What the model is for
The Okanagan Raven model of Mission Creek runs today, from the headwaters to Okanagan Lake. This page says what it can already do, what it could do next, and what it needs from partners. It is adapted from the OBWB primer Raven for Mission Creek: A primer for local governments and water suppliers (October 2026).
What the model can already do
- It knows the system you operate. Reservoirs, release rules, utility intakes, the Stirling Creek import and crop water demand are inside the model, not only natural flow.
- It replays history. Every year from 1993 to 2024 can be rerun, including recent droughts.
- It compares choices side by side. Demand, crop watering and reservoir operations can be changed and compared in the same weather year. The difference between choices is what the model shows best.
- It shows who wins and who waits. Each run reports reservoir storage, water delivered to intakes, and days the creek falls below the fish-flow targets.
- It looks ahead. Projections from nine global climate models, under two emissions pathways, are run through to 2100 (F55).
- It has found the big levers. Of the choices tested, two matter most for late-summer flow: total water withdrawn, and how Graystoke Reservoir releases follow irrigation demand (F128, F145).
What it cannot do yet
- It cannot forecast August flow (F58).
- It does not yet reproduce the driest weeks: late-summer flows fall more slowly in the model than in the creek, so summer low-flow results are optimistic (F52, F180).
- It counts fewer days below the fish-flow targets than were observed (F85).
It is weakest where drought decisions are made, mainly because some of the information it needs sits in partners' files, not in public datasets.
What it could do next for Mission Creek
With partner data and one water season of testing, the model becomes two practical tools.
- An in-season drought tool for water suppliers. On April 1 and May 1 it gives the odds that spring and summer runoff will be in the driest third of years. It then lets managers rehearse the season, for example deficit watering for a crop from August 1 or holding more in storage, and see the effect on the reservoir, the intakes and the creek first.
- A planning tool for local government and the Water Use Plan. It tests the long game: new development and changing farmland, larger or smaller reservoir releases, a different fish-flow target, a hotter 2050s climate. It is the shared science base for an ONA-led Mission Creek Water Use Plan.
Who asks what
| Who | Question the model helps answer |
|---|---|
| Water suppliers | When should we move to the next drought stage, and what does waiting two weeks cost? |
| Local governments | Can this watershed supply planned growth in a 2050s drought year? |
| Growers | Which crops can be carried on deficit water, and for how long? |
| ONA and fisheries | How many days below fish-flow targets does each choice add or remove? |
| Provincial regulators | What does full use of existing licences do to the creek and the lake? |
What it takes
The fastest improvements are records partners already hold:
- Reservoir storage and release records after 2010
- Metered intake and diversion records
- Current crop and irrigation mapping
- Upland snow and rain measurements on the plateau
- Fish-flow targets agreed with ONA, with Syilx data used on Syilx terms
Our ask:
- Come to the partner workshop in early 2027 and bring your toughest drought decision.
- Share the records that make the model honest.
- Name one person to test the in-season tool through the 2027 water season.
Sources: Jatel, N. (2026). Raven models, data and gaps for drought decision support: A preliminary assessment for Mission Creek (Draft v0.13, revision 10). Okanagan Basin Water Board. Decision log rows as cited.