MARCEL G.

SCHAAP

TOPICS IN ENVIRON-
M E N TA L P H Y S I C S : T H E
SOLAR SYSTEM AND
EARTH
2 marcel g. schaap
Copyright © 2015, 2016, 2017, 2018 Marcel G. Schaap

All rights reserved. This book or any portion thereof may not be reproduced or used in any manner
whatsoever without the express written permission of the author or publisher except for the use of brief
quotations in a book review.

Selling class notes and/or other course materials to other students or to a third party for resale is not
permitted without the instructor’s express written consent. Providing student email addresses to a third
party is not permitted. Violations to this and other course rules are subject to the Code of Academic Integrity
and may result in course sanctions. Additionally, students who use D2L or UA email to sell or buy these
copyrighted materials are subject to Code of Conduct Violations for misuse of electronic resources provided
by The University of Arizona. This conduct may also constitute copyright infringement.

Printed in the United States of America

First Printing, 2015

ISBN 0-9000000-0-0

published by dthetadt publishing

Address information

books.dthetadt.com
 Contents

1 The Formation of the Solar System and Planet Earth 9

1.1 Introduction 9
1.2 The Early Solar System 10
1.2.1 The Origin of the Earth-Moon System 13
1.2.2 The Origin of Earth’s Water 13
1.3 The Present Solar System 14
1.3.1 Terrestrial (Rocky) Planets 14
1.3.2 Asteroid Belt 15
1.3.3 Gas and Ice Giants 17
1.3.4 Kuiper Belt and Oort Cloud 18
1.4 The Internal Structure of the Terrestrial Planets 18
1.5 Abundance of Elements on Earth 22
1.5.1 Volatiles 22
1.5.2 Lithophiles 23
1.5.3 Chalcophiles 23
1.5.4 Siderophiles 23
1.5.5 Availability of Elements 24
1.6 A Simple Planetary Surface Temperature Model 25
1.7 Earth’s Uniqueness 29

2 Celestial Mechanics, Seasonality and Climate 31

2.1 Introduction 31
2.2 Seasonality and Orbital Parameters 33
 6 marcel g. schaap

2.2.1 Ecliptic Plane 33

2.2.2 Eccentricity 34
2.2.3 Kepler’s Laws and Newtonian Mechanics 35
2.2.4 Axis Tilt 37
2.3 Seasons 38
2.3.1 Eccentricity and Global Insolation 39
2.3.2 Declination 39
2.3.3 Annual Variation in Insolation by Latitude 42
2.4 Orbital Changes 44
2.4.1 Eccentricity Changes 45
2.4.2 Apsidal precession 46
2.4.3 Axis Tilt Changes 47
2.4.4 Precession of Earth’s Axis 48
2.5 Changes in Solar Insolation as a Driver for Ice Ages 49
2.5.1 Building a Conceptual Model 50
2.5.2 Data About Past Climate 52
2.5.3 Comparing Model and Data 54
2.5.4 Interaction Between Ice Sheets and the Asthenosphere 57

3 Bibliography 61

A Critical Elements 65

B Albedo 69

C A Demonstration of the Greenhouse Effect 71

C.1 Equipment 71
C.2 Experiment 72
C.3 Discussion 73
C.4 Simplified Terminology 74
C.4.1 Light Emission 74
 topics in environmental physics: the solar system and earth 7

C.4.2 Light Transmission 74

C.4.3 Light Absorption 74
C.4.4 Light Reflection 75
C.4.5 In Summary 75

D Earth’s Rotation and Tides 77

D.1 Fluctuations in Earth’s Rotation 77
D.2 Interaction with the Moon 78
1
The Formation of the Solar System and Planet Earth

1.1 Introduction

The previous chapter described how the sun emerged from a molecu-
lar cloud, how it produces its energy and discussed the constancy of
this energy source. This chapter will focus on the development and
structure of rest of the solar system and, in particular, Earth. As far
as we know, Earth is the only place where life evolved, though the ex-
istence of microbial life cannot be completely ruled out on Mars and
possibly in the subsurface of some of Jupiter’s or Saturn’s moons. As
we will see in this and other chapters, Earth’s position in the solar sys-
tem is nearly optimal: it is far enough from the sun’s scorching heat,
yet close enough to keep water liquid. In addition, the earth has a mag-
netic field that protects its atmosphere from erosion by the solar wind,
while the angular momentum of our moon serves a vital function in
keeping its rotation axis almost steady in space. Both are needed for
maintaining a stable climate that was suitable for the development of
life.
To get a better understanding of planet Earth, we will therefore
Video how this module fits
study the formation of the solar system, and -as a prelude to under- in the course:https://d2l.
standing the fundamentals of climate models later in this book- de- arizona.edu/content/enforced/
velop a simple model can predict the surface temperature of planets. 950156-867-2204-1ENVS420520001101/
02_Solar_System/HD_Intro_hd_720.mp4

Key points of this chapter are:

• The Solar system formed out of the protoplanetary disk that sur-
rounded the proto-sun. The protoplanetary disk consisted of large
amounts of hydrogen and helium gas with a minor amount of dust
consisting of various kinds of ice, metals and silicates.
Video introduction to this chapter,
• There are four rocky planets in the inner solar system: Mercury, and origin of the solar system: https:
//d2l.arizona.edu/content/enforced/
Venus, Earth, and Mars. These formed relatively close to the sun 950156-867-2204-1ENVS420520001101/
where it was too hot to preserve much gas or ice. 02_Solar_System/M2S1_hd_720.mp4
10 marcel g. schaap

• Beyond the snow line where water ice remained stable, there was
enough gas to form the giants Jupiter and Saturn. The ice giants
Uranus and Neptune formed in the more distant and even colder
part of the solar system where there was abundant gas and ice.

• Smaller metallic, rocky and icy asteroids also formed among the
planets, particularly between the orbits of Mars and Jupiter. Comets
consist predominantly of ice and originated in the outer solar sys-
tem beyond the orbit of Neptune.

• The Earth-Moon system formed by an impact of a large Mars-sized

body with Earth. The presence of a large moon stabilizes Earth’s
rotation axis and climate.

• Large-scale changes in the orbits of the giant planets disturbed the

orbits of asteroids and comets resulting in an intense bombardment
of asteroids and comets onto the inner solar system planets. This
happened during the first 500 million years after the formation of
the solar system

• The abundance of elements found in Earth’s crust is a result of the

formation of Earth in the inner solar system as well as differentia-
tion of Earth in an iron-nickel core, silicate mantle and crust, over-
lain by an ocean and atmosphere containing low-density volatile
compounds

• Volatile elements include H, N, C, and the noble gases He, Ne, Ar,
Kr, Xe. Most high-density siderophile elements (Fe, Ni, Mn, and
some precious metals) sank to the core. Lithophile elements gen-
erally bind strongly with oxygen and form relatively light minerals
that are found in the crust and mantle, while chalcophile elements
prefer sulfur and form denser minerals located in ore bodies.

• A simple model can be developed to estimate a planet’s surface

temperature from the solar luminosity, a planet’s distance to the
sun, and its albedo. This model is able to generate more or less
correct values for some planets, but not for Venus and Earth, both
of which have substantial atmospheres with greenhouse effects.

1.2 The Early Solar System

The planets in our solar system formed in the protoplanetary disk

a few tens of million years after the birth of the proto-sun 4.6 bil-
lion years ago. (Figure 1.1). Unlike the proto-sun, the disk was cool
(Willacy et al., 2015) and contained only about 1% of the mass of the
entire solar system. The majority of the proto-planetary disk (98%)
consisted of hydrogen and helium gas. The remaining 2% was “dust”
 topics in environmental physics: the solar system and earth 11

Figure 1.1: Left: protoplanetary disks

within the Orion Nebula. Their apparent
glow is due to the light of massive stars
outside the field of view. The small dark
blob (middle right) is a protoplanetary
disk without direct illumination. The
tiny red dot in its middle is an emerging
proto-star. The size of the image is about
0.14 ly, making the largest protoplane-
tary disk about 0.01 ly across. Right:
artist concept of the early solar system’s
protoplanetary disk. Credit (left): C.R.
O’Dell/Rice University, NASA; (right):
made of silicates (0.4%), metals (0.2%) and ice (1.4%). In this context, NASA/JPL-Caltech.

“ice” not only means water ice (H2 O), but also other volatile substances
such as carbon dioxide and -monoxide (CO2 and CO), methane (CH4 )
and ammonia (NH3 ), the melting points of which are listed in Table
1.1.
In a process called accretion, electrically charged dust grains merged
into clusters, which produced self-gravitating aggregates that after
many gentle collisions and mergers produced kilometer-sized planetes-
imals (Steinpilz et al., 2019). The composition of these planetesimals
depended on the distance from the proto-sun (Figure 1.2) and hence
the local temperature of the protoplanetary disk.

Terrestrial planets Gas Giants Ice Giants

(Mercury, Venus, Earth, Mars) (Jupiter, Saturn) (Uranus, Neptune) and
coments Figure 1.2: Simplified cross section
through the accreting protoplanetary
disk shown the distribution of rocky and
icy planetesimals. The snow-line indi-
Sun cates the distance from the sun where
water ice remains stable.
Hot Snow line Cold

Silicates and metal Ice (H2O, CO, CO2, NH3, CH4) Gas

Close to the proto-sun, strong solar radiation vaporized most of the

ices into gas. Strong solar winds expelled the vaporized products as
well as H2 and He gas to the outer solar system. Metal and silicate dust
grains in the inner solar system therefore absorbed small amounts of Ice Melting Point
gas and volatiles. Planetesimals that formed here therefore contained (kelvin)
He 0
only small amounts of volatiles before more violently coalescing, into
H2 14.0
the rocky planets Mercury, Venus, Earth and Mars. A considerable CO 63.1
amount of gravitational energy was released as heat, leaving the in- NH3 77.4
CH4 90.7
ner planets in a completely molten state with temperatures of several CO2 217
thousands of degrees kelvin. H2 O 273.15
Temperatures were lower in the outer solar system where solar radi-
Table 1.1: Melting point (MP) of some
ation and solar wind were less intense, allowing gas and ice to remain. volatiles or “ices” present at the solar
The distance from the sun where water ice remained stable is known as system’s formation. H and He were not
present in the proto-planetary disk in
the snow line. Other ices with lower melting points (Table 1.1) were sta- ice form, their melting point tempera-
tures are much lower than temperatures
prevalent in the proto-planetary disk.
12 marcel g. schaap

ble at correspondingly larger distances from the sun. Dust in the outer
solar system therefore was able to attract and retain large amounts gas
and ice, which facilitated a rapid coalescence of planetesimals (Cieza
et al., 2016).
The gas giants Jupiter and Saturn formed near the snow line by grav-
itational attraction of free hydrogen and helium gas onto their rocky
and icy cores. Farther out, where more ice was available the somewhat
smaller ice giants Uranus and Neptune formed. Although these plan-
ets predominantly contain hydrogen and helium, they have a much
higher content of water, ammonia and methane than the gas giants,
owing to their formation in the colder regions of the solar system. The
blueish color of Uranus and Neptune (Figure 1.4) is due to methane in
the atmospheres of these planets.
The planetary accretion process was halted after tens million years
because the increasingly bright sun generated solar winds strong enough
to remove most of the remaining gas from the solar system. Some of
the original planetesimals still remain at distances far beyond the orbit
of Neptune. In 2019, NASA’s New Horizons probe few by the as-
teroid MU69 (also named Arrokoth). This system is a contact-binary
(Figure 1.3), indicating a very gentle merger of two separate planetesi-
mals. The topography of the larger body suggests previous mergers of
smaller bodies.

Figure 1.3: The binary asteroid MU69

Arrokoth (formerly known as Ultima-
Thule) as imaged by NASA’s New
Horizons in 2019. The biggest sec-
The formation of the early solar system was not a quiet time and it is tion resembles a flattened disk with
likely that none of the planets formed in their current orbits. Accord- the largest dimension of 21.6 km, the
smaller body has the largest dimension
ing to the Nice Model1 , Saturn, Uranus and Neptune formed closer to of 15.4 km. Image credit: NASA/Johns
the sun than their present orbits and migrated outward, while Jupiter Hopkins University Applied Physics
Laboratory/Southwest Research Insti-
migrated inward (Gomes et al., 2005; Tsiganis et al., 2005; Morbidelli tute//Roman Tkachenko.
et al., 2005).
The large scale rearrangement of planetary orbits gravitationally
disturbed the trajectories of leftover planetesimals and smaller plan-
etary bodies. Some of these were expelled to the outer solar system 1
Named after the city in southern France
while others were directed inwards, causing a bombardment of the in- where the theory originated.
ner planets during the first 500 million years of the solar system. The
duration and intensity of the bombardment is still somewhat contro-
versial (Mann, 2018) but it certainly happened as evidenced by the
large number of old impact craters present on the Mercury and Mars.
Many of the dark areas visible on the face of the moon are basalt-filled
impact craters from this era. Continual renewal Venus’ crust by vol-
canism and Earth’s crust by plate tectonics has removed all evidence
of early impacts on these planets.
 topics in environmental physics: the solar system and earth 13

1.2.1 The Origin of the Earth-Moon System

The orbital havoc in the early solar system may also yield an expla-
nation of the unique Earth-Moon system. While moons roughly of
the size of our Luna are present around the massive giants and ice-
giants, none of the other rocky planets have large moons (Mars has
two tiny moons, Phobos and Deimos, both of which are less than 30
km in diameter). The most popular theory states that Earth’s moon
formed as a result of the collision between proto-Earth and a single
Mars-sized body, Theia (Stevenson, 1987; Wade and Wood, 2016) or
multiple smaller bodies (Rufu et al., 2017) around 50 million years af-
ter the proto-sun formed. Such a collision would have destroyed the
impactor(s), and have resulted in a total (re-)melting of Earth’s surface
while expelling a large amount of debris that quickly merged into a
large moon.
The formation of the moon was rather fortuitous for us because the
Earth-Moon gyroscope effect provides the earth axis tilt a considerable
stability. The axis tilt (currently: 23.44 degrees) is responsible for the
intensity of Earth’s seasons and changes by only one degree over tens
of thousands of years. Because of the lack of a large moon, Mars’ axis
tilt changes roughly between 10 and 50 degrees (Touma and Wisdom,
1993), causing large climate shifts. Without our Moon, Earth’s axis
would similarly wobble and result in an unstable climate. It is unlikely
that higher life forms would have evolved on a moonless Earth.

1.2.2 The Origin of Earth’s Water

The bombardment by comets after the formation of the moon can also
offer and explanation of the abundance of water on Earth. As men-
tioned earlier, the inner solar system was too hot for water ice to be
stable, and -even if water was present after the formation of Earth- the
heat generated by the violent collision with Theia would have expelled
most volatiles into space. A bombardment of icy comets allowed for
water to be deposited on Earth after the formation of the Earth-Moon
system.
The theory of cometary deposition of water has recently been chal-
lenged by Altwegg et al. (2015) using data collected on comet 67P/Churyumov–
Gerasimenko by ESA’s Rosetta probe. The isotopic ratio of deuterium
(D, or 2 H) and “ordinary” hydrogen (1 H) is very different on this
comet than the ratio found for Earth’s oceans. Unless the comet stud-
ied is different from the ones that would have deposited water on
Earth, the study of Altwegg et al. (2015) makes it unlikely that comets
would have deposited the majority of Earth’s water.
Recent work by Hallis et al. (2015) indicates that D/H ratios of
basalts derived from Earth’s mantle have a similar value as those in the
14 marcel g. schaap

pre-solar nebula. This indicates that most of Earth’s water was derived
from the planetesimals that formed Earth and was not derived from
comet impacts after the collision with Theia. Yet, it appears necessary
to invoke an impact of a Theia-sized body to explain the amount of
carbon in Earth’s mantle and crust (Li et al., 2016b; Budde et al., 2019)
as well as the almost identical isotopic compositions of the earth and
the moon. Additional evidence indicates that primordial water may
have survived the collision of proto-Earth and Theia (Lin et al., 2016;
Mercury
Williams et al., 2019). Some of these theories are contradictory and it is
clear that there are still many unknowns about formation of the early
solar system, including Earth and the origin of its water.
Venus

1.3 The Present Solar System Earth

Things settled down a few hundred million years after the formation Moon
of the solar system with the planets located near their present day
orbits. Figure 1.4 and Table 1.2 provide an overview of the major fea-
tures of the solar system, while its structure is depicted in Figure 1.5. Mars
The distance of a planet from the sun is given in astronomical units Ceres
(AU), which is defined by the International Astronomical Union as
Vesta
149,597,870,700 meters exactly, or nearly 150 million kilometers. The
AU is closely related (but not quite identical) to Earth’s distance from
the center of the sun. Jupiter

1.3.1 Terrestrial (Rocky) Planets

Saturn
There are four rocky planets: Mercury, Venus, Earth and Mars. Venus
and Earth are almost sisters regarding size and mass, while Mercury
and Mars are much smaller (Table 1.2). The density of the three in-
ner planets is considerable with values between 5243 and 5513 kg/m3
(Table 1.2), which is much larger than most silicate materials found at Uranus

the surface (quartz and feldspar minerals have densities of about 2650
kg/m3 ). These planets have sizable cores of iron and nickel. Mars has
a much lower density (3934 kg/m3 ) because it has a comparatively Neptune
much small iron.
As will be discussed in more detail in one of the next chapters,
Venus, Earth and Mars have atmospheres, while Mercury has none
due to its low gravity and exposure to an intense solar wind. Even Charon Pluto
though it has a mass and radius almost like Earth’s, Venus’ atmo-
sphere is extreme: it is more than 96% CO2 and contains sulfuric acid
Figure 1.4: Overview of the solar system.
clouds. The surface pressure of Venus is 92 bar, and its surface tem- The images are not to scale but scaled
perature is well over 400 °C. Liquid water (and life) cannot presently such to show some detail. Jupiter is
about ten times larger in diameter than
exist on Venus, though it is possible that more hospitable conditions Earth. All images indicate roughly how
were present several billion years ago (Way et al., 2016). the planet would look to the human eye,
except Venus which is shown in ultravi-
olet to exhibit some details in its other-
wise featureless cloud deck.
 topics in environmental physics: the solar system and earth 15

Mars, too, has an atmosphere that is mostly CO2 but with a surface
pressure less than 1/100th of that of Earth. As a result, surface tem-
peratures only occasionally rise above the freezing point of water near
equator. Liquid water cannot exist in stable form due to the low sur-
face pressure, though there is evidence for ephemeral channels. Large
water ice and CO2 ice deposits exist in Mars’ subsurface, with perma-
nent icecaps at the poles that are visible through small telescopes (see
Figure 1.4).
While orbiting the sun, each planet also rotates around a rotation Solar system structure and
axis. For Mars and Earth a full rotation takes about a 24 hours (see general characteristics of plan-
ets, including albedo: https:
Table 1.2). Mercury and Venus rotate much slower: it takes Mercury
//d2l.arizona.edu/content/enforced/
nearly 59 days to rotate (i.e., a Mercury “day” lasts 59 Earth days). 950156-867-2204-1ENVS420520001101/
Venus rotates in 243 Earth days but orbits in 0.62 years or 226 Earth 02_Solar_System/M2S4_hd_720.mp4

days. Venus (and also Uranus) further exhibits retrograde rotation,

which is indicated by the negative value for rotation in Table 1.2. If
it would be visible through the permanent global cloud deck, the sun
on Venus rises in the west and set in the East. The other planets which
have prograde (“normal”) rotation. As seen from above Earth’s north
pole, orbits around the sun are all clockwise, while planets with pro-
grade rotation turn clockwise around their axes.

