R is an excellent computer programming language that has lots of powerful tools for working with spatial data. This means data sets that contain information about locations, shapes, and distances.

Spatial data can be in 2D or 3D and you do not need special software to use it. Almost any device (phone, laptop, desktop) will work just as well!

There are many ways to create programs using this toolset in R. You may choose to use one of the built-in functions, or write your function to solve a new problem. Either way, there are plenty of resources available to help you develop your skills more quickly.

This article will go over some basics of how to use spatial data in R. There will also be an example of how to read a set of coordinates into R and plot them using ggplot2, a popular graphing package in R.

Why use R?

This article will talk about why you should consider using R for your spatial data analysis needs. Starting with some basic definitions, we’ll discuss how easy it is to get up and running in this powerful language tool, before diving into some applications of Spatio-temporal analytics that make it particularly useful.

A quick note – while there are many ways to do business analytics, there are only a few good tools that have solid support beyond just their software platform. As such, we’ll be focusing our energy here on those tools that are actively supported and extended by other areas within the company (for example, QGIS or Python libraries used for processing).

Installation of R

First, you will need to install the software that allows your computer to work with spatial data. This is typically referred to as a GIS (geographic information systems) or map-making software package. Some examples are ArcGis, Qgis, Google Earth, ESRI’s mapping program called ArcMap, etc.

There are many free alternatives to this software, but they may not have all of the features needed for complex projects. If you are looking at very simple apps to create maps then look into one of these free programs!

Many universities use either Arcmap or Qgis so students can access them free of cost. Check out what each one offers and determine which one seems more intuitive to you. Also, make sure it is compatible with your OS- Windows, macOS, and Linux all offer different versions of operating systems.

Once you have found your toolbox be sure to take time to learn how to use it properly. There are plenty of tutorials and videos online and in person at academic libraries and research institutes.

Understanding R

There are many ways to create spatial data in R. This article will focus exclusively on rgdal, which is an open-source library that can be installed via conda or through your software’s package manager. It is important to remember that not every function in rgdal requires special input! Some functions such as ggplot2 do not require latitude and longitude information, but rather area-weighted averages or categorical values.

rgdal was forked from proj4R, which has been deprecated due to its expensive price tag. Luckily, rgdal is free and actively maintained by the community!

There are several reasons why using rgdal is a good idea over other methods of creating spatial datasets. First, it is very easy to use. Many people who start working with rgdal quickly become familiar with it and feel comfortable using it. Second, some functions within rgdal are more efficient than those in proj4R or similar paid libraries. For example, the up transform() method is much faster because it uses simple matrix multiplication instead of coordinate transformations.

This article will go into detail about how to install rgdal, what each part of the code means, and some examples using real-world data. To learn more about rgdal, check out our dedicated page here: http://www.mghlabs.com/spatstatisticswithr/.

Basic concepts in R

The second major category of software that can be used to work with spatial data is called an “interactive graphics program” or, more commonly these days, a “spatial visualization tool.” Unlike GIS software like ArcGIS, which is designed for professionals who have significant experience using it, interactive tools are typically simpler and easier to use because they assume you know how to use them.

There are several great free and low-cost interactive software packages available including Qgis (for Linux and Windows), Google Earth Pro (for Android and iOS) and Grasshopper (for macOS). All of these programs offer similar features, making it easy to find one that fits your needs.

Some things to look out for when choosing an interactive software package include if they are open source if there are lots of reviews and if their community is active. There are many ways to achieve the same goal so make sure to do some research!

This article will focus on QGIS since it is a relatively new software option but those familiar with other interactive software may still learn something from reading it.

Variables and data types

The second important concept to understand about spatial datasets is how to define your variables. A variable in this case is like a piece of furniture you have at home, for example, a living room couch, dining table, or bed.

In spatial analytics, two main types of variables get defined – feature (or object) variables and attribute (property) variables. Feature variables are properties of objects in the dataset, such as highways, rivers, or mountains. Attribute variables describe the content of those features, such as the length of a river or the elevation of a mountain.

So, if we were analyzing the shape of countries, our feature variable would be defining what country we want to analyze, while our attribute variables would include things such as total area, perimeter, etc. For our simplified analysis, our only attribute variable will be total height, so it is not needed to update that yet!

Another way to look at it is that a feature variable defines an element in the dataset, and an attribute variable describes something about that element. In our case, the feature is a country so we would use the “country” as our feature variable name and total vertical distance from the sea level as our attribute variable name.

Operations with data

In spatial statistics, two main types of operations can be performed on datasets. These include one-off calculations and longer-term analyses. One-off calculations typically do not require advanced statistical skills but performing them repeatedly to get meaningful results depends heavily on how well you prepare your dataset before calculating.

Longer-term analyses usually involve defining regions or areas in which conditions change or remain stable, identifying transition points between these different states, and determining what factors contribute to those changes or stability. This is where more complex statistical concepts come into play.

There are many ways to perform these tasks depending on the research question being asked. Different methods have their strengths and weaknesses, so it is important to know about all of them before starting any long-term studies. Some examples of spatial analytical techniques include kriging (spreading out average values using weighted averages), cluster analysis (groupings or patterns of variables determined through math), and regression (prediction models).

All three of these concepts rely on an underlying concept known as “distance”. The distance here refers to something like the Euclidean distance formula, which looks at the difference between two numbers and then multiplies this by some constant. For example, if the numbers are both 5, then the product would be 25. The smaller the number, the closer they are together and therefore a larger distance value. It works oppositely for bigger numbers.

Creating plots

Let’s look at some examples of creating spatial plots. First, we will make a scatterplot using ggmap to visualize whether or not there is an association between state capitals and their surrounding suburban areas.

We can use the same package to create our second example plot. This time we will determine how much urban sprawl exists in each state by calculating the ratio of land area for cities over the total land area for the state.

These two graphs show that most states have a relatively little amount of urban sprawl, with only one (Hawaii) having very spread-out suburbs. Only five states had less than 2% of their population living in a city with more than 10k people, while three other states had less than 5%. On average, 11% of all residents lived within 1km of a major city, making this a popular option.

However, 42% of all US citizens live farther away from a large town than they do from a small town, showing that rural America still dominates! These longer distances are due to many reasons- the cost of housing being the main factor.

Understanding spatial data

A fundamental element of GIS is understanding how to work with geographic information. This article will go into more detail about what is meant by “spatial”, and how rasters (images) contain this spatial content.

In general, anything that has a position or location is considered spatial. For example, if you had a ball and a bottle, the ball and the top of the bottle are both spatial objects.

The distance between these two points is not important - only their relative positions in space are! The same goes for people- they are spatial entities.

When talking about maps and geographical images, there are three main types of content used: coordinates, shapes, and colours. These components together make up what is referred to as spatial content. More complex examples include features such as roads and rivers which have separate sets of coordinates but also shapes and colours.

Several software packages use this concept to create interactive, map-based applications. Two major ones are Google Maps and Esri’s ArcGIS. While neither one is better than the other, they do cater to different users. If you are ever looking to learn something new about mapping, try out either one for some practice.

Sources: MapMakingBeginners.com, Wikipedia, about.