Chapter 26: Introduction to GIS

Chapter Overview

This chapter introduces Geographic Information Systems (GIS) as foundational technology for Community Mapping. GIS allows practitioners to store, visualize, analyze, and interpret spatial data with precision and power. You will learn the core concepts — layers, attributes, geometry types, coordinate systems, geocoding, and spatial operations — that underpin all GIS work. The chapter emphasizes practical application, common mistakes, and ethical awareness: GIS is a tool, not a neutral one, and how you use it shapes what communities see and how decisions get made.

Learning Outcomes

By the end of this chapter, you will be able to:

Define GIS and explain how it differs from static maps or simple cartography
Identify the three fundamental geometry types (points, lines, polygons) and when to use each
Explain how layers and attributes work together to represent real-world features
Distinguish between geographic coordinate systems and projected coordinate systems, and recognize when projection choice matters
Apply geocoding to convert addresses into mappable locations
Conduct basic spatial operations including spatial joins, buffers, and catchment analysis
Recognize and avoid common GIS mistakes that distort interpretation or mislead audiences

Key Terms

GIS (Geographic Information System): Software and workflows for capturing, storing, analyzing, and visualizing spatial data.
Layer: A thematic dataset representing one category of features (e.g., roads, parks, service locations).
Attribute: Non-spatial information attached to a spatial feature (e.g., a park's name, size, or amenities).
Geometry: The spatial representation of a feature — point, line, or polygon.
Coordinate System: A standardized framework for locating features on Earth's surface using coordinates (latitude/longitude or projected x/y).
Geocoding: The process of converting addresses or place names into geographic coordinates.

26.1 What Is GIS?

Geographic Information Systems — universally abbreviated as GIS — are software platforms designed to work with spatial data. At their simplest, they let you put dots, lines, and shapes on a map. At their most powerful, they let you analyze complex spatial relationships, model scenarios, predict outcomes, and communicate findings with clarity and precision.

GIS is not a map. A map is a product — an image, a static representation. GIS is a system — a set of tools, data structures, and workflows that produce maps, but also do much more. GIS lets you ask spatial questions: Which neighborhoods have no grocery stores within walking distance? Where do service gaps overlap with high vulnerability? How many people live within 500 meters of a park? What routes do residents actually take to access healthcare, and what barriers do they encounter?

The origins of GIS trace to the 1960s, when Canadian geographer Roger Tomlinson developed the Canada Geographic Information System to manage land inventory data for rural planning. Tomlinson's innovation was recognizing that geographic data could be stored, layered, and analyzed digitally — making spatial analysis faster, more flexible, and more rigorous than hand-drawn overlays. His work earned him the title "father of GIS," and the field he founded now underpins everything from urban planning to environmental science, public health, logistics, and community development.

Modern GIS platforms range from desktop applications like ArcGIS and QGIS, to web-based community-mapping tools like map.ca and Felt, to cloud-based geospatial databases and APIs that power location services in smartphones. Some are proprietary and expensive. Others, like QGIS, are free, open-source, and community-maintained. The choice of platform matters less than understanding the core concepts that all GIS tools share.

For Community Mapping practitioners, GIS offers three critical capabilities: visualization (making spatial patterns visible), analysis (answering spatial questions with data), and communication (presenting findings in ways that support decisions and actions). A well-executed GIS workflow can turn raw data into insight, and insight into change.

But GIS is not neutral. Every choice — what to map, what categories to use, what scale to display, what colors to choose, what to label and what to leave unlabeled — shapes how people interpret the world. A map showing "service deserts" frames the problem one way. A map showing "underserved populations" frames it another. Both may use the same data, but they tell different stories. GIS practitioners must be conscious of these choices and accountable for the narratives their maps create.

26.2 Layers and Attributes

The fundamental organizing principle of GIS is the layer. A layer is a thematic dataset that represents one category of real-world features. One layer might show roads. Another might show parks. A third might show census boundaries. A fourth might show service provider locations. Each layer sits on top of the others, like transparent sheets on an overhead projector, and together they build a composite picture of place.

Layers are not just images. Each layer is a structured dataset. A "parks" layer doesn't just draw green shapes — it stores information about each park: its name, size, amenities, accessibility features, hours of operation, maintenance status. This non-spatial information is called attribute data, and it is stored in a table linked to the spatial features.