1.3.2 Asteroid Belt

Beyond Mars, we find the main asteroid belt (the white dots in the top
panel of Figure 1.5) which is an area of thousands of rocky and some-
times icy bodies that are essentially leftovers from the formation of the
solar system. The largest of these are Vesta and Ceres, which have radii
that are 263 and 473 km, respectively (Figure 1.4). Although these bod-
ies are much smaller than Earth, or even our moon (radius: 1737 km),
Vesta and Ceres had sufficient mass to gravitationally form roughly
spherical bodies Most other asteroids are smaller and less massive,
and consequently these bodies have irregular shapes. The total mass
of the asteroid belt is 0.0006 Earth masses or about two Ceres masses
(Krasinsky et al., 2002).
Jupiter is the main reason for the existence of the asteroid belt be-
tween it and Mars: Jupiter’s gravity caused so much disturbance in
this region that no true planet could ever form by prolonged accretion.
The many bodies that formed established an equilibrium between ac-
cretion and occasional collisions that would again break up the plan-
etesimals.
The bottom panel of Figure 1.5 shows an oblique view of the solar
system, with the asteroid belt as a flattened disk of tens of thousands
of bodies (only 200 are shown here). This disk indicates that the orig-
inal flattened structure of the protoplanetary disk (Figure 1.1) is still
16 marcel g. schaap

Figure 1.5: Distribution of bodies in the

inner solar system up to the orbit of Sat-
urn. Top: vertical view; bottom: oblique
view. For visualization purposes only
distances are to scale, object sizes are
greatly exaggerated even for the sun and
have no accurate relative values. The sun
is the orange object in the center of both
images; Mercury, Venus and Earth are
over exposed, Mars, Jupiter and Saturn
exhibit their natural colors. The small
white bodies grouped in a disk between
the orbits of Mars and Jupiter are 200
selected asteroids. Blue and green ob-
jects are comets and more distant as-
teroids (trans-Neptunian Kuiper-belt ob-
jects), respectively and are less confined
to the plane in which the planets orbit.
The diagonal line is the inclined orbit of
Pluto, which is a Kuiper-belt objects as
well as a “minor planet”. Credit: Marcel
G. Schaap.
 topics in environmental physics: the solar system and earth 17

present in the current solar system.

1.3.3 Gas and Ice Giants

Jupiter is the largest planet in the solar system. It has a mass that
is nearly 318 times larger than Earth’s, and a radius that is 11 times
larger. Compared to the sun, it is small however, having only 0.1% of
its mass and 10% of its radius.
Jupiter orbits the sun at 5.2 AU and consists mostly of H2 and He.
Its center consists of a hot rocky iron-nickel core with a temperature
of 20,000 K (Figure 1.6, NASA/Lunar and Institute, 2003). The core is
surrounded by a shell of high-pressure metallic hydrogen, above which
there is a thick layer of molecular hydrogen and helium. Jupiter has a
distinct banded cloud structure that is easily perceived through a small
telescope. The cloud bands are indicative of a global circulation system
somewhat similar to that of Earth. Colors in the cloud bands are due
to varied concentrations of water, methane, ammonia, and hydrogen
sulfide.

Figure 1.6: Cross-sections of the gas and

ice planets. Credit: NASA/Lunar and
Planetary Institute.

Saturn is a third as massive as Jupiter but, with a radius that is not

much smaller, its mean density is less than that of water (687 kg/m3 ,
see Table 1.2). Similar to Jupiter, Saturn has a rocky iron-nickel core
surrounded by a metallic hydrogen shell above which there is a liquid
hydrogen and helium shell followed by a gaseous atmosphere con-
taining water clouds, hydrogen sulfide clouds and ammonia clouds
(NASA/Lunar and Institute, 2003). Saturn’s banding is not as distinct
as that of Jupiter. Saturn is, of course, best known for its ring system
which consists of micron to boulder sized particles of mostly water ice.
Uranus and Neptune are smaller and less massive ice giants (Table
1.2, Figure 1.6). Their cores are probably also rock and iron-nickel.
18 marcel g. schaap

Because their lower mass limits the internal pressure (necessary for
hydrostatic equilibrium), no metallic hydrogen layer is expected to
exist on Uranus and Neptune. Instead, the cores is thought to be
surrounded by water, ammonia, and methane. The upper layers are
predominantly hydrogen, helium and methane gases. The methane
content of Uranus and Neptune is larger than that of Jupiter and Sat-
urn, which is a result of the colder conditions in the protoplanetary
disk where these ice giants formed (see also Table 1.1).

1.3.4 Kuiper Belt and Oort Cloud

The Kuiper belt is a collection of mostly icy objects beyond the orbit of
Neptune with only a few members (green) visible in the bottom panel
of Figure 1.5. The minor planet Pluto and its largest moon Charon, are
the most well-known members of the Kuiper belt. Compared to Earth,
the density of Pluto is low (2050 kg/m3 , Table 1.2), indicating that
Pluto’s interior probably consists of silicate rock surrounded by a thick
layer of ice. With a mean distance of nearly 40 AU and an insolation
of only 1 W/m2 , it is very cold on Pluto. Imagery collected by NASA’s
space probe New Horizons, which flew by Pluto in 2015, demonstrated
that Pluto’s surface is made of water ice “rocks” and slow-moving
glaciers made of nitrogen ice. Pluto has a very thin atmosphere with a
current pressure of about 1 Pa (i.e. 1/100,000th of the sea level pressure
on Earth). Arrokoth (Figure 1.3) is a much smaller member of the
Kuiper belt.
The last major, but most distant, component of the solar system is
the Oort cloud (not depicted in Figure 1.5) which extends from about
2000 AU to about a light-year from the sun. The Oort cloud is home
to possibly trillions of kilometer-sized cold and icy planetesimals that
were left over after the formation of the solar system. Continuous
small gravitational disturbances by the large gas and ice giants some-
times nudge these bodies into inner solar system. When they come
closer to the sun (typically within the orbit of Jupiter) the bodies warm
up sufficiently for the ices to sublimate and release large amounts of
gas and dust particles. Occasionally these comets approach Earth near
enough to be seen with the naked eye.

Figure 1.7: Comet ISON on Nov. 19,

2013 at 71 and 129 million km from the
1.4 The Internal Structure of the Terrestrial Planets sun and the Earth, respectively. Credit:
NASA/MSFC/Aaron Kingery.
The heat released during the formation of the rocky inner planets left
those bodies in a completely molten state which allowed for the dens-
est elements to sink to the center of the planets. In the case of Earth,
and with some variations also for the other planets, this resulted in
an initially liquid core made of mainly iron and nickel, with less dense
 Body Massa Distanceb Orbita Ecc.b Incl. Radiusa Rotationa Axis g Press. Densitya S(d) Albedoa
Geometric Bond
(Earths) (AU) (year) (-) ° (km) (day) (deg) (m/s2 ) (Pa) (kg/m3 ) W/m2 (A g ) (Ab )

Sun 3.33×105 1.00 - - - 695,000 25.38 7.25 274.0 12.5 1408 - - -

kPak

Mercury 0.055 0.387 0.2408 0.206 7.0 2439.7 58.646 0.034 3.7 ~0 5427 9093 0.106 0.068
Venus 0.815 0.723 0.6152 0.007 3.4 6,051.9 -243.013 2.64 8.87 9.2 5243 2605 0.65 0.77
MPa
Earth 1 1.000 1.0000 0.017 0 6,378.1 0.99727 23.44 9.81 101.3 5513 1361 0.367 0.306
kPa
Mars 0.107 1.523 1.8808 0.093 1.9 3396.1 1.02595 25.19 3.71 ~0.8 3934 586.8 0.17 0.25

topics in environmental physics: the solar system and earth

kPa
Jupiter 317.8 5.203 11.862 0.048 1.3 71,492 0.41354 3.13 24.8 -e 1326 50.28 0.538 0.343
Saturn 95.16 9.537 29.447 0.054 2.5 60,268 0.44401 26.93 10.4 -e 687.1 14.97 0.499 0.342
Uranus 14.54 19.19 84.017 0.047 0.8 25,559 -0.71833 97.77 8.87 -e 1270 3.696 0.488 0.300
Neptune 17.15 30.07 164.79 0.086 1.8 23,764 0.67125 28.32 11.2 -e 1638 1.505 0.442 0.290

Vesta 2.6×10-4 2.362 3.6294 0.099 7.1 262.7d 0.2226 ≈29l 0.25 0 3456 243.9 0.423 0.20j
Ceres 9.4×10-4 2.767 4.6040 0.116 10.6 473d 0.3781 ≈4m 0.29 0 2160 177.8 0.09 0.24i
Pluto 0.002 39.48 247.92 0.249 17.2 1151 6.3872 119.6 0.66 ~1 Pa 2050 0.8730 0.5-0.7 0.4-0.6

Moon 0.0123 383,399 km 27.32d 0.055 5.1 1737 27.32 1.67 1.62 10-7 3346 1361 0.12 0.11
Pa

Table 1.2: Selected properties of the planets in the solar system. Abbreviations: AU: Astronomical Unit (1 AU is 149,597,870.700 km, by definition); Ecc.: eccentricity; Incl.:
inclination of the orbit with respect to the ecliptic plane (Earth: 0 by definition); g: acceleration of gravity; Press: surface pressure (gas and ice giants have no surface); S(d):
planet-specific solar constant as a function of distance, d (Eq. 1.1). References and notes: a : NASA (2015b); b : pdf file aprx_pos_planets.pdf listed on NASA (2015a), for the
moon we list the distance to Earth in km; c : Sheppard (2015); d : mean; e : gas/ice planets have no solid surface and therefore no defined surface pressure; f : NASA (2016); l :
bond albedo from Li et al. (2016a); j : Li et al. (2013); k : bottom of photosphere; ’: Thomas et al. (1997); m : Schorghofer et al. (2016).

19
20 marcel g. schaap

elements settling above this core as a mantle that is made of high-

temperature and high-pressure silicates. The surface crust (or litho-
sphere) is situated above the mantle and is made of low-temperature
and low-pressure silicate rocks.

Figure 1.8: Cross-section of present day

Earth (white: inner core, yellow: outer
core, orange mantle, green: crust (not to
scale). The white lines indicate the mag-
netic field generated by convection in the
outer core (black ovals).

Density differences among elements therefore caused a differentia-

tion of our planet a core-mantle-crust-ocean-atmosphere system as il-
lustrated in Figure 1.8 with some present day characteristics listed in
Table 1.3. The densest material (metal alloys, mantle minerals) is found
deep inside the Earth. These forms of matter are dense not only be-
cause they contain elements with higher atomic numbers (Fe, Ni), but
also because of the immense pressure needed to maintain hydrostatic
equilibrium. Similar to the sun, each of Earth’s layers develops an in-
ternal pressure high enough to balance the overlying weight (=mass
times gravitational acceleration). The intense pressure compresses the
matter to much higher densities than it would have on Earth’s surface.
The iron-nickel core consists of two layers: a solid crystalline inner
core and a liquid outer core. The inner core has a radius of about 1200 Introduction to this chapter: https:
//d2l.arizona.edu/content/enforced/
km, a temperature over 5000 kelvin, and pressure of 360 GPa (giga-
950156-867-2204-1ENVS420520001101/
Pascals, more than 3 million times atmospheric pressure at sea level). 02_Solar_System/M2S2_hd_720.mp4
The outer core is a shell with a radius of 3500 km, making the outer
core a 2300 km thick layer around the inner core.
The outer core is gradually solidifying upon the inner core by about
 topics in environmental physics: the solar system and earth 21

Depth Main elemental composition Density Temperature Pressure

(km) (dominant phase) 103 kg/m3 °C
Ocean/Atmosphere 0 H, C, N, O, Ar, Ne (air/water) 1.2-1000 15 100 kPa
Crust (continental) -8-70 O, Si, Al, K, Na, Fe (Granites) ~2.7 15-500 24 GPa
Crust (ocean) 5-10 O, Si, Mg, Fe, Ca (Basalts) ~2.9 15-500 24 GPa
Mantle 35-2890 O, Si, Mg, Fe (peridotite) 3.4-5.5 500-4000 140 GPa
Outer Core 2890-5150 Fe, Ni (liquid) 9.9-12.2 ~4000 330 GPa
Inner Core 5150-6360 Fe, Ni (solid) 12.6-13.0 ~5400 360 GPa

Table 1.3: General characteristics of

Earth’s core, mantle, crust and its ocean
1 mm per year. This phase-transition releases heat and expels some ele- and atmosphere.
ments that are not compatible with solid iron-nickel core. In turn, this
results in convection in the liquid core that -together with the rotation
of Earth- generates a global magnetic field. This field leads to a magne-
tosphere which extends several Earth radii into space on the sun-facing
side. Except for the strongest solar flares and CMEs, the magneto-
sphere deflects most of the solar wind and prevents erosion of Earth’s
atmosphere.

Tectonic uplift
(mountains and
plateaus) Figure 1.9: Simplified depiction of the
Sediments Mid-Oceanic upper mantle, asthenosphere and the
Ridge oceanic and continental lithosphere (not
Ocean
to scale).
Oceanic crust Continental crust

Astenosphere

Mantle
Sedimentary
basins

Earth’s mantle and lithosphere both consist of silicate rocks. Due to

the high pressure and temperatures, the lower mantle consists mostly
of high-density perovskite minerals while the lower pressure and tem-
perature upper mantle holds lower-density peridotite rock (which con-
tains olivine and pyroxene minerals).
The oceanic lithosphere consists primarily of low temperature and
low-pressure basaltic rock (in which feldspar, pyroxene and olivine
minerals are dominant), while the lower continental lithosphere con-
sists of granitic rocks (containing quartz, feldspar, mica and amphibole
minerals). The lower density of crustal rocks (2.7 − 2.9 × 103 kg/m3 )
allows them to float and slowly move on the viscous top of the man-
tle called the asthenosphere. This allows continental drift to occur along
with faulting in the upper brittle lithosphere (Figure 1.9).
Although subject to faulting, orogeny (mountain forming), igneous
intrusion of magma, and erosional processes, continental crust is usu-
22 marcel g. schaap

ally more stable than oceanic crust. The stablest sections of conti-
nental lithosphere are known as cratons and are billions of years old.
Oceanic crust is much younger than continental crust (less than 200
million years) because it is constantly renewed at mid-oceanic ridges
(1.9) while it disappears by subduction into the mantle under the less
dense granitic continental crust (Chopin, 2003). Extension at mid-
oceanic ridges and sinking motion near subduction zones is in the
order of millimeters to centimeters per year. Continental and oceanic
crusts are usually covered with sediments that are erosion products
of primary crustal rocks or other sediments. Metamorphic rocks are
formed in the lithosphere where temperatures and pressures are high
enough to physically and chemically alter igneous and sedimentary
rock.
The ocean and atmosphere are the least dense of all and therefore
situated above the crust. As discussed in the following section, both
contain volatile elements with low melting points.

1.5 Abundance of Elements on Earth

The formation of the solar system and the differentiation of Earth also
explain the abundance of elements in Earth’s crust. Figure 1.10 shows
that Earth’s elemental abundance exhibits both similarities and differ-
ences with the abundance of elements in the solar system that was
discussed in the previous module.
It is useful to use the Goldschmidt (1937) classification into volatile,
chalcophile, lithophile and siderophile elements as indicated by the back-
ground colors in Figure 1.10. The primary reason for this division is
due to the electron configuration of the elements which determines
their place in the periodical system, their chemical reactivity but also
Goldschmidt N Density
their density (Table 1.4). To normalize cosmic and crustal abundances Classification kg/m3
both series to the same scale, both were by their respective abundance Volatile 8 0.210
of silicon, one of the dominant elements in the crust. Data for Earth’s Lithophile 47 1.124
Chalcophile 17 3.960
inaccessible mantle and core are less reliable and are not shown in the Siderophile 13 7.869
figure.
Table 1.4: Goldschmidt classification
with mean elemental densities. The den-
1.5.1 Volatiles sity is given as an abundance-weighted
quantities (abundance as found for
Earth’s crust). N indicates the number
Volatile elements are noble gases (He, Ne, Ar, Kr, Xe), or form com- of elements in each class.
pounds such as H2 O, N2 , CO2 , all with low melting and boiling points
and therefore subject to degassing from the mantle and crust. Relative
to the cosmic abundance, volatile elements are depleted in the crust
but dominant in the oceans and atmosphere. Carbon is an exception.
Even though it is classified as a volatile, most of Earth’s is present as
organic matter (kerogen) and limestone in Earth’s crust due to biolog-
 topics in environmental physics: the solar system and earth 23

ical deposition. Because of this, CO2 is a minor atmospheric gas on

Earth (currently around 400 ppm). By contrast, CO2 is the dominant
gas in the Venusian atmosphere because of the absence of life. Al-
though the volatile carbon is sequestered in Earth’s crust, Figure 1.10
indicates that it is four orders of magnitude less abundant on Earth
than in the cosmos. The main reason for this is that Earth was assem-
bled from volatile-poor planetesimals that formed near the hot young
sun, as discussed previously.
Although the oxygen concentration in Earth’s atmosphere is high
(nearly 21%), this element is not classified as a volatile because it is
also controlled by life. Without active photosynthesis, the oxygen con-
centration in Earth’s atmosphere would eventually decline to very low
levels because of oxygens reactivity. Because it forms strong bonds
with silicon, oxygen is classified as a lithophile element.

1.5.2 Lithophiles

Lithophile (“rock loving”) elements dominate the composition of Earth’s

crust and typically bind strongly with oxygen. Most lithophile ele-
ments are silicates (silicon bearing minerals) and have a relatively low
density, allowing the crust to float on the mantle. With respect to the
cosmic background, the crust is enriched in the elements Rb to Nb, Cs
to Ta (the “rare earth metals”) and also Th and U. The main reason for
enrichment is that these elements are not compatible with the metal-
lic iron core and fractionated out as low-density oxides. Th and U
are about 1000 times more abundant in the crust than they are in the
cosmos and useful for energy production in nuclear fission reactors.

1.5.3 Chalcophiles

Chalcophile elements form strong bonds with sulfur, rather than the
chemically similar oxygen (both elements belong to group VI in the
periodical system). Chalcophiles tend to form ore bodies of insoluble
sulfides, which concentrates these minerals and facilitates mining op-
erations. Sulfide minerals tend to be more dense than lithophile ele-
ments (Table 1.4) and are generally located deeper in the crust or in
the upper mantle.

1.5.4 Siderophiles

Most of Earth’s iron and nickel settled into core. Siderophile elements
are able to dissolve in the liquid outer core or amalgamate into the
solid Fe − Ni core. Siderophile elements are therefore all depleted in
the crust, relative to the cosmos. For example, gold and platinum as
24 marcel g. schaap

well as some other elements are (very) rare in the crust, but they are
likely enriched in the core because of their siderophillic nature.
Iron itself, however, is not particularly depleted in the crust and
it is actually more abundant than some common lithophile elements
such as K, Ca and Ti. The relative abundance of Fe in the crust is
a consequence of iron-containing ultra-mafic volcanism that occurred
during the first few billion years. The ubiquitous iron ore bodies
presently found in Earth’s crust are actually a result of the evolution of
cyanobacteria that use oxygenic photosynthesis to produce O2 which
reacted with soluble ferric (Fe2+ ) into insoluble iron Fe2 O3 and Fe3 O4
(hematite and magnetite) that precipitated out in oceans as banded
iron formations.

