BioSystems 125 (2014) 1–15

Contents lists available at ScienceDirect

BioSystems
journal homepage: www.elsevier.com/locate/biosystems

Global warming description using Daisyworld model with greenhouse

gases
Susana L.D. Paiva, Marcelo A. Savi *, Flavio M. Viola, Albino J.K. Leiroz
Universidade Federal do Rio de Janeiro, COPPE – Department of Mechanical Engineering, P.O. Box 68.503, 21.941.972 Rio de Janeiro, Brazil

A R T I C L E I N F O A B S T R A C T

Article history: Daisyworld is an archetypal model of the earth that is able to describe the global regulation that can
Received 23 January 2014 emerge from the interaction between life and environment. This article proposes a model based on the
Received in revised form 9 September 2014 original Daisyworld considering greenhouse gases emission and absorption, allowing the description of
Accepted 10 September 2014
the global warming phenomenon. Global and local analyses are discussed evaluating the influence of
Available online 16 September 2014
greenhouse gases in the planet dynamics. Numerical simulations are carried out showing the general
qualitative behavior of the Daisyworld for different scenarios that includes solar luminosity variations
Keywords:
and greenhouse gases effect. Nonlinear dynamics perspective is of concern discussing a way that helps
Daisyworld
Global warming
the comprehension of the global warming phenomenon.
Carbon cycle ã 2014 Elsevier Ireland Ltd. All rights reserved.
Ecology
Nonlinear dynamics

1. Introduction about the literature associated with the Daisyworld, emphasizing

Climate change is a growing global concern, and a topic of
considerable research. Global warming is within this general issue
being related to the increase of earth’s temperature (Houghton,
2005). While the consequences of global temperature changes
remain unknown, it is possible to use mathematical models to
evaluate different scenarios. There are several general modeling
approaches (Alexiadis, 2007). A first strategy makes use of time-
series analysis to build predictive models of future temperature
change (Viola et al., 2010). A second approach makes use of
mathematical models to describe system behavior using governing
equations based on fundamental principles of involved phenome-
na (Jorgensen, 1999). Phenomenological models are a considerably
simpler approach, useful for some descriptions (Kay et al., 2009).
Daisyworld is an archetypal model that represents the earth
describing the global regulation that can emerge from the
interaction between life and environment. Daisyworld is a
mathematical description of the Gaia theory of the earth
representing life by daisy populations while the environment is
represented by temperature. It was originally proposed by Watson
and Lovelock (1983) in order to describe the self-regulation of the
planetary system. Wood et al. (2008) presented a general overview
* *Corresponding author.
E-mail addresses: supaiva@gmail.com (S.L.D. Paiva), savi@mecanica.ufrj.br
? (M.A. Savi), fmviola@gmail.com (F.M. Viola), leiroz@mecanica.ufrj.br
(A.J.K. Leiroz).
the model main characteristics and different analysis approaches.
Spatial aspects and variants of the model are discussed.
Adaptability is an important issue that defines system evolution
by natural selection. In brief, it is possible to say that Daisyworld
allows one to perform a qualitative description of the climate
system, permitting the investigation of different aspects of the
system behavior. Because of that, several researches are dedicated
for variants and extensions of the Daisyworld.
An important discussion about Daisyworld characteristics is
related to the antagonism between altruism and competition of
the daisy populations. In this regard, several researches were
developed trying to investigate adaptive behavior of the Daisy-
world, incorporating competition and mutation aspects of life.
Lenton and Lovelock (2000) discussed some adaptive aspects of
the Daisyworld based on the idea that it is a Darwinian model in
the sense that there is a competition among different types of life
with heritable variation in a trait, which represents a kind of
natural selection (Robertson and Robinson, 1998). Daisy colors (or
planetary albedo) and optimum growth temperature are the main
parameters related to the Daisyworld adaptation.
Restrictions on environmental conditions define some evolu-
tion changes. Cohen and Rich (2000) introduced an extra source of
competition that describes the interaction among daisy species in
order to retain temperature in an appropriate range in order to
make the planet habitable. Boyle et al. (2011) pointed that
homeosthasis of the Daisyworld can be promoted by symbiotic
physiology. Qualitative comparisons are evaluated with and
without the consideration of symbiotic physiology.

http://dx.doi.org/10.1016/j.biosystems.2014.09.008
0303-2647/ ã 2014 Elsevier Ireland Ltd. All rights reserved.
2 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15

Table 1
Parameters of the Daisyworld.

aw 0.75
ab 0.25
ag 0.5
g 0.3
q 2.06  109 K4
Topt 295.5 K

Table 2
Parameters of the Daisyworld considering different timescales.

