Could gauge station data be interpolated over the watershed?

[D_D_]

I don`t know if I`m making the right decision trying to interpolate this type of data? Thiessen polygons shown for better description (picture in the attachment).
My main watershed was divided into gauge contributing areas. The data describes flood end date on tributaries and also on a main river (that flows from North). I have also flood start and flood duration data.

I haven`t googled any helpful information about interpolation techniques for gauge data over a waterheds, especially when there is an objective lack of data.

Thank`s in advice.
[ATTACH=CONFIG]27737[/ATTACH]

[EricKrause]

It looks like you only have 8 points.  I would not try to interpolate with such a small dataset.  If Thiessen Polygons are meeting your needs, you should probably stick with that.

[D_D_]

Eric6346, it`s objective fact, we have a sparse gauging network on tgis river. So I just can`t get more archieve data.

Is there a techics to interpolate surfaces when there is a sparse observation network?

Might the data from small and large watersheds be campared and then interpolated, or I just need to somehow distribute this interpolation just along the river sub-watershed boundaries, so they doesn`t influence the neighborhood areas? This is flood date.

Don`t you know any literature about hedrological data interpolation (you might know more about this), I`ve tried to google purposefully, but didn`t find anything helpful?
...