Introduction

Thematic Maps offer one of the greatest potentials for creativity in maps. You can think of them a lot like info-graphics with location. Making truly great thematic maps is an art form and requires a great deal of skill to fully master. Where most of the maps we have been using in the course so far seek to express information about where things are and how to get there; thematic maps will strive to provide information about a location that is easy to digest, making use of symbol hierarchy, colour choices, and embedded charts.

An ideal thematic map will convey its meaning intuitively, while a legend will still be present it shouldn’t needed to understand the essence of the map. A truly great thematic map would then seamlessly blend into the medium where it is presented, matching styles of surrounding elements.

1. Nominal/Categorical Symbology

Choosing Colours for Nominal Data:

Do not use varying shades of one colour in a nominal symbol fill set (for example, multiple shades of green) since increasingly darker shades of one colour are used to symbolize magnitude (interval) data as “darker” means “more.”

As an example of this, open the (Provincial Boundaries) layer “Provinces with population 2020” from the ArcGIS Portal. This will show up in your Contents pane as ‘gpr_000a11a_e’. If you wish, you can rename it by opening the properties and editing the Name in the General tab.

Next lets take a look at our map projection: you will notice that the file is presented very rectangular compared to how we are used to seeing maps of Canada. We will fix this by changing the map’s projection. Note that layers in your project, and the map in the drawing order can have different projections. ArcGIS will convert the layer projections in real time to match the map.

Double click on Map in the Drawing Order to open properties. Under Coordinate Systems, search for Canada Albers, find Canada Albers Equal Area Conic in the list under Projected Coordinate System, select it, confirm Current XY has been updated (see point 5 in the screenshot below), and press OK.

Now it is time to symbolize the layer, first using nominal coloring.

Select the Layer in the contents pane, and find Symbology in the Feature Layer tab, choose (under ‘Symbolize your layer by category’) Unique Values

Set Field 1 to PRNAME

This doesn’t display any more information than before, but the random colours add more contrast between the provinces and should seem very familiar to the type of national maps you would have seen in primary school. ArcGIS provides a variety of colour schemes that can be used. The default provided is appropriate for nominal data, however there are many more in the drop-down, or you can build your own using the gear icon.

2. Symbolizing Areas by Quantities

Next, we will examine how colour can be used to provide additional meaning to the data on the map. In GIS, the attributes we will be using for styling can come in two primary ways. The first is as an attribute of the polygons that you would find in the attribute table for the layer you are styling. The second way, as is provided to us by Stats Canada, has the attributes in a separate data table that can be linked in GIS software to the geometry.

A great reason that we might want to use data in this way is that the boundary files can be used with a variety of statistics, and for simplicity, we may not want to have hundreds of attributes with the geometry. Instead, we may want to add only the information that we need.

For reference these data can be downloaded from here:

https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/index-eng.cfm

Next, we will get the actual thematic data. This generally comes in the form of a spreadsheet (remember the CSV files from Lab 3), or in this case, we will be using an ESRI DBF table. In shapefiles, the attribute table is stored as a .dbf file.

The CanadaEnergy2020 table is available on the L: drive at L:\GEOG205\lab05\CanadaEnergy2020. Bring it into your project.

At this stage, the Drawing Order of your map should appear as below

Notice that the CanadaEnergy2020 is listed as a stand-alone table. This, like the .csv file we imported in Lab 3, is because there is no spatial information associated with this layer. To draw this on the map, we need to “join” our layers together. Right-click on gpr_000a11a_e (Stats Canada name for the Provincial Boundaries), and find the option to add a join.

The Join tool will take two tables that each have a column with exactly the same text, and will add the columns of the Join Table to the end of the Input Table (see a screenshot of the Join window below). Note that this is case sensitive, so when you download your data you may need to edit names to match the boundary files that you have.

For our case, the PRNAME column exists in both tables so we will join them based on this.

Press OK, then open the attribute tables and confirm that the extra columns have been added to the attributes.

