The Visualize Space Time Cube in 2D and Visualize Space Time Cube in 3D tools can be used to visualize and explore the variables and analysis results stored in netCDF cubes created using the Create Space Time Cube By Aggregating Points, Create Space Time Cube From Defined Locations, or Create Space Time Cube From Multidimensional Raster Layer tools. The outputs vary based on the Cube Variable and Display Theme parameter values specified in each tool. The following table contains brief descriptions and links directly to more information for each Display Theme value below:
Display Themes for 3D | Description |
---|---|
The numeric value of the Cube Variable parameter will be displayed. | |
The statistical significance of each bin will be displayed based on the space-time hot spot analysis run in Emerging Hot Spot Analysis. | |
Bins with estimated values will be displayed. | |
The cluster or outlier type (COType) for each bin determined by Local Outlier Analysis will be displayed. | |
The count of records aggregated into each space-time bin will be displayed. | |
The input time steps and the resulting forecasted values from the Time Series Forecasting tools will be displayed. | |
The results of the Change Point Detection tool will be displayed. The output will contain fields indicating whether each time step is a change point along with estimates of the mean or standard deviation for the current and previous time step. | |
The results of the Outlier Option parameter in the Time Series Forecasting tools will be displayed. |
Display Themes for 2D | Description |
---|---|
All locations that contain data for the Cube Variable parameter will be displayed. | |
The trend of values at each location that were determined using the Mann-Kendall statistic will be displayed. | |
The trend of z-scores at each location that were determined using the Mann-Kendall statistic will be displayed. | |
The results of the Emerging Hot Spot Analysis tool for the specified Cube Variable parameter value will be displayed. | |
The results of the Local Outlier Analysis tool for the specified Cube Variable parameter value will be displayed. | |
The total percentage of outliers at each location will be displayed. | |
The outliers occurring in the most recent time period will be displayed. | |
The results of the Time Series Clustering tool for the specified Cube Variable parameter value will be displayed. | |
Locations that have no spatial neighbors for the last analysis run will be displayed. These locations rely only on temporal neighbors for analysis. | |
The number of bins that were estimated for each location will be displayed. | |
The locations that were excluded from analysis because they had empty bins that did not meet the criteria for estimation will be displayed. | |
The results of the Time Series Forecasting tool used for the specified Analysis Variable parameter value will be displayed. | |
The results of the Change Point Detection tool. The output features display the number of change points at each location, along with pop-up charts that display the original time series, change points, and estimates of the mean or standard deviation of each segment. | |
The results of the Outlier Option parameter in the Time Series Forecasting tools will be displayed. |
Display Themes for 3D
The following themes can be displayed in a globe scene. Descriptions of each Display Theme available in the Visualize Space Time Cube in 3D tool are as follows:
Value
With the Value option, the raw numeric value of the selected Cube Variable value is displayed for each bin in the space-time cube. This can be particularly important to visualize if you have chosen to aggregate during cube creation by either aggregating points into bins or temporally into locations.
A chart is also created displaying the Value over time for the whole space-time cube.
Hot and cold spot results
The Hot and cold spot results option shows the statistical significance of each bin based on the space-time hot spot analysis run in Emerging Hot Spot Analysis. The tool calculates the Getis-Ord Gi* statistic for each bin based on the neighborhood set in the tool run. The Emerging Hot Spot Analysis results in 2D identify trends in these results.
A chart output is also created plotting the z-scores over time for the entire cube.
Cluster and outlier results
The Cluster and outlier results option displays the COType assigned to each bin from the space-time analysis run in the Local Outlier Analysis tool. The tool calculates the Anselin Local Moran's I statistic for each bin and based on the neighborhood parameters set in the tool run. The Local Outlier Analysis results in 2D categorize these results by location.
This parameter option also includes a Moran's Scatterplot chart, which can be useful to identify space-time bins that are outliers or anomalies.
Estimated bins
Upon creation of the cube, some bins may have been estimated using Summary Fields or the Fill Empty Bins with parameters. Bins with estimated values are displayed with this display option. This can be especially useful information for locations that have many estimated bins in sequence either at the beginning or end of the time series at a location, as you may not be able to trust forecasting results in these particular locations.
Temporal aggregation count
The Temporal aggregation count option displays the count of records aggregated into each space-time bin. This can be helpful to gauge the density of records that were input into each bin at the creation of the space-time cube.
Forecast results
The Forecast results option displays the results of the forecast method for the selected Cube Variable value. The original time steps of the Input Space Time Cube parameter and the forecasted values added from time series forecasting are displayed.