Different timescales

W m
2 J day
1 m
2

s
8
5.67  10
4
(K ) 4.90  10
3 (K
4)
S 915 7.91 107
c 950 (K
1) 8.21 107(K
1)

Fig. 1. Gaussian representing the optimum condition and limits of life.

Weaver and Dyke (2012) discussed Daisyworld model from
timescale perspective. Basically, it is considered temperature,
changes in insolation, and self-organisation of the biota using an
analytical approach.
One of the most important characteristics of the Daisyworld is the
capability to describe either global or local phenomena. Global
analysis assumes that the translational movement of the planet is the
main component of temperature variation, being the planet
subjected to temperature changes according to seasons of the year
or aspects of the planet. Under this assumption, Daisyworld is related
to the entire planet. On the other hand, local analysis defines a

Fig. 2. Classical Daisyworld response (Nevison et al., 1999). (a) Daisy dynamics; (b) temperature dynamics; (c) projection of the state space showing the daisy subspace.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
specific region of the planet. Therefore, either rotation or translation
of the planet influence solar energy input to the planet since a specific
region is subjected to temperature changes due to solar luminosity
variations associated with days and nights, and also with seasons.
Staley (2002) and Charlson et al. (1987) discussed the differences
between the global and local effects on the planet dynamics.
The Daisyworld model was also studied considering
spatio-temporal behavior. Adams et al. (2003) proposed a
one-dimensional model incorporating a distribution of incoming
solar radiation and heat diffusion consistent with a spherical planet.
Biton and Gildor (2012) added seasonal solar radiation forcing to the
latitudinal-depended one-dimensional Daisyworld model. Wood
et al. (2006) included a mutation of the optimal growth temperature
and of the albedo in a two-dimensional Daisyworld model
determining the temperature oscillations. Ackland (2004) used a
two-dimensional Daisyworld model considering the temperatures
and albedos on a grid indexed by longitude and latitude positions.
Fig. 3. Daisyworld with greenhouse effect where a = 10
4 and b = 0. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing the daisy-greenhouse effect subspace; (d) subspace projections.
3
Daisyworld can be employed for global warming description. In
essence, greenhouse gases affect Daisyworld behavior in a similar
way of the black daisies in terms of radiation absortion.
Nevetheless, the interaction with environment has an important
difference. Viola et al. (2013) incorporated greenhouse gases in the
Daisyworld considering prescribed time series that alter system
albedo. Besides, a sinusoidal variation of the solar luminosity
represents climate variability. Hence, the Daisyworld is employed
for global warming description and complex dynamical responses
are observed. Chaotic behavior was of concern showing that this
kind of response is possible under certain conditions. Zeng et al.
(1990) and Wood et al. (2008) also investigated possibilities
related to the chaotic behavior of the Daisyworld.
This paper deals with the description of the global warming
using the Daisyworld model considering that greenhouse gases are
included in the model with the aid of an extra equation motivated
by the carbon cycle dynamics and predator–prey models. The
4 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Fig. 4. Daisyworld with greenhouse effect where a = b = 10
4. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space showing
the daisy-greenhouse effect subspace; (d) subspace projections.
temperature dependence of life is represented by a Gaussian
function avoiding life extinction observed on the classical peak
function of the Daisyworld model. Emissions and absorptions of
greenhouse gases are treated showing the capability of the model
to describe global warming scenarios. Numerical simulations show
this general qualitative capability presenting several simulations
related to greenhouse gases effect and also changes in solar
luminosity. Results are analyzed considering a nonlinear dynamics
perspective that provides a comprehension about system behavior.
2. Daisyworld model and the carbon cycle
The climate system is a complex system based on several
phenomena. The carbon cycle is associated with chemical and
biological interactions of great importance to define the origins and
the destinations of this element. Carbon is stored in four main
reservoirs of the earth: atmosphere, lithosphere, biosphere and
hydrosphere. Each reservoir contains a variety of carbon compounds
of organic and inorganic nature. Furthermore, the transfer time and
storage of carbon compounds in each reservoir can vary from years to
millennia. For example, the lithosphere contains a large amount of
carbon trapped in sedimentary rocks formations of mineral
carbonates and organic compounds such as oil, natural gas and
coal. The redistribution from the lithosphere reservoir to another is
associated with millions of years, until all the geological process,
such as chemical decomposition and deposition, occurs. Therefore,
the lithosphere is considered a relatively inactive component of the
carbon cycle when preserved in its natural state. The active reservoirs
are divided between the atmosphere, the terrestrial biosphere and
the hydrosphere. As the absolute sum of the amounts of active carbon
reservoir is maintained close to the steady state value by slow
geological processes, biogeochemical processes that lead to redis-
tribution of carbon between active reservoirs occur more rapidly.
The Daisyworld represents self-regulation of the planet that is
basically composed of the environment, represented by the
temperature, and life, represented by daisy populations.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15 5