Think of it this way: the geometry tells you where something is. The attributes tell you what it is and what you know about it. A point on a map might represent a community center. Its attributes might include the center's name, address, operating hours, programs offered, capacity, and contact information. When you click on that point in a GIS, the software displays its attributes. When you run an analysis — such as "find all community centers that offer youth programs and are open evenings" — the GIS queries the attribute table.

This layer-plus-attribute structure makes GIS extraordinarily flexible. You can turn layers on and off to focus on specific themes. You can filter layers to show only features that meet certain criteria. You can symbolize features based on their attributes — for example, coloring parks by size, or sizing service dots by the number of people served. You can join attribute data from external sources (like census data) to spatial features (like neighborhood boundaries) and visualize patterns.

But the layer model also has limits. Layers flatten complexity. A "road" layer treats all roads the same unless you add attributes to distinguish them. A "service provider" layer shows dots without showing relationships, referral pathways, or trust networks. Layers are powerful for representing what exists, but weaker for representing how things connect or what people experience. This is why Chapter 1 emphasized that Community Mapping integrates GIS with qualitative methods, participatory engagement, and systems thinking.

26.3 Points, Lines, and Polygons

All spatial data in GIS is represented using one of three fundamental geometry types: points, lines, or polygons. Understanding when to use each is essential.

Points represent discrete locations. A service provider is a point. A playground is a point. A bus stop is a point. Points are defined by a single coordinate pair (latitude/longitude or x/y in a projected system). Points are appropriate when the feature's exact location matters and its size or shape is either irrelevant or too small to display meaningfully at the map's scale.

But point representation has limits. A hospital is not truly a point — it is a building with entrances, wings, parking, and departments. Representing it as a point simplifies reality. This simplification is fine at city scale, where you are showing the distribution of healthcare facilities. It becomes problematic at neighborhood scale, where the location of the emergency entrance versus the main lobby might matter for accessibility analysis.

Lines represent linear features: roads, rivers, trails, transit routes, pipelines, power lines. Lines are defined by a series of connected coordinate pairs. They have length but no width (though you can style them with visual width on a map). Lines are appropriate when the feature's path or connectivity matters — for example, analyzing which roads connect to which neighborhoods, or measuring travel distance along a route.

Line data can be simple (a road segment running from one intersection to another) or complex (a transit route with stops, schedules, and service frequency). The attribute table attached to a line layer can store information like road name, surface type, speed limit, or number of lanes. For transit lines, attributes might include route number, frequency, and accessibility features.

Polygons represent areas: parks, neighborhoods, census tracts, flood zones, land parcels, building footprints, service catchments. Polygons are defined by a closed series of coordinate pairs forming a boundary. They have both area and perimeter. Polygons are appropriate when the feature's extent, size, or internal characteristics matter — for example, calculating how much park space exists per capita, or identifying which neighborhoods fall within a flood risk zone.

Polygon data is often used for aggregation. Census data comes aggregated to polygons (census tracts, block groups). Service usage data might be aggregated to postal code polygons. Health outcomes might be aggregated to health regions. This aggregation is useful for analysis but also introduces problems — which we will return to in Section 26.10.

Choosing the right geometry type is not always straightforward. Should a large park be a point or a polygon? It depends on your scale and purpose. At city scale, a point might suffice. At neighborhood scale, a polygon lets you analyze internal features (playgrounds, sports fields, trails) or measure walking distance to the park's edge, not just its center. The rule of thumb: use the geometry type that best supports the spatial questions you need to answer.

26.4 Coordinate Systems

Every location on Earth can be described using coordinates, but which coordinate system you use matters enormously. Get this wrong, and your map will be distorted, your measurements inaccurate, or your data misaligned.

There are two broad categories of coordinate systems: geographic coordinate systems (GCS) and projected coordinate systems (PCS).

A geographic coordinate system uses latitude and longitude — angles measured from the Earth's center. Latitude measures north-south position (0° at the equator, 90° at the poles). Longitude measures east-west position (0° at the prime meridian in Greenwich, England). The most common GCS is WGS84 (World Geodetic System 1984), used by GPS and most web mapping platforms. Canada also uses NAD83 (North American Datum 1983), which is nearly identical to WGS84 for most practical purposes but differs slightly in how it models Earth's shape.