Now for the gpr_000a11a… layer, go into Feature Layer > Symbology > Unclassed Colors


To visualize the energy demand for each province, set the field to demand:

Now when you look at your map you should see the first problem that comes up when working with this type of data. The Territories all appear white, and Saskatchewan and Manitoba both appear in the same shade of light blue. What the map is really showing us at this point is how uneven the spread of population is across Canada. Whenever you are working with this type of data, take a moment to consider what variables may be driving the numbers you are trying to graph.

To make this map a little bit more interesting, change the field to “percapa”. This field shows the energy used per person in each province in 2020 (note that no units of energy were provided with this dataset).

Normalizing Data for Display

This data was provided to us already normalized, however this will not always be the case; an alternative option would be to select ‘demand’ and normalize it against the population.

Conveying meaning with colour

Next we will look at the Color Scheme. You can pick several from the drop-down, or press the ‘Format Colour Scheme” at the bottom of the drop-down to be given a window that allows you to select custom colours at intervals across the gradient. When picking your colour scheme, your primary focus should be that the colours are intuitive! Going from green to red will convey a very different meaning than red to green. Secondly, you should pick a Color Scheme that fits the theme of your map!

Lets look at a couple examples below of how a poor choice can show the wrong message:

This colour scheme provides very little meaning. Without seeing the ramp for reference it is hard to tell what purple means.
This colour scheme is backwards and conveys the wrong message.
This map, while far from the best looking option, shows provinces using less electricity per person as green, providing a more intuitive meaning.

3. Proportional Symbols

  • Symbol size should reflect the values being represented.
  • All symbols should be visible. The smallest symbols should be bigger than dots.
  • Where there is overlap, if any, the smaller symbols should retain their outline to give the visual impression of being on top.
  • The circle is the standard proportional symbol, but in some cases, other shapes may work better.

ArcGIS Pro Proportional Symbols

To take a look at proportional symbols, we are going to add some XY data to our map. You will find the layer under L:\GEOG205\lab05\postsecondary\NewBrunswick Universities.xlsx\Sheet1$ You will need to access this through the catalog pane. Much like Windows views a geodatabase with the .gdb extension as a ‘file’, Excel files are not opened directly in ArcGIS. Rather, the catalog will allow the opening of specific sheets inside of the workbook.

Add the university points to the layer using the XY Point Data tool like we did in Lab 3. Note that you cannot drag-and-drop XY data directly onto ArcGIS Pro.

This layer contains the 3 universities in New Brunswick that have a post-graduate program as listed by Wikipedia (https://en.wikipedia.org/wiki/List_of_universities_in_Canada).

Once you add the layer, go to Feature Layer > Symbology > Proportional Symbols

Next, set the field to Total [number of students], then pick a colour and shape you like by clicking the symbol beside template (circles are generally a good choice!)

Note also that you can constrain the size of the symbols. If symbols are appearing too small, increase minimum size; if too large, reduce the maximum size.

Add labels as we did in previous labs and you should have something as below.

ArcGIS Pro Graduated Symbols

Graduated Symbols work the same as proportional, except they are placed into classes instead of a gradient. You have already used Graduated Symbols on the roads in Smithers for Lab 4, so we will not be revisiting this today.

Charts

Continuing with the New Brunswick Universities, wouldn’t it be cool to know the proportion of graduate to undergraduate students? To do this, under symbology, choose charts:

Set Chart type to Pie Chart

Fields will be Undergrad, and Postgrad. You will also need to increase the size under Appearance to make the charts easy to read.

Finally, you may wish to set Display options to Clockwise so that the chart starts at the top:

Discussion: What is wrong with the legend below?

Using both Scale and Colour

The keen among you may be wondering why there is a need to choose between Graduated Colours and Graduated Symbols. The answer is that you don’t need to choose. Before we continue, add the layer Populated Places from the portal.

Start by styling based on one of the attributes above (graduated colours or symbols), then click on the Vary Symbol by Attribute button in the Symbology pane.


If your primary symbol type is colour based, you will be able to set a proportional size field. If your primary symbol is size-based, you will be able to set an unclassed colour ramp field.

*Note that only the primary symbol type allows for graduated classes. Secondary fields will always be a continuous data range.

In this case there is only one attribute available, however this is a great way for quickly showing multiple attributes.

