Trigger & Intervention Map for Forecast-based-Action#

đź”™Back to Homepage

🚧This training platform and the entire content is under ⚠️construction⚠️ and may not be shared or published! 🚧

Aim of the exercise:#

This exercise is based on the monitoring and triggering process used by the Somalia Red Cresent Society (SRCS) in the framework of a drought Early Action Protocol (EAP).

Within this exercise, you will build a simplified version of the monitoring and trigger mechanism for the FEWSNET projection pillar.

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 FbF, 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).

For the development of the Somaliland-Somalia Drought Trigger mechanism, various datasources were thoroughly analysed. Finally, the main parameters chosen for the trigger based on the historical impact assessment are the twelve month Standard Precipitation Index (SPI12) and the IPC acute food insecurity classification. The exact data used are the documented and forecasted SPI12 (source: ICPAC) and the forecasted IPC classification (8 month forecast, source: FEWSNET), that is used to calculate a population weighted index of food insecurity. The trigger thresholds for both components were optimised towards the most favourable proportion of hit rate and false alarm rate. The emerging thresholds were <-1 for the SPI12 and >=0,7 for the IPC based index. The triggering is done on district level and per district just one trigger initiation per year is possible.

Trigger Statement

When ICPAC issues a SPI-12 forecast of less than -1 for a district AND the current FEWSNET food insecurity projection reaches at least 0.7 in its derived population weighted index in the same district, then we will act in this district. We expect the lead-time to be 90 days.

Relevant Wiki Articles#

Data#

Food Insecurity Projection data#

The drought trigger mechanism is based on two variable monitoring datasets. One of them Food Insecurity projection produced by FEWSNET which is updated ruffly every month.

Dataset

Source

Description

IPC Projections

FEWSNET

five-phase scale providing common standards for classifying the severity of acute or anticipated acute food insecurity.

What is IPC Food Security Projection Data?#

The IPC is a commonly accepted measure and classification to describe the current and anticipated severity of acute food insecurity. The classification is based on a convergence of available data and evidence, including indicators related to food consumption, livelihoods, malnutrition and mortality. Food Insecurity is one of the prioritized impacts of droughts in Somalia which is why it is also used for the triggering mechanism, in a population-weighted index.

Three times a year (February, June, and October) FEWSNET estimates most likely IPC classes for the upcoming 8 month (near-term and mid-term projection), available from 2019-current. The near-term projection is called ML1 and is a projection for the upcoming 4 month, the mid-term projection is called ML2 and projects the IPC classes for the 4 subsequent months. For the triggering ML1 (near-term) as well as ML2 (mid-term) projections will be considered.

Outlook updates are produced almost every month and are also taken into account.

Colour

Phase

Descriptions

1. Minimal

Households are able to meet essential food and non-food needs without engaging in atypical and unsustainable strategies to access food and income.

2. Stressed

Households have minimally adequate food consumption but are unable to afford some essential non-food expenditures without engaging in stress-coping strategies.

3. Crisis

Households either have food consumption gaps that are reflected by high or above-usual acute malnutrition OR are marginally able to meet minimum food needs but only by depleting essential livelihood assets or through crisis-coping strategies.

4. Emergency

Households either have large food consumption gaps which are reflected in very high acute malnutrition and excess mortality; OR are able to mitigate large food consumption gaps but only by employing emergency livelihood strategies and asset liquidation.

5. Famine

Households have an extreme lack of food and/or other basic needs even after full employment of coping strategies. Starvation, death, destitution, and extremely critical acute malnutrition levels are evident. (For Famine Classification, area needs to have extreme critical levels of acute malnutrition and mortality.)

Training Data#

Download the data folder here and save it on your PC. Unzip the .zip file!

For this particular exercise, we will use a combination of pre-processed data and the download of real data from FEWS.net. The preprocessed datasets are:

Dataset

Source

Description

Adminstrative bounderies

HDX

The administrative bounderies on level 0-2 for Somalia and Somaliland can be accessed via HDX. For this trigger mechanism we provide the administrative bounderies on level 2 (district level) as a shapefile. We have added the population number for each district derived from Worldpop.

Population Counts

Worldpop

The worldpop dataset in .geotif rasterformat provides population estimates per hectar for the year 2020

Whereas the IPC-Projections data will be downloaded by the participants directly from FEWS.net.

Dataset

Source

Description

IPC Projections

FEWSNET

five-phase scale providing common standards for classifying the severity of acute or anticipated acute food insecurity.

Task#

Step 1: Setting up folder structure#

Purpose: In this step, we set up the correct folder structure to make the analysis easier and to ensure consitent results.

