logo
Ecosystem Adaptation & Nature-Based Solutions

From Drain to Spine: What Nairobi’s Rivers Framework Gets Right, and What Will Decide Its Success

Nairobi has a new blueprint for turning its three degraded rivers into green infrastructure. Its diagnosis is sharp and its priorities sound. Comparisons with rivers restored elsewhere show that success will turn on four things the framework only begins to settle: sewer connections, secure tenure, long-term maintenance finance and measurable outcomes.

By AFRIHECOM Research Team24 October 20269 min read

Illustration: AFRIHECOM. A restored river corridor with native trees and reeds running through the city: the vision of rivers as green infrastructure.

32% → 20%
forest cover across Nairobi, 1985–2022, as sealed surfaces rose from 7% to 26%
Nairobi Rivers NbS Framework, 2026
59 m → 12.5 m
width of the Mathare River channel and riparian reserve at one stretch, 2005–2025
Nairobi Rivers NbS Framework, 2026
15
priority clusters, with two catalytic projects budgeted at US$14m over 36 months
Nairobi Rivers NbS Framework, 2026
4.1%
of target sewer connections recorded at completion of an earlier AfDB-financed Nairobi rivers sewerage project
AfDB completion report

Key takeaways

  • The framework’s diagnosis is sharp: forest cover fell from about 32% to 20% and sealed surfaces rose from 7% to 26% in 1985–2022.[2]
  • Its strengths are risk-based prioritisation, equity built into design, links to affordable housing and a phased start with two catalytic projects.
  • Comparators from Seoul to Durban show that river restoration can cool cities and cut flood risk, but is costly and depends on maintenance.[7][12]
  • Success will turn on household sewer connections, secure tenure and resettlement, funded maintenance and monitoring of heat and health outcomes.
  • Our view: an excellent map; delivery should be measured by connections, cooler streets and fewer people displaced.

On 30 July 2026, in Nairobi, the Nairobi Rivers Commission and the World Resources Institute launched the Nairobi Rivers Nature Based Solutions Strategic Framework, alongside a WRI brief on flooding in Nairobi’s informal settlements.[1] The framework, owned by the Commission, sets out how riparian restoration, wetlands, tree corridors and green–grey infrastructure could reduce flood, heat, biodiversity and air-quality risks along the Nairobi, Ngong and Mathare rivers.[2] It describes itself, candidly, as “a strategic prioritization tool rather than a detailed feasibility study”.[2] That is the right lens for reading it.

A clear diagnosis: the rivers have been squeezed

The framework’s own land-cover analysis is stark. Between 1985 and 2022, forest cover across the city fell from an average of 32% to about 20%, while impervious surfaces rose from 7% to 26%.[2] On one stretch of the Mathare River, the combined width of channel and riparian reserve shrank from about 59 metres to 12.5 metres between 2005 and 2025, largely through encroachment.[2]

Figure 1

In under four decades, sealed surfaces have nearly quadrupled as forest has retreated

Share of Nairobi’s land area, 1985 and 2022 (framework’s own land-cover analysis)

Forest cover
1985
2022
32% → 20%
Impervious (sealed) surfaces
1985
2022
7% → 26%
Source: Nairobi Rivers NbS Strategic Framework (2026), land-use change assessment, 1985–2022.[2] Forest cover is reported as falling “from an average of 32% to approximately 20%” and impervious surfaces as rising “from 7% to 26%”.
View the data as a table
Land cover19852022
Forest32%~20%
Impervious surfaces7%26%

The framework defines a Nairobi Rivers zone extending 60 metres from the high-water mark on each bank: a 30-metre riparian corridor and a 30-metre development area, declared a Special Planning Area in March 2025.[2] Across the wider study area the framework maps around the three rivers, residential land takes 41% of the area; forest and recreational land together take less than 9%; open water, 0.03%.[2] Across 378 cities monitored by the UN, open public space took only about 2% of urban land in 2025; with streets, the total was about 17%, far below UN-Habitat’s recommended 30–45%.[3]

Figure 2

Across the framework’s mapped river study area, residential land outweighs green space almost five to one

Land use across the framework’s mapped Nairobi river-basin study area (CSUD 2010 data updated with 2025 OpenStreetMap), % of area