4. Heat Maps

Sometimes you simply have too many points. An example of this would be the Lower Mainland (Vancouver BC) crash data from ICBC.

Turn off all layers except the base-map and add Lower_Mainland_Crashes_data.csv (from the L: drive > GEOG205 > Lab 5 folder) to your map as XY point data.

Once this data is added, the points are so dense that you can’t really tell what is happening.

To fix this, select Heat Map from the symbology pane.

The settings in here are

  • Radius – Determines how far to search for nearby points
  • Weight Field – This is set to crash count as each point can represent more than one crash.
  • Colour ramp – Same concept as we have seen in the lab so far
  • Method – Choosing Dynamic recalculates the colour ramp based on the data visible on the screen. (Increased contrast as you zoom in).

Here is a trick on how to soften the edges of your color map:

Inside ‘Format Colour Scheme’, this ramp has the rightmost colour chip as red with 15% transparency to help the labels be slightly more readable.

The left-most chip is 100% transparent, the chip just slightly to the right is at 85% transparent. This positioning causes the transparency to shift rapidly at low densities into fully transparent, ‘feathering’ the edge of the heat map.

Before:

After:


Assignment 5

Due at the beginning of next week’s lab:

For your assignment this week, you will be producing two thematic maps. Pay particular attention to the overall aesthetic of your maps.

Along with your maps, include 1-2 paragraphs about why you made the design decisions you have made.

  • What meaning do colours and fonts convey?
  • How do colours work together?
  • What prompted your decisions on where to place elements such as legend and scale bar?

If you are looking for a way to find colours that work well together consider using a colour wheel like this one https://color.adobe.com/create/color-wheel

Marking Rubic:

A – To achieve an A on this assignment, see B below, with the addition that colours and fonts work well to convey the meaning of the map. A base map that matches your colours and labels has been chosen. Design is cohesive and conveys the ‘feeling’ of the map. Elaborate on your design choices in your paragraph provided with the maps.
B – Map contains all required elements. It is laid out well, easy to read, free of formatting mistakes on labels and scale bars. The symbols that are used make logical sense, and are not overly messy.
C – Map contains all required elements listed in the assignment
D – Maps are missing required elements
F – Maps do not serve the intended purpose

Reminders:

  • Export your maps as a pdf following naming conventions for the course:
    <username>_geog205_lab5_<map>.pdf
  • All maps for this course must contain a Title, Your Name, and the Date you produced it
  • All maps must include a scale bar and legend
  • Make sure your map is visually pleasing. You have colours, fonts, borders, and multiple base maps at your disposal. (Some of the base maps are very simple in design with very few to no labels, this can be a benefit on thematic maps in building your overall look)
  • Proofread your PDF before emailing to ensure that the formatting survived the export, if something isn’t looking the way you intend to, reach out to your TA for help

1. Quantity Map of Canada Median Income 2015

Data: L:\GEOG205\lab05\Asignment\canada_median_income_2015.csv

Produce a map that uses a colour gradient to show Median Income in Canada by province.

  • One optional component to help your map would be to locate the Census Canada cartographic boundaries (left to your research to find). One note is due to the complexity of these boundaries any ArcGIS Pro takes about 2 minutes to draw the map with labeling enabled. If you go this route a possible solution is to manually label by adding text the same way you did the title on layout, using abbreviations can also help to keep the map clean.

2. Heat Map of Peru Seismic Events

Data: L:\GEOG205\lab05\Asignment\eventos_sismicos

Add this data to a new map by inserting a New Map (button next to inserting a New Layout in the Insert tab on the ribbon).

You can use the official projection in Peru (WGS 1984 UTM 18S). After adding eventos_sismicos.shp, right click on the Map and go to Properties/Coordinate systems. Select the projection from the shapefile.

Produce a heat map of the Seismic events in Peru. Map elements should include a legend showing the intensity scale (not absolute values, so labeling high and low would be appropriate). You may choose to use Magnitude as the weight (as we used count for crashes), however, your title should reflect this choice (are you mapping frequency of events, or total activity?).

Categories: GEOG 205Labs