COMNAP Sea Ice Challenges Workshop Hobart, Tasmania, Australia Council of Managers of National Antarctic Programs 12–13 May 2015

WORKSHOP REPORT

COMNAP Sea Ice Challenges Workshop 1 COMNAP Sea Ice Challenges Workshop Volker Siegel Photo: Volker cover p hoto: Jan L Lieser and b ack Front COMNAP Sea Ice Challenges Workshop

Hobart, Tasmania, Australia 12–13 May 2015 Workshop Report

The Council of Managers of National Antarctic Programs The Council of Managers of National Antarctic Programs (COMNAP)

COMNAP Secretariat Private Bag 4800 Christchurch, New Zealand www.comnap.aq

© COMNAP and contributors 2015

COMNAP does not endorse or recommend any commercial products, processes, or services nor does it grade or rank specific research programmes. Therefore, mention or photo of any of these in this publication, in discussions and outcomes cannot be construed as an endorsement or recommendation from COMNAP nor any of the COMNAP Member national Antarctic programs. The designations employed, including any geographic names, and the presentations in this material does not imply the expression of any opinion whatsoever on the part of COMNAP concerning the legal status of any country, territory or authority.

ISBN 978-0-473-34368-2

Photo: Jan L Lieser 6 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 7 Background to the Workshop

Sea ice conditions are one of the many together in addressing a multi-faceted challenges for vessels operating in the issue of critical importance to everyone who Antarctic marine area. In recent years, operates in the area. Workshop presenters in some parts of the Antarctic, sea ice were invited from the science community and conditions have been such that those there was an open call for presentations from conditions have affected science, science COMNAP Members organisations. support and logistics operations in the The overall aim of the Sea Ice Challenges Table of Contents regions. Some parts of the Antarctic are Workshop was to discuss the latest scientific experiencing expanding sea ice extent Background to the Workshop...... 9 advice on the causes and likely future trends while other areas are seeing multi-year sea in sea ice expansion, to scope the challenges Workshop discussion...... 10 ice remain in areas where, previously, that was not the norm. The nature of changing sea ice will pose for national Antarctic Session 1: Recent National Antarctic Program Experiences sea ice conditions creates problems with programs and other operators, and to identify with Changing Sea Ice...... 10 resupply of both coastal and inland Antarctic and discuss potential solutions. research stations and also poses threats to The specific aims of the workshop were to: Session 2: Sea Ice Trends...... 11 human safety and has the potential to create emergency situations quickly. • Understand the causes of the expansion Session 4: Operational Implications...... 12 in sea ice in the waters around Session 5: Sea Ice Navigation: Operational Requirements Given the challenging nature of recent sea Antarctica; ice conditions in many parts of the Antarctic and Technologies...... 12 region, COMNAP agreed at its Annual • Scope the challenges sea ice expansion Session 6: Alternative Solutions...... 14 General Meeting (AGM) 2014 to convene pose for sustaining Antarctic stations the Sea Ice Challenges Workshop in Hobart, that depend on over water refuelling and Further Work...... 15 Tasmania, Australia, on 12 and 13 May 2015. resupply; and Appendices...... 16 On behalf of COMNAP, the workshop • Identify and explore potential solutions. convenor was Dr. Rob Wooding, Australian Appendix 1: Sea Ice Challenges Workshop Programme.....16 Twenty-five speakers from nine different Antarctic Division (AAD) General Manager countries presented. In addition, there Appendix 2: Abstracts...... 19 Support & Operations and COMNAP Vice was one poster presentation and two Chair. The workshop was co-hosted by the presentations that were sent by way of Appendix 3: Workshop Participants...... 67 AAD with thanks to its Director, Dr. Tony information papers to the workshop. Fleming, and by the Antarctic Climate and Appendix 4: Links to Further Information...... 70 Ecosystems Cooperative Research Centre The following sessions were part of the Appendix 5: Acknowledgements...... 71 (ACE CRC) with thanks to its Chief Executive, workshop programme (Appendix 1): Professor Tony Worby. Dr. Yves Frenot, Session 1: Recent National Antarctic Director of Institute Paul Emile Victor (IPEV) Program Experiences with Changing Sea Ice and COMNAP Vice Chair, provided oversight Session 2: Sea Ice Trends of the workshop. The Chair of COMNAP, Session 3: Sea Ice Technology (poster Professor Kazuyuki Shiraishi, Director- session) General of the National Institute for Polar Session 4: Operational Implications Research (NIPR), presented the welcome and Session 5: Sea Ice Navigation: Operational opening address at the workshop. Requirements and Technologies Session 6: Alternative Solutions (Including The workshop brought together scientists, Icebreakers) Antarctic program managers, sea ice forecasters, ice navigators and shipping This report presents the key outcomes of the experts from around the world to work workshop.

Photo: Jan L Lieser 8 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 9 Workshop discussion

SESSION 1: RECENT NATIONAL When we envision sea ice, many of us picture requirements on stations (of a behavioural strengthen and push cooler water and sea ANTARCTIC PROGRAM solid, flat, continuous platforms which form and technical nature) and the introduction of ice further north. This results in growth of EXPERIENCES WITH CHANGING and extend for kilometres out from the increased station self-sufficiencies (such as sea ice extent. Localised changes in wind SEA ICE Antarctic coast. But while sea ice extent is introduction of hydroponics) can have other patterns also affect the distribution of sea ice certainly a challenge, there are four major benefits in the long-term. and partially explain the increase in sea ice Session Chair: Rob Wooding factors making operating conditions difficult: extent off the Ross Sea and decrease west Session Presenters: Yves Frenot, Takeshi the extent of ice cover, icebergs (that may of the Antarctic Peninsula. Another factor Tamura, Robb Clifton, Lei Ruibo have sea ice built up around them), rafted and SESSION 2: SEA ICE TRENDS affecting surface winds is the depletion of ridged ice, and multi-year fast ice-especially ozone in the stratosphere, which cools the National Antarctic programs have operated Session Chair: Tony Worby vessels in the Antarctic marine area for more in areas where such ice has not usually stratosphere and leads to stronger gradients accumulated in the past. These obstacles Session Presenters: Rob Masson, in temperature to the surface – propagating than 50 years. Vessel operators are first-hand Marilyn Raphael, Sharon Stammerjohn, witnesses to changing sea ice conditions, impact shipping schedules which creates its a signal of strengthening westerly winds at own set of challenges given the relatively Will Hobbs, Xavier Crosta, Mark Curran, Petra the surface during summer. Sea ice models patterns and trends. Changes are regional, Heil, Phil Reid and there is no one overarching statement short Antarctic summer season. are not currently capturing many of these processes very well and are therefore not in regards to sea ice that would fit a “whole All this means that national Antarctic National Antarctic programs are interested particularly good tools for predicting future of Antarctic” situation. But, certainly for programs can no longer rely on what they in knowing if the changes they have been trends. most areas beyond the Peninsula and West thought they knew. They now have to observing in sea ice condition will continue Antarctic regions, sea ice conditions have develop contingency plans, especially for for a period of time, or whether sea ice While first-hand experience confirms sea proved challenging for the past several years. delivery of fuel supplies. Such contingencies challenges are on the decline, thus they ice challenges are increasing in parts of are interested in understanding trends. Such challenging conditions have meant that often cost more money and fuel to implement Antarctica, the climate models are showing no than is usual, an example being the need to Keeping an eye on Antarctic sea ice year- detectable trends, or a reduction in mean sea national Antarctic programs have had to: by-year, one might be correct in thinking implement alternative solutions for resupply burn aviation fuel in order to deliver the diesel ice extent. There are strong feedback actions fuel which would normally be delivered by sea ice extent appears to be very similar occurring with differing coastal icescapes of their Antarctic stations, accept increases in each year. However, duration and thickness their operational costs, work with an impacted ship. This adds to the challenges also, as ship and differing continental shelf bathymetry off-load of fuel from greater distances away are increasing in many regions, and taken also having a profound influence on sea ice ability to deliver Antarctic science and science together it is this trend that is worrying support, and consider assistance to a range from the station leads to greater risk of fuel trends. Model resolution is not as good as it spill or accident. Even the most experienced in terms of national Antarctic program needs to be and there is a need for increased of other vessels which have encountered operations and science support. problems due to sea ice conditions. national Antarctic programs and the most resolution of global-coupled models. capable ice-breaking vessels are being faced Understanding trends is important but Scientific understanding of conditions can be All this increases risk to personnel safety, with challenging conditions, so lesser vessels complicated and requires consideration of assisted through first-hand knowledge from introduces delays in service - including delays will certainly be challenged. This diverts many factors, including waves, grounded vessel operators who log positions where to planned construction, and inhibits ability resources from resupply and science to iceberg numbers and extent, information on ships stop on each voyage due to thick, fast to transport into the Antarctic necessary emergency situation. extreme events, and melt and accumulation of supplies, scientific equipment and personnel, ice. Software to assist with standardised While national Antarctic programs have snow rates. Collecting such data requires high recording of observations has been used and to transport out of the Antarctic wastes resolution remote sensing capabilities over and cargo to be retrograded. always built flexibility into planning schedules, for many years, and is currently undergoing new methods are being sought. Sea ice extensive areas at regular (daily to sub-daily) an update as part of the “Antarctic Sea The impact is felt not only at coastal Antarctic challenges may not appear to have impacts intervals. Ice Processes and Climate” (ASPeCt) stations, but in regards to inland station beyond ship movements, but they do. There is strong spatial, temporal and regional program. There is a critical need for more, resupply, where regular resupply is critical to Contingency planning often means thinking variability in Antarctic sea ice, driven by many standardised observations including bridge- maintaining and operating such stations year- about cargo container sizes and increasing drivers. For example, when the Southern based observations to be collected on all round. Inland stations cannot be closed and fuel and food storage capacity on stations. Annular Mode (SAM) is positive (something voyages into the southern pack-ice zone. reopened easily since winter temperatures Such contingency planning must also include that appears to be happening more and Along with information on pack-ice conditions plunge to temperatures close to or below -80 increased renewable option considerations, more), it strengthens westerly winds and and total sea ice extent, there is also a degrees Celsius. power usage strategies to decrease power as they strengthen, the surface currents need to understand landfast ice extent and

10 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 11 the processes controlling the presence or No matter what the sea ice situation is, the Arctic and the Antarctic. Such specialists New technologies, like CubeSat, mean better absence of landfast ice. Landfast ice can national Antarctic programs are required work on board the expedition vessels in resolution imagery is becoming available to present the final, impenetrable barrier to to provide support to their research the Antarctic and at Antarctic stations. In deliver near real-time data and information. Antarctic operations if it persists along the programme. Understanding sea ice extent addition to receiving and processing satellite Examples include sea ice deformation data coast where Antarctic stations are located. and duration will therefore allow national images of sea ice, personnel from Antarctic and new SAR imagery, GF-2 satellite imagery Antarctic programs to make smart decisions stations also carry out coastal observations Even with uncertainty in models, what we which has an 0.8m resolution, UAV use about when and where to operate and to of the state of landfast ice, assessing all its know is that historic records related to sea and OSSI software which uses Sentinel-1 have confidence in investments related age stages (beginning of seawater freeze ice extent going as far back as 30 years data that is high resolution and less than to infrastructure and to any necessary up, establishment of landfast ice, onset of its are being regularly broken. We also know two hours old. The IICWG is developing a contingencies. decay and complete or partial destruction). large-scale or extreme events influence sea collaborative ice chart for safer navigation in ice conditions. Such events may become Having a definitive information paper Examples of post-processing technologies Antarctic waters. During the period October more common as our climate changes, so produced periodically (perhaps annually) include new work which focusses on 2015 through April 2016, Weddell Sea ice there is an urgent need to improve our ability which outlines what the current scientific albedo measurements of sea ice from aerial charts will be made available. First-hand to understand the likelihood of such events consensus on the “state of play” in regard photography to identify “types” of sea ice. observations from vessel operators would be and how they influence sea ice growth. to sea ice could therefore be very helpful in While such methods are proving useful welcome to inform those charts during that Proxy records from marine and ice cores planning. Also, national Antarctic programs to operations, there is often limited time period. allow reconstruction of Antarctic sea ice need to explain clearly to the science and available to conduct such science on-board As critically important as understanding sea extent from before the instrumental period. forecasting community what their needs are, the ship, since most national Antarctic ice is, understanding weather and provision Measurements from ice cores reveal decadal the timings of their requirements and the program ships serve the dual purpose of of weather-forecasts is equally important. cycles in sea ice extent and show that sea areas where operations are planned. Open resupply and conducting science support. National Antarctic programs rely on accurate ice extent may have declined by up to 20% lines of communications across communities New ships, dedicated to science programs between the mid and late 20th Century, at are therefore very important. are in planning stages and will be on-line as weather forecasts, but there may be an least in the East Antarctic sector near Casey early as 2018. inconsistency across service levels, and therefore, there may be a need for Antarctic Station where the study was conducted. It is There are a range of products that are forecaster competency training programs therefore possible that coastal stations along SESSION 5: SEA ICE NAVIGATION: available for “now-time” forecasting. These coupled with a commercial service provided the East Antarctic coast were established OPERATIONAL REQUIREMENTS include an iceberg database, the Global during a period of reduced ice cover, and that AND TECHNOLOGIES Ocean Forecasting System (GOFs) 3.1 (7- to national Antarctic programs which could be the extent is now increasing on multi-decadal day forecast with up to a 3km resolution), based on their ship use days and areas. Session Chair: Yves Frenot scales. seasonal statistics model predictions of Session Presenters: Lei Ruibo, Caryn The International Programme for Antarctic sea ice and Polar View which uses SAR Panowicz, Lin Zhang, Andrew Fleming, Xiao Buoys (IPAB) has recognised a crucial imagery to deliver near real-time data for Cheng, Penelope Wagner, Thomas Krumpen, need for more co-ordinated launches of SESSION 4: OPERATIONAL ship operations. Other “now-casting” services Neal Young, Jan Lieser, Scott Carpentier sophisticated sea ice buoys to provide real- are being developed. These include such IMPLICATIONS time information on the evolution of ice and Understanding trends and long-term products as “Tie-Points”, but such services Session Chair: Rob Wooding snow thickness and temperature in space forecasting is important, but so is information are currently only available to the national Open discussion and time, ice drift and key meteorological in real-time. During an Antarctic season, Antarctic program of the organisation which information. Costs, ship communications The uncertainty of the models, the highly vessel operators and national Antarctic provides the service, in this case the USA, limitations, lack of buoys in place or lack regional variation and the need for higher programs require the ability to select the best but there are other examples provided of ship-time for placement of buoys, and resolution and temporal data mean that route through sea ice in terms of safety and by China and India. Joining together in placement of buoys which is science-driven the current science cannot give national timely delivery of supplies, people and science collaborative organisations and groups, such and not operations-driven, all impact the Antarctic programs the certainty they require support. as the International Ice Charting Working ability of service providers to deliver timely for annual implementation and planning Group (IICWG) can mean that resources and Some national Antarctic programs provide of their programmes. Even knowing the products can be shared amongst members and robust sea ice products for real-time training to ice pilots/ship navigators. These trends with certainty would not be the end of such groups. In the Arctic, the Sea use by national Antarctic program vessel specialists have vast practical experience in itself. Of greatest importance to national Ice Prediction Network (SIPN) connects operators. Parallel information is required in performing ship-based and airborne ice Antarctic programs is reliable information at scientists and stakeholders to improve sea from increased investment in sophisticated observations and in receiving, processing a regional level at a “now-cast” timescale. ice predictions in a changing Arctic sea ice measurement stations and Automatic and interpreting satellite sea ice images in environment. Weather Stations deployed in fast ice.

