Geoff Eddy’s Climate Cookbook

Geoff’s Climate Cookbook

Written by Geoff Eddy
All text in green added by The Astrographer

Last updated: 27 February 2006

With grateful thanks to krinnen, aka Gonzalo, who drew the pictures.

This should provide more reliable hosting for the page, but the formatting suffers a bit in adaptation to wordpress. I’ll do what I can, but I’m not really a wp guru…



This page is part of my essay about creating an
Earthlike planet
; it is intended to guide the creator of such a
planet, after he or she has drawn a Map, through the process of
working out the climates which characterise a particular area. As far
as learning about the physical causes of climates goes, there’s no
substitute for a good textbook; however, textbooks tend to work
backwards from observed phenomena to inducing the causes, whereas the
typical conworlder needs to know the causes before he or she can
deduce the observed phenomena – which is what this page is for.

Please note that predicting climates is notoriously complicated and
full of approximations, which is why there are no equations on this
page and very little quantification. Ideally, I would be able to offer
a program which would convert a Map of a planet and its physical data
– such as axial inclination and distance from the sun – into a diagram
showing the climate at every point of interest on the planet’s
surface; when I’ve written this program I will be able to retire for
good on the money. In the meantime, the best I can do is talk in
generalities without going into too much specific detail.

If you find this page useful, please let me know! As ever, I welcome
corrections and suggestions for improvements.

Basic principles

Virtually everything important about climates can be deduced from the
following physical principles, which are referred to in [square

  1. All heating comes from the sun.
  2. Water heats and cools much more slowly than land; water thus
    acts as a stabilising effect on temperature.
  3. Hot air rises, cold air sinks; this is because air
    expands as it heats up and thus becomes less dense.
  4. Cold air gives rise to areas of high pressure, and
    hot air gives rise to areas of low pressure.
  5. Wind flows from areas of high pressure to areas of low
  6. Due to the Coriolis effect – the effect of the rotation of the
    earth on the flow of air – winds are deflected to the right in
    the northern hemisphere, and to the left in the southern.
  7. Rising air is conducive to the fall of precipitation, sinking
    air is not.
  8. Warm air carries more moisture than cold air.


You will need the following items before you can proceed any

  1. The axial inclination of your planet, which is 23.5
    degrees for the Earth. The lines of latitude at this distance
    from the equator are known as the tropics, and those at
    the same distance from the poles are called the polar
  2. Two identical copies of your Map, which should show the
    locations of as much land as you know about, the
    locations of the mountains, the lines of
    latitude in increments of no greater than fifteen
    degrees, and the tropics and polar circles. Label one copy
    “January” and the other “July”.
  3. A transparent drawing medium which can be marked and erased
    without damaging the Map. In the physical world, this means
    several sheets of tracing paper or something made of clear
    plastic; on a computer, the equivalent is a drawing program
    which can handle layers, such as the GIMP.
  4. Something erasable with which to draw on the transparent
    medium; for tracing paper, coloured pencils (not pens)
    are suitable. You will need several colours.
  5. Something with which to erase the above, because you
    will make mistakes, and lots of them.

The following assumptions have been made:

  • Your planet rotates from west to east, like the Earth.
  • Your planet has a similar diameter and rotation period to the
    Earth. These quantities are respectively 12750 km and 24

For ease of reference, “January” and “July” refer respectively to the
periods shortly after the sun reaches its furthest south and north
respectively, and “April” and “October” to those just after it passes
directly above the equator northwards and southwards respectively. The
“just after” is necessary because the atmosphere acts as a drag on the
heating and cooling processes; thus the hottest time of the year in
the northern hemisphere is typically around mid-to-late July, some
weeks after the summer solstice on 21 June.

pencilThis image lets you know when you
should think about drawing something.


The first stage consists of locating the large-scale areas of high and
low pressure.

