Flow Data Interpretation for Steelheaders

Flow is a vitally important factor for understanding fish movement in the grand scheme of steelhead fishing. Flow is arguably the single-most important variable when taken in the context of fish movement, just as temperatures dictate when fish begin to run, when they spawn, where they locate within a stream, and when they are the most vulnerable to bait or fly presentations. But even there, you'll be looking for a certain type of "flow".

Many people are initially baffled when it comes to interpreting flow charts from USGS and others. Fortunately, the folks at Minnesota Steelheader can walk you through interpretation and when you are done with this article, you’ll be able to read the charts like a pro.

To understand the charts, it is helpful to understand a few terms that are used on any chart you might encounter:

CFS: Cubic Feet per Second or "CFS," is the rate of flow in streams and rivers. It is equal to a volume of water one foot high, one foot long and one foot wide (a cube) flowing a distance of one foot in one second as measured by one of the various instrument packages. The higher the number, the higher the water volume flowing past the station at a given point in time. By way of comparison, 1 cubic foot per second of water volume is equal to approximately 7.5 gallons per second of water volume.

Discharge: The total volume of water which passes a given location within a given period of time expressed in cubic feet per second. Think of discharge as the total CFS measurement passing a distinct point in a tributary (wherever the station is located); the higher the number, the higher the flow. Discharge readings are displayed in CFS typically as in Figure 1 below:

Figure 1

Exceedence: This is the tough one to wrap your brain around however, exceedence is an important way to describe the percentage of time for which a measured stream-flow is greater than, equal to, or less than all of the measured stream-flow values taken on a particular date. Exceedence is used because discharge is not normally distributed, i.e. flows go up and down, but don’t typically follow classic bell-shaped curves. Also, low-flows generally have high exceedence percentages, and high-flows generally have low exceedence percentages.

    Low-flow events have a high exceedence percentage because most of the time, the measured flow exceeds the low flow value. Similarly, high-flow events generally have low exceedence percentages because most of the time, measured flows are lower than the high-flow value. In other words they rarely exceed the value given.
Exceedence then is useful in understanding when a tributary is low, normal or high for a given time of year, although the mean (read average) value is a more intuitive way to get a feel for where a tributary is at in the context of a particular time of year. This is arguably the most important concept to grasp with regards to fishing migratory salmonids. More on that later. There are two measurements generally given for exceedence: 20% and 80%.

Figure 2: The Exceedence chart for a hypothetical stream

In the example illustrated by the Figure 2, a hypothetical stream has an average (mean) flow of 45 cfs, and a median flow of 38cfs. The 50% exceedence value is the median flow of 38cfs. This means that the median flow -the mathematical “middle” of all measured flow values for this date- will be greater than or equal to 38cfs 5 out of every 10 years. Another way to put it is that one-half of all measurements were higher, and one-half of all measurements were lower than 38cfs.
The stream has an 80% exceedence flow of 25cfs. An 80% exceedence flow means that the measured flow will be equal to or greater than 25cfs 8 years out of 10. The stream has an 10% exceedence flow of 72cfs. A 10% exceedence flow means that the measured flow will only be equal to or greater than 72cfs 1 year out of 10.

O.k., I know, you’re looking at 10% when I said 20%. Generally USGS only provides the 20% and 80% exceedence values, but you get the point, and we're talking about a hypothetical stream anyway. The other thing to remember is that when any chart provides exceedence, mean and median values, those apply only to the date you are looking at. Yesterday was different and so too will tomorrow’s values be.

So why is this important? When looking at any USGS chart, you need to be able to interpret what is low, normal or high for that date AND that specific time of year. Most fall readings on any given tributary are vastly different from spring readings, particularly on the North Shore. What might seem like absurdly low flow in fall in comparison to spring flow, could really be a discharge rate that will draw fish like crazy. This is because migratory fish “adjust” as it were to individual tributary’s annual flow cycles. Somehow the fish seem to know just what works for them in terms of flows with regards to time of year on a specific tributary. On one upper shore trib I fish, a flow rising to 11-15cfs in fall will draw Coasters, Pink, Coho & Chinook Salmon, while a spring steelhead would simply yawn and wait for flow values in 85-100cfs range.

