How To Tell Spotted Bass from Largemouth at a Glance

Some anglers have difficulty telling spotted bass (Kentucky bass) from largemouth, and since in many places, including my home Ozark streams, the regulations are different for the two species, every angler should know how to tell them apart.  Even though many fishermen release their bass no matter what the species, in the Meramec river system in Missouri, spotted bass are invasive and have caused a serious decline in native smallmouth population, and those who care about the smallmouth should actually keep and kill all the spotted bass they legally can.

I remember the first spotted bass I ever caught.  I was in my early teens, perhaps 12 years old, and my grandfather had taken me to Whitewater River to trap minnows.  It was a long drive just to trap some bait, but Grandpa grew up in the area and I'm sure he was just wanting to revisit some old haunts.  I was, as always, fishing while he watched his minnow traps, and caught a bass that even back then, I knew was different from either the largemouth or smallmouth I'd caught previously.  Granpa just called it a "lineside", but I knew it wasn't a largemouth, and when I got back home I tried to find out what it was.  The Conservation Department had published a book on Missouri game fish with black and white photos, and it was there that I saw that my bass was a spotted bass.

Maybe, because I'm an artist and was an artist of sorts from the time I was a small child, I had trained myself to see details that others miss.  But I've always been able to tell spotted bass from the very similar appearing largemouth at first glance, as I lift them out of the water.  Indeed, I've often said I can tell them apart blindfolded, by feel.  Spotted bass scales are slightly saw-toothed on the outside edge, while largemouth scales are smooth, so when you feel them, the spotted bass feels slightly rough, the largemouth very slick.  But you don't have to feel them to see the difference.

It seems that some fishermen misidentify them because they have learned to feel the tongue.  Spotted bass have a small patch of teeth on their tongue, a spot that is slightly raised and feels like coarse sandpaper.  Largemouth supposedly don't.  But I have a largemouth mounted and hanging on my wall that has a tooth patch.  It seems about 10% of largemouth have a tooth patch.  It is raised like spotted bass but the teeth are much smaller so it feels only slightly rough, but it can be there.

The other thing that many anglers seem to remember is that spotted bass have smaller mouths.  Their upper jaw (maxillary) extends only to about even with the rear of their eye, while largemouth maxillaries reach well past the eye.  However, the mouth must be almost fully closed to check this characteristic, and if the largemouth is under about 10 inches, its maxillary might not be obviously past the eye.

So how do I tell them apart just at a glance (and I'm certainly not the only fisherman who can do so)?

Below is an illustration I did of both species.  I used good, clear photos to begin with, and enhanced the details--photos of fish usually have reflections and glare from their wet, slimy skins that often obscure details.  What is immediately obvious is that the two fish actually do look quite similar.  They are both basically greenish, and they both have a blotchy horizontal bar running from eye to tail down the middle of their sides.  You can also see in this illustration that the maxillary is slightly smaller in the spotted bass, but the size difference is really not too obvious, and could be missed or misread.

The upper fish is the spotted bass, the lower is a largemouth.  Now, here is the same illustration, with the differences I find very noticeable and that I use to identify the two species.

The area of the body marked A on both fish is the first thing I see.  Spotted bass have obvious rows of small spots in this area that, at a glance, form thin, dark lines.  Largemouth may or may not have some spots, but they are never as dark and noticeable and they do not form those thin lines.

The second thing I see is the area of the body marked B.  This is the portion between the lateral line and the dark band.  On spotted bass, the scales in this area have dark edges and each scale is plainly visible as a result.  On largemouth, these scales are not edged in darker colors and thus are not obvious.

Those two things immediately tell me whether the fish I'm lifting out of the water is a spotted bass or a largemouth.  I've also noted a couple other differences.  The notch between the spiny dorsal fin and soft dorsal fin, marked C, is deep on largemouth, with the membrane barely connecting the two fins.  On spotted bass it is significantly shallower, the membrane broadly connecting.  

And finally, the area marked D.  The scales on the lower cheeks and in front of the pectoral and pelvic fins on spotted bass are very small, much smaller than those on the upper gill cover and the rest of the body.  On largemouth, these scales are larger, a little smaller than the other scales but not nearly as small as they are on spotted bass.