Pm

Am
Tm
Mn

Ga
Na

Co
Cu

Rb

Nb

Rh

Ho

Re

Np
Ag

Sb

Cs

Eu

Au

Pa
Sc

As

Tb

Ac
Tc

La

Lu
Ta
Br

Pr
Cl

At
Fr
Al

Bi
In

Tl
Li
H

N
B

Y
F

Ir
I
He
Be
C
O
Ne
Mg
Si
S
Ar
Ca
Ti
Cr

Ni
Zn
Ge
Se
Kr
Sr
Zr
Mo
Ru
Pd
Cd
Sn

Xe
Ba
Ce
Nd
Sm
Gd
Dy
Er
Yb
Hf
W
Os
Pt
Hg
Pb

Rn
Ra
Th
U
Pu
Fe

Po
Te

Figure 1.10: Cosmic and Earth’s crustal

Solar system
abundance of the elements relative to
1e+03
that of silicon. The Goldschmidt
Abundance (mole fraction relative to Si)

Earth’s crust
Si=1 classification into volatile, chalcophile,
Volatile
Chalcophile
lithophile and siderophile is provides as
1e+00 background color. A white background
Lithophile
Siderophile indicates that the element is unstable
Unstable
and does not occur in significant quan-
1e−03 tities. The downward pointing trian-
gles under the element labels (top) in-
dicate that the U.S. Government consid-
1e−06 ers these elements (or their compounds)
critical for national security and econ-
omy (Appendix A)
1e−09
1 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Atomic number (Z)

1.5.5 Availability of Elements

Our modern society thrives on being able to use elements for particular
technological purposes. However, as indicated in Figure 1.10 not every
element is equally abundant and shortage of certain elements would
severely hamper global and national economies. Shortages of select
elements would further pose a security risk because they are essential
for weapons systems.
The triangles at the top of Figure 1.10 indicate which elements are
deemed critical for the U.S. national security and economy (USGS,
2019). These are mostly elements which are of essence to national
defense or without which the economy would be severely affected,
but which must be imported because no, or at most, limited pro-
duction of processing facilities exist within the USA. The availabil-
ity of these elements may therefore be affected by international trade
(dis)agreements, (inter)national disasters, politics, or conflicts. To avoid
 topics in environmental physics: the solar system and earth 25

sudden shortages, the USA has created stockpiles of elements such

as beryllium, germanium, helium, lithium, and rare-earth elements
(REE), among others. Appendix A provides a more in-depth discus-
sion of some of these elements and their technological applications.

1.6 A Simple Planetary Surface Temperature Model

The sun is the single energy source in the solar system. Some of the
Introduction and context of
sun’s energy is absorbed by the planets and converted into heat, giving a simple planetary surface
each a characteristic mean surface temperature. It is therefore interest- temperature model: https:
//d2l.arizona.edu/content/enforced/
ing to digest the data in Table 1.2 and to formulate a simple model
950156-867-2204-1ENVS420520001101/
that predicts the surface temperature of a planet using the solar en- 02_Solar_System/M2S3_hd_720.mp4
ergy received, the planet’s albedo, and the law of Stefan-Boltzmann
for black-body radiation. With such a model, we can quickly analyze
why Earth is so unique, but also immediately indicate why Earth re-
quires an atmosphere greenhouse effect to support liquid water and life.
The central assumption in our model is that the amount of energy
received from the sun by a planet equals the amount of energy that
is re-radiated back into the universe as infrared black-body radiation.
This is also known as thermal equilibrium. A planet would warm up if
it receives more solar radiation than it emits as infrared (IR) back into
space. Conversely, the planet would cool down if its IR emission is
larger than the radiation it receives from the sun. In both cases, the
change in temperature would cause a change in IR emission and re-
sult in a new thermal equilibrium. We therefore have a simple balance:
energy absorbed by the planet equals the energy emitted as IR radia-
tion, which is controlled by the temperature of our planet. Figure 1.11
provides a schematic overview of the model.

Figure 1.11: Graphical overview of the

simple surface temperature model. The
open triangular arrows indicate visible
Inf
n
ed radiatio

light (received or reflected solar radia-

r
ared radiatio

tion), the filled barbed arrows represent

infrared radiation.
ec t fl

n
Re

day night

Solar radiation Earth, Re and Tsurf

The amount of energy received (in J/[m2 s], or W/m2 ) by a planet

falls off with the square of the distance (d in meters) from the sun:
26 marcel g. schaap

L⊙
S(d) = (1.1)
4πd2
where L⊙ is the sun’s luminosity (3.828 × 1026 J/s). Planet-specific
values of the solar constant S(d) are listed in Table 1.2.
The amount of energy absorbed by a planet is proportional to its
solar constant times the area that intercepts the radiation (i.e. the day-
light side), less the amount that is reflected back into space without
ever being “used”. Albedo (A) is the fraction of radiation reflected
back into space.
There are two types of albedo: Geometric albedo (A g ) used for Figure 1.12: From the perspective of the
sun, spherical planets look like disks
imaging purposes, and Bond albedo (Ab ), which is useful for energy (shadows). The intercepted solar radi-
balance calculations, such as done here (also see Appendix B). Values ation is therefore proportional to πR2 ,
with R being the radius of the planet.
for the Bond albedo listed in Table 1.2 indicate that no planet is a per-
Shown are (left to right): Mercury,
fect reflector (Ab = 1), nor a perfect absorber (Ab = 0). Figure 1.11 Venus, Earth and Moon, and Mars.
represents incoming solar radiation and reflected solar radiation with
triangular arrows. The amount of energy absorbed by Earth is:

Ein = (1 − Ab )S(d)πR2e (1.2)

Where Re is the radius of Earth, and πR2e is the area that intercepts
the solar radiation (see also Figure 1.11). Note that this corresponds to
the area of a disk and not the area of a hemisphere because the planets
“look” like disks from the perspective of the sun (see Figure 1.4).
Let us now assume that the amount of absorbed solar radiation is
Video about the development of
converted into heat that gives our planet a surface temperature (Tsur f ). the temperature model. https:
Most solid or liquid materials objects emit black body radiation that is //d2l.arizona.edu/content/enforced/
proportional to the fourth power of their temperature. This energy is 950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S5_hd_720.mp4
radiated into space as infrared (IR) radiation and allows the planet
to maintain thermal equilibrium by balancing the absorbed solar ra-
diation. We will assume that there is no storage of absorbed energy
(e.g. heat stored in oceans, rocks, soils or the atmosphere) and, impor-
tantly, we assume that the entire spherical surface of our planet has
one uniform temperature. The IR emission is therefore uniform over
the surface as indicated by the red barbed arrows in Figure 1.12.
We must, however, immediately realize the limitations of this model.
Especially the assumption of isothermal conditions is unrealistic be-
cause it ignores the usually lower temperatures at the night side of the
planet (Figure 1.11). There would also be a problem near the poles
where the rays of the sun will come in more obliquely, thus reducing
the intensity of solar radiation which results in less heating and less IR
emission. We also ignore the axis tilt of a planet, which would cause
opposite seasons on each hemisphere as the planet goes around its or-
bit. However, a uniform planetary temperature is an assumption we
are presently willing to make in order to keep our model simple.
 topics in environmental physics: the solar system and earth 27

The total amount of energy emitted as black body radiation (Ebb ) is

governed by the law of Stefan-Boltzmann:

4
Ebb = ǫσsb Tsur f (1.3)

where σsb is the Stefan-Boltzmann constant2 ; Tsurf must be given in de- 2

5.6705 × 10−8 kg s-3 K-4
grees kelvin. We will assume that we have a perfect “black body” with
an IR emissivity (ǫ) of 1, which is not unreasonable for the infrared
wavelengths at which the energy is emitted. We will leave the emis-
sivity out of consideration for simplicity. Equation 1.3 gives the black-
body radiation for one square meter and the total IR output (Eout ) for
a spherical planet is therefore:

4 2
Eout = σsb Tsur f 4πRe (1.4)

where 4πR2e is the area of the planetary sphere.

We assume thermal equilibrium, which means that the total ab-
sorbed energy (Ein ) equals the amount of emitted black body radiation
(Eout ), therefore:

Ein = Eout
(1.5)
4 4πR2
(1 − Ab )S(d)πR2e = σsb Tsur f e

The term πR2e on the left-hand side (lhs) and right-hand side (rhs) can-
cels and can be removed. Because we would like to know the surface
temperature, we must now solve for Tsur f . After some simple rear-
rangement of the symbols, we arrive the following expression:
1
(1 − A b ) S ( d )

4
Tcalc = Tsur f = (1.6)
4σsb
This model tells us that the surface temperature depends on the solar
constant (S(d)), which is different for each planet, see Table 1.2) and
the planet’s Bond albedo (Ab ) and no other variables (σsb is a physical
constant).
How useful is our model? This can be evaluated by comparing
Video about an evaluation of
predicted model outcomes with reality (observations). To this end we the temperature model. https:
make explicit that Tcalc = Tsur f (Eq. 1.6) is a model prediction, while //d2l.arizona.edu/content/enforced/
Tobs is the observed surface temperatures (i.e reality). Table 1.5 lists the 950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S6_hd_720.mp4
values for eight planets and three other solar system bodies as well
as the disagreement between observations and calculations, Tdi f f =
Tobs − Tcalc . A positive value indicates that the model overestimates
the real temperature, a negative value implies an underestimate.
The temperature difference should be zero for a perfect model, but
by glancing at Table 1.5 it is clear that the model certainly is not per-
fect (e.g., Venus). On the other hand, the small differences for some
28 marcel g. schaap

Planet S(d) Ab Tcalc Tobs Tdi f f

Table 1.5: Calculated (Tcalc ) tempera-
(W/m2 ) (K) (K) (K) tures, Observed (Tobs ) temperatures, and
Merc. 9088 0.068 440 440 0 temperature differences (Tdi f f = Tobs −
Tcalc ) for all the bodies listed in Table 1.2.
Venus 2604 0.77 227 737 510 a : at the 1 atmosphere level, these plan-

Earth 1361 0.306 254 288 34 ets do not have a solid surface.
Mars 586.8 0.25 210 208 -2
Jup.a 50.28 0.343 110 163 53
Sat.a 14.97 0.342 81 133 52
Uran.a 3.696 0.3 58 78 20
Nep.a 1.505 0.29 47 73 26
Ceres 177.8 0.24 156 168 13
Vesta 243.9 0.20 171 177 6
Pluto 0.8730 0.5 37 44 7

bodies indicate that our simple model works quite well in some cases
(Mercury, Pluto, Ceres, Mars). So, the model has at least some merit.
It is evident that the model works poorly for larger bodies. We
already mentioned Venus where the observations are 510 K hotter than
modeled, while Earth is 34 K warmer. Jupiter and Saturn are also
substantially warmer by about 50 K. As will be discussed further in
a next chapter, both Venus and Earth have atmospheres that exhibit
greenhouse effects (which is extreme for Venus), allowing the trapping
of heat. The situation is different for Jupiter and Saturn: even after 4.6
billion years these planets have still not reached thermal equilibrium
state and have large internal sources of heat. This make these planets
warmer than calculated from a solar energy supply alone.
Even though our surface temperature model is not very accurate (it
is much too simple, really), in some cases it is useful because the differ-
ence with the observations tells us that we have overlooked something,
namely a planet’s greenhouse effect or internal heat production. The
discrepancy between the model outcomes and observations can there-
fore give us hints where the model can be improved.
The model is also useful in a qualitative sense. It predicts that an
increase in S will lead to an increase in temperature while an increase
in albedo (Ab ) will lead to a temperature decrease. Both are demon-
strated with an experiment on a scale model in Appendix C. This
aspect of the model is relevant when studying Earth because more
clouds and larger ice-caps lead to more reflection of sunlight (increase
in albedo) and therefore lower temperatures. Conversely, shrinking ice
caps, snow fields, and sea ice tend to decrease the albedo and lead to
more absorption of solar radiation. In the following chapters we will
discuss refinements to the model, such as the greenhouse effect, global
circulation in atmosphere and oceans.
The usefulness of the model is also demonstrated by a major recent
 topics in environmental physics: the solar system and earth 29

finding of a planetary system around a red dwarf star, TRAPPIST-1.

Gillon et al. (2017) the model given by Eq. 1.6 to calculate the surface
temperature for each of the seven discovered planets. To approximate
a greenhouse effect, the study set the albedo to zero and found that
three planets could host water oceans, provided that they have an at-
mosphere similar to that of Earth. Whether this is indeed the case
cannot (yet) be observed as the system is nearly forty light-years away
from us.

1.7 Earth’s Uniqueness

No planet in the solar system is quite like another, but it appears that
Earth is truly unique in the sense that life evolved there. Several factors
are important. First of all Earth has an enormous amount of liquid wa-
ter with an active hydrological cycle. The water is not only necessary
for life, but also allows for efficient, mixing and transport of nutrients
and dissolved gases. Water allows for weathering of rocks and (facili-
tated by oceanic micro-organisms) deposition of excess carbon-dioxide
as limestone.
Earth has liquid water due to its distance from the sun and an at-
mosphere with a pressure that is sufficiently high. As a result of a
moderate greenhouse effect, its surface temperatures are neither too
cold such that all water freezes (Mars), nor too hot (Venus) such that all
water eventually vaporizes. Without a greenhouse effect, Earth’s water
would probably frozen solid too, because our temperature model in-
dicated that Earth’s mean temperature “should” be 254 K (-18 °C). The
atmospheric pressure is important because it raises the boiling point of
liquids. Mars’ surface pressure (about 0.8 kPa, or less than 1 millibar)
is too low to permit liquid water at its surface.
Earth’s magnetic field is also crucial as it protects our atmosphere
from the solar wind. Mars, on the other hand, lost its magnetic field,
most of its atmosphere and most of its liquid water early in its his-
tory. The water that remained froze solid in its subsurface or its ice-
caps. The erosion of its atmosphere continues into the present as doc-
umented recently by NASA’s Maven orbiter (Jakosky et al., 2015) and
further analysis by Fedorova et al. (2020). Although after Earth, Mars
is the most hospitable planet for human settlement, it is unlikely that
we will be able to create Earth-like conditions (“Terraform”) on Mars
- even if we had the technology to do it. Recent evidence indicates
that there is insufficient CO2 in Mars’ icecaps to create a sufficiently
strong greenhouse effect for plants or humans to survive at the surface
(Jakosky and Edwards, 2018).
Based on mass and size, Venus is the most Earth-like planet in the
solar system (Table 1.2), yet its atmospheric conditions are character-
30 marcel g. schaap

istic of a runaway greenhouse effect. Even though Venus is closer to the

sun with a solar constant that nearly twice that of Earth, its calcu-
lated surface temperature is lower than that of Earth (227 K vs 254
K, respectively), yet its observed surface temperature is no less than
510 K hotter than calculated (1.5). Several factors have played a role.
Like Earth, most of Venus’ carbon was out-gassed as CO2 by its man-
tle through volcanism. Unlike Earth, however, this CO2 was not se-
questered into the Venusian crust as limestone as there was no life or
liquid water. The CO2 accumulated in the atmosphere and caused the
runaway greenhouse effect.
Venus probably did have liquid water in its early history (Way et al.,
2016) but lost nearly all of it to space because the planet lacked a mag-
netic field. The absence of a magnetic field is explained by the planets
low rotation rate, and possibly the suppression of core convection due
to low heat transport caused by the absence of plate tectonics (Nimmo,
2002) or stratification of the core in layers of different compositions
(Jacobson et al., 2017). Intense solar ultraviolet radiation in the upper
Venusian atmosphere would have dissociated water into hydrogen and
oxygen, which were then removed by the solar wind.
Even if we had the technology to lower the temperature of the Venu-
sian surface and atmosphere, the enormous amount of CO2 and asso-
ciated pressure would remain. If terraforming Mars is hard for lack of
volatiles, terraforming Venus is nearly impossible because there is too
much CO2 . It appears that Earth is just right.
2
Celestial Mechanics, Seasonality and Climate

2.1 Introduction

It would be very convenient if Earth and other bodies in the solar sys-
tem orbited the sun with constant circular motions without any mutual
effects and without any change. Unfortunately, this is not the case. We
have already seen that Jupiter’s gravity had a dominant role in shaping
of the structure of the early solar system by affecting the orbits of other
planets. This continues to this day: all planets exert small gravitational
forces continuously change the orbital characteristics of, especially, the
less massive bodies in the solar system. The result of this is that all
planets follow quasi-periodically changing elliptical (non-circular) or-
bits around the sun (Figure 2.1).
Planets also spin around their respective rotation axes, which are
tilted (oblique) with respect to the plane of their orbits, thus caus-
ing the seasons that are so apparent on Earth. Slow orbital and ax-
ial changes can change the solar intensity (insolation), particularly at
higher northern and southern latitudes which has the potential to
cause cold and warm climatic periods (glacials and interglacials, re-
spectively).
The interplay of orbits, axial tilt and their changes is complicated,
and we will discuss these one by one. To this end, we will first discuss
the most important characteristics that are important for seasonality on
Earth. With seasonality we mean the variation of solar energy input
throughout the year which drives the annual pattern in weather, ecol-
ogy, and agricultural production. We will see that there are global ef-
fects and latitude-specific effects. We will then study the slow changes
in orbital parameters and discuss their significance for long-term quasi
periodic changes in our global climate.

Key points of this chapter are:

• The energy received by the entire planet depends on Earth’s dis-

32 marcel g. schaap

tance from the sun, which is controlled by time and the eccentricity
of Earth’s orbit.

• The energy received during a day at a given point on the planet is

controlled by the tilt of its rotation axis (23.44 degrees), latitude and
position of Earth in its orbit (i.e. time of year).
• Earth’s closest (farthest) distance from the sun is known as perihe-
lion (aphelion). Perihelion and aphelion are also known as apses.
• A solstice occurs when Earth’s rotation axis aligns with the sun. A
summer solstice occurs when the axis maximally tilts toward the
sun, a winter solstice happens when it tilts away from the sun.
There are two solstices per year. A summer solstice on the north-
ern hemisphere (June 21st ) implies a winter solstice in the Southern
hemisphere.
• A major axis of an orbit connects perihelion and aphelion. A minor
axis is perpendicular to the major axis and cuts it in half at the center
of the orbit. Half of the major axis is known as the semi-major axis
and is used to characterize orbits.
• The sun is offset from the center of the orbit in a location named the
focus. The offset divided by semi-major axis is known as eccentric-
ity of an orbit. Earth’s orbital eccentricity is 0.0167.
• Orbital changes caused by gravitational interaction between Earth,
the moon, the sun and planets slowly affect axial tilt and orientation,
as well as the eccentricity and orientation of Earth’s orbit.

• The orbital dynamics affect Earth’s climate by causing quasi-periodic

changes in insolation with time scales in the order of tens of thou-
sand of years and offer a partial explanation the waxing and waning
of ice ages during the past 2.5 million years.
• Energy from the sun drives Earth’s surface processes, however there
are many process intrinsic to the atmosphere, oceans and geology
that determine short and long term climatic changes.
• The rotation of Earth around its axis varies slightly with the redistri-
bution of mass on its surface. Changes in of ocean currents, melting
of glaciers and tectonic action (earthquakes) cause a small drifts in
the length of day. To prevent the solar day from going out of sync
with civilian time, irregularities in Earth’s rotation are corrected by
adding leap seconds.

