Exercise 5: Larkana Flood response

Aim of the exercise

In 2024, the provinces of Punjab, Sindh, and Balochistan in Pakistan experienced devastating floods due to intense and prolonged rainfall. The following analysis will utilize actual data from this natural disaster. The objective is to pinpoint the specific medical centers and healthcare facilities that were impacted by the flooding. Additionally, we will assess the viability of road access to the city of Larkana throughout the flood period.

Participants will work with multiple layers and conduct spatial queries. Additionally, they will learn how to create their own geodata. The exercise is divided into three tasks. In the first part, we will export the administrative boundaries in our area of interest (AOI). In the second task, the health facilities located in our AOI will be extracted and we will identify which health facilities are located inside the flood extent layer. In the third task, the road access will be determined by identifying which roads are potentially flooded.

Larkana flood response exercise track:

This exercise is part of the Larkana flood response exercise track

• Next exercise

Competences covered in this exercise

• Select features and export to new file

• Extract by Location

• Reproject Layer

• Editing the Attribute Table

• Classification

• Digitising a point Layer

Estimated time demand for the exercise

• The exercise takes around 3 hours to complete, depending on the number of participants and their familiarity with computer systems.

Relevant wiki articles and module chapters

• Geodata Import in QGIS

• Layer Concept

• Geodata Classification- Categorized

• Geodata Classification - Graduated

• Spatial Queries

• Table function - Add field

• Digitisation- Point data

Instructions for the trainers

Trainers Corner

Prepare the training

• Take the time to familiarise yourself with the exercise and the provided material.

• Prepare a white-board. It can be either a physical whiteboard, a flip-chart, or a digital whiteboard (e.g. Miro board) where the participants can add their findings and questions.

• Before starting the exercise, make sure everybody has installed QGIS and has downloaded and unzipped the data folder.

• Check out How to do trainings? for some general tips on training conduction

Conduct the training

Introduction:

• Introduce the idea and aim of the exercise.

• Provide the download link and make sure everybody has unzipped the folder before beginning the tasks.

Follow-along:

• Show and explain each step yourself at least twice and slow enough so everybody can see what you are doing, and follow along in their own QGIS-project.

• Make sure that everybody is following along and doing the steps themselves by periodically asking if anybody needs help or if everybody is still following.

• Be open and patient to every question or problem that might come up. Your participants are essentially multitasking by paying attention to your instructions and orienting themselves in their own QGIS-project.

Wrap up:

• Leave time for any issues or questions concerning the tasks at the end of the exercise.

• Leave some time for open questions.

Available Data

Download all datasets here and save the folder on your computer and unzip the file.

Dataset	Original title	Publisher	Downloaded from
PAK_Sindh_adm1.gpkg	Subnational Administrative Boundaries	UN OCHA	HDX
PAK_Sindh_adm2.gpkg	Subnational Administrative Boundaries	UN OCHA	HDX
PAK_Sind_Health_Facilities.gpkg	Pakistan Health Facilities (OpenStreetMap Export)	Humanitarian OpenStreetMap Team (HOT)	HDX
VIIRS_20240721_20240803_MinimumFloodExtent_PAK.shp	Satellite detected water extents from 08 to 12 August 2024 over Pakistan)	UNO SAT	HDX
Roads_Larkana.gpkg	Roads Larkana	Humanitarian OpenStreetMap Team	HOT Export Tool (Export created in September 2024)

Hint

Reprojected and fixed Flood extend layer can be downloaded here

Hint

Folder structure To keep your data organized and easily accessible, it’s important to establish a clear folder structure on your computer for your QGIS projects and geodata. Ensure that your exercise data are saved in a location that allows for easy retrieval and association with the corresponding QGIS project.

Task 1: Gain an overview of the situation around Larkana

Context

You have been deployed as an information manager to the flood-affected regions of Pakistan. Upon your arrival you received reports from the operations team indicating that the city of Larkana and its surrounding areas have been severely affected by the floods. The team needs a general overview of the location of the city.

Tip

You cannot interact with a base map!

1. Open QGIS and create a new project by clicking on Project -> New

2. Once the project is created save the project in the folder of the exercise “Modul3_Exercise_2_Flood_Larkana”. To do that click on Project -> Save as and navigate to the folder. Name the project “PAK_Larkana_flood_2024”.

3. First, we want to add the OpenStreetMap as a base map for orientation. To add the OSM as a base map click on Layer -> Add Layer -> Add XYZ Layer…. Choose OpenStreetMap and click Add.

4. Next, load the GeoPackage “PAK_Sindh_adm2.gpkg” in your project by drag and drop (Wiki Video). Or click on Layer-> Add Layer-> Add Vector Layer. Click on the three points and navigate to “PAK_Sindh_adm2.gpkg”. Select the file and click Open. Back in QGIS click Add (Wiki Video).