Gage Height: The height of the water surface above the gage datum (zero point). Gage height is often used interchangeably with the more general term stage, although gage height is more appropriate when used with a gage reading. Gage heights generally correspond to distinct discharge readings and are how Rating Curves are developed. This is important to understand because many people think that gage and CFS are not related, they are. A distinct gage reading corresponds to a distinct CFS rate. In a perfect world, they would always be the same however, in the real world they are not quite the same, but that gets into a technical argument regarding stream cross-section and is not highly relavent to the discussion. You Cliff Clavin types can feel free to e-mail me if you want to know why... At any rate, it really doesn't matter whether you intuitively understand stage or CFS better, the point is to be able to know what readings mean that it's time to go fishing. Figure 3 illustrates Gage Height and is the corresponding sonde reading to Figure 1.

Figure 3:

Mean: The “average” in plain English. For a data-set such as discharge, the mean is the sum of all readings divided by the total number of observations for a given date. This value is then used as the basis for all other annual data. Figure 4 shows Means, Medians and Exceedence as depicted on USGS charts in the following manner:

Figure 4:

Here you can see that for May 11th, the flow is actually quite low. Well below the average of 219cfs, and very close to the lowest historically recorded flow value. This graphic is very telling with regards to what’s going on with the fish.

Median: The number dividing the higher half of a discharge data-set from the lower half. The median of a list of numbers is simply calculated by arranging all the observations from lowest value to highest value and using the middle value. Medians are depicted pictorially on USGS charts as a Triangle (See Figure 1) as well textually in the summary method depicted in Figure 4.

Sonde: A type of instrument package ideal for profiling and monitoring water conditions in lakes, rivers & wetlands. Sondes may have multiple sensors that record a range of water quality data such as: Gauge Height, Discharge, Turbidity (how much dirt etc. is suspended in the water), Temperature, Dissolved Oxygen etc. If the sonde has “on board” battery power, it can be left unattended for weeks at a time, with measurement parameters sampled at pre-programmed intervals and data securely saved in the unit's internal memory. Some sondes transmit data automatically while others require physical retrieval from the water. The sonde's information is downloaded to a computer and specialized software is used to graph and interpret the data.

Getting Started
There are three very good resources for obtaining flow on North Shore and other area tributaries. The first is the US Geological Survey Site
This link is specific to Minnesota and displays tributaries in a table format. To get other State’s tributary data, simply select that State from the dropdown menu located in the upper right-hand corner of the screen, then click on the trib you wish to view. A useful feature of this table is that you can quickly scroll through all available tributaries to see what they are doing once you get a feel for each tribs flow cycle.

The second resource is: Lake Superior Duluth Streams
This link has data on a few lower shore tribs along several upper-shore tribs. The data is displayed in a slightly different format, but still shows discharge and stage (gauge height). These are important, but greatest tools available here are temperature and precipitation. Another great feature of this site is the Data Visualization Tool and Color Data Plot. It does require Java to display, but you can download it from the site. If you don’t want Java on your machine, you can still load the data in an Excel format although temperatures will be displayed in degrees Celsius.
The third resource is: MPCA - MN DNR Cooperative Stream Gaging Program. This one is great because you can get limited temp data along with access to at least one mid-shore tributary.

One note on available Minnesota tributary data: Because of the relative lack of available sondes, you will have to make assumptions about flows between stations. Generally this means watching weather stations in addition to looking at tribs with sondes for precipitation events. The general rule of thumb is that 15-20 miles is about the maximum distance you can assume general conditions of flow will be similar when comparing tribs with sondes to tribs without sondes. Meaning that if flow is going up or down at a given station, it will be going up or down on tribs within 15-20 miles either direction. This is why you need to look at radar in particular to get a feel for how widespread a precipitation event is.

For our discussion, we’ll us the data displayed above from the Knife River. You can follow along however by opening the USGS link and performing the following steps:

1. Open the USGS link to the Minnesota table
2. Click on the Knife River Station ID: 04015330
You should now see the graphic displays containing the gauge height backup, discharge and primary gauge height. The current display is for a full month of data. The very top of the page has selections to specify parameters for each of those displays. Generally one week (7 days) is a more useful display. This is also where you can specify that you would like the means, medians and exceedence displayed.

