Transcript Lecture 1
Lecture 2
ASTR 111 – Section 002
Introductory Astronomy:
Solar System
Dr. Weigel
Reading for this week
– The reading for this week is Chapter 1 (all)
and Chapter 2 (sections 2.1-2.2 only)
– The quiz will cover this reading and the topics
covered in this week’s lectures
– The quiz will be available on BlackBoard at
10:15 am
Outline
•
•
•
•
•
Angular Measurements Review
Accuracy, Precision, and Bias
The Scientific Method
Astronomical Distances
Ancient Astronomy
A
B
1. What is the angular distance between points A
and B on this slide (In degrees and
arcminutes). Each student should take their
own measurement. Each student should also
write down their height alongside of their
measurement.
2. Predict what will happen if you made your
measurement in two different parts of the
room.
3. Next week you sit in the same chair but weigh
30 pounds less. Will your (angular)
measurements change?
4.
Do you think there will be a relationship between a
persons height and the angle they measure?
http://antwrp.gsfc.nasa.gov/apod/ap071025.html
Outline
1.
2.
3.
4.
5.
Angular Measurements
Accuracy, Precision, and Bias
The Scientific Method
Astronomical Distances
Ancient Astronomy
Precision, Accuracy, and Bias
• Whenever you take measurements, you
should account for them
• Fundamental terms that you must
understand when interpreting
measurements
Accuracy vs. Precision
Target is red
Shots are black
Accuracy vs. Precision
Target is red
Shots are black
High precision
Low accuracy
High precision
High accuracy
Accuracy vs Precision
Mnemonic:
You’ll get an A
for Accuracy
Bias
Target is red
Shots are black
• The left target shows bias – the
measurements were made with high
precision, but the were consistently
“off” in the same direction.
Summary
• Accuracy – all measurements or
values are clustered around the
true value (you’ll get an A for
accuracy, because you are on the
true value)
• Precision – all measurements are
clustered but are not centered on
true value
• Bias – measurements are not
centered on true value
Center of
red dot is
true value
No bias
Group work (~ 4 minutes)
1. Draw a diagram like the ones on the
previous slide that show
1. Low precision and high bias
2. High accuracy and very low precision
2. On a piece of paper, write down
•
Possible causes of low accuracy – be
specific! (Don’t say “human error”)
• Possible causes of bias – be specific!
associated with your angular measurements
Group work (~ 3 minutes)
• Which diagram best represents the
statement: “Preliminary polling
results indicated that Obama won
Virginia by a landslide because the
preliminary poll results were all from
Northern Virginia”.
A
B
C
D
Outline
1.
2.
3.
4.
5.
Angular Measurements
Accuracy, Precision, and Bias
The Scientific Method
Astronomical Distances
Ancient Astronomy
The Scientific Method
What is Science?
1) A set of facts
2) Something that professional
scientists do
3) The underlying Truth about the
Universe
4) The collection of data and formation
of a hypothesis
5) None of the above
The Scientific Method
the process
•
•
•
•
characterization of existing data
formulation of a hypothesis
formulation of a predictive test
experimental testing, (important: error
elimination and characterization)
• report and peer review
• validate or revise hypothesis
Outline
1.
2.
3.
4.
5.
Angular Measurements
Accuracy, Precision, and Bias
The Scientific Method
Astronomical Distances
Ancient Astronomy
Parallax
Parallax
10
20
30
40
50
60
70
80
90
Parallax
• When the apparent position of an object
(numbers on speedometer) changes
because of the change in position of the
observer (driver’s seat to passenger’s
seat).
The Parsec
Astronomical distances are never measured
in Car hours, dotsecs, and Moon Units
• Car Hour (ch)
– the distance a car can travel in one hour at a speed of
about 60 miles/hour. How far is Baltimore? About an
hour.
A time
• Car Year (cy)
A distance
– the distance a car can travel in one year at a speed of
about 60 miles/hour
• dotsec (ds)
– the distance at which the two dots on the screen
subtend an angle of 1 arcsec
• Moon Unit (MU)
– One Moon Unit is the average distance between
Earth and the Moon
Astronomical distances are often measured
in astronomical units, parsecs, or light-years
• Light Year (ly)
– One ly is the distance light can travel in one year at a
speed of about 3 x 105 km/s or 186,000 miles/s
• Parsec (pc)
– the distance at which 1 AU subtends an angle of 1
arcsec or the distance from which Earth would appear
to be one arcsecond from the Sun
• Astronomical Unit (AU)
– One AU is the average distance between Earth and
the Sun
– 1.496 X 108 km or 92.96 million miles
Earth
Sun
Observer’s view of Sun and
Earth from outer planet
Gods-eye view Looking down on
Sun and Earth
Observer’s view
“
Gods-eye view
Observer’s view
Group Problem
• Form groups of exactly 4
• Optimal configuration is two
students in one row and two
students in another row
No
Yes
1.
2.
3.
4.
5.
6.
Imagine that you are looking at the stars from
Earth in January. Use a straightedge to draw a
line from Earth in January, through the nearby
star (Star A), out to the Distant Stars. Which of
the distant stars would appear closest to Star A in
your night sky in January. Circle this distant star
and label it Jan.
Repeat Question 1 for July and label the distant
star “July”.
In the box below, the same distant stars are
shown as you would see them in the night sky.
Draw a small x to indicate the position of Star A
as seen in January and label it “Star A Jan.”
Distant Stars
Nearby Star
(Star A)
In the same box, draw another x to indicate the
position of Star A as seen in July and label it “Star
A July”.
Describe how Star A would appear to move
among the distant stars as Earth orbits the Sun
counterclockwise from January of one year,
through July, to January of the following year.
Consider two stars (C and D) that both exhibit
parallax. If Star C appears to move back and
forth by a greater amount than Star D, which star
do you think is actually closer to you? If you’re
not sure, just make a guess. We’ll return to this
Earth
question later in this activity.
(January)
Based on Lecture Tutorials for Introductory Astronomy, Prather et al., pg 35
1 AU
Earth
(July)
Outline
1.
2.
3.
4.
5.
Angular Measurements
Accuracy, Precision, and Bias
The Scientific Method
Astronomical Distances
Ancient Astronomy
http://www.google.com/sky/
Naked-eye astronomy had an important
place in ancient civilizations
• Positional astronomy
– the study of the positions of objects in the sky
and how these positions change
• Naked-eye astronomy
– the sort that requires no equipment but
human vision
• Extends far back in time
– British Isles Stonehenge
– Native American Medicine Wheel
– Aztec, Mayan and Incan temples
– Egyptian pyramids
Stonehenge
http://archaeoastronomy.wordpress.com/2005/06/15/stonehenge-astronomy-ii-solar-alignments/
See also http://news.bbc.co.uk/2/hi/uk_news/england/wiltshire/7465235.stm
Aztec, Mayan and Incan temples
Eighty-eight constellations cover
the entire sky
• Ancient peoples looked at
the stars and imagined
groupings made pictures
in the sky
• We still refer to many of
these groupings
• Astronomers call them
constellations (from the
Latin for “group of stars”)
Modern Constellations
• On modern star charts,
the entire sky is divided
into 88 regions
• Each is a constellation
• Most stars in a
constellation are nowhere
near one another
• They only appear to be
close together because
they are in nearly the
same direction as seen
from Earth