12 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 13 Particular attention should be paid to operations and technology employed for Further Work stepping up such activities during the Year of safety, such as Ground Penetrating Radar Polar Prediction (YOPP) 2017/19. (GPR), Electro-Magnetics (EM) and Remotely Based on workshop discussions it appears • One or two collaborative Piloted Aircraft (RPA). there are three key items of further work forecasting and analysis services More and more vessels and traverses are required, which, if implemented, would to cover the entire marine area SESSION 6: ALTERNATIVE being supported by remotely piloted aerial assist national Antarctic programs when around the Antarctic continent SOLUTIONS technologies, since they can be easily Antarctic sea ice conditions prove particularly (noting this would need to be Session Chair: Michelle Finnemore deployed with an on-board camera that can challenging. They are: resourced by national operators relay, in real-time, images of the forward sea not from research funding). Session Presenters: Patrice Godon, Doug 1. All sea ice scientists engaged in ice conditions and sea ice traverse routes. Thost, Guy Williams, Ted Maksym research, monitoring or forecasting in 2. Antarctic Treaty countries to work Such technologies can be tethered and can relation to the southern hemisphere together through COMNAP to provide In situations where icebreakers are unable to therefore be powered for long duration and should work together, with input from access to greater telecommunications advance through sea ice to approach land- can be wind-aware and ship-aware. based targets, national Antarctic programs COMNAP where appropriate, to build a bandwidth across the continent and are resorting to traverse operations from Technologies such as this, deployed for an global network to produce, inter alia: surrounding marine areas, to facilitate operational situation, can also be important the real time upload of sea ice data the ship to stations, associated facilities and • An annual report on the scientific data collectors for scientists, that is, they to research, monitoring and modelling field camps. In recent years, in some parts consensus on major drivers can provide data to the scientific community. facilities and researchers, and thereby, of the Antarctic, the number of traverses and trends for patterns of sea For example, GPR data from Global Position improve the accuracy of monitoring, required has increased, as has the length ice coverage in the southern System (GPS)-referenced sea ice traverses forecasting and analysis. of the required traverse. That is, in some hemisphere; and are valuable for the sea ice community who circumstances the minimum distance that a 3. Sea ice research and analysis to be can use such data to understand sea ice ship is able to approach is now farther away pursued by interested scientists from thickness which can ground truth satellite due to the presence of impassable sea ice. Antarctic Treaty countries as a priority. data and models. All ice edge data and fast Traverse operations require a range ice edge data can also be GPS-referenced of vehicles to support various terrain by ship’s captain and can also be a useful encountered and to cope with lift source of information to the scientific requirements. Traverse operations also community. require trained personnel to adequately support long distance and long duration

14 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 15 Appendices APPENDIX 1: SEA ICE CHALLENGES WORKSHOP PROGRAMME WEDNESDAY 13 MAY: 11.40–12.00 Thomas Krumpen, Lasse WORKSHOP DAY TWO Rabenstein, Stefan Hendricks, Paul Cochrane TUESDAY 12 MAY: 11.30–11:50 Marilyn Raphael (University of (Drift & Noise Polar Services GmbH, 9.00am: Session 5: Sea Ice Navigation: WORKSHOP DAY ONE California Los Angeles (UCLA), USA) Germany) Operational Requirements and Antarctic sea ice: variability, trends and 21st On-Site Sea Ice Information (OSSI): a system 8.45 am: Welcome and Opening of Technologies (Session Chair: Yves Frenot) century projections to support operations in polar regions Workshop Professor Kazuyuki Shiraishi, 9.00–9.20 Lei Ruibo (Polar Research COMNAP Chairman 11.50–12.10 Sharon Stammerjohn (University 12.00–12.20 Scott Carpentier, Neal Institute of China) of Colorado, USA) Young & Jan Lieser (Australian Bureau of Ship-based measurements on sea ice 9.00am: Session 1: Recent National Comparing and contrasting regional sea ice Meteorology) morphology during the CHINARE Antarctic Antarctic Program experiences with challenges around Antarctica A case for coordinating Antarctic marine and Arctic cruises changing sea ice weather services 12.10–12.30pm Will Hobbs (Institute for (Session Chair: Rob Wooding) 9.20–9.40 Caryn Panowicz (US National Marine and Antarctic Studies, Australia) See also information paper which will not be Naval Ice Center) 9.00–9.20 Yves Frenot (Institute Paul Emile Antarctic sea ice changes – natural or presented but is sent to share information by: US National Ice Center – Sea Ice Analysis in Victor, France) anthropogenic? D. Ram Rajak, R. K. Kamaljit Singh, Antarctic Waters The impact on the French Antarctic Program Jayaprasad P., Sandip R. Oza, Space 12.30pm: Lunch and official opening of sea ice conditions around Dumont D’Urville 9.40–10.00 Lin Zhang (National Marine Applications Centre, ISRO, Ahmedabad, since 2011 Session 3: Sea ice technology (display) Environmental Forecasting Center, Polar India) and M. Javed Beg (National Centre for Environmental Research & Forecasting Antarctic and Ocean Research, Goa, India) 1.30 pm: Session 2 (continued): 9.20–9.40 Takeshi Tamura, Shuki Ushio Division) Sea Ice Advisories for Indian Research & Sea Ice Trends & Daisuke Simizu (National Institute of Polar Operational sea ice forecasting and Supply Vessels Operating in central Dronning 1.30–1.50pm Xavier Crosta (University of Research, Japan) navigation service for the Chinese National Maud Land & East Antarctica Bordeaux, France) Recent tough sea ice conditions around Antarctic Research Expedition 12.00–12.30 Summary of morning Syowa Station Sea ice dynamics off George V Land, East 10.00–10.20 Andrew Fleming (British session/questions and discussion Antarctica: beyond the instrumental period Antarctic Survey, UK) 9.40–10.00 Robb Clifton 1.50–2.10 Mark Curran (Australian Antarctic 12.30pm: Lunch Polar View – Developing and delivering (Australian Antarctic Division) Division) operational sea ice information for the polar 1.30pm: Session 6: Alternative Adapting to a new normal for sea ice access A 100-year reconstruction of Antarctic sea regions Technologies (Session Chair: Michelle and operations at Australian Antarctic ice extent from ice cores Finnemore) stations 10.20am: Morning coffee/tea break 2.10–2.30 Petra Heil (Australian Antarctic 1.30–2.00 Patrice Godon (Institute Paul 10.00–10.20 Wang Jianzhong, Yuan 11.00–11.20 Xiao Cheng (Beijing Normal Division) Emile Victor, France) Shaohong & Lei Ruibo (Polar Research University, China) Fast-ice variability in East Antarctica Dealing with the sea ice around Dumont Institute of China) Navigational sea ice analysis for the RV 2.30–2.50 Phil Reid (Australian Bureau of d’Urville China’s experiences with impenetrable and Xuelong in Prydz Bay, East Antarctica Meteorology) changing sea ice in the Antarctic 2011/12 to 2013/14 2.00–2.20 Doug Thost (Australian Antarctic Summation of Antarctic sea ice: What we Division) & Guy Williams (Institute for Marine 10.30am: Morning coffee/tea break 11.20–11.40 Penelope Wagner (Norwegian know and where we should go and Antarctic Studies) Meteorological Institute) 2.50–3.00 Questions/discussion Using remotely piloted aircraft (RPA) in ice 11.00am: Session 2: Sea Ice Trends International Ice Charting Working Group reconnaissance roles: from crystal ball gazing (Session Chair: Tony Worby) (IICWG) collaborative Antarctic sea ice 3pm: Afternoon coffee/tea break & some practical results product 11.00–11.30 Rob Massom (Australian 3.45pm: Session 4: Operational Antarctic Division) Implications (Session Chair: Rob Wooding) An overview of sea ice and challenges for Responses from national Antarctic programs navigation to the latest scientific findings. 5.00pm: Close of Workshop Day One

16 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 17 2.20–2.40 Ted Maksym (Woods Hole Oceanographic Institution, USA) Appendix 2: Abstracts Autonomous platforms: emerging technologies for sea ice research and SESSION 1: RECENT NATIONAL ANTARCTIC PROGRAM EXPERIENCES operations WITH CHANGING SEA ICE See also information paper which will not be Recent Tough Sea Ice Conditions Around Syowa Station presented but is sent to share information by: Takeshi Tamura, Shuki Ushio, and Daisuke Simizu A. Korotkov, V. Korablev and V. Lukin (Arctic & Antarctic Research Institute of National Institute of Polar Research, Japan Roshydromet) [email protected] Ice navigation support for marine operations of the Russian Antarctic Expedition 2.40–3.30 Questions/answer session Recent sea ice condition around Syowa Station is very tough. Rafted (and ridged) drifting sea ice, huge icebergs, ridged (and rafted) fast ice, and multi-year fast ice are four major factors to 3.30 pm: Afternoon coffee/tea break make this difficult situation for our icebreakerShirase . 4pm: Summing up/future directions (Session Chair: Rob Wooding) 1. Rafted (and ridged) drifting sea ice 5.00pm: Close

During 2011/12 season cruise, Shirase spent more than two weeks to go through the heavy rafted first-year ice zone (black circle in the above figure) which are considered to be formed by the strong prevailing southward wind during December 2011.

18 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 19 2. Huge icebergs and ridged (and rafted) fast ice SESSION 2: SEA ICE TRENDS Overview of Antarctic Sea Ice and Challenges Rob Massom Australian Antarctic Division and Antarctic Climate and Ecosystems Cooperative Research Centre [email protected]

Due to its a vast coverage (ranging seasonally Understanding the scale-dependency of the from ~3-4 to ~19-20 million km2) of the air-sea-ice interaction system is crucial to seas surrounding Antarctica and highly- understanding Antarctic sea ice (Massom dynamic nature, sea ice in the form of both and Stammerjohn, 2010). At hourly to moving pack ice and stationary fast ice daily scales, the notorious dynamism of the abutting the coast (both with an accumulated Antarctic sea ice zone is strongly associated snow cover, which is an important factor) with the frequent passage of storms around represents a considerable challenge for the high-latitude S Ocean. These drive logistical operations/shipping activities in the rapid changes in air temperature and wind Southern Ocean. By the same token, the direction and strength (while dumping large ice represents a unique natural laboratory amounts of snow on the ice – typically under for crucially-important scientific research blizzard conditions), and result in a complex During 2014/15 season, Shirase spent approximately 5 days to go through the heavy ridged towards unravelling the key role of the ice interplay of thermodynamic (freeze/melt) fast ice zone close to the boundary of drifting ice and fast ice (red circle in the above figure), and its snow cover in high-latitude ecological and dynamic (ice motion and deformation) and a huge iceberg has passed around the rote of the vessel within 1 week. and biogeochemical processes and the processes. Alternating synoptic-scale periods Earth’s climate system. Sea ice is not only a of ice divergence and convergence thicken 3. Multi-year fast ice sensitive indicator of climate variability and the pack ice beyond ~1-2 m (Worby et al., change (which is amplified at high latitudes), 1998), with constant reworking of the ice by virtue of its intimate association with the leading to a high degree of heterogeneity in atmosphere and ocean; it also contributes to ice concentration and ice and snow thickness climate change through complex and poorly- and properties on horizontal scales of metres understood interactions involving the coupled to kilometres. ice-ocean-atmosphere system (see other Snow is a key factor that both limits and abstracts). contributes to thermodynamic ice growth At circumpolar and seasonal scales, – the former due to its strong insulating patterns of annual sea ice advance, retreat properties, and the latter by “snow-ice” and duration closely mirror climatological formation (freezing of slush created by temperature, wind and ocean current patterns, the flooding of the ice surface following and vary considerably between regions and depression below sea level by a snow years. Crucially, the current overall trend in overburden). Snow also exhibits complex Antarctic sea ice coverage (since 1979) is internal structure with layers of variable made up of different and contrasting regional density and properties, and can have a first- changes in seasonality (Stammerjohn et al., order effect on the efficiency of icebreakers 2012). An issue/challenge is the need for (depending on snow type and wetness). seasonal ice forecasting capability tuned to Enhanced precipitation predicted in future different regions pan-Antarctic, to support (Bracegirdle et al., 2008) may lead to both voyage operations and science experiments greater snow-ice formation and shielding During 2012/13 season, Shirase could not go through the heavy multi-year fast ice zone (black (planning). of the ice surface to delay summer melt. circle in the above figure) and gave up reaching the Syowa station.

20 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 21 There is, however, considerable uncertainty, concentrations, larger floe sizes (due to the considerable challenge. Overcoming this • Understanding of key processes e.g., given our poor knowledge of the current damping effect of sea ice to the north on challenge requires: i) region-specific, high- wave-ice interaction (notably in the MIZ), contribution of snow-ice to overall Antarctic encroaching wave energy) and thicker sea resolution modelling that couples the different regional air-sea-ice interaction, and sea ice mass balance. Major challenges ice and snow cover. It is typically traversed by interactive cryosphere elements in a realistic seasonal evolution of the coupled ice revolve around the need to better measure, networks of leads that are largely ephemeral. fashion; ii) improved bathymetric data and and snow system etc. monitor and model precipitation and During freezing periods, leads are areas of information on (small grounded) iceberg This will involve (amongst other things) accumulation/loss over sea ice (Leonard rapid new ice formation, and become areas distribution; iii) frequent high-resolution and dedicated field campaigns involving careful and Maksym, 2011). Please see Sturm and of enhanced ice melt in the melt season. near real-time satellite coverage involving coordination to upscale in situ observations Massom (2010) for a review of snow on sea Lead opening and closing is a challenge to SAR-derived ice motion/deformation and fast with remote sensing (also involving cross- ice. observe and model, requiring regular high- ice mapping over more extensive areas than calibration/validation of satellite products), resolution information over an extensive area. is currently the case; and iv) an understanding While Antarctic sea ice is characterised by and extensive use of state-of-the-art Key to successful operations is improved of the regional setting in the longer term strong small-local scale variability, it exhibits autonomous surface, airborne and under-ice timely information on the local-regional ice – the Mertz/B9B event was, for example, recurrent patterns on the larger scale of technology (Maksym et al., 2012). Within convergence and divergence fields; this will a decadal-scale event that was difficult/ hundreds of kilometres in the form of zones this is a crucial need for more launches of require enhanced space borne Synthetic impossible to predict. with distinctive characteristics but variable sophisticated sea ice buoys coordinated Aperture Radar (SAR) coverage over all key width depending on geographic location and A critical knowledge gap remains our within the International Programme for areas around Antarctica. season (Massom and Stammerjohn, 2010). inability to accurately & routinely measure/ Antarctic Buoys (IPAB, http://www.ipab. These are: i) the highly-dynamic marginal ice In the coastal zone, complex interactions monitor Antarctic sea ice (and snow cover) aq) to provide crucial real-time information zone (MIZ); ii) the inner pack; and iii) coastal occur between pack and fast ice, polynyas, thickness on the large scale from space. on the evolution of ice and snow thickness zone. In the MIZ, the ice is strongly affected ice sheet coastal promontories and icebergs Satellite altimetry holds the key, but thickness and temperature in space and time, ice by ocean swell and waves. The ice edge itself (including assemblages of small icebergs derivation requires independent knowledge drift and key meteorological information seldom forms a clear-cut boundary, unless grounded on offshore banks <~350-400 of sea ice density and snow thickness and for transmission via the GTS. Parallel persistent northerly winds persist; more m deep) (Massom et al., 2001). Recurrent density, and is undermined by the typically information is required from increased typically, it constitutes a diffuse zone up to regional patterns once again occur, but small freeboard of Antarctic sea ice. investment in sophisticated measurement tens of kilometres wide and comprising series stations and AWSs deployed in fast ice these can be rapidly disrupted by persistent Given the challenges and complexities of ice bands. within the international Antarctic Fast Ice changes in atmospheric forcing, the influx outlined above, combined with the Network. Particular attention should be paid Ocean waves play a key role in both ice of large icebergs and/or change in coastal vast coverage of Antarctic sea ice and to stepping up such activities during the Year formation and destruction (including fast configuration. For example, persistent the difficulties involved in carrying out of Polar Prediction (2017-19, http://www. ice breakup), and can penetrate and affect northerly winds can create hazardous research there, there is a pressing need polarprediction.net/yopp.html), with a view to the entire sea ice zone at times e.g., in compaction of sea ice against the coast/ for collaborative international and cross- improving sea ice modelling and forecasting East Antarctica (Kohout et al., 2014). They offshore obstructions accompanied by disciplinary sea ice research in support of capabilities. can also make fieldwork and operations extreme ice thickening – with the same enhancing our: hazardous. Given these factors, there is a northerly winds also bringing in enhanced • Observational, modelling and forecasting strong need for improved information on amount of snowfall e.g., Massom et al. capability; and sea-state and wave-ice interaction (both (2006). Also, “wildcard” events involving in real-time and for research purposes) i.e., abrupt change in coastal cryosphere from models, satellite SAR, ship’s radar and “elements” can have a dramatic effect on wave buoys. A current issue is that wave- the coastal “icescape” - as highlighted ice interaction processes are typically not by the 2010 calving of the Mertz Glacier included in coupled climate models. These Tongue and repositioning of iceberg B9B off challenges/issues are underpinned by Commonwealth Bay (Tamura et al., 2012). the predicted future scenario of increased Given the complex nature of the interactions storminess (and “waviness”), particularly in and their scale i.e., kilometres to tens of summer and autumn (Turner et al., 2013). kilometres, observing, modelling/forecasting To the south, the inner or central pack ice and understanding sea ice conditions in zone is generally characterised by higher ice the near-coastal icescape represents a

22 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 23 References Antarctic Sea Ice: Variability, Trends, Drivers and 21st Century Bracegirdle, T.J., W.M. Connolley, and J. Turner Stammerjohn, S., R. Massom, D. Rind and D. Projections (2008). Antarctic climate change over the twenty Martinson (2012). Regions of rapid sea ice change: Marilyn Raphael first century. Journal of Geophysical Research, An inter-hemispheric seasonal comparison. 113(D3), D03103, doi:10.1029/2007JD008933. Geophysical Research Letters, 39, L06501, UCLA Department of Geography, Los Angeles, California, USA doi:10.1029/2012GL050874. [email protected] Kohout, A.L., M.J.M. Williams, S.M. Dean and M.H. Meylan (2014). Storm-induced sea-ice breakup Sturm, M. and R.A. Massom (2010). Snow & Sea Unbounded by land, Antarctic sea ice extent century. A positive SAM is associated with and the implications for ice extent. Nature, 509, Ice. In D. Thomas & G. Dieckmann (editors): Sea grows from an average minimum of near stronger westerlies. The resulting enhanced 604-607. Ice. 2nd Edition, Wiley-Blackwell, New York (USA) & Oxford (UK), pp. 153-204. NB A new edition is 5x106 km2 in March to an average maximum Ekman transport causes surface cooling by Leonard, K.C., and T. Maksym (2011). The currently in press. near 20x106 km2 by September. While moving cold surface waters northward from importance of wind-blown snow redistribution to the whole sea ice system experiences this Antarctica, creating conditions for enhanced snow accumulation on Bellingshausen Sea ice. Tamura, T., G. D. Williams, A. D. Fraser, and K. pronounced regular annual cycle, as might sea ice growth. ZW3 induces preferred Annals of Glaciology, 52(57), 271-278. I. Ohshima (2012). Potential regime shift in be expected within such a vast system, there regions of equatorward and poleward flow decreased sea ice production after the Mertz is variability in space and time in the size thereby influencing the meridional transport Maksym, T., S.E. Stammerjohn, S. Ackley and Glacier calving. Nature Communications 3: 826. and timing of the maximum and minimum of heat into and out of the Antarctic with R. Massom (2012). Antarctic sea ice – A polar doi:10.1038/ncomms1820. extents. Antarctic sea ice can be divided into resulting impact on temperature and sea ice opposite? Oceanography, 25(3), 140-151, http:// regions of internal coherent variability – each extent. Research suggests that ZW3 has dx.doi.org/10.5670/oceanog.2012.88. Turner, J. et al. (2013). Antarctic climate change region displays distinct differences in timing the potential to influence the regionality of and the environment: an update. Polar Record, of advance and retreat and the length of time Antarctic sea ice trends (Raphael, 2007). Massom, R.A. and S.E. Stammerjohn (2010). 50(03), 237-259. Antarctic sea ice change and variability – Physical that they remain at maximum extent that may Like ZW3, the ASL variability influences the and ecological implications. Polar Science, 4, Worby, A. P., R. A. Massom, I. Allison, V. I. Lytle, and underlie the differences in variability that the climate of West Antarctica by controlling the 149-186. P. Heil (1998). East Antarctic Sea Ice: A Review of regions exhibit (Raphael and Hobbs, 2014). meridional component of the large-scale Its Structure, Properties and Drift. In Antarctic Sea In recent years temporal Antarctic sea ice atmospheric circulation, with consequences Massom, R.A., Stammerjohn, S.E., Smith, R.C., Ice: Physical Processes, Interactions and Variability, variability has been marked by a positive for meridional wind velocity, surface air Pook, M.J., Iannuzzi, R.A., Adams, N., Martinson, ed. O Jeffries, 41–67. Washington, D.C.: AGU. trend in annual total sea ice extent. This temperature, precipitation, and sea ice D.G., Vernet, M., Fraser, W.R., Quetin, L.B., Ross, doi:10.1029/AR074p0041. positive trend in the total sea ice extent is concentration. A persistent and deep ASL R.M., Massom, Y. and Krouse, H.R. (2006). Extreme led chiefly by the positive trend in the Ross over the Amundsen–Bellingshausen Seas anomalous atmospheric circulation in the West Sea in autumn (Turner et al, 2009). Other (ABS) sector leads to enhanced northerly Antarctic Peninsula region in austral spring and regions around Antarctica are experiencing airflow across the western Antarctic summer 2001/2, and its profound impact on sea negative trends in sea ice (for example the Peninsula sector, resulting in higher surface ice and biota. Journal of Climate, 19, 3544-3571. Bellingshausen Sea) and in the other seasons air temperature, and reduced sea ice extent the negative trend is more dominant than the in the Bellingshausen and eastern Amundsen positive trend (Simpkins et al, 2012). Seas. Conversely, the enhanced southerly There are several proposed contributors to flow of cold continental air along the western the trend in Antarctic sea ice and perhaps flank of the ASL results in increased sea that fact is a measure of the complexity of the ice extent in the western Amundsen and system. These contributors include the high Ross Seas. While the ASL has deepened, latitude, largescale atmospheric circulation analysis suggests that it is the location of system (SAM, ZW3, ASL), Tropical influences, the ASL that is most important because the freshwater influx from basal melting of location determines where the associated ice shelves, winds on ice motion and drift, winds would prevail (Hoskings et al, 2013). ice-ocean feedback, and atmosphere-ocean Turner et al, (2009) indicate that the annual feedback. The index measuring the strength increase in sea ice is led by autumn increases of the Southern Annular Mode (SAM) in the Ross Sea associated with stronger has a positive trend in the later decades ASL. Ongoing analysis suggests that the of the 20th century and in the early 21st ASL location has a significant negative trend

24 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 25 suggesting that it lies west of its average There are few published studies on recent References Raphael, M. N., 2007: The influence of location. The influence of winds on ice motion CMIP5 model results that focus on Antarctic atmospheric zonal wave three on Antarctic sea Bintanja, R., G. J. van Oldenborgh, S. S. Drijfhout, and subsequently on the trends in sea ice is sea ice but those that do show that climate ice variability. J. Geophys. Res., 112, D12112, B. Wouters, and C. A. Katsman, 2013: Important discussed in Holland and Kwok (2012) who models simulate a weak decrease of sea ice doi:10.1029/2006JD007852. role for ocean warming and increased ice-shelf show that large and statistically significant extent (e.g. Mahlstein et al., 2013; Turner et melt in Antarctic sea-ice expansion. Nat. Geosci., 6, Raphael, M.N. and W. Hobbs, 2014: The influence changes in ice motion are driven by changes al, 2013; Zunz et al., 2013). These studies 376–379, doi:10.1038/ngeo1767. of the large-scale atmospheric circulation on in the winds. conclude that the observed SIE trend is Antarctic sea ice during ice advance and retreat within the range of the models’ internal Ferreira, D., J. Marshall, C. M. Bitz, S. Solomon, Tropical influences on Antarctic sea ice seasons. Geophys. Res. Lett., 41, 5037–5045, variability and that the simulated decrease and A. Plumb, 2015: Antarctic Ocean and Sea Ice include ENSO and the AMO (Atlantic doi:10.1002/2014GL060365. in Antarctic SIE since1979 is also within the Response to Ozone Depletion: A Two-Time-Scale Multidecadal Oscillation). While the effect of range of internal variability (e.g. Swart and Problem. J. Climate, 28, 1206–1226. doi: http:// Swart and Fyfe, 2013 Swart, N. C., and J. C. Fyfe, ENSO is on the atmospheric circulation and Fyfe, 2013). However Hobbs et al (2015) dx.doi.org/10.1175/JCLI-D-14-00313.1 2013: The influence of recent Antarctic ice sheet is concentrated largely in the Bellingshausen suggest that the differences between the retreat on simulated sea ice area trends. Geophys. Sea and on the Peninsula, it is suggested Goosse, H. and V. Zunz, 2014: Decadal trends in model simulations and observations cannot Res. Lett., 40, 4328–4332, doi:10.1002/grl.50820. that sea surface warming related to the be explained by internal variability alone. That the Antarctic sea ice extent ultimately controlled by AMO reduces the surface pressure in the Simpkins, G. R., L. M. Ciasto, D. W. J. Thompson, all the models fail to reproduce the observed ice–ocean feedback. The Cryosphere, 8, 453–470, Amundsen Sea thereby contributing to the and M. H. England, 2012: Seasonal relationships increase in SIE may indicate that there are 2014, doi:10.5194/tc-8-453-2014 observed sea ice dipole between the Ross between large-scale climate variability and common failings in the representation of sea Hobbs, W.R, N. L. Bindoff, and M. N. Raphael, and Amundsen, Bellingshausen and Weddell Antarctic sea ice concentration, J. Clim., 25, ice in the models. The disagreement with 2015: New Perspectives on Observed and seas (Li et al, 2014). 5451–5469, doi:10.1175/JCLI-D-11-00367.1. the observed increase might be due, among Simulated Antarctic Sea Ice Extent Trends Using Bintanja et al (2013) proposed that other things, to the misrepresentation in Optimal Fingerprinting Techniques (doi: 10.1175/ Turner, J., and Coauthors, 2009: Non-annular freshwater influx from basal melting of climate models of some important feedbacks, JCLI-D-14-00367.1) atmospheric circulation change induced by Antarctic ice shelves leads to freshening or because model resolution is too largescale stratospheric ozone depletion and its role Holland, P.R. and R. Kwok, 2012: Wind-driven seawater, which then allows ice to form to allow accurate representation of sub- in the recent increase of Antarctic sea ice trends in Antarctic sea-ice drift. Nat. Geosci., 5, more easily. However Swart and Fyfe (2013) gridscale processes in the ice and ocean. extent. Geophys. Res. Lett., 36, L08502, st 872–875, doi:10.1038/ngeo1627. find that this input is not significant enough Projected sea ice trends for the 21 century doi:10.1029/2009GL037524. to produce sea ice to match the observed suggest continued decrease of sea ice Hosking, J. S., A. Orr, G. J. Marshall, J. Turner, and Turner, J., T. J. Bracegirdle, T. Phillips, G. J. increase in extent. extent. T. Phillips, 2013: The influence of the Amundsen– Marshall, and J. S. Hosking, 2013: An initial Bellingshausen Seas low on the climate of West The two major feedbacks proposed are an assessment of Antarctic sea ice extent in the Antarctica and its representation in coupled atmosphere-ocean feedback where a warmer CMIP5 models. J. Climate, 26, 1473–1484, climate model simulations. J. Climate, 26, 6633– climate leads to more precipitation (snow) doi:10.1175/JCLI-D-12-00068.1. and more snow-ice formation (Powell et 6648, doi:10.1175/JCLI-D-12-00813.1. Zhang, J. L., 2007: Increasing Antarctic sea al 2005; Zhang, 2007) and an ice-ocean Li, X., D. M. Holland, E. P. Gerber, and C. Yoo, 2014: ice under warming atmospheric and oceanic feedback where an increase in SIE is Impacts of the north and tropical Atlantic Ocean on conditions. J. Climate, 20, 2515–2529, doi:10.1175/ associated with decreased mixed layer depth the Antarctic Peninsula and sea ice. Nature, 505, JCLI4136.1. and stabilization of the water column due to 538–542, doi:10.1038/nature12945. the net inflow of water and brine rejection. Zunz, V., H. Goosse, and F. Massonnet, 2013: How Mahlstein, I., P. R. Gent, and S. Solomon, 2013: The stratified water column becomes very does internal variability influence the ability of Historical Antarctic mean sea ice area, sea stratified, limiting the vertical transfer of the CMIP5 models to reproduce the recent trend in ice trends, and winds invCMIP5 simulations. oceanic heat flux that would melt the ice from Southern Ocean sea ice extent? Cryosphere, 7, J. Geophys. Res. Atmos., 118, 5105– below, thereby maintaining a larger sea ice 451–468, doi:10.5194/tc-7-451-2013. extent (Goosse and Zunz, 2014). While it is 5110,vdoi:10.1002/jgrd.50443. not clear which of the proposed mechanisms Powell, D.C., Markus, T., Stössel, A., 2005: Effects is chiefly responsible for the observed of snow depth forcing on Southern Ocean sea trends, it seems clear that they are due to a ice simulations. J. Geophys. Res.: Oceans 110, multiplicity of contributing agents. C06001.

26 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 27 Comparing and Contrasting Regional Sea Ice Changes details). Once computed, the mean ice season convergence of the sea ice cover drive sea duration map (Slide 4) for the period 1979-80 ice production and deformation (mechanical Sharon Stammerjohn to 2012-13 shows, as expected, the longest thickening by rafting and ridging). Overall, Institute of Arctic and Alpine Research, University of Colorado Boulder ice season along most coastal regions, with the Antarctic pack is thin (relative to Arctic [email protected] exceptions being in the polynya areas (e.g., in pack ice), showing a mean thickness < 1 m the southwestern Ross Sea) and along the (Slide 6), with thinner sea ice (< 0.5 m) along Acknowledgements: The presenter thanks the many esteemed colleagues referenced north-western side of the Antarctic Peninsula. the outer pack ice during autumn and winter throughout the presentation and NSF, with special thanks to Dr. Scott Borg and to the Palmer Although there is only one region showing, in (Worby et al., 2008; Kurtz and Markus, 2012; LTER (ANT-1440435) the mean, year-round (or perennial) sea ice Holland et al., 2014). The thin sea ice in the cover (located in the western Weddell Sea), outer pack ice is maintained by wind-driven there is indeed summer sea ice in many of divergence, which creates areas of new sea This presentation gives a quick overview maximum sea ice extent (the area inside the other coastal regions and embayments. ice (Worby et al.,2008), and a high ocean heat of the regional and seasonal differences the ice edge containing both sea ice and However, summer sea ice varies considerably flux, which limits basal ice growth (Martinson in Antarctic sea ice concentration and open water areas) is about 19 million square in its extent and location from year to year; and Iannuzzi, 1998). Sea ice is thickest in annual ice season duration, and how those kilometres. This seasonal wax and wane of thus, coastal areas with summer sea ice fall summer, when the thinner, outer pack has differences relate to winds, waves, sea ice sea ice is more than a six-fold change in within the 330-360 day range when averaged melted, leaving behind thicker, mostly heavily motion and sea ice thickness. ice-covered area and is one of the largest over all years. deformed sea ice (i.e., thick enough to survive seasonal signals on earth (Lieser et al., 2013). The seasonal growth and decay of Antarctic the summer melt). As mentioned, ice season duration is the sea ice is a function of both thermodynamical Temporal variability in sea ice concentration time elapsed between the autumn ice edge We next explore regional trends in sea ice, and dynamical processes. For example, (Slide 3) is highest at and inside the ice edge advance and subsequent spring ice edge winds, ice motion, waves, and ice thickness. in addition to seasonal cooling in autumn where concentrations range from 0% to retreat at a given location. Variability in those Trends in ice season duration (Slide 9) and warming in spring, winds, waves and ~75% (Simpkins et al., 2012). Here is where two metrics (ice advance and retreat) shows show large contrasting regional trends, with ocean currents also affect ice growth and the influence from storms (thus winds and high variability along the outer ice edge most decreases ranging from 2 to 3 months in melt processes, as well as ice edge location waves) is highest in terms of variability. The everywhere except in the Weddell Sea and the western Antarctic Peninsula, southern and ice motion, and thus ice concentration width of this high-variability zone depends Indian Ocean sectors (~30W to 90E) and in Bellingshausen (Stammerjohn et al., 2008) (here meaning percent ice cover) and ice both on the season and on the direction and the coastal regions in the western Weddell and eastern Amundsen Seas (Stammerjohn thickness. There can also be seasonal ocean magnitude of the storm tracks. Seasonal Sea, Amundsen Sea and eastern Ross Sea. et al., 2015), as well as in isolated areas feedbacks between spring and autumn. means show high variability throughout most along East Antarctica (Massom et al., 2013). Finally, regional differences in geography (e.g., of the pack ice in summer and autumn. The Antarctic sea ice variability is largely wind- Increasing trends ranging from 1 to 2 months low latitude coastlines versus high latitude high variability zone then follows the ice edge driven (Assmann et al., 2005; Massom et al., are most pronounced in the western Ross embayments; narrow versus wide continental northward in winter and spring, such that the 2008; Holland and Kwok, 2012), particularly Sea (Stammerjohn et al., 2008; Turner et al., shelf areas) lead to regional differences in interior pack ice during these seasons show in spring, with possible ocean feedbacks 2009; Stammerjohn et al., 2012), with weaker local atmosphere and ocean forcing that also low variability. In spring, the high-variability contributing to autumn sea ice changes increasing trends in the Weddell Sea and contribute to regional differences in Antarctic zone becomes slightly more compact and (Nihashi and Ohshima, 2001; Stammerjohn Indian Ocean sectors. sea ice. In the following, we first focus zonally symmetric. et al., 2012; Holland, 2014). An analysis of on regional differences in mean seasonal winds and ice motion (Slide 5) show them Regional trends in sea ice concentration To determine the annual ice season duration conditions and then we explore how regional to be highly correlated through most of the largely reflect trends in winds and wind-driven (Stammerjohn et al., 2008; Massom and and seasonal conditions have changed over winter pack ice area, with exceptions being ice motion (Holland and Kwok, 2012) (Slide Stammerjohn, 2010; Stammerjohn et al., 2012; time. in the western Ross Sea and East Antarctic 10), and these trends in turn project onto Massom et al., 2013), the ice year begins and sector (~120E to 180E), where respectively the trends in ice season duration (discussed Antarctica consists of a vast glaciated ends during the mean summer minimum, convergent zones (in strong outflow areas) above). The largest trends in northward ice polar continent (about 14 million square defined here to begin on 15 February and end or strong coastal currents play a greater role motion and increases in sea ice concentration kilometres), which is surrounded by the on 14 February of the following year. Within in determining ice motion variability (Holland during the ice growth period (here defined Southern Ocean, a large portion of which that period, the annual ice season duration and Kwok, 2012). from April to June) are in the outer pack ice becomes covered by seasonal sea ice (about is the total number of days between the day zones of the western Amundsen and Ross 16 million square kilometres), with a much of ice edge advance in autumn and the ice Winds not only contribute to the spatial and Sea sector (~120W to 160E) and eastern smaller portion covered by year-round sea ice edge retreat in spring (both reported in year temporal variability of sea ice extent but Weddell Sea and Indian Ocean sector (~0E (about 3 million square kilometres). In winter, day; see Stammerjohn et al., 2008 for more also sea ice thickness, as divergence and to 60E). Southward trends in ice motion