Understanding and Using River Gauges, Part 3

In many ways, the rivers of the Ozarks are all similar.  They have much the same characteristics, so much of what holds true for one stream is true for all of them.  That makes using the river gauges even more useful, because there are universal "rules of thumb" that you can rely upon in attempting to figure out the water conditions on a given stream.

The first thing you need to remember is that the "normal" flow for any of these streams is considerably higher at certain times of the year than at other times.  Typically, from late winter through late spring, these rivers are at their highest "normal" flows.  By June those normal flows are dropping, until August through October the streams are at their lowest normal flows.  So a flow that might seem quite high in August will be close to normal in April.  To get the most out of the river gauges, it helps to know what the normal flow is in April and in August.  In the case of the Jacks Fork near Mountain View gauge I referenced in Part 2, I went back to the beginning and end dates, and changed them to a week in mid-April of this year, to see what the median flows were during that time period:

Look at the little triangles that denote the median flow, not the blue line that shows what the river was actually doing that week.  They show a "normal" flow between about 140 cfs and 200 cfs.  Let's call it an average of 180 cfs.  So that's about what the normal flow for the upper Jacks Fork is in the spring.  Remember in Part 2, we saw that the Jacks Fork was flowing 184 cfs on the day in August when we were investigating it.  While that seemed high for August, we now know that in the spring it would be considered a normal flow.  What that tells me is that while the river is higher than normal, it wouldn't be so high that it was up in the willow bushes.  Indeed, that's almost the last piece of the puzzle we need to know to determine the floatability of the upper Jacks Fork on the given date.

The last piece of information is one you can't get from the gauges.  I call it "Al's rule of thumb on river flows".  Since the rivers of the Ozarks are similar, a given flow in cfs that is high enough to make one stream floatable is enough water to make any stream in the Ozarks that is reasonably close to that stream in size floatable.  By looking at the flow in cfs and following my rule of thumb, you can determine immediately whether a stream has adequate water for paddling--or jet boating, for that matter.  Here it is:

Streams with median flow less than 300 cfs:

Flow less than 50 cfs--not easily floatable; a canoe or kayak loaded with a day's gear will scrape bottom in most riffles and a good portion of the riffles will have to be walked, dragging the boat.

Flow 50-75 cfs--still not easily floatable; some narrower riffles without obstacles will be runnable without scraping bottom, and if you don't care much about your boat bottom you will be able to paddle most riffles.

Flow 75-100 cfs--floatable; canoe or kayak will still scrape bottom on some of the wider riffles and split channels around islands, but the majority of riffles will be runnable without touching bottom if the right lines are chosen.

Flow 100-150 cfs--quite floatable; if the right lines are chosen, a canoe or kayak will only scrape bottom in the widest riffles and some split channels.

Flow 150-350 cfs--optimal floating level; the river will be moving well, the riffles all deep enough to run without scraping.

Flow 350-600 cfs--floatable but high; riffles all runnable, but obstacles such as sweeper and strainer trees become more difficult to avoid and the consequences of mistakes are worse.  At this flow, streams that are ordinarily too small for jetboats may become navigable by jetboats if there aren't too many obstacles.

Flow 600-1000 cfs--upper limits of navigability for most paddlers;  only get on the stream if you are very confident in your paddling abilities in heavy, fast water with obstacles.  Anything over this is probably too high and dangerous.

For streams with median flow over 300 cfs:

Flow less than 100 cfs--real drought conditions with dead low water, these ordinarily floatable and jetboatable streams will be low enough that jetboats are out and paddlers will scrape bottom in many riffles.

Flow 100-200 cfs--floatable with few problems, but jet boating may require considerable experience and somewhat of a disregard for your motor, because it will be sucking up gravel.

Flow 200-400 cfs--easily floatable, experienced jet boaters should encounter few problems except for obstacles like downed trees, but riffles with larger rocks may be dangerous and difficult to run.

Flow 400-800 cfs--floatable and usually fairly easy jetboating.

Flow 800-1500 cfs--floatable but heavy, fast water; usually very easy to float but the consequences of mistakes begin to be dangerous.  jetboating optimal.

