|
POST
|
In ArcGIS Pro, there are no explicit "ESDA tools" like there are in ArcMap. Instead, all graphical exploration is run through charting. Charts were first released in ArcGIS Pro 1.2, and each release, they are adding new charts. They already have a Histogram, and in the upcoming 2.1 release, they will include a QQ Plot chart that will work both as a Normal QQ Plot and a General QQ Plot. They will continue to implement new charts, but I can't promise when each of the ESDA tools from ArcMap will be implemented in ArcGIS Pro. If possible, it would actually help if you could specifically say which charts you want, and how you generally use them in your day-to-day work. There are many potential charts that can be implemented, and the more we hear from you about charts you need, the easier it is to decide which ones get made before others.
... View more
11-07-2017
10:55 AM
|
1
|
2
|
3707
|
|
BLOG
|
Spatial interpolation is one of the most common workflows in GIS, and the Geostatistical Analyst extension is built specifically to solve this problem. However, there is often confusion about how exactly interpolation should be done. Which interpolation method should I use? Which parameters should I use? How do I know if the interpolated surface can be trusted? These questions frequently seem daunting to people when they first approach spatial interpolation, and the purpose of this blog is to help you get started towards your goal of accurate spatial interpolation. Don’t have Geostatistical Analyst? Try Geostatistical Analyst for free. As a starting point, we suggest the Geostatistical Analyst tutorial: Download the ArcGIS Tutorial Data for Desktop through my.esri.com. You must have an up-to-date license for ArcGIS for Desktop to download the tutorial data. There are five tutorials that you should complete. It will take a few hours to finish all of them: Introduction to the ArcGIS Geostatistical Analyst Tutorial Exercise 1: Creating a surface using default parameters Exercise 2: Exploring your data Exercise 3: Mapping ozone concentration Exercise 4: Comparing models Exercise 5: Mapping the probability of ozone exceeding a critical threshold Once you have seen the videos and done the tutorial exercises, you should be ready to start using Geostatistical Analyst on your own data. If you need additional education material, the following Web Courses are available: Modeling the Unknown: Spatial Interpolation with ArcGIS Pro Exploring Spatial Patterns in Your Data Using ArcGIS 10 Creating Prediction Surfaces in ArcGIS If you have reviewed the training material above but still have questions, feel free to post your questions to GeoNet at the Geostatistical Analyst Place. Good luck and happy kriging! ------------------------------------------------------------------------------------ Geostatistical Analyst also has extensive documentation that is available for free online. Here are a few select topics about some of the most important features in the extension: Learn about the different types of kriging. Learn about Empirical Bayesian Kriging, a modern kriging technique that automates many of the most difficult aspects of kriging. Learn about Empirical Bayesian Kriging 3D, the first interpolation method developed by Esri for data collected in 3D. Watch of video of Empirical Bayesian Kriging 3D. Learn about Areal Interpolation, an interpolation method that interpolates data collected in polygons and allows you to predict values in a different set of polygons. ------------------------------------------------------------------------------------ UPDATE (September 2017) - Free textbook and data I am happy to announce that we have made Spatial Statistical Data Analysis for GIS Users available as a free download. This textbook, written by Konstantin Krivoruchko, was previously available through Esri Press: "This book presents a practical introduction and guide to spatial statistics for researchers, statisticians, academics, and college students who want to expand their knowledge and skills in geographic information system (GIS) technology to new areas of analysis. More than 1,000 full-color illustrations are included, along with lessons and sample data to help organize courses and lectures." Download link for the PDF book: https://downloads.esri.com/esripress/pdfs/spatial-statistical-data-analysis-for-gis-users.pdf Download link for datasets used in the book: https://downloads.esri.com/esripress/pdfs/spatial-statistical-data-analysis-for-gis-users.zip ------------------------------------------------------------------------------------ UPDATE (May 2018, February 2019) - LearnGIS exercises using Geostatistical Analyst Several LearnGIS exercises have been created that make heavy use of Geostatistical Analyst. LearnGIS lessons are free online exercises that teach different concepts related to geographic analysis in real-world workflows. They are a great way to see geostatistical workflows from beginning to end. Analyze Urban Heat Using Kriging - Use Simple Kriging, Empirical Bayesian Kriging, and EBK Regression Prediction to interpolate maps of urban heat in Madison, Wisconsin. Model Water Quality Using Interpolation - Use Kernel Interpolation With Barriers to interpolate dissolved oxygen levels in Chesapeake Bay. Interpolate 3D Oxygen Measurements in Monterey Bay - Use Empirical Bayesian Kriging 3D to interpolate oxygen measurements taken along vertical columns in the ocean. ------------------------------------------------------------------------------------- UPDATE (July 2019) - "Evaluation of empirical Bayesian kriging" has been peer-reviewed and published in the August issue of the Journal of Spatial Statistics. This paper presents the theory and principles behind EBK as well as results from controlled simulations. The results show EBK is accurate and valid for a wide variety of data and generally performs very well compared to other modern interpolation methods. https://www.sciencedirect.com/science/article/pii/S2211675319300168 ------------------------------------------------------------------------------------- UPDATE (June 2020) - "Empirical Bayesian kriging implementation and usage" has been published in the June 2020 edition of "Science of The Total Environment." This peer-reviewed paper contains the details of the mathematical underpinnings of Empirical Bayesian kriging. https://www.sciencedirect.com/science/article/pii/S0048969720308007
... View more
09-12-2017
02:48 PM
|
8
|
0
|
7780
|
|
POST
|
I think the link will exist indefinitely. If it ever breaks, just post here and we'll try to get it corrected or rehosted. As far as our preference for referring students to the link versus distributing it yourself, do whatever is easiest for you and your students.
... View more
09-11-2017
11:10 AM
|
2
|
0
|
12223
|
|
POST
|
I am happy to announce that we have made Spatial Statistical Data Analysis for GIS Users available as a free download. This is a PDF textbook written by Konstantin Krivoruchko, one of the founding members of Geostatistical Analyst. This textbook was previously available through Esri Press: "This book presents a practical introduction and guide to spatial statistics for researchers, statisticians, academics, and college students who want to expand their knowledge and skills in geographic information system (GIS) technology to new areas of analysis. More than 1,000 full-color illustrations are included, along with lessons and sample data to help organize courses and lectures." Download link for the PDF book: https://downloads.esri.com/esripress/pdfs/spatial-statistical-data-analysis-for-gis-users.pdf Download link for datasets used in the book: https://downloads.esri.com/esripress/pdfs/spatial-statistical-data-analysis-for-gis-users.zip Enjoy! -The Geostatistical Analyst team
... View more
09-11-2017
09:51 AM
|
9
|
14
|
20097
|
|
POST
|
Did you use the same semivariogram parameters for simple and ordinary cokriging? If any of the parameters are very different, the two surfaces could end up looking very different. Remember that in cokriging, you must estimate three different covariance models: the semivariogram for the primary dataset, the semivariogram for the secondary dataset, and the cross-covariance between them. You can change which model you are viewing by using the control at the top of the Wizard. "Var1 - Var1" shows the primary semivariogram. "Var1 - Var2" shows the cross-covariance. "Var2 - Var2" shows the semivariogram for the secondary dataset. Unless all three of these models are set to the same (or similiar) values, simple and ordinary cokriging may produce very different maps.
... View more
07-10-2017
12:40 PM
|
1
|
0
|
2093
|
|
POST
|
Hi Clement, Geostatistical Analyst does not treat categorical data any differently than continuous data. It doesn't know that your categorical field is a categorical field, and it is just operating on the numbers in the field the same way it does with any other field. This is true for all types of kriging, simple and otherwise. As for why you are seeing better maps with simple cokriging than with regression kriging and ordinary cokriging, I honestly do not know. As you said, you should expect regression kriging to outperform cokriging for categorical covariates. We often see simple kriging outperform ordinary kriging, but that is mostly due to the very flexible transformation, which you said you didn't use. The counter-intuitive results may have something to do with how the categories are coded (how are they coded?) in your database. Before I write this off as just "some datasets are weird", I want to talk to a couple of colleagues. However, they are on vacation right now. I will see them next week at the User Conference, but we may not have time to discuss this. If I don't get a chance to discuss it at the User Conference, I will talk to them the week after and get back to you. Sorry I couldn't be more help right now, but this is a bit of a mystery, and there might not be any rational explanation other than just quirks in this particular dataset. -Eric Krause
... View more
07-06-2017
12:10 PM
|
2
|
0
|
2093
|
|
POST
|
Hi Naci, You will have a better chance of getting this question answered if you ask it at the https://community.esri.com/community/gis/analysis/spatial-statistics place.
... View more
02-28-2017
09:33 AM
|
1
|
0
|
1474
|
|
POST
|
Glad to hear that it is working for you now. Create Geostatistical Layer does accept both layer files and xml files as input. I don't know why you could only get the layer file to work, but I guess it isn't worth worrying about now that you have your script working.
... View more
02-17-2017
08:04 AM
|
0
|
0
|
679
|
|
POST
|
I think you have a couple problems. First, use the field name, not the field alias. Change this line: FieldName = field.aliasName to FieldName = field.name Later in the code, you need to pass the variable into the tool with: arcpy.GACreateGeostatisticalLayer_ga("C:/Arc_Workspace/Metals_Workspace.gdb/KernelS_Template.xml", "C:/Arc_Workspace/Metals_Workspace.gdb/BH_Points_export " + FieldName + "; C:/Arc_Workspace/Metals_Workspace.gdb/Water_Boundary", "outGL") If you pass "field" inside the string of the parameter, it won't be recognized as a variable. It will look for a field called "field" and fail when it doesn't find one. You will also want to store the output geostatistical layer (the last parameter) as a variable. If you keep the same name ("outGL") for each iteration, it will just get overwritten after each iteration. Or you can save each one as a layer file, but those will also need unique names. I am not the greatest Python programmer, and I can't directly test this without your data, but I think what I just outlined will work. Optionally, you may want to also consider storing all of the parameters as variables in order to make it easier to read and understand. Something like: inXML = "C:/Arc_Workspace/Metals_Workspace.gdb/KernelS_Template.xml" inPoints = "C:/Arc_Workspace/Metals_Workspace.gdb/BH_Points_export" infield = "<your field>" inBarrier = "C:/Arc_Workspace/Metals_Workspace.gdb/Water_Boundary" inDataset = inPoints + " " + infield + ";" + inBarrier outGAlayer = "outGL" arcpy.GACreateGeostatisticalLayer_ga(inXML, inDataset, outGALayer)
... View more
02-16-2017
12:59 PM
|
1
|
2
|
3667
|
|
POST
|
Thanks for the clarification. You'll need to alter the statement to something like: arcpy.GACreateGeostatisticalLayer_ga("C:/temp/myModelSource.xml", "C:/temp/myDataset.shp myField ; C:/temp/myBarriers.shp", "myNewGALayer") The difference is that you need to pass the feature class, the field, and the barriers in a single string. You separate the two datasets with a semicolon like in the example above.
... View more
02-16-2017
10:00 AM
|
1
|
5
|
3667
|
|
POST
|
Hi again, Your syntax should be something like: arcpy.GACreateGeostatisticalLayer_ga("C:/temp/myModelSource.xml", "C:/temp/myDataset.shp myField", "myNewGALayer") The first parameter should be a file path to the xml. The second parameter will be a string where you first specify the file path to the dataset (shp and fgdb both work) followed by a space, then the name of the field. The third parameter is the name of the new geostatistical layer that is created by applying the parameter from the model source to the new dataset. Since geostatistical layers are saved in memory, you probably also want to save them as a layer file on disk so that you can easily open them in maps. After the code above, use something like: arcpy.SaveToLayerFile_management("myNewGALayer","C:/temp/galayerfile.lyr") Regarding the X, Y, and F1 options that appear in the example, these options are not required in your case. They are used when the X and Y coordinates of the points are stored in fields rather than in the SHAPE attribute of the feature class. If your datasets were, for example, csv files (which don't have a SHAPE attribute), you could pass the X coordinates, the Y coordinates, and the field (F1) using those options. One final question... are you using any barriers in your interpolation? If so, I'll need to give a slightly different code example of how to do that. If you aren't using barriers, the code above should work.