Tool: No special tools or programs are needed.

Instruction

Folder Structure

  1. Open the Folder “Modul_5_Exercise2_Drought_Monitoring_Trigger”

  2. Open the subfolder “Monitoring”

  3. Copy the Template folder “TEMPLATE_Year_Month” and change the name to the current year and month 2024_01.
../../_images/Exercise_Folder_structure_Drought_Monitoring_Trigger.drawio.svg

The Video below shows the process for setting up the folder for december 2023.

Video: Setting up folder structure

Step 2: Download of the forecast data#

Purpose:

Tool: Interent Browser

The IPC data will be pulled from the FEWSNET website. FEWS NET publishes IPC data on its website. The main data publications, as well as the updates of the IPC data, amount to the publication of new data almost monthly.

IPC Data#

The IPC Projection data is provided and regulary updated on the FEWSNET Website. On the website, you will have to click on Somalia to acess the data. Alternativley, you can navigate through Data -> Acute Food Insecurity Data and enter „Somalia”. In the menu you will see different dataformats for different timestamps. Once you find out which timestamp is the most current one, find the ZIP download. We need the data in shapefile (.shp) format, which is only included in the ZIP file and not provided as single download file.

Warning

The FEWSNET pages change often!

  1. Go to FEWSNET Website. Click on Data -> Acute Food Insecurity.

  2. Scroll down. In Geograhic Area, type in “Somalia” and click Apply

  3. Choose the newest dataset.

../../_images/IPC_Projections_website.png

  1. Download the one with the ZIP Data

  2. When you have downloaded the data, right-click on the file and click on Extract all -> Extract

  3. Open the extracted folder and copy the ML1 data in the IPC_ML1 folder you have created in step 1.

  • The filename is composed of “SO” for Somalia, year and month of the report month e.g SO_202308_ML1.shp Example path: .../Modul_5_Exercise2_Drought_Monitoring_Trigger/Monitoring/Year_Month_template/IPC_ML1

Warning

Make sure to not use the ML1_IDP data which comes in the .zip folder as well!

Warning

Remember that you need to copy over all components that the respective shapefile is composed of. Most probably it has 5 components: .cpg, .dbf, .prj, .shp, and .shx.

../../_images/IPC_zip.PNG

Tip

On the main FEWSNET page, you can also sign up for information on latest updates via email. For this option scroll down to the end of the page and click on Sign up for Emails. You will get the option to choose updates only for Somalia.

../../_images/IPC_Newsletter.png

Step 3: Loading data into QGIS#

Purpose: In this step, all the data needed will be loaded into QGIS.

Tool: No specific tools are needed, only QGIS.

  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 you created in Step 1 (e.g. 2022_05). To do that, click on Project -> Save as and navigate to the folder. Give the project the same name as the folder you created (e.g. 2022_05). Then click Save

  3. Load all input data in QGIS by drag and drop. Click on Project -> Save

  • From the folder you created in step 1

    • ML1

  • From the Fixes_data folder:

    • district_pop_som

    • WorldPop_som.tif

Result: QGIS project with all necessary data ready to be analysed.

Step 4: Intersection of ML 1 data with the district polygons#

Purpose: The goal is to receive polygon layers which share both the borders and the attributes of both input layers.

Tool: Intersection

Instruction

Screenshot of the Intersection window

  1. Click on Vector -> Geoprocessing Tools -> Intersection

  2. Input Layer: ML 1

  3. Overlay layer: district_pop_sum

  4. Under Intersection click on the three points -> Save to File and navigate to you monitoring folder [Year_Month]. Give the output the name “ML1_Intersection” and click Save

  5. Click Run
../../_images/SRCS_Trigger_step_4_Intersection.png

Result: After doing this for ML1, you should have one polygon layer containing all columns of ML1 and district_pop_sum.

Note

The resulting layer can have more rows than the original layers.

Video: Intersection of ML 1 data with the district polygons

Step 5: Calculation of Population per Intersection Polygon#

Purpose: Here we calculate the population in each polygon of the intersection layer from step 4.

Tool: Zonal Statistics

Instruction

Screenshot of the Zonal Statistics window

  1. In the Toolbox -> Search for Zonal Statistics

  • Tip: If the Toolbox is not open, click Processing-> Toolbox

  1. Input Layer: “ML1_Intersection”

  2. Raster Layer: “som_ppp_2020_UNadj_constrained.tif”

  3. Statistics to calculate: Only Sum

  4. Under Zonal Statistics, click on the three points -> Save to File and navigate to you monitoring folder [Year_Month]. Give the output the name “ML1_zonal_statistic” and click Save

  5. Click `Run
../../_images/SRCS_Trigger_step_5_zonal_statistic.png

Result: The result should be the “ML1_zonal_statistic” as a polygon layer. This layer should have the same columns in the attribute table like ML1_Intersection plus the column “_sum”, which is the number of people living in the single parts of the polygons.

