UK Polar Early Warning Hub

Data Explorer Tool

Select a storyline to continue exploring the supporting data

Rapid Polar Sea Ice Decline

1. Data Climate Product

2. Dataset

Sub-polar Gyre Collapse

Accelerated Snow Cover Decline

Sudden Reduction in Southern Ocean Carbon Uptake

Welcome to the UK Polar Resilience & Early Warning Hub

Portal developed by the British Antarctic Survey, UK Polar Data Centre

The climate and geopolitics of the polar regions are changing fast. There is an increasing need for government and the wider public to better understand the implications of climatic changes in the polar regions for the UK and the UK’s climate security risks. There are a number of aspects of polar change that are of particular concern for the UK. The UK Polar Resilience and Early Warning Hub is a central resource for providing policy-relevant information on these concerns; in particular the current status, potential future changes and repercussions for the UK. If a major event occurs, or is anticipated, this can help to inform policy responses and communication to the wider public.

The UK Polar Hub brings together UK Research organisations and the Met Office alongside relevant ongoing grant-funded projects to coordinate the availability of the latest key polar datasets and policy-relevant scientific knowledge for informing decision-making in relation to domestic resilience and adaptation needs.

Rapid Polar Sea Ice Decline Overview

What is sea ice?

Sea ice is the frozen ocean: it forms when the sea freezes and floats on the ocean surface.
The areal extent of sea ice varies dramatically between summer and winter.
Minimum in late summer, maximum in late winter.
Sea ice is highly reflective and so reflects large amounts of sunlight during the polar day.
Sea ice is very mobile and so responds to winds and ocean currents, not just temperatures.
The thickness of sea ice is varies throughout the year, by up to several metres.
Since 1978 satellites have provided estimates of sea ice extent.
These data have shown a long-term retreat in the Arctic and a significant reduction in Antarctic sea ice over the last decade.

Sea Ice retreat matters to weather and climate outisde the polar regions

RRS Sir David Attenborough moored at Gromit's Creek, Antarctica 2024/25.
This shows the contrast between thick ice shelves that are previously land ice that has flowed out to sea and broken sea ice.
(credit: Pete Bucktrout, BAS)

Arctic Sea Ice as an Indicator

Sea ice is an important component of the climate system because it regulates the amount of energy entering the climate system from the sun and alters the transfer of heat and momentum between the atmosphere and the ocean. Sea-ice loss is linked to faster warming in the poles than the global average, a phenomenon known as Arctic or polar amplification. The weather at mid-latitudes, where the UK is positioned, is fundamentally linked to the difference in temperature between the cold Arctic to the north and the warm sub-tropics to the south. Therefore, changes in this temperature gradient are a key control over present and future trends in weather and climate over the UK.

Prediction Capability and Research

Approaches to sea ice prediction

Accurate prediction of sea ice conditions can provide valuable insight and decision support at various time and space scales.

Two broad approaches can be used for prediction of sea ice, as is done for weather forecasting. Dynamical models solve physical equations, often derived from first principles, on a three-dimensional grid, to iterate forwards atmosphere, ocean and sea ice conditions. Data-driven models derive relationships between variables (for example, past sea ice and current sea ice) and use this to make predictions. While relatively simple statistical models such as linear regression have long been used in forecasting, data-driven models now include advanced AI models. In 2021, Andersson et al. published the first AI-based sea ice forecasting model, IceNet, which outperformed the state-of-the-art dynamical predictions at lead-times of two to six months. Since then, data-driven models have proliferated with various authors exploring different domains, architectures and input variables.

Current capability of monthly-to-seasonal forecasting

IceNet: a machine learning forecasting system

IceNet (https://icenet.ai/) is a UK-led sea ice AI forecasting system which has previously been demonstrated to have skill exceeding that of dynamical systems for summer Arctic sea ice. The original version predicts sea ice conditions based on recent sea ice cover and atmospheric conditions.

