Introduction to Surfer
Surfer generates contour and surface maps from spatial data that varies with location (or any function of 2 variables: Z = f(x,y). Examples include elevation as a function of latitude and longitude, precipitation as a function of latitude and longitude, % infestation of emerald ash borers as a function of latitude and longitude, and so on. You input data and the program attempts to generate contours that may reveal patterns or trends not obvious from just looking at the table of values.
Surfer is published by Golden Software (home office is in Golden, Colorado, home of the Colorado School of Mines) so the files used to open the application is usually in a "Golden Software" folder, probably in the Programs folder.
Surfer is copy-protected by means of a license code. The BO 3051 has 20 licenses. Fortunately, any one of the codes works with any one of the BO 3051 computers. Once you have logged in and entered a valid Surfer code, you will not have to re-enter that code on that desktop.
Link to BO 3051 codes – write one down, please.
Today we will work with 2 data sets:
Using your Internet Explorer browser, right click on each of these data links and "save target as" to a location that you will access later.
Once you have these data sets downloaded and stored, return to Surfer.
The first step in Surfer is using the data to generate a grid.
Data dropdown menu, Grid.
Navigate to your data and open the 1 km data.
Surfer picks the high and low X and Y values, making these the default limits of the grid, and selects a grid spacing that gives you more or less 100 rows or columns, whichever has the larger range. Never use the defaults except to get a quick and dirty look at the data displayed. Before resetting grid geometry, look at methods for generating a grid. Kriging is best for widely scattered, irregular data as is commonly developed in field surveys. This method attempts to fit a surface through 4 points in 4 quadrants about a grid point where the value is to be interpolated. When data are dense, you will get about the same contour map no matter which contouring method you use. Sparse data can yield a surprising variety of contour maps depending on the contouring method. For this grid, use kriging.
Know your data: are the data points close together or far apart? Regular or scattered? You can more or less accurately interpolate one or two grid points between data points, a more dense interpolation may introduce details unsupported by the data.
If your grid limits are not nice even values, reset the maximum and minimum X and Y to round values. Then set the grip spacing to a round value that, divided into the range of X and Y, yields an integer (number of grid lines). For this 1 km data, 0.5 km is a good choice. Unless your data are anisotropic (densely spaced readings along widely spaced profiles), always use the same spacing for X and Y grid interval values.
Note that the location of the output grid file is in the same folder and has the same name as the data file. Click the folder button to change one or the other. Remove the check from the ‘Generate grid report’ box, I have yet to discover what grid reports are good for.
When you click "OK" Surfer begins to generate a grid. Lots of data and lots of grid nodes makes for lots of time needed to make the grid. But 15 years ago things were even slower. It took 10 hours to produce a single 100 x 100 grid from a couple hundred data points.
Once the grid is ready you are ready to make maps. Map dropdown menu reveals a selection of map types – begin with Contour Map - New Contour Map.
I will demonstrate some ways to make contour maps more suitable. Double click the contour map opens a dialog box.
Beware smoothing contours – sometimes this results in maps that violate contour map rules. Never smooth unless you are absolutely certain that it is appropriate and necessary.
In order to display locations of data points, under the Map menu, select Post Map. Drag the post map to one side and double click it for that dialog box.
Now select both the contour map and the post map (select on, hold down the Shift key and select the other). Under the Map menu, chose Overlay Maps. The data points and contours are now part of a single map. You can edit one or the other by selecting the appropriate item from the control column on the left side of the screen.
Axis labels and fonts
Titles and text boxes
Under the File menu, you may Export your map as an image. This is important because you cannot plant a Surfer document directly into Word or other Microsoft application.
Now try a Surface map (Map menu).
Double click to get the dialog box. View, lighting, scale are useful for making your surface map look more interesting.
Assignment: generate contour and surface maps using the dense data set. Increase the number of nodes (decrease grid interval – note that this data set uses meters, not kilometers, for X and Y). This grid has data at about 90 meter intervals – if you use a grid interval of 100 meters or more, Nearest Neighbor yields a valid result without the need to search and interpolate – much faster than kriging!
Do a side-by-side comparison of contour maps generated from the 1 km and the 3-arc-second data. Tell me what you see in one that is not clear in the other. Do the same for surface maps. Show both surface maps from the same view with the same lighting and the same scales.
I find making maps and surface displays of topography is fun (must be the nerd in me). Later this semester you will select some feature you think interesting and go through the entire process of going after the data and turning it into a surface map.