Field Observations of Sea Ice Thickening by Artificial Flooding

[Geoengineering News]

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2025JC022738

Authors: T. C. Hammer, L. L. van Dijke, A. Shestov, C. Haas, H. Hendrikse

First published: 08 April 2026

Abstract

Arctic sea ice is retreating at a high rate, also due to the positive ice-albedo feedback loop: as ice melts and disappears, it reflects less sunlight, further accelerating ocean warming. One proposed way to slow the retreat is by thickening sea ice in winter, increasing its chances of surviving summer melt. This could be achieved by artificially flooding existing sea ice with seawater pumped from below, allowing it to freeze at the surface through exposure to cold air and thicken the ice layer. However, the effectiveness of this approach remains uncertain, as numerical models show contrasting results and few field experiments have been conducted. This study examines the growth and melt of ice through spring and summer after artificial flooding covering ~1500m^2, resulting in thickened (+26 cm) snow-covered first-year sea ice. Observations were carried out in Vallunden Lagoon (Van Mijenfjord), Svalbard, from 20 March to 24 June 2024, with flooding and intensive in situ measurements from 11–15 April. Artificial flooding significantly heated the upper two-thirds of the original 90 cm thick ice, increasing salinity. Surface albedo evolution was influenced by specific events such as slush formation, snow drift, and a major meltwater drainage event in spring. Artificial flooding resulted in thicker ice and delayed rotten ice formation by 6 days, but did not delay the disappearance of ice in summer compared to a non-flooded reference site. Experiments at other scales and locations could help reveal how local conditions and flooded area size influence results and the potential of this method.

Plain Language Summary

As Arctic sea ice melts and disappears, less sunlight is reflected from the Earth's surface, accelerating ocean warming. One proposed way to slow this process is to artificially thicken sea ice in winter, increasing its chance of surviving into summer while reflecting sunlight. This can be done by flooding ice with seawater when air temperatures are far below zero, allowing it to freeze at the surface and add thickness. However, the effectiveness of the method remains uncertain because few real-world tests have been performed. This study tested artificial flooding over ~1500m^2 of sea ice in Svalbard in 2024 April, monitoring it remotely until 2024 June. Ice temperature, salinity, and solar reflection were measured before and after flooding. The results show flooding warmed most of 90 cm thick ice and allowed salty water to soak in. Events such as snow drift, slush, and a major spring drainage affected how much sunlight the surface reflected. Although flooding produced thicker ice (+26 cm) and delayed rotten ice formation by 6 days, it did not make the ice last longer in summer than nearby non-flooded ice. Tests at other scales and locations could reveal how local conditions and flooded area size influence results.

Source: AGU