Hi there from Australia

Yes, as suggested the first sensible and easy step is to test the
correlation between the complete station and each of the other two
incomplete stations. This can only be done for the respective periods of
common data record of course. If the correlation is reasonable, you can be
moderately happy that the rainfall occurring between the complete station
and the other(s)is caused by the same climatological events. You can then
either do a simple regression as has already been suggested, or use the
normal ratio method. The latter is where you assume that the ratio between
any given reading for the complete station and the "normal" (i.e. the
mean) for that station is the same as for the incomplete station (e.g. the
ratio of the corresponding unknown reading to the mean for the incomplete
station. Of course, the mean for the complete station would again only be
the mean for the period when the two stations both have readings.

so Rc/Nc = Ri/Ni and Ri is the unknown reading.

Hence the record may be completed, and this is a simple process and
probably sufficiently accurate given the nature of the data. I still
prefer to use a slide rule rather than a computer where feasible!

I hope this helps.

Dr Richard Faulkner
Associate Professor of Environmental Engineering
University of New England
Armidale, NSW Australia


> Respected sir,
>
> i do not think if the problem is very complicated.
>
> i think you can as well develop a correlation between the stations and
> extend the data. i feel there are good no. of methods to estimate the
> data. the catchment is hilly and the precipitation is mainly orographic.
> as i see, the precipitation is higher at xanthi, medium at gerakas and
> again slightly higher at oraiou. its better if you can have the
> toposheets.
>
> if you think i can help you in any way, please send me the data with all
> the information. i shall provide you with the final data.
> i am currently working with hydro power projects.
>
> if your accuracy limit is not very high, i can solve your problem.
>
> Thanking you sir,
> Ranjit,
> Hydrologist.
>
>
> geopapaioa <geopapaioa@...> wrote: The
> first rainfall station (xanthi station)is at 100 m altitude and i
> have all the rainfall data from 1980-2000. The second rainfall station
> (gerakas) is at 340 m altitude and i have the data from 1980-1998 but
> there are missing many years and a lot of months from this data(only 6
> years are full). The third station (oraiou) is at 640 m altitude and i
> have the data from 1980-2007 but there are missing many years and a
> lot of months from this data. The watershed area is about 260 km2 .
> The highest level of the watershed is 1300 m and the lower 100 m. I
> have no other stations in the research area. What i want to do is to
> complete the rainfall data of station gerakas and oraiou from the data
> of xanthi station(for their same years 1980-1998). The fact is that
> there are no other stations in the area working and the most near
> station is about 50 km in other watershed with data off enough
> difference(so i can not use the data, non homogeneous).I would like to
> thanks all off you that have reply on my message. I am waiting for new
> ideas and answers to my problem. Thank you very much dear Dr S.H.R.
> Sadeghi for the tips