• Exposed_Cyclone_Area

• Exposed_Population

• Admin1_Impact_Assessment_Map already loaded and style in QGIS

Right click on each layer > Duplicate this layer and move the copies to the group “Map_Cyclone_Impact_Assessment”

Map 2: Styling of the layers

1. Deactivate all the layers except gor the group “Map_Cyclone_Impact_Assessment” and the OpenStreetMap Basemap.

2. Right click on the “exposed_population - copy” layer -> Properties -> Symbology

3. In the down left corner click on Style -> Load Style

4. In the new window click on the three points . Navigate to the “AA_Cyclone_Monitoring_Trigger_MAD/layer_styles” folder and select the file “exposed_population_style.qml” style layer.

5. Click Open. Then click on Load Style

6. Back in the “Layer Properties” window click Apply and OK

Repeat this process for the following output layers, along with their corresponding style sheets:

Layer name	Style	Comment
Admin1_Impact_Assessment_Map	adm1_style.qml	pre-loaded
CRM_warehouses	CRM_warehouses_style.qml	model output
exposed_cyclone_area	exposed_cyclone_area_style.qml	model output
cyclone_track	storm_track_cyclone_style.qml	loaded by user

Attention

Ensure that all relevant output layers are properly added to the QGIS project. If any layers are missing, try re-running the model or check your Model Outputs folder to see if the files were created successfully.

To maintain a clear and organized workspace, group the output layers in the Layers panel under the appropriate group (e.g., Map_Cyclone_Impact_Overview). This helps keep your project structured and makes navigation easier during the map creation process.

Other Impact Assessment Maps

The documentation covers the exposed population impact assessment map. However, the model also estimates the exposed buildings, landcover, and health and education facilities. These variables can also be displayed on the map using the following style files. To keep the map easily understandable, use only one of the

Layer name	Style	Comment
exposed_population	exposed_population_style.qml	model output
exposed_building	exposed_building_style.qml	model output
exposed_health_facilities	exposed_health_facilities_style.qml	model output
exposed_education_facilities	exposed_education_facilities_style.qml	model output
exposed_agricultural_landcover	exposed_agriculture_landcover_style.qml	model output
exposed_health_facilities_points	points_exposed_health_facilities_style.qml	model output
exposed_education_facilities_points	points_exposed_education_facilities_style.qml	model output
relevant_warehouses	CRM_warehouses_style.qml	model output
exposed_cyclone_area	exposed_cyclone_area_style.qml	model output
cyclone_track	storm_track_cyclone_style.qml	loaded by user

Map 2: Making the Print Layout

Tip

The same workflow applies to all five impact variables: population, buildings, education facilities, health sites, and agricultural landcover. The following example demonstrates the process for creating the population impact map. The remaining maps can be generated by following the same steps.

1. Open a new print layout by clicking on Project -> Layout Manager. A small new window will appear. Here you can select an existing layout or create a new layout from a template.

2. We want to create a new layout from a template. Click on the Empty Layout dropdown menu and select Specific.

3. Below, click on the three dots and navigate to the folder ../AA_Cyclone_Monitoring_Trigger_MAD/map_templates/ and select the file with the name cyclone_impact_population_map_template. Click Open, then Create.

4. QGIS will ask you to name the new layout. Give it a name such as “Cyclone_Overview_Map_Freddy_2023”. Click OK. A new window will open. This is the print layout composer. It should look similar to the figure below.

5. Update the Map Title

   • Click on the title text element at the top of the map.

   • In the Item Properties panel, edit the Main Label text to match your event, e.g. Cyclone Harald – 2025.

   • Adjust font size or alignment as needed.

6. Update the Attribute Table on the Right-Hand Side of the Map
   To update the attribute table displaying the exposed districts:

   • In the Item Properties panel, select the exposed_populationOr any other layer you are working with layer and click Refresh Table Data

   • Click on Attributes...

   • In the Columns section:

     • Click Clear

     • ➕ Add the columns: ADM1_EN, ADM2_EN, ADM2_PCODE and exposed_population or any other layer you are working with

   • In the Sorting section:

     • ➕ Add ADM1_EN and set the sort order to Ascending

   • Click OK to apply

Note

If too many districts are affected, the attribute table might not fit the page. Reduce the font size in the table’s item properties to make everything visible — but be aware that this may reduce readability.

7. Adjust the Legend by clicking on it in the map layout and have a look at the Item Properties tab and scroll down until you see the Legend items field. If it is not there, check if you have to open the dropdown. Make sure Auto update is not checked.

