POST
|
Opps I think I was reading the wrong axis. The power looks about 1.6-1.75. The empirical transformation provided the better qqplot compared with a logempirical transformation but both were not quite as accurate on the prediction front as the default EBK. I checked the data to see if the deviations might be caused by measurement errors but they are accurate and represent relatively large differences in groundwater level head over short distances. This is because one well is representing the shallower unconfined aquifer and the other wells an artesian aquifer with much higher head. I get very good predictions using ordinary and simple kriging with trend removed and a better fit using these models for the middle data along the qqplot, however the transformed EBK does forces more points to fall closer to the normal line. The issue with the simple and ordinary kriging, despite for the large deviations at the tail, is that the RMSSE is about 3 and not 1. For EBK it is near to 1. Using the transformed EBK and looking at the prediction statistics and qqplot am I able to use the prediction standard error maps or would this be misleading for some reason? Simon
... View more
06-26-2014
04:26 PM
|
0
|
0
|
964
|
POST
|
I'm having some trouble deciding how best to model my groundwater level data. I have evaluated it using the ESDA tools and I have a positively skewed non-normal data set with a WNW to ESE global trend. I have tried a couple of different kriging models and the EBK provides the best overal result in terms of prediction error statistics. My power function is not between 0.25 and 1.75 as recommended in the help files so I'm wondering what does this mean. Also I'm not entirely sure what my normal QQ plot is telling me- does the line have to be 1:1 or does the slope not matter just as long as the data points fall along the line? Thanks Simon. Ps: I also noted there are more options for different semivariogram models in the 10.2 files..I assume these options aren't available in the 10.1 version.
... View more
06-25-2014
07:36 PM
|
0
|
4
|
4266
|
POST
|
Hi, I have some questions relating to kriging that I???m hoping you might help me with?? I want to redesign my groundwater level monitoring network using the kriging standard error as the criterion for optimising my design. There are many journal articles to help guide me in my work. However few comment on the assumptions and limitations behind the methods used. My questions relate to variogram analysis and defining an appropriate spatial structure for my redesign. Groundwater levels in Hawke???s Bay fluctuate seasonally and possess a long-term decline, therefore the spatial structure or variogram model must change accordingly. As such, how can I produce a valid variogram model for my redesign if the model parameters are not static? (I.e. change with time). Is this just part of the limitation of the method? Or is there another way to combat the temporal component of the data? Kind regards, Simon
... View more
07-19-2012
08:46 PM
|
0
|
1
|
1315
|
POST
|
Hi Steve, I'm quite new to this so bear with me. I understand the reason for making a weight raster but not sure how to tell the densify sampling network toolbox to you use it. Do you specify it under the Environmental settings>raster analysis>mask??? or do I specify it under the processing extent?? Cheers
... View more
06-04-2012
12:53 PM
|
0
|
0
|
423
|
POST
|
Hi, I want to use the Densify Sampling Network to show the optimum locations to position 10 new sites based on a reduction in the standard error. However I want to constrain the analysis to an irregular area (shapefile/raster). How do I do this?? I figure you either have to modify the geostatistical layer or change something in the Environments tab? Cheers, Simon
... View more
05-31-2012
03:52 PM
|
0
|
4
|
2688
|
POST
|
hi, I'm hoping someone might be able to tell me whether I'm on the right track or be able to help. Eric from ESRI has been helping me with many of my questions and has been fantastic but I thought I would add this to the thread so not to bother him with my endless emails. I am trying to replicate work I have read in Journal papers to determine the optimum density of sites for a network. The method requires altering the number of sample points and examining the change in the variance of the estimation error. Because the location of the point and the variogram model only effect the variance (i.e and not the actual measured value at the point) I should be able to examine the effectiveness of adding or removing fictious sample locations on the variance. I'm not entirely sure how the steps are performed to do this but this is what I think: Step one: determine the best variogram model and parameters from my known measured values Step two: using the model parameters and location of the measured values calculate the average kriging standard deviation or variance for my measured values (as a global indicator of performance) Step three: Add fictious points to the sample set and recalculate the average kriging standard deviation or variance using the sample model parameters in step one - only x,y, are needed for the fictious points to caclulate the standard deviation. Step four: repeat this for a number of different site additions - say 10, 20, 50, 100 new sites Step five: plot the average kriging standard deviation against the number of observation wells assessed- the graph should flattened where the addition of new wells makes little difference to the reduction in standard deviation. At this point I will say this is my optimum number of sites needed for the biggest reduction in error. Do these steps sound right to anyone? How do extract the model parameters after variogram analysis How do I set the model parameters for subsequent analysis and addition of fictious points? Where do I find the standard deviation of each point added to the network? Hope this make some sense. Cheers, Simon
... View more
06-09-2011
06:57 PM
|
0
|
1
|
679
|
POST
|
hi, I'm looking to krige my groundwater levels. When I examine the distribution I notice my data isn't normall distributed and even when I attempt to transform the data I get no improvement (okay maybe a slight improvement on my histogram but none on my q-q plot). Does that mean I shouldn't krige?
... View more
05-25-2011
09:00 PM
|
0
|
1
|
3102
|
Online Status |
Offline
|
Date Last Visited |
11-11-2020
02:24 AM
|