Flow 1500-2500 cfs--at the limits of floatability; you'd better be good and prudent.  Jetboating becomes easy but mistakes could be very costly.

Flow over 2500 cfs--stay off it. The only possible exceptions to this rule are the very largest Ozark streams like lower Current and Gasconade rivers.

Not all Ozark float stream stretches are covered well by gauges.  Our Jacks Fork example is one of the few streams that are very well covered.  The Mountain View gauge, at Buck Hollow, is an excellent indicator for the river above Alley Spring.  Then there is the Alley Spring gauge, which actually is located at the bridge above Alley Spring, so it is most useful for the stretch above Alley--it's pretty exact for Bay Creek to Alley; and the Eminence gauge, which is perfect for Alley Spring to the mouth of the Jacks Fork.  But on many streams, you must understand that tributaries (or large springs) coming into the river downstream of the gauge location can render the gauge useless for the river below them.  Current River, for instance, has a gauge at the headwaters at Montauk, one above Akers, at Van Buren, and at Doniphan.  You would think the Van Buren gauge would serve well for the river down to Doniphan, but Big Spring comes in not far below Van Buren and adds an average of something like another 350 cfs.  So actually the Doniphan gauge is better for the river below Big Spring.  Since the Jacks Fork comes in well below the Akers gauge but well above the Van Buren gauge, the Akers gauge is most useful for the river between Akers and Two Rivers, the Van Buren gauge is better for the river between Two Rivers and Big Spring.

And then there are the rivers that are not served well at all by gauges.  The Gasconade in a 250 plus mile long river with two major tributaries and many smaller but significant streams feeding it, yet it only has three gauges; at Hazelgreen, at Jerome, and at Rich Fountain.  The Jerome gauge is good for much of the river downstream, and the Rich Fountain gauge gets more useful the farther downstream you go.  But Jerome is only useful upstream to the mouth of the Big Piney, just 8 or so miles above.  The Hazelgreen gauge is below the mouth of the Osage Fork, and good from the Osage Fork to the Big Piney.  But there is no gauge at all that will tell you anything about the 75 miles of floatable water above the Osage Fork.  And there isn't a gauge at all on the Osage Fork, either, and the only two Big Piney gauges are both on the lower section of the river, and aren't too accurate when considering the upper half of the floatable length of the Piney.

So if you regularly float a stream that isn't well covered by gauges, you can still make some assumptions by looking at the gauge closest to where you want to float, but they won't be perfect.  If you're floating an upper portion of a stream and the only gauge is far downstream, if that gauge is close to normal you can make a good guess that the upper river will be close to normal as well.  But what if it rained hard on the upper watershed and the upper river had a big rise that just hasn't gotten down to the gauge on the lower river?  In other words, reading gauges is often not an exact science, and you must use other sources of information.  In the case above, a weather report from somewhere in the upper watershed might tell you that there was that big rain that probably affected the upper river.

Another bit of useful info that you can't get directly from the gauges--if there has been rain in the area in the last few days, you can assume that the river may have had a significant rise.  But what constitutes "significant"?  Al's rule of thumb on rises is that if the rise is less than 6 inches and the river was near normal before the rise, the river may have gotten a little murkier than it was but probably not muddy, and fishing should still be okay.  If the rise was around a foot, there's a better chance the river got a slug of muddy water, and it might take a couple days for it to get back to normal clarity, but the chances are still fairly good that it will only be murky and fishable.  If the rise is between a foot and two feet, there's a real good chance the river got muddy, and fishing prospects aren't good.  If it's more than two feet, forget it.

If you read the gauges as I have outlined, paying close attention to the flow in cfs and using the height in feet only for determining how significant a rise was, and you remember the rules of thumb on floatability, you should be able to have a good idea of river conditions on most of the rivers of the Ozarks.

Following is a listing of gauges in the Missouri Ozarks and the stream sections for which each is most useful. An (*) in front of the gauge means the median flow of that gauge is under 300 cfs.  See above for the differences between streams with median flows under and over 300 cfs when considering "Al's rules of thumb for river flows".

Meramec River

*At Cook Station--upper river above Maramec Spring only.

