I am working to develop the water poverty in a river basin in Nepal, and it need river discharge data but the problem is we do not have enough coverage of the stations. Do you have any suggestion how to estimate the discharge for ungauged basin?
There are different ways of estimating runoff from ungauged catchments:
1. Area proportion method: if you have a gauged catchment with similar watershed characteristic then you can simply predict the flow from the ungauged catchment by area proportion. This is very simple but might be helpful.
2. Using hydrological modelling: If you have few gauged river basins and if meteorological data is available, you can calibrate those gauged catchments. After calibration, you can develop a regional model which relates model parameters with watershed characteristics. Then knowing the watershed characteristics of the ungauged catchment, you can predict the plausible model parameter value of the ungauged catchment. Then once, you determined the model parameters, you can simulate flow of the ungauged basin by forcing it a meteorological data.
You can see this research that we did Wale et al 2009
There is also a discussion on RG, follow this link
Perhaps it does not fit your circumstances but is it possible for you to apply a constant flow of a conservative tracer and measure the dilution downstream?
A long time ago I managed to estimate a stream discharge using the difference in temperature between the river upstream and downstream of a constantly pumped mine water discharge which was significantly warmer than the river.
There are different ways of estimating runoff from ungauged catchments:
1. Area proportion method: if you have a gauged catchment with similar watershed characteristic then you can simply predict the flow from the ungauged catchment by area proportion. This is very simple but might be helpful.
2. Using hydrological modelling: If you have few gauged river basins and if meteorological data is available, you can calibrate those gauged catchments. After calibration, you can develop a regional model which relates model parameters with watershed characteristics. Then knowing the watershed characteristics of the ungauged catchment, you can predict the plausible model parameter value of the ungauged catchment. Then once, you determined the model parameters, you can simulate flow of the ungauged basin by forcing it a meteorological data.
You can see this research that we did Wale et al 2009
There is also a discussion on RG, follow this link
an easy method to estimate the discharge is the salt dilution method. But it works only in smaller rivers and brooks because other wise you have to use tens of kg of NaCl.
Steffen Bender: I think the method you are suggesting is for one time data collection. I need at least 10 years of flow data, anyway thanks for the suggestion.
Try Snyder approach(1938) in which you will be directed towards the so called "Hydromorphometrical approach".but prior to the application of Snyder equation you must locate the center of gravity of the basin,length of main stream from center of gravity to outlet.....you will find more than one equation to use in the handbook of "Applied HYDROLOGY "BY Chow.
We too have faced similar kind of problem when we are working for Environmental Flows Computation of Tawang Basin rivers in Arunachal Pradesh. If you are having the discharge data of similar kind of catchment then Area- Proportionate method gives fairly good estimates.
Distributed hydrological modelling based on satellite observation is one way to go, however you would need to quantify the uncertainties coming from the model and the obs. You may also use NWP output for the predictive mode, EFAS is nice success example (https://www.efas.eu/about-efas.html ). There is also a dedicated page under the IAHS site with few interesting studies that can be stimulating for your application: http://iahs.info/pub/pubs.php
Yea there is one method in Nepal, which is being used by some scholars. As Dhiraj Sir said, it is developed by the WECS and DHM and later on modified somehow. It is called hydest method. It just depend upon the area of the basin and monsoon wetness index. Once you estimate the discharge you can check with the gauged basin. I have tried once for the Imja catchment, it works somehow good. But there are many uncertainties still, which is still better than non.
There are many research findings that have been published see e.g the summary by
1. Hrachowitz, et al., 2013 A decade of Predictions in Ungauged Basins (PUB)—a review. Hydrological Sciences Journal, 58(6): 1–58.
Further more see also
2. Montanari et al., 2013. The continued initiate to the PUB, the Panta Rhei—Everything Flows”: Change in hydrology and society—The IAHS Scientific Decade 2013–2022.
3.Book:
Runoff Prediction in Ungauged Basins (Synthesis across Processes, Places and Scales)
4.Depends on your purpose you can estimate the discharge in the ungauged part using SWAT model
5. I think , ensemble methods are practiced now a days, so you can also excess information related these articles
In my opinion you just need to use one of the synthetic methods available in the literature. However, you need to try to get reasonable values of its parameters.
If you need some excel sheet for some calculations you may go to these links: