30 Day Map Challenge 2022

Following the bicycle theme; a quick grab of Brisbane City Council's Open Data spatial datasets for the cycle plan network and street parking lines.

A very crude analysis ensued, with a comparison of kms of bikeway versus space for parking. Unsurprisingly, many more kms of roadspace are given to parking than to prioritising safe off road cycleways.

Data is somewhat incomplete/nuanced and open to interpretation. Some aspects I've noticed:

• Parking lineway dataset is increasingly incomplete the further out from the inner city area (ie there is plenty of designated street parking in neighbourhood streets that isn't captured as lines). Parking line dataset hasn't been filtered by type, but generally represents kms of streets with space for parking.

• Bikeways dataset has been filtered to priority/safe bikeways (as opposed to 'just brave it with cars' routes). Whether this is the right category to compare with parking is probably questionable.

Nonetheless, a fun little exercise in data exploring.

A slight departure from bicycles to have a play with the  Queensland Police Crime Data AP I.

I ran a query in  Postman  to first extract crimes for period 1st January 2021 to 1st January 2022.

I joined the JSON to the 2021  ABS Meshblock boundaries  based on meshblock code and filtered for the most violent crime types (assault, homicide or homicide (other)) and counted occurences of these crimes by meshblock.

Data hasn't been normalised by population or area as I....kind of ran out of time to figure out how to do this dynamically in  MapBox Studio , so displayed are the raw counts of crimes.

This one was a bit revelatory to me in that I didn't expect to see higher numbers of violent incidents in hospitals - but as a wise colleague pointed out, often people are at hospitals during the lowest/most difficult times of their lives.

A frog map!

This one was a quick subset of the WildNet public species data from the  Open Data Portal 

I filtered it by the Family corresponding to toad and frog species and customised a  MapBox Studio  style using the sum of counts at each sighting.

Nothing fancy on this one, just a visualisation.

Had grand plans for this one, but sadly ran out of time and got overwhelmed with the choice of ideas and data options.

In the end, I stumbled across a list compiling the countries that have banned flights from Russia (and have been reciprocally banned) and opted to hastily transform these into a map.

Not the best map I've ever produced, but an interesting insight into the politics and ramifications of bans.

More information on flight ban situation here:  https://ops.group/blog/ukraine-russia-update 

And list of countries maintained at  MAK Aviation Services 

As a quick refresher in digitising skills, I traced linework corresponding to what used to be the tram and trolleybus network in Brisbane until the late 60s.

The maps I used as a reference can be found here:

 Wikipedia 

 Tundria 

Some wonderful history and archival material is available online in various caches:

 ABC News 

 State Library of Queensland 

 Queensland Historical Atlas 

 Brisbane Tramway Museum 

Aka a steep learning curve on using  Blender  and DEMs and limited time.

The DEM source is from the global Copernicus DEM, accessed through  eurostat 

I ran out of time to figure out a whole raft of things, but really enjoyed getting to know Blender.

I'd never before made a map in  QGIS  using  OSM  data, so this was a super exciting mini-dive into how to do it.

I initially wanted to do a map of all features that came under the 'castle' category in Australia (because why not), but unfortunately there were only 3 of them :(

So I defaulted back to a familiar theme - bicycles :) This time I queried the data for bicycle repair stations and any linework corresponding to cycleway to create a basic map with a  Stamen  basemap background.

All credit to the  OpenStreetMap  contributors and  Stamen Design .

The final front ear.

Sort of. For this one I browsed NASA's open data offerings and stumbled across this Meteorite Landing dataset.

Seemed like a fun thing to try and visualise so after some minor data wizardry (removing records with null or 0,0 coordinates or no mass /year values) I then loaded it into  Kepler.gl  and  Flourish  to explore their offerings.

Delightfully atrocious, thought to try and map the locations covered by the tune 'I've been everywhere', of which there are multiple editions.

List of locales was snaffled from  Wikipedia 

There are several versions/iterations of the tune/locales, and I'm currently still trying to decipher Athol Wightman version from this  recording 

My progress so far can be found here, all advice welcome.

Inspired by this work of the utmost scientific rigour, I thought to render the map in  figure 1  of the paper 'Indirect Tracking of Drop Bears Using GNSS Technology' in a colour that more accurately instils the emotion associated with the creature in question.

Further detailed information and images on the drop bear can readily be found online through your search engine of choice.

This theme evokes a genre of request that many a spatial analyst has been presented with during their careers.

Client:

Can I please have a map showing all the lots (labelled) in for my meeting this afternoon? I need it in A4 so it fits in my folder.

Me:

For this one, I grabbed a data-source I was vaguely familiar with - the  Queensland Traffic API 

I grabbed the  geojson file  directly in the browser (representing a point in time), downloaded/saved and then added to an ArcGIS Online map and quickly symbolised the points and lines.

Quick and simple one today.

I was keen to try this in QGIS, so I followed a tutorial by Maria Ruehringer  here .

The data is from OSM and contributors.

The one thing I ran out of time to sort out was the legend/classification categories - QGIS symbology seemed to keep defaulting to shared upper/lower boundaries when... Mental note to find out why later/next attempt.

And yes, there are many more cartographic adjustments I'd have liked to make... but time.

Food is never far from thought... in fact, I eat as I type this.

For today's challenge I'm taking the slightly lazy option and rehashing a small piece of work I did in another role quite some time ago, as a bit of a value-adding exercise during a quieter period.

An important part of the food paradigm is the collecting (harvesting) of it. In Australia, labour to carry out harvesting follows seasonal patterns of crops and is intimately linked with labour from overseas - namely back-packers and a handful of specialised labour programs with a number of nations in the Pacific region.

To aid/attract backpackers in particular, the government produces something called the 'Harvest Trail' - a booklet of information on where back-packers can go to find harvest work and when.

Details can be found  here  and an example of the current 'original' PDF of the information can be found  here .

I essentially digitised locations that appeared either in the PDF or in the online map, and added (where I could find it) the crop(s) and anticipated labour demand by month.

The idea was to try and visualise labour demand spatially across the country.

Anyhoos, here is the raw/rough result from several years ago, overlaid on the  ABARES  land use cover raster, filtered for agriculture.