Benchmarking IceNet for both hemispheres

Recently, IceNet’s performance in the southern hemisphere has been assessed for the first time, alongside a re-assessment of its performance in the northern hemisphere.

Models have been trained and evaluated on test periods selected to characterise both ‘extreme’ and ‘normal’ sea ice conditions. These periods were 2007 (extreme low in Arctic), 2018 (‘normal’ in both hemispheres) and 2023 (extreme low in Antarctic). In these test cases, the IceNet consistently outperforms a linear trend forecast at short (1-2 month) lead times, with mixed performance at longer lead times. IceNet shows improvement against a linear trend forecast, although still large errors, for long leadtimes for the 2023 extreme Antarctic case. This is demonstrated in the Figure below; for the forecasts produced at 3 month leadtime (bottom), the IceNet forecast largely follows the linear trend prediction, whereas at 1 month leadtime IceNet is providing substantial added information to the linear trend forecast.

Figure: Example sea Ice ‘hindcast’ predictions from IceNet without ocean variables for Arctic and Antarctic, September 2023. Models are initialised 1 month (top) and 3 months (bottom) ahead. Left hand panels: sea ice edge (15% sea ice concentration contour) in model (red), observations (blue) and linear trend baseline (black). Right hand panels: sea ice concentration error.

Research extensions: ocean variables

Prior research suggests that there may be a source of predictability at either or both poles due to heat stored below the ocean surface (Bushuk et al, 2021); the thickness of the ice; and the salinity (salt content) of the surface waters.

Therefore, ongoing research is exploring the performance of IceNet incorporating ocean reanalysis (ORAS5) variables into the IceNet training and prediction pipeline, alongside atmospheric reanalysis (ERA5) variables and sea ice concentration. Specifically, surface salinity, mixed layer depth, the heat content of the upper 300m and sea ice thickness have been added. Preliminary results suggest that, with the current IceNet architecture and configuration, these variables do not improve prediction skill for the test periods discussed above, or overall performance in training.

Future IceNet extensions: towards operational forecasting

The current IceNet infrastructure uses as input sea ice concentrations from a product which uses discontinued source data from US satellites. This data was discontinued in 2025 meaning it can no longer support future operational forecasting capabilities. Current work is therefore developing the infrastructure to use an ongoing product based on a different data source. This data is higher-resolution (6.25km rather than 25km spatial resolution) and ongoing. The previous product began in 1979 whereas this begins in 2001. The shorter data record provides less data for the AI models to learn from, and it is unclear what impact this will have, if any, on model performance. Existing and new approaches are being explored to optimise skill and move towards operational forecasting capability that can be used to inform UK polar resilience.

Data Provider and Documentation

OSI-420 Sea Ice Index

Provided by EUMETSAT OSI SAF

The EUMETSAT Member States, Satellite Application Facilities (SAFs) are dedicated centres of excellence for processing satellite data. They form an integral part of the distributed EUMETSAT Application Ground Segment. SAFs develop, generate and distribute operational satellite products including long-term data record generation.

Dataset Access Get Dataset DOI Dataset Documentation

Dataset Description:

The Sea Ice Index is a climate indicator of sea-ice coverage. The index covers both the sea-ice extent and area over the last 45+ years and is continuously updated to present time.. both OSI SAF & NSIDC include products based on multiple different satellite sources and both "near-real-time" (i.e. less accurate but timely data for forecasting) and "climate data record" (high quality consistently processed data). You wouldn't calculate long-term trends from the NRT data but would struggle to forecast using the CDR.

Daily records

Daily SIE and SIA values are computed from daily SIC maps. Sea Ice Concentration (SIC) is the fractional area (of an ocean grid cell) that is covered with sea ice. All sea-ice covered ocean is included, lake ice is not. SIE is computed as the sum of the total area of the grid cells that have a SIC value larger than 15%. SIA is the sum of the area of all grid cells weighted by the SIC values (no threshold on SIC).

Monthly records

Monthly SIE and SIA values are computed as the monthly average of the daily SIE and SIA values. The interpolated and extrapolated daily data are used in the monthly average. When interpolation and extrapolation are not possible (e.g. too many missing missing days), the monthly SIE and SIA are missing in the product files. The missing months are April, May, and June 1986, as well as December 1987.

Please select a theme to view mapping features.

Use "View in browser" to load interactive chart.

Use "View in browser" to show ASCII response here.

UK Impacts

Declining polar sea ice is an integral part of shaping how weather and climate over the UK is changing and will change in the coming years and decades.

Current understanding of impacts of polar sea ice loss on UK weather and climate is summarised here. This is divided into two timescales. Longer-term multi-decadal ‘climate’ timescales and shorter-term annual-decadal timescales. Key concerns on these timescales are:

On shorter timescales there is a possibility of a decadal timescale surge in sea ice loss, which could be large and rapid – a Rapid Ice Loss Event (RILE). The causes of short-term surges in sea ice loss combine both climate forcing and natural climate variability, making it challenging to predict timing and magnitude of rapid ice loss events.
On longer-term climate timescales there is strong consensus on overall loss, but the specific rate is uncertain with some climate models showing higher rates of loss than others.

Short-term Change and Rapid Ice Loss Events (RILEs)

RILEs are characterised by abrupt sea-ice reductions over short periods (~3-10 years), which are quicker than the background long-term trend.
There is currently low confidence in predicting the timing of the occurrence of a RILE
Climate models indicate that a recent slow-down in Arctic sea ice decline has increased the chances of a period of more rapid decline in the near future, although not necessarily fast enough to be categorised as a RILE.
As climate change progresses RILEs are increasingly likely to occur due to increased year-to-year variability of the thinner sea ice

RILEs Impact on the UK

The impacts of RILEs on UK weather and climate is a topic of active scientific research. It is challenging to identify the effects of RILEs on lower latitudes, since on the shorter timescales involved there are a wide range of other influences. Framing around the following questions is a useful way to set out what is known scientifically (following Barnes and Screen, 2015).

Could they?

Yes, either through impacts on the north-Atlantic atmospheric jet stream or by ocean pathways. See Long term node for more details on jet stream and resulting UK impacts.

Have they?

An effect is likely there, but it is difficult to ‘see’ due to contemporaneous changes in other modes of climate variability that affect the UK such as El Niño.

Will they?

Future RILEs are projected to be more rapid under climate change and therefore have a bigger effect but, as mentioned above, their impact needs to be considered in the context of a range of climate influences on the UK.

Resources

Causes and consequences of mid–21st-century rapid ice loss events simulated by the Rossby centre regional atmosphere-ocean model

Paquin, J. P., Döscher, R., Sushama, L., & Koenigk, T. (2013)

View Article

Long-term Change and Sea Ice Decline

Polar sea ice is projected to decline significantly over the 21st century under continued global warming. These estimates are based on projections from many climate models. Although there is broad cross-model agreement that sea ice will decline, different models show differing amounts of decline. Therefore, best estimates of the amount of sea ice loss sit within a range between relatively low sea ice loss to high sea ice loss. This range of possible outcomes in sea ice decline is essential to explaining uncertainty in projected future changes in the North Atlantic jet stream and implications for UK weather and climate (Screen et al., 2025).

Multi-decadal Arctic sea ice decline is a major contributor to Arctic amplification – the Arctic is warming 3-4 times faster than the global average – with its exact role a matter for ongoing research. For example, climate model simulations show that sea ice loss reduces the Arctic’s ‘pull’ on the North Atlantic jet stream, which consequently shifts equatorward together with the associated mid-latitude storm track.

