topic Re: Geostatistical analyst layer in ArcGIS GeoStatistical Analyst Questions

Geostatistical analyst layer

ErnestDunwoody — Mon, 22 Nov 2010 05:41:56 GMT

can any one tell me how to create a geostatistical analyst layer?

Sorry to have to ask such a dumb Q but ArcMAP help was helpless, yet again, on this topic.

Using the Create Geostatistical layer Toollbox requires a Geostatistical model source and an input layer.

What is this Model source? A misleading name for an ordinary layer or is it something else? Where do you get or create it?

The next box wants an Input dataset but I can't get anything to go in there, probably because the Layer I put in the first box doesn't have what it is needed.

Sure would be helpful if the Arc Map tool help files contained helpful information.

Sincerely,
edi151

Re: Geostatistical analyst layer

EricKrause — Mon, 22 Nov 2010 17:16:18 GMT

Geostatistical layers are fundamentally tied to interpolation methods and their associated parameters. Interpolation is generally done in the Geostatistical Wizard, and the output will be a geostatistical layer.

The Create Geostatistical Layer tool is used for creating a new geostatistical layer from an existing one. It requires a model source because the tool needs to know which interpolation method and parameters you are using. It sounds like you don't have an existing geostatistical layer, so the tool isn't going to work for you until you make one in the Geostatistical Wizard.

If you tell me what exactly you're trying to do (some kind of interpolation, I assume), I can provide more help.

Re: Geostatistical analyst layer

EricKrause — Mon, 22 Nov 2010 18:12:29 GMT

After reading your post again, it sounds like you are trying to convert a layer into a geostatistical layer. That can't be done. A geostatistical layer can be exported to a raster but not the other way around.

Re: Geostatistical analyst layer

ErnestDunwoody — Tue, 23 Nov 2010 08:47:57 GMT

Thank you very much for offering to help. I've read the Help files extensively but am having trouble getting started. I have a modest classical statistical background.

What I seem to have stumbled across so far is that you can use Geostatistical Analyst/Explore Data to get an insight into the spatial distribution of point vector data and you can convert raster layers into geostatistical layers using the Geostatistical Wizard in Geostatistical Analyst. Then you can use these Geostatistical Layers as input to the Geostatistical Tools in Arc ToolBox. Am I right so far?

I have two tasks.

1. I need to find the strength of association between my species distribution data (Greater Bilby distribution features such as burrows, feeding sites, fecal pellets etc) and a number of features in the landscape that I have analysed such as landcover class, elevation, aspect, soil type, soil density, soil texture and pH. I can do these by a series of spatial joins on pairs of data and extract the positive joins and count them and build up a table of frequencies this way. Long and tedious. No measure of the covariance. Was wondering if I could analyse the strength of association between each species feature and the landscape features using Geostatistical Analyst somehow?

2. In another project I have to develop measures of the directional variance of specific landscape features (e.g. clumps of trees, clumps of shrubs etc) for different landscape types. Basically I want to know the minimum transect length I have to use in each cardinal direction in each landscape type to capture 95% of the variance in the landscape. What I was hoping to do was to analyse the images into indices, e.g. NDVI as a simple example, and then measure the frequency of different sized NDVI clusters in each direction using semivariograms. How would I do this? Do I have to classify the NDVI pixel values into objects, create polygons out of each type and size class and then anayse the ditribution of polygon centroids? I was hoping I could do it by analysing the distribution of the raster of Indice values.

Thank you for your interest in my request. I will very much appreciate your advice.
Sincerely,
edi151

What I would now like to do is to get a measure of the directional variance of the landscape cover data.

Re: Geostatistical analyst layer

EricKrause — Mon, 29 Nov 2010 23:24:04 GMT

Sorry for the delay in response. You caught us on a particularly busy week.

I think you may have a few misconceptions about what Geostatistical Analyst is and what it can do. I'll do my best to clear up any confusions.

What I seem to have stumbled across so far is that you can use Geostatistical Analyst/Explore Data to get an insight into the spatial distribution of point vector data and you can convert raster layers into geostatistical layers using the Geostatistical Wizard in Geostatistical Analyst. Then you can use these Geostatistical Layers as input to the Geostatistical Tools in Arc ToolBox. Am I right so far?

Geostatistical Analyst deals almost exclusively with interpolation methods, with special emphasis given to kriging. Many variables change continuously across space (like temperature and elevation). The goal of interpolation is to predict the value of the variable at any location based on measurements taken at particular locations.

The general workflow is to start with point vector data, where each point contains a measurement of the variable at that location. Before proceeding with interpolation, the Exploratory Spatial Data Analysis (ESDA) tools allow you to explore the structure of your data. You can see histograms, apply transformation, explore trends, etc. Note that these "tools" are not geoprocessing tools (they are not in ArcToolbox); they are GUI-based and interactive, and they're found in the Geostatistical Analyst toolbar.

