4th Report

FOURTH REPORT (08-11-07)
Kriging with external drift


Sorry again, I am late. From Wednesday to Sunday I am working in the last Spanish F3 meeting and I don’t have internet in the circuit. I have just sent you an excel file and a manual explaining you how it woks (it`s very simple).

In anyway, it´s just a first version program and I am waiting for your feedback to improve it. Please send me your ideas or write them on your website.
My spreadsheet is limited because it`s designed for a given number of known points and for aeromaps with equidistant points.
To calculate the trend I need at least two points to create a line. The problem now, it`s the estimation of trends in the aeromap corners (red cells in the next figure). How you can see in the following figure, in the places where it isn’t possible to estimate (well I know it`s possible, but it isn’t easy to obtain reliable estimations) I don’t know which it`s the better option.



4th Report - White-Smoke































I have used the trend calculated manually from the universal kriging. It means that I use the red marked values (in the UK aeromap) from next picture to calculate the trend of aeromaps corner however the blue marked values cannot be used because I will need an extra point. This is the main problem that I have got.4th Report - White-Smoke

During the week I have also done Simple Least Squares Data Fitting for the Gauss variogram model, in this way the results will be more accurate. For the next version of the excel that I have sent you, you will have this feature. I have also been investigating your proposed methods in the links, but it isn`t easy to obtain some information about them because they are private companies. In anyway I will keep on searching new methods of extrapolation. Probably, I will have more time to work on this the next week.


For the next week, I will try:

- to improve the KED spreadsheet, taking into account your feedback.

- to investigate alternative methods.


I look forward to hearing from you

José Miguel Martínez Turégano.

More pages