Time is supported in spatial data in a variety of ways. Time information can be stored as an attribute (feature classes and mosaic datasets), or it can be stored internally (such as in netCDF data). The following sections describe the data that can be visualized through time.
With feature layers, features can be visualized over time in two ways:
- The shape and location of each feature are constant, but attribute values can change over time.
- The shape and location of each feature change over time.
Features that change in shape or location over time need to be stored as separate features. For example, for hurricane tracks that are being visualized over time, the point feature representing the location of a hurricane at a particular time needs to be stored as a separate feature.
Features that do not change in shape or location can also be represented in the table as separate features, for example, for population values per city. Each city can be represented by multiple features. Each feature representing the same city has the same location with a different population value for each date.
However, in cases where you have many time stamps for the same static feature, you can use a one-to-many join where the spatial information is stored in the base table, and the duplicate information is stored in a separate table.
Mosaic datasets can be used to store rasters representing a change over time. For example, a mosaic dataset can contain aerial images representing land-use change over time, which can be visualized over time. As with feature layers, you need a date field in your mosaic dataset's attribute table to indicate the valid time for each raster.
With netCDF layers, you can choose a dimension for visualizing the data. Time values are stored as one dimension of the netCDF layer. For netCDF feature layers, you can specify the layer time using a time dimension or the attribute fields (start time and end time fields) containing the time values. For netCDF raster layers, however, you can only specify layer time using the time dimension that allows you to step through the data over time.