{"id":7327,"date":"2025-01-13T09:48:46","date_gmt":"2025-01-13T08:48:46","guid":{"rendered":"https:\/\/nersc.no\/?p=7327"},"modified":"2025-01-13T09:57:16","modified_gmt":"2025-01-13T08:57:16","slug":"nansen-guest-lecture-andrew-mcdonald","status":"publish","type":"post","link":"https:\/\/nersc.no\/en\/events\/nansen-guest-lecture-andrew-mcdonald\/","title":{"rendered":"Nansen Guest Lecture: Andrew McDonald"},"content":{"rendered":"

[vc_row][vc_column]

<\/div>[\/vc_column][\/vc_row][vc_row][vc_column width=”2\/3″ css=”.vc_custom_1693479579188{background-color: #ffffff !important;}”]

Nansen Guest Lecture 16.01.2025<\/h1><\/div>

Andrew McDonald, University of Cambridge and British Antarctic Survey<\/h2><\/div>[vc_column_text css=””]<\/p>\n

<\/h2>\n

“Short-term, high-resolution sea ice forecasting with diffusion model ensembles”<\/h2>\n

Sea ice plays a key role in Earth\u2019s climate system and exhibits significant seasonal variability as it advances and retreats across the Arctic and Antarctic every year. The production of sea ice forecasts provides great scientific and practical value to stakeholders across the polar regions, informing shipping, conservation, logistics, and the daily lives of inhabitants of local communities. Machine learning offers a promising means by which to develop such forecasts, capturing the nonlinear dynamics and subtle spatiotemporal patterns at play as effectively\u2014if not more effectively\u2014than conventional physics-based models. In particular, the ability of deep generative models to produce probabilistic forecasts which acknowledge the inherent stochasticity of sea ice processes and represent uncertainty by design make them a sensible choice for the task of sea ice forecasting. Diffusion models, a class of deep generative models, present a strong option given their state-of-the-art performance on computer vision tasks and their strong track record when adapted to spatiotemporal modelling tasks in weather and climate domains. In this talk, I will present preliminary results from a IceNet-like [1] diffusion model trained to autoregressively forecast daily, 6.25 km resolution sea ice concentration in the Bellingshausen Sea along the Antarctic Peninsula. I will also touch on the downstream applications for these forecasts, from conservation to marine route planning, which are under development at the British Antarctic Survey (BAS). I welcome ideas and suggestions for improvement and look forward to discussing opportunities for collaboration between NERSC and BAS, given the excellent recent work (e.g., [2, 3]) from researchers at NERSC applying deep learning to sea ice modelling.<\/p>\n

[1] Andersson, Tom R., et al. “Seasonal Arctic sea ice forecasting with probabilistic deep learning.”\u00a0Nature communications<\/em>\u00a012.1 (2021): 5124.\u00a0https:\/\/www.nature.com\/articles\/s41467-021-25257-4<\/a><\/span><\/p>\n

[2] Finn, Tobias Sebastian, et al. “Towards diffusion models for large-scale sea-ice modelling.”\u00a0arXiv preprint arXiv:2406.18417<\/em>\u00a0(2024).\u00a0https:\/\/arxiv.org\/abs\/2406.18417<\/a><\/span><\/p>\n

[3] Palerme, Cyril, et al. “Improving short-term sea ice concentration forecasts using deep learning.”\u00a0The Cryosphere<\/em>\u00a018.4 (2024): 2161-2176.\u00a0https:\/\/tc.copernicus.org\/articles\/18\/2161\/2024\/<\/a><\/span><\/p>\n

[\/vc_column_text][\/vc_column][vc_column width=”1\/3″ css=”.vc_custom_1695033709348{padding-right: 0px !important;}”]

\n

When and where?<\/strong><\/h2>\n

Thursday, 16.01.25 kl. 13:15 – 14:00.<\/p>\n

Copernicus lecture room, 1st floor, Nansen Center, Jahnebakken 3, Bergen<\/p><\/div>

\n

About Andrew McDonald<\/strong><\/h2>\n

Andrew McDonald is a PhD student in the\u00a0AI4ER Centre for Doctoral Training<\/a><\/span>\u00a0<\/span>at the University of Cambridge and\u00a0British Antarctic Survey\u00a0working with\u00a0Scott Hosking<\/span><\/a>\u00a0and\u00a0<\/span>Rich Turner<\/a><\/span>\u00a0<\/span>to advance the intersection of machine learning and climate science. His research aims to improve sea ice forecasts using diffusion models as part of the broader\u00a0<\/span>IceNet<\/a><\/span>\u00a0<\/span>project.<\/p><\/div>[\/vc_column][\/vc_row]<\/p>\n<\/div>","protected":false},"excerpt":{"rendered":"

[vc_row][vc_column][\/vc_column][\/vc_row][vc_row][vc_column width=”2\/3″ css=”.vc_custom_1693479579188{background-color: #ffffff !important;}”][vc_column_text css=””] “Short-term, high-resolution sea ice forecasting with diffusion model ensembles” Sea ice plays a key role in Earth\u2019s climate system and exhibits significant seasonal variability as it advances and retreats across the Arctic and Antarctic every year. The production of sea ice forecasts provides great scientific and practical value to […]<\/p>\n","protected":false},"author":4,"featured_media":3364,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_relevanssi_hide_post":"","_relevanssi_hide_content":"","_relevanssi_pin_for_all":"","_relevanssi_pin_keywords":"","_relevanssi_unpin_keywords":"","_relevanssi_related_keywords":"","_relevanssi_related_include_ids":"","_relevanssi_related_exclude_ids":"","_relevanssi_related_no_append":"","_relevanssi_related_not_related":"","_relevanssi_related_posts":"","_relevanssi_noindex_reason":"","footnotes":""},"categories":[82],"tags":[],"class_list":["post-7327","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-events"],"acf":[],"publishpress_future_action":{"enabled":false,"date":"2026-05-09 12:46:06","action":"change-status","newStatus":"draft","terms":[],"taxonomy":"category","extraData":[]},"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/posts\/7327","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/comments?post=7327"}],"version-history":[{"count":2,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/posts\/7327\/revisions"}],"predecessor-version":[{"id":7330,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/posts\/7327\/revisions\/7330"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/media\/3364"}],"wp:attachment":[{"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/media?parent=7327"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/categories?post=7327"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nersc.no\/en\/wp-json\/wp\/v2\/tags?post=7327"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}