Attention

GeoPackage files can contain multiple files and even entire QGIS projects. When you load such a file in QGIS a window will appear in which you have to select the files you want to load in your QGIS project.

5. To get an overview of the situation, we want to export the administrative boundaries for our Area of Interest (AOI). To do this, we want to export the district Larkana, as well as the neighbouring districts from the PAK_adm2_Sindh-layer.

   • Right-Click on the PAK_adm2_Sindh-layer and select Open Attribute table.

   • Find the the column ADM2_EN and find the row for the district of Larkana.

   • Click on the numbers all the way to the left of the attribute table window to select the Larkana feature. The row will appear blue and the area of Larkana will turn yellow on the map canvas.

   • Click on Zoom Map to selected rows in the top bar of the attribute table window.

   • Close the attribute table.

   Tip

   Holding Ctrl while selecting features let’s you add more features to your current selection. Otherwise, you will deselect the previous polygon. Keep in mind that you can only select features from the layer you have currently selected in the layers panel.

   • In the toolbar at the top of the QGIS-window, select the tool Select Feature(s) . Hold Ctrl and Click on the districts that are surrounding the district of Larkana (Wiki Video). The four selected districts should appear yellow on your map canvas.

   • Deactivate the Select Feature(s)-tool by clicking on the icon in the toolbar at the top of your QGIS-window.

   • We can now export the selected features and save them to a new file. Right click on the PAK_adm2_Sindh layer and select Export > Save Selected Features as.

   • A new window will open. Here, you can select how and where the selected features are saved.

   

   Make sure to save the geopackage in the correct location by clicking on the three dots to the right of the File name-field

   • Under Format, select geopackage.

   • To the right of the File name-field, click on the three dots. Navigate to the folder with the exercise data and save it in the data/temp/ folder.

   • Enter a layer name. For example, “Flood_2024_AOI”.

   • Click Ok. The exported layer should appear in your layers panel.

Achievement

Now you have an overview of where the district of Larkana is located in Sindh. The operations team can use this information.

Tip

Do not forget to save your project from time to time!

Task 2: Estimation of Flood Impact on the Health Sector in Larkana

Context

Posts on social media have indicated a significant impact on the healthcare system in the region. You have been tasked to find out as much as you can about the situation and, if feasible, to estimate the impact on the health system.

1. First, we need to find out where the health facilities are located in the area. We can find datasets by doing a quick search on the humanitarian data exchange (HDX). Here you can find the dataset “Pakistan Health Facilities (OpenStreetMap Export). This will do for now. The dataset is already available in the download folder for this exercise.

   • Import the GeoPackage PAK_Health_Facilities_complete.gpgk into your project. You can either drag it onto the map canvas, or open the import window by clicking on Layer > Add Layer > Add Vector Layer in the top bar of QGIS (see wiki page). A new layer with point data will appear on your map canvas.

   • Once we have the imported the healthsites, we can extract the healthsites that are located inside of our Area of Interest. We can achieve this with the tool Extract by Location.

   • In the Processing Toolbox (opening the toolbox), search for the tool “Extract by Location”. Double-Click on it. A new window will open.

   

   Extract by location Pakistan

   • As Input Layer, select the layer PAK_Health_Facilities_complete.

   • The Geometric predicate should be set to Intersect

   • Under By comparing to the features from, select the area of interest layer “Flood_2024_AOI”.

   • Under Extracted (location), click on the three dots and select Save to Geopackage.

   • Navigate to the /Module_3_Exercise_5_Larkana_Flood/data/temp/-folder.

   • Give the file the name “Health_Facilities_Flood_2024_AOI” and click Save.

   • You will be prompted to enter a layer name. Give it the same name as the file and click Ok.

   • Click Run. The new layer will appear in your layers tab.

Extract by location Pakistan

Now we have an overview of the location of health facilities. However, we want to know which health facilities are impacted by the flood. Fortunately, the UN has just shared a dataset about the extent of floods from the 8th to the 12th of August, which we can overlay with our layer with the healthsites to identify the healthsites that are in the flooded area.

2. Import the dataset “VIIRS_20240721_20240803_MinimumFloodExtent_PAK.shp” into your QGIS project.

3. Once you have loaded the layers in QGIS, you can see that they are correctly displayed. However, upon checking the layer information, you can see that the new layers have a different Coordinate Reference System (CRS). They have the EPSG Code 9707 whereas our project has 4326 (Wiki Video).

   • Right click on the data layer, click on “Properties”.

   • The “Layer Properties” Window of the data layer will open. Click on “Information”.

   • Under the headline “Coordinate Reference System (CRS)” you find all information about the CRS. The most important are:

   • Name: Here you find the EPSG-Code.

   • Units: Here you find information about the units of measurements used in the dataset. For example, either meters or degrees.