28 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 29 and decreases in sea ice concentration are potential trends in sea ice thickness (Slide geography of where these warm deep References centred in the eastern Bellingshausen Sea. 12). Most models show increases in sea ice waters have access to the continental shelf Assmann KM, Hellmer HH, Jacobs SS. 2005. The decreases in sea ice concentration in the thickness in the western Weddell Sea and regions of Antarctica (Martinson, 2012), e.g., Amundsen Sea ice production and transport. J East Antarctic sector (~90E to 120E) are in some coastal regions (in the southern along the Bellingshausen and Amundsen Geophys Res 110: doi:10.1029/2004jc002797. related to the increase in southward winds, Bellingshausen and Amundsen Seas) and continental shelf break areas and along but ice motion shows a mix of trends due to decreases in the western Antarctic Peninsula some continental shelf break areas of East Goosse H, Zunz V. 2013. Decadal trends in the this area being a convergent zone with strong and Bellingshausen Sea regions (Massonnet Antarctica (e.g., ~90-100E) (Slide 14). Since Antarctic sea ice extent ultimately controlled by coastal currents. et al., 2013; Holland et al., 2014). Otherwise, we do not have sufficient in situ observations ice-ocean feedback. The Cryosphere Discuss 7: model results are mixed, some showing of contemporaneous ice-ocean changes, we 4585-4632. doi:10.5194/tcd-7- 4585-2013. Although the influence of winds and waves on increases in ice thickness in the Ross Sea cannot assess the space/time variability in sea ice will be most pronounced near the ice or in Amundsen Sea or in the outer eastern ocean heat flux to the surface mixed layer. Holland PR. 2014. The seasonality of Antarctic edge, large storms can generate long period Indian Ocean. But the geographic similarity between ice sea ice trends. Geophysical Research Letters 41: waves, and those waves can travel hundreds season duration trends and the proximity 4230-4237. doi:10.1002/2014gl060172. of kilometres into the pack ice, affecting ice Regional trends in ocean surface properties of deep ocean heat to the continental shelf concentration and thus ice growth or melt (temperature and salinity) (e.g., Meredith and Holland PR, Bruneau N, Enright C, Losch M, Kurtz areas is striking. Where these warm deep processes well into the pack ice interior King, 2005) and in seasonal feedbacks also NT, et al. 2014. Modeled Trends in Antarctic Sea waters have access to the continental shelf (Kohout et al., 2014) (Slide 11). For example, correspond to regional sea ice changes. For Ice Thickness. Journal of Climate 27: 3784-3801. regions along West Antarctica is also where the relationship between changes in ice example, regional changes in sea ice can doi:10.1175/jcli-d-13-00301.1. the strongest trends in ice shelf thinning have edge latitude and modelled significant wave lead to increases (or decreases) in surface been observed (Rignot et al., 2013; Paolo et Holland PR, Kwok R. 2012. Wind-driven trends in height by longitude was examined over the ocean warming that were caused by an earlier al., 2015). Antarctic sea-ice drift. Nature Geoscience 5: 872- ice growth period (here defined from March (or later) ice edge retreat spring, which then 875. doi:10.1038/ngeo1627. to August) and ice decay period (September caused a later (or earlier) ice edge advance In summary: to February), which showed high regional in autumn (Nihashi and Ohshima, 2001; Kohout AL, Williams MJ, Dean SM, Meylan MH. • Regional differences in sea ice changes correspondence between trends in waves and Stammerjohn et al., 2012; Holland, 2014). 2014. Storm-induced sea-ice breakup and the largely reflect regional differences in ice edge location (for details see Kohout et Evidence for this feedback is indicated implications for ice extent. Nature 509: 604-607. atmospheric circulation patterns, which al., 2014). by the high correlations between yearly doi:10.1038/nature13262. in turn can lead to differences in wind- anomalies in the spring ice edge retreat and In addition to winds, trends in air and ocean and wave-forcing on ice motion, ice Kurtz NT, Markus T. 2012. Satellite observations the subsequent autumn ice edge advance temperature and in precipitation (and hence concentration and ice thickness. of Antarctic sea ice thickness and volume. (i.e., over summer) (Slide 13) as observed snow depth) and ocean freshwater content Journal of Geophysical Research 117: throughout most of the pack ice, in particular • Seasonal sea ice changes exhibit strong also exert considerable control on trends doi:10.1029/2012jc008141. the interior pack ice areas in the West feedbacks between spring and the in sea ice growth and melt processes Antarctic sector (Stammerjohn et al., 2012). subsequent autumn, consistent with ice- Lieser JL, Massom RA, Fraser AD, Haward MG, (e.g., Meredith and King, 2005; Sturm Conversely, very low correlations are detected albedo/ocean feedbacks, accentuating Heil P, et al. 2013. Position analysis: Antarctic sea and Massom, 2009; Maksym et al., 2012). between yearly anomalies in the autumn ice regions of strong sea ice changes. ice and climate change 2014. Hobart, Tasmania, Due to weak ocean stratification and the edge advance and the subsequent spring ice Australia: Antarctic Climate & Ecosystems presence of warm deep waters, seasonal sea edge retreat (i.e., over winter), consistent with • Continental shelf regions differ in Cooperative Research Centre. Report no. PA07- ice thickness is also limited by high ocean this type of ice-albedo-ocean feedback being their bathymetry and proximity to the 131115. heat flux (Martinson and Iannuzzi, 1998). operative during the sunlight periods. There Antarctic Circumpolar Current, thus Given these factors, thick Antarctic sea ice Maksym T, Stammerjohn S, Ackley S, Massom are indications of other ocean feedbacks differ with respect to shelf currents and evolves either through wind driven rafting R. 2012. Antarctic Sea Ice—A Polar Opposite? affecting the outer ice edge due to changes proximity to warm Circumpolar Deep and ridging or from the top-down through Oceanography 25: 140-151. doi:10.5670/ in net freshening or the inner pack ice area Water. surface flooding (caused by a thick snow oceanog.2012.88. due to changes in net sea ice production as cover), freezing and thickening (rather than • Although not directly discussed in this well (Goosse and Zunz, 2013). presentation, continental shelf regions Martinson DG. 2012. Antarctic circumpolar current’s the more usual way of thickening from the role in the Antarctic ice system: An overview. bottom-up) (Maksym et al., 2012). Thus, As mentioned, weak ocean stratification also differ in their coastal icescapes (e.g., polynyas, ice tongues, fast ice), Paleogeog, Paleoclim, Paleoecol: doi:10.1016/j. trends in winds and precipitation can affect and high ocean heat flux (caused by the palaeo.2011.04.007. trends in ice thickness. Since we do not presence of warm deep waters) can limit which in turn contribute to regional have long enough observations of sea ice sea ice thickness (Martinson and Iannuzzi, differences in sea ice. thickness, we must rely on models to explore 1998). There is regional variability in the

30 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 31 Martinson DG, Iannuzzi RA. 1998. Antarctic Rignot E, Jacobs S, Mouginot J, Scheuchl B. 2013. Antarctic Sea Ice Changes – Natural or Anthropogenic? ocean-ice interaction: implications from ocean bulk Ice-shelf melting around Antarctica. Science 341: property distributions in the Weddell gyre. Pages 266-270. doi:10.1126/science.1235798. Will Hobbs 243-271 in Jeffries MO, ed. Simpkins GR, Ciasto LM, Thompson DWJ, England Antarctic Climate and Ecosystems Cooperative Research Centre Antarctic Sea Ice: Physical Processes, Interactions MH. 2012. Seasonal Relationships between Large- [email protected] and Variability, vol. Antarctic Research Series 74. Scale Climate Variability and Antarctic Sea Ice Washington, D.C.: American Geophysical Union. Concentration. Journal of Climate 25: 5451-5469. Confidence in short and long term projections forced response is quantified, and the doi:10.1175/jcli-d-11-00367.1. of the future Southern Ocean sea ice state is models are simultaneously tested against the Massom R, Reid P, Stammerjohn S, Raymond only possible with a complete understanding observed climate. B, Fraser A, et al. 2013. Change and variability Stammerjohn S, Massom R, Rind D, Martinson of the processes involved, and an evaluation The work presented here is an overview of in East Antarctic sea ice seasonality, 1979/80- D. 2012. Regions of rapid sea ice change: of whether climate models adequately the current state-of-the-science of Antarctic 2009/10. PloS one 8: e64756. doi:10.1371/journal. An inter-hemispheric seasonal comparison. represent those processes. For the sea ice cover Detection and Attribution work, pone.0064756. Geophysical Research Letters 39: L06501. Southern Ocean, the situation is particularly along with suggested directions for future doi:10.1029/2012gl050874. complicated since sea ice variability in Massom RA, Stammerjohn SE. 2010. Antarctic sea progress. ice change and variability – Physical and ecological Stammerjohn SE, Maksym T, Massom RA, Lowry different regions is affected by quite different implications. Polar Science 4: doi:10.1016/j. KE, Arrigo KR, et al. 2015. Seasonal sea ice modes of atmospheric variability (Raphael and Almost all coupled climate models, when polar.2010.05.001. changes in the Amundsen Sea, Antarctica. Elem Hobbs, 2014). For long term logistics planning driven by realistic estimates of natural and Sci Anth: accepted. that is influence by ice cover changes, there is anthropogenic 20th century climate forcings, Massom RA, Stammerjohn SE, Lefebvre W, a clear need to understand whether observed show a decrease in Antarctic sea ice cover Harangozo SA, Adams N, et al. 2008. West Stammerjohn SE, Martinson DG, Smith RC, Yuan changes in sea ice cover are anthropogenic since 1979, which is the exact inverse of what Antarctic Peninsula sea ice in 2005: Extreme X, Rind D. 2008. Trends in Antarctic annual sea ice and likely to continue in the future, or simply is observed. Are the models then incorrect? compaction and ice edge retreat due to strong retreat and advance and their relation to ENSO and the result of natural multidecadal variability. Several studies say no, because the internal anomaly with respect to climate. J Geophys Res Southern Annular Mode Variability. J Geophys Res variability of Antarctic sea ice is so high that Detection and Attribution is the branch of 113: doi:10.1029/2007JC004239. 113: doi:10.1029/2007JC004269. neither the observed nor modelled trends can climate science that seeks to determine be distinguished from ‘noise’ (Mahlstein et Massonnet F, Mathiot P, Fichefet T, Goosse H, Sturm M, Massom RA. 2009. Snow and sea ice. whether an observed change: König Beatty C, et al. 2013. A model reconstruction Pages 153-204 in Thomas D, Dieckmann G, eds. al, 2013; Polvani and Smith, 2013; Zunz et al, of the Antarctic sea ice thickness and volume Sea Ice, 2nd Edition. Wiley-Blackwell: Oxford. 1. Is outside the range of internal variability 2013). However, those studies used total sea changes over 1980–2008 using data assimilation. (i.e. Detection). ice extent, whereas it is well established that Turner J, Comiso JC, Marshall G, Lachlan-Cope TA, the observed changes have a strong spatial Ocean Modelling 64: 67-75. doi:10.1016/j. 2. Is directly attributable to some external Bracegirdle T, et al. 2009. Non-annular atmospheric pattern. In particular a strong increase in Ross ocemod.2013.01.003. forcing or combination of forcings (i.e. circulation change induced by stratospheric ozone Sea cover is counterbalanced by the strong Attribution). Meredith MP, King JC. 2005. Rapid climate change depletion and its role in the recent increase of decrease in Amundsen/Bellingshausen Sea in the ocean west of the Antarctic Peninsula during Antarctic sea ice extent. Geophys Res Lett 36: The methods used rely heavily on model ice cover. By using the spatial pattern of sea th the second half of the 20 century. Geophys Res doi:10.1029/2009GL037524. simulations. Given the short length of most ice trends and applying formal Detection and Lett 32:doi:10.1029/2005GL024042. Worby AP, Geiger CA, Paget MJ, Van Woert ML, observational records (especially in the polar Attribution methods, (Hobbs et al, 2014) show Nihashi S, Ohshima KI. 2001. Relationship Ackley SF, et al. 2008. Thickness distribution of oceans) models are usually necessary to that: characterise the system’s internal variability between the sea ice cover in the retreat and Antarctic sea ice. Journal Geophysical Research • Observed winter sea ice changes on multidecadal to century timescales. An advance seasons in the Antarctic Ocean. 113: doi:10.1029/2007JC004254. are small compared to model internal expected theoretical response of the system Geophysical Research Letters 28: 3677-3680. variability. doi:10.1029/2001gl012842. to an external forcing is also required, which is usually only obtainable from climate model • Very few coupled climate models are Paolo FS, Fricker HA, Padman L. 2015. Volume loss simulations. Therefore, the Detection and able to replicate the observed changes, from Antarctic ice shelves is accelerating. Science: Attribution method is also a comprehensive even accounting for internal variability. doi:10.1126/science.aaa0940. means of model evaluation. Applying these • The discrepancy between models and methods to the question of Southern Ocean observations occurs largely in the Ross sea ice change is invaluable for validating Sea. model projections, since the level of external