   • Remove all items in the legend by clicking on each item and then the red minus icon

     • In the pop-up, check Only show visible layers to help you find the correct ones

     • To rename a legend item, double-click on the layer name in the legend item list and enter the new name

   • ➕ Add the following layers by clicking the green plus:

   • In the pop-up, check Only show visible layers to help you find the correct ones

   • 💡 To rename a legend item, double-click on the layer name in the legend item list and enter the new name

   • Ensure all legend entries use clear and meaningful labels

Exposed Population

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_population → rename to

Exposed Population

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Buildings

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_building → rename to

Exposed Buildings

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Health Facilities

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_health_facilities → rename to

Exposed Health Facilities

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Education Facilities

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_education_facilities → rename to

Exposed Education Facilities

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Agriculture in Hectare

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_agricultural_landcover → rename to

Exposed Agriculture in Hectare

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Health Facilities Points

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_health_facilities_points → rename to

Exposed Health Facilities Points

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Exposed Health Education Points

• Admin1_Impact_Overview_Map → rename to

Regions

• exposed_education_facilities_points → rename to

Exposed Health Education Points

• Cyclone Track → rename to

Projected Cyclone Track

• Exposed_Cyclone_Area → rename to

Exposed Cyclone Area

• CRM_warehouses → rename to

CRM Warehouses

• Background Map: OpenStreetMap → rename to

Background Map:
OpenStreetMap

Your final output should look like this after styling the layer

The map now clearly displays the exposed population within the affected districts, along with the locations of relevant warehouses. The original storm track line — used as input data — is highlighted, as well as the buffered impact area, which serves as a proxy for identifying exposed districts.

On the right-hand side of the map, a list shows all exposed districts, including data on total population and exposed population. The districts (Admin 2) are organized under their corresponding regions (Admin 1).

Exporting the Map

When you have finished the design of you map you can export it as pdf or image file in different data formats.

Export as Image

1. In the print layout click on Layer -> Export as Image

2. Choose the map_outputs folder. Give the file the name of the event e.g MDG_Trigger_Impact_Overview_Map_Freddy_2023. For the specific impact assessment change the name to something like MDG_Trigger_Impact_Population_Map_Freddy_2023.

3. Click on Save

4. The window Image Export Options will appear. Click Save. Now the image can be found in the result folder.

Export as PDF

1. In the print layout click on Layer -> Export as PDF

2. Choose the map_outputs folder. Give the file the name of the event e.g MDG_Trigger_Impact_Overview_Map_Freddy_2023. For the specific impact assessment change the name to something like MDG_Trigger_Impact_Population_Map_Freddy_2023.

3. Click on Save.

4. The window PDF Export Options will appear. For the best results, select the lossless image compression.

5. Click Save. Now the image can be found in the result folder.

Working with the warehouse isochrones

The project includes isochrones for each warehouse. The warehouse isochrones correspond to one warehouse and are identifiable by their location name. If you want to add an isochrone to one of the It is possible to add individual isochrones to the map templates by simply duplicating the isochrone layer and moving it to the desired map group.

Historical Analysis of Cyclone Impacts

To run the full trigger process using historical cyclone track data, you can assess the impacts of past events and gain insights into what occurred in similar scenarios. The storm track data is available from the International Best Track Archive for Climate Stewardship (IBTrACS). Instructions on how to access this data are provided in the following section.

Download of historical storm track data

The International Best Track Archive for Climate Stewardship (IBTrACS) v04r01 data is updated three times a week (usually on Sunday, Tuesday, and Thursday), and could be updated more frequently to address specific needs and use cases. The latest updates in the correct file format can be found on their website:

1. Look for the Access Methods section and click on Shapefiles. The link leads to the following website which can also be seen in the figure below.

2. Since we don’t need storm track data for the entire world or the full archive, we will download only a relevant subset. Locate for the file named IBTrACS.ACTIVE.list.v04r01.lines.zip and click on it - the download should begin automatically.

3. Unzip the file and open it in QGIS.

4. Open the attribute table and delete all the storm tracks that are not relevant for this analysis. Safe the updated storm track file.

Note

The storm track subset IBTrACS.ACTIVE.list.v04r01.lines.zip contains all storms active in the last 7 days. If more comprehensive data is needed, it is advisable to download a subset by basin. For Madagascar, the most relevant region is SI – South Indian, which includes our Area of Interest. This dataset can be downloaded from the same website under the name IBTrACS.SI.list.v04r01.lines.zip.