The default

The most important is the low-pressure belt called the
inter-tropical convergence zone, or ITCZ, about which the
temperature and pressure characteristics are theoretically
symmetrical; this zone is caused by the rising of hot tropical air
[3][4]. In April and October, the ITCZ lies more or less along the
equator. In the northern summer, it moves northwards, reaching its
farthest north in July; its most southerly position is attained in
January. The range of movement on Earth is about 5 degrees of latitude
over the oceans, and up to 40 degrees over land.

About one-third of the way from the ITCZ to the poles is the
high-pressure belt known as the subtropical high-pressure zone,
or STHZ, which is caused by air from the ITCZ cooling and sinking back
to the ground [3][4]. Between the STHZ and the poles is the polar
or PF, a band of low pressure where cold air from the poles
meets warm air from the STHZ. The interaction between these air masses
at the polar front is responsible for the rain-bearing low-pressure
areas familiar from weather forecasts.

If the surface of the planet was uniformly water, the distribution of
these pressure belts and the prevailing winds
would be as shown below, allowing for seasonal movements, which would
be slight.


Adding land

The presence of land has two effects on the pressure distribution,
both results of principles [2][3][4]: the pressure belts front bend
northwards over land in July and southwards in January, and they are
broken up by seasonal pressure-areas over the land. In general, the
larger the area of land, the more noticeable the effect.

In winter, the cooling of the land creates a high-pressure area over
the interior, which merges with the high pressure area around the
STHZ and leaves low-pressure systems over the oceans:


while in summer the land warms to create a low-pressure area, which
joins up with the ITCZ and the PF, leaving high-pressure areas over
the oceans:


In general, these pressure areas are located east of the longitudinal
(east-west) middle of the continent, and are more intense when the
surrounding land mass is larger. This is particularly noticeable with
Asia; if the Eurasian landmass was reversed laterally, the pressure
areas would be considerably less intense. Correspondingly, the
pressure gradient is greater on east coasts than on west coasts; the
precise difference depends on the shape of the continent.

Figures 7p-4 and 7p-5 on this
show how this works out for the Earth; the animation, one of
many from here,
is also here. Note
particularly the considerable northward movement of the ITCZ in July
over Africa and Asia, the continuous low-pressure zone over the
Antarctic Ocean where there is no land to disrupt the southern PF, and
the change in the air pressure over the interior of eastern Asia.

pencilYou need to draw similar diagrams
showing the pressure for January and July. Start by drawing with the
ITCZ, STHZ, and PF, then locate the continental pressure-areas, and
finally join them up as in the diagrams. Different colours for each
stage are a good idea.


Wind, in meteorological terms, is a flow of air from an area of
high pressure to an area of low pressure [5]; the strength (speed) of
the wind increases with the difference in pressure. Winds have two
important effects on climate: they transport moisture, and –
for our purposes – they are the principle cause of the ocean
. Winds pick up moisture as they blow over the oceans and
deposit it as rain or snow over land. Obviously, a wind can only carry
a finite amount of moisture, so it wil become dry after blowing across
a large area of land.


The winds we are interested in here are those which blow at the
surface. Because of the Coriolis effect [6], the winds do not blow
directly from high pressure to low pressure, but are deflected to
blow, in the northern hemisphere, clockwise around high-pressure areas
and anticlockwise around low-pressure areas. In the southern
hemisphere the deflection is in the opposite direction. This
deflection gives rise to the trade winds over the oceans; in
the northern hemisphere they are south-westerlies in mid-latitudes and
north-easterlies otherwise, and in the southern hemisphere
north-westerlies and south-easterlies respectively.

The monsoon

On the east and south-east coasts of sufficently large land masses,
pressure gradient will be sufficiently extreme that the resulting
winds will override the prevailing trade winds; they will blow
offshore into the ocean in winter, while the summer low-pressure area
will pull in moisture-laden air from the ocean. This important
seasonal reversal of the winds is, of course, the monsoon; it
is prototypically observable in south-east Asia. The two pictures
below show the general directions of the prevailing winds in winter
(above) and summer (below). Note particularly the monsoon effect on
the east coast.



A good question is: How large is “sufficiently large”? North America
has no monsoon as such, so somewhere between the size of it and of
Asia is probably as good an answer as any.