• The length-of-day is slowly increasing because of the transfer of

Earth’s rotational momentum to the moon’s orbital momentum (mainly
trough tides). This makes Earth’s rotation slower while the Earth-
Moon distance becomes larger.
 topics in environmental physics: the solar system and earth 33

2.2 Seasonality and Orbital Parameters

If you could go high above the North Pole you would observe that
the earth is orbiting the sun in a counter clockwise (CCW) direction
(Figure 2.1) and that the orbit is not circular, but slightly elongated.
You would further notice that the earth is also spinning in a CCW
fashion around its rotation axis. This axis connects north and south pole
and is perpendicular to the equatorial plane that separates the northern
and southern hemispheres (abbreviated as NH and SH, respectively).
It would also be clear that Earth’s rotation axis is tilted with respect
to the plane of the orbit. The elliptical orbit combined with the axis
tilt brings about seasons on Earth. The following sections will describe
the most important concepts in more detail.

Figure 2.1: Schematic overview of

Earth’s orbit around the sun and its ro-
tation around its axis.
Sun

Ecliptic plane

Earth orbit Eq
ua
tor

ne
la
lp
ia
or
t
ua
eq

rotation axis

2.2.1 Ecliptic Plane

The ecliptic plane is the plane in which Earth orbits around the sun
(Figure 2.4). All other planets circumscribe the sun in planes that make
a small angle (inclination) with the ecliptic plane (see the “Inclination”
column in Table 1.2). The ecliptic plane is therefore a convenient, but
geocentric reference plane. As seen from Earth, the sun would appear
to travel past the stars that are fixed on an infinitely far celestial sphere.
From the perspective of the sun, Earth would appear to travel past the
same stars half a year later.
Because the orbital planes of the other planets have only small an-
gles with the ecliptic plane (see again Table 1.2) they travel roughly
along similar paths on the celestial sphere. Their speeds are different
and consistent with their orbital periods thus causing an ever-changing
configuration of planets along the “ecliptic”. Traditionally, star group-
ings along the ecliptic were divided into 12 constellations (signs) of the
zodiac which, however, no longer align with the 13 constellations that
are presently recognized. The term “ecliptic plane” finds its origin in
34 marcel g. schaap

the fact that solar or lunar eclipses occur when the moon crosses the
ecliptic while in line with Earth and Sun.

Figure 2.2: Ellipse with relations among

semi-major (a) and semi minor (b) axes

semi minor
and eccentricity (e = 0.5), r: the distance
between Earth and the sun, A: Aphelion,

axis (b)
Earth P: perihelion. The red arrows indicate
relative velocities.

Sun r
(focus)
center θ
A P
semi major axis (a)
ae

a(1+e) a(1-e)

2.2.2 Eccentricity
Eccentricity is an index of non-circularity of an elliptical orbit, which is
characterized by a major (longest) axis (a) and a perpendicular minor Video about elliptical orbits. https:
(shortest) axis (b, see also Figure 2.2). Eccentricity, e, is defined as: //d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
r 02_Solar_System/M2S8_hd_720.mp4
b2
e= 1− (2.1)
a2
A zero eccentricity indicates a perfectly circular orbit while larger val-
ues cause an increased “stretching” of the ellipse. Note in Figure 2.2
that the sun is not in the center of the ellipse (the intersection of the
major and minor axes), but located in the focus which is offset from
the center by a factor ae. There is a second focus on the other side of
the center that is empty. Eccentricity therefore determines how much
the focus is displaced from the center. Table 1.2 provides the eccentric-
ity of the planets orbits as well as their characteristic “distance” from
the sun, which is the semi-major axis (i.e., one-half of the major axis).
The orbit’s elliptical shape and the off-center position of the sun
cause the Sun-Earth distance to vary throughout the year according to:

a 1 − e2


r= (2.2)
1 + e cos θ
 topics in environmental physics: the solar system and earth 35

Where θ is the angle of Sun-Earth (focus) with the major axis (Figure
2.2). The closest point is reached around January 3rd and is called the
perihelion (P in Figure 2.2). Around July 4th the Sun-Earth distance is
largest and is called the aphelion (A in Figure 2.2).
The fact that Earth has an elliptical orbit around the sun means that
the Sun-Earth distance varies throughout the year, according to Equa-
tion 2.2. On January 3, when θ equals 0° and Earth is at perihelion, the
Sun-Earth (center-center) distance is therefore 147.1 million km, while
on July 4’th the distance is 152.1 million km (when Earth is at aphelion
and θ equals 180°, Figure 2.2).

2.2.3 Kepler’s Laws and Newtonian Mechanics

The fact that the orbits of planets are elliptical was discovered in 1619
by Kepler who re-analyzed data collected by Tycho Brahe during the
16’th century and are refinement of a heliocentric model with circular
orbits proposed by Copernicus in the 15’th century. Kepler’s three
laws are:

• Planets orbit the sun in elliptical orbits.

• In equal periods of time, planets carve out equal area slices from
the elliptical “pie” (Fig. 2.3).

• The square of the orbital period is proportional to the cube of the

semi-major axis.

Kepler’s laws are actually a consequence of Newtons law of gravity

which was proposed later in 1687:

Gm1 m2
Fg = (2.3)
r2
Where Fg is the mutual gravitational force (in Newtons) of two bodies
with mass m1 and m2 separated by distance r. The force of gravity
becomes the centripetal force that holds body 2 (say, Earth) in orbit
around body 1 (the sun). The centripetal force depends on the square
velocity v and distance r:

mv2
Fc = (2.4)
r
When we set Fg = Fc , we obtain:

m2 v2 Gm1 m2
= (2.5)
r r2
and thus:
Gm1
v2 = (2.6)
r
36 marcel g. schaap

This simply states that the square of the orbital speed of body 2 de-
pends on the mass of the body 1 and the distance between the centers
of mass of both bodies. This is only true when m1 ≫ m2 . When this
is not the case m1 and m2 orbit a common center of mass known as
the barycenter. The Earth-Moon barycenter is inside the Earth, 4,671
km from its center. The Pluto-Charon barycenter is outside Pluto and
about 1000 above Pluto’s surface.
Equation 2.6 tells us that the square of orbital speed increases pro-
portionally to the mass of the parent body and decreases inversely
proportionally with their distance. Planets further from the sun (or
satellites further from Earth) move with slower speeds.
The orbital period of a body is the time it takes to complete a full
orbit. Assuming a circular orbit with a circumference 2πr, the period
P is:

2πr
P= (2.7)
v
Kepler’s third law can be found by squaring Eq. 2.7 and combining
this with Eq. 2.6:

4π 2 r3
P2 = (2.8)
Gm1
Although this equation was derived for a circular orbit, it remains
valid when the orbit is elliptical, in which case the semi-major axis
(a) should be used instead of r. This results in P2 ∝ a3 as stated in
4π 2
Kepler’s third law with a constant of proportionality equal to Gm . A
1
more practical version of this equation is:
s
a3
P = 2π (2.9)
Gm1
This implies that the orbital period only depends on the semi-major
axis and not on the eccentricity (e) or semi-minor axis (b). This conclu-
sion is important later when we discuss changes in Earth’s orbit.
Kepler’s first law (elliptical orbits) can be derived by allowing ki- 2

netic energy ( 12 m2 v2 ) to be exchanged for gravitational potential en- A

3
1
P
ergy. Kepler’s second law comes down to realizing that Eq. 2.6 im- 4

plies that the planet will move faster during perihelion and slower
during aphelion. When a planet orbits the sun in an elliptic orbit, nei-
ther the distance nor its speed are constant. However, the area of the
“pie-slices” carved out in equal time are constant (see Figure 2.3).
The difference in velocity near perihelion and aphelion causes the Figure 2.3: Kepler’s second law ex-
plained in graphical form. Because the
time between the fall solstice and spring solstice to be shorter than the distance to the sun and the planet’s or-
time between the spring solstice and fall solstice (by about two days). bital velocity changes, the colored “pie”
section near the perihelion (P) has the
This is the main reason why February has 28 (or 29) days while July
same area as the narrow colored area
and August have 31 days. near the aphelion (A). The time it takes
to go from 1 to 2 and from 3 to 4 is the
same. The red arrows indicate relative
velocities.
 topics in environmental physics: the solar system and earth 37

2.2.4 Axis Tilt

Axis tilt is the angle between the earth’s axis and an imaginary line
that is perpendicular to the ecliptic plane (Figure 2.4). The current
axis tilt angle is 23.44 degrees. The rotation axis points to two fixed
points in space (the celestial north or south poles) around which all
celestial bodies appear to orbit (See Figure 2.5). Polaris, the “pole star”
is currently situated about 45 arc minutes from the celestial north pole
and has historically been crucial for nautical navigation. There are no
bright stars near the celestial south pole and finding this exact point is
rather cumbersome.
Earth’s rotation axis is perpendicular to the equatorial plane that
separates the earth into a northern and a southern hemisphere. The
equatorial plane therefore makes a 23.44 degree angle with the ecliptic
plane (see Figures 2.1 and 2.4).

NH summer NH spring
equinox NH winter Figure 2.4: Solstices, aphelion and peri-
solstice (June 21)
solstice (Dec 21)
helion dates in Earth’s orbit.
Perihelion (Jan 3)

Aphelion (July 4)
l plane
Equatoria

23.44 °

NH fall
Earth axis tilt
Ecliptic plane equinox

It may surprise the reader that Earth’s rotation period is not ex-
actly 24 hours, but listed in Table 1.2 as 0.9927 days instead (23 hours,
56 minutes and 4.1 seconds). This is called sidereal rotation, which is
the time needed for an exact 360 degree rotation. This rotation time
is determined by measuring the time it takes between two meridian
crossings (i.e., stars being exactly south).
The sidereal rotation of Earth is subject to slight quasi-random fluc-
tuations due to changes on Earth (melting glaciers, ocean and wind
currents, tectonics) and because of the occurrence of tides in Earth’s
oceans (Appendix D). A smaller effect is caused by tides in the oceans
which gradually slow down the rotation of our planet, but increase the
semi-major axis of the moons orbit around Earth.
An Earth day (or solar or synodic day ) is 24 hours (on average) be-
cause for most daily life we use the sun as a reference for our time
(not the stars). The nearly four minute difference is due to Earth’s or-
bit around the sun: while a full (360 degree) rotation it takes 23 hours,
Figure 2.5: Star tracks around the north
56 minutes and 4.1 seconds, it takes an extra 4 minutes if we take the celestial pole in a long-exposure photo-
sun as our reference star. This apparent discrepancy is caused by the graph. Credit: LCGS Russ.
fact that during a day Earth also moves in its orbit (Top left diagram
in Figure 2.6), which means that it must rotate about 361 degrees for
a solar day, rather than the 360 degrees necessary for a sidereal day.
38 marcel g. schaap

There are some further complications caused by Earth’s elliptical orbit

and axis tilt which cause length of the solar day to change throughout
the year.
The images at the bottom (from the SOHO spacecraft) of Figure 2.6
illustrate the motion of the sun with respect to the distant stars. On
July 16 the sun lines up with Pollux (the star in the red circle), but has
moved east of Pollux on July 17. The relative daily motion is about one Video about Earth’s rotation
around its axis, and a discussion
degree (approximately two solar diameters), which can be verified in of sidereal and synodic days.https:
Figure 2.6. The nearly four minute difference between the sidereal day //d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
and the solar day is also the reason why you see a slightly different
02_Solar_System/M2S9_hd_720.mp4
part of the celestial sphere each night.

Figure 2.6: Sidereal year, synodic and

A
sidereal day. Top right: illustration of the
Pollux Pollux sidereal year, which is the return of the
(Gemini) sun to the same background stars. The
ellipse was rotated to avoid confusion
Apsidal Precession:
111,597 years
with the anomalistic year which longer
than the sidereal year because of apsi-
Sun dal precession. Top left: with respect
23h, to the stars (e.g. Pollux in Gemini, red),
56m,
Sun Earth rotates in 23 hours, 56 minutes and
24h 4.1s
4.1 seconds (a sidereal day). A synodic
(solar day) day is 24 hours on average.
Bottom left and middle images: it takes
more than 365 days for the sun to re-
P
turn to the same background star. Bot-
Sidereal Year:
tom middle and right: illustration of a
365.25636 days Day 0 Day 1 sidereal year day (red dashed) and syn-
odic day (white line). The synodic day is
longer than the sidereal day, because due
its orbit around the sun, the earth must
rotate more than 360 degrees. Bottom
image credit: ESA/NASA SOHO collab-
E W E W E W oration.
July 17 12:00 UTC

July 17 11:56 UTC

July 16 05:51 UTC July 16 12:00 UTC

2017 2018

1 Synodic day (24 h, or 86,400 s)

1 Sidereal year: 365.25636 synodal

days (365 d, 6 h, 9m, 9.5 s) 1 Sidereal day (23h 56m 4.1s)

2.3 Seasons

The axis tilt and the elliptical orbit explain the occurrence of seasons on
Earth. Both hemispheres always have opposing seasons (if it is winter
in one, it is summer in the other), but there are also strong latitudinal
differences in the intensity of the seasons as will be discussed next.
 topics in environmental physics: the solar system and earth 39

2.3.1 Eccentricity and Global Insolation

The varying Sun-Earth distance means that during one complete orbit
the global solar insolation on Earth will change: at perihelion in Jan-
uary the global solar insolation is about 7% stronger than at aphelion
in July.1 This is visible in the left diagram of Figure 2.10: the red line 1
you can verify this by taking the ratio
of the squared longest and shortest dis-
which displays the December insolation versus latitude does not ex-
tances as discussed in Section 2.2.2.
actly mirror the blue line which shows the June insolation. The far
southern latitudes receive more radiation in December (SH summer,
which occurs near perihelion) than do the far northern latitudes in
June (NH summer, near aphelion).

2.3.2 Declination
Because Earth’s axis is tilted by 23.44° the sun will make an angle (or
declination) with the equatorial plane that varies between 23.44° and
-23.44° as the earth goes around its orbit (Figures 2.4, and 2.6). There
are four distinct dates in the year:

• The northern solstice occurs around June 21 when the solar decli- Video about solar declination with
Stellarium demonstration. https:
nation reaches 23.44°. At this date the sun is above the latitude of //d2l.arizona.edu/content/enforced/
Cancer (the tropic of Cancer) and it is summer in the northern hemi- 950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S11_hd_720.mp4
sphere, but winter in the southern hemisphere. Days are longer
than 12 hours in the NH but shorter than 12 hours in the SH.

• The southern solstice around December 21 when the solar declina-

tion is -23.44° and the sun is situated above the latitude of Capricorn
(the tropic of Capricorn) and it is summer in the southern hemi-
sphere but winter in the northern hemisphere. Days are shorter
than 12 hours in the NH but longer than 12 hours on the SH.

• The two equinoxes (around March 21 and September 21) occur

when the solar declination is exactly zero and the sun is situated
above the equator. On these dates nights are exactly 12 hours long
everywhere on Earth (also see Figure 2.8).

The tropics of Cancer (Capricorn) are named such because the north-
ern (southern) solstice occurred when the sun was in the constella-
tion of Cancer (Capricorn). However, that was 2500 years ago, due to
axis precession (discussed later) the northern solstice occurs in Gem-
ini, while the southern solstice occurs in Sagittarius. Note that the
northern (southern) solstice is close in time to -but does not coincide
with- the aphelion (perihelion) of Earth. This distinction is important
when studying orbital changes and ice ages.
The angle of the sun with the equatorial plane varies between -23.44
and 23.44° irrespective of the location on Earth. The effect of the solar
40 marcel g. schaap

declination is visible in the second and third diagrams of Figure 2.7

which show part of the celestial hemisphere in equatorial coordinates.
In January, the solar declination is 23.44 degrees (second diagram),
while it is -23.44 degrees in December (third diagram).
A convenient approximation for this solar declination (δ in degrees)
is:

t + 10
δ (t) = −23.44 cos(360 ) (2.10)
365
where t is the time in (possibly fractional) days since midnight on Jan-
uary 1 (=0.0); t+10 merely approximates the number of days since the
NH winter solstice, which occurs around December 21. This equa-
tion assumes that Earth’s orbit is perfectly circular and that one orbit
is exactly 365 days. As long as it is not used for applications that
need high-precision values (e.g. navigation) this equation is practical
because Earth’s orbital eccentricity is small.
Solar declination is a global value, but because we live on a sphere
we are commonly interested in the angle that the sun makes with the
local horizon. This is called the elevation angle and is useful because
it provides us with the angle by which the sun’s rays will strike the
ground. To further pinpoint the position in the local sky we need az-
imuth which provides the angle of the sun with the geographic north.
For any given NH latitude north of the tropic of Cancer the daily
noontime solar elevation angle is given by:

ψ(t) = 90 − λ + δ (t) (2.11)

where λ is the latitude and all angles are in degrees. For latitudes
south of the tropic of Capricorn the noon-time elevation angle with
the northern horizon is:

ψ(t) = 90 + λ − δ (t) (2.12)

where λ is a negative value. For latitudes between the tropics of Can-
cer and Capricorn the minimum of Equations 2.11 and 2.12 must be
used to prevent elevation angles greater than 90 degrees. The bottom
diagrams in Figure 2.7 show that noontime elevation angle changes
between the solstices: a minimum is reached in December, while a
maximum is reached in June.
Equations 2.11 and 2.12 provide the noon-time elevation angle, when
the sun is exactly south (or north). For other times of the day, eleva-
tion angles change from zero degrees at sunrise to the maximum at
noon, and back again to zero a sunset. Because of their complexity,
we will not provide an exact mathematical specification how elevation
angle and azimuth vary throughout the day for any given latitude (for
details see Stull 2016).
 topics in environmental physics: the solar system and earth 41

June 21
Figure 2.7: Simulated view of the noon-
time sky in Tucson, Arizona (+32 de-
grees NH latitude) for June 21 and De-
cember 21 (NH summer and winter sol-
stice). The top diagram shows the blue
sky that prevents use from seeing the
stars. The second and third diagrams
show the sky with atmospheric light
scattering taken out, showing the back-
ground stars with an equatorial grid.
Solar declination is +23 and -23 de-
grees in June and December 21, respec-
June 21 tively. The yellow lines indicate the
path taken by the sun on those days,
+40
showing that sunrise/sunset is not in
the east/west. The bottom two dia-
Betelgeuse grams show the same dates and times,
+30 but now with an azimuthal grid, indi-
cating that the elevation angle changes
during the day, reaching a maximum
+20 at noon. These images were generated
+10 0 -10 -20 -30 -40 by the open-source package Stellarium
(http://stellarium.org/).

December 21
+40

+30

+20
+10 0 -10 -20 -30 -40

+30 +40 +50 +60 +70 +80 June 21

Z

+20

+10

+30 +40 +50 +60 +70 +80 December 21

Z

+20

+10

0
42 marcel g. schaap

The interplay between declination and latitude mean that, except

for the equinoxes, the sun does not rise exactly east or set exactly
west. When the declination is positive the sun rises north of east (and
sets north of west) by an amount that varies by latitude. The opposite
happens when the declination is negative: the sun rises (sets) south
of east (west). This also implies that the length-of-day changes with
declination and latitude, as shown in Figure 2.8. All these factors affect
the total amount of energy received from the sun for a given latitude
and a day of year.

