Including the ones that failed, because a project list containing only successes tells you the organisation edits its list.
Sites
| Site | District | Started | Status | Notes |
|---|---|---|---|---|
| Biharwe bank, upper | Mbarara | 2022 | Established | First site. 400m of buffer, survival above 70%. |
| Biharwe bank, lower | Mbarara | 2023 | Active | Extension downstream. Replanting after dry-season losses. |
| Katete crossing | Mbarara | 2023 | Monitoring | Monitoring point RW-02. No planting; used as an urban reference. |
| Rwentondo stretch | Rwampara | 2024 | Active | Buffer agreements with 11 households. Nursery on site. |
| Rwampara wetland margin | Rwampara | 2024 | Active | Rest period agreed with 6 households. Papyrus recovering. |
| Nyakayojo trial | Mbarara | 2023 | Failed | Planted without confirming who farmed the land. Lost entirely within a season. |
| Isingiro school tanks | Isingiro | 2025 | Established | Three institutions. Annual maintenance checks ongoing. |
| Control stretch RW-05 | Rwampara | 2024 | Monitoring | Deliberately unrestored. Needed to know whether our work causes anything. |
Why we keep a control stretch
RW-05 is a length of bank we have deliberately not touched. Without it, an improvement at a restored site tells you nothing about whether the restoration caused it. It costs us nothing and it is the difference between a claim and a finding.
What each status means
| Status | Definition |
|---|---|
| Survey | Walked and recorded, nothing planted. Some sites stay here permanently and that is a legitimate outcome. |
| Active | Planted within the last twelve months. Survival is not yet known, and no success is claimed. |
| Established | Past the twelve-month count with survival recorded. The figure is published whatever it says. |
| Monitoring | No planting. The site exists to produce readings, including the control stretch that is deliberately left alone. |
| Failed | Lost, with the reason recorded. Failed sites stay in the directory and in the denominator of every survival figure we publish. |
How a site gets onto this list
A stretch is proposed, usually by a field officer or a district environment officer, and walked on foot at the waterline. The survey record — bank condition, current cultivation, who farms it, GPS points — exists before anything else happens, and a good number of proposals stop here.
Then the conversations. Buffer agreements are made with the households actually farming the bank, in Runyankole, in writing, and before any seedling is ordered. This is the step that takes longest and the step that most restoration projects skip; skipping it is what killed Nyakayojo.
Planting happens when the season has actually broken, not on a date fixed in January. Everything that goes into the ground is counted at the site on the day, and that count is the denominator for everything reported about the site afterwards.
Then the returns, at six months and twelve. A site moves from Active to Established on the strength of a count, not on the strength of how it looks.
Nyakayojo is on this list on purpose
We planted a trial stretch in 2023 without confirming who farmed the land, and lost it entirely within a season. It cost us a cohort of seedlings and taught us the thing this organisation now does differently from most: the negotiation is the intervention, and the planting is the part that happens afterwards. Removing it from the directory would improve our numbers and delete the most useful thing on this page.
Sites are added as they start
The directory is maintained in the admin system and updated when a site opens, changes status, or fails.