GRASS GIS: Getting Started (NCGIS 2019)

GRASS GIS

Timeline:
- 1983 started at US Army CERL as a land management system
  - William Shatner narrates a promotional video doi.org/10.5446/12963
  - evolved into a general purpose GIS
- 1999 international team of developers + GNU GPL
- 2008 OSGeo Project
- 2018 35 years of GRASS GIS development
- 2019 latest release Jan 21, 2019 (7.6.0)
- check out last change (probably just few hours ago)

Licensing

According to Stallman, there are four freedoms that you should have as a user:
- the freedom to use the software for any purpose,
- the freedom to change the software to suit your needs,
- the freedom to share the software, and
- the freedom to share the changes you make.

Stallman, R. M. (2002). Free software, free society: selected essays. Ed. by J. Gay. 1st. ed. OCLC: 253840339. Boston, Mass: Free Software Foundation. 220 pp. ISBN 978-1-882114-98-6.

Processing Capabilities Overview

raster and 3D raster: map algebra, DEM, interpolation, flow, neighborhood, solar, cost surfaces, ...
imagery (remote sensing): rectification, multispectral, classification, segmentation, PCA, FFT, ...
vector and database: buffers, overlays, networks, topology, attribute management, SQL, ...
spatio-temporal data (time series): sampling, statistics, aggregation, temporal topology, algebra, ...

General Capabilities Overview

visualization: 2D display, 3D visualization, temporal visualization, PNG, PS and PDF maps, ...
interoperability: import from and export to different formats, WMS, JSON, ...
extensibility: scriptable (typically Python or Bash), custom tools (typically Python or C), ...

Software Description

portable:
- many operating systems: MS Windows, macOS, Linux, ...
- 32bit and 64bit versions
- high-end desktops, low-end laptops, Raspberry Pi, High-Performance Computing clusters, ...
single integrated software
written in C and Python

Modules

Functionality is divided into modules (over 500)

Prefix	Functionality	Example
r.	raster processing	r.mapcalc: raster map algebra
v.	vector processing	v.surf.rst: interpolation from points
g.	general management	g.remove: removes maps
d.	display and rendering	d.rast: display raster map

There is more: i., t., r3., m., ps.

User Interfaces

Modules can be accessed using:
- Graphical User Interface (GUI)
- Command Line Interface (CLI)
- Python (API)

GUI

The interface for desktop

GUI for modules

All modules have standardized, unified interface layout.

Standard dialog (form) for the r.neighbors module

Graphical Modeler

Visual programing tool

Command Line

The baseline interface for efficiency and reproducibility
Available in GUI as Console with autocomplete functions

r.fill.dir input=elev output=fill direction=dir

Python

grass package part of GRASS GIS
access to modules, but also to internal C functions
integrated Python editor in GRASS GIS
integrates with Jupyter Notebooks
combine with PyPI grass-session package for external use (pip install grass-session and import grass_session)

import grass.script as gs
gs.run_command('r.fill.dir', input='elev',
               output='fill', direction='dir')

Duality between GUI and commands

Command line (Bash):

Python:

Examples in the documentation and class instructions are usually provided as commands which can be used to fill in the GUI, write Python code, or run them directly.

3rd Party Interfaces

QGIS Processing Plugin
QGIS GRASS Plugin
R rgrass7 package

3rd Party Interfaces

Actinia – The GRASS GIS REST API
- Used in the OpenEO GRASS GIS driver

3rd Party Interfaces

Tangible Landscape
- tangible user interface to GRASS GIS and Blender
- by NC State University, Center for Geospatial Analytics

Modules running: v.surf.rst, r.contour, r.fill.dir, r.sim.water

APIs

native:
- C (usable in C++)
- Python
- command line (Bash, cmd, …)
3rd party:
- R (rgrass7 package)
- Ruby (grassgis gem)
- Java (JGrasstools library)
as backend:
- QGIS Processing (Python and R APIs)
- PyWPS, ZOO-Project (OGC WPS)
- Actinia (REST)
- gvSIG JGrasstools Spatial Toolbox under development

GRASS GIS in the FOSS Ecosystem

GRASS GIS: Geospatial processing, algorithms, and workflows
Inkscape, Gimp, ImageMagic, GMT, QGIS: Graphics and refined the cartography
PostgreSQL with PostGIS: Relational database (spatial and non-spatial data storage)
R, Octave, Python, Jupyter, LibreOffice: Statistics, non-spatial data analysis
OpenLayers, Leaflet, GeoExt, Django, Flask: Web interface
PyWPS, ZOO-Project, Actinia: Processing server
MapServer, GeoServer, QGIS Server: spatial data publishing server

Concepts

locations and mapsets
data management
projections
computational region

Spatial Database

Example of GRASS GIS database structure into locations (based on SRS) and further to mapsets (for organization)

Selecting Mapset

Selecting database, location, and mapset

Import, export, linking

Data need to be imported or linked (GDAL)
Vector and raster data can be linked
- r.external, r.external.out
- v.external, v.external.out
Export to various formats (or linking)
GRASS GIS native exchange format (v.pack, r.pack, v.unpack, …)

Computational Region

extent and resolution for raster computations

Science & Research

Used in research platform and as scientific code repository
Example: Solar radiation model (r.sun)
- available since 1993
- major changes in 1996, 1999, 2002, 2003, 2006, 2008, 2017
  - parallelized version in 7.4
- by Jaroslav Hofierka, Marcel Suri et al. (2004, 2007, ...)