Create a Raster using Triangulation

On the Raster ribbon tab, in the Interpolation group, expand Create Raster.
Select the interpolation method from:
- Triangulation
The Create Raster) panel displays.
Select the Input point table.
Select the Fields to interpolate.
Set the Output raster, where the raster is saved.
Select the Color ramp for the raster.
If there are Breaklines select the Source file and the Attribute field to load the breaklines from.
Note: Breaklines are defined as multi-segment lines in which the slope is monotonically increasing or decreasing along each segment. Examples of breaklines include stream or river traces, cliff faces, dredged channels, topographic ridge lines or roadways. By incorporating break lines into the gridding process the output grid is forced to conform specific slope requirements in critical areas.
Expand Triangulation.
1. Select whether to check Distance specified in data units, This parameter is used to control the units of distance that the maximum triangle side length property is measured in. By default this control is disabled and the maximum distance unit is expressed as a ratio of the tile (or patch) size. If you wish to constrain the Maximum triangle side length to a fixed value that is measured in absolute data units (e.g. 100 m) then enable this control and enter in the appropriate value. If the entered value is large and exceeds the size of an individual tile of data then it may have no effect on the output grid.
  Note: If the coordinate system of the input data is Longitude/Latitude then the absolute distance units need to be specified in fractions of a degree (Arc seconds).
2. Set the Maximum Triangle size, this parameter applies to the triangulation phase of gridding and is used to minimize or eliminate long thin triangles that may be created across large holes or gaps in the data or between widely separated points that lie around the perimeter of the dataset. By default triangles that are created with a length that is greater than half the diagonal length of a tile is discarded. The size of the tiles used during triangulation are determined automatically by the software after it analyses the spatial statistics of the input points, however you can modify the size of the tiles by applying a triangle patch multiplier.
3. Set the Triangle patch multiplier, the triangle patch multiplier is used to modify the number and size of tiles (or patches) that the software segments the input data into before sequentially triangulating it. The tile size is automatically determined by the software following a detailed analysis of the spatial statistics of the input data. Under certain circumstances the patch size can be modified by applying a triangle patch multiplier. Increasing the default value of 1 to a higher number creates larger patches and may assist in the infilling of large holes or gaps in the dataset, however it also increases peak memory usage during the gridding phase. For very large datasets increasing the Triangle patch multiplier reduces the number (but not the storage requirements) of temporary files that are created during the gridding phase. For most datasets a patch multiplier of 1 or 2 is sufficient. Increasing the patch multiplier beyond a value of 4 would be rarely necessary.
Expand Geometry and set Cell size.
Note: Cell size is the size (or resolution) of each cell in the output grid file. The size is measured in the spatial units of the output grid coordinate system.
Expand Advanced.
- If Define maximum memory usage is checked, this control is used to constrain the amount of physical memory (RAM) that the software attempts to use during the gridding process. The default control setting is unchecked and in this state the software attempts to use up to 80% of available physical memory (to a maximum of 2 GB) during gridding. While it is possible to constrain the amount of RAM that is available to the software during the gridding process; doing so reduces the gridding performance significantly if the software has to repeatedly page tiles of data between disk and memory in order to perform the gridding operation. To achieve best performance on large datasets it is advisable that you close all running applications and free up as much physical memory (RAM) as possible before commencing gridding. For very large datasets (>50 million points) it is recommended that the software is run on a machine with between 2 and 4 GB of RAM.
- To Scan data extents controls the resolution at which the software initially scans the input data files to establish the preliminary spatial statistics during the first phases of gridding. The default behaviour is to scan all lines of the input data. It is possible to speed up the initial scan of the input files by adjusting the scan data extents control to one of the following settings:
  - Complete—Scans every line of each input data file.
  - Fine—Scans approximately 12% (1-in-8) of the lines from each input data file.
  - Course—Scans approximately 3% (1-in-32) of the lines from each input data file.
  - Overview—Scans approximately 0.75% (1-in-128) of the lines from each input data file.
  - Bounds—Acquires the data extents from information stored in the files if available or performs an overview scan if unavailable.
    Note: For datasets that have a relatively even spatial distribution of input points setting the scan data extents control to Overview provides the best compromise between speed and a representative statistical sample.
- Select the Temp Folder, this folder is used to temporarily store the spatially sorted input data tiles which are used during the gridding process. If all of the input data can fit into system memory then no temporary files are created and the entire process occurs in RAM. If the input dataset is very large (>10 million points) then it is necessary to store a copy of the input data on disk during the gridding process. By default the temporary directory is set to the Windows system temporary directory; however it may be necessary to map it to an alternative storage location depending on the size of the input dataset. Note You should always ensure that you have at least the same amount of free temporary storage space as the total size of the input dataset. A good rule of thumb is to set the temp path to a location that has 2x as much storage space as the size of the input dataset.
- Select the Output Raster data type, the data type control is used to set the numeric storage type for the interpolated values in the output grid. It is advisable to select the appropriate data type that most efficiently represents the range of data that is stored in the output grid. For example, a signed 2-byte integer is generally suitable for storing typical elevation data at 1m vertical resolution. The data types available are:
  - Automatic
  - Unsigned byte
  - Signed byte
  - Unsigned short
  - Signed short
  - Unsigned int
  - Signed int
  - Float
  Note: The automatic option sets the output format to an appropriate data type based on an analysis of the input data range.
Select whether to Clip output raster.
Click Run.
The grid should display in the Map window.