3. Under Available Parameters located at the top of the page, select “Graph w/Stats” then type the number 7 in the “Days” box. Now click the “Go” button. This will refresh the displays to give you the full range of available data from the latest reading going back a full seven days.

Using the examples provided above, you can see that the flow and gage readings are trending slightly downward meaning that the flow on the Knife is dropping. The actual current discharge is 37cfs, or about 277 gallons per second. This is well below the mean or average flow of 219cfs, as well as the median flow of 144. This tells you that the flows are very low for this particular time of year. You can now draw several important conclusions from the available data:
It indicates that there is probably very little fish movement, and that the Knife is about as clear as it’s going to get. You can also infer that other area tribs are also similarly low and clear, and that you’ll have to adjust your tactics and presentations to the current conditions. Fishing deeper holes and runs or broken pocket water in low light periods with long leaders and light tippets, along with small natural flies or baits and a stealthy approach will be the order of the day. You can also see by the date that the run is probably winding down, and any remaining fish have likely been pressured.

If the current reading were closer to the mean flow, the water would be more discolored. Fish movement would most likely be greater. Less stealthy approaches could be used along with shorter leaders, heavier tippets and larger brighter flies and baits should now be employed. All of these conclusions can be drawn from a simple glance at the current data and go a long way towards making any outing that much more successful.

One final note: I’ve mentioned the term flow cycles several times. Again, each tributary has it’s own natural rhythm which the fish somehow adapt to. Each tributary also has a point at which the fish seem to shut down, and looking elsewhere can make or break a day. Because this is highly variable between Spring and Fall based on the cycle, the only way to know for sure is to watch for those means along with keeping track of your observations and success, or lack thereof, while comparing this against the available flow data. On the Knife in Spring for example, a 450-500cfs reading tells me that the river is basically what is referred to as, “blown out.” In other words it is high, dirty and just about un-fishable, not to mention dangerous to attempt wading. The DNR studies also show that fish movement on the Knife pretty much shuts down as you approach 500 cfs.
And if you do see these high readings, but it's the only time you can fish, at least you are going in armed. Fishing close to the bank between the points of outside turns and the mid-point of the inside bends, or any other sheltered area with really big, really bright or really dark baits or flies is going to be the the order of the day.
Similarly, a 150cfs spring reading on the Knife tells me that it is getting pretty low based upon my own past observations. However, I’ve had some pretty good days fishing it at this level so I won’t count the Knife out, I just know I have to change tactics. In Fall, a 150cfs reading by way of comparison would put a huge smile on my face.
But what if you have never fished a particular tributary before? A good rule of thumb is that the closer a discharge reading is to the mean, the greater your chances are of finding active fish and being successful. Keep in mind that Spring and Fall means are normally very different and that the actual flow may not be as important as how close the current flow is to the mean value. This is why it is important to consider them since the means indicate how close the current reading is by comparison to where it should be for that specific time of year. Again, the fish adapt to the cycle.
Being able to interpret flow data goes a long way towards making you a more successful steelheader. Once you know how to interpret the data, you can quickly visualize what the current conditions are, as well as to formulate a plan on where to fish and what presentations to try first. And now that you know how, it’s all a matter of putting these concepts into practice.
One item of significant import for North Shore Steelheaders: Keep in mind that while flow is very important to fish movement, temperature also plays a huge role in initial run activity. Early on in late March or early April, you can have the most ideal flow in the universe, but if the temps aren't where they need to be, things can be mighty slow. That's not to say we don't get early fish forays into the tribs, particularly loopers, because we do; it's just that the initial push is highly dependant upon temperature. Here's a quote found in multiple DNR Adult Trap Operation Reports typically located under Spring Environmental Variables. "Upstream movement of adult steelhead in the spring initiates when maximum daily water temperatures exceed 4.4*C (39.92*F) and mean daily water temperatures exceed 3.3*C (37.94*F)." Schreiner et.al. Minnesota DNR. Once daily temps hit these ranges and maintain, flow once again takes over as the factor that dictates upstream fish movement.
Regards and Good Fishing!
NMF