• Project
  • Host
  • How to Apply

Overview

The Arctic is a unique, but fragile ecosystem that is increasingly threatened by changes to our climate. In this project, we will develop novel techniques for identifying small icebergs in satellite radar images. Algorithms will be developed and used to study the distribution and drift of icebergs providing valuable information to climate and ocean circulation studies. Furthermore, the output results will provide a valuable aid to navigation in ice infested waters. Collisions with iceberg pose a threat not just for lives and goods, but also for the environment, since they may result in oil spills or other ecological disasters. The reduction in sea ice across the Arctic has driven an increase in navigating these waters, especially through the North West and North East Passages which promise large reductions in travel.

This project will initially focus on the regions surrounding Greenland where accelerated ice flow is leading to changes in iceberg production. For example the Helheim Glacier is one of the largest tidewater glaciers in Greenland, where due to the highly crevassed front of the glacier, most calving events create small icebergs that enter the Denmark Strait [1]. The Helheim Glacier highlights the importance of detecting small icebergs, i.e. less than a hundred meters in diameter (for the visible part of the iceberg). After the algorithms are trained, the analysis will be extended to the full Arctic. In this project we will address two scientific questions: a) can we reliably detect small icebergs using the Synthetic Aperture Radar (SAR) onboard the Sentinel-1 satellites?; b) are the icebergs’ distribution and drift changing in time?

Unfortunately, past iceberg detection methodologies have struggled, especially for small icebergs and in circumstances when the icebergs are surrounded by sea ice. Recently, Dr. Marino (the lead supervisor) proposed a new algorithm which is able to increase the separability between icebergs and sea ice. This algorithm will be the starting point of this project. Once the algorithm is finalised the student will exploit freely available European Space Agency (ESA) Sentinel-1 data [2] covering the entire Arctic to use the detector to carry out environmental studies. 

Sentinel-1 SAR image acquired on 02/03/2015 in East Greenland. The land is in the top right corner and the bright spots in the dark background are icebergs.

Methodology

In this project, we will be using freely available Sentinel-1 data to detect small icebergs with a novel detection methodology.

The project includes two main stages:

1) The starting point is the iDPolRAD algorithm developed by Dr. Marino [3]. The algorithm shows unprecedented separability between icebergs and sea ice. However it does not currently run automatically and is therefore not optimal for very large datasets. The student will automate the detector for Sentinel-1 data designing an appropriate statistical test and a Geographical Information System (GIS) procedure.

2) The automatic detector will then be applied to very large historical datasets. There are historic SAR data starting from the 1980s (however older data have a coverage that is more sporadic than Sentinel-1). The student will analyse the data and extract eventual temporal trends.

Training and Skills

NERC CENTA students are required to complete 45 days training throughout their PhD including a 10 day work placement. In the first year, students will be trained as a single cohort on environmental science, research methods and core skills. Throughout the PhD, training will progress from core skills sets to master classes specific to CENTA research themes.

This is a multi-disciplinary project including topics related to (a) satellite Earth Observation; (b) radar; (c) image analysis; (d) glaciology and oceanography; (e) data handling and programming.

The successful candidate will have the opportunity to gain valuable skills in the context of: (a) analysing and processing satellite data using Python or Matlab; (b) developing statistical methodologies; (c) using Geographical Information Systems (GIS) software.

Timeline

Year 1: Preparing a literature review on the topics of microwave scattering from ocean, sea ice and icebergs and iceberg detection using SAR. Working on automation of the iDPolRAD.

Year 2: Finalising the automatic iceberg detector. Designing a GIS system to process and stack automatically data downloaded from the ESA data Hub. Expected submission of a journal paper on the automatic algorithm.

Year 3: Analysing the processed data linking results to climate changes or environment protection. Formulation of the thesis chapters. Expected submission of journal paper on the data analysis.

Partners and collaboration (including CASE)

This is a CASE project and the student will spend approximately 3 months at eOsphere Ltd, sited within the growing Space Cluster in Harwell. eOsphere has wide expertise in working with end users for exploitation of satellite data including many years’ experience utilising satellite imagery for ice monitoring for both scientific and practical applications. eOsphere also have extensive experience in working with PhD and Internship students.

As a collaboration, the project also includes a visit to the Alfred Wagener Institute in Bremerhaven, Germany, where the student will be working with Prof. Wolfgang Dierking, who is one of the World leading scientist in the field of iceberg observation.

Further Details

Students should have a strong background in Engineering and Physical Sciences or related subjects. They also should have enthusiasm for research, data processing and travelling. Experience of programming (e.g. Matlab, Python, etc) are desirable. The student will join a well-established team researching Earth Observation at the Open University.

Please contact Dr Armando Marino for further information (e-mail: armando.marino@open.ac.uk).

Applications should include:

Applications should be sent to

STEM-EEES-PhD-Student-Recruitment@open.ac.uk  

by 5 pm on Monday 22nd January 2018