Rainfall data preparation

[Sokly Siev]

Dear Dr. Tom, 

Greeting from the Philippines, 

I am new to the CAESAR-lisflood. I would like to model the hydrological and sediment transport in Tonle Sap River in Cambodia using CAESAR-Lisflood. I have time series daily rainfall data of all rainfall stations in the Tonle Sap River basin but there are some missing rainfall data in some years. I have the following questions; 

1. Regarding missing rainfall data, how can I fill this missing rainfall data? 

2. After filling the missing rainfall data, Do I have to get the average distribution rainfall data of the whole basin for the data input in CAESAR-Lisflood? Since I have read some questions and answers regarding to the rainfall input format which is single column of rainfall data in the text file, Am I right the rainfall data here is the average distribution rain fall over the whole basin ? 

Thank you for your time to answer my questions

Best regards, 

Sokly

[Tom Coulthard]

Hi Sokly,

OK, CAESAR-Lisflood (CL) will only use one rainfall input – so it will assume the rainfall is the same over the whole basin. This may be an issue on larger basins – but for smaller its usually not a problem. If you are trying to make an average – a spatial average – I would recommend using a weighted interpolation between the rainfall gauges – with the weighting dependant upon the area of the catchment covered by the rain gauge.

 

Hope that helps – how large is your catchment?

 

Tom

--
You received this message because you are subscribed to the Google Groups "caesar-lisflood" group.
To unsubscribe from this group and stop receiving emails from it, send an email to caesar-lisflo...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Sokly Siev]

Dear Dr. Tom, 

Thank you for your prompt reply, Yes it helps a lot. In my case, my study area is quite large, the length of the Tonle Sap River is 126km and its catchment is 15380km2. What do you think CL can handle with this large basin? I am thinking to run CL separately for subcathment. Any advice?

One interesting thing of Tonle Sap River is that it has a reversal flow. The water coming from Mekong River will flow into Tonle Sap Lake during rainy season and flow out from the Tonle Sap Lake to Mekong River during dry season via Tonle Sap River. Based on your experiences what would you recommend? or I have run CL seasonally?  

Anyways, How to deal with missing rainfall data? because there are some missing data in some months in some years of rainfall data. 

Best regards, 

Sokly

[Tom Coulthard]

Hi Sokly, OK thats quite a large basin! It can certainly handle large areas but the more grid cells – the slower it will run. So you’ll have to consider running it at a coarse grid scale.. Often the best way is to start running the model on a smaller subcatchment (as you suggested) to get a feel for how well the model works and to get experience of using it etc..

 

Reversing flow. Yes – it should handle that fine – the flow model works fine in tidal environments where you have twice daily reversals so I imagine it would work OK.

All the best,

[Sokly Siev]

Dear Dr. Tom, 

Thank you so much for your reply. I think I am going to use bigger grid size (90m). I will be back when I encountered the problem or have questions. 

Best wishes, 

Sokly 

[Sokly Siev]

Dear Dr. Tom, 

I just have another question regarding to the rainfall data. I am wondering if CL can compute for the average distributed rainfall the same as WMS: GSSHA model does by inputting the dates, coordinates of rain stations, and rainfall values? Since I have to compute the average distributed rainfall for a time step (day) one by one and I have to do so for 10 years of time steps. It will take me much time to compute. Any advice and recommendations ?

Thank you very much for answering my question. 

Sincerely Yours, 

Sokly 

[Tom Coulthard (gmail)]

Hi Sokly - 

No, afraid CL will not compute the rainfall in this way - it needs to have just one rainfall input! What you could do is work out he weighting for each of your rainfall gauges (usually based on the percentage/proportion area of the catchment they occupy and then just use that to calculate a total rainfall from all of the guauges..

Tom

[Sokly Siev]

Dear Dr. Tom, 

Is it how to calculate rainfall for one time step? So I have to calculate a total rainfall from all the of gauges for every time step? if so it's gonna be too much time in calculating the rainfall input. Is there any faster methods in calculating this? 

Sincerely yours, 

Sokly