While raster data can be stored in many different formats, GeoTIFF is one of the most common. GeoTIFF is a standard .tif image that also contains spatial data (e.g., coordinates and projections) and attributes. In general, however, any standard image (e.g., jpeg), if given proper spatial data, a practice called georeferencing, can act like a raster file.
Vector data includes data in the form of points, lines, and polygons along with any connected attributes. This type of data can come in several different formats, such as GeoJSON, CSV, and KML. Below you can see an array of data formats (found in Boston Open Data), GeoJason, CSV, KML, Shapefile, ESRI Rest API, and ArcGIS Hub Dataset. All contain the same food truck information:
While most platforms allow for some interoperability (the ability to work with several file formats), it is helpful to understand the basic distinction between them.
A CSV (comma-separated value) file is one of the most well-known formats for both spatial and non-spatial data and is created by using commas to separate each value of an attribute, with each line of the file equal to a single data record. If we return to thinking about structured data sets, comma-separated value file simply puts a comma in between attribute value, returning an output that looks something like this:
CSV data is useful simply because it is easy to create and manipulate in software like Excel or GoogleSheets, and these programs often allow you to export your data as a CSV to make it more interoperable. It is also easy to transform a CSV file into other formats, like GeoJSON, making it an easy "go-to" for sharing spatial data
CSV files are especially useful if you simply have a collection of point data. A file of this kind, sometimes called "XY data" or "long/lat data" can be easily imported into software like ArcGIS or QGIS in order to visualize it. (Note that it will not have any attributes).
GeoJSON is an open standard format for spatial data based on the JSON file format, which includes spatial information (points, lines, polygons) along with other qualitative and quantitative attributes. If you are using Python, a coding language, or platforms like Leaflet to manipulating spatial data, a GeoJSON export is probably your best choice.
The example below is a selection of GeoJSON data. It is organized similarly to the well-known CSV, with each feature containing a series of attributes, including (in this case) the geometry for point data, latitude, and longitude values for where the food truck will be located.
It's worth noting that there are tools for converting CSV to GeoJSON and vice versa.
A KML file (Keyhole-Markup-Language) is an open standard spatial data format most commonly associated with displaying geographic data in an Earth browser such as Google Earth. Although not necessarily as easily compatible with other platforms such as ArcGIS or Leaflet, there are a variety of conversion tools and plugins built-in within these platforms that may help ease the process.
A Shapefile is a simple format for storing the geometric location and attribute information of geographic features represented by points, lines, or polygons (areas), and is most commonly associated with ESRI's ArcGIS program suite. From within ArcGIS/QGIS, you may then export it as other file types, though this sometimes requires purchasing the Interoperability extension.
The following examples of vector and raster data demonstrate some of the many forms the data can take.
Vector data in mapping generally appears either as points, lines, or polygons. In this example of polygons as raster data, spatial information tied to the data defines the shape's boundaries while further information (such as a title and image) is included as additional attributes. Attributes can include any type of additional information that is useful for a viewer to know about a particular location appearing in your dataset.
A basemap is a georeferenced raster image. They help give a larger context to vector data sets, which would otherwise simply appear as points, lines, or polygons on a blank space. Satellite imagery and other top-down images (orthophotos) are the most common raster data of this kind, seen, for example, when using satellite imagery from GoogleMaps as seen below.
Raster surface maps contain attributes that mark change over a particular landscape instead of representing the visual world as basemaps do. Common attributes indicated on surface maps include elevation (which is specifically termed a digital elevation map, or DEM), rainfall, or temperature. Once these types of surface maps are imported into a GIS platform like ArcGIS or QGIS, they can be analyzed in a myriad of ways.
Raster thematic maps are similar to surface maps as they use attributes of a particular landscape, be they physical or cultural. Combining several types of data to create what are called thematic data sets, thematic maps group together certain specified attributes from vector or raster data into specific categories, and then mapping out the categories.
The example below shows vegetation types from a dataset that breaks land-cover types into categories. The data is multispectral, meaning it is the acquisition of reflected wavelength data in the visible, near infrared, and short-wave infrared spectrums.
Spatial datasets, in general, come in two distinct forms, vector data and raster (or pixel data). Raster and vector data can come together in the creation of a wide variety of mapping projects, from a traditional figure with an explanatory legend and caption, such as might appear in an academic text, to an online interactive platform that allows for the searching or filtering of thousands of pieces of spatial data or hundreds of historical maps.
Vector data includes points, lines, or polygons (shapes made up of straight lines) containing spatial information that represent some sort of feature or event in a physical or imagined landscape and may contain other types of qualitative or quantitative information, called attributes. A point may represent a tree, a city, or a moment in time. Lines might indicate the street grid of a town, the path someone traveled across the world, or a social link between two communities. Polygons can mark the boundaries of a country or voting district, the catchment area of a river, or a single city block.
Raster consists of "cells" of data covering a specific area (its extent), with attribute values in each cell representing a particular characteristic. It may still consist of points, lines, and polygons, but these shapes are themselves composed of pixels (the way a jpeg or other image file type is).
Data of this type may take many forms, such as satellite imagery containing vegetation or elevation data, precipitation maps, or even an historical map, which has been given a spatial reference. Unlike vector data, raster data has a particular resolution, meaning each pixel represents a particular geographic region of a specific size.
