Wednesday, January 8, 2014

AAS 223: Wednesday, January 8.


Morning was mostly spent wandering around poster session, so more tweeting than blogging took place. Today's the exoplanet session, so it's pretty much all the posters I've been looking forward to. I'll be blogging from the late morning plenary session (in 20ish minutes) and the afternoon talks, including one from our very own Ben Nelson (after which I'll ditch the exoplanet talks in favor of Astronomy Education Outreach.)


318 Plenary Session Mark Krumholz: The Origin of Stellar Masses
(Just took Kevin Luhman's ISM course, so I should be able to understand what's going on!)

IMF = initial mass function, which is the distribution of stellar masses at birth (for you non-astronomers reading this).
One of the most fundamental problems in astrophysics, and not a new one. Also underlies most extragalactic observations (just ask Alex Hagen). Determines energy/chemical balance of the universe. Determines suitability of the universe for life (need stars that can support planets that support life).
(Also, turns out that letters complaining about not being cited in journals goes back at least 100 years).

Observing the IMF:
You can measure galactic field stars (good for numbers/statistics, and not useful above a few times the mass of the Sun) or young clusters (worse statistics, but fewer systematics, no corrections for stellar evolution).
Best case scenario: Orion Nebula Cluster
Most IMFs show peak between .1 and 1 solar masses, and power-law drop off at higher masses
Unresolved stellar populations observed in dwarf and large elliptical galaxies.
Dwarf galaxies don't actually have a different IMF, just normal IMF coupled with low SFR and clustering.
In giant ellipticals, IMF peak appears to be lower masses (from kinematics and spectroscopy).

So where does the power-law high mass tail come from?
Turbulence described by a power-law (alpha roughly -2). Generates nearly log-normal (logarithmic Gaussian distribution) of gas densities. Now some derivation of the IMF slope based on the distribution from turbulence, gonna skip on the details here because no one likes derivations. Predicts we should get dense 100 solar mass regions, so why don't they form more often?
Problem #1: Well, first off, massive stars are bright as hell, so they exert radiation pressure on surrounding material. Radiation pressure alone restricts us to 20 or so solar masses as the most massive stars.
Problem #2: Fragmentation. How do we make a 100 solar mass cloud collapse and not have it fragment into small pieces?
"Aww crap, we need a computer."
Cue nasty equations of stellar structure.
"We have a technical term for 6-dimensional objects: really really bad."
Also, simulations of star formation always look really cool.
"Real world is not a spherical cow."
Heating from protostars can inhibit fragmentation in surrounding gas. So what happens when you take this into account? (cue simulation) With radiative heating, accretion still happens onto massive star, doesn't form a bunch of tiny stars as well.

Now how about the peak?
Clouds fragment due to Jeans instability. Except that while Giant Molecular Clouds have roughly constant temps, densities are not remotely uniform. Argument described as "bullshit".
MORE UGLY EQUATIONS! (describing isothermal gas)
Isothermal collapse doesn't actually work.
More equations that I can't really type here, but basically building a toy model of a star.
Radius of accreting protostar set by deuterium burning. Burning sets characteristic core temperature of roughly 10^6 K. Core temperature => escape speed => luminosity
Bunch of math gives stellar mass in units of fundamental contests. Pressure still unknown, but has exponent of -1/18.
"I plead guilty to bullshit to the -1/18th power."

Summary
Power-law tail plausibly produced by statistics of supersonic turbulence.
Peak mass likely comes from effects of stellar heating.


325.01 Ben Nelson: Remastering the RV Classics: Self-Consistent Dynamical Models for the 55 Cnc and GJ 876 Planetary Systems

55 Cancri A: solar-like stat hosting 5 planets
1418 RV observations over 23 years, 40 model parameters, 5 minute integration timestep.
Mass estimate for planet e, 7.99 +/-0.25
Can't be misaligned with outer planets or it tends to get accreted by the star.
planets b and c strongly interacting "not in a mean-motion resonance", strongly interacting on observable timescales.

GJ 876: red dwarf with 4 planets, 2, 30, 60, 120 day periods.


(Holy crap, Astro Education Research is packed to the gills!)
322.03 Douglas Duncan: Digital Devices and Student Learning: Faculty Policies Make a Difference

Worked with sociologist whose specialty was college student behavior.
Does digital distraction (texting) affect grades? Frequency of cell phone use correlates with worse grades, but also dependent on instructor's policy. Harsh anti-texting policies work. Students recognize lack of policy concerning cell phones.

Laptop use (how oddly appropriate):
Multitasking laptop users suffered full letter grade drop, those behind users suffered BIGGER drop.
Learning limited to being superficial, more easily forgotten.
Need to find most effective uses of technology, not just use it haphazardly.


322.04 Angela Speck: "Assessment" of Teaching Methods and Critical Thinking in a Course for Science Majors

Assessment is even trickier than they thought. How do you test critical thinking?
Critical thinking is important for students to overcome previous knowledge. "critical thinking is the correct assessment of a statement." Can you adjust your thinking when you learn new stuff?
Course used: Solar System Science, for seniors and grad students.
Ennis-Weir Critical Thinking Essay Test (1985)
Designed to "evaluate a person's ability to appraise an argument and to formulate in writing an argument in response."
Re-done for course-relevant material, using video "What if We Had No Moon?"
Critique essay "What if Earth Had Two Moons?"

322.05 Kate Follette: Science Literacy's Neglected Twin: Numeracy

Started off with a number of numerical mistakes, some of which I've definitely made on tests in the past.
Pre-test scores ranged from rather good to guessing. 50% of students get simple math questions wrong and think they got them right. Very hard to make a statistically significant change in students skills.


322.06 Kathryn Williamson: Didn't catch the full title. Shit was long.

Newtonian Gravity Concept Inventory: can measure student understanding of gravity on a linear scale.
Biggest significant predictors were pre-instruction ability and the instruction received. Physics students consistently slightly better. Significantly significant under-performance of women (as usual :( )

What we want to know:
How *should* we teach Newtonian Gravity?
What types of interactive engagement strategies are most effective?
How to students in the study compare naturally?


322.07 Ed Prather: The Collaboration of Astronomy Teaching Scholars (CATS) – Reporting from the Nation’s Largest College-Level, Astronomy Education Research Initiative.

Hard to disentangle CATS from Center for Astronomy Education work. Positive effects of interactive learning affects all students equally. Lot of projects, can't write them all down.
50 publications, 200 talks and posters at professional meetings, 100 professional developement workshops, lot of work being done.
"150 years ago you had to use an outhouse. Now you have a cell phone. That's science."


322.08 Sanlyn Buxner (covered by Ed Prather again): Findings from Five Years Investigating Science Literacy and Where Students Get their Information about Science

College preparation probably best indicator of science literacy.
No significant changes in mean and standard deviation over time.
Faith and belief based factors have strongest effect on where people end up on science literacy test.
Pseudoscience not necessarily at odds with functional science literacy.

Sources of science info


322.09: Seth Hornstein: A Research-Informed Approach to Teaching About Light & Matter in STEM Classrooms

Mostly a developmental talk. Develop research-validated, student-centered interactive engagement classroom activities. Brought together astro education specialists and NRAO scientists. Scientists identified hot topics while educators identified needs in Astro 101 courses. Managed to find common ground. All topics required understanding of light and matter.
Created lecture slides with embedded think-pair-share questions.


My battery died at this point.

No comments:

Post a Comment