Figure 2.8: Number of daylight hours as

J F M A M J J A S O N D a function of day of year (doy) and lati-
tude.
24
perihelion

aphhelion
solstice

solstice
Lat.
90
Day length (hours)

18 60
30
0
12 −30
−60
−90
6

0
0 50 100 150 200 250 300 350
Day of Year (doy)

2.3.3 Annual Variation in Insolation by Latitude

As discussed there are three factors that control the seasons on Earth:
the axis tilt, latitude, and eccentricity of Earth’s orbit. The solar dec-
lination in combination with latitude has a much stronger local effect
on the insolation than eccentricity of Earth’s orbit, which determines
variability of the total energy input to the planet.
It probably does not need much explanation that at noontime the
Video about orbit, axis tilt,
sun is always near the zenith at low latitudes (i.e. tropical latitudes be- latitude and insolation https:
tween the tropics of Cancer and Capricorn) and always rather low in //d2l.arizona.edu/content/enforced/
the sky at far northern or southern latitudes. The amount of solar en- 950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S10_hd_720.mp4
ergy that reaches the surface of the earth is proportional to the sine of
the elevation angle. When the sun’s elevation angle is low, the sunlight
is spread out over a larger area, as indicated by the yellow bars striking
Earth in Figure 2.9. This diagram ignores significant and variable ef-
fects in Earth’s atmosphere, such as reflection (clouds), scattering and
 topics in environmental physics: the solar system and earth 43

NH Summer NH Winter Figure 2.9: Seasonality due to axis tilt

Axis and orbital eccentricity (not to scale).
AU is astronomical unit (about 150 mil-
lion km), the sun in the center is only
Ca Ca
nce
r
nce
r partially drawn because of its size.
Ecliptic

Ca Ca
Night
pri
cor Sun pri
cor
n n
Day Day Night

~1.017 AU ~0.983 AU
SH Winter not to scale SH Summer

absorption of light.
At the equator, the variation is relatively small as shown in the right-
hand panel of Figure 2.10. For example, at the equinoxes when the
declination is 0° the noon-time elevation angle is 90° and the incom-
ing energy is S (the solar constant). During the northern solstice the
amount of energy received is equal to sin(90 − 23.44) × S = 0.92 × S,
thus only 8% less. However, at 60 degree latitude the incoming energy
at noon is sin(90 − 60 + 23.44) × S = 0.80 × S at the summer solstice,
but only 0.11 × S during the southern solstice (an 85% variation, see
Eq. 2.11). In addition, there are larger variations in day length the
farther north or south you go from the equator (see Figure 2.8).
Latitudes outside the tropics do not only see less insolation because
the incoming sunlight is spread out over a larger area, but also experi-
ence a larger annual variability. Seasonality therefore tends to increase
at more extreme northern and southern latitudes. The situation is, of
course, most extreme at the poles: here we essentially have days and
nights that last up to half a year (Figure 2.8).
A surprising fact is that during the summer solstice, the total daily Video about solar power dis-
tribution on Earth and climate
incoming energy is greatest of all latitudes (Figure 2.10). The 24-hour differences between northern and
day length at the pole (2.8) compensates for the rather low elevation southern hemispheres. https:
angle. Conversely, near the winter solstice no solar energy is available //d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
whatsoever, and temperatures at the surface and in the atmosphere 02_Solar_System/M2S12_hd_720.mp4
can reach very low values, especially at higher altitudes near the south
pole. These conclusions are not just trivia as they are highly relevant
in explaining the melting of arctic ice, or the dynamics of antarctic
stratospheric ozone.
While insolation dynamics are driven by latitude, Earth’s axial tilt
and orbital eccentricity, these are not the only factors that determine
seasonality at any given terrestrial location: altitude, proximity to large
water bodies, and land cover also play a significant role in reducing or
enhancing the effect of isolation-driven seasonality. For example, one
would expect that the southern hemisphere experiences more season-
44 marcel g. schaap

By Latitude By Month Figure 2.10: Insolation, latitude and

600 600 time of year. Top: daily average mid-
December (red) mid-March (black) and
500 500
mid-June insolation versus latitude. Bot-
Mean insolation (W/m2)

Mean insolation (W/m2)

400 400 tom: daily average insolation versus
time for 90 (purple), 60 (blue), 30 (green)
300 300 and 0 black) degrees north. Data credit:
Berger and Loutre (1991).
200 200
90 N
Dec. 60 N
100 Mar. 100 30 N
June 0
0 0
−90 −60 −30 0 30 60 90 J F M A M J J A S O N D
Latitude (degrees) Month

ality because the southern solstice roughly aligns with the perihelion
(occurring on December 21st and Jan 4’th, respectively); the northern
solstice aligns with the aphelion. However, measurements indicate
that temperature fluctuations are less extreme than that of the north-
ern hemisphere (bottom panel in Figure 2.11). The explanation for this
contradiction is threefold:

• Due to Kepler’s second law Earth actually moves faster in its orbit
near its perihelion and slower near its aphelion causing the period
with low global insolation (during NH summer and SH winter) to
last longer than the period with high global insolation (NH winter
and SH summer).

• The SH is significantly different from the NH in that it has a much

larger area of oceans. Water has a much higher specific heat ca-
pacity than land which mutes temperature fluctuation on the SH.
Solar radiation striking land therefore leads to larger variation in
temperature as shown in Figure 2.11.

• The albedo of snow and ice on Antarctica is rather high, which al-
lows the SH reflect solar radiation more efficiently and which limits
absorption of solar radiation during the SH summer (compare the
polar regions in the top and middle panels of Figure 2.11 and see
also equation 1.6).

2.4 Orbital Changes

 topics in environmental physics: the solar system and earth 45

Figure 2.11: Measured reflected radi-

ation in W/m2 during the NH winter
solstice (top left, December 22, 2004), the
NH summer solstice (bottom left June
20, 2004) and the annual temperature
swing. Credit: left-hand images adapted
from http://earthobservatory.
Dec. 22, 2004
nasa.gov/IOTD/view.php?id=6125;
right-hand image: UCAR.

June 20, 2004

0 250 500 750

What we have seen so far is that the sun is crucial in keeping the
earth warm. Although our simplistic radiation balance model (Eq.
1.6) is not very good, it provides a conceptual understanding that the
Earth is kept warm by solar insolation. We can also make it plausible
that variation in insolation is caused by Earth’s axis tilt, latitude and
orbital eccentricity as the main drivers of seasons.
On a human time scale it is possible to assume that eccentricity Introductory video about relation
between orbit, axis and climate. https:
and axial tilt are constant. However, these parameters vary consider- //d2l.arizona.edu/content/enforced/
ably over periods of thousands of years, along with the celestial di- 950156-867-2204-1ENVS420520001101/
rection with of Earth’s rotation axis as well as orientation of the orbit 02_Solar_System/M2S13_hd_720.mp4

itself. Variation in these features over time will cause significant sea-
sonal variation in insolation and is one reason for the occurrence of ice
ages in the past few millions of years. These orbital changes are com-
monly known as Milankovitch cycles, after the Serbian scientist who, in

ow )
(n ky
the early 20th century, identified the complex gravitational interaction
)
67 27
01 (+

among Earth and other celestial bodies that result in slow changes
0. 27

)
20 3
ky
(- 50
00

7
0
0.

0.

in the orbit, axis direction and tilt. Modern computation techniques A e=0 P
have superseded Milankovitch’ original hand-calculations and in the
following we will use work by Berger and Loutre (1991) and Berger
(1992) to discuss four major orbital factors that control the intensity
and timing of insolation.

2.4.1 Eccentricity Changes Figure 2.12: Extremes in eccentricity

in the “recent” past (-209 ky BP) and
“near” future (27 ky AP) compared to
The massive planets Jupiter and Saturn cause changes in the eccentric-
the present. For display purposes, ec-
ity of Earth’s orbit around the sun. There are several superimposed centricity has been exaggerated by a fac-
effects that cause cycles with timescales of 413 ky (1 ky is one kilo tor of 10, leading to more squashing
along the minor axis (vertical) and an in-
year, or 1000 years) and 95-125 ky (see top panel in left-hand side of creased displacement of the focus (sun)
along the horizontal axis.
46 marcel g. schaap

Figure 2.21). Currently, Earth’s eccentricity is 0.0167, but it was much

larger (0.0503) around 207 ky before present (BP, relative to the year
1950; AP is after present). Two minimums were reached around 400 ky
before present (BP). The eccentricity is decreasing to another minimum
(0.0027) which will be reached in 27 ky (not visible in Figure 2.21).
The significance of eccentricity changes is that summer solstice in-
solation can be enhanced if these are synchronized with the perihelion. Video about eccentric-
ity changes. https://d2l.
Conversely, there will be a lower insolation at the summer solstice of arizona.edu/content/enforced/
the other hemisphere because it will occur at aphelion. Variation be- 950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S14_hd_720.mp4
tween insolation at perihelion and aphelion is currently about 7%, but
was 22% 207 ky BP. It will decrease to nearly 1% at 27 ky AP at which
point eccentricity will have almost no effect on seasons since the Sun-
Earth distance is nearly constant; seasons will still occur, however, due
to Earth’s axis tilt.
Figure 2.12 illustrates the relative effect of eccentricity with an exag-
geration factor of 10 (otherwise effects of eccentricity would be nearly
imperceptible). Even with this exaggeration the orbit change for 27 ky
AP (e = 0.0027) cannot be distinguished. The focal displacement (ae) Year Orbital Period
(day)
is rather striking for the large eccentricity (7.5 × 106 km) but relatively
Sidereal 365.25636
small for the minimum eccentricity at 27 ky AP (4 × 105 km). At that Tropical 365.24219
time the focal displacement is smaller than the radius of the sun. Anomalistic 365.25963

The duration of the year does not change when eccentricity varies. Table 2.1: Sidereal, Tropic and Anomal-
The orbital period for an elliptic orbit around the sun is defined by Eq. istic years in days. One day is defined as
2.9 (Kepler’s third law). Eccentricity does not appear in this equation 86,400 seconds.

and it also has no effect on the semi-major axis, as illustrated in Fig-

ure 2.12; only the minor axis b and the focal displacement ae change
when eccentricity changes. Equation 2.9 therefore provides Earth’s
sidereal year which is defined as an exact 360 degree orbit around the
sun. Equivalently, this is the alignment of the sun with the same back-
ground stars (see Table 2.1 and the left-hand side of Figure 2.6).

2.4.2 Apsidal precession

Earth’s elliptical orbit changes in another way: it rotates (or precesses)

in a counter clockwise (CCW) fashion around the sun. To be clear:
as Earth goes around its orbit, the orbit itself also rotates (or precesses)
around the sun with respect to distant stars. To say this in another way:
the line that connects perihelion and aphelion (the “apses”) rotates
around the sun, which means that the perihelion moves around the
sun and is therefore not constant with respect to the background stars.
Apsidal precession is caused by gravitational interaction of Earth with
Jupiter and Saturn.
The period of the apsidal precession is 111,597 sidereal years. Both
Video about apsidal pre-
the orbital motion and the orbital precession move in a CCW man- cession. https://d2l.
arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S15_hd_720.mp4
 topics in environmental physics: the solar system and earth 47

Figure 2.13: Orbit (or apsidal) preces-

sion. The earth orbits the sun in one
sidereal year (the time it takes to line
up the sun and the earth to the same
stars) in a CCW motion, seen from above
A
2000 Distant Earth’s northern Hemisphere. However,
Sun stars the orbit itself also precesses in a CCW
motion as shown by the position of the
,000
elliptical orbit in the year 2000, 12,000
12

0
,00
22 and 22,000.

ner (see Top right-hand diagram in Figure 2.6). Therefore, the time
between perihelion passages (the anomalistic year Table 2.1) is a lit-
tle longer than the sidereal year. In other words: Earth needs to play
“catch up” with the perihelion (because the motion is in the same di-
rection), which takes almost 5 minutes per year. Figure 2.13 shows the
relative orbital positions in the year 2000, 12000 and 22000, with Earth
positioned at the aphelion in 2000 but lagging behind in 12,000 and
22,000.
The significance of apsidal precession is that the perihelion moves
with respect to the sidereal year. The effect of apsidal precession is that
perihelion (or aphelion) would occur 5 minutes later each sidereal year.
This implies that the timing of the maximum in global insolation (at
perihelion) changes slowly through the years. See below for a stronger
and opposite effect due to precession of Earth’s rotation axis.

2.4.3 Axis Tilt Changes

Earth’s axis tilt (or obliquity) with respect to the ecliptic is currently
23.44 degrees (°), but changes between 22.1 and 24.5° with a period
of approximately 41 thousand years due to gravitational interaction
between Earth, the moon and other planets. Notice that the change
in obliquity is distinct from the annual rhythm in declination as ap-
proximated by Eq. 2.10 and which is caused by Earth’s orbit around
the sun. To complicate matters further: build-up and melting of polar
icecaps and rearrangements of continents as well as large earthquakes
also have a small effect on the axis tilt as well as the geographic location
of the rotation axis.
Compared to the other rocky planets, Earth’s axis tilt has been re-
markably stable due to the large angular momentum of the Moon. Video about axis tilt changes. https:
//d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S16_hd_720.mp4
48 marcel g. schaap

Even so, the 2 degree variation does have a significant effect on Earth’s
climate: a larger value means that the axis is tilted more towards or
away from the sun at the solstices. An increased tilt therefore means
more insolation extremes during the summer solstices. Conversely, a
22.1°
decreased tilt means that there is less variation in insolation between 24.5°
winter and summer solstices. Seasonality in solar energy input will
therefore be increased at high angles and reduced at low tilt angles.
As the tilt of rotation axis changes, so does the angle of the equatorial
plane with the ecliptic (since the axis is perpendicular to the equator).
As a result, the tropics of Cancer and Capricorn will shift in latitude
Eq
ua
according to the axis tilt, but the equator itself will not be displaced. tor

2.4.4 Precession of Earth’s Axis

If you could extend Earth’s rotation axis into space you would notice
that it points to a location close to the pole star, no matter where Earth Figure 2.14: The angle of the earth’s
axis changes between 22.1 and 24.5 de-
is in its orbit. Although Earth’s rotation axis appears to be stable with gree with respect to the ecliptic plane.
respect to the stellar background (Figure 2.5), it is actually gyrating
(precessing) around a second axis that is perpendicular to the ecliptic
(Figure 2.15). The direction of the rotation axis in space is gradually
changing. A full precession cycle takes 25,772 years and is due to
gravitational interaction with the sun and the moon and the slightly
non-spherical shape of Earth as well as tides in the oceans.
As we saw earlier, the solstices and equinoxes, and therefore the
seasons, depend on the direction of Earth’s axis with respect to the sun.
Earth’s axis precesses in a clockwise (CW) fashion as seen from above
the north pole and is opposite to the CCW orbital motion around the
sun that defines the sidereal year. This makes the time of occurrence of
the equinoxes and solstices earlier by about 20 minutes each (sidereal)
year. A return to the same equinox or solstice is called a tropical year.

Figure 2.15: Axis precession is oppo-

Constellation site to the CCW orbital motion around
of Gemini
the sun and the rotation of the earth
around its axis. A full cycle takes 25,772
6/21 6/20 6/19
2015 2087 2159
Earth orbi
t years. Shown here are the location of the
summer solstice for the year 2015, 2087
and 2159, relative to the constellation of
1° 1° Gemini.
Earth orbit

Rotation (1 day)
CW Axis precession
26,000 y

There are, therefore, three different ways for expressing a “year”

Video about axis precession. https:
//d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
02_Solar_System/M2S17_hd_720.mp4
 topics in environmental physics: the solar system and earth 49

(Table 2.6): the sidereal year, the anomalistic year, and the tropical
year. Of these, the sidereal year has no effect on the timing of Earth’s
seasons (it is simply a full, 360 degree orbit), but the other two do.
The anomalistic year relates to apsidal precession and therefore the
timing of the perihelion of Earth’s orbit. The tropical year depends
on axial precession. Because axis tilt has a much stronger effect on
Earth’s seasonality than when perihelion and aphelion occur, we use
the tropical year for Earth’s civilian calendar. If we used the sidereal
year or anomalistic year the solstices would slowly occur at a later or
earlier date. Because the tropical year is shorter than the sidereal year
the solstices and equinoxes move relative to the fixed stars and are the
main reason why the zodiac no longer lines up with the astrological
signs. Figure 2.15 illustrates the effect of axis precession at the aphelion
in the context of Earth’s orbit around the sun.

Video about the three ways

2.5 Changes in Solar Insolation as a Driver for Ice Ages to define a year. https:
//d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
The head of the reader is now surely spinning because of the complex- 02_Solar_System/M2S19_hd_720.mp4
ities of changes in eccentricity, apsidal precession, axial tilt and axial
precession. All four effects have their own characteristic timescales
(104 to 105 years) and influence the intensity and timing of solar inso-
lation at particular latitudes on Earth. The changes in orbit and axis
parameters can be calculated with precision to about 5 million years
into the past and future (Berger, 1992). Longer time scale changes
become unpredictable because minuscule uncertainties in present so-
lar system body interactions (masses, positions, velocities) magnify to
large errors and to overwhelm the accuracy of gravitational calcula-
tions far into the past or future.
One distinct feature on Earth has been the recurrence of ice ages
during the Quaternary (the last 2.5 million years) and it is likely that
the timing of the ice ages is influenced by the Milankovitch cycles
(Paillard, 2015). In this context, paleo-climatologists like to calculate
the July insolation at 65 degrees Northern latitude. Compared to the
Southern Hemisphere, the Northern hemisphere has more landmass
that can support ice sheets, such as the one still present on Greenland.
The general idea is that at a low NH summer insolation (e.g. aphelion
in NH summer at high eccentricity) there is not enough energy avail-
able to melt all the ice and snow that has built-up during the previous
winter. Some ice will therefore remain throughout the summer and
because it has a high albedo it will reflect solar radiation back into
space (rather than expose darker ground or water). A positive feedback
cycle is therefore started that will not only lead to a buildup of ice with
more reflection of solar radiation (thus colder temperatures), but also
to the vertical growth of ice sheets to heights where temperatures are
50 marcel g. schaap

colder and the ice is therefore less likely to melt.

Lower NH summer insolation at 65 degree north should therefore
lead to a build-up of land ice and start an ice age. Conversely, when the
orbital parameters conspire to produce high NH summer insolation
the melting “wins out” on build-up of ice and the ice-sheets should
diminish. This is the basic idea, although we will also see that there
are several complicating factors. Nevertheless, it is interesting to study
the drivers of historic “climate change” as a context in which we can
compare present-day climate change.

