It's an interactive 3D map of commuting data from the 2018 New Zealand census.

These data show the numbers of people making trips between a home locations and a usual place of work or study on census day. Data are at the level of Statistical Areas, roughly equivalent to neighbourhoods.

These connections are shown as 'ribbons' whose colours relate to the purpose and transport mode, and widths to the volume of trips.

To navigate the 3D map use mouse, keyboard arrows, or touch gestures. Most operations are as you would expect, but changing pitch and is done using CTRL or Cmd mouse-click, or right-click and drag. If you get lost, use a tour desination to find your place again, or if an area is selected focus it with the ⚲ button in the chart view.

Clicking on an area will display a chart with details of its overall transport mix with transport modes as car (any private road-transport), transit (bus, school bus, train, ferry), or active (walking, cycling, other modes).

To deselect an area, click on it again or use the chart's ⨂ button, and the view will switch back to showing trips for all areas. You can also focus on the selected area using the ⚲ button.

Change which trips are displayed using the Trips filters.

Origins of trips (i.e. where people live) are blue-ish for work and green-ish for study trips. Trip destinations are pink-ish for work and orange-y for study.

When there are both work and study trips between places the colours mix accordingly. Major centres of employment appear bright pink and major centres of education bright orange. Mostly residential places appear blue-green.

Ribbon widths change based on the total number of trips included using the Trip filter controls.

You can switch on and off different trip purposes (work/study), trip modes (car/transit/active), and also limit the range of trip volumes, or point-to-point distances displayed.

Changing the ranges of trip volumes and distances shown can help clear map clutter so you can focus on a particular place.

Distances are measured point-to-point from area centre to area centre, so large rural areas, they do not reflect actual distances travelled very well.

Depending on your screen and viewing conditions you might prefer a dark or light basemap (the other colours stay the same).

Use the ◐ button to switch, but be warned, the whole page will reload, which might take a little time...

The site defaults to the light theme, because it seems to work better in more viewing conditions, but the dark theme is very cool to look at it if it works for you. Bookmark one of the links below to ensure you can immediately revisit the version you prefer in future!

• Light theme (the default)
• Dark theme

This was built as an entry to the There and back again contest announcement by Stats NZ - Tatauranga Aotearoa.

Many free tools and resources were used including—but not limited to:

• deck.gl and Mapbox GL JS for the overall 3D visualization framework,
• R, RStudio, mapshaper, and QGIS for data tidying,
• basemaps and data from Carto and OpenStreetmap,
• Plotly for the charts,
• the neat and clever noUiSlider and autocomplete widgets, and
• numerous answers and code snippets at W3Schools and StackOverflow.

I learned a ton building it. Enjoy! David O'Sullivan.

Kaikoura is isolated, with larger flows only between Kaikoura itself and the Ranges. Switch to the Trip filters tab to experiment with switching between work and study trips. You should see work trips go from blue to pink, and study trips from green to orange.

There are more trips from the Ranges to Kaikoura than in the other direction. Flow direction is also shown by the shape of the arcs which swing out left and back in the direction of travel.

If you switch among Car, Transit and Active modes you'll see that most trips are by car, with only a small number on transit (probably school buses).

If you click on either Kaikoura or the Ranges this help will be replaced by a chart showing total trip volumes by purpose mode, and direction (flip back to this help using the ⇆ View controls button). The chart also shows local trips, where home and place of work/study place are inside the area, and home, which is people working or studying from home. The local and home numbers tend to be proportionately greater in large rural districts like Kaikoura Ranges.

Almost 20,000 people travel to work at Auckland Airport, nearly all of them in cars. Use this stop to get a feel for how to reduce clutter in crowded parts by filtering trips, to see local details. This example is good to explore using the Volume and Dist(km) filters.

First ensure that all distances are visible by dragging the upper and lower limits on the Dist(km) filter to the limits. This may seem to make no difference if the lower limit on trip volumes is above 0. Change this by dragging the low number on the Volume filter. You should see even more inbound work trips to the Airport. Zoom out a bit and you'll see that these are from as far as Aotea - Great Barrier Island to the north, and Huntly to the south!

If you switch between work and study trips you'll see almost nobody goes to school at the airport, and that these trips are fairly local.

The biggest clusters of study related travel tend to be the universities, and unlike work travel, these trips are often distributed across modes.

The first thing to notice about the Campus South trips is that the single highest volume flow in Aotearoa New Zealand is into this area, with 2526 work and study commutes from Gardens (Dunedin City). Try using the Volume filter to view only this flow. The impressive thing is how many of these trips use active modes, walking, biking, and... we can assume, skateboards, scooters and probably roller skates. Look at the chart and notice the contrasting mix of transport modes between the work and study trips.

Something else to note here is that the scaling of the width of ribbons is not linear, but instead limits the width when zoomed in. This is because viewing a volume of 2500 trips and one of just 6 at the same scale would not be practical.

Paekakariki is an unusual case in Aotearoa New Zealand as a place with a reasonable balance between car and transit based commuting.

This is as good a place as any to note one difficulty with the data. You might notice in the popup tables for ribbons, that the numbers don't add up! If you are particularly perceptive you'll see that all the numbers divide exactly by 3... and no... it's not because Kiwis travel around in groups of 3!

Both effects are because Statistics New Zealand confidentialises the data. Numbers under 6 are suppressed, and all others numbers are rounded to a near number divisible by 3 (so 37 might become 36 or 39). For more on this, see here. This makes small numbers in the data unreliable, and a lot of the zeros might not be zero at all! In this map data have been grouped. For example the Active category includes walking, cycling and other modes. It's' that all were 5 in the raw data so that 15 trips have 'disappeared' from the map.

When non-car modes of transport are potentially so important for a low-carbon future it's unfortunate we don't have better data to keep track of progress, even if it is for sound, privacy reasons.

Canterbury, and the next stop on the tour (Auckland region) show the contrast between a monocentric urban region and a polycentric one. Looking at Canterbury it is useful to switch off work trips, when regional centres come to the fore with many inbound school trips and the overall hierarchy of places in the region is clear.

Turning work trips on (and study trips off) the picture is less clear, but the dominance of the region by Christchurch City as a centre of employment is very apparent.

By contrast with Canterbury, the picture in Auckland is chaos! Looking first at education, the University dominates. travel patterns. If you reduce the maximum trip distance included in the view, you can see the travel patterns associated with schools emerge.

Turning work trips on (and study trips off) the polycentric structure of the region is clear—people are travelling everywhere to everywhere! There are major centres of employment at the Airport, in the CBD (represented by several areas in this map), Penrose, Manukau, New Lynn, North Harbour, and others besides. This helps explain the traffic mayhem everyone talks about so much!

It might seem strange to compare the Wanaka-Queenstown-Cromwell-Alexandra area to Auckland Region or Canterbury, but from the perspective of trips to work and school there are some similarities! There are multiple centres, with no one centre dominant, even if Queenstown is clearly the busiest.

There aren't many cities in Aotearoa New Zealand as remote as Gisborne and it shows in the patterns of trips in and out of Gisborne Central—a single center of both work and study.