To the list of courses || RASA2018 || To the theme || Estonian

**Data**: lumikatte_kestus.xlsx ### Instructions

The exercise is compiled by Taavi Virro and Kalle Remm.

## Exercise 4140. Points 3, theme: Interpolation |
Open exercise |

The worksheet *andmeid_pole* in the attached file contains names and coordinates of 13 Estonian weather stations with missing snow cover duration data. Predict snow cover duration for these stations using the observed data from other stations in the same worksheet using Inverse distance weighted interpolation.

- Present the root mean squared error (RMSE) of this prediction.
- Compare values and standard deviations (SD) of predicted and observed values on worksheet
*andmed_on*. - Explain why is variability in observed and interpolated values so different.

- Find Interpolation from the SDC.
- Select
*Calculate single points*and method*Inverse distance weighted*(figure). - Copy existing snow cover data as [X] [Y] [Value] from the worksheet
*andmeid_pole*in the attached file to*Given locations*. - Copy to
*Target locations*the stations with no snow cover data (as [Station name] [X] [Y]). - Press Calculate.
- Copy results to the sheet
*andmed_on*. - Open Difference tests and check
*RMSE*. - Copy the observed values to one cell and the interpolated values to other.
- Press
*Calculate*and find sample SD and RMSE values from the results. You can also use Excel function STDEVP() to find the sample SD.

The exercise is compiled by Taavi Virro and Kalle Remm.

Log in to send your results and to see the expected answer and responses from other students.