32 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 33 The short record of passive microwave References Sea Ice Dynamics off George V Land, East Antarctica, Beyond the observations of sea ice cover is a source of Instrumental Period significant uncertainty in these conclusions. Hobbs, W. R., N. L. Bindoff, and M. N. Raphael, 1 1-2 3 4 1 5 2 However, new work presented here that 2014: New Perspectives on Observed and Crosta X , Campagne P , Dunbar R , Escutia C , Etourneau J , Houssais M-N , Massé G , 1 compares century-scale proxy reconstructions Simulated Antarctic Sea Ice Extent Trends Using Schmidt S Optimal Fingerprinting Techniques. Journal of of sea ice cover is consistent with these 1 Climate, 28, 1543-1560, 10.1175/jcli-d-14-00367.1. UMR 5805 EPOC, Université de Bordeaux, 33615 Pessac Cedex, France findings. Where proxies are available they 2 UMI 3376 TAKUVIK, Université Laval, Québec City, Canada show a long-term pattern that agrees with Mahlstein, I., P. R. Gent, and S. Solomon, 2013: 3 UMR 7159 LOCEAN, Université Pierre et Marie Curie, 75252 Paris Cedex, France the models in the E. Antarctic, Weddell Historical Antarctic mean sea ice area, sea ice 4 IACT, CSiC-Universidad de Granada, 18100 Armilla, Spain Sea and Amundsen/Bellingshausen Sea. trends, and winds in CMIP5 simulations. Journal of 5 Department of Environmental Earth Systems Science, Stanford University, USA Both the models and reconstructions show Geophysical Research: Atmospheres, 118, 5105- [email protected] a decrease in ice cover from the early to 5110, 10.1002/jgrd.50443. mid-1960s. However, the magnitude of this Polvani, L. M., and K. L. Smith, 2013: Can natural change is small compared with the internal Antarctic sea ice is the most seasonal started to melt back to its modern position variability indicated by both the models and variability explain observed Antarctic sea ice trends? New modeling evidence from CMIP5. physical parameter on Earth, which waxing at ~18 kyrs BP in phase with the last simulations. Projections using only models and waning is of major importance for deglaciation. Off George V Land, deglaciation that are consistent with the observed sea ice Geophys Res Lett, 40, 3195-3199, 10.1002/ grl.50578. global climate through modulation of the was initiated at ~12 kyrs BP and lasted until climate indicate that the small response is Southern Hemisphere radiative balance, ~9 kyrs BP when a modern-type seasonal unlikely to be significant for the next two to Raphael, M. N., and W. Hobbs, 2014: The influence transfer of energy and gas at the ocean- sea ice cycle set up. Sea ice duration was three decades. of the large-scale atmospheric circulation on atmosphere interface, atmospheric and shorter during the 9-4 kyrs BP period (mid- A confounding factor is the Ross Sea, where Antarctic sea ice during ice advance and retreat oceanic circulation and regional and remote Holocene hypsithermal) and subsequently there are clear and significant discrepancies seasons. Geophys Res Lett, 41, 5037-5045, oceanic productivity. Antarctic sea ice cover increased during the 4-0 kyrs BP period (Late between the models and observations. A 10.1002/2014gl060365. slightly increased over the last decades, Holocene Neoglacial). This pluri-millennial opposite to numerical models’ output that trend resulted from long-term changes in number of hypotheses have been suggested Zunz, V., H. Goosse, and F. Massonnet, 2013: How infer a global decrease. Reasons of such an local seasonal insolation modulated by the to explain the Ross Sea changes, none of does internal variability influence the ability of increase, in the context of global warming, is memory effect of the ocean. Centennial and which are adequately represented in global CMIP5 models to reproduce the recent trend in still under debate but may rely on Southern pluri-decadal variations were superimposed coupled climate models. It is suggested that Southern Ocean sea ice extent? The Cryosphere, 7, Ocean atmospheric reorganization forced by onto the Holocene trend in sea ice duration, Antarctic Detection and Attribution efforts 451-468, 10.5194/tc-7-451-2013. should focus on using long-term model the anthropogenic-induced recent trend to including the last 2 kyrs. Off George V Land, experiments using high-resolution regional positive Southern Annular Mode (SAM) or the strong variations in sea ice duration over models, to overcome these uncertainties. on natural variability. The instrumental and the last 2 kyrs were out-of-phase compared historical observations are unfortunately too to the Northern Hemisphere climatic periods. short to robustly document the relationships The Dark Ages and Little Ice Age were between Antarctic sea ice and climate. generally warm while the Medieval Warm Proxy records from marine and ice cores Period and Current Warm Period were allow to reconstructing Antarctic sea ice mainly cold and icy as a result of changes cover beyond the instrumental period and to in the timing of spring sea ice melting and documenting the forcings of sea ice dynamics autumn sea ice freezing. Changes in the and their predominance and interactions timing of spring sea ice melting probably from geological to annual timescales. It is responded to the pluri-centennial expression worth noting that these forcings dictated of the Southern Oscillation Index (SOI) while sea ice dynamics mainly through changes changes in the timing of autumn sea ice in ocean and atmosphere temperatures and freezing responded to the pluri-centennial circulations. expression of the SAM. Variations in both sea ice proxies, SOI and SAM present Winter sea ice cover was twice the modern periodicities similar to solar activity cycles one during the last glacial period (30.000 to (Gleissberg and Suess cycles) showing an 18.000 years before present, kyrs BP) and influence of solar activity on atmospheric and

34 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 35 oceanic circulation through the modulation of creating a polynya downstream of the MGT. Fast-Ice Variability off East Antarctica & Challenges for Sea Ice the SOI and SAM. Last decades monitoring Concurrently, icier conditions are observed Science and geological proxy data have demonstrated off Dumont d’Urville (DDU). After a calving, that these two climate modes interact to the MGT cannot act as a dam anymore and Petra Heil shape inter-annual variations of Antarctic sea fast ice covers the formal polynya region. In Australian Antarctic Division and ACE CRC, University of Tasmania, Hobart, Tasmania, Australia ice cover. the same time, more open conditions prevail [email protected] off DDU. This “natural” opposite response At the pluri-centennial to pluri-millennial between the Mertz Polynya and DDU regions timescales, proxy records therefore indicate East Antarctic fast-ice variability Regional variability exists across scales and is not observed today, whereby the 2010 that the main forcings of sea ice cover and Maritime accessibility of the Southern is largely driven by boundary conditions, calving conducted to heavy sea ice conditions duration off George V Land are precessional Ocean all the way to the Antarctic coastline such as shape of coastline (including in both regions. insolation, solar activity and thermohaline is a basic requirement to all engaged in islands) and bathymetry as well as nearby circulation. Other processes such as volcanic Investigation of several sediment cores off exploration of the Southern Ocean and the topographic features (including slope of activity and, more locally, glacial discharge George V Land and Adélie Land suggests Antarctic coastline, in moving expeditioners nearby ice sheet, which affects local wind may have had a secondary influence. that regional sea ice evolution results from and supplies to/from their Antarctic stations, systems). Within the pack, the ice drift is the non-linear interaction of different forcing and conducting marine-based science. highly mobile and largely driven by synoptic At a shorter timescale, glacial processes are Operations off East Antarctica benefit from factors taking action at different timescales atmospheric systems. Surface ocean conversely of prime importance for sea ice the largely seasonal sea ice in that region. At (Figure above). Of special interest, the heavy currents and tidal forcing are important history off George V Land. Spectral analysis annual maximum extent, a belt of moving pack sea ice conditions observed today ensue drivers of sea ice motion in some regions, of a 250-year long record of local sea ice ice may extend in excess of 300 km north from the combination of the 2010 calving such as over the Antarctic Slope Current conditions reveals a ~70-year periodicity, from the coast off East Antarctica [Worby and the highly positive SAM. However, more or near shore [Heil et al., 2011]. Close to associated with the Mertz Glacier Tongue et al., 1998], while during late summer the paleo-data are needed to understand whether the coast land-fast ice is encountered. It (MGT) calving and regrowth dynamics. When sea ice retreats close to the East Antarctic these modern conditions represent a unique forms in situ and remains immobile to its long enough (~110-160 km long) the MGT coast. Nevertheless, in the so-called situation or already occurred in the past and, melt or until it breaks out and resumes life acts as a barrier to westward drifting ice Antarctic Paradox, contrary to accelerated if so, at which periodicity. within the pack. The extent of fast ice from and funnels katabatic winds, both processes loss of Arctic sea ice, the annual maximum Antarctic sea ice extent has been increasing the shore is highly variable and depends on slightly during recent years [Comiso and the local bathymetry, coastal protrusions or Nishio, 2008]. On 17.09.2014 the Arctic islands and any beset icebergs. Offshore sea ice reached its annual minimum extent, from Davis, the fast ice may extend up to 2 down to 5.02 million km , making it the 15 km from the coast, off Mawson the fast sixth lowest on record, while on 19.09.2014 ice has been observed more than 80 km the Antarctic sea ice reached 20.07million offshore [Fedotov et al., 1998], while the 2 km , a new maximum since routine fast ice off Casey is generally considered observations commenced in the 1970s as not reliable. Taken together, factors like [NSIDC, 2014]. A number of hypotheses these shape the logistical approach best have been suggested to explain the suited for any coastal Antarctic station. increase in Antarctic sea ice extent. Wind- driven changes, inducing both, direct ice Weekly or monthly fast-ice and snow advection (Bellingshausen and Amundsen thickness (plus auxiliary measurements) have seas) and changed thermodynamic ice been taken intermittently off Mawson and advance (remaining Antarctic pack-ice area) Davis stations since the mid-1950s [Mellor, Composite sea ice record off George V Land and Adélie Land (dark blue = diatoms, light blue = diatom have been proposed to drive expanding 1960]. Analysis of these data together with biomarkers) over the last 11,000 years along with the main forcing factors acting at the millennial to annual Antarctic sea ice [Holland and Kwok, 2012]. oceanic and atmospheric observations timescales. On the other hand, increased glacial and ice- suggest that fast-ice thickness off Mawson sheet melt forms expansive pools of newly Station is determined by both, the oceanic melted cold freshwater, which insulates sea heat transported onto the nearby shelf by ice from the underlying warmer salty deep intrusion of relatively warm Circumpolar water, aiding the expansion of Antarctic sea Deep Water and the strength and timing ice [Bintanja et al., 2013]. of local katabatics [Heil et al., 1996]. Off

36 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 37 Davis Station, however, the increased • Collaborations between the scientific In light of the shared need to understand Fraser, A.D., R.A. Massom, K.J. Michael, B.K. cyclonic activity, due to changes in the community and the (logistics) the changing Antarctic sea ice conditions, Galton-Fenzi, and J.L. Lieser, East Antarctic large-scale circulation in the troposphere operators are crucial to advance our developing active collaborations between landfast sea-ice distribution and variability, 2000– and lower stratosphere give rise to large understanding of current sea ice science and the national Antarctic programs 2008, . J. Clim., 25, 1137 — 1156, 2012. interannual variability in fast-ice thickness, changes and providing a future outlook. presents itself as key to address challenges Giles, A.B., R.A. Massom, P. Heil and G. Hyland, peaking during the 1990s, while the annual • A common language, shared faced by both parties. Semi-automated feature-tracking of East Antarctic maximum ice thickness has changed little understanding and common fast ice and pack ice from Envisat ASAR imagery, Rem. Sens. Envirnm., 115, 2267 — 2276, 2011. over our record [Heil, 2006]. However, dates approaches are required. of annual maximum ice thickness and final Acknowledgements • Standardised data collection format need Heil, P., R.A. Massom, I. Allison, and A.P. Worby, fast-ice breakout are delayed (each by +0.43 Physical attributes of sea-ice kinematics to be applied. Fast-ice observations off Mawson and Davis d yr-1), the latter consequently contribute during spring 2007 off East Antarctica, Deep- stations have been carried out under a number • Critical scientific issue to be addressed Sea Research II, 58, 1158 — 1171, 10.1016/j . to a prolonged persistence of the fast ice of projects, most recently under AAS2500 and are: dsr2.2010.12.004, 2011. (+0.67 d yr-1) [Heil, 2006]. Overall the AAS3328. All contributors are thanked for their fast-ice characteristics co-varied largely with • Consideration of scales: most sea assistance. Holland, P. R. and R. Kwok, Wind-driven trends atmospheric changes. This is also supported ice processes are scale-dependent! in Antarctic sea ice drift, Nat Geosci., 5, 827- MODIS data were obtained from the NASA by MODIS-derived fast-ice extent [Fraser 875,10.1038/NGEO1627, 2012. • Need to understand measurement Level 1 Atmosphere Archive and Distribution et al., 2012], which exhibits high year to year and model uncertainties. System (http://ladsweb.nascom.nasa.gov/). Mellor, M., Sea ice measurements at Mawson and as well as regional variability. For 2000 to Davis, 1954-58, ANARE Interim Report, 19, 34pp., 2008 their MODIS record shows a small but • Opportunities for operations to assist sea New data acquisition software for ASPeCt sea- 1960. NSIDC, Arctic sea ice reaches minimum ice research: ice observations is currently under development. statistically-significant increase (1.43±0.30% extent for 2014, 22.09.2014, http://nsidc.org/ Prototype data have been collected during yr-1) increase, which arose from increasing • Antarctica and the Southern Ocean arcticseaicenews, 2014. fast-ice extent in the sector from 20E to 90E are data sparse (including a lack Australian Antarctic voyages V1, V2 V3 2014/15 (under AAS5032: Dr. J. Lieser, Mr L. Symons, Mr S. (4.07±0.42% yr-1), which was counteracted of high-resolution satellite data). Worby, A.P., and I. Allison, A Technique for Langdon), during the Alfred-Wegener Institute’s Making Ship-Based Observations of Antarctic by a (statistically not significant) decrease in There is a critical need for more, PS89/ANT XXX-2 (Dr S. Schwegmann) and Sea Ice Thickness and Characteristics PART I standardised observations (e.g., the sector from 90o to 160oE (-0.40 ±0.37% during CHINARE -31 (S. Kong). All contributors Observational Technique and Results, Antarctic yr-1). These changes in East Antarctic fast ice ASPeCt [Worby and Allison, 1999] are gratefully acknowledged. CRC Research Report No. 14, ISBN: 1 875796 are likely contributors to the recent changes bridge-based ice-pack observations 09 6, 1999. in (East) Antarctic maximum pack-ice extent. to be collected on ALL voyages into References the southern pack-ice zone.) http:// Worby, A.P., R.A. Massom, I. Allison, V.I. Lytle, and Bintanja, R., G.J. van Oldenborgh, S.S. Drijfhout, For the wider Mawson region our record aspect.antarctica.gov.au/home/ P. Heil, East Antarctic sea ice: A review of its B. Wouters, and C.A. Katsman, Important role for structure, properties and drift, In: Antarctic Sea from the mid-1950s to current suggests that conducting-sea-ice-observations/ ocean warming and increased ice-shelf melt in Ice Physical Processes, Interactions and Variability severe fast-ice conditions (i.e., no breakout) ice-observation-software occurred only during the recent decade. Antarctic sea-ice expansion, Nat Geosci., 6, 376- (Ed. M.O. Jeffries), Antarct. Res. Ser., 74, AGU, 379, doi:10.1018/NGEO1767, 2013. However, there is anecdotal evidence that the • Event logging system to be Washington, 41— 68, 1998. integrated within standardised fast ice survived summer twice before, once Comiso, J. C. & Nishio, F. Trends in the sea ice in the early 1960s and again in the 1990s. underway (nautical and cover using enhanced and compatible AMSR-E, Recent progress in the analysis of high- meteorological) database. If no SSM/I, and SMMR data, J. Geophys. Res. 113, resolution satellite imagery [Giles et al., 2011] underway system, then adopt with C02s07, 2008. event logging system. might aid to identify causalities and assist Fedotov, V. I., N. V. Cherepanov, and K. P. Tyshko, in providing short-term outlooks of fast-ice • Hosting sites of observation Some features of the growth, structure and conditions. networks, such as AWSes, metamorphism of East Antarctic landfast ice, in: Antarctic Fast-Ice Network [AFIN; Antarctic Sea Ice Physical Processes, Interactions Challenges in sea ice science http://seaice.acecrc.org.au/afin/] and Variability (Ed. M.O. Jeffries), Antarct. Res. Ser., The summary above draws on a number observatories. 74, AGU, Washington, 141— 160, 1998. of in situ measurements and remotely- sensed observations. Here we expand on • Making available shipping plans, observational and related challenges in sea and deploying own or scientific ice science. drifting (sea ice) buoys.

