2024 Nigeria Floods
| General Information | ||||||||||||||
Context of the Activation Alau Dam, located just over 10 miles to the south of Maiduguri, the capital of Borno State, collapsed in the middle of the night on Tuesday 10 September 2024, forcing hundreds of thousands from their homes.Homes, infrastructure, roads, schools, hospitals have been submerged by water. The dam collapse saw river water overrunning 50% of Maiduguri and state authorities issued evacuation orders to residents in the affected areas, appealing for humanitarian support. More than 1,000,000 people have been displaced across northeast Nigeria, according to Associated Press. MSF and the Nigerian Red Cross are both responding and in need of data. | ||||||||||||||
2024 Nigeria Floods Response
Contacts
- Coordination: Pete Masters - pete.masters@hotosm.org
- Community engagement: Wilson Munyaradzi - Wilson.Munyaradzi@hotosm.org
- Partner engagement: Sam Colchester - sam.colchester@hotosm.org
Hashtag
All contributions through HOT's Tasking Manager are tracked with a unique change set comment tag: #2024NigeriaFloods
ohsomeNow Stats page for tracking overall contributions
Timeline
- Start: September 13, 2024
- End: November 14, 2024
Remote Mapping Coordination
Tasking Manager project prioritization
Prioritising neighbourhoods in Maiduguri using the following criteria:
- If they intersect the flood extent analysed by UNOSAT (light blue outline on on uMap)
- If they contain priority points shared by Humanitarian Mappers (green on uMap)
- If they intersect priority points/areas shared by Nigerian Red Cross (blue points/areas on uMap)
- If they don't intersect with areas mapped (or already being mapped) by volunteers (red areas on uMap)
List of active projects in HOT's Tasking Manager
Map and Data Services
Exporting OpenStreetMap data
About This Disaster Activation
About HOT
To learn more about the Humanitarian OpenStreetMap Team (HOT), explore more of our wiki-pages (root: HOT) or our website hotosm.org. HOT is a global community, mostly of volunteers, and it is a US registered nonprofit able to contract with organizations (email info at hotosm.org to contact our staff), we are also a 501(c)(3) charitable organization.
History of this Activation
- 11 Sep 2024: HOT activated
- 13 Sep 2024: Existing MSF Tasking Manager project that happened to overlap flooded areas in Maiduguri was bumped to 'Urgent' priority and a priority area (see red polygon) in the north east was defined to focus mapping efforts in flooded zone (see UNOSAT)
- 13-16 Sep 2024: HOT worked with Nigerian Red Cross and local organization Humanitarian Mappers to geolocate a list of priority areas and points in Maiduguri
- 15 Sep 2024: HOT requested recent pre-flood images from Maxar's catalogue covering Maiduguri
- 17 Sep 2024: HOT posted in OpenStreetMap Community Forum to share Wiki and raise awareness
- 17 Sep 2024: Using priority areas HOT published the first dedicated project in HOT's Tasking Manager focusing on Gwange 1 Ward in Maiduguri
- 20 Sep 2024: Additional priority areas added to MSF project based on Nigerian Red Cross and Humanitarian Mappers geolocated priority points
- 23 Sep 2024: All priority areas in MSF project fully mapped and validated so project moved down from 'Urgent' to 'High' priority
- 29 Sep 2024: Additional priority wards shared by Nigerian Red Cross: Bolori, Gwange I & II, Lamisula and Gomari - used to re-prioritize Tasking Manager projects in order
- 9 Oct 2024: Based on earlier request Maxar released up-to-date imagery under CC-NC license for impacted areas in Nigeria through the Maxar Open Data Programme
- 10 Oct 2024: Two pre-flood 2024 images from Maxar's catalogue added to OpenAerialMap. However, on comparison it appears identical to the Esri World imagery. This indicates that the Esri World Imagery was captured in early 2024 and so is relatively up-to-date.
- 14 Nov 2024: HOT contacted Nigerian Red Cross to consider the disaster response mapping phase complete and suggested that any additional areas could potentially be mapped by the volunteer community as longer term disaster recovery mapping efforts.
Data Quality
Validation
Validation permissions for Tasking Manager projects are restricted to particular Tasking Manager Teams / users to ensure high quality validation:
- HOT Global Validators
- Validator Trainees
- MSF Validators
- Data Quality Interns
- Data Quality Interns 2022
OSMCha
A specific OSMCha filter can be used to detect data quality issues across a wider area.
For Mappers
How You Can Contribute
Learn to Map
- Most of our volunteer needs are for remote OSM contributors, visit LearnOSM.org to get started.
Mapping Priority
- Please choose from highest priority first.
- We call for local mappers to prioritize local knowledge on the area.
- The call is only for intermediate and experienced mappers.
- Experienced mappers are also asked to participate in validating completed tasks. Information on validating can be found here.