Geographic coordinate systems are good for storing data and for global-scale mapping, but they are not ideal for measurement or analysis. Why? Because Earth is a sphere (technically an oblate spheroid), and latitude/longitude lines are curved. A degree of longitude at the equator represents about 111 kilometers. At 60° north latitude (roughly the latitude of northern Alberta), it represents only 55 kilometers. If you calculate distances or areas using latitude/longitude directly, you will get wrong answers.

This is where projected coordinate systems come in. Projection is the mathematical process of flattening the Earth's curved surface onto a flat plane. All projections distort something — distance, area, shape, or direction. The trick is choosing a projection that preserves what matters for your analysis and minimizes distortion in your area of interest.

For Canadian Community Mapping work, common projections include:

UTM (Universal Transverse Mercator): Divides the world into 60 narrow north-south zones, each with its own projection optimized for that zone. Canada spans UTM zones 7 through 22. UTM preserves distance and area within a zone, making it excellent for local and regional analysis. If your study area crosses UTM zones, you will need to reproject or accept some distortion.
Canada Albers Equal Area Conic: A national projection optimized for Canada. It preserves area, making it ideal for calculating densities (e.g., population per square kilometer) or comparing regions.
Web Mercator (EPSG:3857): The projection used by Google Maps, OpenStreetMap, and most web mapping tools. It preserves shape and direction (useful for navigation) but grossly distorts area, especially near the poles. Greenland looks bigger than Africa on a Web Mercator map; in reality, Africa is 14 times larger. Web Mercator is fine for visualization but should never be used for area calculations.

The ethical implication: maps that use Web Mercator exaggerate the size of northern regions and shrink equatorial ones, reinforcing colonial-era visual hierarchies that centered Europe and marginalized the Global South. This is not an accident — the Mercator projection was designed for 16th-century navigation, not for representing the world fairly. As Chapter 4.9 discussed, all maps are representations, and representations carry political weight.

When working in GIS, always check your data's coordinate system. Most GIS platforms will display it in the layer properties. If your layers are in different coordinate systems, the software will often reproject them on the fly for display — but this can cause subtle alignment errors. For analysis, reproject all layers to a common system appropriate for your study area and your analytical goals.

26.5 Geocoding

Geocoding is the process of converting addresses or place names into geographic coordinates that can be mapped. A street address like "123 Main St, Vancouver, BC" is meaningless to GIS software until it is geocoded into a latitude/longitude pair or projected coordinate.

Geocoding engines work by matching your address against a reference dataset of known addresses with coordinates. Common geocoding services include OpenStreetMap's Nominatim (free, open), ESRI's ArcGIS World Geocoding Service, the Google Maps Geocoding API (paid, with usage limits), and national postal services. Map.ca uses an open geocoder for the platform's pin-on-place workflow. Some are free; others charge per query. Some require internet access; others can run offline with local data.

Geocoding quality varies. In dense urban areas with well-maintained address databases, geocoding is usually accurate to within a few meters. In rural areas, where addresses may be sparse or ambiguous, geocoding can fail or return incorrect locations. A rural address might geocode to the center of a large parcel, a kilometer from the actual building. An apartment address might geocode to the building entrance, even though the unit is at the back. Newer developments may not yet be in the reference database.

Geocoding also fails when addresses are incomplete, misspelled, or formatted inconsistently. "123 Main St" and "123 Main Street" and "123 Main St." should all match, but some geocoders are finicky. Standardizing your address data before geocoding — cleaning up abbreviations, correcting typos, adding postal codes — dramatically improves match rates.

For Community Mapping, geocoding is essential. You might geocode:

Service provider addresses to map where resources are located
Client addresses (anonymized and aggregated) to understand where demand originates
Community asset addresses collected in participatory workshops
Event locations for a cultural mapping project

But geocoding has privacy implications. If you geocode individual home addresses, you create a dataset that could be used to identify people, track movements, or target vulnerable populations. Best practice: geocode to the smallest necessary scale (often postal code or neighborhood centroid, not exact address) and apply privacy safeguards (Chapter 25's validation and anonymization principles apply here).

