“hernanz” · 2017. 7. 10. · separability)in)astronomical)phase)space) 19 coverage, spread...

“Hernanz”

M. Shara -‐ 16 June 2017

•  the Spanish form of the Old German name Ferdinand, meaning "bold voyager," from the elements farð, meaning "journey" and nanð / nanth, meaning "courage" or "daring.”

•  So… Hernanz = “Courageous Voyager”

Margarita … personally tesNng eAstrogam in 2019


A few memorable quotes…

•  It was a nice insNtute…with a beach…and biologists •  Maybe some of these models are fake •  I think the curtains are making a comment about these slides

•  We don’t need atmospheres…all novae are black bodies

•  This morning there was the suggesNon to do trial by combat (for observing Nme)

•  There is a proposal to change the name of Messier 31 to MarNnHenze 31


A. Coc -‐ Boldest PredicNon

•  Classical Novae will soon become the first explosive sites for which all nucleosynthesis is based on experimental data


What is the Common Theme of this CelebraNon? And of

Margarita’s Career?

•  Extreme physics/Collapsed macer/γ rays •  ExploraNon of new physical regimes •  Daring PredicNons

! DISCOVERY IN ASTROPHYSICS


So…

• “Let me talk about the details…and …I’ll stop at some point”


Overview

*What is an Astronomical “Discovery”?

*ScienNfic Discovery PotenNal =

SDP of Surveys/Telescopes of 2020+

Including eXTP/LOFT EASTROGAM


Understanding what comprises a major discovery…and how astronomers make “discoveries”…

is essenNal to furthering the discovery process

cf MarNn Harwit “Cosmic Discovery” Basic Books 1981

Astronomical Discoveries


What Is A Major Astronomical “Discovery”? Part 1

A new “class of object” – differing by a factor of >1000x in at least one parameter from the next most similar object

Same Class of Object K dwarf star (0.5 Msun, 0.1 Lsun) = G dwarf star (1.0 Msun, 1 Lsun)

Not the Same Class of Object Red Giant Star ≠ White Dwarf Star ≠ Neutron Star Densi4es & sizes differ by > 3 orders of magnitude M. Shara -‐ 7 June 2017

33 “Discoveries” unNl 1961 (adapted from Harwit)

Pre telescope

1600s

1700s

1800s

1900-‐1961


36 “Discoveries” 1961-‐2017

1961-‐1980

1981-‐2017


What Is A Major Astronomical “Discovery”? Part 2

A new astronomical or physical process

Masers Neutrino oscillaNons (“solar neutrino problem”) AccreNon onto compact objects (X-‐ray astronomy)

Stellar mergers/promiscuity (Blue stragglers, Red Novae, “impossible” binaries in globular clusters)

Jets… and

NOVA γ RAYS M. Shara -‐ 16 June 2017

The Rate of Astronomical Discoveries is AcceleraNng

•  ~1000 BCE-‐ 1609 ~5 Discoveries •  1609-‐1961 ~29 Discoveries •  1961-‐2017 ~36 Discoveries


Hard Choices

•  OpNon 1: OpNon 2:


Astronomical Discovery Part 3

TELESCOPE/DETECTOR 1,000x “becer”

! DISCOVERY

*Since WWII: Highly novel technologies from physics, And/or military– (IR, satellites,UV, X-‐ray…) Probe new observaNonal “phase-‐space”

UHURU…HST…SWIFT…Chandra…Fermi HESS…LSST…JWST… SKA…

Equally true of eASTROGAM, LOX, JPAS/JPLUS, CTA, HiSCORE M. Shara -‐ 16 June 2017

Astronomical Phase Space 5 INFORMATION CARRIERS: (E.M., GRAVITY, NEUTRINOS,COSMIC RAYS) +METEORITES, LUNAR/MARTIAN/COMET SAMPLES

NON-‐”SAMPLE” CARRIERS’ 7 PARAMETERS:

ANGULAR RESOLUTION TEMPORAL RESOLUTION WAVELENGTH RANGE SENSITIVITY AREAL COVERAGE POLARIZATION CONTRAST


2D CUT #1: ANGULAR RESOLUTION VS. WAVELENGTH

BS


2D cut #2 TEMPORAL RESOLUTION vs. WAVELENGTH


Astronomical Phase Space (from Djorgovski 2011)

  15 

The dimensionality of the OPS is given by the number of characteristics that can be defined for a given type of observations, although some of them may not be especially useful and could be ignored in a particular situation. For example, time domain axes make little sense for the observations taken in a single epoch. Along some axes, the coverage may be intrinsically discrete, rather than continuous. An observation can be just a single point along some axis, but have a finite extent in others. In some cases, polar coordinates may be more appropriate than the purely Cartesian ones.

Some parts of the OPS may be excluded naturally, e.g., due to quantum limits, diffraction limits, opacity and turbulence of the Earth’s atmosphere or the Galactic ISM on some wavelengths, etc.; see Harwit (1975, 2003). Others are simply not accessible in practice, due to limitations of the available technology, observing time, and funding.

We can thus, in principle, measure a huge amount of information arriving from the universe, and so far we have sampled well only a relatively limited set of sub-volumes of the OPS in general, much better along some axes than others: we have fairly good coverage in the visible, NIR, and radio; more limited X-ray and FIR regimes; and very poor at higher energies still. The discrimination becomes sharper if we also consider their angular resolution, etc.

Figure 1. A schematic illustration of the Observable Parameter Space (OPS). All axes of the OPS corresponding to independent measurements are mutually orthogonal. Every astronomical observation, surveys included, carves out a finite hypervolume in this parameter space. Left: Principal axes of the OPS, grouped into four domains; representing such high-dimensionality parameter spaces on a 2D paper is difficult. Right: A schematic representation of a particular 3D representation of the OPS. Each survey covers some solid angle (Ω), over some wavelength range (λ), and with some dynamical range of fluxes (F). Note that these regions need not have orthogonal, or even planar boundaries.


Separability in Astronomical Phase Space

  19 

coverage, spread among a number of surveys, that covers as much of the accessible OPS hypervolume as possible and affordable.

The OPS and the PPS may overlap in some of the axes, and new discoveries may be made in the OPS alone, typically on the basis of the shapes of the spectral energy distributions. This generally happens when a new wavelength window opens up, e.g., some blue “stars” turned out to be a new phenomenon of nature, quasars, when they were detected as radio sources; and similar examples can be found in every other wavelength regime. The unexpectedness of such discoveries typically comes from some hidden assumptions, e.g., that all star-like objects will have a thermal emission in a certain range. When astronomical sources fail to meet our expectations, discoveries are made. Improvements in angular resolution or depth can also yield discoveries, simply by inspection. Of course, noticing something that appears new or unusual is just the first step, and follow-up observations and interpretative analysis are needed.

Figure 3. A simple illustration of a physical parameter space (PPS), viewed from two different angles. Families of dynamically hot stellar systems are shown in the parameter space whose axes are logs of effective radii, mean surface brightness, and the central velocity dispersion in he case of ellipticals, and the maximum rotational speed in the case of spirals. In the panel on the left, we see an edge-on view of the Fundamental Plane (FP) for elliptical galaxies (Djorgovski & Davis 1987). On the right we see a clear separation of different families of objects, as first noted by Kormendy (1985). They form distinct clusters, some of which represent correlations of their physical parameters that define them as families in an objective way. This galaxy parameter space (Djorgovski 1992ab) is a powerful framework for the exploration of such correlations, and their implications for galaxy evolution. Data for ellipticals are from La Barbera et al. (2009), and for dwarf spheroidals and globulars are compiled from the literature.

Sometimes we expect to find objects in some region of the parameter space. For example, most quasars (especially the high-redshift ones) and brown dwarfs are found in particular regions of the color space, on the basis of an assumed (correct) model of their spectral energy distributions.


Size Velocity dispersion Surface brightness

Improved Angular ResoluNon and Higher Contrast…1000x


SNIa – The expansion of the

universe is acceleraNng -‐ Dark Energy

Improved Temporal/SynopNc ResoluNon and SensiNvity/Precision…1000x

MicroTransits – 3,000 TransiNng Exoplanets –

Planets are ubiquitous


Discoveries Part 4


Discovery Part 5

Serendipity in Discovery? Many discoveries were not predicted but discovered serendipitously. e.g. Cosmic Microwave Background RadiaNon, Dark Macer, Large Scale Structure

Others were predicted and long-‐sought in large, carefully planned surveys

e.g. GravitaNonal RadiaNon, Brown Dwarfs, Exoplanets, GravitaNonal Lensing…and


Key Lessons for Discoveries

•  Work with the most innovaNve telescopes/instruments available

•  Start using them as soon as they come online •  Be alert to unexpected objects/outliers •  Collaborate with a variety of physicists/engineers/astronomers/IT experts

RESULT = “Discoveries”


QuesNon:Which Telescopes Will Be 100-‐1000x “Becer” Than ALL Others By 2022?

•  SKA – sensiNvity AND angular resoluNon AND wavelength coverage

•  JWST – IR sensiNvity AND angular resoluNon •  LSST – temporal resoluNon and sensiNvity and areal coverage

•  CTA – sensiNvity and wavelength coverage •  LIGO, ICECUBE – new physics regimes •  And…TESS, Gaia, Euclid, eRosita, WFIRST…

Answer: All of them… including eASTROGRAM


Figures of Merit for ScienNfic Discovery PotenNal=SDP

Etendue = A x Ω (Primary Area x solid angle) does NOT measure SDP

**Single-‐pass surveys ! SDP= How deep x How wide x wavelength baseline

**MulN-‐pass surveys SDP =How deep x How wide x How oyen x How long x Wavelength baseline


NOVA γ RAYS’ PLEDGE


*We believe Margarita, that they exist *We will do all in our power to prove it *We will never rest unNl their existence is demonstrated … Then we will argue like cats and dogs about what they REALLY tell us

Margarita has ALWAYS followed these principles rigorously...


…and taken Nme to smell the flowers along the way

Happy Birthday, Margarita!

•  Per molts anys! •  Moltes felicitats!


“hernanz” · 2017. 7. 10. · separability)in)astronomical)phase)space) 19 coverage, spread...

Documents