Between 2009 and 2012 I worked on addressing a data gap for supporting lidar surveys on drifting pack ice. The issue was that no ‘good enough’ data existed to account for ice floe drift and rotation between passes for aerial surveying instruments, so any alignment of data between different passes over the same spot was aligned manually – usually in an ad-hoc fashion.

My colleagues and I developed a new system using two GNSS receivers and a robotic total station to set up a basline on a drifting ice floe which formed the ‘target’ for multiple on-ice observations covering multiple disciplines. The robotic total station was then used to survey in key points, and monitor/maintain the baseline on a routine basis.

I developed a battery box system to power the GNSS instruments, using converted cooler bins acquired secondhand from an outdoor education organisation. On advice from glacial geodesists, we also developed a ‘foot’ system, timber platforms supporting tripod feet buried in snow. This was critical, we often encountered snow deeper than 1m, so digging to ice would have been completely impractical.

Finally, I developed an internal guy system using a bamboo deadman between each tripod leg pair, anchoring to the opposite tripod leg. This ensured easily adjustable tripod securing system without occupying a lot of space around each tripod – seen in the feature image.

I was responsible for deploying the system on the SIPEX-II expedition in 2012, leading a small team of volunteer helpers drawn from the pool of researchers on the ship. Using this system we were, for the first time, able to map all on ice activities – essential if we want to look closely at relationships between observations.

It also supported the intended purpose: alignment of multiple airborne datasets, on ground data and even under-ice data!

In turn, this permitted analyses like the following plot – showing ice draft (how far ice pokes into the ocean) observed using upward looking sonar mounted to a free swimming AUV (tracks shown in cyan on the ice floe map), ice draft modelled from lidar surface elevations (and some inferences about ice density, snow thickness and snow density), and the differences between observed and modelled ice draft.