GIS Concepts
Geodesy and Cartography
Background material
- An introduction to Geodesy from NOAA
- Wikipedia's Geodesy entry
- Wikipedia's GIS entry
- Bowditch's American Practical Navigator - (especially chapter 2)
- NGA Geodesy and Geophysics publications
- UK Ordnance Survey primer on coordinate system concepts (PDF)
- Clifford J Mugnier's Grids and Datums column from PE&RS containing detailed descriptions of many national projections & datums
Map projections
- MapRef - The Collection of Map Projections and Reference Systems for Europe
- Projections Transform Lists (PROJ4)
- UTM Zones
EPSG:
Projection galleries:
- Map projection concepts by Carlos Furuti
- Map projection gallery by Paul Anderson (old link)
Map datums
An extra calculation is needed when re-projecting maps and data between two different co-ordinate systems (in addition to the re-projection) if the two co-ordinate systems are based on different models of the curvature of the earth. E.g. OSGB36 uses the Airy ellipsoid and WGS84 uses the WGS84 ellipsoid, which have slightly different sizes and shapes. The error is not large - generally a few hundred metres at most on the ground. The datum transformation parameters describe this adjustment mathematically.
As the transformation between any two datums is approximate and varies by location, different sets of parameters are often offered to give improved accuracy in different regions of a country. In general there is no one "correct" set of transformation parameters - indeed the accuracy changes over time due to tectonic movements in the Earth.
- A brief history of map datums for the layman
- An introduction to geodetic datums by Peter Dana
- How WGS 84 defines the Earth
- A discussion of 3-term, 7-term, and NTv2 grid datum transformations by Land Information New Zealand
- (besides the web page have a look at the PDF fact sheet and guide linked therein)
How GRASS deals with geodetics
As far as GRASS is concerned, an ellipsoid and a spheroid are the same thing, and ellipsoid is the prefered name.
As far as GRASS is concerned, a datum is made up of an ellipsoid and an origin. Setting a datum is optional, but highly recommended.
GRASS uses a modified version of the PROJ.4 library.
Modules controling a location's map projection
Modules for reprojecting GIS maps and data
- r.proj for reprojecting raster maps
- v.proj for reprojecting vector maps
- m.proj for reprojecting a list of coordinate pairs
Modules for georectifying images
- r.region for resetting a raster map's bounds information
- gis.m GIS manager GeoReferencing tool (File menu)
- i.points and i.vpoints for setting GCPs
- i.rectify for georectifying imagery
- GDALwarp (use with gdal_translate, see the i.warp script in the wiki AddOns page)
GIS Data types
Raster Data
Data which occurs in a regularly spaced grid. e.g. a satellite image or digital terrain map.
Region settings determine the spatial extent and resolution of the grid.
3D Raster Data (Voxel)
A stack of 2D raster maps.
Vector Data
Data which occurs as a series of coordinates. e.g. a GPS position or coastline map. May be a point, line, area, etc in either 2D or 3D space. Generally independent of region settings.
Imagery Data
Pixelated photographic or satellite images, often imported from a GeoTIFF or PNG image file.
As far as the GIS is concerned this is just another raster map, but there are several modules specially tailored for rectification and processing common imagery types. e.g. ortho-photos or multi-channel LANDSAT data.
- The GRASS Image processing wiki page
Site Data
Old versions of GRASS (5 and earlier) treated point data separate to line and polygon data. GRASS 6 classes all vector data features the same. Convert old sites file data into GRASS 6 vector format with the GRASS 6 v.in.sites or v.in.sites.all modules.
Conversions between data types
The following table is intended to catalog transformations from one type of data to another:
[table is currently incomplete!]
| From / To | Raster | 3D Raster | Vector |
|---|---|---|---|
| Raster | r.mapcalc | r.to.rast3 | r.to.vect, v.sample, r.volume |
| 3D Raster | r3.to.rast, r3.cross.rast | r3.mapcalc | |
| Vector | v.to.rast, v.surf.rst, v.surf.idw | v.vol.rst, v.vol.idw, v.to.rast3 | v.clean |
| Data | r.in.* | r3.in.* | v.in.* |
How a GRASS project is organized
GRASS data is stored in a three level structure, the database, location and mapset. These can be found in a series of nested directories on the user's computer. All three must exist and are set at GRASS startup time.
The Database
The directory in which all GIS data is to be stored.
e.g. ~/grassdata/
The Location
A location is a GRASS project consisting of an area, projection definition (or unprojected), a grouping of mapsets, all with the same projection settings. A location is a subdirectory of the GRASS database.
e.g. world_lat_lon, utm_zone_59, or west_coast
A location contains one or many mapsets.
The Mapset
A mapset contains map(s), it is a subdirectory of a location. Conceptually, if several mapsets are used in a location, they may be assigned to different users (each has one or several own mapsets to work in and cannot modify thos of other users), and/or it they are used to organize a project (location) by subareas or subprojects. There are no specific organizational limitations.
There is always a PERMANENT mapset which is readable from all other mapsets within the same location. Read access to maps in other mapsets is managed with the 'g.mapsets' command or by adding the "@" symbol and mapset name (e.g. map@othermapset).
Raster GIS Analysis
Simple Raster Math
Sometimes when analyzing the relationship between two or more raster data sets, a relatively simple mathematical approach is best. One example using the r.mapcalc tool would be to look at changes between two raster data sets. By subtracting the values in these two data sets you can assume that resulting cells with a positive value have a positive change and those with a negative value have negative change. If the cell values have a zero value then there would be no change.
GRASS comes bundled with the r.mapcalc command line tool as well as a GUI interface for the tool accessible using the r.mapcalculator command. This GUI allows the user to easily assign raster maps to the variables used in the formulas and easily create mathematical strings that will result in a new raster data set containing the results.