Near Steelville--Maramec Spring to Onondaga.

Near Sullivan--Onondaga to Moselle Access at the mouth of the Bourbeuse.

At Pacific--mouth of Bourbeuse to Hwy. 66 State Park.

Near Eureka--below Hwy. 66 State Park.

Big River

*At Irondale--Irondale to Leadwood Access.

*Below Desloge--Leadwood Access to St. Francois State Park.

*Below Bonne Terre--St. Francois State Park to Washington State Park.

Near Richwoods--Washington State Park to Brown's Ford Access.

At Byrnesville--Brown's Ford to the Meramec.

Bourbeuse River

*Near High Gate--upper river to Noser Mill (Hwy. 185).  Note that this gauge is very far upstream and isn't really very accurate over most of the floatable portions of the upper river.

*At Union--Noser Mill to the Meramec.  This gauge is most accurate for the stretch beginning at the mouth of Spring Creek, which is about 8 miles below Noser Mill.

Huzzah Creek

*Near Steelville--only gauge on either Huzzah or Courtois, and most accurate from Highway 8 on down on the Huzzah.  It can be used to make a decent guess as to conditions on Courtois Creek, since it is very close to the Huzzah and is about the same size, so usually has much the same water conditions.

Gasconade River

Near Hazelgreen--accurate from the mouth of the Osage Fork, a few miles above the Interstate 44 bridge and Hazelgreen Access, to the mouth of the Big Piney, though keep in mind that enough springs and tributaries come into the river in that stretch that it will be flowing significantly more water by the time you reach the Piney.  It is of marginal use for the 75 miles of river above the Osage Fork, and there are no gauges covering that long reach.

At Jerome--mouth of the Big Piney to about Paydown Access.

Near Rich Fountain--Paydown to the Missouri River.

Big Piney River

Near Big Piney--accurate from about Mason Bridge Access to Fort Leonard Wood.  No gauges serve the river upstream from that stretch.

Below Fort Leonard Wood--Fort Wood to the Gasconade.

Little Piney River

*At Newburg--fairly accurate for the floatable portion of the stream.

Niangua River

*At Windyville--use this gauge for the Niangua above Bennett Spring.

Above Lake Niangua near Macks Creek--Bennett Spring to Lake Niangua.

At Tunnel Dam near Macks Creek--Tunnel Dam (Lake Niangua) to Lake of the Ozarks.

James River

*Near Springfield--above Lake Springfield.

*Near Boaz--Lake Springfield to the mouth of Finley Creek.

At Galena--Finley Creek to Table Rock Lake.

Bryant Creek

*Near Tecumseh--most accurate for the lower portion of the stream, only gauge on Bryant.

North Fork

Near Tecumseh--good for the stretch from Rainbow and Double Springs to Norfork Lake, of marginal usefulness upstream from there.  Only gauge on the North Fork.

Eleven Point River

Near Bardley--Greer Spring to the Missouri state line.  Of marginal usefulness above Greer.  Only gauge on the Missouri portion of the river.

Current River

*At Montauk--Tan Vat to Akers.

Above Akers--Akers to Two Rivers.

Above Powdermill--this is a new gauge and so far is not showing the flow in cfs.  Right now it is of marginal usefulness, but eventually it will be useful for the river between Two Rivers and Van Buren.

At Van Buren--Two Rivers to Big Spring.  It is less accurate than the Powdermill gauge will be above Logyard.

At Doniphan--Big Spring to the state line.

Jacks Fork

*Near Mountain View--Prongs to Bay Creek.

*At Alley Spring--Bay Creek to Alley.

At Eminence--Alley to Current River.

Black River

Near Annapolis--Lesterville (mouth of Middle Fork) to Clearwater Lake.

Below Annapolis--oddly enough, this gauge is only a couple miles below the one above; the only difference is that it is within the flood control pool reach of the upper river.  It often registers significantly different from the Near Annapolis gauge, but there is little reason for it.  I prefer using the Near Annapolis gauge for the upper river.  Note that many floats start on the West Fork above the mouth of the Middle Fork, and neither of these gauges is particularly accurate for that stretch.

At Leeper--Clearwater Dam to Leeper.