In winter, the continental high-pressure areas are responsible for
cold waves, which are flows of very cold air eastwards to the
offshore oceanic low. These cold winds pick up moisture as they pass
over the sea, which will be deposited as snow on any mountains they
encounter; western Japan is a terrestrial example.

Ocean currents

The formation and movement of the ocean currents is a complicated
subject, much of which is not of interest here; for our purposes we
are only concerned with currents on the surface of the oceans, which
are caused wholly or mainly by the winds. The Coriolis effect comes
into play again here, deflecting the currents from the path of the
wind; the deflection is greatest (up to 45 degrees) at high latitudes
and least (about 5 degrees) at the equator.

Ocean currents come in two flavours, depending on the direction in
which they flow: poleward currents, which carry water from hotter
areas to colder areas, are classified as warm, while
equatorward currents are similarly classified as cold. Note
that these are relative terms, thus a particular warm current flowing
to a cold region may actually be colder than a cold current which
flows to a warm region.

The oceanic high-pressure areas of the STHZ give rise in low latitudes
to warm currents along the east coasts of continents and cold currents
along the west coasts. The reverse distinction obtains in
mid-latitudes, because the wind blows around the oceanic low-pressure
areas in the opposite direction. The currents affecting the sample
continent shown above would thus be as follows, with warm currents
shown in red and cold currents in blue:


The Gulf Stream, which keeps western Europe much warmer in winter than
the north-eastern USA and south-eastern Canada, is a classic warm

pencilNow is a good time to add the
prevailing winds and ocean currents to your Maps for both January and
July. The currents are easy; don’t forget that the winds will blow
more or less in S-shaped double spirals.

… add water…

The annual distribution of the fall of precipitation in the form of
rain and snow is one of the factors which characterise a
particular climate. Rain and snow result from four processes:

  • Moist winds blowing onto land, as previously mentioned.
  • Orographic lifting of moisture-carrying winds as they
    blow over mountains and are forced to rise; the air cools as
    it rises, depositing its moisture on the windward side of
    the mountains [8].
  • Convection due to the heating of the air. Again, the air
    cools as it rises and loses its moisture [8]. (Although the
    ITCZ passes over the Sahara Desert, it does not cause rainfall
    because of the dryness of the air; there is very little
    moisture for the rising air to pick up.)
  • Frontal lifting along the polar front. Here the warm air
    from the STHZ is lifted up by the colder air from the
    poles, causing the low-pressure areas which weather forecasts
    warn about; further details are beyond the scope of this page.

An important detail about orographic lifting should be observed: after
the wind crosses the mountains it sinks, expands, and warms back up
again. These winds on the leeward sides of mountains (the
rain-shadows) are thus characteristically warm and dry, and are
known as chinook or Fˆhn, or colloquially as
“snow-eaters” after their ability to melt snow in otherwise cold

Finally, cold currents cool and stabilise the air, inhibiting the
formation of precipitation, while warm currents heat and destabilise
it, encouraging precipitation [2][7]. The relative amounts of
precipitation due to various factors are shown in the following

Factor High precipitation Low precipitation
Pressure ITCZ, on or near the equator STHZ
Mountains Windward sides Leeward sides, in rain-shadow
Prevailing winds Onshore Offshore or parallel
Coastal currents Warm Cold, especially in low latitudes
Location West coasts subject to the PF,and some way inland Interiors

pencilYou should now be able to work out,
for both January and July, the relative amounts of precipitation on
your Map.

… and place in the oven.

The annual variation in temperature is the other characteristic
feature of a climate. As a first approximation, the temperature is
highest at the equator and decreases steadily towards the poles [1],
subject to the following modifications.

Effect of the oceans

Variations in temperature are lowest along the coasts and highest in
areas remote from maritime influence [2]. The variation increases with
the distance from the oceans, and less so with distance from the west
coast; the eastern regions of continental interiors thus experience
the greatest variations in temperature. Incidentally, another
consequence of [2] is that the hottest and coldest times of the year
occur two to three weeks earlier in these regions than at the coasts.

