The Argo dataset contains spatial data in three dimensions (3D). In this tutorial, you will learn how to model a spatial data in 3D using spatial statistics using the R software program. The following data analysis and modeling was conducted on a 2020 MacBook Pro with Apple M1 chip with 8 cores, 8GB in memory, and aarch64-apple-darwin20 (64-bit) platform running under macOS Sonoma 14.0. The source code can be found on GitHub at
3D_Spatial_Modeling_Argo_Dataset_tutorial_codes.R
Prepare the Argo dataset
Download the 3D spatial Argo residuals dataset produced by Yarger et al. (2022)1 from this link: 3D spatial Argo residuals. This will download a file named jan_march_residuals.RData which contains the 3D spatial Argo residuals saved in an RData format.
Open R or RStudio. A typical RStudio session looks like this:
To keep track of R codes and also to prepare them for execution later, open the R Script window. This can be done by clicking on the 'New File' icon located at the topleft most portion of the RStudio window. Click on the 'R Script' option.
The R Script window will appear on top of the R Console window.
Locate where the .RData file was saved.
On a MacBook, this can be done by right-clicking on the file and choosing the 'Get Info' option.
Another window will appear containing several information about the file. The location and nested folders that contain the file is displayed in 'General -> Where:'.
Highlight the location information and right-click on it. A dialog box will appear which you can click to copy the location information and save it as a Pathname (the exact address of the file on the computer). On my computer, the Pathname is /Users/laisalvana/Downloads
Store the Pathname as a variable in R. For example, define the variable filepath and paste the Pathname. This can be done by running the following codes on the R console:
filepath <- '/Users/laisalvana/Downloads'
1. Yarger, Drew, Stilian Stoev, and Tailen Hsing. "A functional-data approach to the Argo data." The Annals of Applied Statistics 16.1 (2022): 216-246.
https://doi.org/10.1214/21-AOAS1477.
↩