4. This will be a problem as soon as we do something different then just displaying the layers. Since we want to manipulate the layers in the next step we need to reproject them first (Wiki Video).

   • Click on the Vector tab -> Data Management Tools -> Reproject Layer or search for the tool in the Processing Toolbox.

   • As Input layer select “VIIRS_20240721_20240803_MinimumFloodExtent_PAK.shp”

   • Select as target CRS/EPSG-Code 4326.

   • Save the new file in your temp folder by clicking on the three dots next to Reprojected, specify the file name as “2024_MinFloodExtend_reprojected”.

   • Click Run.

   • Delete the old layer from the layer panel by right-clicking on the layer -> Remove layer.

   • Adjust the opacity of the flood layer by right-clicking on layer “2024_MinFloodExtend_reprojected” in the Layer Panel and selecting Properties. A new window will open up with a vertical tab section on the left.

   • Navigate to the Symbology-tab.

   • Adjust the opacity to around 60% by moving the slider.

We have observed that certain health facilities have been impacted by the flood. In order to visualise this information on the map, we plan to include a new attribute called “Flood_affected” in the attribute table of “Health_Facilities_Flood_2024_AOI”. To accomplish this, we will select all the health facilities that are located inside the flood extent by using the tool “Select by Location”. In a next step, we will add a new column to the attribute table and add information into our selected healthsites.

5. Open the Processing Toolbox (Wiki video) and search for the tool “Select by Location”.

   • Select features from = “Health_Facilities_Flood_2024_AOI”.

   • As Geometric predicate we use intersect.

   • For By comparing to the features from we use the layer “2024_MinFloodExtent_reprojected”.

   • Modify current selection by = creating new selection.

   • Click Run.

Select flood affected health facilities

Possible Error Message

In case you encounter the error:

Feature (1) from “2024_MinFloodExtend_reprojected” has invalid geometry. 
Please fix the geometry or change the Processing setting to the “Ignore 
invalid input features” option.
Execution failed after 0.07 seconds

You need to first use the tool “Fix Geometry” before repeating the previously failed step 5 of using the tool “Select by Location”.

• To do so open the Processing Toolbox and search for the tool “Fix Geometries”.

• Input layer = 2024_MinFloodExtend_reprojected

• Save the new file in your temp folder by clicking on the three dots , specify the file name as “2024_MinFloodExtend_reprojected_fix”.

• Click Run.

The error message indicating invalid geometries

6. We can edit the attribute table for the features that we have selected:

   • Open the attribute table of “Health_Facilities_Flood_2024_AOI” by right clicking on the layer -> Open Attribute Table(Wiki Video)

   • Activate the editing mode by clicking on (Wiki Video). Now you are able to edit the data directly in the table.

   • We want to add a new column with the name “Flood_affected” to identify which healthsites are located inside of the flood extent. Click on . A new window will open.

   • In the Add field-window, add the name of the column and specify the Type to Text (string) (Wiki Video).

   • Click Ok .

Adding a new column to the attribute table

7. Next, we want to edit the rows in the attribute table for the features we have selected.

   • Look for the Show all Features option in the lower left corner and click on it.

   • Select the option Show selected features (Wiki Video). This will filter the table to display only the rows that represent the health facilities directly impacted by the flood.

   • Write Yes in the “Flood_affected” column in each selected row.

   • When you are done, click to save your edits and switch off the editing mode by again clicking on (Wiki Video).

   • Click on the icon in the toolbar to end the feature selection.

8. We can display the enriched dataset by visualising it using the categorised classification symbolisation method. This means that we select a column from the attribute table and use the values/content as categories to sort and display the data (Wiki Video):

   • In the layers panel, right-click on the layer “Health_Facilities_Flood_2024_AOI” and select Properties. A new window will open up with a vertical tab section on the left.

   • Navigate to the Symbology tab.

   • At the top of the window, there is a dropdown menu. Open it and choose Categorized.

   • Under Value, select the column we added, “Flood_affected”.

   • Further down in the window, click on Classify. The unique values of the column “Flood_affected” should appear.

   • You can adjust the symbolisation and colour of each value by double-clicking on each color in the table.

   • Once you are done adjusting the colours, click Apply, then Ok to close the symbolisation window.

Flood affected health facilities classification

Achievement:

We’ve pinpointed the specific health facilities that have been inundated by the floods. Our findings indicate that a total of four facilities have been completely flooded and are currently non-operational. Considering we assessed the minimum flood impact, it’s highly probable that more health facilities will also be impacted. This data is crucial for our operational team as it will enable them to strategize and execute an effective response.

Task 3: Logistical access to Larkana City

Context

The operations team is making plans to deliver much-needed supplies to the affected region around Larkana. Currently, there is uncertainty about how the supplies can be transported there. The operations team has asked for more information on this topic.