Effect of moisture

Heat is more readily transmitted through clear skies than cloudy
skies; consequently, the less cloud an area receives, the greater will
be its temperature variation during a single day. The higher the
temperature, and the clearer the skies, the more moisture will be lost
during the day through evaporation, which is the opposite of
precipitation. The greatest amounts of evaporation are found in land
areas influenced by the STHZ, where the high-pressure belt is not
conducive to precipitation and thus cloud-formation [7]. These areas
are thus the hottest of all during the day, and cold at night.

pencilYou should now be able to work out,
for both January and July, the relative levels of temperature on your


On both of your Maps you should now have indications of the

  • The main pressure-belts (ITCZ, STHZ, and PF);
  • The oceanic and continental areas of high and low pressure;
  • The prevailing winds;
  • The main ocean currents;
  • Temperature and precipitation, on land at least.

The final stage consists of identifying the closest matching climate
from the table below; it uses a classification system similar to the
widely-used system developed by Wladimir Köppen.

Temperature Precipitation Location, for
Name Köppen Summer Winter Summer Winter latitude in degrees
Tropical rainforest Af Hot Hot Wet Wet 0-10
Tropical monsoon Am Hot Warm Very wet Short and dry 5-15; east and south-east coasts only
Savannah Aw Hot Warm Wet Long and dry 5-15
Hot desert BWh Very hot Warm Dry Dry 10-30, especially on west coasts with cold currents
Hot steppe BSh Hot Warm Low to dry Low to dry 10-35; typically next to deserts
Cold desert BWk Hot Cold Dry Dry Interiors, rain shadow
Cold steppe BSk Warm Cold Low to dry Low to dry Interiors, rain shadow
Maritime east coast Cfa Hot Warm to mild Wet Moderate 20-40; east coasts only
Maritime west coast Cfb, Cfc Warm to mild Cool to cold Wet Wet 40-60; west coasts only
Mediterranean Csa, Csb Hot Mild Dry Moderate 30-45, west coasts only
Temperate monsoon Cwa, Cwb Hot Mild to cold Wet Dry 20-40; east coasts only
Laurentian Dfa, Dfb Warm to mild Cold Moderate Low 40-60; not on west coasts
Subarctic Dfc, Dfd Mild to cold Very cold Moderate Very low 60-80; not on west coasts
Manchurian Dwa, Dwb Warm to mild Cold Moderate Dry 40-50; east coasts only
Subarctic east Dwc, Dwd Mild to cold Very cold Moderate Dry 45-70; east coasts only
Tundra ET Cold Very cold Low Dry 60-80
Icecap EF Very cold Very cold Low Dry 75+

The climates given in italics are those which, generally
speaking, are subject to the same influences throughout the year. The
other climates may be regarded as transitions between these; for
example, the mediterranean climate is a combination of hot desert in
the summer and maritime west coast in the winter.

Note the following:

  • Steppe and desert climates experience large
    diurnal variations in temperature, which means cold nights.
  • In the subarctic and tundra climates, winters are
    long, dark, and cold, and the other seasons are short.
  • Some sources mention Köppen climate types As and Ds, which are
    like Aw and Dw but with the dry season in summer rather than
    winter. I don’t know what causes these particular climates;
    they are very rare anyway and can probably be safely ignored.

The progression of climates

Moving from the equator to the poles, the climates appear in
the well-defined sequences described below. It is instructive to
compare these found on the Earth.

The climates appear on the west coast in the following

  • Tropical rainforest.
  • Savannah.
  • Hot steppe, with dry winters. The boundary between this
    and the savannah is the line where evaporation equals
  • Hot desert, due to the influence of the cold
    current, which is also responsible for coastal fog on the
    west coasts of desert climates.
  • Hot steppe again, this time with dry summers.
  • Mediterranean. The boundary between this and the steppe
    is, again, the line where evaporation equals
    precipitation. Coastal fog is often experienced in summer.
  • Maritime west coast, cooling steadily poleward. These
    climates are warmed by the ocean currents.
  • Tundra.
  • Icecap.