Above Williamsville--Leeper to Hwy. 67.

At Poplar Bluff--Hwy. 67 to Poplar Bluff.

St. Francis River

*Near Mill Creek--most accurate from the mouth of the Little St. Francis to the mouth of Marble Creek, not a very long stretch.  This gauge is used for the whitewater sections of the river at Silvermines and above, but those stretches are above the mouth of the Little St. Francis and it often adds considerable water to the river.  You can get a slightly better idea of the river through the whitewater sections by checking this gauge and the Little St. Francis gauge, and subtracting the flow of the Little St. Francis from this gauge, but there is no good gauge reading for the whitewater stretches.

*Near Saco--mouth of Marble Creek to Sam A. Baker State Park.

Near Patterson--Sam A. Baker Park to Wappapello Lake.

Understanding and Using River Gauges, Part 2

So let's say you are planning to paddle, and perhaps fish, a particular Ozark river, and you want to know what the water conditions are before you leave the house.  Back in the old days before the internet, your only real choice was to call one of the float outfitters or campgrounds along the river close to where you wanted to float and ask them.  But now we have this great resource in the U.S. Geological Survey's real time river gauges, where water flows and levels are updated (usually) every 15 minutes. 

However, it seems that most people don't really know how to use the gauges to their best advantage.  There is a wealth of information on them that is seldom utilized, even by avid anglers and paddlers.  Most people look at the level in feet, and then wonder what it really means, because unless you have floated the river many times under many different water conditions and made notes of what the level was after every time you floated it, you have no idea whether 2.2 feet on the gauge is high, low, or normal.  For the average person, that figure means nothing.  So let's run through how to really use the gauges.

I'm going to say for the purposes of this article that you are wanting to paddle or fish a river in the Ozarks of Missouri or Arkansas, but you can get to other states just by inserting the state abbreviation in the appropriate place on the web address.  For Missouri, the address is:  waterdata.usgs.gov/mo/nwis/rt.  For Arkansas: waterdata.usgs.gov/ar/nwis/rt.  This will bring you to the main state page:

From there, click on the Statewide Streamflow Table.  This will bring you to a listing of all real time gauges in the state, grouped by river system:

If you know which river system your river is in, you can scroll down until you reach that river system.  Most Missouri Ozark rivers are in the Osage (Niangua River being the most popular), Gasconade (Gasconade and Big Piney notably), Meramec (Meramec, Huzzah, Courtois, Big, Bourbeuse), St. Francis (St. Francis and Big Creek), Arkansas River (Elk and Big Sugar), and White (almost every other Ozark river, including Black, Current, and Eleven Point).

Then you find the gauge or gauges on the river you're interested in.  More about that in a bit...

I bookmark this page on my phone and computers, because in itself it gives you some of the most important information.  You'll note that on the right side of the table, there is a column for "Discharge, ft3/s" and one for "Long-term median flow (date)".  The discharge figure is what that stream is flowing right now, and the long-term median flow is a good approximation of what the normal flow for this time of year is for the stream at that gauge.  If those two figures aren't far apart, then the stream is near normal.  So if that's the case, this may be all the farther you need to delve into it; if the stream is near normal it's almost certainly good to paddle and fish.  But it is still best to click on the individual gauge, not only a gauge that appears to be close to the stretch you want to float, but also other gauges if there are any for that stream.  

So let's say you are interested in floating the Jacks Fork.  There are three Jacks Fork gauges, "near Mountain View", "at Alley Spring", and "at Eminence".  The order in which they appear on the table is the order of their locations on the stream, with the first one being the farthest upstream.  Here is what the top portion of the "near Mountain View" gauge page looks like:

Most people probably skip over this part of the page, but there are several useful features.  First of all, if you are not sure where the gauge is located--is it above where you plan to float, for instance--you click on the drop down menu at the top for "Available data for this site".  One option on the menu is "Location map", which will bring up a map showing the exact location of the gauge.  Second, under "Available Parameters" on the bottom left, you will see all the available pieces of information for this gauge.  Only the discharge and gage height are checked, which means only those two things are showing on down the page, but you can click on the other parameters, which show the water temperature and precipitation for this location.  Note that these two parameters are not available for all gauges, however.

