Thinking of a career as a Geographer?

Geospatial analysis requires a range of complementary technical skills. The more of the following skills you can learn and nurture, the more effective I find your geospatial analysis will be:

GIS – Geographic Information Systems (or Solutions), are software programs that enable geospatial analysis. Esri’s Arc Pro is the best known proprietary tool, QGIS is the open-source and free to use alternative. I personally use QGIS to do most of my GIS work, though familiarity with both is helpful, as each has its strengths and weaknesses. When it comes down to it, both can do 90% of GIS processes very well.

Remote Sensing – is the science of remotely studying phenomena on the Earth’s surface, whether using camera traps and drones, light aircraft flyovers that capture aerial imagery, or satellite sensors that capture great swathes of imagery. I use camera traps to study animal behaviour, drones to capture high-resolution imagery of the landscapes I am studying, and most commonly, a range of satellite imagery sources to measure change in landscapes over time. You’ll encounter some physics and statistics whilst processing imagery, whilst some creativity is needed to depict and communicate what are highly visual outputs. Remote sensing capability has dramatically improved over the last 10 or so years with the advent of cloud computing, largely replacing software programs such as Envi and Idrisi with browser-based platforms such as Google Earth Engine (GEE). Which brings me nicely onto…

Coding – at the heart of geospatial analysis is data. Efficent analysis of large and regularly updated datasets requires coding skills, in what are known as high-level programming langauges (user-friendly and closer to human language), such as Python or R. Google Earth Engine’s web-based platform uses its own Earth Engine language which is based on JavaScript, but GEE can also be accessed in QGIS via a plugin using Python. GIS and coding, or Remote Sensing and coding are highly complementary skills.

Field Survey – nothing beats getting into the field to build that intimate familiarity with the features or phenomena that you are trying to map. For remote sensing, ground-truthing is an important part of calibrating machine-learning based models (this sounds scarier than it actually is!), which means visiting and verifying on the ground examples of the different land cover classes you are mapping, whether tropical rainforest, cropland, rivers, or even football pitches! I’ve been lucky enough to visit the beautiful mangrove forests of Madagascar and Singapore to improve and refine my mangrove maps.