Continental interiors, and areas in the rain-shadows of
north-south mountain ranges, will experience dry versions of the
climates to the west. The equivalent order of climates would be:

  • Tropical rainforest or savannah.
  • Hot steppe, with dry winters.
  • Hot desert.
  • Hot steppe.
  • Cold desert in areas far from the west coast.
  • Cold steppe.
  • Laurentian in its colder incarnations. Round about here,
    the colder temperatures reduce evaporation to the point that it
    no longer exceeds precipitation.
  • Subarctic.
  • Tundra.
  • Icecap.

On the east coast, there are two cases to consider, depending
on whether the land mass is large enough to generate monsoons. East
coasts not subject to the monsoon will feature the following

  • Tropical rainforest, or savannah if the land is
    high enough, as in east Africa.
  • Maritime east coast. The difference between this and
    the preceding is largely one of winter temperatures.
  • Laurentian, becoming steadily colder polewards.
  • Subarctic.
  • Tundra.
  • Icecap.

East coasts of continents where there is a monsoon will feature the
following climates:

  • Tropical rainforest, which may be absent.
  • Tropical monsoon, prototypically.
  • Temperate monsoon. This is the same as tropical monsoon,
    but with colder winters; equivalently, it is equivalent to
    maritime east coast with dry winters.
  • Manchurian. Effectively a laurentian climate with dry
  • Subarctic east. Similarly, this is the subarctic climate
    with dry winters.
  • Tundra.
  • Icecap.


One of the reasons for being interested in climate is to discover the
types of vegetation which grow in a particular region. This section
describes, in general terms, the vegetation types asociated with the
climate types. More detail, with information about the fauna, can be
found with a Google for “biomes”; for example

to biomes
, Habitats
and biomes
, Blue Planet
, World
, and – the most detailed – Kˆppen biomes.

The vegetation of the icecap climate is the simplest to
describe: there is none at all, because the temperature is below
freezing for most or all of the year. Tundra climates similarly
discourage growth for most of the year, but some vegetation grows in
the short summer, typically small mosses, lichens, and alpine
plants. Equatorward, where the climate borders subarctic, stunted
trees may grow.

The characteristic vegetation of the subarctic, subarctic
, and manchurian climates is extensive coniferous
forest known as taÔga, typically made up of spruce, fir, scots
pine, and larch; larch is commonest in the coldest and driest
climates, and the deciduous birch, aspen, and alder are also found in
the lower altitudes. Despite the low amounts of precipitation, even
lower evaporation means that enough moisture is retained to allow the
growth of vegetation. Conifers have needle-like leaves to preserve
water and strong branches to endure the snow which lies on them for
much of the winter.

A mixture of coniferous forests and broadleaved forests characterises
the maritime and laurientian climates; the dominant type
of forest depends on the proportion of the year in which the
temperature is less than 5.5 degrees centigrade (this is 42 degrees
Fahrenheit, interestingly). The progression is from evergreen
broadleaved through deciduous broadleaved to coniferous as the winters
become colder; thus if the temperature is always above 5.5 degrees
(i.e. the proportion is zero), the forest wil be mainly or entirely
evergreen broadleafed. The dominant type of tree will be coniferous if
the proportion is greater than 50%, and deciduous broadleafed if it is
between 0% and 50%.

Mediterranean vegetation needs to guard against losing water in
the dry summers, and tends towards scrub made of small plants with
hard leaves, similar to the chaparral familar from many Western
movies. The trees are either coniferous or evergreens with small waxy
leaves and thick bark; evergreen oak, pine, cedar, and above all olive
are typical mediterranean trees.

