GIS Basic Principles What is GIS Geographical Information

GIS Basic Principles

What is GIS? � Geographical Information Systems…GI Science (NB rebrand taking hold: Spatial Data Science) � Conceptualising reality in a computer model � Not just maps or manipulation � Why is GI important? � � � Everything happens somewhere Ability to attach multitude of information Relationship between phenomena spatial co-incidence relate/transfer info (attributes) between layers where co-incide

Functionality � Data acquisition/integration � Data management/database management � ‘electronic � Data analysis � Decision � filing cabinet’ making Visualisation/cartography

Data Types I: Vector � � Discrete entities with specific location Multiple attributes for each feature Point Longley et al. , 2005: 77. Polyline Polygon

Data Types II: Raster � Continuous surface with changing values � Elevation and derivatives � Satellite imagery � Photographs � Scanned maps � Attributes � Each cell has single value � This can relate to Value Attributes

Examples: Vector � Mapping schools, hospitals, retail outlets, etc. � Points � Multiple attributes can be recorded Type of facility Number of users, staff, etc. Turnover, vol. of sales, success rates, infection rate, etc. Date built, condition of buildings Record number � And of course implicitly – spatial reference/location Permits spatial analysis, incl. of all the attributes above Point-pattern analysis

Examples: Raster � Display density of alcohol outlets in Scotland � Grid (of usually square cells) � Each grid cell can show value relating to density of outlets per unit square � Aggregate (zonal) measures of one variable per grid

GIS – Data Capture � In the field – data capture and recording; primary data � � Vector Fieldwalking Point locations �Incidents �Street Furniture Line Surveys �Transects �Elevation Profiles Polygons �Land Parcel Capture �Exclusion Zones �Conservation Areas � � Contour surveys Full Plans www. english-heritage. org. uk � In the lab – (automated) digitising, geo’l text analysis

GIS – Data Sources � Maps � Modern Can be vector or raster OS (Open. Data, Master. Map) Enviro, e. g. CEH, BGS � Historical Old OS BGS? Other Historic Maps/Plans � Topography � Geophysical imagery � Remote sensing � Optical/RADAR/Li. DAR Caracol Li. DAR (www. britannica. com)

Databasing � Asset, Data, Facilities, Resource management � Land Property Gazetteers (National/Local) � Large Environmental/Scientific Datasets e. g. Climate, Flood risk, Land Capability, Forestry � More flexible storage and querying � Topology preserved � Sophisticated querying/spatial testing � Handling of large data volumes (Big Data? ) � Attaching detailed attribute data to spatial objects � Analysis of sites/processes within broader context

Databases – Public Access � Public dissemination � � Search databases online Examples Data. gov. uk – goldmine! � Met Office Rainfall, DTI/BERR Windspeeds, BGS Geology Data � ONS � OS – GPS network, Open. Data, Open. Space API � � Or restricted to Academia e. g. UK (Census) Data Service � EDINA Digimap (OS, CEH), MIMAS Landmap � CEH/NERC datasets � � Lack of standardisation. Not all may qualify as DBMS.

Data Extraction – APIs/ Screenscraping � Many services publicly accessible but limited � � � In number of points that can be retrieved/processed In number of tasks which can be achieved, thus… APIs/Screenscraping: Automation of data retrieval across web Using a mashup (our custom web page code) we can 'run' a target web service with different inputs and extract desired information from the returned web page � E. g. Twitter, Foursquare, Facebook…. "Big Data" � See: Twitter. Map, Tweet. Map, etc. � But also gov, enviro, health – see Open. Census proj in R � � Smart Cities/City Science/Sensor Networks VGI/Crowd-Sourcing

The one million tweet map

Trendsmap – mapping the #hashtags

Visualising Movement/Flow

Interesting Maps on the Web � These just appeared as I was preparing today’s materials: � http: //matadornetwork. com/life/57 -worlds-interesting- maps/ � And, we can have interactivity � http: //earth. nullschool. net/ � Or, even a 'full' GIS in a browser…

Distributed GIS – GIS online BGS Geology of Britain viewer

GIS in your Browser � Google Maps (and Google Earth) � Basic Functionality Online � Full Power via Java. Script powered web pages � Over-reliance on commercial megabusiness? � Open. Layers � Free � and Open alternative (NB Open <> Free) Other web map tools: � E. g. Map. Server, Geo. Server

Full GIS Software – ESRI’s Arc. Map � � � Market leader in GIS software One of several ESRI Arc. GIS packages – Arc. Scene, Arc. Catalog… Can handle most data types At a basic level, is helpful for cartography and visualisation Many forms of analysis available � � � However – � � Closed, proprietary software; (tho Arc. GIS Online to compete with Google) Not the only program available, and expensive (tho some other free components) Alternatives: Map. Info, Geo. Media, FME, Cad. Corp Open Source GIS � � Multi-criteria analysis (prediction modelling) Visibility Cost surfaces Networks… Much open source software available; modifiable, extensible, fixable! GRASS, Quantum GIS, gv. SIG – many make use of GDAL/OGR libraries http: //opensourcegis. org/ Or e. g. R or Python – stats, programming – both oft used for maps

Büyük Bedesten http: //www. shc. ed. ac. uk/projects/longwalls/Method ology/Visualisation. htm Casal de Freiria Rua and Altivo, 2011: 3302 USGS (Sept 2011) – Li. DAR particularly good for trees (top and bot = first and last)

(Now) We’ll Have Manhattan… U. S. Army JPSD/NOAA / www. britannica. com – a decade earlier (2001)

3 D (2. 5 D) fly-through animations � Previous examples may require GIS data/model be fed to 3 D modelling software for detailed work (or that model be constructed in dedicated CAD/3 D software) � We can however very quickly create effective visualisations of 2. 5 D landscapes in e. g. Arc. Scene (and we can import 3 D models from CAD/3 D software) � We can also very quickly render animated visualisations of landscapes in GIS software � One example! (NB may not be made with Arc!): � http: //www. satimagingcorp. com/gallery/quicktime-northkorea. html

Think! � We must be cautious when using computers � Demonstrate how models are developed, and provide information indicating our depth/lack of knowledge � Full publication (data, methods, limitations) � ‘Knowledge representation’-what we think we know � Don’t be blinded by their ‘scientific aura’ � Don’t use them for the sake of it: question them � Technological determinism � Don’t tacitly accept their results…

Questions to ask before starting GIS analysis � � � What am I trying to get the GIS to do? How does this relate to my aim? What data are available and will more need to be created? What about data quality? Completeness? What is the state of scientific/area knowledge? Therefore, how appropriate are the data? What is the most appropriate scale at which to work? And what scale/resolution are the data?

Limitations of GIS � Data quality Different bodies hold data; different standards � Currency, Completeness � � � Representation of reality; and not 3 D but 2. 5 D Technological determinism Packages can only perform certain operations � Algorithms themselves can be restricting � � Experiential/Subjective difficult to analyse… GIS/RS works fast over large areas, but final decisions made at local scale/scale of day-day human experience Big Data Volumes may allow behavioural analysis, modelling of supply and demand, etc. but temporal GIS still limited (as is true 3 D, 4 D, 5 D, etc…)

Questions? � If not, then…

Practicals!