Methodology

How our river scores work

Every score says what it is based on, how sure we are, and where the numbers came from.

The short version

A flow number on its own does not tell you much. Fifty cfs is a normal October day on one river and a drought on another. So we rank each river against its own history for the same calendar day, then add how the flow is moving, how warm the water is, the weather forecast at the river, and which hatches are in their usual window. The result is a score from 0 to 10, a confidence level, and a plain list of reasons.

What goes into a score

Safety comes before the number

Confidence

Confidence shows how much of the score rests on fresh, measured data. It drops when a gauge has no temperature sensor, when there is no forecast, when the reading is a few hours old, or when the gauge has a short history. Rivers without a water temperature reading cannot score above 6.9 and cannot be the day's top river. If a gauge reports exactly the same daily flow for two weeks or more, we say so and lower confidence: a steady reservoir release looks like that, and so does a stuck sensor. Charts only draw days the agency actually reported; missing days are left blank.

Outlook

Tomorrow and the weekend use that day's forecast at the river, with the flow assumed to hold. Releases from reservoirs can change without notice, so outlook confidence is always lower than today's.

What we never do

Rivers outside Colorado

Our river pages for other states, and for the Colorado rivers that do not have a full page here yet, use a simpler score built only from what the USGS gauge measures. It does not yet include the weather forecast, flow trends, or hatch windows that go into our Colorado scores, so read it as a first look at the water rather than a full fishing report.

Water-quality lines. Some states give anglers no temperature advice, but their environmental agency sets a water-quality limit for streams that hold cold water fish. For Wisconsin, Oregon, Michigan, New Jersey, North Carolina, Tennessee and New Mexico we use that limit, and the river page calls it the state's water-quality limit for cold water fish. It is a rule for keeping streams healthy, not a fishing closure, so the most it can do is hold the score to 5 or less. It never counts as a stop line.

We check each limit the way the state writes it. Most are compared with the day's highest reading. Oregon's is an average of the daily highs over 7 days, so we average the last 7 daily highs at the gauge, and if 7 days in a row are not available the limit is left out. Wisconsin's limit changes by month, and Michigan's applies only in June, July and August. A limit only counts where it applies: if the state puts a stream in a different class, or we could not confirm its class, temperature stays out of the score, as before. New Hampshire and Connecticut have no number like this, so their rivers still cannot score above 6.9.

The same safety-first rules apply. Water at or above the agency's stress line holds the score to 5 or less, and water at or above its stop line holds it to 2 or less. When a state's rule needs several warm days in a row, as Montana's 73°F line does, we check that many days of readings before applying it. Flow at or below the state's own low-flow criterion holds the score to 4 or less. Each page says in plain words when one of these limits applied.

If a gauge has no reading, or its newest reading is more than 48 hours old, the page says "Not available yet" instead of showing an old number. USGS marks recent readings as provisional, which means they can still be revised after review, and we show them as reported. Every daily national score is archived along with its inputs, so it can be checked against what the river did next before we change the method.

Sources

See today's river scores