Previous Seminars

Poisson lines and applications

Speaker: 
Christian Lantuejoul
Affiliation: 
Ecole des Mines de Paris
Seminar Date: 
4. June 2013 - 11:00 - 12:00

Poisson line simulations are a different kind of geostatistical simulations from the random fields previously presented by Hans Wackernagel. Random Poisson lines (for example the STIT tesselations) are random objects that can be used to simulate the random structure of cracks in rocks, or possibly the random arrangement of cracks and leads in sea ice.

The lecture will present the principles of random simulations, some algorithms and applications.

Trends in the Barents Sea sea ice using satellite data from the Arctic ROOS portal

Speaker: 
Natasha Ridenour
Seminar Date: 
29. May 2013 - 11:15 - 12:00

For countries such as Norway, that depend heavily on the ocean, information about oceanic conditions and sea ice is extremely valuable. This report provides an example of the application of operational oceanography data by utilizing sea ice area and extent data from the Arctic-ROOS portal (http://arctic-roos.org/observations/satellite-data/sea-ice/sea-ice-varia...) to analyze the trend in the Barents Sea ice cover for the time period 1978-2011. Data was collected by satellite remote sensing, where the measured variable was brightness temperature.

The Indian Ocean: Variability and modelling

Speaker: 
Syam Sankar
Affiliation: 
NERCI
Seminar Date: 
23. May 2013 - 12:30 - 13:00

1, An overview of the decadal to multidecadal variability of the Indian summer monsoon for the period 1400-2010 using observations and proxy data.

2, GOTM-ERSEM (one dimensional) model set up for the Indian Ocean region.

Climate variability seen in paleo-observations

Speaker: 
Katja Lohmann
Affiliation: 
Max Planckk Institute, Hamburg
Seminar Date: 
23. May 2013 - 11:15 - 11:45

The dynamic and structure atmospheric boundary layer and sea surface during offshore wind

Speaker: 
Irina Repina
Affiliation: 
A.M. Obukhov Institute for Atmospheric Physics
Seminar Date: 
7. May 2013 - 12:30 - 13:15

The structure of the atmospheric boundary layer over sea after abrupt change in surface characteristics in the coastal zone was studied. For the analysis, field measurements of turbulence characteristics in the layer 1 – 21 m above the sea surface and wind waves at the fetch of about 1 km were used under the wind from the shore having mountain terrain. It is shown that the profiles of mean wind speed and turbulence intensity are very different from those typical for the open sea conditions or coastal zones of flat topography.

About the Agulhas Current system, HYCOM and the EnOI: Progress report by Bjørn Backeberg & François Counillon

Speaker: 
Bjørn Backeberg (with contributions from François Counillon)
Affiliation: 
Nansen-Tutu Centre for Marine Environmental Research, Department of Oceanography, University of Cape Town Mohn-Sverdrup Center for Global Ocean Studies and Operational Oceanography, Nansen Environmental and Remote Sensing Center
Seminar Date: 
3. May 2013 - 11:00

The outline of the presentation is given below:
- A bit of background on the Agulhas Current system.
- Modelling the Agulhas with HYCOM
- Motivation for doing data assimilation.
- About the Ensemble Optimal Interpolation
- Some validation of the historical ensemble.
- Some limitations of the EnOI and how we deal with these.
- Some results & animations of a the first few assimilation cycles.
- The “to do” list (your input required)

The Mozambique Channel Eddies: Characteristics and Mechanisms of Formation

Speaker: 
Dr Issufo Halo
Affiliation: 
Nansen-Tutu Centre for Marine Environmental Research, Department of Oceanography, University of Cape Town, South Africa
Seminar Date: 
26. April 2013 - 12:30 - 13:30

A regional model configuration was designed for the South-West Indian Ocean to study the mesoscale eddy field in the Mozambique Channel. A good agreement between the model and climatological observations was found. The model reproduced well the main oceanographic features of the region. However, it overestimated the seasonal cycle of the volume transport, the eddy kinetic energy and the variability of sea surface height in the narrows of the channel. The model also captured the main water masses of the region, while failing to reproduce accurately their spreading.

How sea ice moderates atmosphere-ocean exchange thereby timing climate variability

Speaker: 
Torge Martin
Affiliation: 
Polar Science Center / Applied Physics Laboratory University of Washington
Seminar Date: 
25. April 2013 - 10:15 - 11:00

Sea ice retreat is considered a striking indicator of global warming. However, while Arctic sea ice is shrinking at an accelerating pace the ice cover of the Southern Ocean is slowly expanding. The inability of climate models to capture both observed trends demonstrates some lack of understanding of the coupled processes of atmosphere-ice-ocean interaction. Climate change forces the sea ice cover to change but sea ice also actively shapes large-scale variability.

Matlab and CUDA: Computing on a Graphics Card

Speaker: 
Brian Dushaw
Affiliation: 
Applied Physics Laboratory, University of Washington Seattle, WA USA
Seminar Date: 
19. April 2013 - 12:30 - 13:15

The latest generation of high-end video cards offer considerable computing power using their 100–2000 on-card processors, 0.3–4.0+ GB of RAM, and fast inter-processor communications. Nvidia has developed a compiler system called CUDA (Compute Unified Device Architecture) for using their graphics cards for computing. One promising application of this Graphics Processing Unit (GPU) computing capability is through Matlab and Matlab mex functions. With a properly developed mex function, the user-friendly Matlab interface can be used to perform behind-the-scenes parallel computations on the GPU.

Drift, dispersion and deformation of sea ice: A physical perspective

Speaker: 
Jérôme Weiss
Affiliation: 
CNRS - Laboratoire de Glaciologie et Géophysique de l'Environnement (Grenoble, France)
Seminar Date: 
11. April 2013 - 12:30 - 13:00

In this presentation, I will discuss the properties of sea ice motion: sea ice velocity fields and their gradients, i.e. strain-rate fields, mainly from lagrangian passive tracers (buoy trajectories). I will show that the sea ice velocity field can be decomposed into a mean field (i.e. a general circulation) + stochastic fluctuations. The statistical properties of these fluctuations cannot be explained by the turbulent properties of the forcing field (the winds), but also result from the strongly non-linear response of sea ice.