Tasking Manager
Using the prioritization process outlined above the projects have been (and will be) published in the following order:
| Tasking Manager Projects | |||||
| Project # | Priority | Name | Task Mapping Status % | Task Validation Status % | Completion date |
|---|---|---|---|---|---|
| 14735 | Done | Missing Maps: Maiduguri, Nigeria (3) - priority area | 100% | 100% | Priority areas completed on 23 Sep 2024, fully completed on 10 Nov 2024 |
| 17598 | Done | Nigeria Flooding: Maiduguri, Gwange 1 | 100% | 100% | 20 Sep 2024 |
| 17605 | Done | Nigeria Flooding: Maiduguri, Gamboru | 100% | 100% | 25 Sep 2024 |
| 17685 | Done | Nigeria Flooding: Maiduguri, Gwange 2 | 100% | 100% | 07 Oct 2024 |
| 17731 | Done | Nigeria Flooding: Maiduguri, Lamisula | 100% | 100% | 17 Oct 2024 |
| 17796 | Done | Nigeria Flooding: Maiduguri, Gomari | 100% | 100% | 25 Oct 2024 |
| 17911 | Done | Nigeria Flooding: Maiduguri, Bolori 2 South | 100% | 100% | 3 Nov 2024 |
| 17646 | Done | Nigeria Flooding: Maiduguri, Mashamari | 100% | 100% | 16 Oct 2024 |
| 17683 | Done | Nigeria Flooding: Maiduguri, Gwange 3 | 100% | 100% | 31 Oct 2024 |
Project-Specific Mapping Notes
Imagery
Please use Esri World Imagery to map and validate. It is the most recent imagery available. In October 2024 Maxar shared some of their latest available clear imagery over Maiduguri (captured in January 2024) and on comparison it appears identical to the Esri World imagery. To check the date of the Esri imagery that you are mapping with, zoom into this map.
If there is a misalignment between the imagery and the existing data, please adjust the imagery to align with mapped roads before mapping buildings.
- In iD: Open the background panel by pressing 'b' or clicking the layer stack icon on the right side of the screen. Scroll to the bottom of that panel and expand the 'Adjust imagery offset' section. Drag to adjust
- In JOSM: select Imagery > Imagery Offset > New offset and drag to adjust
Please mention the offset that you have used in your task comment when submitting. This will make it much easier for the validator to check your work.
Buildings
Please map all buildings. Buildings are generally rectangular, so make buildings rectangular even if a shadow, tree, or roof pitch makes it appear otherwise.
- In iD: draw an area and label it as Building, then correct for right angles with the Q key.
- In JOSM: draw an area and use the tag building=yes, then correct for right angles using the Q key - OR use the building plugin.
To check whether you missed any buildings or are in doubt about any features, try the following:
- Zoom out a bit and toggle the data layer off and on a few times (type 'U' in iD to access the menu to do this)
- Comparing with Bing may be useful as buildings may show differently compared to Esri
- Increase the saturation of the imagery (hit 'B' in iD and scroll down to display options)
If you see buildings that look like they are under construction or are missing a roof, these should be tagged as normal buildings. It is likely that the construction work has been completed since the Esri imagery was captured.
Mapping hints
- Parts of the area have been mapped in 2017 and 2019 using old Mapbox and Bing imagery and need updating. If you find it too difficult for you to fix, just map the buildings that not have been mapped yet and leave the rest of the task for another mapper. If you are confident that there are buildings that have been mapped that no longer exist in the Esri imagery you can consider deleting them, but please treat this as a last resort and avoid mass deletions.
- The local mapping community has collected extremely important local information such as place names, locations of services and road names. So please do not delete any of these! If you have personal knowledge of the area yourself, please don’t hesitate to add that information as well.
- Buildings may be very close to each other, but do not actually touch each other. Try to map them close to each other without letting them connect or share nodes with each other, roads or residential area outlines.
- If you have questions, please ask. You can post questions in the Questions and Comments tab of this task or if you are at a mapping event, check with your neighbour or just raise your hand and ask.
After Action Review
Used this as a template: https://www.openstreetmap.org/user/pedrito1414/diary/402943
Debrief Conducted by Sam Colchester (HOT) partner engagement lead for activation.
Relevant statistics
In total 171 contributors made over 125,000 edits to OSM including mapping over 96,000 buildings as part of this campaign. (See ohsomeNow Stats page for tracking overall contributions)
Narrative summary
Shortly after the severe flood event in Nigeria on Tuesday 10 September 2024 the director of the Open Mapping Hub - West and Northern Africa requested support from HOT's central team to respond.
On Thursday, 12 September a size-up was completed by Pete Masters (HOT) and a search for affected areas revealed a UNOSAT flooded area polygon for Maiduguri. The extent of this polygon was manually digitised from a georeferenced screenshot because UNOSAT’s download did not include the Maiduguri flood extent in a spatial file format. MSF staff flagged to HOT that OSM road and building features would be important to their operations in Maiduguri. HOT conducted a manual visual check of Maiduguri, revealing that most buildings were not digitised in OSM but that the road network was fairly well mapped.
Search for existing Tasking Manager projects
On Friday, 13 September, HOT compiled a list of relevant active or completed Tasking Manager projects near Maiduguri and added the extents to a uMap for ease of sharing. The most relevant project that this revealed was a partially complete MSF Tasking Manager project for mapping buildings in western Maiduguri. It happened to intersect the UNOSAT flooded area and so was bumped up to 'Urgent' priority by HOT (it initially had a different focus and was about 50% mapped and 25% validated). HOT defined a 'Priority area' in the eastern section of this project using the manually digitised UNOSAT flooded extent. This was done in close contact with MSF.