A ¼ a w aw þ a b ab þ a g ag (4)
Daisyworld model represents life by daisy populations, which
evolution is described by the following equations: The functional form for b is usually assumed to be a symmetric
  single-peak function. Here, a Gaussian function is assumed as
a_ w ¼ aw ag bðT w Þ
g (1)
follows:

bðT i Þ ¼ Be
s ðT i
T opt Þ

1 2
  (5)
a_ b ¼ ab ag bðT b Þ
g (2)
where B is a parameter related to environmental characteristics, s
For the sake of simplicity, only two daisy populations are is the variance and Topt is the optimal temperature, usually
considered: white (aw) and black (ab) daisies. The dot represents considered as Topt = 295 K = 22.5  C. The Gaussian form of this
time derivative and b = b(Ti) represents the growth rate that is function is shown in Fig. 1 and it is convenient to avoid a drastic
temperature dependent and g is the population death rate. The population extinction observed on the classical peak function of
variable ag is the fractional area coverage of the planet represented the Daisyworld model.
by The local temperature of each population is defined as follows:
ag ¼ p
aw
ab (3) T 4w ¼ qðA
aw Þ þ T 4 (6)
where p = 1 represents the portion of land suitable for the growth
of daisies. The mean planetary albedo of the Daisyworld, A, can be
estimated from the individual albedo of each daisy population (aw T 4b ¼ qðA
ab Þ þ T 4 (7)
and ab), and from the ground albedo (ag):

Fig. 5. Daisyworld with greenhouse effect where a = 10
4 and b = 10
3. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing daisy-greenhouse effect subspace; (d) subspace projections.
6 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15

emission while b represents the absorption. Note that when b

T 4g ¼ qðA
ag Þ þ T 4
where T is the globally-averaged temperature of the Daisyworld,
and q is a constant used to calculate local temperature as a function
of albedo. Finally, it is important to establish the thermal balance of
the Daisyworld (Foong, 2006). Therefore, the absorbed energy is
given by (Nevison et al., 1999):
1
T_ ¼ ½SLð1
AÞ
s T 4 ð1
EÞ
c defines a specific region of the planet. Numerical simulations are
where L is the solar luminosity and S is the solar constant that
establishes the average solar energy represented by SL; s is the
Stefan–Boltzmann constant; c is a measure of the average heat
capacity or thermal inertia of the planet. Finally, E represents the
greenhouse effect that alters the thermal balance, described by the
following differential equation.
E_ ¼ a
bEðaw þ ab Þ
This new equation establishes the interaction between gases
and daisy populations being motivated by the general predator–
prey equation and the carbon cycle where life is responsible for the
absorption of the gases (Novak, 2006). Parameter a is related to gas
vanishes, no interaction occurs and the effect of greenhouse gases
(8) is only related to the emissions. From now on, the greenhouse
variable E is referred as greenhouse effect.
3. Numerical simulations
The Daisyworld model can represent global or local aspects of
the planet dynamics. Global analysis considers that Daisyworld is
(9) related to the entire planet. On the other hand, local analysis
carried out considering both situations and different scenarios.
Initially, numerical simulations are performed considering a
global analysis. Three different models are treated: the original,
without greenhouse gases; a model with greenhouse gases; and a
model with emission and absorption of greenhouse gases.
Moreover, luminosity variations are also investigated representing
the profile of solar activity. In this regard, three different scenarios
(10)
are considered: constant value; linear increase; and climate
variability represented by a sinusoidal modulation over a linear
growth.
The Daisyworld model can be simulated using classical
procedures for numerical integration. Here, a fourth order