38 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 39 Summation of Antarctic Sea Ice: What We Know and Where We important role in sea ice trends. During fail to simulate the observed regional increase Should Go the season of sea ice advance SSTs south in Antarctic SIE (Bintanja and others, 2013). of 50°S have decreased over the last few It is quite probable that other important Phil Reid decades (Bintanja et al., 2013), although mechanisms are similarly missing from Australian Bureau of Meteorology the Bellingshausen Sea region is a distinct climate models. Table 1 contains an extended [email protected] exception to this. Freshening of the upper- list of questions raised within the SCAR ocean in the high southern latitudes, which Horizon Scan process. Answers to these and Here we will give a very brief outline of by various phases of ENSO (El Niño- acts to enhance sea ice growth by stabilising other questions might help us gain a better the current state of our knowledge of the Southern Oscillation), the strength of the the upper ocean and insulating it from ocean understanding of the complex interactions variability and trends in Antarctic sea ice SAM (Southern Annular Mode) and ozone heat, has been attributed to an increase in and subsequently help us close the gap extent (SIE), reiterating in general some depletion that determine atmospheric synoptic precipitation entering the Southern Ocean between model simulations and observations. (Liu and Curry, 2010) and increased basal of what has been presented so far in this patterns and ocean circulation (Harangozo, While our understanding of Antarctic SIE melting of ice shelves (Hellmer, 2012; workshop. We conclude with providing some 2006; Holland and Kwok, 2012; Liu et al., drivers might currently be incomplete there Pritchard et al., 2012; Rignot et al., 2013). suggestions and highlighting some initiatives 2004; Massom and Stammerjohn, 2010; are a number of national and international Recent research (Li et al., 2014) suggests of where we might go in the future in order to Simpkins et al., 2012; Stammerjohn et al., initiatives that, if supported, might help to a link between trends in the SSTs in the reduce risk to operations within a changing 2008, 2012). Wind, ocean currents, wave reduce the risk for Antarctic operators. These Tropical Atlantic and SIE in West Antarctica sea ice environment. action, iceberg distribution, precipitation, basal initiatives include developing and employing via atmospheric Rossby waves. Record- melt of ice shelves, SSTs and a number of a range of sea ice outlooks or forecasts; There has been an increase in net Antarctic breaking net sea ice extents (post-satellite other variables all play their role in distribution. from short term nowcasts to longer seasonal SIE over the last 30+ years (Comiso, era) over the last couple of years have But many of these variables are hard to outlooks. Much of this is beyond the scope of 2010; Parkinson and Cavalieri, 2012). This been attributed to combined impacts of quantify and even harder to model in relation one individual national Antarctic operator. A net increase, however, masks the strong atmospheric anomalies, SSTs and ocean to sea ice since their individual impacts on the solution to this is cooperative and coordinated contrasting regional differences in extent currents (Massom et al, 2014; Reid et al, ice are often non-linear. When combined, as efforts across nations on global initiatives. trends. Predominantly there is a trend 2015; Turner et al, 2013). in real life, these variables impact on the sea Several initiatives include: towards greater SIE in the Ross and Weddell ice in possibly counter intuitive ways. Kimura It is obvious that complex interactions seas and decrease in extent in the Western • Polar Prediction Project (PPP), whose and Wakatsuchi (2011) examine the large- between a range of drivers are responsible Antarctic Peninsula-Bellingshausen Sea mission is to: “Promote cooperative scale processes that influence the seasonal for the observed trends in Antarctic SIE. region (WAP-BS) (Figure 1). Trends in SIE international research enabling variability of Antarctic sea ice and find that There is not one simple hypothesis that fully are evident throughout the year and are very development of improved weather and there are regional and seasonal differences in explains the trends that we have observed. distinctly regional. These regional differences environmental prediction services for the these processes. Stammerjohn et al. (2008, The authors of the SCAR Antarctic and are similarly reflected in sea ice seasonality, Polar Regions, on time scales from hours 2012) suggest that there is a relationship Southern Ocean Science Horizon Scan particularly in the total duration of sea ice to seasonal.” between the variability of sea ice retreat and report state: Our understanding of the drivers (Stammerjohn et al., 2012). The positive trend sea ice advance in the subsequent year. and impacts of Southern Ocean and sea • Polar Climate Predictability Initiative in net SIE in the Antarctic is in contrast to ice change remains incomplete, limiting our (PCPI) which contributes to the the rapid decline in the Arctic (Stroeve et al., Various mechanisms are suggested for the ability to predict the course of future change development of GIPPS (the Global 2011). recent observed trends and their regional (Kennicutt et al., 2014). This incomplete Integrated Polar Prediction System) on distribution. Results from Holland and Kwok To put the recent Antarctic SIE trends into understanding is to some degree reflected time scales of a season or beyond. (2012) suggest that changes in atmospheric a longer-term perspective, there is some in climate model results. Simulation of net • International Ice Charting Working dynamics are impacting on regional sea ice evidence, based on ice-core proxies, that Arctic SIE from the latest CMIP5 climate Group (IICWG) which was formed in extent: wind-driven ice advection around regionally sea ice extent in the decades models, as analysed by Shu et al. (2014), order to promote cooperation between much of West Antarctica and wind-driven immediately prior to the satellite era was are consistent with the decreasing trend the world’s ice centres on all matters thermodynamic changes elsewhere. Turner more extensive than it has been in the last in observed SIE and, broadly, the spatial concerning sea ice and icebergs. et al. (2009) find that a link between ozone 3 decades (Curran et al, 2003; de la Mare, distribution of this change. However this is depletion and atmospheric circulation in • The Sea Ice Prediction Network which has 1997, 2008). not the case for Antarctic simulations, where as a mission to “Network scientists and autumn might play a role in the recent the sign of the trend of net SIE is incorrect. stakeholders to improve sea ice prediction We know that large-scale variability in sea ice increase in SIE. Other research suggests It has been suggested that not including ice- in a changing Arctic”. It is proposed that distribution, seasonality and concentration that various changes in SSTs and upper- shelf basal melt in climate models is one of a similar network be established for the on a year-to-year basis is largely modulated ocean freshening may also be playing an the reasons global coupled models currently Antarctic (SIPN South).

40 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 41 Trends in Antarctic sea ice extent: 1979-2014 Figure 2: Questions from SCAR Antarctic Holland PR and Kwok R, 2012 Wind-driven trends (x103 km2 per decade) and Southern Ocean Science Horizon in Antarctic sea-ice drift. Nature Geosci., 5(12), Scan (Kennicutt et al., 2014) that relate to 872–875 (doi: 10.1038/ngeo1627) sea ice. Our current understanding of the Kennicutt C and 69 others, 2014 A roadmap for processes controlling the mass, properties Antarctic and Southern Ocean science for the next and distribution of Antarctic sea ice (Q.15 and two decades and beyond. Antarctic Science, 1 - 16. Q.18) and its interaction with the atmosphere doi:10.1017/S0954102014000674 and ocean is inadequate to predict future Li X, Gerber EP, Holland DM, and Yoo C, 2015 A conditions with confidence (Q.6, Q.7, Q.19 and Rossby Wave Bridge from the Tropical Atlantic to Q.20). West Antarctica. J. Climate, 28, 2256–2273 doi: http://dx.doi.org/10.1175/JCLI-D-14-00450.1 References Liu J, Curry JA and Martinson DG, 2004 Bintanja R, van Oldenborgh GJ, Drijfhout SS, Interpretation of recent Antarctic sea ice Wouters B, and Katsman CA, 2013 Important role variability. Geophys. Res. Lett., 31, L02205, for ocean warming and increased ice-shelf melt (doi:10.1029/2003GL018732) in Antarctic sea-ice expansion. Nature Geosc., 6, Liu J and Curry JA, 2010 Accelerated warming 376-379 of the Southern Ocean and its impacts on the Cavalieri DJ, Parkinson CL, Gloersen P, and Zwally hydrological cycle and sea ice. Proc. Nat. Acad. Sci. H, 1996 updated yearly: Sea ice concentrations (PNAS), 107(34), 14 987–14 992 (doi:10.1073/ from Nimbus-7 SMMR and DMSP SSM/I-SSMIS pnas.1003336107) passive microwave data (1981-2011). National Maslanik J, and Stroeve J, 1999 updated daily: Snow and Ice Data Center, Boulder, CO, digital Near-Real-Time DMSP SSM/I-SSMIS Daily Polar media. [Available online at http://nsidc.org/data/ Gridded Sea Ice Concentrations. National Snow nsidc-0051.html] and Ice Data Center, Boulder, CO, digital media. Comiso JC, 2010 Variability and trends of the [Available online at http://nsidc.org/data/docs/ Figure 1: Trends in Antarctic sea ice extent for the period 1979-2014 (x103 km2 per decade). Stippiling global sea ice cover. In: Thomas, DN, Dieckmann, daac/nsidc0081_ssmi_nrt_seaice.gd.html] shows regions of statistical significance (p < 0.01). Distinct regional differences in the signs of trends GS (Eds.), Sea Ice, 2nd Edition. Wiley-Blackwell, Massom RA and Stammerjohn SE, 2010 Antarctic exist, with predominantly the Ross and Weddell seas showing positive trends while the Amundsen and Oxford, 205-246 Bellingshousen seas experiencing negative trends. Some interesting things to note are the general sea ice change and variability – Physical and propagation over time of trends (positive and negative) to the east during the sea ice mid-advance season Curran MAJ, van Ommen TD, Morgan VI, Phillips ecological implications. Polar Sc., 4, 149-186 KL and Palmer AS 2003 Ice core evidence for (March/April through to August). This eastward propagation is particularly evident in the: Ross Sea; Massom RA, Reid P, Stammerjohn S, Barreira S, Antarctic sea ice decline since the 1950s, Science. Amundsen into the Bellingshausen seas; and Weddell Sea into the western Indian Ocean sector. A similar Scambos T & Lieser J, 2014 [Antarctic] Sea ice Vol. 302(5648), pp. 1203-1206. doi:10.1126/ westward propagation is present in the sea ice retreat season. Data are NASA Team, based on Cavalieri et extent and concentration, in “State of the Climate in science.1087888 al (1996) and Maslanik and Stroeve, (1999) 2013”. Bull. Amer. Met. Soc., 95(7), S150-S152 de la Mare WK, 1997 Abrupt mid-twentieth-century 6. What controls regional patterns of atmospheric and oceanic warming and cooling in the Antarctic and Parkinson CL and Cavalieri DJ, 2012 Antarctic sea decline in Antarctic sea-ice extent from 41 whaling Southern Ocean? ice variability and trends, 1979–2010. Cryosphere, records, Nature, 389(September), 57–60 7. How can coupling and feedbacks between the atmosphere and the surface (land ice, sea ice and ocean) be 6(4), 871–880 (doi:10.5194/tc-6-871-2012) better represented in weather and climate models? de la Mare WK, 2008 Changes in Antarctic Reid P, Stammerjohn S, Massom RA, Scambos 15. What processes and feedbacks drive changes in the mass, properties and distribution of Antarctic sea ice? sea-ice extent from direct historical observations T and Lieser J, 2015: The record 2013 and whaling records, Climatic Change, 92(3-4), 16. How do changes in iceberg numbers and size distribution affect Antarctica and the Southern Ocean? Southern Hemisphere sea-ice extent maximum. 461–493, doi:10.1007/s10584-008-9473-2 17. How has Antarctic sea ice extent and volume varied over decadal to millennial time scales? Annals of Glaciology, 56(69), 99-106 (doi: Harangozo SA, 2006 Atmospheric circulation 10.3189/2015AoG69A892) 18. How will changes in ocean surface waves influence Antarctic sea ice and floating glacial ice? impacts on winter maximum sea ice extent in Shu Q, Song Z, and Qiao F, 2015 Assessment of 19. How do changes in sea ice extent, seasonality and properties affect Antarctic atmospheric and oceanic the west Antarctic Peninsula region (1979- sea ice simulations in the CMIP5 models, The circulation? 2001). Geophys. Res. Lett., 33(2), L02502. 4, pp. Cryosphere, 9, 399-409, doi:10.5194/tc-9-399- 20. How do extreme events affect the Antarctic cryosphere and Southern Ocean? 10.1029/2005GL024978 2015 23. How will changes in freshwater inputs affect ocean circulation and ecosystem processes?

Figure 2

42 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 43 Stammerjohn SE, Martinson DG, Smith RC, Yuan SESSION 5: SEA ICE NAVIGATION: OPERATIONAL REQUIREMENTS X, and Rind D, 2008 Trends in Antarctic annual AND TECHNOLOGIES sea ice retreat and advance and their relation to El Niño–Southern Oscillation and Southern On-Site Sea Ice Information (OSSI): A System to Support Annular Mode variability. J. Geophys. Res., 113, Operations in Polar Regions doi:10.1029/2007JC004269 Thomas Krumpen, Lasse Rabenstein, Stefan Hendricks and Paul Cochrane Stammerjohn S, Massom R, Rind D, and Martinson D, 2012 Regions of rapid sea ice change: an inter- Drift & Noise Polar Services GmbH, Germany hemispheric seasonal comparison. Geophys. Res. [email protected] Lett., 39(6), L06501, doi: 10.1029/2012GL050874 Stroeve J, Serreze M, Holland M, Kay J, Maslanik Decision making for activities in ice-covered oceans can be improved by an increased level J and Barrett A, 2011 The Arctic’s rapidly shrinking of information. The current potential of available remote sensing products is far from being sea-ice cover: A research synthesis. Clim. Change, exploited due to limited usability for non-experts. Main usability aspects are data unification, 110, 1005–1027 automated processing and ordering, early availability, smaller data packages and improved Turner J, Hosking JS, Phillips T and Marshal visualization functions. The OSSI (On Site Sea-ice Information) data-stream was developed to GJ, 2013 Temporal and spatial evolution of address these issues. the Antarctic sea ice prior to the September 2012 record maximum extent. Geophys. OSSI is characterized by 100% automation, a modular structure designed for implementation of Res. Lett., 40(22), 5894–5898 (doi: any data product, data compression and a “with one order many data” principle. Improvement of 10.1002/2013GL058371) the availability of information needed for navigation or other activities in ice covered areas can be achieved for some remote sensing data by the usage of single swath data in comparison Turner J, Phillips T, Hosking JS, Marshall GJ to daily composites. For instance, swath sea ice concentration data is capable of visualizing and Orr A, 2013 The Amundsen Sea low. Int. J. Climatol., 33, 1818–1829. doi: 10.1002/joc.3558 sub-daily variations in the sea ice cover, such as tidal effects or changes in ice concentration in highly variable ice-edge zones. Furthermore, swath data is made available within two hours of acquisition; whereas daily composites are sent on board with a more than a 24-hour delay.

44 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 45 Developing and Delivering Operational Sea Ice Information for the To support the Copernicus programme, a number of established features remain, Polar Regions the Sentinel-1 satellite was launched on 3 new functionality and a redesigned layout, April 2014. Early in 2015, Sentinel-1 will including an expanded map view, provide a Andrew Fleming complete its commissioning phase and greatly improved interface for all users. begin to deliver SAR imagery to support British Antarctic Survey In the near future a number of new data a range of activities including sea ice [email protected] products and imagery from the Sentinel-1 monitoring in the polar regions. These new SAR satellite will be integrated into the images are freely available through the Polar website. Other options for accessing Polar Developed by an international consortium & Commonwealth Office and commercial View service. of ice charting experts led by the British income fund development and ongoing View information by email or the low- Antarctic Survey, Polar View in the Antarctic operations. The updated Polar View Antarctic website bandwidth interface are still available. provides a near-real-time sea ice information (www.polarview.aq) provides easy access SAR (Synthetic Aperture Radar) satellite The new and improved Polar View website service for ship operators. The service helps and better visualisation of information. While imagery provides excellent information about available at www.polarview.aq. them minimise delays, improve efficiency, and sea ice due the high spatial resolution and take action to avoid life-threatening safety ability to provide images despite cloud cover hazards, damage to vessels and potentially and during the polar winter. Since the loss of severe consequences for the environment. the ENVISAT satellite, a reduced coverage of It is widely used by commercial, fishing SAR data has been available thanks to the EC and tourist shipping, and by national polar Copernicus Marine Monitoring Service. research programmes. The European Space Agency, European Commission, UK Foreign

Ship operators in the Antarctic will also The project will also improve the delivery benefit from the EC Polar Ice project, and integration of information into user’s on- which aims to develop next generation board systems by providing an integrated sea ice information service by integrating service (including new information products and building on a wide range of existing and imagery from the new Sentinel-1 European and national-funded activities. This satellite) for delivery to Arctic and Antarctic project will develop a number of new sea ice marine operators. More information is monitoring services, specifically for the Arctic available at www.polarice.eu. and Antarctic regions, including sea ice pressure, thickness and forecast products.

