|
POST
|
1. If no transformation is applied, the underlying model is an intrinsic random function, specifically a Power semivariogram model. If a transformation is applied, it is a simple kriging model with an exponential semivariogram. 2. The prior distribution is estimated from the data using restricted maximum likelihood (REML). This is the difference between "Bayesian" and "empirical Bayesian." In empirical Bayesian models, the prior is estimated from the data, and the posterior distribution is generated through simulations. 3. There are several reasons to think EBK will, in general, be more accurate. First, REML is known to be a better estimator of semivariogram parameters than weighted least-squares (which is what is used in other kriging methods). Second, EBK works on subsets, so it eases the stationarity assumption of kriging. Other kriging methods assume global stationarity. EBK, however, only assumes stationarity within subsets, so it can work effectively even in the presence of global nonstationarity, as long as the subsets are close to being stationary. Third, EBK does a better job of estimating kriging variances because it can account for error in estimating the underlying semivariogram (it does this with simulations); other kriging models assume the semivariogram is modeled absolutely correctly, which often results in standard errors that are too small. Here is a reference for EBK (a longer paper with all the details is in the works): Krivoruchko K. and Gribov A. (2014) Pragmatic Bayesian kriging for non-stationary and moderately non-Gaussian data. Submitted. In Mathematics of Planet Earth. Proceedings of the 15th Annual Conference of the International Association for Mathematical Geosciences. Eds: Pardo-Igúzquiza, E.; Guardiola-Albert, C.; Heredia, J.; Moreno-Merino, L.; Durán, J.J.; Vargas-Guzmán, J.A. Springer 2014, pp. 61-64. (We're not sure why the date is 2014 because it's already been published)
... View more
09-24-2013
09:57 AM
|
0
|
0
|
2620
|
|
POST
|
This topic explains what you need to know: http://resources.arcgis.com/en/help/main/10.2/index.html#/Examples_of_using_iterators_in_ModelBuilder/00400000001n000000/ESRI_SECTION1_33EDB1C8829F4CF4BBD5DABAA3ED34DD/
... View more
09-16-2013
02:37 PM
|
0
|
0
|
2620
|
|
POST
|
Looking at your graphic, it looks like you've decided to use quadrants 1 and 4. The shadow of the (1,1) bin may extend into the 2nd and 3rd quadrants. The shadow in the 2nd quadrant needs to be inflected into quadrant 4, and the shadow from quadrant 3 needs to be inflected to quadrant 1. These weights need to be added to the weights that are already in those quadrants. Here's a quick test: make two points with the same coordinates so that their vector plots at the origin. The (1,1) and (1,-1) bins should each get a weight of 0.5 (they'll each get 0.25 naturally, then another 0.25 from the inflection). If that doesn't work, I don't think I can help you. You've pretty much exhausted my knowledge of our binning mechanism.
... View more
09-12-2013
01:35 PM
|
0
|
0
|
5559
|
|
POST
|
I talked this over with the developer that programmed the semivariogram functions, and we're not sure why your red bands aren't matching ours. We suspect that you have a bug in the part of the algorithm that inflects weights that cross over the axes, but we're not sure. When you're inflecting the weights, make sure that you are rotating them rather than just mirroring them. That could be the problem.
... View more
09-12-2013
11:44 AM
|
0
|
0
|
4323
|
|
POST
|
You don't need to do any adjustment. Ozone is a continuous phenomenon, so its value changes everywhere. When you extract the points from a raster, all you're doing is asking for the ozone value at the closest cell center. You don't need to do any corrections for the size of the grid cell. Again, you can eliminate any imprecision from using the closest cell center by using GA Layer to Points. It will calculate the ozone value at the exact point rather than looking to the closest cell center.
... View more
09-12-2013
11:04 AM
|
0
|
0
|
3441
|
|
POST
|
Remember that the vector between two points, A and B, can be defined two ways: from A to B and from B to A. For this reason, the bins in the 1st and 3rd quadrants are identical, and so are the bins in the 2nd and 4th quadrants. This is why the semivariogram map is symmetrical about the two diagonals. Keeping this in mind, you need to pick two quadrants to use (we use the 1st and 2nd), then you'll need to inflect any weights that fall into the 3rd and 4th quadrants. Be careful not to double-count each pair.
... View more
09-12-2013
10:21 AM
|
0
|
0
|
4323
|
|
POST
|
Regarding the boundary cells, the description is on page 6. Also, there's a picture of the triangular kernels in Figure 6 (in the appendix). And the averaging of the bins is the average of the semivariances of all pairs within that bin. The trick is that there are actually two binning mechanisms: one involves gridded cells and the other uses concentric circles (Figures 2 and 3 in the appendix). In the semivariogram UI, the red dots come from the gridded cells, and the blue crosses come from the concentric circles. These circular bins are what are actually used to fit the semivariogram (the red dots are more for diagnostic information and for calculating the semivariogram map). The semivariogram parameters are estimated using weighted least-squares, where the weights come from the (modified) number of points that fall within each circular bin.