Fig. 6. Daisyworld with greenhouse effect where a = b = 0. (a) Daisy dynamics; (b) temperature dynamics; (c) projection of the state space showing the daisy subspace.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Runge–Kutta method associated with an iterative process is
employed. Time steps smaller than 10
2 are assumed for the
global analysis while local analysis assumes time steps smaller
than 10
3. These time-step values are defined after convergence
analysis. Tables 1 and 2 show the values of the parameters
employed in all simulations.
3.1. Global analysis
This section presents the global analysis of the Daisyworld.
Three different luminosity scenarios are discussed: constant value;
linear increase; and climate variability. Each of these analyses may
represent a scenario related to different timescales, for instance.
For each one of them, three different situations are treated: the
original model, without greenhouse gases; a model with green-
house gases; a model with emission and absorption of greenhouse
gases.
3.1.1. Constant luminosity
A situation where greenhouse gases are not involved is used to
start the analysis. Therefore, emissions or absorptions of
Fig. 7. Daisyworld with greenhouse effect where a = 10
4 and b = 0. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing the daisy-greenhouse effect subspace; (d) subspace projections.
7
greenhouse gases are not considered in the model, which is
represented by a = b = 0. Fig. 2 presents Daisyworld response
showing consistent results compared with the classical Daisy-
world (Nevison et al., 1999). Fig. 2a shows the dynamics of daisy
populations while Fig. 2b shows the temperature dynamics. Fig. 2c
presents a projection of the state space in daisy subspace. Note
that the interaction between the two daisy species, white and
black, limits temperature values within a proper range for life
existence. The increase of white daisies tends to decrease of
temperature, while the increase of black daisies causes the
increase of temperature due to the respective albedo.
A scenario with greenhouse gases emissions is now investi-
gated, but the planet is not able to absorb these gases. This
analysis is accomplished by considering a = 10
4 and b = 0. Fig. 3
presents results of this case showing that the increase of gases
causes a general increase of planet temperature. Basically, Fig. 3a
shows the time history of daisy populations while Fig. 3b shows
the temperature and the greenhouse effect dynamics (repre-
sented by variable E). Fig. 3c shows a projection of the state space
considering daisy populations and greenhouse effect. Fig. 3d
presents some projections of the subspace of Fig. 3c. Results show
8 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15

Fig. 8. Daisyworld with greenhouse effect where a = 10
4 and b = 10
1. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing the daisy-greenhouse effect subspace; (d) subspace projections.