European Sentinel-1a satellite

46 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 47

B Collaborative Antarctic Sea Ice Chart Penelope Wagner Norwegian Ice Service [email protected]

Over the years, the ice services of the This allows weekly Antarctic sea ice charts phase because ice charting services need International Ice Charting Working Group to be available for navigational planning information on how they can improve their (IICWG) have recognized the value of and potential short-term sea ice forecasting products for operational purposes. However, cooperative activities in ice services methods. The revised Antarctic sea ice the consistent sea ice archive has also supporting maritime navigation and product also includes information on sea ice shown potential to be used in the scientific polar environmental awareness. With the Stage of Development (SoD) and up-to-date community as another sea ice data source implementation of the World Meteorological iceberg information generated by the US for ingesting real-time data (i.e. ship-based Organization (WMO) SIGRID3 code used in NIC. During the beta testing phase, the new observations or buoy data) and short-term sea ice forecasting. geodatabases, the possibility of collaborative Antarctic sea ice chart product has shown to ice charting has made attaining this goal be an excellent upgrade to what was already From the operational community operational more realistic. Previously, the Arctic and available because the continuity allows for a community we know sea ice charts are not Antarctic Research Institute (AARI) and the better understanding for ice analysts when useful for real-time navigation but can provide U.S. National Ice Center (US NIC) produced assessing sea ice behaviour, timely updates useful information for navigational planning. a comprehensive Southern Ocean analysis on the ice shelves, improved iceberg tracking, The problem remains that there’s not a lot biweekly, and the Norwegian Ice Service and more satellite-derived sea ice information of information available on how to use the (NIS) analysed sea ice for the Antarctic included in the ice charts. This is especially data. From user surveys the community Peninsula and Weddell Sea on a weekly useful in the Antarctic Peninsula and Weddell would like sea ice charts to be updated basis during the austral summer (October- Sea areas where there is a lot of ship traffic more frequently and with timely delivery, April). Because of this overlapping area because the weekly Antarctic sea ice charts however sea ice charting organizations are of responsibility, these three ice services will produce analysis to this region twice a limited by infrequent high resolution satellite implemented a regular collaborative method week. cover needed for real-time data requests. to analyse and disseminate a weekly The science community recognizes the Antarctic sea ice analysis hence contributing Additionally, the US NIC is also making efforts importance of sea ice charts, especially to development of safer ice navigation in the towards improving iceberg analysis. The US because ice charts for the Southern Antarctic waters as well as regional sea ice NIC is responsible for the naming of the Ocean have had a consistent archive since charts for improved resolution. Per agency individual icebergs used by anyone tracking 1972 from the US NIC. It is used by some specific requirements, all three agencies icebergs including British Antarctic Survey researchers as a ground-truth proxy for sea agreed to use the SIGRID3 to code sea ice and British Navy. They track all icebergs over ice edge, extent, and climatology, however, concentration, sea ice type, sea ice form, and 18 km but are preparing to implement iceberg other opinions of researchers are that the icebergs. This new product requires AARI tracking limits > 5km using the Brigham bias in the charts from subjective nature from and the US NIC to alternate the weeks when Young University (BYU) Scatterometer Image ice analysts are not yet quantified. The current they provide comprehensive sea ice charts, in Reconstruction data product. Feedback from charts can be found at: http://ice.aari.aq/ addition to ingesting the NIS sea ice charts. the community is important during this testing antice/

48 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 49 Antarctic Weather Service Delivery – A Focus on Best Practice and Sustainability Scott Carpentier and Elizabeth Donovan Antarctic Meteorological Section, Australian Bureau of Meteorology [email protected]

Making the most from the weather evolving science and who has many other forecast considerations at the forefront of their mind. It is well recognised that timely, accurate and Where are Antarctic operators sourcing fit for purpose weather services underpin their weather information from? the safe and efficient running of Antarctic operations. However, weather and sea ice A 2015 World Meteorological Organisation information is not consistently made available (WMO) survey to the Council of Managers to all Antarctic operators. For example, of National Antarctic Programs (COMNAP) some National Antarctic programs are revealed that those COMNAP members who supported with tailor made products on an responded obtain their weather information Figure 1: WMO survey to COMNAP (18 responses). Where do you gather your weather information from? expedition specific basis by affiliated national from a near equal mix of products sourced meteorological or research agencies, whilst from National weather services and free Antarctic tourist, cargo and fishing vessels internet sites (Figure 1). COMNAP has The 18 respondents from COMNAP reported Daily to sub-daily updates and high resolution are mainly left to their own devices to acquire 29 member countries and 18 responses wind, horizontal visibility, precipitation, weather products are reported as most weather and sea ice information. to the survey were received. Personal temperature, sea state and sea ice as the desired. Included in the mix of routinely communications and a 2014 survey from the main forecast elements that are imperative accessed weather products are cutting edge Moreover, a key challenge to gaining International Ice Charting Working Group for reducing cost and risk to their activities but often non-validated products such as the most from the information at hand is (IICWG) to the International Association of (Figure 2). Around 80% of the respondents sea ice motion and thickness analyses and having an underlying understanding of the Antarctic Tour Operators (IAATO) indicated reported that these forecasts are imperative forecasts. limitations of the forecasts. In other words, that IAATO members rely less on National for tactical decision making (i.e. < 5 days). is the model well-grounded with accurate weather services, preferring instead to observations at the time of analysis? Is there use free internet sites and ship based sufficient horizontal and vertical resolution visualisations of raw model weather data in the model to capture moisture exchange which can be freely emailed from websites through cracks in the sea ice; ocean waves such as zygrib (http://www.zygrib.org/) or through the marginal ice zone; the extent sailmail (http://www.sailmail.com/). of katabatic winds slumping onto the coast; barrier winds; tip and gap jets; blocked flow; standing waves and lee vortices? How well do the model physics and parameterisations handle cloud phase, such as super cooled droplets, which is an important consideration for cloud and precipitation forecasts? What about the effects of turbulence? Placed in the context of continuous changes and ameliorations in observing, computing and modelling techniques, the challenge of really understanding the information at hand is considerable for the professional meteorologist, let alone the Antarctic operator who is less familiar with the ever Figure 2: WMO survey to COMNAP (18 responses). How useful are the following forecast elements?

50 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 51 Satellite images, manual and automatic organisation would benefit from a weather weather station observations are highly service provided from Antarctica whilst only regarded by COMNAP (Figure 3) and IAATO ~30% of respondents currently benefit from members alike. Also of note is that around such a service (not shown). 70% of respondents indicated that their

Figure 4: WMO survey to COMNAP (18 responses). Please grade the impact of atmospheric weather information on your marine operations?

Due to the harsher winter environment, weather data is limited by communication most commercial, private or governmental bandwidth. This figure is likely on-par if not Antarctic operations occur over the summer larger for the smaller tourism and commercial season. Tourism is concentrated over the operators represented by IAATO. Graphical Antarctic Peninsula region whilst National products viewed or delivered via the world- programs sporadically cover the breadth of wide-web and email is preferred, though plain the continent. language text forecasts and on-site briefings are also important weather service delivery Around 70% of COMNAP respondents to methods for COMNAP members (Figure 5). the WMO survey noted that their access to

Figure 3: WMO survey to COMNAP (18 responses). How useful are the following observations?

Weather information becomes less critical for of respondents, with a particular emphasis operational and strategic planning as opposed on sea ice outlooks (not shown). Only one to shorter time-scale tactical decision making respondent answered that their service (Figure 4), though a requirement for seasonal requirement for seasonal to decadal to decadal atmospheric and oceanographic atmospheric and oceanographic outlooks is outlooks is still noted as useful by >50% currently being met.

Figure 5: WMO survey to COMNAP (18 responses). How is your weather service currently provided?

52 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 53 The current state of Antarctic analysis schemes such as 4-dVar and Though inefficient, the current service weather service delivery non-linear modelling techniques carried delivery model that has National Antarctic out by super-computers. Each step of this Historically and to this day most National programs relying on their respective and process is characterised by errors and Antarctic weather services were/are built often remote homeland National weather assumptions. Given the complexity of the from small dedicated groups operating at the services does have historical momentum, field, weather knowledge is arguably best periphery of their homeland national weather addresses language barriers and satisfies gained through being informed and guided service. Some National meteorological national pride. Successful international co- services have forecasters with limited or no by competent Antarctic forecasters who investment into (a) consolidated Antarctic Antarctic specific experience or knowledge are cognisant of the weather impacts and weather service provider(s) will be conditional issuing the Antarctic forecasts. Given some critical weather thresholds that concern the to not only improving service delivery as of the unique attributes of Antarctic weather, Antarctic operator. Such skilled forecasters per the WMO model but also ensuring that it is no surprise that Antarctic operators look should employ various briefing techniques stakeholder nations are engaged throughout to augment their weather knowledge beyond to ensure that the user fully comprehends the service design process. the information being provided. User their National weather services by seeking It is suggested that the nascent fields of out research centre and other free website education and training materials and good Figure 6: The WMO model for service Antarctic sea ice charting, sea ice forecasting communication techniques underpin the weather products that they may then interpret delivery and implementation. and climate services would benefit in transformation of weather information into with their own site specific knowledge and (Source: www.wmo.int/pages/prog/amp/pwsp/documents/ particular from considered development via experience. weather knowledge. WMO-SSD-1129_en.pdf) International collaboration and co-investment. However, by dealing with multiple service A model to build and expand Antarctic The service model is then expandable to providers, Antarctic operators carry more weather service capacity and user It is suggested that service capacity be other fields if found suitable, like aviation risk of misinterpreting the information at knowledge built up at the National and International forecasting for example. The World hand and are more susceptible to breaks in level through co-investment, commercial Meteorological Organisation, associated In early 2015, the WMO released service quality and continuity. For example, development and improved coordination to agencies and the Antarctic Treaty system National weather services are typically their strategy for service delivery and minimise duplication of effort and maximise provide an unparalleled framework for staffed 24/7 with quality assurance systems implementation, which cites key qualities to on skill and resource sharing. Consider that coordinating such international collaboration in place that mitigate risk of breaks in service. service delivery as shown in Figure 6. in the Prydz Bay region in East Antarctica, in Antarctic weather service delivery. Research products usually do not come with To meet the WMO criteria of authenticity, Indian, Chinese, Russian and Australian forecasters are all supplying their unique a guarantee for service timeliness as they are credibility, availability, timeliness, dependability interpretation of weather conditions for not designed for service to operators making and reliability in particular, it is suggested stations within a 100km distance. Such tactical decisions. Research products are also that current service-providing organisations multiplication of effort exists across several experimental by nature, so operators should collaborate to define best practice in the Antarctic regions. not expect continuity in product quality from science and implement quality management such sites. principles in both their training and product There is an important and often overlooked delivery systems. This can ensure that the step between weather information and highest-quality service is offered across weather knowledge. No matter how good Antarctica. the supplied weather information might be, Some Antarctic weather services are only its usefulness will always be limited by the staffed during the peak summer months user’s ability to interpret that information by forecasters on one-off temporary (i.e. turning information into knowledge) and placing it in the context of the operation assignments. An operational model that under consideration. Weather modelling offers continuity in employment and that is a fast evolving and highly complex field engages forecasters in competency training, based on multiple high-tech observing assessment and service building over the platforms (satellites, drifting buoys, Automatic quieter winter months would help build more Weather stations, aircraft, radar…), complex experience and knowledge into the pool of Antarctic forecasters.

54 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 55 Sea Ice Advisories for Indian Research and Supply Vessels SESSION 6: ALTERNATIVE TECHNOLOGIES Operating in Central Dronning Maud Land and East Antarctica Ice Navigation Support for Marine Operations of the Russian D. Ram Rajak,1 R. K. Kamaljit Singh,1 Jayaprasad P.,1 Sandip R. Oza,1 M. Javed Beg2 Antarctic Expedition 1Space Applications Centre, ISRO, Ahmedabad, India Korotkov A., Korablev V. and Lukin V. 2National Centre for Antarctic and Ocean Research, Goa, India Arctic and Antarctic Research Institute of Roshydromet , St. Petersburg, Russian Federation [email protected] [email protected]

Approaching Antarctica through formidable vigil on changing sea ice scenario near sea ice has always been challenge. With Antarctic coasts adjacent to Indian Antarctic Marine fleet remains the most cost-effective “Akademik Tryoshnikov” (2012), which were modern ice breakers and ice class vessels research stations Bharati and Maitri periodic type of transport for intercontinental cargo widely used in operations of the Soviet and things were eased out a little but effects Sea Ice Advisories (SIA) are relayed to supply and passenger shipping at the beginning from 1992 – Russian Antarctic Expedition st of global climate change are affecting vessel and the research bases. of the 21 century. Transport operations for (RAE). supporting activity of the National Antarctic Polar Regions differently and thus making The ice information-prognostic support of The advisories are based on the information Programs are not exclusion. Sea ice of the navigation easy in the Arctic and difficult shipping in the Antarctic is performed at the derived from multi-satellite data and products. Southern Ocean remains however a serious in Antarctica especially in the eastern part. present time through delivery of consultation Indian satellites capture Antarctic features obstacle for ship navigation in this Earth’s Satellite as well as navigational observations services of specialized ice centers or satellite in optical as well as microwave regions of region, like in the early 19th century when over past couple of years suggests increase receiving stations and ice experts onboard the electromagnetic spectrum at different spatial the sixth continent was discovered. Technical in lateral extent and thickness of sea ice and expedition ships and at the Antarctic stations. resolutions. While Resourcesat-2 LISS-IV characteristics of modern ships and their thus posing serious hurdle in the logistic The RAE applies in its operations the second (5m), Resourcesat-2 LISS-III (24m), and radio-navigation equipment ensure to a great operations and supply vessels not being approach for addressing this practical task. Resourcesat-2 AWiFS (56m) provide data in extent safety of shipping. Nevertheless, able to navigate to designated landing sites The Arctic and Antarctic Research Institute optical region; RISAT-1 Synthetic Aperture serious incidents also occur in our days with in Antarctica. For the Indian Station, Maitri (AARI) of Roshydromet (St. Petersburg) Radar (SAR) at multiple polarizations and ships beset in ice and damages of ship hull season 2010-11 being the worst as the has a multiyear school for training ice spatial resolutions, and SARAL AltiKa provide and propeller-rudder system. In the second supply vessel could not reach the Indian pilots providing practical assistance to ship data in microwave region. MODIS mosaic at part of the 20th century when planned Barrier and Station was starved of jet fuel navigators for ice shipping. These specialists 250m, sea ice concentration at 3.125km, and large-scale investigation of the Antarctic but managed to survive on the reserve stocks have vast practical experience in performing models predicted atmospheric parameters began involving international cooperation, accumulated over years. ship- and airborne ice observations and are the other major inputs which are used in the National Antarctic Programs of different in receiving, processing and interpretation Confronted with the sea ice challenges, a preparing near real time SIA. The SIAs are countries used two types of ship support for of satellite sea ice images in the Arctic programme on providing Sea ice advisory prepared by integrating information related their activity on the ice continent. The first and the Antarctic. Such specialists work was initiated in 2012. Satellite derived to sea ice concentration, thickness, drift, type applied by USAP was in using a powerful onboard the RAE expedition vessels in the information related to the current status of deformation, melting/refreezing trend, status USCG icebreaker for escorting transport and Antarctic and at the coastal Mirny, Progress, sea ice, fast ice, and icebergs en-route are of fast ice, predicted weather conditions etc. research vessels and tankers to the coast of Novolazarevskaya and Bellingshausen acquired, processed and verified with limited India has successfully made use of earth Antarctica. The second type was in employing Antarctic stations. In addition to receiving and ground controls wherever possible. Space observation data for sea ice monitoring ice-strengthened ships-suppliers that were processing satellite ice images, personnel Applications Centre (SAC) at Ahmedabad required for SIA and plans to include more widespread in the Arctic shipping. Later, such of Antarctic stations also carry out coastal and National Centre for Antarctic and Ocean satellite for refinement of advisories. countries as the USSR, Germany, Australia, observations of the state of landfast ice, Research (NCAOR) at Goa keep a constant France, RSA, Norway and Great Britain assessing all its age stages (beginning of began construction of special multipurpose seawater freeze up, establishment of landfast ships. Technical characteristics of such ships ice, onset of its decay and complete or partial combined capabilities of cargo, research, destruction). passenger and tanker vessels equipped with helicopter deck complexes to meet objectives Modern satellite information is presented in of their national Antarctic expeditions. In the different ranges of TV, IR, microwave and USSR and Russia, these were specially built radar sounding of the underlying surface. In research-expedition vessels “Mikhail Somov” the summer navigation season, continuous (1975), “Akademik Fedorov” (1987) and

56 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 57 ice observations from the pilot bridge at the for the Antarctic ice belt, which presents an In September-November each year, the not melt in summer and much later dates transit routes of research-expedition vessels agglomeration of different age types of ice. AARI develops a long-range forecast for the of landfast ice decay up to its remaining are added to satellite information. Primary entire forthcoming navigation season from unbroken. All data obtained are analyzed by forecaster, processing of the aforementioned mutually December to April. The forecast is based who assesses the character of development Remaining landfast ice in the area of Mirny complementary types of observations results on large-scale features of variability of the of ice processes compared to multiyear Observatory in 2002 for the first time in its in issuing summary composite ice charts Antarctic ice cover – seasonal change of the averages of sea ice characteristics (Figure half a century history had extremely negative at the AARI twice a month and in weekly sign of ice anomalies, prevailing quasi-biennial 1) and records the individual peculiarities logistical implications for RAE (Figure 2). quantitative assessments of sea ice extent in periodicity of sea ice extent fluctuations and of the year. The results of the analysis are Resupply of Novolazarevskaya station was the Southern Ocean. Shipborne data serve as opposition of the Atlantic and Pacific Ocean published in the RAE Quarterly Bulletin “State very difficult due to unbroken landfast ice in a unique material for verification of satellite ice massifs. A tendency for the increased of Antarctic Environment”. Belaya Bay in 2000, 2011, 2012 and 2014 information, which is especially important sea ice extent of the Southern Ocean, finally (Figure 3), which was used for unloading from manifested in the new millennium, and related 1986 and was practically annually cleared worsening of navigation conditions is also from landfast ice. taken into account. The main cause is the increased amount of residual ice that has

Figure 2: Variability of the dates of breakup of landfast ice at the roadstead of Mirny for a multiyear period (1956–2015).