Video: Calculation of Population per Intersection Polygon

Step 6: Weighting of the Population based on IPC-Phase#

Purpose: The purpose of this step is the weighting of the population in the five IPC phases as described in IPC Data.

Note

The IPC Index represents low-population districts equal to high-population districts. No underrepresentation of high food insecurity of small districts occurs.

IPC-Population Weighted Index

To better operationalise the IPC data, a simple population-weigthed index was developed. Relative population numbers are weighted based on the respective IPC class they fallin, in order to give the amount of people in a certain IPC class the importance instead of the IPC class only. Furthermore, population located in a higher IPC class is more important than population located in a lower class. The index is calculated as follows:

$ IPC\ Index =  Weights \times \frac{District\ Pop\ per\ IPC\ Phase}{Total\ District\ Pop}$

Where the weights are defined as:

IPC Pahse

Weight

IPC 1

0

IPC 2

0

IPC 3

1

IPC 4

3

IPC 5

6

The IPC Index represents low-population districts equal to high-population districts. No underrepresentation of high food insecurity of small districts occurs.

Tool: Field Calculator

  1. Right-click on the layer “ML1_zonal_statistic” -> Open Attribute Table-> click on Field Calculator to open the field calculator

  2. Check Create new field

  3. Output field name: Name the new column “pop_sum_weighted”

  4. Result field type: Decimal number (real)

  5. Add the code block from Input into the Expression field Click ok

CASE

WHEN "ML1" = 3 THEN "_sum" * 1
WHEN "ML1" = 4 THEN "_sum" * 3
WHEN "ML1" = 5 THEN "_sum" * 6
ELSE "_sum"

END

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

Step 7: Calculation of Population Proportion per Intersection Polygon#

Purpose: In this step, we calculating the IPC-Population Weighted Index for every small part of the polygon layer.

Tool:Field Calculator

  1. Right-click on “ML1_zonal_statistic” layer -> “Attribute Table”-> click on Field Calculator to open the field calculator

  2. Check Create new field

  3. Output field name: Name the new column “Index_per_IPCPolygon_ML1”

  4. Result field type: Decimal number (real)

  5. Add the following code into the Expression field

"pop_sum_weighted"/"districtpo"

  1. Click ok

  2. Save the new column by clicking on in the attribute table and end the editing mode by clicking on

../../_images/SRCS_Trigger_step_8_field_calculator.png

Result: The layer “ML1_zonal_statistic” should now have the column “Index_per_IPCPolygon_ML1”. The numbers in this column have to be smaller than in the “district” column.

Video: Calculation of Population Proportion per Intersection Polygon

Step 8: Calculate IPC Index per District#

Purpose: The purpose of this step is to calculate a population-weighted mean over the IPC classes per district. In this way, the amount of people living in a certain IPC class will be given more importance than just the area affected by a certain IPC class. The result is an IPC Index value for each district.

Tool: Join attribute by location (summary)

Instruction

Join attribute by location (summary)

  1. In the Toolbox-> Search for Join attribute by location (summary)

  • Tip: If the Toolbox is not open, click Processing-> Toolbox

  1. Input Layer: Select your “district_pop_som” layer

  2. Input Layer 2: Select “ML1_zonal_statistic”

  3. Geometric predicate: Select “Intersection”

  4. Field to summarise: Select “Index_per_IPCPolygon_ML1”

  5. Summaries to calculate: Only check the option “mean”

  6. Under Join Layer, click on the three points -> Save to File and navigate to you monitoring folder [Year_Month]. Give the output the name “ML1_IPC_Index” and click Save

  7. Click Run
../../_images/Exercise_trigger_join_attributes_location.png

Result: As a result, your layer “ML1_IPC_Index” should have the column “Index_per_IPCPolygon_ML1_mean”. Furthermore, the number of rows should be the exact number of districts in Somalia and the polygons should have the exact shape of the districts.

Video: Calculate IPC Index per District

Step 9: Evaluate Trigger Activation#

Purpose: The purpose of this step is to gain a quick overview of possible trigger activation without having to revise the actual data. Instead, we will have a binary column with trigger = yes or trigger=no values.

