Search Events

The 14th Symposium on Polar Science will be hosted by the National Institute of Polar Research (NIPR). The NIPR is organizing this annual symposium to present and promote a wide variety of polar scientific research and interdisciplinary studies.

The symposium will be conducted in-person, meaning participants are expected to attend at the physical venue. However, some sessions may be hosted in a hybrid format, combining in-person and virtual participation. Please stay tuned for future announcements regarding this possibility. Please note that the official language of the symposium will be English.

Extended deadline for abstract submission: 15 September 2023.

Abstract: Foundation models are artificial intelligence (AI) models that are pre-trained on large unlabeled datasets through self-supervision and then fine-tuned for different downstream tasks. There is increasing interest in the scientific community to investigate whether this approach can be successfully applied to domains beyond natural language processing and computer vision to effectively build generalist AI models that make use of different types of data. Here, IBM and NASA present the first end-to-end framework for pre-training and fine-tuning foundation models efficiently from a large source of geospatial data. We have implemented and applied this framework to produce Prithvi, a geospatial foundation model pre-trained on multispectral satellite imagery from the NASA Harmonized Landsat-Sentinel 2 (HLS) dataset. The framework supports automated statistical smart sampling strategies based on whether, land cover and other datasets to maximize impact and minimize waste of resources (e.g., avoiding areas and time ranges that would not bring any new information). Prithvi is a Temporal Vision Transformer that includes positional and temporal embeddings, which was trained on IBM Cloud Vela cluster (NVIDIA A100 GPUs) using a Masked Auto Encoder approach and Mean Squared Error loss function for a total of 10k GPUs hours. We demonstrated using the fine-tuning workflows built in our framework that Prithvi could be successfully fine-tuned to produce state-of-the-art AI models for Earth observation tasks: flood mapping, burn scar identification and multi-temporal crop classification. We carefully studied the impact of Prithvi's pre-trained weights on the downstream tasks by comparing learning curves for: (1) fine-tuning the whole model, (2) fine-tuning only the downstream task decoder, (3) training the model without taking advantage of Prithvi's pre-trained weights. Furthermore, given the scarcity of labeled data for Earth observation tasks, we progressively decreased the amount of labeled data available for fine-tuning the model to assess data efficiency. This analysis showed that using Prithvi we could achieve peak performance on test data quicker and with less training data (i.e. up to 50% less). Finally, in order to increase the impact of this work, the pre-trained model and fine-tuning workflows have been made publicly available through Hugging Face (https://huggingface.co/ibm-nasa-geospatial).

Cyber2A is a newly funded NSF effort aiming to provide cybertraining to geoscientists, in particular Arctic scientists, in modern artificial intelligence (AI) techniques. The effort’s long-term goal is to establish a sustainable and scalable learning community. This will be accomplished through customized training in cutting-edge AI for both current and next-generation geoscientists who are interested in applying AI to solve scientific problems.

The European Union-funded Arctic Black Carbon impacting Climate and Air Pollution (ABC-iCAP) project is organizing a 3rd online research collaboration with European and North American fire, smoke, and climate experts. This 90-minute webinar will take place 13 November 2023 at 10:00 a.m. Eastern Time/6:00 a.m. Alaska Time (15:00 UTC).

The organizers’ aim is for the projects and people to learn about ongoing work and describe current and near-term future trends of boreal and Arctic fire and smoke.

The webinar will be recorded and shared with participants, as well as posted online.

This session will include briefings from federal agencies on their polar activities and updates from US delegates to the International Arctic Science Committee and the Scientific Committee on Antarctic Research.

CATCH Seminar Series

This is a monthly online seminar series which hosts talks on CATCH related topics with the aim of fostering engaging scientific discussions between researchers on outstanding questions for our communities. This is open to researchers at all stages of their careers and we hope this seminar series provides researchers with the opportunity to interact and encourage potential future collaborations.