Residential
41.36%
Transportation
9.78%
Special purpose
7.8%
Industrial
7.41%
Educational
6.16%
Forest
5.52%
Agriculture
4.48%
Public purpose
4.24%
Undeveloped
3.96%
Recreational
2.92%
Commercial
2.13%
Public utility
0.39%
Water bodies
0.03%
Source: Nairobi Rivers NbS Strategic Framework (2026), basin land-use distribution.[2] Green categories (forest and recreational) are highlighted. The mapped area is wider than the 60-metre Nairobi Rivers zone, which Nairobi City County declared a Special Planning Area in March 2025.
View the data as a table
Land useShare of study area
Residential41.36%
Transportation9.78%
Special purpose7.8%
Industrial7.41%
Educational6.16%
Forest5.52%
Agriculture4.48%
Public purpose4.24%
Undeveloped3.96%
Recreational2.92%
Commercial2.13%
Public utility0.39%
Water bodies0.03%

What the framework gets right

It prioritises by overlapping risk. Fifteen clusters were delineated where two or more climate risks coincide, and each is matched to land-use typologies and a menu of 11 NbS tools, from riparian floodplains and creek daylighting to rain gardens, green roofs and constructed wetlands.[2] This is how leading cities have approached river corridors: Singapore’s Bishan–Ang Mo Kio redesign and Medellín’s green corridors both targeted specific hazards (flooding at Bishan, urban heat islands in Medellín) rather than generic greening.[4][5]

It treats equity as design, not an afterthought. Gender equality and social inclusion run through every stage: mandatory GESI assessments, at least 50% women and youth on basin committees, cash-for-work, accessible early warning and explicit safeguards against forced eviction.[2] Its review of 13 past interventions suggests that community-led and NGO-supported projects showed “higher social sustainability than large-scale, engineering-driven programs”.[2]

It links river restoration to housing. The framework screens 21 government affordable-housing sites and argues for building nature-based features in before construction rather than retrofitting them later.[2] With the Nairobi River regeneration programme planning tens of thousands of homes on reclaimed land, this is potentially the framework’s most consequential idea.[6]

It starts small and specific. Two catalytic projects, upstream in Kangemi–Kawangware and midstream in Kariobangi–Korogocho, are designed to test the approach over 36 months before replication.

Table 1

The framework’s two catalytic projects

Objectives and budgets as set out in the Strategic Framework

ClusterMain risksHeadline targetsBudget
Kangemi–Kawangware (upstream)Heat, biodiversity, social vulnerability; 4.55 km of river and 47 schoolsRestore 52.8 ha of riverbank; plant 300,000 native trees; 1,500 green jobs; 5,000 householdsUS$6.0m over 36 months
Kariobangi–Korogocho (midstream)Flooding and pollution beside the Kariobangi treatment works10 km of buffers; 8 km of vegetated levees protecting about 25,000 people; 5 km of re-meandering and 50 ha of wetlands; BOD reduced from 60 to 40 mg/LUS$8.0m over 36 months
Source: Nairobi Rivers NbS Strategic Framework (2026), catalytic projects 1 and 2.[2] The remaining 13 clusters are planned for years 4–10 without published cost estimates.

Lessons from rivers restored elsewhere

River restoration in cities is not new, and the evidence on what it can deliver is now substantial.

Table 2

What urban river restoration has delivered elsewhere

Selected comparator projects: scale, cost and reported outcomes

ProjectScale and costReported outcomes
Seoul, Cheonggyecheon (2005)5.84 km of buried stream daylighted; about US$380m3.3–5.9°C cooler than a parallel road; species counts up from 98 to 626 (2003–08); small-particle pollution down 35%; designed for a 200-year flood[7]
Singapore, Bishan–Ang Mo Kio (2012)2.7 km concrete canal turned into about 3 km of meandering river; S$76.7mAbout 40% more conveyance capacity; biodiversity up about 30%; about 15% cheaper than a rebuilt concrete canal[4][8]
Medellín, Green Corridors (2017– )30 corridors along roads and streams; US$16.3mCity reports a 2°C reduction in the urban heat island; maintained by about 150 gardeners[5]
Kigali, Nyandungu Eco-Park (2022)121 ha of degraded urban wetland restored; Rwf 4.5bn17,000 trees of 55 indigenous species; about 4,000 green jobs[9]
Dar es Salaam, Msimbazi Basin (2022–28)US$260m total, including US$200m from IDAAims to reduce flood exposure for more than 300,000 people; US$72.1m spent by February 2026[10][11]
Durban, Transformative Riverine ManagementProgramme for 7,400 km of rivers; about US$470mBusiness-case estimate of a minimum benefit–cost ratio of 1.8–3.4; at least 10,000 jobs targeted; maintenance by local co-operatives[12]
Addis Ababa, Beautifying Sheger (first phase to 2020)56 km of riverside corridor; about 29 billion birrThousands of residents feared displacement in the second phase[13]
Figures as reported by each source; several are city- or agency-reported and not independently evaluated. Costs are in nominal terms.

