Hands-on Introduction to Spatial Data Analysis in Python
In this tutorial, attendees will learn how they can geoprocess and analyze spatial data using Python and how it compares to other available options such as desktop GIS options (ArcMap or QGIS) or R. The hands-on tutorials will explore two interesting Python projects (PySAL and Rasterio), and give attendees the head-start needed to move forward for independent exploration and learning of more advanced geoprocessing skills using Python.
Presentation: Review of Spatial Data Options (15 mins)
Data formats (vector – point/line/polygon |raster –continuous/discrete surfaces)
Data types (vector: shapefile, database geometry, HDF5, tables| raster: GeoTIFF, Image, JPEG, MrSID, HDF5 | syntax-based: KML, GeoJSON)
Intro to Spatial Autocorrelation
Hands-on Tutorial #1: Advanced Vector Data Analysis with PySAL (75 mins)
ex: Spatial Autocorrelation: calculate global and local Moran's I to quantify spatial autocorrelation in crime data across 78 counties
Break (15 mins)
Hands-on Tutorial #2: Introduction to Raster Geoprocessing with Rasterio (60 mins)
ex: Raster Calculation: calculate a vegetation index of greeness using two bands of a satellite image (Red and Near Infrared)
Presentation: Next Steps (15 mins)
Quick recap of what was completed/learned
Comparison of Python options to other geoprocessing options (ex: R, open source GIS Desktop software like QGIS, non-open source GIS Desktop software like ArcGIS)
Quick introduction to open source web mapping options to publish spatial data (including Python-based GeoDjango, but others that are more plug-and-play like CartoDB)
NOTE: please see installation guide in the attachments. This is particularly important for Windows users, which may need more guidance to get the necessary Python components installed.