Geographers - expeditions need you!

It was some time in October 2016, and my friend James asked me if I was up for an adventure. Without hesitating, I agreed. We had met at the Royal Geographical Society, at the annual ‘Explore’ weekend several years ago. We had stayed in touch, become good friends, and the previous year I had joined an expedition that he had led to Madagascar. There we studied the forest ‘edge effect’ i.e. the effect of proximity to forest edges, on amphibians and reptiles. On that expedition I had been appointed GIS and Data Specialist - a role not dissimilar to my then-day job with Amphibian and Reptile Conservation in the UK. I think it was somewhat to James’ surprise when it turned out that I knew relatively little about herpetofauna, but my saving grace was that I did know about GIS. Evidently I had proved my worth because now - one year later - James was inviting me on another expedition. This time it was botanical. This time it was Oman. I had just enough leave left to take the three weeks that I needed to go. I began furiously downloading data and preparing maps for the next adventure…

How do you use GIS to prepare for an expedition?

I bought a laptop specifically for the expedition - a Lenovo ThinkPad X220 - which is still going strong having now survived the montane forests of Madagascar as well as the burning desert of Oman. On the previous expedition to Madagascar I had used Google Earth extensively, and I stand by it as a very reliable tool for expedition. Google Earth allows you to very easily cache (but not download) satellite imagery and elevation data for use offline - which lends itself to remote expeditions. It also consumes common files like GPX, KMZ and SHP. I used Google Earth to navigate to our chosen fieldwork location, make dynamic logistical decisions on the go, and to manage and visualise points of interest and all of our GPS data. Having a system capable of managing GPS data from many sources, I was able to maintain a centralised GIS or ‘geographic information system’. Points of interested included the locations of our forest transects, navigational aids such as ‘path through swamp’ and the locations of important animal sightings. I was then able to distribute these locations back to all of the field GPS units every night, ensuring that every expedition team member had the latest spatial information necessary to manage the fieldwork effectively.

I also used the free and open source GIS software ‘QGIS’, which allowed us to visualise data from more diverse sources, and to analyse this data to tell us things about the environment. About one month before the expedition we met at James’ house in London, and I was already able to show to the team the area that we were going to in great detail. On my little field laptop we pored over elevation models, satellite imagery and environmental data, all of which helped us to put our study site in context within the wider environment. I downloaded some OSM data showing roads and settlements, which I overlayed on top of the satellite imagery and the elevation model. These maps I also printed out to take into the field, and they proved invaluable for navigating to our remote field study site which lay beyond navigable roads, ox-cart tracks and mountain paths. I realised during that expedition that we could be doing much more with the OSM data - we could contribute to it, as well as benefit from it, and I resolved that on my next expedition I would aim to do just that.

What do expeditions stand to gain from OSM?

So it was that while preparing the maps and GIS for the recent expedition to Oman, I did a few things differently. First of all, I asked the other members of the expedition team to do some mapping in their spare time. I explained that by creating an OSM account they could map roads and settlements from the aerial imagery, and that I would download this data onto the laptop just before heading out into the field. This way, everyone on the team could actively contribute to our mapping resource before we’d even left home. Unfortunately, none of the team actually did this, and so on the next expedition I plan to run a quick intro session to get everyone set up in OSM and demonstrate just how easy it actually is to map in the iD editor.

The night before departure, I downloaded the entire OSM vector dataset for Oman onto my laptop. This may sound like overkill, but last year in Madagascar we had had to change our study site at the last minute, and all of the map data that I’d downloaded was suddenly redundant. Two of the team had then been forced to make an arduous day’s hike across the savanna in order to get enough mobile signal to cache the Google Earth imagery for our new study site. This time I wasn’t taking any chances. This time I was using ArcGIS, and I set about symbolising some of the key features such as roads and cities. I believe there’s an ArcGIS style file for symbolising OSM data in ArcMap, but I wasn’t able to get this to work in time for the expedition, so that’s something I’ll work to resolve next time.

When we arrived in the field, I began updating our OSM data based on local knowledge. There was one day in particular when we were stranded in Al-Jazer awaiting the fuel tanker to come and replenish the petrol station. We were sitting outside a coffee shop across the road from the petrol station and I got talking to a local gentleman named Said. It wasn’t long before I had pulled out the laptop to show him where we were going and where we had been, with the aid of the OSM data and the satellite imagery which I had also downloaded in preparation for the trip. He pointed excitedly at the fishing settlements that punctuated the coastline, calling out their names one after another. He would spell them out while I typed them each into the map, and then make me say them repeatedly until I had perfected their pronunciation. It was evidently a source of great pleasure to him to see the map updated in real time with the place names that he was providing me. Sometimes I wasn’t quite sure of Said’s information, and I would verify the place names with the two young English-speaking coffee shop proprietors.

As well as mapping roads which don’t yet exist on the map, it can be just as useful to remove features which have been added erroneously, or to update information about existing features. One afternoon James didn’t meet us at our agreed rendezvous. We couldn’t know for sure which direction he would arrive from, and eventually we decided to set up camp at this location and await his arrival. Darrach, one of our collaborators from Oman Botanic Garden, went off in search of him and by a great stroke of luck was able to make contact. Later that evening, James arrived back at camp with a truckload of weary botanists. They had followed a road on the map which turned out to be unsuitable even for their 4x4 vehicle. After this realisation, they had been forced to make a great detour to get back around to our rendezvous location. We’ve now updated that road to prevent others from falling into the same trap.

How can expeditions contribute to OSM?

After returning home, we are planning to sync our changes back to the OSM server. Anyone looking at this remote region of Oman will have access to better map data, and as a consequence will be better informed about the region’s geography. On future expeditions, I’ll be using OSM again, and I’ve been advocating this to friends and contacts who are planning their own expeditions. Every day, expeditions are going to remote corners of the globe. What an incredibly powerful tool we have here; to be able to openly access the latest geographic information and to contribute back to it for others to benefit from. I’m convinced that OSM is the future of global cartography; get mapping!