... View more
02-16-2017
09:26 AM
|
1
|
1
|
3667
|
|
POST
|
Hi Bryan, You are correct that the Kernel Interpolation with Barriers geoprocessing tool does not have a search neighborhood option. This option is only available in the Geostatistical Wizard and only for Exponential, Gaussian, and Constant kernel types. However, you can automate your process using the Create Geostatistical Layer geoprocessing tool. This tool takes a model source as input and allows you to apply that model source to a new dataset. To create the model source, use the Geostatistical Wizard and perform Kernel Smoothing with the parameters that you want (it doesn't matter which dataset you use). When you click Finish, there will be a "Save..." option on the Method Report dialog. This will allow you to save an xml file that will serve as the model source in Create Geostatistical Layer. In your Python script, run Create Geostatistical Layer for each of your datasets, and use the xml from before as the model source for all of them. Since you are using ArcMap 10.1, you will need to provide your datasets and fields as strings (more recent versions can use an arcpy class). See the script examples in the tool documentation above, and feel free to post more questions if you run into any problems with the script.
... View more
02-15-2017
02:17 PM
|
1
|
1
|
3667
|
|
POST
|
If you have ArcGIS 10.4, you should also have access to ArcGIS Pro 1.2 (or even a more recent version of ArcGIS Pro). If you have a Geostatistical Analyst license, there is a tool in ArcGIS Pro starting in version 1.2 called EBK Regression Prediction. It allows you to perform Empirical Bayesian Kriging using rasters as explanatory variables. In your case, you would provide the temperature feature class and the field to interpolate, then use the DEM as an explanatory variable raster. This will build local simple kriging models where the mean value of the model is defined by building a regression equation between the temperature values and the DEM values. You could also try cokriging, as Dan_Patterson recommended, but I strongly suspect that you will get a more accurate interpolation from EBK Regression Prediction than from cokriging.
... View more
02-02-2017
08:47 AM
|
3
|
1
|
4182
|
|
POST
|
It is simulating from a Gaussian distribution, and negative values are valid simulations from this model. The phenomenon that you are interpolating may not be able to be negative (like rainfall or pollution levels), but the model doesn't know this, and some of the simulations will likely contain negative values just due to chance. I only know one way to guarantee that all simulated values will be greater than zero, but it requires changing your simple kriging geostatistical layer. In the Geostatistical Wizard, make sure that you are using simple kriging, and use a Normal Score Transformation. On the transformation page (with the histogram on the left), choose Lognormal, Gamma, or Log Empirical for the Base distribution (found on the right). To use these options, all your data values must be greater than zero, but these options will guarantee that all predictions and simulations will also be greater than zero.
... View more
01-05-2017
12:43 PM
|
1
|
0
|
1284
|
|
POST
|
I think Steve is right. You are probably doing unconditional simulation when you want to be doing conditional simulation. To do conditional simulation, you must provide the dataset and the field that you use to create the kriging layer into the Gaussian Geostatistical Simulations tool. You provide them in the "Input conditioning features" and "Conditioning field" parameters. The idea here is that there are an infinite number of surfaces that all have the same covariance structure (ie, the same semivariogram). When you do an unconditional simulation, you simply create several of them at random. However, these surfaces do not pass through the same set of points, and their high/low values will not occur in the same places. When you average over all these simulations, you will get close to a constant raster. But there are also an infinite number of surfaces that all have the same covariance structure and are conditioned to pass through a given set of points. You can specify these conditioning points to be anything that you want, but the most common thing to do is to condition that the simulations must pass through the input points from the kriging layer. When you do this, all simulations will resemble the original kriging layer, and when you take an average of the simulations, you will get something close to the original layer. The more simulations that you perform, the closer the average will look to the original layer. Steve's picture shows how this works. Both conditional and unconditional simulations have uses, but it can sometimes be tricky to tell which one you should use. In your case, it sounds like you want to be doing conditional simulations and conditioning on the features that were used to create the kriging layer.
... View more
12-15-2016
03:11 PM
|
1
|
1
|
2194
|
| 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
|