Trend Polynomial Gridding

This gridding method is used to illustrate regional trends in your data. It tries to fit a polynomial trend surface to your data. This is a 3-dimensional surface represented by a polynomial equation.

In automatic mode, the program will find the surface with the best fit to your data points, trying different orders of the trend surface equation (e.g. a flat surface, a surface with one bend, two bends, etc.). How well the surface fits is determined by a "correlation factor" that the program computes for each surface that is fit to the data. The greater the correlation factor the better the fit.

Note that equally important to the trend surface information are the Trend Surface Residuals. These are the localized differences from the regional trend, and can be helpful in isolating anomalous areas of interest.

Advantages: This method can project trends into areas with no data. It also shows regional structures otherwise hidden by "high frequency" features.

Disadvantages: This mapping method does not work well with data sets that have no discernible regional trend. It does not honor the control points.

Automatic: Choose this option if the program is to pick the best fit polynomial order for your data set.
1st-Order - 6th-Order: If Automatic is not selected, you can choose from one of the listed orders by clicking in its radio button. The "order" of a trend surface equation refers to the highest values of the exponents used in the equation.
- For example, an equation with x and y (to the first power) is a "first order" equation, and creates a flat surface.
- An equation with x2 and y2 is a "second order" equation, and creates a surface with one bending moment.
- An equation with x3 and y3 is a "third order" equation, and creates a surface with two bends.
For a summary of how well each polynomial order fits your data, you can run, separately, a trend surface report.

Back to Grid Method Summary

RockWare home page