Selection of Roughness Coefficients

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This video describes the field selection of roughness coefficients, or Manning's "n" values. Both the comparative (photos) and analytical (Cowan's) methods are discussed.


Date Taken:

Length: 00:08:34

Location Taken: Las Vegas, NV, US

Video Credits

Carly Venghaus, Hartley Delvalle, Todd Geiger, Mark Smith, Office of Employee Development


Hi, this is Megan Poff and I’m the Field Office Chief at the USGS in Las Vegas, Nevada.  During some of the other videos in this series, we mentioned the selection of roughness coefficients, or Manning’s n-values.  This video will describe some of the primary techniques for field estimation of roughness coefficients.  I’m going to cover both the comparative and analytical methods.  There is a plethora of information available on roughness value estimation, such as Water-Supply Paper 2339, Water-Resources Investigations Report 85-4004, Professional Paper 1584, and Water-Supply Paper 1849.  Keep in mind that roughness coefficient selection is subjective and the only way you’ll get better at it is through experience, but everyone starts somewhere.   

I’m going to cover the comparative method of roughness coefficient selection first.  This involves looking at pictures of sites with known roughness coefficients (that is, determined through roughness verification studies), and then assigning your site a roughness coefficient based on the pictures.  For this method, I’m going to use Professional Paper 1584 – Verification of Roughness Coefficients for Selected Natural and Constructed Stream Channels in Arizona.  Southern Nevada has similar channels to those shown in Professional Paper 1584, so we have found that to be a good reference for us.  These types of studies have been done all over the country though, so if you happen to live in a place with things like trees, I can guarantee that there will be a good reference available to you. 

Let’s take a look at our field site.  Looks like it’s a pretty uniform channel with sand and small gravel and some desert brush on the edges.  I’m going to scan through Professional Paper 1584 here and find a channel that looks similar.  By similar, I mean not only similar in terms of substrate size and type and vegetation, but I’m also talking similar in terms of the depth and width of the flow.  This is important, because in many cases, deeper depths translate to smaller roughness coefficients.  Okay – Skunk Creek above Interstate 17 looks similar to this channel, although the substrate is definitely larger than what I’m seeing at this site.  I’m going to keep looking.  Oh, here we go – Hassayampa River near Morristown looks very promising!  Let’s check the depths of the verified coefficients before we get too excited.  They calculated four roughness coefficients for this site for flow depths ranging from about three to six feet.  My depth of flow for this channel was about three feet, so it appears we have something comparable.  Let’s look at the widths.  The comparison site is nearly 200 ft wide, and my site is not nearly as wide as that.  Therefore, my site should have a slightly higher roughness than the comparison site.  The verified roughness coefficient for Hassayampa River near Morristown was 0.018.  By looking at a similar site, I make a judgement that my site probably has a roughness coefficient of around 0.022, which I increased based on the fact that my site is not nearly as wide as the Hassayampa River site.  I’m going to write that in my field notes.  Repeat this exercise for every cross section in your indirect measurement.  It’s completely okay to have different roughness coefficients estimated for each cross section – that’s why we’re going to do this for all the cross sections.  If you haven’t already done so, take photographs of each cross section – looking across the section from one bank and looking downstream at the section; document each view and perspective. Remember that the roughness value you assign to each cross section should reflect channel conditions upstream and downstream for a distance about halfway to adjacent cross sections.

I’m going to cover the analytical method of roughness coefficient selection now.  The analytical method is the method that is often emphasized in indirects classes, because it tends to result in more repeatable results among different analysts.  One of the most well-known of the analytical methods is known as Cowan’s method, and it is detailed in Cowan, W.L., 1956, Estimating hydraulic roughness coefficients: Agricultural Engineering, v. 37, no. 7, p. 473-475.  For this method, you will use a simple equation: n = m (n0 + n1 + n2 + n3 + n4).  Let’s start with the n-variables in our equation – as we did for the comparative method, consider channel conditions upstream and downstream for a distance about halfway to adjacent cross sections.  The first variable, n0, will be the base roughness.  You will determine the base roughness by looking at a table and choosing a value based on your predominant bed material.  Looking at our channel here, this is definitely a sand channel.  I don’t know the grain size, but it looks like a fine-to-medium sand to me.  I’m going to use 0.020 for the n0. 

Next up is an adjustment for channel irregularities, or n1.  As you can see in the table, we can have adjustment values from 0 to 0.020.  This channel is in excellent condition, so I will say that it has a smooth degree of irregularity and assign it a value of 0 for n1.

Next, we’ll adjust for cross-sectional variations, or n2.  This adjustment value ranges from 0 to 0.015 based on how our channel changes as we move downstream.  It looks like we have very gradual changes in this channel if any, so I’m going to assign a value of 0 for n2.

Our next variable, n3, is an adjustment for obstructions.  I typically think of obstructions as non-living things like debris deposits, piers, or boulders.  The values range from 0 to 0.050 for obstructions.  This channel has no obstructions, so I will assign a value of 0 for n3.

n4 is the adjustment for vegetation.  This is different than the obstruction value, which was n3.  Remember, obstructions are not living things.  Vegetation is a living thing, so it gets its own variable.  Our possible values range from 0.002 for small vegetation like grass or bushes where the depth of flow is three times the height of the vegetation, all the way up to 0.100 for things like dense cattails growing in the channel bottom.  This channel has pretty minimal vegetation, so I will assign a value of 0.002 for n4.

Our last variable is m, and that is the adjustment for channel meandering.  This is a multiplier, and its values range from 1 to 1.3.  The channel we’re looking at here is perfectly straight, so there are no meanders.  I will assign this variable a value of 1. 

Plug in all of your assigned variables into the equation: 1 (0.020 + 0 + 0 + 0 + 0.002) = 0.022.  Look at that, it’s the same result that I obtained using the comparative method.  Obviously, there are many channels that are not nearly as simple as this one.  The main thing to remember with the analytical methods is not to “double count” certain aspects, because you will bias your roughness value high.  For example, remember when I mentioned that obstructions are non-living things and vegetation is a living thing?  That distinction helps me not double-count the same thing for two separate variables.  If you have a log jam in your cross section, only count the log jam as an obstruction (the logs are dead) rather than both an obstruction and vegetation. 

Whatever method you choose, make sure you repeat that method for each cross section.  If you choose an analytical method like Cowan’s method, it’s helpful to have a form with you so you can remember all of the variables and document your choices for each one. Template forms are available to make the process easier and ensure you’ve documented everything you need for a complete roughness determination.

I also recommend that you have relevant roughness coefficient references with you in the field to make your job easier.  You can save PDF copies of the published references on your computer, or you can stow some printed copies in your field vehicle.  Always do your estimates in the field instead of waiting until you get to the office.  The pictures you’ll take on-site will help with evaluation in the office, and you might decide to modify the roughness coefficients during analysis – but nothing beats your first-person perspective while you’re at the site.  Plus, the only way to get better at roughness estimation is to do it over and over and get feedback from more experienced people. 

One last comment about the scope of this video. I’ve explained determination of roughness coefficients for a site with a simple trapezoidal channel. For sites with more complicated geometry, where the flood overtops the main channel and extends onto the flood plain, we need to subdivide our roughness determinations based on channel shape.  If you have a situation like that, make sure you consult with experienced personnel.  A good rule to follow in the field is to document cross-channel roughness changes and identify where those changes occur in your field sketch – final n-values for measurement computation can be determined in the office if you’ve made detailed observations in the field.

If you need help in the field, call your supervisor, surface-water specialist, or indirect measurement specialist.