that in this situation, the balance between life and environment is

lost, and daisy populations cannot compensate the greenhouse
effect. It is noticeable the persistence of white daisies for a longer
time, acting to delay the life extinction caused by the excessive
increase of temperature. It should be highlighted that the
extinction is related to an escape imposed by the variable E,
clearly shown in state space projection (Fig. 3d).
The planet capacity to absorb greenhouse gases is now of
concern by assuming a = b = 10
4. In this scenario, daisy
populations contribute to gas absorption. Fig. 4 presents the
response of the Daisyworld for this case. It is clear the increase
of the self-regulation is caused by the absorption capacity.
Nevertheless, this is not enough to avoid the extinction of life on
the planet. The increase of the absorption capacity can be
represented by considering a = 10
4 and b = 10
3. This increase
avoids the extinction as shown in Fig. 5. Note that this new
scenario is related to a stabilization of the system response for a
specific variable E, which is clear in the state space projections
(Fig. 5d). Fig. 9. Luminosity representing climate variability.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
3.1.2. Linear increase of the luminosity
In order to represent a more realistic solar activity, a linear
increase of the luminosity is considered as:
L ¼ 1:9  10
4 t þ 0:75
Once again, three scenarios are treated: the original model,
without greenhouse gases; a model with greenhouse gases
emissions; and a model with greenhouse gases emission and
absorption. Initially, the dynamic behavior of the Daisyworld
without greenhouse gases, represented by a = b = 0, is observed.
Fig. 6 presents the general behavior of the Daisyworld where an
increase of luminosity causes a temperature increase, which leads
to unfavorable environmental conditions, causing the life extinc-
tion. The emission of greenhouse gases (a = 10
4 and b = 0)
accelerates this tendency even more, as shown in Fig. 7. Once
again, it is possible to associate the extinction with the greenhouse
effect variable escape observed in the projections of the state
space, Fig. 7d.
The absorption capacity can change the scenario associated
with emissions. By considering a = 10
4 and b = 10
1 (Fig. 8), the
Daisyworld presents similar results of the one presented in Fig. 6,
Fig. 10. Daisyworld without greenhouse effect where a = b = 0. (a) Daisy dynamics; (b) temperature dynamics; (c) projection of the state space showing the daisy subspace.
9
showing the system capacity to avoid extreme situations. Note
however, that there is still an extinction of life that occurs when the
luminosity reaches a critical value.
(11) 3.1.3. Luminosity representing climate variability
Climate variability is now in focus assuming a sinusoidal
variation of the luminosity over a linear increase as follows:
ðSL Þglobal ¼ S þ ðNsinwtÞ (12)
where (SL)global is the global solar energy and corresponds to SL,
where S is the solar constant and L is the luminosity; N is the
amplitude and w is the forcing frequency. Here, it is considered
N = 2.89  1012 and w = 0.01 that is related to a behavior presented
in Fig. 9.
Initially, situations without greenhouse gases (a = b = 0) are
considered. Under this condition, the dynamics of daisy popula-
tions and temperature levels are presented in Fig. 10. Note that
Daisyworld presents a more complex behavior due to sinusoidal
excitation.
A situation with greenhouse gases emission but without
absorption, represented by a = 10
4 and b = 0, is now considered.
10 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Fig. 11. Daisyworld with greenhouse effect where a = 10
4 and b = 0. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing the daisy-greenhouse effect subspace; (d) subspace projections.
Fig. 11 presents the Daisyworld behavior showing the same
complex response but causing the extinction event earlier when
compared with the previous case, showed in Fig. 10.
The inclusion of absorption, assuming a = 10
4 and b = 10
1,
changes the system dynamics as shown in Fig. 12. Note that life is
able to persist for a longer time, causing a regulation of the
temperature, similar to the situation without greenhouse gases.
For later times, when luminosity reaches critical values, the life
ability for system regulation saturates and life extinction is
achieved.
3.2. Local analysis
One of the characteristics of the Daisyworld model is the
capability to represent either global or local analysis (Wood et al.,
2008). This section treats the local analysis by assuming that the
solar energy input occurs in an isolated point of the planet. A
proper description of this luminosity needs to be proposed. For
example, in a region near the Equator, the solar luminosity is zero
over the nights and varies from zero to its maximum value during
the day. For the sake of simplicity, it is possible to consider that
luminosity intensity vanishes for twelve hours and varies from
zero to its maximum value through the day. Here, it is assumed the
following function to represent day–night behavior together with
seasons:
ðSL Þlocal ¼ S½sinð2ptÞ½M1 sinðwtÞ þ M2  (13)
where (SL)local is the incident solar energy and corresponds to SL,
where S is a solar constant equal to 7.9  107 J day
1 m
2 and L is the
luminosity; M2 = 2.8  107 and M1 = 32  107 are the amplitudes;
w = 0.0175, such that sin(wt) corresponds to one cycle of the year.
Fig. 13 shows the general behavior of the luminosity.
Once again, three scenarios are of concern: a model without
gases; a model with gases emissions; and a model with emission
and absorption of gases. Initially, consider a situation without
greenhouse gases, a = b = 0. Daisyworld behavior is presented in
Fig. 14. Fig. 14a shows the time history of daisy populations while
Fig. 14b presents the projection of the state space showing the
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15 11