Once you are satisfied that your data is appropriate for interpolation, the points are used as input for the Geostatistical Wizard. Inside the Wizard, the GUI environment allows you to visually fit an interpolation model (interpolation methods like kriging have many parameters, and the quality of your predictions depends on the quality of your model). The Wizard provides many interactive diagnostics for fitting an appropriate model.

When you are satisfied with the model, the output of the Wizard is a geostatistical layer. This layer is a smooth surface of predictions for the variable. It is visualized as a contour map, but the predicted value of the variable can be calculated at any point on the surface.

Geostatistical layers are different than raster layers in that they are dynamic. A geostatistical layer contains all the information it needs to make predictions at any specified location, but it doesn't actually store the predictions themselves. Instead, it stores the mathematical model needed to make the predictions, and then it calculates them when you ask it to.

Geostatistical layers can be converted to raster, and this is done by simply applying the mathematical model to the center of each cell in the raster. Ignoring memory and time constraints, you can make the cell size of the raster as small as you want.

You cannot convert a raster layer into a geostatistical analyst layer because the resolution of the raster is limited by the cell size, whereas geostatistical layers have infinite resolution (limited only by the precision of the computer). You could, however, convert the raster to points, then use these points to build an interpolation model in the Wizard. The Wizard would then create a geostatistical layer from the model (though I would be very cautious before doing this).

The tools in the Geostatistical Analyst toolbox are generally used for post-processing after a geostatistical layer is created in the Wizard (such as converting a geostatistical layer to a raster or getting predictions for a set of points). In addition, the interpolation methods that do not require a high level of interactive model fitting are offered as tools. Like the Wizard, they take point vector data as input and produce geostatistical layers of predictions.

All that being said, it doesn't sound like interpolation is what you need. I wish I could be more help, but my specialty is interpolation, and I'm just not sure how to go about solving your particular tasks. You may want to look into Geographically Weighted Regression, which is available in the Spatial Statistics toolbox.

Let me know if you have any other questions, and I'll do my best to help.

Re: Geostatistical analyst layer

XIANWANG — Fri, 05 Jun 2015 14:14:29 GMT

Hi, Eric

When I read your post, I am confused about these two layers, what is the difference between two layers: the existing geostatistical layer and the new one, which we create using the Create Geostatistical Layer tool. I mean we already created on using Geostatistical Wizard, why we need to use Create Geostatistical Layer tool to create a new one.

Thank you.

Re: Geostatistical analyst layer

EricKrause — Fri, 05 Jun 2015 15:51:04 GMT

The key is that you can apply the interpolation parameters to a new dataset. Indeed, it doesn't make much sense to apply it to the same data (this would just make a copy of the layer). Here is a typical scenario that shows how this tool can be useful:

You have a monitoring network set up to take daily measurements of pollution levels. Each day, these sensors measure the pollution and store it into a database, and each day you want to create an interpolated map of pollution levels. On the first day, you take a lot of time to create a quality kriging model in the Geostatistical Wizard and output a geostatistical layer. On the next day, you can use this layer as the model source in Create Geostatistical Layer and apply the kriging parameters to the measurements of the new day. By reusing the original geostatistical layer for the data of each new day, you can fully automate this process and not have to manually go into the Geostatistical Wizard every day to create your map.

Re: Geostatistical analyst layer

XIANWANG — Fri, 05 Jun 2015 16:01:20 GMT

Thanks a lot. Here I have a question. How can I apply the geostatistical layer which created by Create Geostatistical Layer to the measurements of the new day?

Re: Geostatistical analyst layer

EricKrause — Fri, 05 Jun 2015 16:18:01 GMT

You'll have to create the first geostatistical layer in the Geostatistical Wizard (or from a geoprocessing tool in the Interpolation toolset). To interpolate a new dataset, open Create Geostatistical Layer and provide the first geostatistical layer as the model source. The tool will analyze the model and decide what kinds of datasets are required. In the simplest case, this would be just a dataset and a field. In a more complex case like cokriging, it might require multiple datasets and multiple fields. You then put the new data into the "Input datasets" parameter of the tool. Give the output geostatistical layer a name and run the tool. It will then create a new geostatistical layer that applies the parameters from the model source to the new dataset.

Re: Geostatistical analyst layer

XIANWANG — Fri, 05 Jun 2015 17:17:17 GMT

Thanks a lot.

To create the template layer is to use the Geostatistical Wizard. In the Wizard, you can specify the interpolation parameters for the template layer. With ArcGIS 10, you can even use the Wizard to optimize various interpolation parameters.

1. Now if I am using Geostatistical Wizard to create a template layer, and I use optimize various interpolation parameters. Then I use this template layer as model resource to create a geostatistical layer for the measurements of the new day. These optimize various interpolation parameters also works for the measurements of the new day.

2. In Geostatistical Wizard, you optimize various interpolation parameters for the variance and semivariogram. What is the difference, and how do I know which one should be chosen.

Re: Geostatistical analyst layer

EricKrause — Fri, 05 Jun 2015 18:32:54 GMT

This is difficult question to answer in an online setting. The differences are very technical and mathematical, but the key difference is that the covariance view requires knowledge of the mean value of the kriging surface, and the semivariogram does not. If you are unsure which one to use, I would suggest the semivariogram.