Geocoding also has accuracy-equity implications. Low match rates in marginalized neighborhoods — due to informal addresses, recent immigration, or inconsistent data collection — can make those communities invisible in spatial analysis. If your geocoding only works for 90% of addresses, and the missing 10% are disproportionately low-income, you have introduced bias. Always report match rates and investigate patterns in unmatched records.

26.6 Spatial Joins

One of the most powerful GIS operations is the spatial join — combining data from two layers based on their spatial relationship, not on a shared ID or key field.

Imagine you have two layers:

A point layer of community centers
A polygon layer of neighborhoods

You want to know how many community centers are in each neighborhood. A spatial join answers this. The GIS checks which points fall inside which polygons and aggregates the count. The result: a new version of the neighborhood layer with an added attribute: "Number of Community Centers."

Spatial joins can be based on different spatial relationships:

Intersects: Features that touch or overlap (e.g., which roads intersect which neighborhoods?)
Within: Features completely contained inside another (e.g., which parks are entirely within the city boundary?)
Contains: The inverse of within (e.g., which census tracts contain at least one school?)
Nearest: The closest feature (e.g., for each household, what is the nearest grocery store?)

Spatial joins are essential for community mapping analysis. You might use them to:

Count how many seniors live within 500 meters of a recreation center (join senior population data to buffered service areas)
Identify which service providers operate in which neighborhoods (join provider points to neighborhood polygons)
Aggregate emergency calls by postal code (join incident points to postal code polygons)

But spatial joins can be misleading if boundaries are drawn poorly. Chapter 4 introduced the idea that boundaries are constructed, not natural. A neighborhood boundary drawn by a municipality might not match how residents define the neighborhood. A census tract boundary might split a cohesive community in half. Spatial joins treat these boundaries as truth, which can distort findings.

26.7 Buffers and Catchments

Buffers are one of the simplest and most useful spatial operations. A buffer creates a zone of a specified distance around a feature. For example:

A 500-meter buffer around a park shows the "walkable catchment" — the area within a 10-minute walk
A 1-kilometer buffer around a school shows the neighborhood it serves
A 100-meter buffer around a hazardous site shows the area at immediate risk

Buffers can be created around points, lines, or polygons. They are typically circular (for points) or parallel (for lines and polygons). Most GIS platforms let you specify buffer distance in meters, kilometers, feet, or miles.

Buffers are foundational for access analysis. To determine whether a neighborhood has adequate park access, you might:

Buffer all parks by 500 meters (a reasonable walking distance)
Merge overlapping buffers into a single "park access zone"
Overlay this zone with census population data
Calculate what percentage of residents live inside the zone

This approach is widely used in public health (access to healthcare), food security research (access to grocery stores), and urban planning (access to transit, green space, or services).

But buffer-based access analysis has significant limitations:

Straight-line vs. network distance. A 500-meter buffer measures "as the crow flies," but people don't fly — they walk along sidewalks, roads, and paths. Actual walking distance can be much longer than straight-line distance, especially in neighborhoods with curving streets, dead ends, or barriers like highways or rivers. Better GIS analysis uses network distance — calculating distance along the actual street network, accounting for one-way streets, crosswalks, and pedestrian infrastructure.
Ignoring barriers. A park may be 400 meters away in straight-line distance, but if a busy highway with no crosswalks separates it from the neighborhood, it is functionally inaccessible. Buffers do not capture this.
Uniform catchment assumption. A 500-meter buffer assumes everyone experiences that distance the same way. In reality, a senior with mobility challenges, a parent with a stroller, or a wheelchair user may have very different "walkable" distances. Accessibility is not just distance — it is barrier-free distance.

Advanced GIS tools address some of these limits. Network analysis tools calculate distance and travel time along street networks. Isochrones show the area reachable within a given time (e.g., "everywhere you can walk in 10 minutes"). Accessibility-aware routing considers slope, curb cuts, and sidewalk quality. These tools are more complex and data-intensive but produce more realistic access estimates.

26.8 Choropleth Maps

A choropleth map colors polygons (like census tracts, neighborhoods, or postal codes) based on an attribute value. For example, a choropleth might color neighborhoods by median income, with darker colors representing higher incomes.

Choropleth maps are ubiquitous in Community Mapping. They are used to show poverty rates, health outcomes, service usage, demographic patterns, election results, and more. They are visually powerful — patterns jump out. They are easy to produce. And they are easy to misinterpret.

The biggest problem with choropleth maps is the modifiable areal unit problem (MAUP). MAUP refers to the fact that patterns you see on a choropleth map depend heavily on the boundaries you use. Change the boundaries, and the pattern changes — even if the underlying data is identical.

Imagine mapping poverty rates in a city. If you use large boundaries (like wards), variation is smoothed out — you see broad patterns but miss local detail. If you use small boundaries (like census dissemination areas), you see fine-grained variation but the map becomes cluttered and harder to interpret. If you use postal codes, you are mapping mail delivery zones, which may or may not correspond to meaningful community boundaries.

A second problem: uneven population density. Choropleth maps give equal visual weight to all polygons, regardless of how many people they contain. A large rural area with 50 people gets the same color intensity as a small urban block with 5,000 people. This can create misleading impressions. A map showing "percentage of population experiencing food insecurity" might highlight rural areas with high percentages but small absolute numbers, while underplaying urban areas with lower percentages but thousands of affected people.

The solution is not to avoid choropleth maps — they are too useful. The solution is to pair them with complementary visualizations. Show a choropleth alongside a dot density map (where each dot represents a certain number of people), so viewers can see both rates and absolute numbers. Add population labels to polygons. Use bivariate mapping to show two variables simultaneously (e.g., both poverty rate and total population).

Also, be cautious with color choice. Red-to-green color ramps imply "bad" to "good," but colorblind viewers cannot distinguish them. Single-hue sequential ramps (light blue to dark blue) are safer for showing "less to more." Diverging ramps (red-white-blue) are good for showing data with a meaningful midpoint (like above/below average).

26.9 Heat Maps

Heat maps (also called density maps or kernel density maps) visualize the concentration of point data. They are commonly used to show crime hotspots, disease clusters, service demand, or event locations.

A heat map works by placing a "kernel" — a smooth probability surface — around each point, then summing overlapping kernels to produce a continuous surface showing density. The result looks like a gradient of color, with "hot" areas (high density) in bright or saturated colors and "cool" areas (low density) in muted colors or transparency.

Heat maps are visually striking. They communicate concentration at a glance. They are especially useful when you have many overlapping points that would be illegible as individual dots.

But heat maps are also easy to misinterpret:

They show density, not absolute numbers. A bright red hotspot might represent 50 incidents in a small area. A cooler yellow area might represent 200 incidents spread over a larger area. Which matters more depends on your question.
They are sensitive to kernel size (bandwidth). A small kernel produces a detailed, noisy map. A large kernel produces a smooth, general map. The "right" kernel size depends on your data and context, but there is no objective answer. Different kernel sizes produce different impressions, and most viewers do not understand this.
They can mask spatial heterogeneity. A heat map showing "crime density" might lump together burglaries, assaults, and vandalism, even though these have different causes, patterns, and responses. Aggregation simplifies — sometimes too much.
They can stigmatize neighborhoods. A crime heat map labeled "danger zones" or "hotspots" can reinforce stereotypes, reduce property values, and justify over-policing. A service demand heat map can make neighborhoods look "needy" or "high-cost." Heat maps are never neutral.

Best practice: use heat maps for exploration and pattern detection, but supplement them with counts, tables, and contextual data. Be transparent about methodology (kernel size, classification breaks). Avoid sensationalist color schemes (black-to-red "crime maps"). Consider who will see the map and what narratives it might reinforce.

26.10 Common GIS Mistakes

GIS is powerful, but power brings risk. Here are the most common mistakes that distort analysis or mislead audiences — and how to avoid them.

Mistake #1: Wrong coordinate system for analysis.
You run a "calculate area" operation on census tracts, but your data is in WGS84 (latitude/longitude). The areas are wrong because lat/long doesn't preserve area. Solution: Always reproject to an equal-area coordinate system (like Canada Albers or a local UTM zone) before measuring areas or distances.

Mistake #2: Geocoding errors in rural areas.
You geocode 500 service provider addresses. Urban addresses match with 95% accuracy. Rural addresses match at 60%, and many geocode to the wrong side of the county. You proceed with analysis, unaware that your map systematically underrepresents rural services. Solution: Always validate a sample of geocoded locations, especially in rural or under-mapped areas. Ground-truth geocoding by checking a few addresses manually on satellite imagery or street view.

Mistake #3: Using Web Mercator for area calculations.
You create a choropleth map of population density, calculate density as population/area, but your data is in Web Mercator projection. Areas at high latitudes are inflated, so densities are artificially low. Solution: Never use Web Mercator for any calculation. Use it only for web-based visualization.

Mistake #4: Ignoring the modifiable areal unit problem (MAUP).
You map poverty rates by postal code and conclude that the northwest quadrant of the city has the highest poverty. But postal codes are arbitrary mail-delivery zones. If you had used census tracts, the pattern might look different. Solution: Test your analysis with multiple boundary definitions. Report which boundaries you used and acknowledge the limitation.

Mistake #5: Choropleth maps without population context.
You map percentage of seniors by neighborhood. A large, low-density neighborhood shows 30% seniors (300 people). A small, dense neighborhood shows 15% seniors (1,500 people). The map makes the low-density area look "older," but the dense area has five times as many seniors. Solution: Always pair choropleth maps with absolute counts, dot density overlays, or population-weighted visualization.

Mistake #6: Buffers instead of network distance.
You draw 500-meter buffers around parks and declare that 80% of residents have "access." But half the buffered area is separated from the parks by a highway with no crosswalks. Actual access is much lower. Solution: Use network-based accessibility analysis whenever possible. If you use buffers, state the limitation clearly.

Mistake #7: Treating boundaries as truth.
You aggregate health data by census tract and report findings by tract. But census tracts do not correspond to neighborhoods, service areas, or community identity. Your findings are statistically accurate but socially meaningless. Solution: Supplement boundary-based analysis with community-defined geographies (Chapter 3.10). Involve residents in validating whether your spatial units make sense.

Mistake #8: Ignoring data quality.
You download a "complete" dataset of community services from a government portal. It was last updated in 2018. One-third of the listings are outdated — services have closed, moved, or changed eligibility. Your map is systematically wrong. Solution: Always check data provenance, update dates, and completeness (Chapter 25). Validate a sample. Triangulate with other sources.

Mistake #9: Over-interpreting precision.
Your map shows service locations with GPS coordinates accurate to eight decimal places — sub-millimeter precision. But the addresses were geocoded from a database with inconsistent quality, and half the coordinates are actually building centroids, not entrance locations. The appearance of precision is false. Solution: Be honest about data accuracy. Don't display false precision. If your data is accurate to 10 meters, don't display coordinates to the centimeter.

Mistake #10: Publishing sensitive data.
You map the home locations of vulnerable individuals — say, clients of a domestic violence shelter or people experiencing homelessness — at precise coordinates. The map is published publicly or shared widely. This endangers people. Solution: Apply privacy safeguards. Aggregate to postal code or neighborhood. Jitter points (add random noise to coordinates). Require access controls. As Chapter 9 established, some places must not be mapped publicly, and some knowledge must not be shared without consent.

26.11 Synthesis and Implications

GIS is foundational infrastructure for Community Mapping. It turns spatial data into insight, patterns into maps, and questions into evidence. It makes the invisible visible — revealing service deserts, access barriers, clustering of risks, and distribution of assets. It supports planning, advocacy, coordination, and accountability. No modern Community Mapping practice can ignore GIS.

But GIS is not the whole story. Chapter 1 established that Community Mapping integrates quantitative data with qualitative knowledge, spatial analysis with participatory engagement, and maps with stories. GIS gives you the "where" and the "how many." It does not, on its own, give you the "why" or the "what does this mean to people living here."

The implications for practice:

GIS is a tool, not a substitute for community engagement. A GIS analysis showing that a neighborhood has "adequate park access" based on 500-meter buffers does not mean residents feel the parks are safe, welcoming, or usable. Spatial proximity is not the same as lived accessibility. GIS findings must be validated with community voice (Chapter 25).

GIS choices are political. What you choose to map, what categories you use, what scale you display, what colors you choose, and what you name the layers — all of these shape interpretation. A map showing "high-crime neighborhoods" tells a different story than a map showing "neighborhoods with high police enforcement." Both might be based on the same data. GIS practitioners must be conscious of how their choices construct narratives.

GIS requires rigor. Wrong coordinate systems, poor geocoding, inappropriate projections, and unvalidated data produce misleading maps. Mistakes are easy to make and hard to catch unless you develop habits of checking metadata, validating samples, and testing sensitivity. Community Mapping practitioners must build GIS literacy — not necessarily deep technical expertise, but enough understanding to recognize when something is wrong and ask the right questions.

GIS is increasingly accessible, but access is uneven. Open-source tools like QGIS and browser-based community-mapping platforms like map.ca lower barriers to entry. But GIS still requires time, training, and spatial thinking skills that are not evenly distributed. Organizations with resources can hire GIS specialists. Grassroots groups often cannot. This creates power imbalances: who gets to map, whose maps are trusted, whose spatial claims are taken seriously. Ethical Community Mapping involves building capacity, sharing tools, and supporting communities to do their own spatial analysis.

Looking ahead: Part V continues with open data sources (Chapter 27), mobile data collection (Chapter 28), remote sensing (Chapter 29), AI-assisted workflows (Chapter 30), knowledge graphs (Chapter 31), and platform design (Chapter 32). GIS is the foundation. The chapters that follow show how to populate GIS with data, extend it with new technologies, and integrate it into community-controlled systems.

26.12 GIS Starter Lab

Purpose: This lab gives you hands-on experience with GIS fundamentals: loading data, exploring layers and attributes, geocoding addresses, creating buffers, and producing a simple map. You will create a basic community resource map showing service locations, neighborhoods, and access zones. By the end, you will have a working GIS project and familiarity with essential operations.

Materials Needed:

Software: QGIS (free, open-source, available for Windows, Mac, and Linux). Download from qgis.org. Alternative: if QGIS installation is not feasible, use a browser-based tool such as map.ca (recommended starter — community-first, no Google account required) or Felt.
Data: You will need three datasets:
1. Neighborhood boundaries (polygon layer) — available from your municipality's open data portal, or use census tract boundaries from Statistics Canada.
2. Service provider addresses (list of 10-20 addresses in your study area — community centers, libraries, clinics, food banks, etc.). You can compile these manually from web searches or use an open dataset if available.
3. Basemap (for context) — QGIS includes built-in options (OpenStreetMap, satellite imagery).
If you cannot access local data, use sample datasets from QGIS tutorials or OpenStreetMap extracts.

Steps:

Install and open QGIS. Launch the software and create a new blank project. Save it with a meaningful name (e.g., "Community_Resources_Map").
Load neighborhood boundaries.
- Go to Layer > Add Layer > Add Vector Layer (or drag and drop a shapefile or GeoJSON file into QGIS).
- Your neighborhoods should appear on the map.
- Right-click the layer, choose Open Attribute Table, and explore the data. What fields are included? Are there names, population counts, or other attributes?
Set a basemap.
- Go to Browser Panel > XYZ Tiles > OpenStreetMap (or add via Layer > Add Layer > Add XYZ Layer).
- Drag OpenStreetMap into your project. This provides street context for your map.
Geocode service provider addresses.
- Prepare your address list as a CSV file with columns: Name, Address, City, Postal Code.
- In QGIS, go to Layer > Add Layer > Add Delimited Text Layer.
- Load your CSV, but do NOT check "geometry" options yet (the file has no coordinates).
- Use a geocoding plugin like MMQGIS (install via Plugins > Manage and Install Plugins).
- Run MMQGIS > Geocode > Geocode CSV with Web Service. Choose OpenStreetMap (Nominatim) as the geocoder — it is free, open, and globally usable without an API key.
- The plugin will generate a new point layer with geocoded coordinates. Inspect the results: did all addresses match? Are there errors?
Alternative (simpler, if plugin issues arise): Manually geocode a few addresses using map.ca's pin-on-place workflow (drop a pin, copy the latitude/longitude from the pin's metadata) or any street-map tool with a coordinates-on-click feature. Create a CSV with Name, Latitude, Longitude. Load this CSV as a point layer (Layer > Add Delimited Text Layer, specify lat/long fields).
Style the service provider layer.
- Right-click your service points layer > Properties > Symbology.
- Choose a simple marker (e.g., circle). Set the color to a bright, distinct color (e.g., red or orange).
- Optionally, label points with their names: Properties > Labels > Single Labels, choose the Name field.
Create 500-meter buffers around service providers.
- Go to Vector > Geoprocessing Tools > Buffer.
- Select your service provider layer as input.
- Set buffer distance to 500 meters (or 0.5 kilometers, depending on your coordinate system).
- Run the tool. A new polygon layer will be created showing 500-meter zones around each service.
- Style the buffer layer with a semi-transparent fill (e.g., light blue at 30% opacity) so you can see overlaps.
Analyze access.
- Visually inspect: which neighborhoods are well-covered by service buffers? Which have gaps?
- (Optional advanced step) Use Vector > Geoprocessing Tools > Intersection to identify which parts of neighborhoods fall within service buffers.
Create a simple map layout.
- Go to Project > New Print Layout. Give it a name (e.g., "Resource Access Map").
- Add a map frame: Add Item > Add Map. Draw a rectangle on the page — your map appears.
- Add a legend: Add Item > Add Legend. QGIS auto-generates legend entries.
- Add a title: Add Item > Add Label. Type your title (e.g., "Community Service Access – 500m Walkable Catchments").
- Add a scale bar and north arrow if desired.
- Export your map: Layout > Export as Image (PNG) or Export as PDF.
Reflect on what you learned.
- Did geocoding work perfectly, or were there errors?
- Do the 500-meter buffers feel realistic, or do you see barriers (highways, rivers) that would limit actual access?
- What additional data would improve this map?

Deliverable:

A saved QGIS project file (.qgz)
An exported map image (PNG or PDF)
A 1-page reflection (300-400 words) answering:
- What did you learn about your study area from this map?
- What challenges did you encounter (geocoding errors, data gaps, interpretation questions)?
- How would you improve this analysis if you had more time or data?

Time Estimate: 2-3 hours (includes software installation, data gathering, and execution)

Safety and Ethics Notes:

If you geocode real addresses, do not publish precise locations of sensitive services (e.g., domestic violence shelters, harm reduction sites) without permission.
Do not map individuals' home addresses.
If you share your map publicly, include a disclaimer: "This map is for educational purposes and may contain errors or omissions. It should not be used for navigation or decision-making without verification."

Key Takeaways

GIS is a system for storing, analyzing, and visualizing spatial data — not just a map-making tool, but a framework for asking and answering spatial questions.
Layers and attributes work together: geometry tells you where, attributes tell you what and what you know about it.
Points, lines, and polygons are the fundamental geometry types; choosing the right one depends on scale, purpose, and the spatial question you are answering.
Coordinate systems matter: always reproject to an appropriate system for your study area and analytical goals; never use Web Mercator for calculations.
Geocoding converts addresses into mappable coordinates but requires validation, especially in rural or under-mapped areas.
Spatial joins, buffers, and catchment analysis are essential operations for community mapping, but all have limitations and require careful interpretation.
Choropleth maps are powerful but easily misinterpreted; always pair them with population context and be aware of the modifiable areal unit problem.
Common GIS mistakes — wrong projections, poor geocoding, inappropriate buffers, ignoring MAUP, publishing sensitive data — can distort findings or cause harm; rigor and validation are essential.

Plain-Language Summary

GIS stands for Geographic Information Systems — software that lets you work with maps and spatial data. It's more than just drawing maps; it's a tool for asking questions like "Where are the grocery stores?" or "How many people live far from a park?" and getting answers with data.

GIS works by organizing data into layers — one layer might show roads, another parks, another neighborhood boundaries. Each layer has information attached (like a park's name or size). You can stack layers on top of each other to see patterns, measure distances, or find gaps in services.

There are three basic shapes you use in GIS: points (like dots for bus stops), lines (like roads), and polygons (like shapes showing neighborhoods). You also need to pay attention to coordinate systems — different ways of measuring locations on Earth — because using the wrong one can make your measurements wrong.

GIS is powerful, but it's easy to make mistakes. You can draw buffers around services to show "walkable distance," but that doesn't account for busy roads or hills. You can make colorful maps showing poverty rates, but if you're not careful, the map can mislead people about where the real need is. Good GIS work means being careful, checking your data, and being honest about what the map shows — and what it doesn't.

End of Chapter 26.

Chapter 26. Introduction to GIS