Fig. 12. Daisyworld with greenhouse effect where a = 10
4 and b = 10
1. (a) Daisy dynamics; (b) temperature and greenhouse effect dynamics; (c) projection of the state space
showing the daisy-greenhouse effect subspace; (d) subspace projections.

Fig. 13. Dynamics of local luminosity and its highlight in 3 days.

12 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15

Fig. 14. Daisyworld without greenhouse effect (a = b = 0). (a) Daisy dynamics; (b) temperature and luminosity dynamics; (c) projection of the state space showing the daisy
subspace; (d) projection of the state space showing the black daisy-temperature subspace;(e) projection of the state space showing the white daisy-temperature subspace.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Fig. 15. Daisyworld with greenhouse effect where a = 5  10
5 and b = 0. (a) Daisy dynamics; (b) projection of the state space showing the daisy subspace; (c) temperature and
luminosity dynamics; (d) temperature and greenhouse effect dynamics.
daisy subspace. Fig. 14c shows temperature and luminosity
dynamics. Fig. 14d–e presents different projections of the state
space showing respectively, the black daisy-temperature and
white daisy-temperature subspaces. Two important aspects
should be observed: the stabilization of the system behavior
and the change in representing the changes through the year.
The introduction of greenhouse gases in the analysis (a = 5  10
5
and b = 0) changes the general behavior, causing an extinction event
(Fig. 15). Now temperature increases dramatically after years of
small increases in temperature. The state space shows a tendency to
go to a stationary response at the origin, associated with the life
extinction.
The effect of gas absorption changes again this general
behavior, bringing a proper balance through the system. By
assuming a = 5  10
5 and b = 10
3, Daisyworld behavior
is presented in Fig. 16. The temperature regulation of the
environment is noticeable, but temperature also has a general
13
increase over time. The state space allows one to identify a
creation of an equilibrium point with preponderancy of white
daisies.
4. Conclusions
The Daisyworld model aims to illustrate the biota influence in
their environment promoting self-regulation to benefit life. This
paper proposes a way to include greenhouse gases emission and
absorption into the model, allowing the description of the global
warming and other related effects. Two different analyses are of
concern: global and local. For each one of these, three scenarios of
the Daisyworld are compared: a model without greenhouse gases;
a model with greenhouse gases emissions; a model with
greenhouse gases emissions and absorptions. Solar luminosity is
also changed in order to contemplate different scenarios. Basically,
global analysis considers three situations related to solar
14 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Fig. 16. Daisyworld with greenhouse effect where a = 5  10
5 and b = 10
3. (a) Daisy dynamics; (b) projection of the state space showing the daisy subspace; (c) temperature
and luminosity dynamics; (d) temperature and greenhouse effect dynamics.
luminosity: constant value; linear increase; sinusoidal variation
over a linear increase. Local analysis, on the other hand, simulates
day and night luminosity and season variations. The greenhouse
effect acts to retain heat in the environment and reducing the
lifetime of daisies. On the other hand, the absorption capacity of
the planet tends to extend daisy lifetimes, increasing the
self-regulation. Daisy populations tend to vary in order to regulate
temperature in benefit of life. White daisies tend to increase when
greenhouse effect increases while the opposite occurs with the