Both methods are designed to estimate covariance matrices that are used in the kriging equations, but they both do this estimation in different ways, and their "optimal" parameters will be different (though hopefully not very different).

Re: Geostatistical analyst layer

XIANWANG — Fri, 05 Jun 2015 18:39:14 GMT

Hi,

When I use the Geostatistical Wizard, I set some parameters, which confused me a lot.

In step 2, the order of trend removal. my data include precipitation data from several years. In Some months, the data has trends, but in some data, there is no trends. When I do the interpolation, for these data, I should use the some criteria. I mean for all the data, I should choose the some parameters. Here, should I choose to remove the trend? If so, it will influence the data which don't have any trend.
In step 3, General Properties, there is Kerenel Function, which is a local polynomial and is a qick interpolator. And in step 4, we have semivariogran model which is also a interpolator. Can anyone tell me difference between them.

Thanks a lot!

Re: Geostatistical analyst layer

EricKrause — Fri, 05 Jun 2015 19:34:31 GMT

This workflow is really only suggested when you know that the new datasets will have the same general structure. Applying the same parameters to datasets that are different (ie, the presence and absence of trends) is not a good idea.

As for trend removal, the idea is to first do a very basic polynomial interpolation to remove the general trends, then you do kriging on the autocorrelation that is left over. The idea is that there is a general trend in the data, but there is also correlated variability around that trend. Trend removal takes care of the first part, and kriging takes care of the second part. In essence, you are interpolating twice, but the first time is only for general trends in the data.

Re: Geostatistical analyst layer

XIANWANG — Sat, 29 Aug 2015 16:17:09 GMT

Hi, Erik

I have another question for Kringing Geostatistical Wizard. Can I set the lower and upper values for the results of Kriging interpolation? Thanks!

Re: Geostatistical analyst layer

EricKrause — Mon, 31 Aug 2015 15:45:07 GMT

No, you cannot control the maximum (or minimum) value of the interpolation. Whatever the maximum is (if there even is a maximum) will come from the kriging equations. These equations are constructed based on all the provided parameters, and the maximum/minimum will be whatever the equations say they are.

In fact, if we could detect the maximum and minimum, my life would be a lot easier. Currently, we symbolize geostatistical layers based on the min/max of the input data, not the actual interpolation. If we knew the min/max of the interpolation, we could symbolize the geostatistical layer based on these values. If you have ever noticed that geostatistical layers often look very different after being exported to raster, it is because of this issue. If we could detect the min/max of the geostatistical layer, we could make the symbology closely match the raster.

Re: Geostatistical analyst layer

XIANWANG — Mon, 31 Aug 2015 17:18:52 GMT

Thanks a lot. Yes, I find the the range of geostatistical layer is different from the raster one. So now I am thinking if I can set the range of geostatistiacal layer. What I mean is that I got the geostatistiacal layers, and I will set the range of these geostatistiacal layers, for example the minimum is 0, then I can use GA Layer To Points tool to get the values of the points, which supposed to be larger than 0.

Re: Geostatistical analyst layer

BillGrimm — Thu, 26 Sep 2024 17:35:57 GMT

So what you're essentially saying is that the tool does the whole Geostatistical Wizard process (including semivariogram/covariance modeling, selecting the optimal lag size/other parameters, etc.) for you based on the model you feed it (i.e., you're essentially just telling it, "use XYZ interpolation method on this dataset"?)?

OR are you telling it to use XYZ interpolation method on this dataset with this specific nugget value, this specific lag size, etc. (where the specific nugget value, lag size, etc. are the actual numbers from the model you feed it, which is based on a different dataset)?

It seems like the second option would produce inaccurate results, but it would be helpful to have clarification as to how the tool actually works on its help page.

Re: Geostatistical analyst layer

EricKrause — Fri, 27 Sep 2024 13:06:48 GMT

I mean the second. And, yes, if the new dataset has fundamentally different properties than the first, it will be quite inaccurate. This is only appropriate in cases like daily measurements of the same data, where the changes in the data values will be relatively small, and you're willing to sacrifice a small amount of accuracy in order to not have to manually perform kriging in the Geostatistical Wizard every day.

Kriging is never particularly safe to do in an automated environment, but if it's something you really need to do, we recommend using Empirical Bayesian Kriging for it. It is available as a geoprocessing tool, so setting up automation is relatively simple.

Re: Geostatistical analyst layer

BillGrimm — Fri, 11 Apr 2025 21:28:30 GMT

Hi @EricKrause, this was very helpful, thank you. We've developed a workflow for our process (creating 3D models of subsurface sand and gravel) using EBK3D. Not knowing the actual math behind the kriging (I've taken multivariable calculus but not linear algebra), I have a couple of questions about the kriging that I need answers to in order to make our workflow as robust as possible. Is there a chance we could set up a Zoom call with you or someone on your team to show you what we're working with and get your insight? Thanks so much!