Orbit precession
Figure 2.16: Orbit and Axial precession
combined. The earth orbits CCW around
4000 5000 the sun in one sidereal year, while the
6000 earth orbit itself also precesses CCW in
3000 112 ky. Axis precession makes the NH
-14,000 winter solstice move (precess) CW along
-13,000 the orbit (26 ky) NH winter solstice hap-
2000 -12,000 pens somewhat earlier in each year, at a
-11,000 different distance from the sun. The ec-
-10,000 centricity of this diagram is exaggerated
-9000 for display purposes. The position of the
-8000 summer solstice relative to the perihelion is
ω measured by the angle ω for between the
1000 -7000
-6000 year -14,000 and + 6000 CE.
-5000
-4000
-3000
0
-2000
-1000

Axis precession

2.5.1 Building a Conceptual Model

It turns out that we can simplify our problem somewhat by combin-
Video about a conceptual
ing orbital eccentricity, orbital precession and axial precession. Orbital ice-age model driven by or-
and axial precession are referenced with respect to the background stars. bital and axis changes. https:
//d2l.arizona.edu/content/enforced/
As illustrated in Figures 2.13 and 2.15, both control where the perihe-
950156-867-2204-1ENVS420520001101/
lion (or aphelion) will occur relative to the equinoxes or the solstices. 02_Solar_System/M2S20_hd_720.mp4
However, with regard to NH summer insolation, we only care when
the perihelion occurs relative to the summer solstice. We do not care
where the solstice or the perihelion occurs with respect to the back-
ground stars. Since axial precession is a clockwise effect and orbital
precession a counter clockwise one, both conspire to shorten the 25,772
year cycle of axial precession to a combined effect with a period of
20,936 years. You can use the analogy of a race track: if you and a
friend start running in opposing directions you’ll meet sooner than
when walking in the same direction (at different speeds). You do not
 topics in environmental physics: the solar system and earth 51

change your time around the race track, but there is a difference with
relation to each other.
Figure 2.16 illustrates how the summer solstice precesses around the
ellipse. Because the perihelion is presently occurring around January
3, there is a large angle (ω) between the summer solstice and the per-
ihelion. NH insolation is more intense when ω = 0 (northern solstice
occurs at perihelion) and less intense when the northern solstice occurs
at aphelion (note this was the case seven centuries ago, Figure 2.16).
When we take the cosine of ω we get a variable that varies between -1
and 1 when the solstice occurs at aphelion or perihelion, respectively
(see Figure 2.21, second panel).
Eccentricity can now be used as a scale factor that modulates the per-
ihelion and NH solstice alignment as e cos ω. When e is low, the effect
of perihelion and NH solstice alignment is negligible (0). In this case
it doesn’t really matter where the NH summer solstice occurs in the
orbit, because the orbit is almost circular (see Figure 2.12). However,
when e is large, e cos ω reaches a maximum value when NH solstice
occurs around the perihelion and a minimum (negative) value when
the SH solstice occurs at the perihelion.
With the above we therefore managed to “collapse” three effects
(eccentricity, orbit and axis precession) into one compound variable:
e cos ω. This is shown in the third panel of Figure 2.21, which now has
the periodical patterns of eccentricity and precession interfering with
each other.
All that remains now is to incorporate the effects of axial tilt (also
known as obliquity and displayed in the fourth panel of Figure 2.21),
which can further enhance e cos ω when obliquity is high (more pro-
nounced differences between seasons) or mute its effect when obliq-
uity is low (less extreme seasons across the globe).
We do not discuss the complex calculation of July insolation at 65
N (see Berger and Loutre, 1991 and Berger, 1992), but it appears in
the fifth panel of Figure 2.21. July’s insolation during the past 500 ky
varied by about 12% around a mean value of 440 W/m2 . July and Jan-
uary insolation near the tropics varied somewhat less by 10% around
a mean value of 455 W/m2 (not visible in the figure). These numbers
are much smaller than the solar constant (1361 W/m2 ) because they
factor in the sun’s elevation angle (which varies throughout the day)
and also the fact that the sun doesn’t shine at night.
Also shown here in blue is the January insolation at 65 South, which
-as might be expected- has an opposite effect (or phase): when July in-
sulation at 65 N is at a maximum, January insolation at 65 S (summer)
is at a minimum. This poses a direct problem: the figure indicates
that Milankovitch stimulation of ice ages should alternate between the
hemispheres. According to these data, a cold period (glacial) in the
52 marcel g. schaap

SH should coincide with a warm period (interglacial) in the NH. Ob-

servations, however show that glacials in the NH and SH occurred at
the same time. This is a first indication that our conceptual model may
not be complete.

18 ky BP Present
Figure 2.17: Ice extent at 18 ky BP (left)
and in the present (right). Land ice is
white and sea ice is gray. Light green
colors denote coastal plains, while light
blue indicate shallow oceans. The sea
level is significantly lower during ice
ages because part of the ocean water
is “stored” as land ice. Adapted from:
http://earthobservatory.nasa.gov/
Features/BorealMigration/boreal_
migration2.php

Glacial ice
Sea ice

2.5.2 Data About Past Climate

The variation in insolation at 65 NH can now be seen as representative
of the external stimulus to the earth system, particularly its climate.
We know that during the Quaternary (the geological time period that
covers the last 2.6 Million years) the earth system alternated between
glacials and interglacials. During glacials, enormous bodies of ice ex-
isted in large areas in North America, northern Europe and Siberia,
as well as most mountainous regions in temperate zones in the north-
ern and southern hemispheres (see Figure 2.17). Interglacial periods
are comparatively short warm periods, such as the present Holocene
(which started about 12 ky BP) and the previous warm period the
Eemian (from 130 to 115 ky BP).
Ice-core records from the Antarctic and Greenland are able to paint Video about data that are avail-
the history of Earth’s climate in high detail. The Antarctic and Green- able to quantify climate over the
past 800 thousand years. https:
land ice sheets have been so stable that snow that fell more than
//d2l.arizona.edu/content/enforced/
800,000 (Antarctica) or 100,000 (Greenland) years ago is still preserved 950156-867-2204-1ENVS420520001101/
as ice. Along with the compaction of snow into ice, atmospheric gases, 02_Solar_System/M2S21_hd_720.mp4

elemental isotopes, and dust was stored in annual layers (Figure 2.19).
From these cores it is possible to reconstruct the dramatic shifts be-
tween the warm and cold periods. For example, the temperature dif-
ference between the deepest lows (e.g. about 20 ky BP) and the highest
maximums is more than 10 centigrade (see black curve in the bottom
left panel in Figure 2.21). We refer the reader to NICL (2017) for an ex-
cellent overview of ice-core sampling, storage and the procedure used
 topics in environmental physics: the solar system and earth 53

1.0
to get data from the collected ice.
Temperature swings on land and in the atmosphere were responsi- 0.8

CO2 Solubility (g/m3)

ble for the buildup of icecaps, but changing temperatures in the oceans
had a dramatic effect on the CO2 content of the atmosphere because 0.6

cold ocean water can dissolve more CO2 , as shown in Figure 2.18 for an
0.4
atmospheric CO2 concentration of 280 ppm (parts per million). When
conditions cool, CO2 will be transferred from the atmosphere to the 0.2

ocean, leading to lower CO2 concentrations in the atmosphere and

0.0
lower CO2 concentrations in the gas bubbles captured in the ice cores 0 10 20 30 40
(the red line in bottom left panel in Figure 2.21). An additional effect Water Temperature (°C)
of transfer of CO2 from the atmosphere to the oceans is that the green-
Figure 2.18: Solubility of CO2 in seawa-
house effect is decreased, leading to further global cooling. A little ter for an atmospheric concentration of
cooling due to orbitally induced changes in insolation may therefore 280 ppm.
lead to positive feedback that further enhances cooling.
The exchange of CO2 between ocean water and atmosphere is the
explanation that the glaciations are synchronized for both hemispheres.
The dissolution of CO2 into the ocean water is a global effect (because
mixing of air between the two hemispheres has a time-scale of about a
year. If one hemisphere dissolves more CO2 than the other releases, the
greenhouse effect for both hemispheres will decrease and vice versa. 19 cm
Colder ocean water also leads to isotopic fractionation of water that Figure 2.19: A section of an ice core ob-
evaporates from the ocean surface. Oxygen in ocean water is 99.76% tained at 1855 meters depth from the
16 O and 0.20 % 18 O (the 0.04% remainder is the rare 17 O). Water Greenland ice sheet. The arrows indi-
cate brighter summer layers that are sep-
molecules containing 18 O have more mass and are consequently harder arated by darker winter layers. Credit:
to evaporate, especially when the water is cold. In colder situations rel- GISP2 (http://www.gisp2.sr.unh.edu/)
and NOAA.
atively more H2 16 O will evaporate from the oceans and be locked up
as land ice, thus leading to a relative enrichment of H2 18 O in the ocean
water.
Foraminifera are small (about 1 mm) ocean organisms which build
calcium carbonate (CaCO3 ) shells during their life cycle (Figure 2.20).
When foraminifera (“forams”) die, they sink to the ocean bottom thus
sequestering the CaCO3 in ocean sediments which also preserves the
isotope composition of the ocean water because:

2CO2 + 2H2 O → 2HCO3− + 2H + (2.13)

and

Ca2+ + 2HCO3− → CaCO3 ↓ +CO2 + H2 O (2.14)

One of the oxygen atoms in CaCO3 is therefore derived from the ocean
water.
When it is colder foraminifera will therefore build an increased
amount of 18 O into their calcareous skeletons, allowing the 18 O en-
richment history to become part of the geological record. Careful ex-
54 marcel g. schaap

amination of ocean cores reveal the 18 O history beyond the Quaternary

and can therefore be used to get information about ocean temperature.
At the same time, increasing 18 O concentrations can be used to make
estimates of the amount of land ice on Earth because 18 O enrichment
implies that 16 O is stored as land ice. The blue line in Figure 2.21
displays the last 500 ky of 18 O record (Lisiecki and Raymo, 2005).
A somewhat similar argument can be made for deuterium (D or2 H)
which makes up 0.015 % of the hydrogen in ocean water. Similar to Figure 2.20: Examples of foraminifera.
18 O, water molecules with deuterium will evaporate at lower rates in Credit: U.S. Geological Survey Depart-
ment of the Interior/USGS U.S. Geologi-
colder situations. In addition, altitude and temperature-dependent at- cal Survey/photo by Randolph Femmer.
mospheric processes will further fractionate deuterium, resulting in
temperature and altitude-dependent deuterium levels once the water
molecules fall as snow on ice sheets (Petit et al., 1999). By measuring
deuterium levels in ice-cores it is therefore possible to determine the
temperature and with careful examination and sampling it is possible
to reconstruct a high-resolution temperature record that is referenced
to present day values. This is shown by the black line in the bottom
panel of Figure 2.21, which also highlights the remarkable synchronic-
ity with the CO2 and 18 O record.

2.5.3 Comparing Model and Data

The data in the bottom panel 2.21 are invaluable in comparing the
present-day climate with more typical values in the late Quaternary.
There is something odd, however, when you compare the pattern of
insolation with the actual response in temperature, CO2 and d18 O: the
pattern of insolation at 65 NH has a dominant period of about 20 ky
(there are about 4 to 5 full sine waves per 100 ky). Earth’s temperature,
CO2 and 18 O responses, however, appear to have dominant period of
about 100 ky. It therefore appears that -even though there is stimulus
in insolation with a period 20 ky- Earth’s climate, ice sheets and ocean
systems respond with a time constant that is about 5 times slower.
To investigate the differences in orbital stimuli and the earth system Video about a comparison be-
response, it is interesting to look at the data patterns in an alternative tween cycles in model predictions
and those in observed data. https:
way. The panels on the left-hand side of Figure 2.21 all display time
//d2l.arizona.edu/content/enforced/
series, which means they give the pattern of a variable or phenomenon 950156-867-2204-1ENVS420520001101/
over time. Such patterns can also be thought to consist of periodic 02_Solar_System/M2S22_hd_720.mp4

patterns of (co)sine waves of different frequencies. An analysis like

this is called Fourier spectral analysis, which essentially decomposes a
time-based signal into its constituent frequencies -much like speech or
music is build up out of signals of different frequencies. For example,
a sinusoidal signal like cos ω (second panel on the left-hand side) can
be thought of as an almost pure single frequency tone that has a period
(recurrence of the cosine) of about 20 ky.
 −500 −400 −300 −200 −100 0 10 20 40 100 200 400
0.06
0.04 ecc. (e)
0.02
0.00 1
0 cos ω
0.06 −1
0.00 e cos ω
−0.06 25 obliquity
24 (degrees)
23
525 22
(W/m2)

65N insolation

topics in environmental physics: the solar system and earth

475
425
65S
375 400
4
350 Rel. T
−10 * d18O

2 CO2
300 d18O
0
250
−2
200

CO2 (ppmv)
Rel. T (oC)

−4 150
−6 100
−8 50

−10 0
−500 −400 −300 −200 −100 0 10 20 40 100 200 400
time before present (1950, in ky) period (ky)

Figure 2.21: Orbital drivers of insolation and earth system response. The left panel shows (from top to bottom) the last 500 thousand years of: eccentricity, cos ω, e cos ω,
obliquity (axis tilt), insolation at 65 N and 65 S (black and blue); and combined in the bottom diagram: relative temperature (black), CO2 concentration (red), and d18 O (blue).
The right-hand panel shows the spectra of the items in the left-hand panel, normalized to the strength of the strongest peak. Vertical gray lines in these diagrams correspond
to periods of 23, 29, 37 and 103 thousand years. The right-hand panels break down the time-based signals of the left-hand panels into their constituent periodic cycles. Peaks
thus show dominant periods, valleys absence of cycles. The right-hand essentially contain the same information as the left-hand panels, except that they did not preserve the
relative phase of the cycles. One striking effect is the absence of the 103 ky peak in the insolation spectrum which appears as a dominant peak in temperature, CO2 and d18 O.
See text for further explanation. Data credits d18 O: Lisiecki and Raymo (2005), CO2 : Lüthi et al. (2008), deuterium: Jouzel et al. (2007), orbital parameters: Berger and Loutre
(1991) and Berger (1992)

55
56 marcel g. schaap

The right-hand set of panels in Figure 2.21 display the spectra of

the time-series signals on the left-hand side; for simplicity, all spectra
were normalized such that their largest peak has a value of 1. In the
case of cos ω, there is a large peak at 23 ky, a smaller peak at 19 ky and
an even tinier peak around 30 ky. The two smaller peaks indicate that
the period in ω has varied somewhat from the present value of 21 ky.
This is also reflected in the stretched and compressed sinusoids in the
left-hand panel for cos ω.
Eccentricity shows a different spectrum, with a strong peak at about Video discussion about why northern
103 ky, with an even more dominant feature at longer periods. This and southern hemisphere glacia-
tions are synchronized. https:
is consistent with the time series of eccentricity on the left-hand side
//d2l.arizona.edu/content/enforced/
where there appear to be at least two superimposed sinusoidal waves. 950156-867-2204-1ENVS420520001101/
Previously, we discussed there were a 95-125 ky and a 413ky patterns 02_Solar_System/M2S23_hd_720.mp4

in eccentricity. The latter period cannot accurately be resolved in the

comparatively “short” time interval of 500 ky.
The term e cos ω should display the compound effect of eccentricity
and ω, but here eccentricity appears to be such a minor effect that it
hardly shows up. Similar to ω, obliquity shows a “clean” spectrum
with a period of about 41 ky, consistent with the almost “pure” sinu-
soidal change in the corresponding left-hand side panel.
The fifth panel on the right-hand side shows the spectrum of inso-
lation, with a dominant peak at 23 ky, 19 ky (both from ω) and 41 ky
(obliquity), but a conspicuous absence of a peak at 103 ky belonging to
eccentricity changes. It is especially this peak that should be present
in the insolation spectrum to explain the pattern in temperature, CO2
and d18 O which all show a pronounced peak at 103 ky and relatively
minor peaks at 23, 29, and 41 ky. While the 23 ky stimulus signal is
present in the earth system response, it is definitely not a strong signal.
By contrast while the 103 ky signal almost has no contribution in the
insolation, it now produces the major peak in temperature, CO2 and
d18 O. Other unusual things are the absence of a peak at 19 ky peak,
and a mysterious appearance of a 29 ky peak that is not present in the
insolation stimulus.
While insolation does have a significant effect Earth’s seasonality,
its longer term effect as a stimulus of ice ages and interglacials is -if
anything- rather complicated. It seems that a “shorter” (19-41 ky) term
variation in insolation stimuli is transformed into a 103 ky response.
This appears to indicate that variations in eccentricity really rule the
glacials and interglacials after all, even though this cycle hardly shows
up in the spectrum of insolation. It is perhaps too simplistic to sup-
pose that insolation alone can initiate and terminate glacial cycles and
that a combination of orbital forcing and geochemistry (Paillard, 2015)
and delayed response (Abe-Ouchi et al., 2013) in Earth’s systems are
responsible.
 topics in environmental physics: the solar system and earth 57

2.5.4 Interaction Between Ice Sheets and the Asthenosphere Litho

sphere
eno-
Asth
re
sphe
The discrepancy between the dynamics in insolation stimulus and the
Mantle
earth-system response (Figure 2.21) has been rather enigmatic (Imbrie
and Imbrie, 1986). Why do the ice ages not occur with a same time
Early Glacial
scale as their orbital stimulus (i.e. tens of thousands of years), but in-
stead in the order of 100 thousand years? Recent computer simulations
(Abe-Ouchi et al., 2013) have shed some more definitive light into the
interaction among the insolation stimulus, response by ocean (currents
and exchange of CO2 ), our climate and dynamics of glaciers. A key
aspect in this work is the role of isostasis of continents (Figure 2.22).
As sheets of land-ice build up to several km height, they become Full Glacial
massive enough to depress the solid continental crust downward. This
process is slow because it requires some the viscous material of Earth’s
asthenosphere to flow away from the region. The effects of isosta-
sis can be rather surprising: while the ice cap on Greenland reaches to
heights over 3000 meters, the solid crust that supports this cap is below
sea level (Figure 2.24). The sluggish response of isostasis is evident in
parts of Canada, and Scandinavia which are still rebounding upward Post Glacial
from the depression of long gone ice-sheets of 20 ky ago (Figure 2.23).
Much of Canada is rebounding up by as much as 10 mm/year, sur- Figure 2.22: Cartoon depiction of slow
rounding regions (including the U.S.) are actually sinking with rates isostatic response to glaciations and
deglaciations.
up to 2.0 mm/year. West Antarctica is also rebounding upward in ev-
idence that it contained a much larger ice mass during the last glacial
than in the present.

Figure 2.23: Isostatic uplift at 500 km res-

olution: credit NASA-JPL, Paulson et al.
(2007), and Geruo et al. (2013)

-2 1 4 7 10 mm/y
The slow response of especially the N. American crust is a major
factor in transforming the 19-41 ky cycles in insolation into the much
longer 100 ky cycle. It simply takes time for the crust to react to a
building ice-sheet and, similarly, it takes time for the crust to rebound
up once an ice-sheet has melted away.
58 marcel g. schaap

The timing of this vertical motion is important because atmospheric

temperature not only depends on orbital triggers of insolation but also
on altitude. Isostatic depression or rebounding may put a land or
ice surface above or below an altitude where the mean temperature
is 0 °C. If an area has a temperature below 0 °C, precipitation will
generally fall as snow and potentially accumulate. If an area is above
0 °C precipitation will fall as rain and runoff to the ocean.
This brings us to another factor: hysteresis, which can be described
as a persistence of a state even when the original stimulus is gone. If
a glacier is present it will grow taller and reach even colder altitudes.
This may allow ice sheets to persist - even during warm periods. How-
ever, if the lithosphere has been depressed (because of a long gone
glacier, e.g. the present situation in Canada), the land surface may
reside below 0 °C and not form an ice sheet, even when there is an
orbital stimulus.

