Pricing the interannual variability in marine accessibility on the Northern Sea Route

A potential shift from prevalent frozen ice regime to a seasonally navigable ocean is likely to arise interests of decision makers and investors to seek opportunities for global trade through the Arctic. While assessments of such economic activity are mostly made on the basis of a progressive linear rate of sea ice retreat that is fundamentally deterministic, the estimates would almost certainly undervalue financial risk in future Arctic investments since actual ice trajectories are subject to inter-annual variability. In this study, we use the Black-Scholes model proposed by Sturm et al. (2017) to price the sea ice edge variance in the Northern Sea Route (NSR) and the response uncertainty in Global Climate Models (GCMs) by evaluating the expected marginal cost of using a longer shipping route via the Suez Canal when a shorter route (NSR) becomes inaccessible due to sea ice variability. We show (1) the central role of ice variance in explaining the potential risk of using the NSR as compared to ice averages, with a standard deviation of detrended season length being 13.5-29.3 days and 5.2 days, respectively. The latter underestimates the risk by up to 9.6% of the needed trade volume; (2) the additional cost responds more strongly to parametric uncertainties that directly influence human activities as indicated by close association between standard deviation of detrended season length and that of detrended ice concentration; and (3) the cost of variance is jointly affected by global trade demands, model structure and parameter values. A nuanced consideration of such relationship, their tradeoffs and relative importance could save billions of dollars for business decisions to Arctic sea ice change.