On my researchgate list in ~2005, we used the USGS/NRCS numbering system to number the hydrologic units in SC, USA from region, subregion, basin, subbasin, watershed, subwatershed with each category having two digits.  

Separately, a system to order channels and streams and number them by stream from headwaters to watershed was used at barious times.  I believe I uploaded the Chattooga River Water Quality report that included some of this description.  We called these Drainage Basin Response Units (DBRUs), and in an EIS I uploaded in 2013, it shows how I used this information to analyze a large project, and its estimated effects to sediment and water yield from timber management, temporary roads, prescribed burning, etc. The information was all spatially analyzed in GIS using RUSLE approach for estimating erosion.  If you cannot find, let me know and I will send from my files.

On my researchgate list in ~2005, we used the USGS/NRCS numbering system to number the hydrologic units in SC, USA from region, subregion, basin, subbasin, watershed, subwatershed with each category having two digits.  

Separately, a system to order channels and streams and number them by stream from headwaters to watershed was used at barious times.  I believe I uploaded the Chattooga River Water Quality report that included some of this description.  We called these Drainage Basin Response Units (DBRUs), and in an EIS I uploaded in 2013, it shows how I used this information to analyze a large project, and its estimated effects to sediment and water yield from timber management, temporary roads, prescribed burning, etc. The information was all spatially analyzed in GIS using RUSLE approach for estimating erosion.  If you cannot find, let me know and I will send from my files.

avid

its very nice interesting simple looking but indeed a very difficult question, because of the simple reason that 'computer system' or computer generated system can never be perfectly matched with. natural system or naturally created system i.e. catchments. 

Catchments are like the thumb of a person from one angle, as it is rare where 2 thumb impression matches or joined together entirely ....as they are always different ...similar can be said about natural catchments ..two catchments can never be same n similar...

now comes the question of joining ..yes it is not impossible but it is not simple to join them easily in the line of the computer system or computer generated system

It require a great deal of understanding knowledge tools interlinkages ..etc to join them, because they are 3 dimensional systems.....the 3rd direction leading to invisible underground domain is just a hit n trial ....so making a move to join catchments just in line of computer-based joining philosophy is a great risk..what I think..

sorry I may be wrong and I never contradicting your good views....it's just a spark to ignite this important question, based upon mine 32 years exposures to that so-called natural catchments ..I dealt more than 25 such catchments in the different part of the India.

and came to this expression at last.

Hello David, perhaps the Pfafstetter method by Verdin and Verdin could come in handy

Here's a link but I'm sure you'll find lots of literature about it so you can evaluate if it fits your needs.  

http://ponce.sdsu.edu/pfafstetter_system.html

Regards, 

K.

The USGS has a variety of existing tools for this purpose: see https://wwwbrr.cr.usgs.gov/projects/SW_MoWS/GeospatialFabric.html and https://datagateway.nrcs.usda.gov

If you want to get deeply into this, take a look at:

the Australian Hydrological Geospatial Fabric (Geofabric): www.bom.gov.au/water/geofabric/

Water ML 2.0 a WMO standard produced through a collaboration between WMO and the Open Geospatial Consortium: www.waterml2.org

Maybe the Pfafstetter method is a good way as Karin Musalen first said. If you think it is helpful we may share with you a database classified following it. 

The three questions you have proposed are answered by the Pfafstetter method. All the databases that represent hydrographic networks in our Agency are coded with this method, both catchments and related reaches. I've written an article about it on Wikipedia: Codificação de Otto Pfafstetter (it's a pitty it's only in portuguese, I hope the translators can help). I've added to this article many references of works all around the world. You can download many examples of these databases from the ANA's Metadata Portal.

http://metadados.ana.gov.br/geonetwork/srv/pt/main.home

https://pt.wikipedia.org/wiki/Codificação_de_Bacias_Hidrográficas_de_Otto_Pfafstetter

Huge thanks for your answers. Most informative. Amazingly, I'd not come across the Pfafstetter method before. I can see how the beauty of the method is that the name of the catchment itself becomes the hash list containing all the information needed to identify contained/containing catchments. I was thinking of a separate information table containing this linkage information where the catchment name is simply a key into this information. The Pfafstetter does away with the need for an additional table. I knew someone would have thought deeply about this!

Again many thanks. More thoughts and comments are always welcome. 

Hi Mr David Leedal, I am working in the area of remote sensing and GIS applications in water resources management. Therefore I think I am eligible to answer your question. You mentioned that you have different many chachments and you want to store those features in a particular format so that you want to do data quiry. I think the best way is handling this data with GIS (QGIS-open source software). Here we can store the data in features as well as attributes format. My suggestion is storing the all river details in chachments, sub-catchment, watershed, sub-watershed, etc by assigning a separate numbering system to each features makes easy while doing queries among them.

Hope answer is useful. Thank you.