Figure 2.24: Greenland bedrock altitude

(left) and ice thickness (right). Credit:
National Snow and Ice Data Center
(Morlighem et al., 2015).

Hysteresis can be illustrated with Greenland (Figure 2.24). Even

though we are presently in a warm period, the ice sheet on Greenland
is relatively stable (ignoring present anthropogenic climate change).
Because the ice sheet reaches altitudes well over 3000 m, most of the
precipitation falls as snow. This allows the Greenland ice sheet to
persist (right-hand diagram in Figure 2.24) even under current warm
conditions. However, suppose now that Greenland did not have an
icecap (left-hand diagram in Figure 2.24). Because of the isostatic de-
pression, most of Greenland’s central and western lithosphere is at or
 topics in environmental physics: the solar system and earth 59

below sea level. In such a case precipitation is more likely to fall as

rain and an ice cap would not build up. This might even be the case
if there was an orbital stimulus that would normally have initiated a
cold period with an ice cap. Only when Greenland’s lithosphere re-
bounds to sufficiently high altitudes (0 °C) an ice-cap would be able to
form. However, since this isostatic uplift may take tens of thousands
of years, orbital conditions may have changed once again.
The complex interplay of orbital stimulus, isostatic rebound, and
hysteresis of glacial conditions may therefore be responsible for the
slower than expected signals in global temperature (deuterium), ice-
volume (18 O) and atmospheric CO2 concentrations. This is still an ac-
tive area of research and not everything is fully understood, however.
For example, before 800 ky BP it appears that ice ages were much more
frequent and lined up with the 41 ky cycle of obliquity, rather than the
100 ky cycle of eccentricity (Lisiecki, 2010).
60 marcel g. schaap
3
Bibliography

Abe-Ouchi, A., Saito, F., Kawamura, K., Raymo, M. E., Okuno, J., Takahashi, K., Blatter, H., 2013. Insolation-driven 100,000-year glacial
cycles and hysteresis of ice-sheet volume. Nature 500 (7461), 190–193.
URL https://doi.org/10.1038/nature12374

Altwegg, K., Balsiger, H., Bar-Nun, A., Berthelier, J.-J., Bieler, A., Bochsler, P., Briois, C., Calmonte, U., Combi, M., De Keyser, J., et al.,
2015. 67p/churyumov-gerasimenko, a jupiter family comet with a high d/h ratio. Science 347 (6220), 1261952.

Berger, A., 1992. Orbital variations and insolation database. IGBP PAGES/World Data Center-A for Paleoclimatology Data Contribution
Series 92 (007).
URL https://www.ncdc.noaa.gov/paleo-search/study/5776

Berger, A., Loutre, M.-F., 1991. Insolation values for the climate of the last 10 million years. Quaternary Science Reviews 10 (4), 297–317.
URL https://doi.org/10.1016/0277-3791(91)90033-q

Budde, G., Burkhardt, C., Kleine, T., 2019. Molybdenum isotopic evidence for the late accretion of outer solar system material to earth.
Nature Astronomy.
URL https://doi.org/10.1038/s41550-019-0779-y

Chapront, J., Chapront-Touzé, M., Francou, G., 2002. A new determination of lunar orbital parameters, precession constant and tidal
acceleration from llr measurements. Astronomy & Astrophysics 387 (2), 700–709.
URL https://doi.org/10.1051/0004-6361:20020420

Chopin, C., 2003. Ultrahigh-pressure metamorphism: tracing continental crust into the mantle. Earth and Planetary Science Letters
212 (1-2), 1–14.
URL https://doi.org/10.1016/s0012-821x(03)00261-9

Cieza, L. A., Casassus, S., Tobin, J., Bos, S. P., Williams, J. P., Perez, S., Zhu, Z., Caceres, C., Canovas, H., Dunham, M. M., et al., 2016.
Imaging the water snow-line during a protostellar outburst. Nature 535 (7611), 258–261.
URL https://doi.org/10.1038/nature18612

EOC, 2016.
URL http://hpiers.obspm.fr/eop-pc/index.php

Fedorova, A. A., Montmessin, F., Korablev, O., Luginin, M., Trokhimovskiy, A., Belyaev, D. A., Ignatiev, N. I., Lefèvre, F., Alday, J., Irwin,
P. G. J., Olsen, K. S., Bertaux, J.-L., Millour, E., Määttänen, A., Shakun, A., Grigoriev, A. V., Patrakeev, A., Korsa, S., Kokonkov, N.,
Baggio, L., Forget, F., Wilson, C. F., Jan. 2020. Stormy water on mars: The distribution and saturation of atmospheric water during
the dusty season. Science, eaay9522.
URL http://science.sciencemag.org/content/early/2020/01/08/science.aay9522.abstract

Geruo, A., Wahr, J., Zhong, S., 2013. Computations of the viscoelastic response of a 3-d compressible earth to surface loading: an
application to glacial isostatic adjustment in antarctica and canada. Geophysical Journal International 192 (2), 557–572.

Gillon, M., Triaud, A. H. M. J., Demory, B.-O., Jehin, E., Agol, E., Deck, K. M., Lederer, S. M., de Wit, J., Burdanov, A., Ingalls, J. G.,
Bolmont, E., Leconte, J., Raymond, S. N., Selsis, F., Turbet, M., Barkaoui, K., Burgasser, A., Burleigh, M. R., Carey, S. J., Chaushev,
A., Copperwheat, C. M., Delrez, L., Fernandes, C. S., Holdsworth, D. L., Kotze, E. J., Van Grootel, V., Almleaky, Y., Benkhaldoun, Z.,
Magain, P., Queloz, D., Feb. 2017. Seven temperate terrestrial planets around the nearby ultracool dwarf star trappist-1. Nature 542,
456.
URL https://doi.org/10.1038/nature21360
62 marcel g. schaap

Goldschmidt, V. M., 1937. The principles of distribution of chemical elements in minerals and rocks. the seventh hugo müller lecture,
delivered before the chemical society on march 17th, 1937. Journal of the Chemical Society (Resumed), 655–673.
URL https://doi.org/10.1039/jr9370000655

Gomes, R., Levison, H. F., Tsiganis, K., Morbidelli, A., 2005. Origin of the cataclysmic late heavy bombardment period of the terrestrial
planets. Nature 435 (7041), 466–469.

Hallis, L. J., Huss, G. R., Nagashima, K., Taylor, G. J., Halldórsson, S. A., Hilton, D. R., Mottl, M. J., Meech, K. J., 2015. Evidence for
primordial water in earth’s deep mantle. Science 350 (6262), 795–797.
URL https://doi.org/10.1126/science.aac4834

Imbrie, J., Imbrie, K. P., 1986. Ice ages: solving the mystery. Harvard University Press.

Jacobson, S. A., Rubie, D. C., Hernlund, J., Morbidelli, A., Nakajima, M., 2017. Formation, stratification, and mixing of the cores of earth
and venus. Earth and Planetary Science Letters 474, 375–386.
URL https://doi.org/10.1016/j.epsl.2017.06.023

Jakosky, B. M., Edwards, C. S., 2018. Inventory of co2 available for terraforming mars. Nature Astronomy 2 (8), 634.
URL https://doi.org/10.1038/s41550-018-0529-6

Jakosky, B. M., Lin, R. P., Grebowsky, J. M., Luhmann, J. G., Mitchell, D. F., Beutelschies, G., Priser, T., Acuna, M., Andersson, L., Baird,
D., et al., 2015. The mars atmosphere and volatile evolution (maven) mission. Space Science Reviews 195 (1-4), 3–48.
URL https://doi.org/10.1007/s11214-015-0139-x

Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., Falourd, S., Hoffmann, G., Minster, B., Nouet, J., Barnola, J., Chappellaz,
J., et al., 2007. Epica dome c ice core 800kyr deuterium data and temperature estimates. IGBP PAGES/World Data Center for
Paleoclimatology data contribution series 91, 2007.

Krasinsky, G. A., Pitjeva, E. V., Vasilyev, M. V., Yagudina, E. I., 2002. Hidden mass in the asteroid belt. Icarus 158 (1), 98–105.
URL https://doi.org/10.1006/icar.2002.6837

Li, J.-Y., Le Corre, L., Schröder, S. E., Reddy, V., Denevi, B. W., Buratti, B. J., Mottola, S., Hoffmann, M., Gutierrez-Marques, P., Nathues,
A., et al., 2013. Global photometric properties of asteroid (4) vesta observed with dawn framing camera. Icarus 226 (2), 1252–1274.
URL https://doi.org/10.1016/j.icarus.2013.08.011

Li, J.-Y., Reddy, V., Nathues, A., Le Corre, L., Izawa, M. R. M., Cloutis, E. A., Sykes, M. V., Carsenty, U., Castillo-Rogez, J. C., Hoffmann,
M., et al., 2016a. Surface albedo and spectral variability of ceres. The Astrophysical Journal Letters 817 (2), L22.

Li, Y., Dasgupta, R., Tsuno, K., Monteleone, B., Shimizu, N., 9 2016b. Carbon and sulfur budget of the silicate earth explained by
accretion of differentiated planetary embryos. Nature Geosci advance online publication, letter.
URL http://dx.doi.org/10.1038/ngeo2801

Lin, Y., Tronche, E. J., Steenstra, E. S., van Westrenen, W., Nov. 2016. Evidence for an early wet moon from experimental crystallization
of the lunar magma ocean. Nature Geoscience 10, 14.
URL https://doi.org/10.1038/ngeo2845

Lisiecki, L. E., 2010. Links between eccentricity forcing and the 100,000-year glacial cycle. Nature geoscience 3 (5), 349.
URL https://doi.org/10.1038/ngeo828

Lisiecki, L. E., Raymo, M. E., 2005. A pliocene-pleistocene stack of 57 globally distributed benthic δ18o records. Paleoceanography 20 (1).
URL https://doi.org/10.1029/2004pa001071

Lüthi, D., Le Floch, M., Bereiter, B., Blunier, T., Barnola, J.-M., Siegenthaler, U., Raynaud, D., Jouzel, J., Fischer, H., Kawamura, K., et al.,
2008. High-resolution carbon dioxide concentration record 650,000–800,000 years before present. Nature 453 (7193), 379–382.
URL https://doi.org/10.1038/nature06949

Mann, A., 2018. Bashing holes in the tale of earth’s troubled youth. Nature 553 (7689).
URL https://doi.org/10.1038/d41586-018-01074-6

Meyers, S. R., Malinverno, A., 2018. Proterozoic milankovitch cycles and the history of the solar system. Proceedings of the National
Academy of Sciences.
URL http://www.pnas.org/content/early/2018/05/30/1717689115

Morbidelli, A., Levison, H. F., Tsiganis, K., Gomes, R., 2005. Chaotic capture of jupiter’s trojan asteroids in the early solar system.
Nature 435 (7041), 462–465.
URL https://doi.org/10.1038/nature03540
 topics in environmental physics: the solar system and earth 63

Morlighem, M. E., Rignot, E., Mouginot, J., Seroussi, H., Larour, E., 2015. Icebridge bedmachine greenland.
URL https://doi.org/10.5067/AD7B0HQNSJ29

NASA, 2015a. Keplerian elements for approximate positions of the major planets.
URL http://ssd.jpl.nasa.gov/?planet_phys_par

NASA, 2015b. Planets and pluto: Physical characteristics.

URL http://ssd.jpl.nasa.gov/?planet_phys_par

NASA, 2016. Planetary fact sheet - metric.

URL http://nssdc.gsfc.nasa.gov/planetary/factsheet/index.html

NASA/Lunar, Institute, P., 2003. Gas giant interiors: 2003.

URL https://solarsystem.nasa.gov/resources/677/gas-giant-interiors-2003/

NICL, 2017. About ice cores.

URL http://www.icecores.org/icecores/index.shtml

Nimmo, F., 2002. Why does venus lack a magnetic field? Geology 30 (11), 987–990.
URL https://doi.org/10.1130/0091-7613(2002)030<0987:wdvlam>2.0.co;2

NIST, 2016.
URL http://physics.nist.gov/cuu/Units/second.html

Paillard, D., 2015. Quaternary glaciations: from observations to theories. Quaternary Science Reviews 107, 11–24.
URL https://doi.org/10.1016/j.quascirev.2014.10.002

Paulson, A., Zhong, S., Wahr, J., 2007. Inference of mantle viscosity from grace and relative sea level data. Geophysical Journal Interna-
tional 171 (2), 497–508.
URL https://doi.org/10.1111/j.1365-246x.2007.03556.x

Petit, J.-R., Jouzel, J., Raynaud, D., Barkov, N. I., Barnola, J.-M., Basile, I., Bender, M., Chappellaz, J., Davis, M., Delaygue, G., et al., 1999.
Climate and atmospheric history of the past 420,000 years from the vostok ice core, antarctica. Nature 399 (6735), 429–436.
URL https://doi.org/10.1038/20859

Rufu, R., Aharonson, O., Perets, H. B., 1 2017. A multiple-impact origin for the moon. Nature Geosci advance online publication, article.
URL http://dx.doi.org/10.1038/ngeo2866

Schorghofer, N., Mazarico, E., Platz, T., Preusker, F., Schröder, S. E., Raymond, C. A., Russell, C. T., 2016. The permanently shadowed
regions of dwarf planet ceres. Geophysical Research Letters 43 (13), 6783–6789.
URL https://doi.org/10.1002/2016gl069368

Sheppard, S. S., 2015. The jupiter satellite and moon page.

URL http://home.dtm.ciw.edu/users/sheppard/satellites/

Steinpilz, T., Joeris, K., Jungmann, F., Wolf, D., Brendel, L., Teiser, J., Shinbrot, T., Wurm, G., 2019. Electrical charging overcomes the
bouncing barrier in planet formation. Nature Physics.
URL https://doi.org/10.1038/s41567-019-0728-9

Stevenson, D. J., 1987. Origin of the moon-the collision hypothesis. Annual review of earth and planetary sciences 15, 271–315.
URL https://doi.org/10.1146/annurev.earth.15.1.271

Stull, R. B., 2016. Practical Meteorology: An Algebra-based Survey of Atmospheric Science.

URL https://www.eoas.ubc.ca/books/Practical_Meteorology/

Thomas, P. C., Binzel, R. P., Gaffey, M. J., Zellner, B. H., Storrs, A. D., Wells, E., 1997. Vesta: Spin pole, size, and shape from hst images.
Icarus 128 (1), 88–94.
URL https://doi.org/10.1006/icar.1997.5736

Touma, J., Wisdom, J., Feb. 1993. The chaotic obliquity of mars. Science 259 (5099), 1294.
URL http://science.sciencemag.org/content/259/5099/1294.abstract

Tsiganis, K., Gomes, R., Morbidelli, A., Levison, H. F., 2005. Origin of the orbital architecture of the giant planets of the solar system.
Nature 435 (7041), 459–461.
URL https://doi.org/10.1038/nature03539
64 marcel g. schaap

USGS, 2019. Mineral Commodity Summaries 2019.

URL https://doi.org/10.3133/70202434

Wade, J., Wood, B. J., 2016. The oxidation state and mass of the moon-forming impactor. Earth and Planetary Science Letters 442,
186–193.
URL https://doi.org/10.1016/j.epsl.2016.02.053

Way, M. J., Del Genio, A. D., Kiang, N. Y., Sohl, L. E., Grinspoon, D. H., Aleinov, I., Kelley, M., Clune, T., 2016. Was venus the first
habitable world of our solar system? Geophysical Research Letters 43 (16), 8376–8383, 2016GL069790.
URL http://dx.doi.org/10.1002/2016GL069790

Willacy, K., Alexander, C., Ali-Dib, M., Ceccarelli, C., Charnley, S. B., Doronin, M., Ellinger, Y., Gast, P., Gibb, E., Milam, S. N., et al.,
2015. The composition of the protosolar disk and the formation conditions for comets. Space Science Reviews 197 (1-4), 151–190.
URL https://doi.org/10.1007/978-94-024-1103-4_8

Williams, C. D., Mukhopadhyay, S., Rudolph, M. L., Romanowicz, B., Jul. 2019. Primitive helium is sourced from seismically slow
regions in the lowermost mantle. Geochem. Geophys. Geosyst. 0 (0).
URL https://doi.org/10.1029/2019GC008437
A
Critical Elements

The triangles at the top of Figure 1.10 and descriptive information in

Table A.1 indicate which elements are deemed critical for the U.S. na-
tional security and economy (USGS, 2019). These are mostly elements
which are of essence to national defense or without which the econ-
omy would be severely affected, but which must be imported because
hardly any production of processing facilities exist within the USA. Be-
cause of this the availability of these elements (or compounds of these
elements) may be affected by international trade agreements, national
disasters, politics, or conflicts.
For example, China and Russia are the dominant suppliers of at
least 19 and 4 of these elements, respectively. Persistent trade or polit-
ical disputes exist between the USA and these countries and the func-
tioning of the US economy would be hurt if it were unable to import
the elements.
The identification as critical element does not mean that the USA
has no geological reserves of a particular element. This is only the
case for manganese, niobium and tantalum, which exist as uneco-
nomic low-grade ores within the USA. Most of the 35 elements are
not currently mined in the U.S. for economic (not profitable), or en-
vironmental reasons (REE element mining is notoriously environmen-
tally burdensome). Activation of mining or processing facilities may
require some time (sometimes years).
Some elements or materials are deemed so critical that the U.S. gov-
ernment actually stockpiles the material or its compounds (as indi-
cated in A.1), such as beryllium and germanium (defense). Some non-
critical elements are also stockpiled, such as nickel, mercury, and zinc.
Even some relatively common materials such as natural quartz crys-
tals (SiO2 ) are stockpiled, primarily to seed the production of artificial
quartz crystals in the optics industry. Gold and silver are stockpiled
by US Treasury for stabilization of the value of the US dollar.
Unique in the list of Table A.1 are helium and zirconium because the
U.S. is a net exporter of these elements (indicated as “E” in Reliance
66 marcel g. schaap

column). Helium is the only element that can be used to attain tem-
perature conditions that approach zero kelvin and is therefore essen-
tial for super-conducting conditions necessary in many research and
medical applications (e.g. MRI imaging). The helium concentration
in the atmosphere, but some natural gas reservoirs have percentage
levels of He due to the radioactive decay of uranium or thorium in
sediments or crystalline base rocks. The US is the largest supplier of
He in the world with the largest known (or disclosed) reserves. Zirco-
nium finds widespread use as ceramic cladding of fuel rods in nuclear
reactors, and is therefore important in nuclear weapons production,
among many other uses.
 topics in environmental physics: the solar system and earth 67