Figure 1: Mean monthly (1) location of external, northern sea ice edge in February, May, September and December 2013 relative to its maximumо(2), average (3) and minimum (4) spreading in the Southern Ocean for a multiyear period.

58 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 59 Figure 3: Ice conditions in the area of Novolazarevskaya station on 14 March 2014 from TERRA satellite data.

A methodological principle of forecasting is to determine future ice conditions by the character of development of ice processes in the preceding period. The main problem is a reliable choice of the analogue-years. The priority criteria here are the dates of the main ice phases (ice formation, landfast ice formation and breakup) and the landfast ice thickness measured at the coastal stations (Figures 4 and 5).

Figure 4: “Secular” profile at the roadstead of Mirny with monthly measurements of landfast ice parameters every 100 m at 13 points, which is made in the unchanged form from 1964 and partly from 1956.

60 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 61 In general, a great deal of attention is paid development of recurring polynyas, usually • Operational preparation of navigation • A significant part of preparation work to Antarctic landfast ice as it serves as a used for choosing transit routes of ships. recommendations including a variant for unloading to sea ice is performed generalizing indicator of the character of It also regulates the dates of ice clearance of approach to each station and route by oceanographers, who work at the development of ice processes. The forecast in most regions being the only source coordinates. coastal stations. They choose in advance includes by all means the expected dates in summer for supplementing a rapidly the place of ship approach for fuel • Expert decision on the ways of of landfast ice decay in the operation areas. decreasing external belt of drifting ice. discharge, considering iceberg situation unloading. Breakup of landfast ice contributes to (Figure 6). • Direct provision of unloading from ship • In general, the programme of ice and to the shore – surveying of glacial oceanographic observations at the barriers and ramps, choice of the place stations is oriented to addressing as of stay, ice anchoring, marking out on a priority the RAE logistical tasks by landfast ice of cargo sites and runways means of determining the state of the for aircraft and laying of ice routes on ice cover. landfast ice, their equipment and safe operation. • The issued ice forecasts for the area of RAE ship operations are annually updated on the basis of actual data.

Figure 5: Multiyear variability of landfast ice thickness at the roadstead of Mirny (1956–2015).

Presence of landfast ice is a synonym of prognostic expectations and whether the ship elevated sea ice extent and complicated follows the worked our recommendations conditions of sea operations. On the one which are updated if necessary. Ship hand, this is decisively an obstacle to ship specialists begin to play here the main role. motion to the coast, and on the other, it is the only possibility of full value resupply of polar In complicated or uncertain situations, ice stations where there is no natural glacial or reconnaissance is performed by means of artificially created pier. Runways for aircraft helicopters. and helipads for helicopters are equipped and The work of ice pilots is as follows: pipelines for discharge of fuel and ice roads for loading-unloading of heavy transport • Timely submission to the Master and the vehicles and other bulky cargoes are built. Expedition Leader of comprehensive A forecast is transmitted to ships before their information on the actual ice situation for the operation areas obtained by means Figure 6: Cardinal change of iceberg situation near Progress Station for a quarter of a century. At the top, departure from Capetown to the Antarctic at left, a view from the stations to the northeast on 15 February 1988. At the top, right, on 10 April 2013. At of combining data of all ice observations the beginning of December. From that time the bottom, on 21 December 2012. the RAE Logistical Center only monitors both carried out onboard ship and at the transit of ships mainly checking consistency stations and prognostic and analytical of the actual development of ice events with information received from the AARI.

62 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 63 The use of Remotely Piloted Aircraft (RPA) in a Sea Ice Many Antarctic programs are already using/ included simulators, miniature toy drones, Reconnaissance Role: Pre-commitment Contemplation and Some trialling multicopter technology in various indoor, and outdoor flying with full-scale UAS Practical Results applications, from operational support to both with and without camera payloads, and science: the British Antarctic Survey took under a range of weather conditions prior 1 2 D.E. Thost & G. D. Williams two small RPA (as well as a tethered balloon) to actual flight testing in the field. In short, 1 Australian Antarctic Division on the RV Ernest Shackleton in the 2013-14 adequate training ensures successful and 2 Institute for marine and Antarctic Studies season, with the intention to help the vessel safe field deployments.” [email protected] with sea ice reconnaissance very close to the In March and April 2015, Australian ship; CHINARE has been using RPA to aid in researchers aboard the U.S. Antarctic surveying their runway site in the Larsemann Program (USAP) research vessel With approximate charter costs of $100,000 by speeding or slowing the multiple downward Hills (pers. comm. Cheng Xiao); in the Nathaniel B. Palmer carried out trials per day ($4,200 per hour) for a ship the thrusting motor/propeller units. They rely Australian Antarctic Program, Arko Lucieer of two RPA, with special permission - size of the RSV Aurora Australis, cumulative on an on-board computer (running complex from UTAS has used model helicopters and following a separate review from the navigation decisions that result in delays to mathematical algorithms) for stable flight: no octacopters at Casey for moss bed surveys National Science Foundation, as part of a the shipping programme can prove costly. computer, no flying. (2009-10, 2011-12). Small, inexpensive, easily deployable aerial research cruise in the Southern Ocean. Their advantages include: observation solutions have the potential Already, tourist and private operators are The flights tested the aerial mapping of to revolutionise the way ice navigation is using RPA to capture stunning footage • Ease of deployment/recovery, small Antarctic sea ice to determine floe-size traditionally done. The ability to “go up and of the Antarctic landscape on their cargo footprint. distribution, important for future integrated have a look” may becoming the norm, taking holiday/adventure (e.g. https://vimeo. observation programmes investigating minutes to accomplish, rather than being an • Off-the shelf (Ready to Fly) options, with com/124858722, and https://www.youtube. the interaction of waves and ice at the expensive, labour intensive proposition taking high resolution cameras integrated in com/watch?v=yA16wGvwFk4 ). Some margins of the ice. Preliminary results, and hours. the airframe. other recent RPA deployments in Antarctica experiences in using RPA from the vessel, have been less successful, resulting in • RTF options affordable will be presented. The RPA can be seen as a useful, re- uncontrolled fly-aways or immediate crashes deployable, “eye in the sky” providing timely • Vertical Take-off and Landing (VTOL) on deployment. An Australian commercial On the legal front, most countries’ civil answers during, or prior to, ice transit, not fishing company used two quadcopters in the aviation authorities (e.g. FAA, CASA) are necessarily replacing traditional helicopters • Hexa- and Octacopters have motor 2014-15 season in the Ross Sea, and both playing “catch-up” as the technology, and when they are on-board (with the ability to redundancy as a safety factor ended up “lost”: the first crashed shortly after uptake of it by consumers, outstrips their travel much further to investigate conditions • Unlike an Aerostat, does not require it took off and the other landed on sea ice ability to anticipate the required rules and ahead), but complementing them. On voyages hard-point attachment (winch system) and then fell into the sea. These incidents regulations. In September last year, the US that do not have helicopters, they may prove may be attributed to poor, incorrect, or no Antarctic Program prohibited the use of to be the best means of assessing areas of • With experience, may be flown away compass calibration while attempting to fly any RPA without specific authorization from heavy ice prior to committing to them with the from the vessel to investigate areas of the RPA in flight modes that rely heavily on the NSF, while at the same time developing vessel. This may be particularly timely with an interest (endurance and broadcast range compass input to the flight controller, coupled an entire chapter on RPA use for their Air ageing vessel, and reduced enthusiasm from dependent). with pilot error and minimal training. Operations Manual. Both COMNAP and the operator for engaging with heavy ice. With some disadvantages too, including: IAATO are keenly aware of the pressure to Researchers from NOAA successfully use this technology that both researchers There is a huge variety of rotary wing • Endurance generally <15 minutes deployed a small hexacopter in the and tourists are bringing to bear (with IAATO RPA options available for the task of ice South Shetland Islands in 2011 and • Operations limited to wind speeds <20 announcing in May a ban on the recreational reconnaissance. Their primary advantage is 2013 for penguin and seal census knots, temperatures >-15 °C use of RPA). their Vertical Take-off and Landing (VTOL) work. One of their most important ability, and small cargo footprint. Traditional • Learning curve to operate successfully recommendations (http://link.springer.com/ single rotor designs range from toy model and fly in manual mode (flying in full article/10.1007%2Fs00300-014-1625-4 ) helicopters to expensive long-range, long GPS + compass dependent mode may based on their field experience was that “Pilot endurance machines (Figure 1a). Multicopters not be possible at high polar latitudes) training is essential and should include virtual (with 4, 6, or 8 engines/props; Figure 1b) simulations, indoor missions, and supervised • Quadcopter (cheapest option) becomes are mechanically simple, but aerodynamically outdoor missions. All pilots in [their study] uncontrollable if one motor/prop fails. unstable aircraft whose motion is controlled completed multiple hours of training that

64 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 65

A A b Appendix 3: Workshop Participants

Surname: First Name: Organisation: Email Address: Beg Javed National Centre for [email protected] Antarctic & Ocean Research Bryson Rob Australian Antarctic Division [email protected] Budd Kerridwen Bureau of Meteorology [email protected] Carpentier Scott Australian Bureau of [email protected] Meteorology Figure 1: Two extremes (a) A $US400,000 Schiebel CAMCOPTER S-100, used by the French, Brazilian, Cheng Xiao Beijing Normal University [email protected] and Italian Navies, capable of autonomous missions of 10 hours endurance with 30 kg payloads, 200 km Cincotti Vincenzo ENEA [email protected] from its point of origin (b) An octocopter (DJI S1000) of the type used on the RV Nathaniel B. Palmer voyage in 2015. Approximate cost is $AU10,000 with an estimated endurance of 20 minutes, payload Clifton Robb AAD [email protected] dependent. Crosta Xavier Bordeaux University [email protected] Curran Mark AAD and ACE CRC [email protected] De Rossi Giuseppe ENEA [email protected] Donovan Elizabeth Bureau of Meteorology [email protected] Doyle Murray P & O [email protected] Fenton Gwen Australian Antarctic Division [email protected] Finnemore Michelle COMNAP [email protected] Fleming Andrew British Antarctic Survey [email protected] Fleming Tony Australian Antarctic Division [email protected] Frenot Yves IPEV [email protected] Frost Cameron Australian Antarctic Division [email protected] Fukamachi Yasushi Hokkaido University [email protected] Gales Nick Australian Antarctic Division nick.gales.aad.gov.au Godon Patrice IPEV / France [email protected] Heil Petra AAD [email protected] Hobbs Will ACE CRC [email protected] Hughes Nick MET Norway Johnston Greg Rescue Coordination Centre [email protected] New Zealand Knuth Margaret NSF/USAP [email protected] Krumpen Thomas DriftNoise.com Polar [email protected] Services Labudda Sharon Australian Antarctic Division [email protected] Lei Ruibo Polar Research Institute of [email protected] China

66 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 67 Surname: First Name: Organisation: Email Address: Surname: First Name: Organisation: Email Address: Lenel Sarah CCAMLR [email protected] Wang Jianzhong Polar Research Institute of [email protected] China Lieser Jan ACE CRC [email protected] Waterhouse David Australian Antarctic Division [email protected] Maksym Ted Woods Hole Oceanographic [email protected] Wicks Nicole AAD [email protected] Institution Massom Rob AAD and ACE CRC [email protected] Williams Guy IMAS [email protected] Mauderer Reg Australian Antarctic Division [email protected] Wilson David Rescue Coordination Centre [email protected] New Zealand McCarthy Peter Antarctica NZ [email protected] Wooding Robert Australian Antarctic Division [email protected] Millhouse Leanne AAD [email protected] Worby Tony ACE CRC [email protected] Mundy Jason Australian Antarctic Division [email protected] WU Yilin Chinese Consulate-General [email protected] O’Farrell Siobhan CSIRO Ocean and [email protected] in Sydney Atmosphere Flagship Yuan Shaohong Polar Research Institute of [email protected] Panowicz Caryn US National Ice Center [email protected] China Payne Bridget AAD [email protected] Zhang Lin National Marine [email protected] Environmental Forecasting Ramm David CCAMLR [email protected] Center Raphael Marilyn University of California [email protected] Reeve Jono Australian Antarctic Division [email protected] Reid Phillip Bureau of Meteorology [email protected] Reid Keith CCAMLR [email protected] Shiraishi Kazuyuki National Institute of Polar [email protected] Research Slocum Gill Australian Antarctic Division [email protected]

Stammerjohn Sharon Institute of Arctic and Alpine [email protected] Research, University of Colorado Boulder Sutton Matthew Australian Antarctic Division [email protected] Symons Lloyd Australian Antarctic Division [email protected] Tamura Takeshi National Institute of Polar [email protected] Research Thost Doug [email protected] Trousselot Chrissie Antarctic Tasmania Chrissie.Trousselot@stategrowth. tas.gov.au van Ommen Tas Australian Antarctic Division [email protected] Wagner Penelope Norwegian Meteorological [email protected] Institute Wallis Ben Ocean Expeditions & IAATO [email protected]

68 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 69 Appendix 4: Links to Further Information Appendix 5: Acknowledgements

The NOAA/BAMS Annual State of the The Year of Polar Prediction (YOPP), COMNAP wishes to thank the following and acknowledge their contributions (global) Climate Assessment Reports http:// scheduled to take place from mid-2017 to to the COMNAP Sea Ice Challenges Workshop, which would not have been www.ncdc.noaa.gov/bams-state-of-the- mid-2019, is one of the key elements of the possible without their support. climate/ Polar Prediction Project. YOPP will cover an The Sea Ice Prediction Network (SIPN), extended period of coordinated intensive The following organisations provided funding or in-kind support for the event: launched in fall 2013, builds and expands on observational and modelling activities in the Sea Ice Outlook project. The goal is to order to improve polar prediction capabilities The Australian Antarctic Division (AAD) develop a collaborative network of scientists on a wide range of time scales in both polar The Antarctic Climate and Ecosystems Cooperative Research Centre (ACE CRC) and stakeholders to advance research on regions and strongly engage in forecast- The Tasmanian Polar Network sea ice prediction and communicate sea ice stakeholder interaction, verification and a knowledge and tools. http://www.arcus.org/ strong educational component. The British Antarctic Survey (BAS) sipn/sea-ice-outlook/2014/post-season- http://www.polarprediction.net/yopp.html Institute Paul Emile Victor (IPEV) highlights The participants of the WCRP/SCAR The National Institute for Polar Research (NIPR) The International Ice Charting Working Group International Programme for Antarctic (IICWG) was formed in October 1999 to Buoys (IPAB) work together to maintain a Polar Research Institute of China (PRIC) promote cooperation between the world’s ice network of drifting buoys in the Southern Australian Bureau of Meteorology centres on all matters concerning sea ice and Ocean, in particular over sea ice, to provide Institute for Marine and Antarctic Studies (IMAS) icebergs. https://nsidc.org/noaa/iicwg/ meteorological and oceanographic data for real-time operational requirements and National Marine Environmental Forecasting Center, Polar Environmental Research & Antarctic Sea Ice Process and Climate research purposes. Forecasting Division (ASPeCt) is an expert group on multi- http://www.ipab.aq/ disciplinary Antarctic sea ice zone research US National Naval Ice Center within the SCAR Physical Sciences Woods Hole Oceanographic Institution programme. Established in 1996, ASPeCt has the key objective of improving our understanding of the Antarctic sea ice zone through focussed and ongoing field programmes, remote sensing and numerical modelling. The programme is designed to complement, and contribute to, other international science programmes in Antarctica as well as existing and proposed research programmes within national Antarctic programs. ASPeCt also includes a component of data rescue of valuable historical sea ice zone information. http://aspect.antarctica.gov.au/

70 12–13 May 2015 COMNAP Sea Ice Challenges Workshop 71 72 12–13 May 2015