Essential 3D Analyst vocabulary

Below are some of the most common terms you will encounter when using 3D Analyst, specifically ArcGlobe and ArcScene, for your 3D visualization and analysis tasks. Some definitions are related to other documents with more thorough descriptions.



Base heights

A feature's base height is the elevation where it or its vertices are displayed in 3D space. Base heights can come from within the feature geometry (for z-aware features), an attribute value or expression, or a referenced elevation data source. Combinations are also supported, such as features that have a height relative to a surface. Base heights are also commonly referred to as "elevation values."

Functional surface

A continuous, 2.5D surface representation, where all locations on the surface have only one elevation value, or z-value, per x,y coordinate. Functional surfaces are most commonly used to model terrestrial data representing the earth's surface, though they can also be used to model many other types of surfaces such as bathymetric data, individual geologic strata, or statistical surfaces describing geographic concentrations. Terrain datasets, TINs, and raster DEMs are all examples of functional surfaces.


Textures, sometimes referred to as "facades" or "materials," are the images that are pasted onto the sides of 3D models, such as 3D buildings. Textures are supported for multipatch features stored in the geodatabase only. Note that the source texture imagery is saved in the Shape column, together with the multipatch's geometry definition.


A z-aware object knows about its elevation, or stored z-values. Features that are z-aware store their z-values inside their geometry in the geodatabase (or shapefile), while feature classes or feature datasets that are z-aware store the units and datum those z-values represent. Using the Catalog window, you can create new z-aware feature classes, as well as load data stored in a table or spreadsheet, to create z-aware feature classes.

Level of detail (LOD)

Refers to the complexity of an object represented in a 2D or 3D view. Commonly in computer graphics, a feature's level of detail is reduced as it gets farther away from the viewer or by some other formula based on object importance, eye-space speed, or position. Reductions in LOD may involve generalizing textures applied to features and/or simplification of their geometry. Reducing feature complexity and detail improves rendering performance by decreasing the workload on the graphics pipeline. Generally, the reduction in detail does not noticeably reduce the visual quality of the view because the feature is in the distance or is moving too quickly through the view.


A method applied to 2D features to generate a 3D object when actual 3D features are not available. Extrusion is a layer property unique to ArcGlobe and ArcScene, where points, lines, and polygons are stretched vertically into lines, walls, and boxes, respectively. The extrusion method is proportional so there is no distortion. It can be applied for many purposes such as generating realism in a 3D view or enhancing statistical/attribute information such as population. Extrusion can also be applied negatively. The only requirement to apply extrusion to geometry features is that an elevation surface must be set to establish the feature's base height. After this, the extrusion can be applied from the known surface location, either by a constant value or calculated expression, using one of four extrusion methods.


Draping is one example of how a layer can exist in 3D space by defining its role with respect to other layers. A draped layer uses other layers as its source for elevation, so it drapes itself over the other layer features, texture (if any), and terrain details. Layers assigned to the Draped category in ArcGlobe can further be organized with a top-down approach as to which will display on top of the next. Since ArcScene does not distinguish layer categories, to configure a draped behavior, use the layer properties instead of the table of contents. For example, a point layer can reference a raster surface for its base height information, which is the same result as creating a draped layer in ArcGlobe.


Like draping, floating is another way to distinguish how a layer exists in its 3D space. In this case, floating is used to display layers that are not intended to be placed on the elevation surface such as rasters, underground or above ground utilities, aircraft, and atmospheric conditions (clouds). Floating layers usually define their height source separately from other layers in the 3D view. Draped layers usually share the same surface data as other layers.


Rasterization in 3D is a different concept than 2D. ArcGlobe has a technique that allows vector data to be rendered (displayed) as rasterized. As a result, the rasterized 3D layer looks as it would in ArcMap (a flat image) and draped on the globe surface. This is useful for many reasons including faster display, ease of consumption of large vector data sources, and so that polygons can match their interiors to the terrain by being draped on the globe surface.

See Rasterizing features for 3D for more information.

3D model

A 3D model refers to the representation of any three-dimensional object, stored as a digital collection of features and/or rules, that can be displayed as a two-dimensional image through rendering. The object may be stored as a wireframe (composed of different geometric entities such as lines, triangles, and curved surfaces) defining the shell of an object or as a solid (composed of parametrically or explicitly defined 3D objects that are added to or subtracted from each other to form a more complex object). Three-dimensional models are used extensively in the world of computer graphics, such as in motion pictures, animation, medical visualization/simulation, architecture, engineering, industrial design, aeronautics, computer gaming, chemical engineering, and planning. These 3D models can be constructed manually through a variety of software packages such as SketchUp, 3D Studio Max, and Revit. In addition, 3D models can be generated from a variety of other sources such as 3D scanning (lidar, sonar, and so forth) or derived from analysis procedures. COLLADA is an example of a 3D modeling format commonly in use today. In ArcGIS, 3D models may be used to represent points in 3D space or stored as features in a multipatch feature class.


A multipatch is a type of geometry in ArcGIS designed to represent the shell of a 3D object. Multipatches are composed from a series of patches that store geometry, color, transparency, and texture information. The geometry the patch stores may be a ring, triangle, triangle strip, or triangle fan. The shell these patches form might represent a completely enclosed feature, such as a sphere, or an open feature, such as a sloped roof. Multipatches may be used as 3D symbols to represent points or stored in a feature class with attributes. Multipatches are commonly used to store geotypical 3D symbols such as trees, street lamps, and park benches, as well as geospecific features such as buildings, bridges, and subsurface geologic formations.


Caching is a mechanism for managing large amounts of data in ArcGlobe. It improves display performance and may reduce data load times by storing prerendered information. ArcGlobe stores the cache in two places, in memory and on disk. Memory caching is the amount of RAM you can assign to each data type in use. Disk caching saves tiles rendered by ArcGlobe as you navigate the 3D view. It can happen automatically (on demand) or by forcing ArcGlobe to generate a full cache (tiles for an entire dataset) or partial cache (tiles for a specific LOD). Navigating to a location with preexisting cached tiles improves display performance, as it simply swaps out the tiles you have in memory for the tiles on disk. You can view your application cache settings on the Display Cache tab of the ArcGlobe Options dialog box. As well, you can set cache properties for layers on the Cache tab of the Layer Properties dialog box.

Related Topics