This option also includes a chart displaying the Forecast Results value over time for the entire cube.
Time series change points
The Time series change points option displays the detected change points in 3D. Time steps detected as change points are drawn in purple and labeled Change Point, and time steps not detected as change points are drawn in light gray and labeled Not a Change Point. The output will contain fields indicating whether each time step is a change point along with estimates of the mean or standard deviation for the current and previous time step.
Time series outlier results
The Time series outlier results option displays the temporal outliers found when using the Outlier Option parameter in the Time Series Forecasting tools for the specified Cube Variable value.
Display themes for 2D
The following themes can be displayed in a map. Descriptions of each display theme available in the Visualize Space Time Cube in 2D tool are as follows:
Locations with data
The Locations with data theme allows you to see all locations that contain data for the selected Cube Variable value and drops locations for areas where no points were aggregated. This option is always available for every cube created by aggregating points.
Trends
The Trends theme shows you where values in your space-time cube have been increasing or decreasing over time. These trend results are calculated using the Mann-Kendall statistic run on the specified Cube Variable value selected for each location.
Hot and cold spot trends
The Hot and cold spot trends display theme shows you where the z-scores for each location have been increasing or decreasing over time. These results are calculated using the Mann-Kendall statistic run on the z-scores for the specified Cube Variable value selected for each location.
Emerging Hot Spot Analysis results
With the Emerging Hot Spot Analysis results option, the results of the Emerging Hot Spot Analysis tool stored in the space-time cube for each location are re-created and displayed. These results identify trends for each location in your cube including new, intensifying, or diminishing hot spots and cold spots.
Local Outlier Analysis results
With the Local Outlier Analysis results option, the results of the Local Outlier Analysis tool are re-created and displayed. These results indicate significant clusters and outliers in your space-time cube by bin and then categorize each location by results over time.
Percentage of local outliers
The Percentage of local outliers display theme shows the total percentage of outliers at each location over time. This can be useful to identify and investigate which locations were outliers more often compared to their neighbors.
Local outlier in the most recent time period
The Local outlier in the most recent time period option shows all locations that were considered outliers most recently.
Time Series Clustering results
The Time Series Clustering results option re-creates and displays the results of the Time Series Clustering tool for the Cube Variable value specified. The Time Series Clustering tool identifies the locations in a space-time cube that are the most similar and partitions them into distinct clusters in which members of each cluster have similar time series characteristics. Time series similarity can be clustered in a variety of ways, such as similar values across time or similar periodic patterns across time.
Locations without spatial neighbors
For the last analysis run, the Locations without spatial neighbors option shows all locations that have no spatial neighbors and, as a result, rely on temporal neighbors for the analysis. This can be useful, as it indicates locations that rely on less information in the analysis than its neighbors.
Number of estimated bins
The results of the Number of estimated bins option display the number of bins that were estimated for each location. Analysis results for locations with a high number of estimated bins should be investigated to be sure the analysis results can be trusted.
Locations excluded from analysis
The Locations excluded from analysis option displays locations that did not meet the requirement for estimation for the specified analysis.
Time series change points
The Time series change points option displays the number of change points detected at each location.
The output features included pop-up charts displaying the original time series with orange circles representing change points and dashed green lines indicating segment mean values.
Time series outlier results
The results of the Outlier Option parameter in the Time Series Forecasting tools for the specified Cube Variable value are displayed. These locations indicate locations that contain space-time bins considered temporal outliers for the selected Cube Variable value and the analysis that is run. Time series outliers can be values that significantly differ from the patterns and trends of the other values in the time series or possibly indicate data entry errors. Even a small number of outliers in the time series of a location can reduce the accuracy and reliability of forecasts. Locations with outliers, particularly with outliers toward the beginning or end of the time series, can generate misleading forecasts. These locations can help you determine how confident you should be in the forecasted values at each location.
This option also outputs a bar chart showing the number of temporal outliers for each time step.
Additional resources
- An overview of the Space Time Pattern Mining toolbox provides a summary of the available analysis tools.
- See Visualizing the Space Time Cube for more information to help you get started exploring your space-time cube.
- The Space Time Cube Explorer Add-in helps you interact with and explore your space-time cubes by automatically setting up time and range sliders for you. The Explorer also allows you to display your cubes quickly with many preset layer symbology options in the Display Gallery. The Space Time Cube Explorer Add-in is available at www.esriurl.com/SpaceTimeCubeExplorer.