Seminars & Colloquia
SEMINARS
OBSERVATIONAL MULTIMESSENGER ASTROPHYSICS Seminars
Observational High-Energy Gamma-Ray Astrophysics in the Multimessenger Era
March 11, 4pm Room B
Speaker: Elisa Prandini
This seminar provides an overview of observational high-energy gamma-ray astrophysics in the context of modern multimessenger astrophysics. We begin with a brief historical perspective on cosmic-ray research, outlining the key open questions related to their origin, acceleration mechanisms, and propagation, and motivating the role of gamma rays as tracers of hadronic and leptonic processes at the highest energies. We then discuss the main classes of gamma-ray sources, including Galactic and extragalactic accelerators, and survey the currently operating gamma-ray instruments and experiments, both space- and ground-based. We highlight key scientific results from facilities such as Fermi-LAT, imaging atmospheric Cherenkov telescopes, and wide-field air-shower arrays, and discuss their impact on our understanding of high-energy astrophysical phenomena and future prospects in the field.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
High energy neutrino Astrophysics
March 25, 4pm Room B
Speaker:Elisa Bernardini
This seminar provides a comprehensive overview of high-energy neutrino astrophysics, covering both detection methodologies and recent observational breakthroughs. We survey experimental techniques across the energy spectrum, from TeV to EeV, including Cherenkov detectors (IceCube, KM3NeT), extensive air shower arrays (Pierre Auger Observatory), and emerging radio, acoustic, and space-based detection methods. Key topics include event signatures, background rejection, reconstruction algorithms, and the challenges of atmospheric neutrino discrimination and systematic uncertainties.
We then present recent observational results, including the diffuse astrophysical neutrino flux discovery, point-source identifications (TXS 0506+056, NGC 1068, Milky Way), and multi-messenger correlations with electromagnetic and gravitational wave observations. We examine implications for cosmic ray acceleration, AGN jet physics, and astrophysical models, demonstrating how complementary detection strategies and multi-messenger approaches provide unique insights into extreme cosmic accelerators and address fundamental questions in high-energy astrophysics.
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Gravitational Wave Science
April 15, 4 pm Room B
Speaker: Livia Conti
This seminar provides an overview of the science of gravitational-wave detection, covering both the sources and the experimental techniques underlying the various detectors operating on Earth and in space. We review how different classes of sources populate different frequency bands, motivating the need of a variety of experiments, and illustrating how gravitational-wave science spans many orders of magnitude in frequency and length scale.
Special emphasis is placed on the ground-based interferometers of the LIGO–Virgo–KAGRA Collaboration. We discuss their working principles and noise budgets, highlighting technical and fundamental physics limitations. An overview of the main observational results collected by the Collaboration since the first detection of GW150914 in 2015 is presented. The seminar concludes with an introduction to the Einstein Telescope, outlining its detection capabilities, its role in future multi-messenger observations, and the prospects it opens for gravitational-wave science in the coming decades.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
An Introduction to Galaxy Surveys: Instruments, Data and Analysis
April 29, 4 pm Room B
Speaker: Chiara Sirignano
This seminar provides a comprehensive introduction to galaxy surveys, exploring the fundamental aspects of modern observational cosmology. We will examine the sophisticated instruments and telescopes employed in large-scale surveys, including current ground-based facilities such as the Dark Energy Spectroscopic Instrument (DESI) and the Vera Rubin Observatory (LSST), as well as space missions like ESA's Euclid satellite and the NASA Roman Observatory. The session will discuss their technical capabilities and observational strategies, covering the various types of data collected, including photometric and spectroscopic measurements, and addressing the challenges of data acquisition, calibration, and quality control. Finally, we will introduce the key analysis techniques used to extract cosmological information from survey data, including clustering statistics, redshift measurements, and methods for mapping the large-scale structure of the universe. This seminar is designed for students and researchers willing to understand how contemporary galaxy surveys contribute to our knowledge of cosmic evolution and structure formation.
High Performance Computing in the sciences
May 4,8,11,15 8:30am Lab P104
Speaker: Tim Mattson
Short Description:
You are not a well-rounded computational scientist if you do not understand High Performance Computing (HPC). And HPC is built on top of parallel computing … hardware with multiple processing elements that run at the same time using algorithms/data-structure decomposed onto those processing elements so they make forward-progress in parallel. In this seminar series, we will cover the full range of mainstream parallel systems and the core application programming interfaced used in parallel computing.
Detailed Description:
There is a class of problems where performance is an essential feature of the solution. In these problems, your programs need to achieve a high fraction of the available performance from a computing system.
For example: (1) you need to complete weather forecasts before the date being forecasted, (2) a climate model needs massive amounts of memory so memory use across multiple computers must be optimized, or (3) an experiment is producing data at a high rate and you need to process the data fast enough to keep up. The key point is that for each of the cases, performance is not a “nice to have feature”. Performance is an essential part of the solution; without sufficient performance the solution to the problem fails.
We call this class of problems, where performance is a fundamental part of the solution, High Performance Computing or HPC. Most fields of science include problems that depend on HPC, so much so, that in many fields (including chemistry and physics) your background in computing is deficient if it doesn’t include HPC.
In this series of four seminars, we will provide a detailed survey of the fundamental topics in HPC. The common theme is parallel computing; breaking a problem down into components that can make forward progress at the same time. We will cover the following topics:
- Parallel computer systems. This is essential since HPC software needs to map onto the features of the hardware in order to achieve a large fraction of available performance.
- Programming shared memory systems. These ubiquitous systems are based on CPUs. We will program them with the world’s most commonly used application programming interface (API) called OpenMP. The key benefit to starting with OpenMP is that it is also one of the easiest to learn so you can focus on algorithms and not learning a complex API.
- GPU programming. We will use OpenMP once again to cover the key concepts in GPU programming. While we generally do NOT recommend proprietary APIs, we will discuss key elements of CUDA from Nvidia given it heavy use in GPU programming.
- Distributed memory computing. We will provide an overview of MPI, the message passing interface.
With these four lectures, you will understand the full range of mainstream parallel computing systems to sufficient depth to intelligently decide which parallel systems you need to use and, more importantly, possess the foundational knowledge needed to work towards mastery of the APIs you use.