... View more
09-12-2013
09:00 AM
|
0
|
0
|
4323
|
|
POST
|
Rather than converting the geostatistical layer to a raster and extracting the points, you should just use the "GA Layer to Points" geoprocessing tool. It will bypass the creation of the raster, and merging all of the outputs will be easy. Also, you might want to consider using Empirical Bayesian Kriging (EBK) instead of Ordinary Kriging. The results are generally more accurate, and EBK is implemented as a geoprocessing tool, so it is much simpler to automate.
... View more
09-12-2013
08:30 AM
|
0
|
0
|
3441
|
|
POST
|
You can find a full description of binning mechanism and weights in this paper: http://downloads2.esri.com/campus/uploads/library/pdfs/30583.pdf When you hover over a cell in the semivariogram map, the weight you see is the number of pairs of points that fall within that bin. The reason the weights aren't whole numbers is that the individual weight of each pair is distributed into the surrounding bins. For example, if a pair of points falls exactly on the corner between 4 square bins, each of those bins will get a weight of 0.25 from that pair. If the pair falls exactly at the center of a square bin, that bin will get a full weight of 1, and all other bins will get 0 weight. The "Weight" in the semivariogram map is the sum of all weights for all pairs of points within that bin. If that explanation wasn't clear, please read the above paper.
... View more
09-10-2013
03:01 PM
|
0
|
0
|
4323
|
|
POST
|
Glad to hear the method worked! But I want to give a bit of a disclaimer. Looking at Surfer's documentation, their Linear semivariogram does not support a Nugget (which is kind of strange), and they default the variogram slope to 1 without setting a range or sill. This essentially means that the semivariogram starts at 0 and increases with a slope of 1 forever (in ArcGIS, setting nugget=0 and the range and partial sill to the same very large value reproduces this behavior). However, there is no reason to think, in general, that nugget=0 and a slope of 1 are good values. So, the workflow I presented will reproduce the defaults of Surfer, but there is no reason to think this model actually fits your data. You noted this in your first post, but I just want to reiterate it because it's very important.
... View more
08-28-2013
09:30 AM
|
0
|
0
|
3800
|
|
POST
|
I'm not 100% sure, but I think this will reproduce the Surfer output. Choose Ordinary kriging with a Linear semivariogram. In advanced options, leave the lag size blank, set the nugget to 0, then set the partial sill and major range equal to each other. Make the value of these two parameters very large; it must be larger than the distance of the diagonal extent of the input points. For the search radius, leave it as "Variable," and set the maximum number of points to the number of points in your dataset. Leave the maximum distance empty. You'll also need to know the cell size of the raster you created in Surfer, and put that in the output cell size.
... View more
08-02-2013
03:07 PM
|
1
|
0
|
3800
|
|
POST
|
No, we do not use Incremental Spatial Autocorrelation for the search radius. In the EBK geoprocessing tool, the search radius is defaulted to one quarter of the diagonal extent of the input points (as is every other Geostatistical Analyst interpolation tool that uses a searching neighborhood). However, since it also uses minimum/maximum neighbors, the search radius isn't actually very important. It will search as far as it needs to in order to get the minimum number of neighbors, then it will stop when it either hits the maximum number of neighbors or the end of the search radius. Also, I'm not sure why you need template layers for EBK. Using template layers is our suggested way of automating kriging (except EBK) because there is no kriging geoprocessing tool in Geostatistical Analyst. But since EBK has a gp tool, there's no reason to use template layers.
... View more
07-31-2013
11:19 AM
|
0
|
0
|
818
|
|
POST
|
Ok, I see the problem. We have a known issue with using spaces in output feature class names. If you remove the spaces from the name of your shapefile, it should work. Sorry for the inconvenience, and we're working on a fix for this problem.
... View more
07-19-2013
01:06 PM
|
0
|
0
|
2402
|
|
POST
|
I cannot reproduce this problem. Did you get this error when running GA Layer to Points geoprocessing tool?
... View more
07-19-2013
12:22 PM
|
0
|
0
|
2402
|
|
POST
|
This can be done in ModelBuilder or in Python. You only need to make a single template with the Geostatistical Wizard. This template will define the type of kriging (Simple, Ordinary, etc), the semivariogram model (Stable, Spherical, etc), and any advanced options such as transformations and anisotropy. When you click Finish in the Wizard, choose to Save the xml model source in a convenient location. This will be the model source that you will use for all your datasets. Then you need to make a Python or ModelBuilder script that iterates through all your datasets in Moving Window Kriging while keeping the same xml model source. If you need help with Python or ModelBuilder, you should contact Esri Support Services or ask on some forums about Python or MB.
... View more
07-18-2013
08:19 AM
|
0
|
0
|
1975
|
| Title | Kudos | Posted |
|---|---|---|
| 2 | 01-16-2025 04:52 AM | |
| 1 | 10-02-2024 06:45 AM | |
| 2 | 08-23-2024 09:18 AM | |
| 1 | 07-19-2024 07:09 AM | |
| 1 | 08-21-2012 09:47 AM |
| Online Status |
Offline
|
| Date Last Visited |
02-25-2026
06:39 PM
|