Select Page

eRAMS Spatial Interpolation Tool

The purpose of this geostatistical analysis toolbox is to automate and speed spatial interpolation and associated analysis of geospatial data.

The geostatistical analysis on eRAMS performs automated spatial interpolation of a point shapefile

Citation:

Dozier, A., Traff, K., and Arabi, M. (2013). “eRAMS Spatial Interpolation Tool”, eRAMS Resource Center, Colorado State University. Available at http://www.erams.com.

API Notes

Spatial interpolation can be performed through the eRAMS website interface or using a REST protocol (via a URL). The following shows how to build a URL that will return results
as well as download locations of spatially interpolated output rasters:

Input Parameters

Parameter Type Description
cellsize Float The cellsize of the output raster. The cellsize is equivalent to both the width and height of each cell.
column String The attribute within the point shapefile with which to perform the spatial interpolation.
f json object Defines the output file format and location. More Information
geometry json object Defines the input file format and location. More Information
method String Defines the technique used to perform the spatial interpolation. Currently, the only implemented method is “kriging”, which performs universal Kriging.

Output Parameters

Parameter Type Description
urls json array An array of output (interpolated) raster file descriptions using the name of the raster file and a URL at which to download the raster file.
method String The technique that was used to perform the spatial interpolation. Currently, the only implemented method is “kriging”, which performs universal Kriging.
model String The model that was selected as part of the automated kriging routine. (Kriging only)
sill Float The limit of the variogram tending to infinite lag distances. (Kriging only)
nugget Float The height of the jump of the semivariogram at the origin. (Kriging only)
range Float The distance at which the difference between the sill and the variogram becomes negligible. (Kriging only)
kappa Float The optimal smoothing parameter of the Matern model or M. Stein’s model. (Kriging only)

User Guide

The geostatistical toolbox in eRAMS performs spatial interpolation of any point shapefile. Currently, eRAMS
implements the automap package in R
to perform universal Kriging and automatically select the best model and its best parameters using linear least
squares regression (Hiemstra et al. 2009). The models used within eRAMS are the following:

  • Sph (Spherical)
  • Exp (Exponential)
  • Gau (Gaussian)
  • Mat (of the Matern family)
  • Ste (Matern, M. Stein’s parameterization)

This guide describes how to perform spatial interpolation using eRAMS.

Example

As an example for performing spatial interpolation, follow these steps:

    1. Upload a point shapefile into either the user layers or project layers

Right-click to add shapefile

Click Add → Shapefile to add the shapefile layer to the map

Click Add → Shapefile to add the shapefile layer to the map

Locate and highlight shapefile components

Locate the shapefile in your directory structure

Locate the shapefile in your directory structure

Check that the shapefile loaded by observing the map

Look at the map to ensure that your shapefile was successfully added

Look at the map to ensure that your shapefile was successfully added

    1. Open the spatial interpolation tool

Click the MapTools button on the upper left. Go to Geostatistics→Spatial Interpolation

The following dialog should appear

The following dialog should appear

06_open_spatial

    1. Run the spatial interpolation tool

Select the point layer and the attribute in the layer over which to interpolate, the interpolation method (currently only Kriging is available), and the cell size for the output raster file

Select the point layer and the attribute in the layer over which to interpolate, the interpolation method (currently only Kriging is available), and the cell size for the output raster file

After successful interpolation, the final model and selected parameters are displayed

After successful interpolation, the final model and selected parameters are displayed

 

Output maps for both prediction and variance (σ²) are shown on the map under project layers

Output maps for both prediction and variance (σ²) are shown on the map under project layers

10_map_var

Important Notes

Spatial interpolation on eRAMS can be performed anywhere throughout the world. However, problems occur when data cross multiple UTM zones because the tool projects any input
shapefile into a UTM projection based on the centroid of the point layer of interest. Analysis must be geographically small enough in order to avoid this issue,
and therefore, clipping must be performed prior to performing the spatial interpolation analysis.

References

Hiemstra, P.H., Pebesma, E.J., Twenhofel, C.J.W., and Heuvelink, G.B.M. (2009). Real-time automatic interpolation of ambient gamma dose rates from the Dutch Radioactivity Monitoring Network, Computers & Geosciences, 35(8), 1711-1721.