An unfinished and relatively new (published 4 September) roads and buildings Tasking Manager project from Nigerian organisation Geohazards Risk Mapping Initiative in a settlement 80km east of Maiduguri called Dikwa was also identified. HOT contacted the project creator to notify them of the activation and ask if they had plans for further project creation. Before contact the project had already been bumped to ‘Urgent’ priority and subsequently made very fast progress. After nearing completion the following day the project creator lowered the project priority to ‘Medium’ and notified HOT that they did this so that mapping in Maiduguri city could be prioritised.
Establishing new areas of interest and prioritization
On Friday, 13 September, Nigerian Red Cross shared that OSM roads and buildings in Maiduguri would be crucial for their planning and relief operations and provided a list of priority locations. This was in the form of a list of location names, including point locations and the names of Wards in Maiduguri. Local organisation Humanitarian Mappers also shared a shortlist of priority point locations in Maiduguri with coordinates. The Nigerian Red Cross locations were manually geocoded over the next two or three days, and combined with the Humanitarian Mappers priority points in a collaborative uMap for ease of sharing. Several Nigerian Red Cross location names could not be matched against available location names online and therefore were not geocoded.
On Tuesday, 17 September HOT published the first dedicated project in HOT's Tasking Manager focusing on mapping buildings in Gwange 1 Ward in Maiduguri. This project area was chosen using the ‘Tasking Manager project prioritization’ process outlined in this Wiki (which used the collaborative uMap containing priority locations shared by Nigerian Red Cross and Humanitarian Mappers). Pete Masters notified the wider public of this first dedicated project by posting in the OSM Community Forum. Over the following weeks a further seven Wards were prioritized for mapping using the prioritization process, with a dedicated project for each.
A risk assessment for mapping activities in Maiduguri was started on Tuesday, 17 September and fully completed with feedback from local partners on Wednesday, 2 October 2024. Due to there being ongoing violence in the area being mapped the results of the risk assessment included mitigations to minimize risk, particularly on how mapping activities would be documented and publicly communicated.
On Friday, 20 September the priority area in MSF’s Maiduguri project was adjusted to focus on all the areas where the project intersected priority points shared by Nigerian Red Cross and Humanitarian Mappers. Three days later this project was shifted from Urgent down to High priority because all of the priority areas in it were fully mapped and validated.
On Sunday, 29 September Nigerian Red Cross shared several additional priority Wards in Maiduguri and these were used to re-prioritize Tasking Manager project ordering and to create new project boundaries.
On Wednesday 9 October, Maxar released up-to-date imagery under CC-NC license for impacted areas in Nigeria through the Maxar Open Data Programme. This was in response to a request that HOT had made for pre-flood images from Maxar’s catalogue on Sunday, 15 September. After adding this Maxar imagery to OpenAerialMap it was discovered to be identical to the Esri World Imagery which had been used on all projects up to that time.
Closure
By the start of November 2024 nine Nigeria Flood projects had been completed and project progress had started to trail off after the slowdown in news. Accordingly, HOT contacted Nigerian Red Cross to suggest that the disaster response mapping phase should be considered complete. On Thursday, 21 November 2024 Nigerian Red Cross confirmed that they considered the disaster response mapping complete and passed on thanks for the incredible work from the OSM community. In mid-November 2024 a third-pass validation (final wide-area check) of the MSF Maiduguri project was commenced.
Successes, issues and lessons learnt
[SUCCESSES] What went well?
- Clear coordination between central and hub team (West and Northern Africa) - HOT internal
- Collaboration with partners throughout (launching activation, defining and revisiting priorities)
- Openness on project prioritization allowed outside input (uMap and process documented in Wiki)
- Good quality tasking manager projects (right small size)
- Quick validation progress
- Project closure, no partially completed projects by end of campaign
[ISSUES] What could have gone better?
- Time lost georeferencing and manually digitizing flood extent from UNOSAT
- Received priority locations in non-spatial format and time was lost manually geocoding (and some could not be geocoded)
- Priority locations had no indication of their relative priority, this made prioritization process difficult
- Maxar request was not worthwhile because Esri World Imagery had latest Maxar scenes
- Better understanding how the data was being used would speed up mapping and help HOT to faster adapt to responders’
[SUGGESTIONS] What lessons should influence how we activate in future?
- Encourage community/organizations sharing priority locations to indicate their relative priority (e.g. see ‘Prioridad’/’Priority’ column in this table)
- Always check the Esri Imagery Date Finder before making a request from Maxar for up-to-date imagery — it may already be on Esri World Imagery!
- Encourage community/organizations to share priority locations in spatial format to reduce geocoding workload and reduce chance of miscommunication
Follow up
If you have questions and / or comments related to the activation or this documentation, please contact Sam Colchester at HOT. Or comment on the OSM Diary Post which is a duplicate of this After Action Review.