Tool: Field Calculator

  1. Right-click on “ML1_IPC_Index” layer -> Attribute Table-> click on Field Calculator to open the field calculator

  2. Check Create new field

  3. Output field name: Name the new column “Trigger_activation”

  4. Result field type: Text (string)

  5. Add the code below into the Expression field

  6. Save the new column by clicking on in the attribute table and end the editing mode by clicking on

Code

 
CASE

WHEN "Index_per_IPCPolygon_ML1_mean" >0.7 
THEN 'yes'
ELSE 'no'

END

  1. Click ok

  2. Save the new column by clicking on in the attribute table and end the editing mode by clicking on

Result: A layer with all districts of Somalia with a column of “Yes” and “No” values indicating whether the trigger levels have been reached or not.

../../_images/Exercise_trigger_evaluation.png
Video: Evaluate Trigger Activation

Step 10.: Visualisation of results#

Purpose: Definition of how features are represented visually on the map.

Tool: Symbology

Trigger Activation

  1. Right cklick on the “ML1_IPC_Index” 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 “FbF_Drought_Monitoring_Trigger/layer_styles” folder and select the file “Style_Trigger_Activation_ex.qml”.

  4. Click Open. Then click on Load Style

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

Info: Trigger Activation Layer

You will now see districts where no trigger is activated in green and districts with trigger activation in pink.

The “Style_Trigger_Activation.qml” style layer is configured to show the district names only where the trigger is actually activated. If there is no trigger activation you can activate the admin 1 boundary layer for better map orientation (see Administrative 2 Boundaries below)

../../_images/Map_yes_trigger.PNG

Administrative 2 Boundaries (Regions)

  1. Right click on the “Som_admin1_regions_UNDP.gqkp” (Regions) 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 “FbF_Drought_Monitoring_Trigger/layer_styles” folder and select the file “SOM_regions_style_ex.qml”.

  4. Click Open. Then click on Load Style

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

  6. Add a the OpenStreetMap basemap by clicking on Layer -> Add Layer -> Add XYZ layer... -> Select the OpenStreetMap. Click Add. (Wiki basemap)

  7. Place the OpenStreetMap basemap on the bottom.

  8. Delet all layers exept:

    • Trigger_activation

    • Som_admin1_regions_UNDP

    • OpenStreetMap

Video: Visualisation of results

Intervention Map without Trigger activation

Intervention Map with Trigger activation
../../_images/Map_no_trigger.PNG
../../_images/Map_yes_trigger.PNG

Attention

Remember the layer concept and make sure the basemap layer is at the bottom of your layers panel.

Step 11: Making print map#

Purpose: Viualization of the map features in a printable map layout

Tool: Print Layout

  1. If not done before, delet all layers expect Trigger_activation, Som_admin1_regions_UNDP and OpenStreetMap

  2. Open a new print layout by clicking on Project -> New Print Layout -> enter the name of your current Project e.g “2024_01”.

  3. Go the the Modul_5_Exercise2_Drought_Monitoring_Trigger folder and drag and drop the file Trigger_activation_Intervention_map_ex.qpt in the print layout

  4. Change the date to the current date by clicking on “Further map info…” in the items panel. Click on the Item Properties tab and scroll down. Here you can change the date in the Main Properties field.

  5. If necessary, adjust the legend by clicking on the legend in 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 itemes in the legend be clicking on the item and then on the red minus icon below.

    • Add Trigger_activation to the legend by clicking on the green plus and click on the layer and click ok

    • Add Som_admin1_regions_UNDP to the legend by clicking on the green plus and click on the layer and click ok

Video: Making print map

Attention

Make sure you edit the Map Information on the template, e.g. current date. Also make sure to check the legend items: Remove unnecessary items and eventually change the names to meaning descriptions.

You can also adapt the template to your needs and preferences. You can find help here.

Attention

Make sure you edit the Map Information on the template, e.g. current date. Also make sure to check the legend items: Remove unnecessary items and eventually change the names to meaning descriptions.

Step 13.: Exporting Map#

Purpose: Export the designed and finalized map layout in order tp print it as a pdf or format of your choice.

Tool: Print Layout

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

Export as Image

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

  2. Chose the Result folder in the folder you have created in step 1. Give the file the name of the project e.g 2022_04

  3. Click on Save

  4. The window “Image Export Options” will appear. Click Save Now the image can be found in the result folder in the folder you created in Step 1

Export as PDF

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

  2. Chose the Result folder in the folder you have created in step 1. Give the file the name of the project e.g 2022_04

  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 in the folder you created in Step 1

../../_images/map_output_example_ex.png