Format
Presentation from guest speaker(s) (30 minutes)
Q&A and open discussion (30 minutes)

UNE North hosts the fall meeting of 42˚N: The North Atlantic Network, formerly known as New England Arctic Network (NEAN). Please register for the meeting. In person and remote options are available.

Meeting Objectives

• Reconvene and nurture the 42˚N: The North Atlantic Network
• Continue to provide a forum for community-based research
• Engage students, including the UMaine & UNH NSF Research Traineeships (SAUNA & CARPE)
• Discuss and develop ideas for a community-based research proposal

42˚N: The North Atlantic Network combines the wealth of academic expertise in Arctic research across New England with researchers, stakeholders, and external partners concerned with environmental, economic, and social impacts. The network provides an ideal community for anticipating and responding to change in the Arctic and its impacts on the eastern coast of North America.

A number of meandering rivers in the Arctic, especially in Siberia, have point bars with strikingly structured permafrost polygon patterns. These seem to mimic not just the scroll ‘bar’ lines that are often visible in permafrost-free point bars, but the radial growth lines as well that tend to be perpendicular to the scrolls and have not been observed before. We use a Unet-type convolutional neural network to map the polygons. The network is trained with image tiles derived from manually mapped scroll- and radial lines on nine point bars. Post-processing of the semantic segmentation results allows the identification of individual polygons; a Python ‘networkx’ graph is used to study the relations between the polygons. A simple model of meandering suggests a preliminary explanation of how these beautiful patterns might form.

The Permfrost Discovery Gateway is a NSF and Google.org funded intelligent data management platform created for big data creation and discovery to support knowledge-generation in the Arctic permafrost region. The project is aimed to enable knowledge-generation and community-preparedness by creating big geospatial data products of permafrost thaw features from remote sensing imagery, developing AI tools to identify and track information within the big geospatial data, and building user-friendly online tools to enable scientific discovery, policy, and to empower Arctic communities facing permafrost thaw.

The next IGS Nordic branch meeting is co-arranged by Aalto University and CSC – IT Center for Science Ltd. and held in person on at Aalto University in Ottaniemi, Espoo, Finland.

The Nordic Branch of the IGS meeting provides an opportunity for Nordic-based scientists and students in glaciology to present their latest results and projects.

The deadline for submitting abstracts is 31 October 2023.

Ilisimatusarfik, the US National Science Foundation, and the US Consulate in Nuuk invite you to a presentation about the National Science Foundation and Greenland.

Have you ever wondered about the groundbreaking research that shapes our future?   Are you curious about the critical role of the National Science Foundation in advancing science, and the work they fund in Greenland?  Then please join us for a presentation about the National Science Foundation (NSF) and Greenland. Learn about the NSF generally and its work with and in Greenland including the Joint Science Education Project (JSEP), a field-based science program for students from Greenland, the U.S., and Denmark. The presentation will be offered in English with translation to Greenlandic. Light refreshments will be offered.

When: Tuesday, 7 November, 14.00 - 16.00
Where: University of Greenland – Ilisimatusarfik - Canteen at Campus Ilimmarfik
Who: Students of all ages are invited, as well as the general public. Representatives of the National Science Foundation will attend, as well as members of the U.S.-based and international research community.

The Arctic Research Consortium of the US (ARCUS), is accepting applications for three Arctic Indigenous Scholars to travel to Washington, DC in February 2024 for an opportunity to meet with officials at US government agencies, non-governmental organizations, and other groups working on Arctic issues. Travel expenses and per diem will be provided. While the selected scholars are in DC, ARCUS will arrange meetings with officials at relevant agencies and organizations, where scholars will be able to share their interests, learn of available resources, build toward collaborative relationships, and provide on-the-ground perspectives to key decision-makers. We define a scholar as an expert within their own knowledge system. This includes hunters, fishers, and gatherers; those that process and store food; health aides; and others. It includes youth, elders, and adults. A scholar’s education may have come from the land, the water, or a classroom. This program is supported by the National Science Foundation's Division of Arctic Sciences.