Visually Verify Grid Results Through Surfer Gridding Software

You’ve used Surfer to grid your data and created a great looking map. You put it in the final report for your presentation. And you think, that was easy and I’m all done now.

The final map is prepared and ready for the report.

And then someone asks you, how do you know that map is accurate? You start to wonder. How do I know if it is correct? Is there any way to verify the gridding? If the grid file matches the original data, then I could confidently say the map is correct. But, how do I know if the grid file matches the data?

There are a few ways to determine if a grid file matches the data. You can visually inspect the grid, comparing Z values to known points on map. If you want a more numerical analysis, you can calculate the residual value difference between the grid and the data by clicking the Grid | Residual command. This numerical method will be discussed in another blog later this week.

Visually looking at the difference between the grid and the data is a quick check. To visually inspect the grid, overlay a contour map from the grid with a classed post map from the original data.

To do this:

1. Grid the data with the Grid | Data command.
2. Create a contour map:
   1. Click the Map | New | Contour Map command.
   2. Select the grid file created in step 1 and click Open.
3. Change the contour properties
   1. Click on the Contours layer in the Object Manager.
   2. In the Property Manager, click on the Levels tab.
   3. Note the Minimum contour, Maximum contour, and Contour interval. We’ll use this in determining how many classes to use for the classed post map. In my case, the contours start at 0 and end at 24000 with a contour interval of 1000.

      
      Note the minimum, maximum, and contour interval.
      These will be used later to determine the number
      of classes for the classed post map.

   4. Check the box next to the Fill contours option.
   5. Next to Fill colors, click the Grayscale and select Rainbow.
   6. Check the box next toColor scale to see the range of Z values for each color.

      
      The contour map is used to visually see where the Z values are mapped in the grid.

Add a classed post map:
Click on the existing contour map to select it.Click the

Map | Add | Classed Plost Layer

command. This command adds a new layer to the existing map. Select the original data file and click

Open

. Click

Yes

in the dialog asking to automatically adjust the limits, if prompted. The classed post is added to the contours.
Change the class properties to match the contour properties.
Click on the

Classed Post 

layer in the

Object Manager

. Click on the

Classes

tab in the

Property Manager

. Click on the

Edit Classes

button to open the

Classes

dialog. In the dialog, set the

Binning method

to

User defined

. Double-click on the

>=Minimum

value in the first class. Change it to the

Minimum contour

value from the contour map and click

OK

. In my case, this is 0. Double-click on the

<Maximum

value in the first class. Change it to the

Minimum contour

value plus the

Contour interval

value and click

OK

. In my case, this is 1000. Double-click on each consecutive

<Maximum

value for each additional class. Set each value equal to the previous value plus the

Contour interval

. Note that depending on how many contours were displayed, you may need to click the

Insert Class

button to add additional classes. Click the

Symbol

button. The

Class Symbol Properties

dialog opens.
1. Set the Method to Gradational.
2. Click the Fill color. Set the colormap to Rainbow and click OK.
3. Click the Symbol button and select a single symbol.
4. Click OK in both dialogs and the classes update on the contour map.

Visually compare the original data points and the contour
lines to get an idea of how the grid represents the data.

Visually, you can see that symbols typically fall in the same colored contour areas. This gives a good indication of how well the data is fit by the grid. You can compare specific Z values (or ranges of Z values) between the classed post map and contour map by labeling the classed post points and comparing the values to the color scale object or contour line labels. You can also move the cursor over the location and seeing the Z value of the contours at that point in the status bar. So, now you can say with confidence that the grid looks to be a good representation of the data.