Name Reliance S Purpose

Al Aluminum 50 used almost all sectors of the economy
Sb Antimony 85 batteries and flame retardants
As Arsenic 100 lumber preservatives, pesticides, and semi-conductors
Ba Barium 84 used (as barite) in cement and petroleum industries
Be Beryllium 22 Y used as an alloying agent in aerospace and defense
Bi Bismuth 96 medical and atomic research
Cs Cesium 100 research and development
Cr Chromium 71 Y primarily in stainless steel and other alloys
Co Cobalt 61 Y rechargeable batteries and superalloys
F Fluorine 100 manufacture of aluminum, gasoline, and uranium fuel
Ga Gallium 100 integrated circuits and optical devices like LEDs
Ge Germanium >50 Y fiber optics and night vision applications
C Graphite 100 natural graphite for lubricants, batteries, and fuel cells
Hf Hafnium ? nuclear control rods, alloys and high-temperature ceramics
He Helium E Y used for MRIs, lifting agent, and research
In Indium 100 mostly LCD screens
Li Lithium >50 Y primarily for batteries
Mn Manganese 100 Y steelmaking
Nb Niobium 100 Y steel alloys
Pt Platinum 33(Pa)- Y Platinum group metals (Ru, Rh, Pd, Os, Ir, Pt) are used for catalytic
elements 73(Pt), agents
K Potassium 92 primarily used as a fertilizer
REE Rare earth 100 Y primarily batteries and electronics
elements
Re Rhenium 84 lead-free gasoline and superalloys
Rb Rubidium 100 research and development in electronics
Sc Scandium 100 alloys and fuel cells
Sr Strontium 100 pyrotechnics and ceramic magnets
Ta Tantalum 100 Y electronic components, mostly capacitors
Tl Tellurium >75 steelmaking and solar cells
Sn Tin 78 Y used as protective coatings and alloys for steel
Ti Titanium 91 overwhelmingly used as a white pigment or metal alloys
W Tungsten >50 Y primarily used to make wear-resistant metals
U Uranium >95 mostly used for nuclear fuel
V Vanadium 100 primarily used for titanium alloys
Zr Zirconium E high-temperature ceramics industries

Table A.1: The 2019 list of 35 miner-

als deemed critical to U.S. national se-
curity and the economy, how reliant the
US on import, whether the element is
stockpiled (S) by the U.S. Government
and the general purpose of the element
(USGS, 2019). Although elements are
listed, some of these are used in com-
pound form with other elements. Note
that REE are rare earth elements, a group
of 17 elements. S: a U.S. of this material
exists. E: USA is a major exporter.
B
Albedo

Albedo is a term that should be used with care as different appli-

cations use different definitions (see also http://nssdc.gsfc.nasa.
gov/planetary/factsheet/fact_notes.html). There are two kinds of
albedo: Bond or planetary albedo and geometric albedo.
Bond or planetary albedo is the fraction of incident radiation that is re-
flected back into space without being absorbed by the planet’s surface
or atmosphere. This is the albedo used in energy balance calculations,
such as our planetary temperature model. The incident solar radiation
covers a broad spectrum of wavelengths (most of the sun’s energy is
emitted at wavelength between 100 nm and 3000 nm). In the case of
the Bond albedo we do not care about the angle of incidence, instead
Bond albedo quantifies the fraction of solar radiation that is reflected
by the entire planet.
Geometric albedo is used in imaging applications, in which it is im-
portant to know illumination and viewing angles of a surface. The
value of the geometric albedo with these angles and often also ex-
hibits a wavelength-dependence. Geometric albedo is illustrated in
Figure B.1, not only shows high-albedo clouds and lower albedo land
and -especially- ocean surfaces, but also a patch of reflected sunlight
moving across the equatorial region of the Pacific Ocean. The bright-
ness of this patch depends on the illumination angle (in this case the
time of day) and the viewing angle (as seen by two weather satellites
in different positions).
The left-hand and center images are from NOAA’s GOES 16, which
is stationed in a geostationary orbit at 75.2 degree western latitude
35,786 km directly above the Brazilian rain forest. The right-hand im-
age is from GOES 17 at 137.2 degrees west above the pacific ocean.
The left-hand image is from 20:00 UTC on September 29, 2019, while
the center and right-hand images were taken four hours later at 24:00
hours UTC. The displacement of the bright patch over 4 hours indi-
cates that not only the angle of incoming sunlight is important, but
also that the viewing angle plays a role. For example, reflected sun-
70 marcel g. schaap

light is barely visible near the western coast of Colombia, South Amer-
ica, but much brighter four hours later when it is further out west on
the Pacific Ocean. However, the view from GOES 17, which is above
the pacific is not nearly as bright, even though it was taken at the same
time as the center image. This implies that reflected sunlight is more
intense at lower incoming and viewing angles.

Goes 16, 16:00 Goes 16, 24:00 Goes 17, 24:00

Figure B.1: An illustration of geometric
albedo. In terms of Bond albedo, the
reflectivity of Earth’s oceans is low. In
these NOAA’s GOES 16 and 17 images
it is possible to discern a brighter ar-
eas that directly reflect sunlight, thus il-
lustrating that the intensity of reflected
light depends on the viewing angle. The
yellow bars indicate the x and y posi-
tions of the bright area in each image
The bottom image shows the positions
of the GOES 16 and 17 geostationary
weather satellites (at roughly 16:00 hours
UTC). Looking down from about 100,000
km above the North Pole, the visible
land surface is that of North-America.
C
A Demonstration of the Greenhouse Effect

The effect of light intensity, albedo, and the greenhouse effect dis-
cussed in Section ?? can be illustrated with a simple experiment as
shown in Figure C.1.

Video about the development of

C.1 Equipment the temperature model. https:
//d2l.arizona.edu/content/enforced/
950156-867-2204-1ENVS420520001101/
Four Styrofoam hemispheres were mounted on a Styrofoam board to
02_Solar_System/M2S7_hd_720.mp4
represent a low and high albedo planets, with and without greenhouse
effects. Two hemispheres were left white (high albedo) and two were
painted black; each hemisphere is 10 centimeters across. Styrofoam
was chosen because of its low heat capacity which allows rapid attain-
ment of thermal equilibrium. All four hemispheres were instrumented
with a thermistor affixed with white or black duct tape. A fifth ther-
mistor was mounted on the back of the Styrofoam panel served as a
temperature reference. The sensors were scanned every 10 seconds.
Continuous imagery of the setup was obtained with a visible light
(Raspberry Pi V2 camera module, 8 mega-pixel resolution, binned
to 640 × 480 pixels) camera and a FLIR thermal infrared (IR) cam-
era which is sensitive to thermal radiation with wavelengths between
8 and 11 microns; resolution of this camera was 80 × 60 pixels but
scaled to 640 × 480 in software. IR intensity obtained with the thermal
camera was colored between blue and green for the lowest intensity
(lowest temperatures), while higher emissions were colored as yellow,
red and white (highest temperatures).
A Raspberry Pi Model 2B credit card-sized computer running Linux
was used to obtain and process the thermistor and camera imagery.
The data was streamed through a local network to provide students
present in the classroom a live view of the experiment. This setup
also facilitated students who were enrolled in an on-line class section,
through recording of the classroom projector stream.
A 250 Watt halogen flood light with three settings (off, on-medium,
on-high) was used to illuminate the scene and act as the “sun”. Two 15
72 marcel g. schaap

cm diameter polycarbonate domes were used during the experiment

to simulate a greenhouse effect. Polycarbonate is transparent to visible
light, but absorptive to thermal IR radiation.

Figure C.1: Temperature model as a

laboratory experiment. Four Styrofoam
hemispheres were mounted on a Sty-
rofoam board. Two hemispheres were
a c left white (high albedo) and two were
painted black. All were instrumented
with a thermistor affixed with white or
black duct tape. A visible light cam-
era and a thermal infrared camera (8 to
11 micron wavelength) were used to ob-
serve the hemispheres, while the tem-
b d
perature was registered every 10 sec-
onds. See text for further explanation.

a b c d

C.2 Experiment

At point “a” (Figure C.1) the floodlight was still in its off setting and no
polycarbonate spheres are present. The thermistors indicated that the
temperature of the hemispheres was between 21 and 23 °C as shown in
the graph. The black and white hemispheres are clearly visible in am-
bient light, but infrared imagery is non-distinct because the Styrofoam
hemispheres and the background all have the same temperature. The
IR processing software “stretches” the color spectrum between blue
and white for minimal temperature fluctuations.
Next, the flood light is switched to full power, and temperatures
quickly rise with a thermal equilibrium is reached at “b”. The ther-
mistor readings show that the two thermistors on both high-albedo
(white) hemispheres measure a temperature of 27.5 °C; the low-albedo
(black) thermistors reach a higher temperature: 32.5 °C. This is con-
sistent with our planetary temperature model (Eq. 1.6) which predicts
that under the same intensity of light, low-albedo planets should reach
higher temperatures.
The infrared imagery show that the white hemispheres have the
lowest emission (blue), while both black hemispheres have a much
higher emissions (red) due to their higher temperature. Both black
 topics in environmental physics: the solar system and earth 73

hemispheres also have a “yellow ring” of medium emission (medium

temperatures) where the rays of the light strike the surface of the hemi-
spheres more obliquely. Here the intensity of the light is spread out,
leading to a lower temperature. We not that both pairs of sensors
(“black” and “white”) register similar temperatures.
After reaching this thermal equilibrium at “b”, the upper white and
black hemispheres were covered with polycarbonate domes. This ar-
rangement simulates a greenhouse effect. A new thermal equilibrium
is established at “c”, with the covered black hemisphere warmer by
about 3 °C than the uncovered black hemisphere, while the covered
white hemisphere is about 2 °C warmer than its counterpart. This is
the greenhouse effect, which is stronger for the lower albedo (black)
“planet”.
To illustrate the effect of light intensity on temperature (S(d) in Eq.
1.6), the intensity of the floodlight was lowered by half. Temperatures
indeed dropped and reached a new equilibrium at “d”. The temper-
atures drop somewhat slower for the greenhouse planets than for the
“bare” planets. The decline in temperature is somewhat stronger for
the greenhouse planets.
Finally, the flood light was turned off completely, allowing the hemi-
spheres to cool down to the background temperature. We note that all
“hemisphere” sensors reached temperatures that were lower than the
reference sensor, which warmed up by about half a degree centigrade
because of heat diffusion through the one-inch Styrofoam panel.

C.3 Discussion

The experiment validates the effect of albedo and insolation on temper-

ature, but unlike the mathematical model we developed earlier, it was
also able to incorporate the consequences of a greenhouse effect and
its interaction with albedo and light intensity. Where the temperature
model of Eq. 1.6 assumes that the surface of the planet obtains one ho-
mogeneous temperature, the situation for the illuminated uncovered
hemispheres show that the temperature depends on the obliqueness
of the incoming rays of light.
We further point out the white and black hemispheres can be seen in
through the polycarbonate dome (points “c” and “d”) in visible light.
In this case we see light reflected from the surface of the Styrofoam
hemispheres. However, this is not the case in thermal infrared light, in
which case we see radiation emitted by the domes themselves (visible
as two bigger blue or blue-green hemispheres).
Even though visible light intensity did not change, the thermistors
under the domes clearly warm up relative to their uncovered counter-
parts (point “c”), implying that they emit more thermal infrared radi-
74 marcel g. schaap

ation. The thermal emission from the planet’s surface is intercepted

by the domes, which warm up and send infrared radiation back to the
planet (which therefore experience extra warming, i.e., the greenhouse
effect). Approximately half of the energy is radiated outward from the
dome into “space” and observed by the thermal camera.
The greenhouse effect recycles IR energy emitted by the surface, back
to the surface through absorption and re-emission (as IR). Ultimately,
however, a new thermal equilibrium is reached and the atmosphere
(dome) must radiate the same amount of energy in infrared as is ab-
sorbed by the surface in visible light.
By comparing predicted planetary surface temperatures with ob-
served temperatures (Section ??) we already identified that Equation
1.6 is unable to deal with a greenhouse effect. In one of the follow-
ing modules about radiation we will improve this model to include a
perfect and discuss several other complicating factors.

C.4 Simplified Terminology

C.4.1 Light Emission

Photons produced by a source. There are several kinds of sources.
Almost all objects exhibit black-body emission, the intensity and wave-
length of which is determined by temperature. The sun is a black-
body emitter (most energy is carried by photons with wavelengths in
visible light), but so are the Styrofoam hemispheres and the polycar-
bonate domes (these have lower temperatures and emit photons in the
infrared). Crucially, air is not a perfect black-body emitter which is im-
portant once we discuss global warming. Black-body emission emits
photons across a broad spectrum of wavelengths. This is contrasted
by other forms of line-emission (e.g. laser light), or emission by LEDs,
which emit in narrow bands of the spectrum.

C.4.2 Light Transmission

Photons traveling trough vacuum, or through a transparent material,
such as air or polycarbonate. Light transmission is wavelength-dependent.
The polycarbonate domes transmit visible light, but absorb (and emit)
thermal infrared photons.

C.4.3 Light Absorption

Photons converted into other forms of energy, such as heat (the black
hemispheres), electricity (solar panels), or chemical energy (photosyn-
thesis), among others. Light absorption can be wavelength-dependent
(color) and is related to albedo when considered over a broad spectrum
 topics in environmental physics: the solar system and earth 75

of wavelengths. Polycarbonate absorbs thermal infrared photons, but

not visible light photons.

C.4.4 Light Reflection

If not transmitted, or absorbed, photons can reflect of surfaces and
change direction. In the case of mirrors (specular reflection) the angle
of incidence is equal to the angle of reflection. For non-smooth surfaces
photons attain a more random direction (a very simplistic statement!).
A surface can both absorb and reflect light. In the case of the white
hemispheres most incident photons are reflected and few absorbed.
Conversely, the black hemispheres absorb most visible light photons
(and turn these into heat, which leads to emission of photons with
thermal IR wavelengths) and reflect few. Color: if an object is blue,
it is reflective to photons with blue wavelengths, but absorptive to
photons with other wavelengths.

C.4.5 In Summary
Most natural objects are black-body emitters and therefore produce
photons in a temperature-dependent range of wavelengths and inten-
sity. Depending on material properties and geometry, one or more
apply: photons are transmitted through the object, reflected off its sur-
faces, or absorbed.
D
Earth’s Rotation and Tides

D.1 Fluctuations in Earth’s Rotation

We have already seen that there is a difference in sidereal and synodic

day. The former refers the rotation of Earth to distant stars while the
latter references the rotation to the sun. Since the orbit around the
sun speeds up or slows down, depending on whether we approach
perihelion or aphelion, a synodic (or solar) day is not constant but
varies slightly during the year. A mean solar day, however, is expected
to be exactly 24 hours as otherwise civilian time (the time we are most
used to) would change as the years pass. Our civilian time is fixed as
24 hours of 3600 seconds each totaling exactly 86,400 seconds per day.
Suppose now that the mean solar time was not exactly 86,400 seconds
this would mean that for example, sunrise and sunset would slowly
change with respect to civilian time as the years progress.

0.004
Figure D.1: Earth’s rotation relative to
Day length anomaly (s)

86,400 seconds (positive means a slower

0.002
rotation). Data before 1962 (vertical blue
line) are based on model evaluations,
and smoothed measurements after. The
0.000 small vertical tick marks at the bottom
axis provide the insertions of leap sec-
onds. Data credit: EOC (2016).
−0.002

−0.004
1850 1900 1950 2000
Time (year)

It turns out that the earth does indeed not spin with a constant rate.
Figure D.1 provides the differences between a civilian day (86,400 s)
and the measured rotation period of Earth. It can be seen that the
earth can rotate slower (usually, positive values) or faster than 86,400
seconds (negative values). Before the reader despairs and asks “is
78 marcel g. schaap

nothing really constant?”: the second is standardized in the SI system

as: “The second is the duration of 9,192,631,770 periods of the radia-
tion corresponding to the transition between the two hyperfine levels
of the ground state of the cesium-133 atom.” (NIST, 2016)
Reasons for the variation in rotational speed are the movement of
mass on Earth in the lithosphere (e.g. earthquakes), changes in ocean
currents, melting of glaciers, and even the direction and speed of
winds on Earth. Conservation of angular momentum requires that the
rotation speeds up or down as mass is moved closer or further away
from Earth’s rotation axis, respectively. The proverbial example is the
spinning ice skater. As the ice skater pulls his or her arms inward the
spinning will increase while moving the arms or legs outward will re-
duce the rotation rate. This also occurs on Earth. For example, glaciers
are usually located near the poles, i.e., close to Earth’s rotation axis. If
the glaciers melt, the water will redistribute to oceans that are typically
further away the rotation axis, thus leading to a (small) reduction in
rotation speed and a (small) increase in day length.
Figure D.1 indicates that the rotation has varied between -0.004
(faster) and 0.004 (slower) seconds per day since the mid 1800s. The
effects are cumulative, meaning that after a year with a daily deviation
of 0.004 seconds, civilian time would deviate by 1.34 seconds from so-
lar time. To prevent a de-synchronization of civilian with solar time,
the Earth Orientation Center in Paris introduces leap seconds into the
civilian time, typically at midnight (UTC) on January 1st , but occa-
sionally also at midnight July 1st . The leap seconds do not occur every
year and Figure D.1 indicates that the density of ticks increases as the
deviation from 0 increases.

D.2 Interaction with the Moon

A much smaller effect is the “stealing” of Earth’s rotational angular

momentum by our moon which allows it to increase the semi-major
axis of its orbit around the earth. In effect, Earth’s rotation is slowed
and its day length increased by 15 microseconds per year (Chapront
et al., 2002). The moon’s orbital angular momentum is given by:

2πmmoon re2−m
Lorbit = (D.1)
Porbit
where mmoon is the mass of Moon, re−m is the radius of the moon’s or-
bit, Porbit is its (sidereal) rotation period. Earth’s rotational momentum
is given by:

2
4πmearth rearth
Learth = (D.2)
5Pearth
 topics in environmental physics: the solar system and earth 79

Where mearth is the mass of Earth, rearth is its radius and P is its (side-
real) rotation period (86,164 s). The moon is able to take Earth’s rota-
tional momentum through the tides it generates in the oceans. Because
the moon orbits slower (27.3 days) than the earth turns, the tides move
around Earth and cause a drag on Earth’s rotation. At the same time
the bulging high-tides gravitationally attract the Moon and acceler-
ate it forwards (see Figure D.2, this is also called “torque”). Because
the sum of rotational and orbital momentum must be conserved this
effectively means that a slowing down of Earth’s rotation (longer pe-
riod Pearth ), Earth’s rotational angular momentum is transferred to the
moon’s orbital angular momentum, which is only possible by increas-
ing the radius of its orbit, re−m .

Figure D.2: Transfer of angular momen-

Earth's rotational tum from Earth’s rotation to the moon’s
motion (1 day)
orbit. Notice that there are two tides on
Tide pulls moon Earth: one “between” the earth and the
forward and moon and one on the opposite side of
causes friction
Center
on ocean bottom
Earth. The former can be easily under-
of mass
stood because of the direct gravitational
attraction by the Moon. The antipo-
Tide with Moon's orbital dal tide exists because it is further away
dragging motion (27 days) from the Earth-Moon center of mass.

Tide without
dragging

There are two high tides on Earth. The one most easily understood
is the one that “faces” the moon because the gravitational attraction by
moon counteracts that of the earth. This will “pull” the water in the
direction of the moon. High tides are also found on the opposing side
of Earth because there the ocean water is a little further away from
the center of gravity of the combined Earth-Moon system. Gravitational
interaction of the ocean water with the sun has a similar but smaller
effect.
Laser measurements using reflectors left behind by NASA’s manned
lunar program (Figure D.3) have indeed shown that the Moon is reced-
ing from Earth by 38 mm per year (Chapront et al., 2002). The transfer
of rotational momentum means that the earth rotated faster in the geo-
Figure D.3: Laser reflector left behind
logical past. Recent work has shown that day was only 18.68 hours 1.4 on the moon by the Apollo astronauts.
billion years ago Meyers and Malinverno (2018). Other estimates point Credit: NASA.
to days of only 5 hours just after the collision with Theia (Stevenson,
1987).