AMPHIBIAN: Twelve Days From Greece's Northern Edge to Its Southern Tip, With the Body as the Laboratory

Two timings, two bodies of water, two lines. I put them side by side in a spreadsheet, divided, and the result stopped me for a few seconds. 101 kilometres in 28 hours 16 minutes — 3.57 km/h. 34 kilometres in 9 hours 20 minutes — 3.64 km/h.

Seventeen years sit between them. The shorter one is the English Channel, 2026, England to France. The longer one is the open Aegean, 2026, the Peloponnese to the coast of Chania. Different salinity, different currents, different water temperature. Between the two attempts the body aged eleven years. The output is nearly identical to the second decimal.

A coach reads that as coincidence or as a training signature. An analyst reads it as a reason to open the rest of the file. Because next to that name there are more lines, and the next line is called AMPHIBIAN.

What the project actually is

Start at Greece's northernmost edge, finish at its southernmost point. Ormenio to Gavdos, the southernmost point of Greece and of Europe, with the country's highest ground crossed on the way. The route cuts through all 13 administrative regions.

Movement is not one sport but a rotation of five: cycling, swimming, mountaineering, running and sailing. The operational window is 12 days. Alongside sit fellow athletes, researchers, physicians, technologists and a field team. Everything is streamed at amphibian.online.

In the project's own language there are two routes. The visible one is the drawn line from Ormenio to Gavdos, which anyone can watch. The invisible one is what changes inside the body each day, which no one can see.

The stated purpose is precise: the crossing is not the goal; the crossing is the operational frame that makes it possible to study physiology, fatigue, adaptation, recovery and environmental impact, while testing wearables, smart garments, GPS, telemetry, and the entire chain of data collection and visualisation.

Behind it sits the support of the Ministry of Digital Governance and Artificial Intelligence. Funding to the Hellenic Heritage Foundation was earmarked for the second phase of integrating artificial intelligence into VR and AR.

The data architecture, and one uncomfortable question

Read the variable list straight through and it is hard to look away. Cardiovascular and respiratory function. Thermoregulation. Oxygenation. Glycemic dynamics. Movement. Work output. Fatigue. Recovery. Eight variables, one body, twelve days, five distinct environments.

My whole career has lived in that space. As a data consultant at Chittagong Abahani in 2026 I logged 22 matches, coded 1,148 defensive actions and built a PPDA model. The club pressed at 14.2 PPDA in the first fifteen minutes and dropped to 21.6 after the seventieth. A coach told me tactics were not a woman's department; the club adopted the one-page protocol that October. PPDA was never a ledger to me; it was a contract written with chaos.

So my first question about AMPHIBIAN is not about collection. Collection is a solved problem. The question is the fourth constraint: connectivity. A pool has wifi. The middle of a sea, a mountain slope, rain and salt air do not. On day seven, when both the body and the telemetry are tired, the real test is how honest the buffer-and-reconcile model turns out to be.

What the project states plainly is this: whether such data can be transmitted, stored, visualised and reliably interpreted in real time despite movement, weather, water, terrain and unstable connectivity. It is a working model of out-of-lab telemetry.

Why this subject

The centre of the project sits on a person whose file demands a specific kind of reading.

Georgios Tsianos was born in Athens with Thessalian roots, educated in Florida, took a BA in human physiology at Berkeley, an MSc at King's College London in human physiology in adverse environmental conditions, and a PhD at the University of Glasgow on human physiology at altitude and in cold, with field work in the Scottish Highlands, the European Alps and the Himalayas. He later completed his MD at the University of Ioannina, trained in general practice, emergency medicine and trauma surgery, with experience in South Africa, the United States, England, Scotland and Greece. He is certified in expedition and travel medicine, works professionally in remote parts of the Scottish Highlands, is an honorary lecturer at the University of Thessaly, and teaches human physiology in adverse environments on the Applied Kinesiology in the Armed Forces MSc.

The athletic file is sharper still. He began in swimming, competed at world and European championships with the national team, was a Panhellenic champion and record holder, and a Balkan champion. He was the first Greek to compete at a world open-water marathon swimming championship. In 2026 he swam the English Channel in 9 hours 20 minutes, the fastest time in the world that year, earning the Rolex award from the Channel Swimming Association. In 2026, on the Hellas Everest 2026 expedition as scientific adviser, first-aid officer and climbing member, he became the first Greek to summit Everest via the north Tibet route. In 2026 he swam 101 kilometres from the Peloponnese to Chania in 28 hours 16 minutes, the first human to cross the open Aegean. In 2026 he finished the Marathon des Sables: 250 kilometres, six days, fully self-supported. In 2026 he summited Everest a second time as expedition doctor with a British team. In 2026, as part of a medical team in Antarctica, he swam in the Southern Ocean while logging physiological data.

With those three feats completed — Channel, Everest, Marathon des Sables — he became the first person in the world to finish the Ice Water Fire challenge.

That, for me, is the interesting part. This is not a hobby list; it is a coverage map. Cold water, altitude, heat and sustained multi-day load: four extreme environments, one body, captured across more than two decades. AMPHIBIAN is asking what that body does when all four exposures are compressed into a 12-day rotation.

In physiological terms there are three joints. Thermoregulation is shared between sea swimming and high-altitude mountaineering. Glycemic dynamics is shared between multi-day running and cycling. Recovery is the yardstick for everything else.

Where I stop

The sample size is one. This is a proof of concept, not a population study. Personalisation and generalisation are different things, and the project creates a convenient opportunity to confuse them. The subject himself holds a PhD in the very physiology being measured. He is the best-case version of the sample, and smooth curves from best-case subjects do not transfer to ordinary bodies. I learned that in my own 2026 audit.

That year I was the only woman in the timing booth at the National Athletics Championships in Dhaka. I re-timed archived footage of the men's 100m national record against the official hand-timed mark. The gap was 0.31 seconds. The clock said 0.31, and the whole match changed its mind. Over five months I audited 47 years of federation results and logged 212 men's 100m performances. Two coaches told me a retired athlete should not be doing arithmetic. When the stadium empties, the data can no longer hide behind the noise.

The second discomfort sits on the funding line. The ministry programme cited is about placing artificial intelligence into VR and AR — a visualisation mandate, not a validity mandate. A polished dashboard and a verifiable dataset are two different objects. A dazzling live visualisation can run for five days and still prove nothing scientifically.

The third risk is rhetorical. Northernmost to southernmost, for the first time, thirteen regions — the phrases look good in a headline. On the day the story becomes the headline, the data becomes the footnote. I trust the spreadsheet, but I still audit the story.

What to watch next

The real scoreboard for me is not arrival at Gavdos. It is three questions. First, whether the raw files are published with missing-data flags rather than only the smoothed curves. Second, where the connectivity dropouts went. Third, whether the protocol is documented well enough that someone who dislikes the team can reproduce the result. The best models are janitors: they clean context before they predict. If AMPHIBIAN learns to clean, its real answer will not be the sum of five sports.