Somalia security dashboard
The security picture is set by an armed opposition that still reaches into Mogadishu and regional capitals, a security transition handing districts to forces whose hold is uneven, and federal-member-state politics that periodically turn into armed standoff. Movement risk is concentrated on the arterial roads between government-held towns rather than inside them.
Every judgement on these dashboards is an analyst reading of public reporting. Direction of travel is a judgement, not a measurement. Nothing here is live monitoring: the picture is refreshed when the desk reviews the country, and the review date is shown.
Assessed risks by severity
Analyst ratings of likelihood and impact on a five-by-five scale, scored 1–25. These are judgements from public reporting, not measured values.
- SO-A1 Staff caught in the effects of an explosive attack on a nearby government, security or hotel targetCRITICAL · 20 of 25
Likelihood Likely · Impact Severe
- SO-A2 Ambush or roadside device against vehicle movement on an inter-city corridorSEVERE · 15 of 25
Likelihood Possible · Impact Severe
- SO-A3 Detention, extortion or long delay at an informal checkpointSUBSTANTIAL · 12 of 25
Likelihood Likely · Impact Moderate
- SO-A4 Abduction of staff for ransom or leverageSUBSTANTIAL · 10 of 25
Likelihood Unlikely · Impact Severe
Risk matrix
Where each rated risk sits once likelihood and impact are combined.
| Likelihood / Impact | Insignificant | Minor | Moderate | Major | Severe |
|---|---|---|---|---|---|
| Almost certain | — | — | — | — | — |
| Likely | — | — | SO-A3 | — | SO-A1 |
| Possible | — | — | — | — | SO-A2 |
| Unlikely | — | — | — | — | SO-A4 |
| Rare | — | — | — | — | — |
Residual position: High. Controls reduce likelihood on movement and the impact on sites, but the threat is capable and persistent; residual risk should be formally accepted at director level and reviewed quarterly.
What is driving the picture
Security transition ahead of capacity
Responsibility for districts has moved faster than the forces taking them can hold ground, garrison it and supply it. Recovered towns become islands with contested roads between them, which is a movement problem long before it becomes a control problem.
Federal and member-state competition
Disputes between the federal government and member states over elections, revenue and command of forces are settled politically most of the time and militarily some of the time. Escalation is usually preceded by public statements and force movements that open sources report.
Clan and land contest
Local violence often follows clan, land and water disputes rather than the national conflict, and it affects staff movement, hiring and acceptance in ways a purely conflict-focused reading misses.
External security assistance
Regional troop contributions, bilateral training and air support shape what operations are possible. Changes in mandate or funding are announced publicly and are followed by observable shifts in the tempo of ground operations.
Indicator watchlist
What an analyst looks for in open reporting, and why a change in it matters for people working in the country.
| Indicator | Why it matters |
|---|---|
| Complex attacks on hotels, government compounds or the airport corridor in Mogadishu. | Signals capability and intent against soft targets used by visiting staff, and usually triggers days of movement restriction and checkpoint change. |
| Reported checkpoint activity or taxation on the Mogadishu–Afgooye, Mogadishu–Jowhar and Baidoa corridors. | The clearest single indicator of whether road movement between government-held towns is realistic in a given week. |
| Public disputes between the federal government and Puntland or Jubaland, including force movements toward disputed districts. | Precedes localised armed confrontation and airport or road closure with very little notice. |
| Aid-security incidents involving staff detention, vehicle theft or facility intrusion in the south-central regions. | A more direct read on risk to civilian organisations than overall conflict-event counts. |
Actors that matter
- Al-Shabaab
Armed opposition with sustained intent and demonstrated capability against government, security and hospitality targets.
- Somali federal security forces and Danab units
State forces conducting clearance and holding operations, with uneven capacity to garrison recovered ground.
- Federal member state administrations
Puntland, Jubaland, Galmudug, Hirshabelle and South West, each with their own forces and disputes with the centre.
- Clan militias and community forces
Local armed groups whose disputes drive much of the violence that affects staff directly.
Sources and what they cannot see
- Somali Dispatch, Radio Dalsan, Shabelle, Jowhar and Puntland Post
- ACLED conflict event records for Somalia
- UN OCHA Somalia situation reporting and ReliefWeb
- Insecurity Insight aid-security incident data
- UK FCDO and US State Department travel advice
Rural districts outside government-held towns are thinly reported, so quiet weeks in the record are as likely to mean no reporter as no incident. Attribution of individual attacks is often contested in the first days.
Confidence: Moderate to high on the pattern of urban attacks and route insecurity, which is consistently reported. Lower on the durability of local security arrangements in transition districts.
Operational implications
- Site selection with measurable stand-off, controlled single entry with search, blast film, a designated safe room and a drill rhythm that is actually run.
- Journey management per movement: route assessment, varied timing and routing, day-of-travel road check, named vehicle commander and written abort criteria.
- Checkpoint mapping with transporters, a documented no-payment position, scripted responses and one named negotiation point of contact.
- Crisis management plan rehearsed with named roles, tested contact tree, and a decision log kept from the first hour.
- Acceptance work: transparent selection criteria, community feedback channels, and a public position on independence that partners can repeat.