They need answers to the following three questions:

• Which roads leading into Larkana are blocked, and at what specific locations are they blocked?

• Are there any bridges that can be crossed from the eastern side of the Indus to the western side, and where are these bridges located?

• If transporting supplies by road into the region is not feasible, what alternative method could be used to deliver the supplies?

In order to get a clearer picture, we need to import the road network data for the region into QGIS. Look for the file in the input folder. The road network is initially displayed without showing any road types or other relevant details. We should apply a categorized classification technique only to display the specific roads that we are interested in.

1. Load the dataset “Roads_Larkana.gpkg” from your input folder into your QGIS-project.

2. Lets set up the categorised classification. OpenStreetMap data distinguishes between different road types using the column “highway”.

   • Right-click on the layer “Roads_Larkana” and select Properties. The properties window will open up.

   • Navigate to the Symbology-tab.

   • At the top, select Categorized (Wiki Video).

   • Under Value, select “highway”.

   • Click on Classify. You should see all the unique values for the “highway” column.

   • Remove the ticks for all categories except for motorway, primary, secondary, and trunk.

   • You can adjust the colours by double-clicking on the categories.

   

   The symbolisation window for the Roads_Larkana.gpkg layer.

   • You have the option to customize the width of the main roads’ lines to improve the visualization. Open the Symbology window, then select Symbol. In the new window, you can adjust the width of the lines to your preference.

   

   Adjusting the symbolisation of the different road types

   • Once you are done, click Apply and OK to close the symbology window.

Tip

There are methods to automate the digitisation process which will be covered in module 5: Intermediate GIS Operations.

3. Finally, we want to visualise the roads that are flooded. To simplify the process, we will manually search for the roads intersecting with the flood extent layer and mark them with points. For this purpose, we will create a new point dataset representing the blocked roads.

   • Click on Layer –> Create Layer -> New GeoPackage Layer(Wiki Video)

   • Under Database click on and navigate to temp folder. Give the new dataset the name “PAK_flood_2024_blocked_road”. Click Save.

   • Geometry type: Select Point

   • Under Additional dimension you should always make sure that you check None.

   • Select the coordinate reference system (CRS) “EPSG:4326-WGS 84”. By default, QGIS selects the project CRS.

   • Under New Field you can add columns to the new layer. Add the column “Blocked_road”.

     • Name = “Blocked_road”

     • Type: Select Text Data

     • Click on Add to Fields List to add the new column to the Fields List.

     • Create another field with the name “Blocked_bridge” and the Type: Select Text Data.

     • Click OK.

   • Your new layer will appear in the layers panel.

   

   Creating a new point layer. Make sure to specify a location using the three points at the top.

Tip

If you cannot see the toolbar, click on the tab View -> Toolbars and check Digitizing Toolbar (Wiki Video).

4. Now you can create a point for each place where the flood layer covers the main roads leading out of Larkana (Wiki).

   • Currently the new layer “PAK_flood_2024_blocked_road” is empty. To add features we can use the Digitizing Toolbar.

   • Activate the editing mode by clicking on . Next, activate the option to add new points by clicking on Add Point Feature .

   • Look out for places where the flood layer covers the main roads or bridges leading out of Larkana. Once you have found one, left-click on the location you want to digitise.

   • Once you click on a place, a window will appear. Indicate that the road is blocked by writing Yes in the field Blocked_road.

   • Repeat this step with all the locations your can find.

   

   This pop-up will open once you have selected a location to add a point. Make sure to enter the relevant information in the columns.

   • Once you are done with digitizing click on to save your edits.

   • Click again on to end the editing mode.

5. Now, we have mapped all blocked main access roads into Larkana. We can use icons instead of just points to display the layer “PAK_flood_2024_blocked_road” to visualise this fact better (Wiki).

   • Right-click on the layer “PAK_flood_2024_blocked_road” in the Layer Panel and click on Properties. A new window will open up with a vertical tab section on the left. Navigate to the Symbology tab.

   • Keep the Single Symbol option. Select any symbol from the list that is appropriate for marking blocked roads.

   • Once you are done, click Apply and OK to close the symbology window.

   • After you are done, click on the icon to end the feature selection mode.

   

   Adjusting the symbolisation for the new point layer. Make sure to choose a marker that can be easily identified.

Part of your assignment was to point out possible alternatives to road transport. Can you identify any?

Answer

In the south-west of Larkan City, you can find the Mohenjodaro Airport. Currently, the road from Larkana City to the airport appears to be open and accessible. This means that essential supplies could potentially be transported from the airport into the city without encountering any roadblocks.

Road access to Mohenjodaro Airport

The operations team has now all the information they need to plan their logistics. Good Job!

Continue along this exercise track

This exercise is part of the Larkana Flood Response Exercise track and continues with an exercise in module 4.

Click here if you want to continue to the next exercise of this exercise track.