http://www-math.umd.edu/research/seminars.html http://www-math.umd.edu/research/seminars.html Mon, 19 Sep 2022 14:00:00 EDT Where: ATL 3400
Speaker: Professor James Carton (UMD AOSC) - https://www2.atmos.umd.edu/~carton/

]]> http://www-math.umd.edu/research/seminars.html Mon, 26 Sep 2022 14:00:00 EDT Where: ATL 4301
Speaker: Jonathan Poterjoy (UMD AOSC) - https://poterjoy.com/

]]> http://www-math.umd.edu/research/seminars.html Mon, 07 Nov 2022 14:00:00 EST Where: TBA
Speaker: Jon Poterjoy (UMD AOSC) - https://poterjoy.com

]]> http://www-math.umd.edu/research/seminars.html Fri, 02 Dec 2022 13:00:00 EST Where: TBA
Speaker: Takemasa Miyoshi (RIKEN, Japan) - http://data-assimilation.riken.jp/~miyoshi/

]]> http://www-math.umd.edu/research/seminars.html Thu, 15 Dec 2022 09:00:00 EST Where: ATL 3400
Speaker: () -

]]> http://www-math.umd.edu/research/seminars.html Mon, 06 Feb 2023 14:00:00 EST Where: Atlantic Building 3400
Speaker: Thomas Breunung (UMD Mechanical Engineering) - https://scholar.google.com/citations?user=5acbq_UAAAAJ&hl=de

]]> http://www-math.umd.edu/research/seminars.html Mon, 27 Feb 2023 14:00:00 EST Where: Atlantic 3400
Speaker: Ben Sheppard and Josh McCarty (University of Maryland) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 06 Mar 2023 14:00:00 EST Where: Atlantic 3400
Speaker: Brendan Gramp and David Russell (UMD AMSC Program) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 27 Mar 2023 14:00:00 EDT Where: Atlantic 3400
Speaker: Luyu Sun + Samarpan Chakraborty (University of Maryland) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 03 Apr 2023 14:00:00 EDT Where: ATL 3400
Speaker: Molly Wieringa (Univeristy of Washington) - https://science.gsfc.nasa.gov/sed/bio/101731/
Abstract: We present a rigorous exploration of the sea ice data assimilation problem using a framework specifically developed for rapid, interpretable hypothesis testing. By coupling a single-column sea ice model to the Data Assimilation Research Testbed (DART), we explore the grid-cell response of complex sea ice models to adjustments made by a combination of data assimilation algorithms. We are particularly interested in understanding behavior related to the model’s ice thickness distribution (ITD), as well as the bounded nature of both state and prognostic variables in the sea ice model. We find that assimilating with algorithms that respect boundedness does not necessarily improve the accuracy of the analysis but does minimize non-physical adjustments insofar as the sea ice state bounds can be properly applied. We also find that assimilating observations of the ITD directly notably improves the analysis across all state variables when compared to assimilating aggregate quantities such as mean sea ice thickness (SIT) or sea ice concentration (SIC). The full details of these results elucidate many of the positive and negative findings of previous sea ice data assimilation studies and tackle the challenges intrinsic to assimilating observations of a bounded material in which relationships between variables are non-linear. We anticipate that the insights gained from this work will facilitate better future sea ice reanalysis products.
]]> http://www-math.umd.edu/research/seminars.html Mon, 10 Apr 2023 14:00:00 EDT Where: Atlantic Building 3400
Speaker: Kenta Kurosawa and Frank McBride (UMD Weather-Chaos Group) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 17 Apr 2023 14:00:00 EDT Where: Atlantic Building 3400
Speaker: Shay Gilpin + George Britzolakis (U. Colorado Boulder (Gilpin) + UMD (Britzolakis)) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 24 Apr 2023 14:00:00 EDT Where: Atlantic Building 3400
Speaker: Sam Di Pasqua & Joey Knisely (UMD AMSC & UMD AOSC) -

]]> http://www-math.umd.edu/research/seminars.html Mon, 01 May 2023 14:00:00 EDT Where: Atlantic Building 3400
Speaker: Chu-Chun Chan & Sarah Loughran (UMD & UMD) -

]]> http://www-math.umd.edu/research/seminars.html Wed, 03 May 2023 12:30:00 EDT Where: Atlantic Building 3400
Speaker: Steven Fletcher (Colorado State University) - https://www.cira.colostate.edu/staff/fletcher-steven/
Abstract: The underlying assumption for variational and Kalman filter based data assimilation algorithms is that the associated errors are Gaussian distributed random variables. Over the last 18 years at CIRA we have worked on relaxing this assumption to allow for lognormally distributed, and recently reverse-lognormally distributed errors. The first part of this talk will be an overview of the development of the lognormal and the mixed Gaussian-lognormal variational approachesas well as the recent development of the mixed Gaussian-lognormal based Kalman filter. In the second part, we introduce the reverse-lognormal distribution to be able to include negatively skewed errors. All these ideas can then be combined to develop a mixed version of the maximum likelihood ensemble filter. The main question then remains: how do we decide the underlying distribution of the errors? To answer this question we have developed a basic machine learning algorithm that can help us
]]>