Three lessons stand out. First, the most celebrated projects were expensive and engineering-intensive: Cheonggyecheon cost about US$380 million for under 6 km.[7] Nature-based does not mean cheap, although Bishan–Ang Mo Kio came in about 15% below a rebuilt concrete canal.[8] Second, maintenance is the programme: Durban’s model pays local co-operatives to keep rivers clear, and a cost–benefit analysis for its business case estimated a minimum benefit–cost ratio of 1.8–3.4.[12] Third, displacement can undo legitimacy: Addis Ababa’s riverside programme drew criticism for prioritising development over residents.[13]

Four things that will decide success

1. Sewers and connections, not only river banks

No amount of riparian planting will clean a river that receives raw sewage. A study of the Ngong River, sampled in 2021, found biochemical oxygen demand of up to 450 mg/L and E. coli counts above national (NEMA) standards.[14] The framework’s midstream target is to reduce BOD to 40 mg/L.[2] History counsels caution: an earlier AfDB-financed Nairobi rivers sewerage project built 63.3 km of trunk sewer but had recorded only 4.1% of its target of 72,553 sewer connections at completion, though connection records were incomplete.[15] The current regeneration programme is laying twin trunk sewers along both banks;[16] the decisive metric will be the number of households actually connected.

2. Tenure and resettlement

The 2024 enforcement of the 30-metre riparian setback saw some 40,000 households ordered out of the corridor, with one-off support of KSh 10,000 per household.[17] The WRI brief notes that cleared land was often left without redevelopment plans, inviting re-encroachment.[18] The framework’s safeguards are sound in principle but carry no resettlement plan, budget or numbers.[2] Without secure, nearby housing options, NbS on the riverbank will be contested ground.

A riverbank cleared without a plan is not a restored river. It is a vacancy waiting to be refilled.

3. Paying for maintenance

Only the catalytic projects are costed: US$14 million over three years.[2] The national regeneration programme was launched at KSh 50 billion.[6] Neither figure covers decades of operation and maintenance, which the framework proposes to fund through payments for ecosystem services, user fees and a Nairobi Rivers Resilience Fund.[2] Those mechanisms need amounts, sources and legal form before the catalytic phase ends.

4. Measuring what matters

The framework’s indicator set is strongest on outputs (kilometres of buffer, seedlings at 75% survival, households protected) and on water quality.[2] It has no target for heat reduction, despite heat being one of its four risks, and no health indicators. Field data show why heat matters: in December 2015–February 2016, daily maximum temperatures in Nairobi’s informal settlements averaged 3.1°C above the official weather station, and 4.8°C above it in Mukuru, the least vegetated settlement studied.[19] The IUCN Global Standard for NbS offers a ready template of eight criteria, including adaptive management and trade-offs.[20]

Figure 3

A 30-metre buffer removes about three-quarters of nitrogen overall, but under half when it arrives as surface runoff

Riparian buffer width associated with given nitrogen-removal effectiveness, metres (meta-analysis of 40 studies)

50% nitrogen removalall studies
≈3 m
75% nitrogen removalall studies
≈28 m
90% nitrogen removalall studies
≈112 m
75% removal, surface flow only
≈118 m
Source: Mayer et al., US EPA (2005), review of 66 width–effectiveness relationships.[21] Kenya’s riparian setbacks are typically 30 m; the Nairobi corridor adds a further 30 m development zone. Width explained only part of the variation (R² ≈ 0.14): vegetation, soils and flow paths matter as much. These are temperate-zone studies; tropical urban evidence is thin.
View the data as a table
EffectivenessWidth
50% nitrogen removal≈3 m
75% nitrogen removal≈28 m
90% nitrogen removal≈112 m
75% removal, surface flow only≈118 m

A final technical point concerns buffer width. Kenya’s riparian setbacks are typically 30 metres, though guidelines on widths are unclear.[18] The international evidence suggests buffers of that width can remove roughly three-quarters of nitrogen overall, but less than half where nitrogen arrives as surface runoff, the dominant pathway in a sealed city.[21] Vegetation structure, soils and source control matter as much as width.

