Float method is unprecized and yiels unreliable results as it measure the velocity of the surface top water layer: which has the maximum values. Instead it is beter to use the current meter to measure the mean velocities among the stream profiles.

Well,

Mainly, controlled by density, not by weight, volume or material type.

means the density of floated material should be near or equal to the density of water.

Regards

I agree with Dr. El-Rayes. However, if you prefer to use this method anyway, I would suggest using perhaps 25 oranges, thrown in at various locations across the channel, and measure speed of each over a suitable distance of 2-3 meander bends, or perhaps 10 or more times channel width. Count and record times for all oranges visually. Some will find the fastest water, others will be delayed if there are pools or bends. Add up all the seen orange times to float the section, and divide by that number to get an average time. If water is muddy, probably will not work well. The oranges tend to float just at or below surface, so the velocity may still be higher than if measured with current meter. I believe there could be a reduction coefficient, but I did not use this method. With current meter, fairly shallow waters are measured at 0.6 of depth, and deeper waters average velocity at 0.2 and 0.8 depth, at numerous location across the cross section. Other methods include Doppler meter that is floated across, and constant injection of known dye concentration and measuring dilution downstream. Flumes or weirs installed with calibrated dimension can be helpful.

If you have engineering specs for channel with dimension and gradient, you might be able to check estimate with HECRAS or other program. For research grade work, I would suggest above all, employ standard methods, and training in method or hire an experienced professional with proper equipment.

Surface float method is an approximate method used for flow measurements in the absence of precise instruments. If you are having the dimensions of canals you may even use a hydraulic method (say Mannings Formula) for estimating flow rate.

However, if you still wants to use surface float method, I recommend you to refer 'Stream discharge using the float-area method' which can be downloaded from the internet.

Greetings!

While not as precise as using a current meter, you can still get decent results using the float method. I offer a few suggestions below:

Float Choice. I agree with previous comments: Oranges are an excellent float because you want a float that is just below the water surface. If you use a stick or an empty plastic bottle, for example, any wind you have will likely influence your results. Oranges are relatively cheap (at least in the US) and they are biodegradable if you lose one in the stream. No matter what you use, it needs to be just dense enough that it sits just below the water surface, but doesn't sink or hit the bottom. Oranges fit this requirement. A half-filled plastic bottle might work, too, in an emergency.

Measurement Reach. This is not going to be an issue in a concrete trapezoid, but find a straight reach with minimum turbulence. There shouldn't be any constrictions, bends, or other other obstructions within about 5 channel widths up or downstream of your selected measurement reach. The more uniform the flow, the better the results. If present, you can remove wood, rocks, debris, etc... before you start the measurement. Just make sure you wait a period of time to allow the flow to become uniform again--15 minutes is a good rule of thumb, but it depends upon the size of the obstruction versus the size of the channel. You'll need to read the reach to decide when it's appropriate.

The section should be long enough such that you get at least 20 seconds of travel time for the floats. If you are doing storm event measurements, at the start of the measurement, you need to note the location of the water surface elevation as you'll likely have to come back later and measure the cross sectional area when it's safe to do so. In baseflow conditions, you should be able to measure the cross sectional area at the time you do the work.

If your intention is to go to the same site or sites in fair weather and foul for multiple measurements, I suggest that you monument or mark the measurement section. That way, you don't need to string a tape every time you go to a site. It's a time saver when you're working in the wet, in particular.

Finally, realize this--it is typically very difficult to find the perfect measurement location for a discharge measurement, no matter how you do it. Find the best site you can and do the best you can! Take lots of pictures of your site, take good notes during your work, and caveat your results as necessary in your write up.

Taking the Measurement. You need to sample the entire section, slow parts and fast. Don't just throw five floats in the middle, for example. In your mind or your field book, break the reach into two or three sections (example: Right Side third, Middle, Left Side Third). How many sections you need is best professional judgement.

In each section, I would recommend at least five measurements (so you'd have 15 data points in the example above). Also, make sure you introduce the float upstream far enough so that it has time to accelerate to the stream velocity before it crosses into the measurement reach. Use a stopwatch to measure the time needed for the float to travel to the downstream point of the measurement reach. I'm sure you realize this, but this type of flow measurement is usually a two person job.

Data Processing. Nothing to add here, really. You simply divide your measurement reach distance by the individual float travel times to get the individual surface velocities for each measurement. Then, I typically average all the measurements to compute an average velocity for the entire section. Once you have your average velocity, one important thing that was touched on before is you need to correct surface velocity measurements to average velocity measurements as surface values are always faster. The standard formula is:

Vavg = kVsur

where k is a correction coefficient that ranges from 0.80 in natural streams to 0.90 in artificial channels. Commonly, 0.85 is used--I might go to 0.90 in your case, based on your site description.

Final Thoughts. I don't know what your reason is for measuring discharge here, but if you are working long-term and you want to understand water flow at this location, you might want to install a staff plate and develop a rating curve for this site. Eventually, that would allow you to go to the site, take a reading off the staff plate, and get an instantaneous discharge value without having to measure it every time. That process, however, is too involved to describe here. But there are all kinds of references on-line for developing rating curves.

Good luck!

If you have some parameters such as channel area, spread, height and volume, and empirical graph that is widely used in the literature.

If you have all the dimensions of the canals I would suggest that you have a look into hydraulic methods as in Mannings equation, for tidal canals piezometric difference could also provide a fairly better estimate of the tidal velocity. If you are using float method make sure wind is not affecting the float as it may create unwanted errors in your readings.

I recommend the use of the application of velocimetry of particle images by large-scale (LSPIV). With a minimum HD resolution camera and measurements of six distances on the water surface, the flow rate can be determined.

https://www.youtube.com/watch?v=169-1d_TUkA

https://www.youtube.com/watch?v=k67P78npxlI

http://riverdischarge.blogspot.com/