Bonny P. McClain - Python for Geospatial Data Analysis

• 2021-08-25: First Release

With Early Release ebooks, you get books in their earliest formthe authors raw and unedited content as they writeso you can take advantage of these technologies long before the official release of these titles.

This will be the 1st chapter of the final book.

If you have comments about how we might improve the content and/or examples in this book, or if you notice missing material within this chapter, please reach out to the author at bonny@dataanddonuts.org.

Are you a geographer, geologist, or computer scientist? Impressive if you answered yes, but I am a spatial data analyst. In a nutshell, I am interested in exploring data and integrating location. Now that access to location data and geospatial datasets is fairly ubiquitous, most of us are becoming data curious regardless of professional title or area of study. Appreciating the where in our analyses introduces a new dimension of comprehending the impact of a wider variety of features on a particular observation or outcome. I spend a lot of professional time examining public health data and large open source datasets in healthcare. Once you become familiar with geocoding and spatial files, not only can you curate insights across multiple domains but you will also be able to target areas where profound gaps exist.

This is also the age of citizen scientists. The accessibility of open source tools and massive open online courses (MOOCs) empowers a broader range of not only professionals but also the data curious. Perhaps you have a hobby or interest in a certain species of bird and would like to access spatial data to learn about their habitats. Where are they nesting? Where are they traveling from and to? Which habitats support the most species and how is this changing over time? You might be able to create maps of your sightings or other variables of interest. Well explore this idea a little later when we generate a data question to explore.

Meanwhile, if you are interested in a citizen scientist project look no further than the New York Public Library (NYPL) Map Warper project. As you might imagine, NYPL has a vast amount of historical maps. The challenge is to correct the errors in outdated survey technology by searching for modern matching ground control points (GCPs) and warp the image accordingly. This is known as map rectification.

The ) is a tool for rectifying (digitally aligning) historical maps from the NYPLs collections to match todays precise maps. Visitors can browse already rectified maps or assist the NYPL by aligning a map. Everyone is welcome to participate!

Figure 1-1. Map of Manhattan from 1870

I have used rectified maps to explore development over time in different cities or to accurately reimagine a historical location. How do investments in infrastructure and industrial development impact neighborhoods over time, for example? shows a rectified map.

Figure 1-2. Rectified contemporary Manhattan map

There are many opportunities for professionals across multiple industries to include location intelligence in their analytics. Location intelligence describes actionable information derived from exploring geospatial relationships when formulating data questions and evaluating hypotheses. Open source tools are welcoming new end-users, and what we need is a lexicon applicable to a myriad of diverse interests, resources, and learning pedigrees. Yes, enterprise solutions are powerful, but many have limited access to subscription-based applications and tools. There are a variety of options in geographic information systems (GIS) software with pros and cons associated with all of them. I will mention a few when they come up, but although I have access to both ArcGIS also known as Aeronautical Reconnaissance Coverage Geographic Information System and QGIS, Quantum GIS I like to give QGIS the main stage. It is truly open source, meaning that you dont need different levels of licensing for access to all of the available tools. Since this book is intended for a wide level of interests, I want you to be able to explore all of the tools. In my professional work, yes I will move between both ArcGIS and QGIS -- mainly for access to the abundant catalog of GIS datasets in ArcGIS online. I did have to learn the hard way when geocoding in ArcGIS. ArcGIS uses a credit system, and it was easy for me to unknowingly get on the wrong end of it. I didnt realize that the paid service kicked in fairly automatically when I uploaded a CSV file with location data. QGIS, on the other hand, offers two options we will explore later -- both free.

The polygons on the map in are different colors corresponding to racial categories. What might we determine about these clusters if we also examine other attributes? Lets sit with these questions for a while. I will give you a hint: the devil is in the details, or should I say, is in the layers. The polygons are shaded based on areas where a certain race is the majority population. This is a good place to introduce the expression defaults should not be the endpoints. A deeper understanding of GIS will allow you to move away from default settings and create unique and deeper insights.