new SuperMap.InterpolationKrigingAnalystParameters(options)
This class allows you to set the parameters required for Kriging interpolation analysis.
Kriging, mainly applied in geology, is a method to interpolate spatial data. It uses the variance of data points to solve the weight each known sample point carries for estimating an unknown point.
Then the value at an unknown point is calculated by combining the values and weights of the known sample points. The most unique characteristic of the Kriging method is that it not only provides estimations with minimum prediction errors but also provides the specific error amounts.
Generally, many geological attributes, such as the terrain surface, are continuous. Thus the attribute values at two points within a short distance are related.
By contrast, if two points on an irregular surface are far apart, they are considered statistically independent. This trend of spatial continuity over distance can be depicted by a semivariogram.
If one needs to estimate the value at an unknown point with values at known discrete points, the semivariogram of the point values can be used to model the spatial relationship between the known points and the point to be estimated.
Then the corresponding spatial parameters can be used to calculate the semivariances, which are then used to calculate the weights of the known points for estimating the value at an un-sampled point. Finally the value at the unknown point is calculated.
· nugget: When the interval of sampling points is 0, the semivariance is 0 in theory. But actually the value is not 0 when the sampling point time is very close. So the nugget effect appears.
The corresponding semivariance is nugget. The nugget may be created by the measurement errors or spatial variation.
· sill:With the increase of interval of sampling points, the semivariance tend to a stable constance. This constant is the sill. After reaching the sill, the interval won't change with the interval of sampling points.
The sampling points larger than this interval do not have the spatial correlation.
· Meta-sill: the difference between sill and nugget.
· The autocorrelation threshold value (range): It is the interval of sampling point when the semivariance reaches the sill. The sampling points that exceed the range do not have the spatial correlation. It will not impact prediction results.
Know then, semivariance is the key of Kriging interpolation. So the right semivariance model is very import. SuperMap provides the following three models:
· EXPONENTIAL: The spatial correlation is exponential decrease with the sample interval increase. The spatial autocorrelation in the sample spacing of infinity disappears entirely.
· SPHERICAL: The spatial correlation decreases with the sample interval increase until beyond a certain distance that the spatial autocorrelation is disappear.
· GAUSSIAN: It applies to the situation that the semivariance gradually close to the base station.
In the semivariance, there is a key parameter, namely the average of interpolation fields. The different processing methods for this parameter derive different Kriging methods. The interpolation functions of SuperMap are based on the following three Kriging algorithm:
Simple Kriging: For this method, it is assumed that the expected value (mean) of the field values for interpolation is a given constant.
Kriging:Assume that the filed average used in the interpolation is unknown and constant. This method generates the grid dataset by simulating the grid value according to some given points and a certain math functions. It not only can generate a surface, but also can calculate the accuracy of the estimated result. So, this method has a higher precision.
Universal Kriging: For this method, it is assumed that the expected value (mean) of the field values for interpolation is an unknown variable. Universal Kriging is suitable if there is a certain trend existing in the sample data and this trend can be fit with a function or a polynomial.
The advantages of the Kriging method is that it has spatial statistics as its theory basis. It has clear physical meaning. It not only can assess the spatial variation of an attribute, but also can estimate the distribution of variance in the attribute. However, it requires complicated calculations, and the semivariance function usually needs to be specified by the user empirically.
| Name | Type | Description | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
options |
Object |
The parameters.
|
Example
var myInterpolationKrigingAnalystParameters = new SuperMap.InterpolationKrigingAnalystParameters({
dataset:"SamplesP@Interpolation",
type: "KRIGING",
angle: 0,
mean: 5,
nugget: 30,
range: 50,
sill: 300,
variogramMode: "EXPONENTIAL",
searchMode: "QUADTREE",
maxPointCountForInterpolation: 20,
maxPointCountInNode: 5,
pixelFormat: "BIT16",
zValueFieldName: "AVG_TMP",
resolution: 30000,
filterQueryParameter: {
attributeFilter: ""
},
outputDatasetName: "myKriging"
});Extends
Members
-
anglenumber
-
Sets the rotation angle for the Kriging algorithm. This angle is the anti-clockwise rotation angle of each search extent with reference to the horizontal direction.
-
https://leafletjs.com
-
boundsSuperMap.Bounds L.Bounds ol.extent
-
The range of the interpolation analysis which is used to determine the range of the result raster dataset. If default, then it is the range of the source dataset. Since this is a method of interpolation, the parameter is better less than or equal to the range of the source dataset.
-
clipParamObject
-
The parameter to clip the interpolation analysis result.