Too little moisture is retained in the steppes to allow trees
to grow; the principal vegetation is thus extensive grassland,
including many cereals. Grassland is also characteristic of the
savannah, in which the vegetation dies back in the dry winter
but grows vigorously in the summer, reaching heights of up to six
feet. Trees in the savannah tend to be isolated and adapted to retain
water for the long dry season, such as the baobab. The vegetation of
the deserts is scanty, patchy, and specially adapted to the
conditions; plants tend to be fleshy and leafless, such as the

The characteristic vegetation of the tropical rainforest
climate is, of course, tropical rainforest: lush, abundant forests
with massive trees and an enormous variety of other plants which grow
all year round in the ever-present moisture, The large amounts of
precipitation leach nutrients from the soil, and as a result the trees
have shallow roots and large buttresses at the bases of their
trunks. Monsoon vegetation is intermediate between rainforest
and savannah: the forests are less dense, many varieties of tree
become deciduous to cope with the dry winters, roots are longer, and
the plant types are less diverse.

What if?

The principles described up to now should work well enough for an
Earthlike planet. This section is intended as a catch-all for
questions not otherwise answered.

… my planet rotates in the opposite direction?

Easy – just interchange “east” and “west”.

… my planet rotates very fast?

The three bands of prevailing winds in each hemisphere are due to the
speed of the planet’s rotation. Above a certain speed of rotation, for
which I am unable to provide figures, the three will become five (they
cannot become four), and in between the STHZ and PF there will appear
another belt each of of low pressure and high pressure. These will
still move north and south with the sun, and the principles can be
applied as before.

Bear in mind that above a certain speed of rotation the planet will
disintegrate; I have no idea what limit this fixes on the maximum
number of bands of prevailing winds. A faster rotation will also lead
to shorter days and nights, which will doubtless have other

… my planet has a small axial inclination?

The north-south movement of the pressure belts will be correspondingly
less, and smaller areas will be subject to the climates which undergo
seasonal changes; annual temperature ranges will also be less. The
tropical rainforest, maritime, hot desert, and icecap climates will be

… my planet has a large axial inclination?

The reverse of the preceding; season effects will be increased, and
the areas subject to the tropical rainforest, maritime, hot desert,
and icecap climates will be less. A large enough inclination – about
40 degrees – will eliminate these climates altogether.

Paper references

  • Meteorology and Climatology, fourth edition (revised),
    E.S. Gates; Harrap (London) 1972. I rescued this from a
    wheelie-bin on Raeburn Place one evening.
  • Earth Science, fourth edition; Tarbuck & Lutgens,
    Merrill (Columbus, OH) 1985.
  • The Climate Revealed, by the unenviably-named William
    J. Burroughs, Mitchell Beazely 1999.
  • The Edinburgh World Atlas, John Bartholomew & Son Ltd,
    1970. Has useful diagrams of the relevant climatic


5 Responses to Geoff Eddy’s Climate Cookbook

  1. Michael | says:

    Hi Astrograher! First of all thank you for putting the Climate Cookbok up on WordPress, so the resource is available in a more reliable way. I’m running a German website about Worldbuilding ( and would really like to translate the whole piece into German. Since you’ve apparently been in touch with Geoff Eddy (and I have no idea how to contact him), do you think you could relay this request? Or would you be able to give permission yourself, assuming you approve of the idea? Best, Michael

    • Astrographer says:

      He has used to be on the Zompist Bulletin Board( pretty reliably, but I haven’t been on much of late. I can’t for the life of me remember what handle he was using, and he did change it at least once. He seemed to like female-themed handles and he usually used really pretty women as his avatars. At least I thought they were really cute. Haha! Err… His handle was Bricka a long time ago. The permission he gave me was pretty open-ended, as I remember, but I hate to speak for others.

      My best suggestion would be to open a thread in the Ephemera section with a title addressed to Geoff Eddy or Bricka.

      By the way, I really like your avatar!

      • Michael | says:

        At least I now know where to look for him – thank you! And thanks for the compliment, it’s the logo I’ve designed for the website in an attempt to symbolize worldbuilding.

      • Astrographer says:

        Nicely done. Looking at your blog makes me wish my grasp of the German language extended beyond crude insults and pick-up lines.

        Good luck getting ahold of Mr. Eddy, mate!

      • Michael | says:

        I’ll try my best. If I ever translate my content to English, I’ll make sure to drop yuo a note!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s