How to replace the missing temporal data of the ocean tidal levels for hydrodynamic modeling of an estuary?

Question

I am modeling the hydrodynamics of an estuary with a 3D model, to use it for sediment transport and quality purposes. The model extends from the inland river till the ocean, and the total length of the model is approximately 500 km. I have to run a simulation for 1 year period but one of the boundary conditions of levels in the ocean is missing 15 days of data. There is another station with values in those 15 days at an approximate distance of 100 km (inside the model), and there is overlap of information outside of that time period in order to study them together or correlate them.

Would it be possible to estimate the 15 days missing values with the tidal values of the other station? Both are oceanic stations. Could you please, suggest me which would be the best procedure to follow?

Andres Alcolea Rodriguez · Answer

Dear Santiago, We did apply a decomposition methodology, summarized in the attached paper.
 
Good luck,
 
Andres

Santiago Symonds · Answer

Thanks for your comments Thinakaran and Stefanie. I have compared the values in both stations with a scatter graph and obtained a linear correlation between the values of R^2=0.78. When I apply this correlation to fill the missing values I obtain a good series but I cannot represent the offsets or lags between the waves. Would it be necessary to decompose the tidal information into astronomic and meteorological and compare the stations in order to obtain a better approximation? In that case please indicate best procedure to follow please?

Stefanie Helmrich · Answer

Yes, you can estimate the data of the missing days using the other stations. I would suggest to compare the data of both stations with the data from the station at the boundary. Depending on which station shows better correlation with the station at the boundary, you decide which station you use to extend the data. You can use vertical offset to extend the data.

Thinakaran Elayappan · Answer

Thinakaran
You can use the values, which are inside the model. I mean that another station in 100 km distance.

Taxonomy