Our view

AFRIHECOM Research Centre welcomes the framework as the most coherent plan yet for Nairobi’s rivers. It is evidence-led, equity-conscious and specific about where to start. We would urge the Commission and its partners to:

  1. Track household sewer connections as the headline metric of river recovery, alongside BOD and E. coli, published quarterly by reach.
  2. Agree a resettlement and in-situ upgrading protocol before works begin in each cluster, with budgets, timelines and grievance mechanisms.
  3. Cost the full ten-year programme, including operation and maintenance, and give the proposed resilience fund a legal form and revenue sources.
  4. Add heat and health outcomes to monitoring, using low-cost temperature sensors and health-facility data in the catalytic clusters.
  5. Commission independent evaluation of the two catalytic projects against a baseline, so that replication rests on evidence rather than enthusiasm.

Done well, Nairobi’s rivers could become the city’s green spine. The framework has drawn the map; delivery will be judged by what changes for the people who live along the banks.

References

  1. World Resources Institute. New WRI research shows Nairobi’s flooding crisis spans beyond rainfall, calls for a new approach to urban resilience (press release). 30 July 2026. africa.wri.org
  2. Nairobi Rivers Commission, with technical support from World Resources Institute. Nairobi Rivers Nature Based Solutions Strategic Framework: a toolkit for providing practical guidance, tools, and resources to design, implement, and scale NbS projects. 2026; launched 30 July 2026. Launch release: africa.wri.org
  3. United Nations. Sustainable Development Goals Extended Report 2026: Goal 11. 2026. unstats.un.org
  4. C40 Cities. Singapore: bio-engineering works at Bishan–Ang Mo Kio Park to prevent urban flooding. September 2018. c40.org
  5. PreventionWeb (Thomson Reuters Foundation). Colombia’s Medellín plants green corridors to beat rising heat. 28 July 2021. preventionweb.net
  6. State House Kenya. President Ruto launches KSh50 billion Nairobi River regeneration project. 10 March 2025. president.go.ke
  7. Landscape Architecture Foundation. Cheonggyecheon Stream Restoration Project (Landscape Performance Series case study). 2011. landscapeperformance.org
  8. American Society of Landscape Architects. Kallang River–Bishan Park (2016 Professional Awards). asla.org
  9. KT Press. Nyandungu Eco-Park opens to the public. July 2022. ktpress.rw
  10. World Bank. Msimbazi Basin Development Project (project brief). worldbank.org
  11. The Guardian (Tanzania). Msimbazi project to redefine city look, unlock business and transform daily lives. 26 May 2026. ippmedia.co.tz
  12. GIZ. Transforming river management and shifting mindsets in South Africa and beyond. 2022. giz.de
  13. Climate Home News. Addis Ababa riverside project gives priority to development over residents. 12 March 2020. climatechangenews.com
  14. Ngatia MN, Kithiia SM, Voda M. Effects of anthropogenic activities on water quality within Ngong River sub-catchment, Nairobi, Kenya. Water 2023;15(4):660. mdpi.com
  15. African Development Bank. Kenya: Nairobi Rivers Basin Rehabilitation and Restoration Program, Sewerage Improvement Project (project completion report). afdb.org
  16. The Star. Nairobi River clean-up races to beat the January deadline. 16 October 2026. the-star.co.ke
  17. Citizen Digital. Eviction from Nairobi rivers corridor to continue, says Interior Ministry. 27 May 2024. citizen.digital
  18. Mwangi CM, Mungai O, Nduru J, Samuel W, Wanjohi-Opil HN, Munyambu P, Mahendra A, Shrivastava V, Akinyi E. Flooding in Nairobi’s Informal Settlements: Advancing Equitable and Resilient Solutions. Issue brief. World Resources Institute; July 2026. doi:10.46830/wriib.25.00024. wri.org
  19. Scott AA, Misiani H, Okoth J, et al. Temperature and heat in informal settlements in Nairobi. PLOS ONE 2017;12(11):e0187300. journals.plos.org
  20. Nature-based Solutions Initiative. IUCN Global Standard for Nature-based Solutions (launched 23 July 2020; 8 criteria, 28 indicators). naturebasedsolutionsinitiative.org
  21. Mayer PM, Reynolds SK, Canfield TJ, McCutchen MD. Riparian Buffer Width, Vegetative Cover, and Nitrogen Removal Effectiveness: A Review of Current Science and Regulations. EPA/600/R-05/118. US Environmental Protection Agency; 2005. epa.gov

This article draws on the Nairobi Rivers NbS Strategic Framework and WRI issue brief launched on 30 July 2026, and on other publicly available sources. Figures are reported as published and may differ in date, scope and definition.

Leave A Comment