Hi. One way to answer this is to investigate spatial autocorrelation between the sites sampled for rainfall. ArcMap has a tool called Incremental Spatial Autocorrelation. Essentially it allows you to find distance bands between sampled points at which autocorrelation is highest. Keeping the spatial resolution of your interpolation below the distance at which spatial autocorrelation is most apparent will give you some confidence that the interpolated values are reasonable. Also if you are using Kriging, ArcMap allows you to create a variance raster that will allow you to see where and by what magnitude your interpolated values differ from sampled values (kriging doesn't preserve data points). Hope this helps.

MB

Hi Brady, thank you for your reply. I think this solution will be very helpful. Thank you so much. Best regards

Did you check the Correlation Decay Distance?, precipitation is highly space variable and I suspect that no universal decisión can be take, and distance at which some value of common varince remain.....could vay from región to región.

Hi José, thank you so much for your attention and answer. Best regards

Gustavo, what is the spatial densiy of rainfall stations you are dealing?, and how is the orography of área??

Hi José, attached the map with the spatial distribution of the rainfall stations and the orography of region. For the state area (27.768) and stations number (36), the density is one rainfall station for each 771 km².

Hi Gustavo,

The problem of the optimum size for interpolation depends on the characteristics of the data, this is not a trivial problem. A good discussion of the problem can be found in:

Hengl T. (2006). Finding the Right Pixel Size.Computers & Geosciences 32, 1283-1298.

Good luck with your problem.

Cheers,

Augusto

Hi Sanabria,

Thank you so much for your reply and sending the paper about the optimization of grid size.

Best regards

Gustavo

Hi Aurin,

Thank you so much for your answer and sending of the references.

Best regards

Hello Gustavo, 

During summer and winter the conditions might be completely different. There is a publication (Dzotsi, 2014 - http://onlinelibrary.wiley.com/doi/10.1002/joc.3904/abstract?userIsAuthenticated=false&deniedAccessCustomisedMessage=) showing how the rainfall varies for each season.

Best

Eduardo