Dear Meshesha,

I think these publications can be fruitful and might support Question !!!

1. http://www.sciencedirect.com/science/article/pii/S0169809517300029

2. https://link.springer.com/article/10.1007/s00704-016-1960-1

Thanks

Dear Tesfa,

The answer to your question may depend strongly on the size of your watershed of interest, and on the spatial resolution you expect, or need, for your application. It is also critical to know what is the time period that is relevant for your problem, and what temporal frequency you require for these measurements or synthetic simulations. Lastly, what level of accuracy do you need?

A simple initial approach would be to consider the regional climatology of the area of interest and to interpolate data from a broader area to generate credible data for your watershed.

A more specific idea would be to exploit the data sets generated by re-analyses and typically available in your local meteorological service: these would provide coherent descriptions of the weather and climate for large watersheds (say tens of thousands of square km).

For smaller areas, or for more generic applications, it may be more appropriate to use weather generators, which are software programmes that yield typical weather sequences, subject to some external constraints.

Cheers, Michel.

Dear Tesfa

We have done a lot of careful work with global data sets as we attempt to fill in gaps around the world that suffer from patchy or non-existent records. In so doing we have worked with the following data sets and spent many weeks cleaning and validating them. Be warned they are not perfect! Some data sets are better suited to specific applications. For example, extreme event analysis is problematic with interpolated data sets as through that process the tops/extremes are often clipped and therefore can lead to the systematic under reporting of extremes. 

Several datasets have recently been released or upgraded, and provide subdaily or higher (than previously mentioned) resolution daily data, including:

·       Multi-Source Weighted-Ensemble Precipitation (MSWEP), 3 hourly, 0.25 degree, since 1979 (http://www.gloh2o.org/)

·       MSMAP, one hourly, 0.1 degree, since 2000 (http://sharaku.eorc.jaxa.jp/GSMaP/)

·       CHIRPS, daily 0.05 degree, since 1981 (http://chg.geog.ucsb.edu/data/chirps/)

The newest datasets like the previously listed do need to be consolidated and verified, and cross validated for project areas with observation data where it is available but that is the case no matter which data set is applied (Beck et al., 2016).

Be very careful when directly applying any of this data. Check against some regional empirical data at the very least and be prepared to state the limitations of the data wherever it is applied. 

Beck, H. E., van Dijk, A. I. J. M., Levizzani, V., Schellekens, J., Miralles, D. G., Martens, B., & de Roo, A. (2016). MSWEP: 3-hourly 0.25 global gridded precipitation (1979–2015) by merging gauge, satellite, and reanalysis data. Hydrology and Earth System Sciences Discussions, 1-38.

All the best. Peter

PS. We have also worked with various reanalysis databases but that is a subject for another post. Email me if you wish for more information [email protected]

Hi Tesfa

I recommend you read some works done during the IAHS decade on "ungauged catchments hydrology"

https://iahs.info/Commissions--W-Groups/Working-Groups/Prediction-in-Ungauged-Basins.do