The third feature that I often find useful is the begin date and end date on the lower right, which you can change to any dates.  By default it is set at the last seven days, but let's say you remember floating the river last year on a given weekend.  You remember what the river was like back then, and you want to know what the gauge height and flow was that weekend so you can compare those figures to the current ones.  You just change the dates in the boxes to a range that covers that weekend last year, click on GO, and it will change the graphs below to what the river was doing back then.

I was planning a float on the John Day River in Oregon earlier this summer.  The last time I'd floated the John Day was 20 years ago.  Back then I'd floated with my friend Corey Cottrell.  I couldn't remember exactly when we'd floated the river, but I thought he might so I called him.  Sure enough, he remembered because we'd been on the river over his birthday.  So I typed in a week back in 1999 that covered his birthdate, and found out that the river had been flowing around 400 cfs then, which turned out to be very useful in planning my recent trip.

If you scroll down the page, you will come to this portion--the meat of the gauge:

The graph for "Discharge, cubic feet per second" is what you want to be looking at first.  On it, the blue line shows what the stream has been doing the last seven days.  In this case you can see that it has been doing a lot of jumping up and down, because the area has gotten thunderstorms with a lot of rain almost every day.  Right now it's close to 200 cfs (the figures to the left edge of the graph).  You will also see small yellow triangles on the graph.  These triangles denote the median flow for each date.  Remember that the median is a good approximation of the normal flow.  So you can easily see that the stream is significantly higher than normal.  But how high is it, really?  Let's zoom in on the table below that graph:

You will see that the median--the normal flow for this date--is 35 cfs, and it's now flowing 194 cfs (Most Recent Instantaneous Value).  That's definitely a lot higher.  Now look at a couple other figures.  The 75th percentile figure is the flow which is higher than the stream has been 75% of the time on this date.  It's only 56 cfs.  So the stream is considerably higher than that.  Indeed, it's higher than the maximum for this date up to this point, which was set in 2011 at 69.5 cfs.  But you need to also look now at one other thing; "based on 17 water years of record".  That means that this gauge has only been recording data for 17 years.  The more years of record, the more accurate the figures on this table will be.  I like a gauge to have at least 10-15 years of record to have much confidence in the numbers like the 25th and 75th percentiles.  So I'm not sure whether a flow of 194 cfs is so high that it makes the stream too dangerous to float yet.  So I scroll down a little farther to see the gauge height, feet.  And I compare it to the Discharge graph:

Note that back on August 20 and 21, the Jacks Fork was pretty near normal, as you can see by how close it was to the little yellow triangles on those dates on the Discharge graph.  So look at the gauge height it was then--about 1.3 feet.  Now look at its present gage height--about 2.3 feet.  This tells me that it is a foot higher than normal.  A foot isn't really too significant a rise on an Ozark stream.

So to summarize, I now know the Jacks Fork near Mountain View, which is considered the upper Jacks Fork (the gauge is at the Highway 17 bridge and the Buck Hollow Access) is about a foot above normal, with normal for this time of year being about 35 cfs.

Understanding and Using River Gauges

I belong to a couple groups on Facebook that often have people posting asking about river conditions.  Invariably, someone will reply telling what the conditions were when they floated that stretch weeks earlier, which, of course, is not helpful at all, since river flows are constantly changing.  And finally somebody else will chime in with what the level in feet is right now.  "The upper Jacks Fork is showing 2.2 feet."  Which is also not very helpful unless they qualify it with something like "2.2 feet is marginally floatable".

To the average person, it seems to make sense to talk about the level in feet from the US Geological Survey's real time river gauges.  They can easily picture 2.2 feet.  But at the same time, unless you know what 2.2 feet really signifies on that particular gauge, the figure means nothing to you.  It does not mean the river is 2.2 feet deep.  It doesn't even mean that it's 2.2 feet deep in the riffles.  All it really means is that on the gauge's arbitrary "ruler", the water level at the gauge is standing at the 2.2 foot mark.  You have no idea what the normal level is.  You have no idea whether the river is high, low, or in between, unless you have been floating the river for years, and checking what the level was every time you floated it, so that you can correlate that level to your floating experience on that day.

But there is another measurement on nearly all the gauges; they will show you the flow in cubic feet per second.  It may be difficult for the average person to understand this measurement.  It simply means that, if the river is flowing 100 cubic feet per second at the gauge (abbreviated as "cfs"), then 100 cubic feet of water are going by the gauge each second.  But how is that measured?

First, most real time river gauges, to simplify, use a tube attached to a vertical or steep surface, like a bridge piling, with the bottom of the tube in the water below the lowest level the river is ever expected to reach.  That lowest level is 0.0 feet on the gauge.  There are openings below the waterline that allow water into the tube, and water pressure pushes the water within the tube up to the level of the stream's surface on the outside.  There is a float inside the tube that records the level, which goes to a remote recording station, usually updating every 15 minutes.  There are different mechanisms in some gauges, but they all give the level in feet at the gauge itself.

Measuring stream flow (volume) is more complicated, and thus impractical to measure constantly, so the measurement of level is assumed to signify a certain number of cfs.  In other words, on a given gauge, 2.2 feet is the same as 100 cfs.  Keep in mind, however, that each gauge is different, and 2.2 feet on a different gauge is likely to mean a totally different number of cfs.  So how do they KNOW that a given level in feet corresponds to a certain number of cubic feet per second?  Well, they periodically measure the flow, getting a wide range of gauge height to flow correlations.

The way flow is measured is by taking a line across the channel of the river, and at intervals along that line, measuring the depth and the current speed.  If the gauge is at a bridge, the bridge itself makes a convenient line along which to measure.  Usually, there are 25 to 30 measurements taken across the channel at equal intervals.  The depth at each interval is recorded, and the current speed is taken using a current meter.  At other times, a cable may be strung across the channel as the line by which the depth and current speed is measured.  And there are now more high tech methods of measuring depth and current speed, rather than the old current meters and a measuring stick.  But basically, those measurements are used to compute the number of cubic feet of water going by the measured line at a given time interval.

Usually these measurements are taken at certain time intervals, but they are also usually taken at the height or near the crest of floods, because hydrologists wish to know what the volume of water flowing in high water is.  So the measurements are taken, and correlated with the level in feet on the gauge at that time to give one an accurate measurement of how many cfs are flowing at that given gauge height.  Other than that, the shape of the river bed and valley in profile is measured, and other levels in feet are estimated.  The estimates are usually surprisingly accurate.  Here is a diagram of how it all works:
This diagram shows a bridge, with the river valley cross section delineated in brown.  The gauge is the solid red line on the left bridge piling.  The bottom of the gauge, level 0.0 feet, is about two feet below the river at normal flow (the lowest solid blue line).  The pink lines are the intervals at which the river's flow is measured at the downstream edge of the bridge.  At normal flow, the river's depth varies from 0 feet at the right bank to 6 feet at the deepest part of the channel, and the cross section of the river's flow at that point comes to 167 square feet.  If the average of the current speed in all the measured intervals is 1 foot per second, then the river is flowing 167 cubic feet per second at the normal flow, which is 2.0 feet on the gauge.  The middle blue line shows the river level at 10 feet on the gauge.  The profile of the river and its valley at that level shows 523 square feet.  Since the water probably flows faster in higher water, I've multiplied that by a current speed of 3 feet per second to come out with a flow of 1,569 cubic feet per second.  The highest blue line is the level at a major flood, 18.0 feet.  The cross section of the valley comes to 1,121 square feet, and at an average current speed of three feet per second, it would be flowing 3,363 cfs.

The results of these measurements show up on the web page for each gauge, and are recorded.  With enough years of record, very accurate figures of mean and median flow for that gauge can be computed, along with record highs and lows, and 25th and 75th percentile figures.  Mean flow is the average of all flows measured for that given date, one for each year of record.  Median is the middle value in that list of numbers; there are just as many years where the flow was greater than the median as years when it was less than the median.  The 25th percentile is the flow figure where the flow was only lower in 25% of the years of record, while the 75th percentile is the figure where flows were lower in 75% of the years.

Next: how to use the gauges and get the most information from them.