Key Impacts on the UK

Impacts of polar sea ice loss on UK weather and climate are most prominent in winter in association with shifts in the North Atlantic jet stream. Under future climate change winters over the UK are projected to become wetter and milder overall. However, the strength of these changes will be modulated by the rate of Arctic sea ice decline, as described below using a ‘tug of war’ analogy. Under a storyline of more rapid future sea ice decline, the expected impacts on winter UK weather and climate are as follows:

A more equatorward position and weakening of the storm-track: on average, storms are less intense in the northwest and more intense in the southeast of the UK and slower moving everywhere.
Weaker mean and days with strong winds across the UK.
As a consequence, climate models generally indicate relative drying in the northwest (i.e. less wetting) and more wetting in the southeast of the UK: this spatial pattern is exhibited by both average and days with high precipitation.
There is little associated change in average UK winter temperatures, with a slight warming over England and Scotland and none over Ireland and Wales.

A tug of war on the North Atlantic Jet Stream

The strength and latitude of the North Atlantic jet stream depends fundamentally on the temperature difference between the cold Arctic and warm tropics. Each can be thought of as exerting a ‘pull’ on the position of the jet stream in different directions. Hence, a ‘tug of war’ analogy is often used to describe how changes in temperature can influence the balance between these two factors and result in a shift of the jet stream if one changes more than the other. Under climate change:

Sea ice retreat and associated warming over the Arctic makes it less cold and reduces its ‘pull’.
Low latitude / tropical warming reduces the ‘pull’ towards the tropics.

So both effects weaken under climate change. The relative amount of this weakening is uncertain: some climate models simulate more sea ice retreat, more weakening of the pull and therefore a more southward shift of the jet stream compared to models that simulate less sea ice retreat.

Wider Context

In addition to influencing UK weather and climate, Arctic sea ice loss may have wider impacts on other regions in the Northern Hemisphere, which could potentially be of concern to the UK. Screen et al. (2025) report high confidence for decreased daily temperature variability and cold extremes over northern North America and medium confidence for increased temperature persistence over mid-latitude marine regions potentially affecting the UK, decreased Mediterranean precipitation and increased central European precipitation (all predominantly winter).
Preliminary research suggests additional wider impacts of polar sea ice decline:

Increased occurrence of wildfires over Siberia (Luo et al., 2024).
Increased summer melt and winter accumulation over Greenland and increased accumulation over Antarctica, with implications for sea level (Sellevold et al., 2022; Trusel et al., 2023).
Sea ice loss is associated with a decline in the Atlantic Meridional Overturning Circulation (AMOC), a system of ocean currents that regulates western Europe’s mild climate (Ferster et al., 2022).

In terms of global warming, the Earth’s energy budget has been significantly affected by sea ice loss in recent decades. Global sea ice has lost 13% –15% of its planetary cooling effect since the early/mid 1980s. Polar change separately contributes 5-10% of global warming.

Infographics and Communications Material

A range of communications materials have been developed to provide updates and explain the implications of polar sea ice change for the UK. Those of particular relevance are presented here. Met Office briefing note on ‘The impact of Arctic sea ice loss on winter weather in the British Isles’.

View FAQ

Maps of wind, precipitation and temperature responses to sea ice loss

The effects of polar sea ice retreat on the weather and climate over the UK (and globally) have been isolated using climate model simulations. Some robust effects have emerged across many different studies using many different climate models. The most important source of this information is a major UK-led international study, called PAMIP1, which brought together climate modelling groups from across the world to study the effects of polar sea ice loss.

Here maps showing the impacts of Arctic sea ice loss on North Atlantic winter climate are displayed, based on PAMIP data from 12 different climate models. They show the effects of sea ice loss in isolation, for which it is important to set them in the context of overall climate change, as explained in this

Map 3 — **Surface temperature**

Little change in the UK

Greatest warming over Soctland

Socio-Economic Impacts

This section is included for the potential addition of information on socio-economic impacts. For example opening up of the Arctic to shipping or impacts on fish stocks and other marine resources. Potential to link to the work of other projects, both those commissioned by the FDCO and more widely.

Please select a theme to view its impact on UK.