Cartography ·
From Geographic Data to Cartographic Art: How I Make My Maps
Geographic data supplies the material. Terrain processing, individually styled features, researched labels and sustained compositional judgement turn it into a cartographic artwork.
Andres de Wet
A road can be in exactly the right place and still be drawn badly. Its line may be too heavy for the landscape beneath it, too similar to a river beside it, or too prominent for its importance within the region. Add thousands of other features, each competing for attention, and geographic correctness alone will not produce a readable map.
Much of my work happens in those decisions.
The short film, on the cartography landing page at, https://www.34degreesdesign.com/cartography, showing how I make my cartographic art uses the Western Cape as its example. The process extends across the collection: South African regions, Australian landscapes, and the Pacific Northwest spanning the United States and Canada. The sources and geographic character change. The need to resolve terrain, infrastructure, names and composition remains.
What the film compresses into seconds is a sustained process of research, drawing, positioning and adjustment. A finished print gives little indication of how many decisions were necessary to make its complexity feel coherent. The geography comes before the styling. I begin by sourcing the information that describes a region. Elevation is fundamental, but it is only part of the material. Water, land use, settlement, roads, railways and other features contribute to the picture.
Depending on the geography, that material draws on sources such as NASA’s SRTM elevation data, OpenStreetMap and national mapping agencies. Those sources provide different kinds of information, at different resolutions and levels of completeness. Selecting and preparing them is part of the work.
The geographic extent and intended scale matter from the beginning. A feature that deserves attention in a local map may be impossible to show meaningfully in a regional composition. Including everything available would produce congestion rather than understanding.
My background in urban planning informs what I look for: how settlements occupy the land, how routes connect them, and how natural features remain legible within places we have extensively altered. Those relationships shape the map before a final palette or title is chosen. Terrain has to be developed. In ArcGIS Pro, I assemble elevation data into a terrain model and develop hillshade and contours. These are different ways of describing the same ground.
Hillshade gives slopes and ridges a sense of relief. Contours make changes in elevation readable through linework. Colour helps differentiate the landscape, but needs to sit comfortably with both.
None of these elements can simply be made as emphatic as possible. Strong shading may describe a mountain beautifully while swallowing the names and routes around it. Dense contours can give useful texture or become a dark mass. The work is in finding their appropriate expression together.
That expression changes between regions. The mountain, lake and forest structure of Tasmania presents a different compositional problem from the coastal-to-inland geography of Durban to Berg. Both need terrain, but neither benefits from having the other’s treatment applied without judgement.
The intention is to give the land a convincing presence within the artwork. It should remain part of what the viewer reads, including where settlement and infrastructure become dense. Every layer affects its neighbours. The physical and human geography is prepared layer by layer. Roads need a hierarchy. Railways must remain distinguishable from waterways. Settlement and land-use colours need to explain something without breaking the region into competing patches.
A motorway and a minor track should not carry the same visual weight. Equally, making every major road dominant can leave the natural landscape looking incidental.
I style these elements individually, then export the prepared layers for composition in Photoshop. This is where much of the artistic resolution takes place. I adjust colour, opacity, blending and layer order, repeatedly looking at what each change does to everything around it.
The answer is rarely contained within one layer. If a river disappears, making it brighter may solve the immediate problem while creating another elsewhere. The terrain beneath it, the surrounding land-use colour or an adjacent line may need attention instead.
Across the British Columbia and Washington artwork, water, islands, settlement and mountain systems all need a readable presence. Let one dominate indiscriminately and the relationships between them become harder to see.
The aim is a composition in which the geography can be explored without every feature demanding attention at once. Names are researched and positioned by hand. Place names and annotations are not a finishing decoration added after the serious mapping is complete. They are part of how a viewer understands the work.
I research and position labels manually, considering their size, spacing, prominence and relationship to the feature they identify. A name needs to be associated clearly with its place while remaining readable among surrounding contours, routes and other names.
A technically available position is not necessarily a good position. Text can obscure a junction, interrupt a coastline or appear to belong to the wrong feature. Resolving that requires attention to both geography and typography.
The process involves moving between close inspection and the whole composition. A label that works in isolation may be too prominent when the full region is visible. Reducing its size might help, or the surrounding hierarchy may need to be reconsidered.
This is a substantial part of what I mean by work composed by hand. The computer is indispensable, but placement and emphasis are decisions I make and revisit. Different kinds of map ask different things of their makers A graphic built primarily around a street network can be effective. It may celebrate an address, a neighbourhood or a recognisable urban pattern. Its selective treatment is part of its purpose.
A reference map has a different task, often organising extensive information so that places and routes can be found efficiently. Such maps can involve considerable expertise. Detail and terrain are not exclusive to art.
My own purpose is to bring a region’s natural and human geography into a composition that can be lived with and repeatedly explored. That requires decisions beyond selecting an extent, applying a colour scheme and exporting the result.
The number of copies printed is a separate question. My editions reproduce completed original artworks. The geography is not newly drawn for each order; the authorship lies in developing the master composition from which those prints are made.
Generative imagery introduces another distinction. An image can resemble a map without that resemblance establishing the location or relationship of the features it depicts. Cartographic appearance alone is insufficient evidence of real geography.
My cartographic artworks are created without generative AI. They begin with geographic sources, and I take responsibility for how that material is selected, interpreted and presented. That is not a claim of infallibility or a promise of navigational suitability. It is an account of how the work is made. The artwork must hold together at two distances; Across a room, the viewer should be able to read the overall composition: the shape of the region, its terrain, its water and the broad patterns of settlement.
Closer inspection should offer a different experience. Names become readable. Routes can be followed. Smaller features emerge without requiring the whole image to be understood at once.
These are two demands on the same artwork. Enlarging every label would undermine the distant view. Suppressing everything into an attractive texture would impoverish the close one.
That balance carries through the 34 Degrees Design collection, across different regions and formats. Giclée printing then reproduces the resolved artwork as a physical piece. The printing matters, but the composition has to work before it reaches the printer.
When the film brings the layers together, it makes the sequence visible. The harder part to show is the repeated adjustment between those stages: a line quietened, a name repositioned, relief brought forward, a colour reconsidered.
Those decisions are what allow a complicated geography to become something you can keep looking at.