-
datasetstring
-
The name of the dataset in the datasource for interpolation analysis. This identifier is the format "dataset name@datasource alias".
When the InterpolationAnalystType is dataset, it is required. -
expectedCountnumber
-
[Fixed Point Number Search] Sets the number of the points to be looked up, i.e., the point number involved in the difference operation, and the default is 12; [Fixed length Search] sets the minimum number of the points to be looked up, the default is 12.
-
exponentSuperMap.Exponent
-
[Universal Kriging] The orders of the trend surface equations in the sampling data.
- Default Value: SuperMap.Exponent.EXP1
filterQueryParameterSuperMap.FilterParameter
The property filter condition.The default is null, meaning analysis all points in the dataset
-
https://leafletjs.comhttps://leafletjs.com
inputPointsArray.<(SuperMap.Geometry.Point|L.LatLng|L.Point|ol.geom.Point)>
he discrete point (Z values included) set used in the interpolation analysis. When the InterpolationAnalystType is geometry, it is required. There is no need to specify the datasets inputDatasourceName, inputDatasetName and zValueFieldName through discrete data interpolation analysis.
InterpolationAnalystTypestring
InterpolationAnalystType. The interpolation contains dataset interpolation and the geometry interpolation. "dataset" represents the dataset interpolation; "geometry" represents discrete data interpolation. Default value is "dataset".
- Default Value: "dataset"
maxPointCountForInterpolationnumber
[Block] Sets the maximum count of points involved in the operation within a single block. It is only used in Kriging. Simple Kriging and universal Kriging don't support block search.
maxPointCountForInterpolationnumber
Sets the maximum count of interpolation points. It is only used in Kriging. Simple Kriging and universal Kriging don't support block search.
meannumber
The mean of the interpolation field, i.e., the sum of the interpolation field values of the sample points divided by the sample point count.
nuggetnumber
The nugget value in Kriging.
outputDatasetNamestring
The name of the interpolation analysis result dataset.
outputDatasourceNamestring
The name of the interpolation analysis result datasource.
pixelFormatSuperMap.PixelFormat
Specifies the pixel format of the output grid dataset. The supported pixel format is: BIT16, BIT32, DOUBLE, SINGLE, UBIT1, UBIT4, UBIT8, UBIT24, and UBIT32.
rangenumber
The autocorrelation threshold in Kriging whose unit is identical to that of the original dataset.
resolutionnumber
The resolution of the raster dataset of the interpolation result, the actual distance denoted by a cell, has the same unit with that of the point dataset. This value cannot be larger than the length of the bounds to be analyzed. The setting of this value should depend on the dataset extent. If the value of the cell (result_dataset_extent/resolution) is less than 500, this value will better show the trend of the density.
searchModeSuperMap.SearchMode
The search mode for interpolating includes fixed point number, fixed length search and block search. Required. Simple Kriging doesn't support block search. Details as follows: {KDTREE_FIXED_COUNT} finds points involved in the interpolation analysis with the KDTREE fixed point number way. {KDTREE_FIXED_RADIUS} finds points involved in the interpolation analysis with the KDTREE fixed length way. {QUADTREE} finds points involved in interpolation analysis (block search) with the QUADTREE method.
searchRadiusnumber
Search radius is the search scope of points in the operation, whose unit is identical to that of the point dataset. When calculating the Z value of a location (the default value is 0), the sampling points within the range whose radius is the value of the search extent will involve in the operation. Users need to set this value according to the distribution of points to be interpolated and the extent of the dataset.
sillnumber
The sill in Kriging.
The kriging type, required. Details as follows: {KRIGING} kriging, {SimpleKriging} simple Kriging, {UniversalKriging} Universal Kriging.
variogramModeSuperMap.VariogramMode
Defines the semivariogram mode of Kriging Interpolation method. The selected mode influences the prediction of the unknown values, particularly when the shape of the curve nears the origin differs significantly. The steeper the curve near the origin, the more influence the closest neighbors will have on the prediction. As a result, the output surface will be less smooth.
- Default Value: SuperMap.VariogramMode.SPHERICAL
zValueFieldNamestring
In the dataset interpolation analysis, The field name whose value is used in the interpolation analysis. The field of text type is not supported by interpolation analysis. It refers to the weight of each sampling point. You can set all of this field value to 1. In this case, all points have the same weight. When the InterpolationAnalystType is dataset, required.
zValueScalenumber
The zoom scale of the value to be interpolated whose default value is 1. Specifies the zoom scale of the value to be interpolated.
- Default Value: 1
Methods
-
destroy()
InterpolationKrigingAnalystParameters.js, line 193 -
Disposes the resource. The property that references resource is set to null.