# Kriging questions - trend removal and error statistics

###
#1
Posted 09 May 2012 - 06:11 AM

I just had a couple of questions about some of the steps used in kriging in ArcGIS geostatistical analyst.

Firstly, I am working with data that I know has a trend first order trend, so I have chosen to remove this trend. What are the differences between neighbourhood and global trend removal, and what should I be aiming for when setting the % of each? Minimising my error? I am also not sure whether I should use the anisotropy option if I am already removing the trend - my understanding was this was to account for trends in data?

Secondly, I was wondering if someone could explain the error statistics to me in a simple way (I have relatively little knowledge of statistics). I understand that I want to minimse the mean error, standard error, and rms, and that the rms should be similar to the standard error, but I’m not sure which of these are most useful for communicating the accuracy of my resulting data. I’m slightly confused as on some datasets I seem to get very low mean error but the other values seem quite high. Does this mean that high errors in both directions are effectively cancelling each other out when the mean is taken?

###
#2
Posted 09 May 2012 - 10:20 AM

Hi,

I just had a couple of questions about some of the steps used in kriging in ArcGIS geostatistical analyst.

Firstly, I am working with data that I know has a trend first order trend, so I have chosen to remove this trend. What are the differences between neighbourhood and global trend removal, and what should I be aiming for when setting the % of each? Minimising my error? I am also not sure whether I should use the anisotropy option if I am already removing the trend - my understanding was this was to account for trends in data?

Secondly, I was wondering if someone could explain the error statistics to me in a simple way (I have relatively little knowledge of statistics). I understand that I want to minimse the mean error, standard error, and rms, and that the rms should be similar to the standard error, but I’m not sure which of these are most useful for communicating the accuracy of my resulting data. I’m slightly confused as on some datasets I seem to get very low mean error but the other values seem quite high. Does this mean that high errors in both directions are effectively cancelling each other out when the mean is taken?

Hi Rhienne,

I'm not sure the best way to go about your problem, but I did want to point you in the direction of the GIS Stack Exchange web forum. There are several geospatial stats experts offering lots of good advice there on a daily basis. Good luck!

###
#3
Posted 10 May 2012 - 05:32 AM

Hi,

I just had a couple of questions about some of the steps used in kriging in ArcGIS geostatistical analyst.

Firstly, I am working with data that I know has a trend first order trend, so I have chosen to remove this trend. What are the differences between neighbourhood and global trend removal, and what should I be aiming for when setting the % of each? Minimising my error? I am also not sure whether I should use the anisotropy option if I am already removing the trend - my understanding was this was to account for trends in data?

In your case, your DEM can be thought of as a noise signal (true elevation values) + a background signal (the trend). In this case you want to remove the background signal.

One form of global de-trending is to use polynomial interpolation to remove (subtract) the first-order trend (manifest as a constant gradient of elevation values across the DEM domain).

Local de-trending involves removing (subtracting ) a smoothed version of the DEM from the unsmoothed. The degree of smoothing of the smoothed DEM would need to be determined by trial and error.

#### 0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users