black daisies. In general, the model is able to describe the influence
of greenhouse effect in a qualitative point of view. In this regard, it
allows one to investigate different qualitative scenarios related to
global warming.
Acknowledgements
The authors would like to acknowledge the support of
the Brazilian Research Agencies CNPq, CAPES and FAPERJ
and through the INCT-EIE (National Institute of Science and
Technology – Smart Structures in Engineering) the CNPq and
FAPEMIG. The Air Force Office of Scientific Research (AFOSR) is
also acknowledged.
References
Ackland, G.J., 2004. Maximization principles and Daisyworld. J. Theor. Biol. 227, 121–
128.
Adams, B., Carr, J., Lenton, T.M., White, A., 2003. One-dimensional Daisyworld:
spatial interactions and pattern formation. J. Theor. Biol. 223, 505–513.
Alexiadis, A., 2007. Global warming and human activity: a model for studying the
potential instability of the carbon dioxide/temperature feedback mechanism.
Ecol. Modell. 203, 243–256.
Biton, E., Gildor, H., 2012. The seasonal effect in one-dimensional Daisyworld. J.
Theor. Biol. 314, 145–156.
Boyle, R.A., Lenton, T.M., Watson, A.J., 2011. Symbiotic physiology promotes
homeostasis in Daisyworld. J. Theor. Biol. 274, 170–182.
Charlson, R.J., Lovelock, J.E., Andreae, M.O., Warren, S.G., 1987. Oceanic phytoplank-
ton, atmospheric sulphur, cloud albedo and climate. Nature 326, 655–661.
Cohen, J.E., Rich, A.D., 2000. Interspecific competition affects temperature stability
in Daisyworld. Tellus 52B, 980–984.
Foong, S.K., 2006. An accurate analytical solution of a zero-dimensional greenhouse
model for global warming. Eur. J. Phys. 27, 933–942.
Houghton, J., 2005. Global warming. Rep. Prog. Phys. 68, 1343–1403.
Jorgensen, S.E., 1999. State-of-the-art of ecological modelling with emphasis on
development of structural dynamic models. Ecol. Modell. 120 (2–3), 75–96.
Kay, A.L., Davies, H.N., Bell, V.A., Jones, R.G., 2009. Comparison of uncertainty
sources for climate change impacts: flood frequency in England. Clim. Change
92, 41–63.
Lenton, T.M., Lovelock, J.E., 2000. Daisyworld is Darwinian: constraints on
adaptation are important for planetary self-regulation. J. Theor. Biol. 206,
109–114.
 S.L.D. Paiva et al. / BioSystems 125 (2014) 1–15
Nevison, C., Gupta, V., Klinger, L., 1999. Self-sustained temperature oscillations on
Daisyworld. Tellus 51B, 806–814.
Novak, M.A., 2006. Evolutionary Dynamics: Exploring the Equations of Life. Belknap
Press of Harvard University Press.
Robertson, D., Robinson, J., 1998. Darwinian Daisyworld. J. Theor. Biol. 195, 129–134.
Staley, M., 2002. Darwinian selection leads to Gaia. J. Theor. Biol. 218, 35–46.
Viola, F.M., Savi, M.A., Paiva, S.L.D., Brasil Jr., F.M., 2013. Nonlinear dynamics and
chaos of the Daisyworld employed for global warming description. Appl. Ecol.
Environ. Res. 11 (3), 463–490.
Viola, F.M., Paiva, S.L.D., Savi, M.A., 2010. Analysis of the global warming dynamics
from temperature time series. Ecol. Modell. 221 (16), 1964–1978.
15
Watson, A.J., Lovelock, J.E., 1983. Biological homeostasis of the global environment:
the parable of Daisyworld. Tellus 35B, 284–289.
Weaver, I.S., Dyke, J.G., 2012. The importance of timescales for the emergence of
environmental self-regulation. J. Theor. Biol. 313, 172–180.
Wood, A.J., Ackland, G.J., Lenton, T.M., 2006. Mutation of albedo and growth
response produces oscillations in a spatial Daisyworld. J. Theor. Biol. 242,
188–198.
Wood, A.J., Ackland, G.J., Dyke, J.G., Williams, H.T.P., Lenton, T.M., 2008. Daisyworld:
a review. Rev. Geophys. 46, RG1001. doi:http://dx.doi.org/10.1029/
2006RG000217.
Zeng, X., Pielke, R.A., Eykholt, R., 1990. Chaos in Daisyworld. Tellus 42B, 309–318.