Interpolate Points predicts values at new locations based on values measured at a set of point locations.
The output is a hosted feature service that contains one to three feature layers.
A GIS analyst has a dataset with air quality points across Thailand. The Interpolate Points tool can be used to predict air quality values across the country to create an air quality surface.
Try the Interpolate values tutorial for the complete workflow.
Interpolate Points includes configurations for input layers, analysis settings, result layers, and output settings.
The Input layers group includes the following parameters:
- Input points is used to choose the points that will be interpolated.
- Field to interpolate is used to choose the field to interpolate, such as elevation or temperature.
- Clipping polygons is used to specify a polygon study area that will be used to clip the prediction and standard error layers. Only results within the polygons will be included in the outputs. If a clipping polygon is provided, the tool will use the extent of the polygons by default so that the result layers fill the full extent of the study area.
The Interpolation settings group includes the following parameters:
- Calculation precision is used to choose your preference between accurate predictions and calculation speed. This parameter automatically sets various options and configurations of the statistical model used to interpolate the points. The following options are available:
- Speed—The interpolation model will be optimized for faster calculations by using the fewest number of simulations and employing the most efficient options and configurations.
- Balance—The interpolation model will be balanced between speed and accuracy by using typical options and configurations. This is the default.
- Accuracy—The interpolation model will be optimized for accurate and precise calculations by using the largest number of simulations and the most complicated options and configurations.
- Classification type is used to determine the class break values when contouring the result layers. The following options are available:
- Equal area—Class break values will be calculated so that the number of input data values in each polygon is equal.
- Equal interval—Class break values will be calculated so that the range of predicted values is equal for each polygon.
- Geometric interval—Class breaks will be calculated based on a geometric series. This option ensures that each class range has approximately the same number of values in each class and that the change between intervals is consistent. This is the default.
- Manual—Class breaks will be defined using custom break values. Provide these values in Class break values.
- Number of classes is used to specify the number of class breaks that will be used. This parameter only applies for equal area, equal interval, and geometric interval classification types. The default is 10. The value must be between 2 and 32.
- Class break values is used to specify custom class break values for the manual classification type. For each class break value, enter the value on the dialog box and click Add. You must provide between 2 and 32 unique values.
The Result layers group includes the following parameters:
- Output features name determines the name of the feature service containing the result layers that will be created and added to the map. The name must be unique. If a layer or service with the same name already exists in your organization, the tool will fail and you will be prompted to use a different name.
- Optional layers is used to create optional outputs. Optional layers includes the following subparameters:
Create standard error layer is used to specify whether a layer of standard errors will be calculated for the predicted values. When checked, a polygon layer named PredictionErrors will be included in the output feature service. This layer represents the standard errors of the predicted values. The default is unchecked (no standard error layer will be created).Standard errors are used to quantify the precision and reliability of the predicted values. A common rule is that 95 percent of the time, the true value at a new location will fall within two standard errors of the predicted value. For example, a new location has a predicted value of 50 and a standard error of 5. This means that the best estimate of the true value at that location is 50, but it reasonably could be as low as 40 or as high as 60. To calculate this range of reasonable values (known as a 95 percent confidence interval), multiply the standard error by 2, add this value to the predicted value to get the upper end of the range, and subtract it from the predicted value to get the lower end of the range.
Point prediction locations is used to specify point locations that will be used to calculate predicted values and standard errors. If a value is provided, a point layer named PredictedPointLayer will be included in the output feature service. This layer is a copy of the point prediction locations with the predicted values and standard errors included as new fields.Predicting to point locations is useful when the values of specific locations are the most important, such as predicting air quality levels at school or hospital locations.
- Save in folder specifies the name of a folder in My Content where the result will be saved.
The following limitations apply to the tool:
- At least 10 input points are required.
- The values of the points should be spatially continuous, meaning that the values change smoothly and consistently over the study area. This tool is not appropriate for count data, such as population, or data that changes abruptly over short distances, such as median income.
- If you use a clipping polygon, the result layers may have fewer classes than the value provided for Number of classes. This can occur when a class is entirely outside of the clipping polygon and is clipped out of the result layer.
- The prediction and standard error output layers share the same classification type and number of classes. However, if manual class break values are used, the standard error layer will instead use an equal interval classification. This is because predictions and standard errors are measured on different scales, so you should not use the same raw values for their class breaks.
Analysis environment settings are additional parameters that affect a tool's results. You can access the tool's analysis environment settings from the Environment settings parameter group.
This tool honors the following analysis environments:
This tool consumes credits.
Use Estimate credits to calculate the number of credits that will be required to run the tool. For more information, see Understand credits for spatial analysis.
This tool includes the following outputs:
- ResultLayer—A polygon layer in which each polygon represents a range of predicted values (known as a filled contour). This is the primary result of the interpolation and is always created. The layer contains fields showing the minimum and maximum values of each class range.
- PredictionError (optional)—A polygon layer in which each polygon represents a range of standard error values for the predictions. This layer is created if the Output prediction errors parameter is checked.
- PredictedPointLayer (optional)—A point layer containing the predicted values and standard errors at a particular set of point locations, such as a layer of schools or hospitals. This layer is created if a point layer is provided for the Point prediction locations parameter.
This tool requires the following licensing and configurations:
- Creator or GIS Professional user type
- Publisher, Facilitator, or Administrator role, or an equivalent custom role
Use the following resources to learn more: