Human population as a dynamic factor in environmental degradation

Abstract

The environmental consequences of increasing human population size are dynamic and nonlinear, not passive and linear. The
role of feedbacks, thresholds, and synergies in the interaction of population size and the environment are reviewed here,
with examples drawn from climate change, acid deposition, land use, soil degradation, and other global and regional environmental
issues. The widely-assumed notion that environmental degradation grows in proportion to population size, assuming fixed per
capita consumption and fixed modes of production, is shown to be overly optimistic. In particular, feedbacks, thresholds,
and synergies generally amplify risk, causing degradation to grow disproportionally faster than growth in population size.

Full text

ORIGINAL PAPER
Human population as a dynamic factor
in environmental degradation
John Harte
Published online: 11 May 2007
ÓSpringer Science+Business Media, LLC 2007
Abstract The environmental consequences of increasing human population size
are dynamic and nonlinear, not passive and linear. The role of feedbacks, thresholds,
and synergies in the interaction of population size and the environment are reviewed
here, with examples drawn from climate change, acid deposition, land use, soil
degradation, and other global and regional environmental issues. The widely-
assumed notion that environmental degradation grows in proportion to population
size, assuming fixed per capita consumption and fixed modes of production, is
shown to be overly optimistic. In particular, feedbacks, thresholds, and synergies
generally amplify risk, causing degradation to grow disproportionally faster than
growth in population size.
Keywords Population Feedback Environment Threshold Synergy 
Climate warming
Introduction
For the past several centuries, humanity has been increasingly polluting air and
water, altering Earth’s climate, eroding the soil, fragmenting and eliminating the
habitat of plants and animals, and depleting the natural bank account of non-
renewable resources. Of especially great long-term concern, we are as a
consequence simultaneously degrading the capacity of natural ecosystems to
regenerate or maintain renewable resources and ecosystem services, such as the
provision of clean air and water, the control of flooding, the maintenance of a
Based on a presentation to the Bixby Symposium on Population and Conservation, UC Berkeley, May
2006.
J. Harte (&)
Energy and Resources Group and ESPM, University of California, Berkeley, CA 94720, USA
e-mail: jharte@berkeley.edu
123
Popul Environ (2007) 28:223–236
DOI 10.1007/s11111-007-0048-3

tolerable climate, the conservation and regeneration of fertile soil, and the
preservation of genetic and other forms of biological diversity (Daily, 1997; Harte,
1993; Harte, Torn, & Jensen, 1992; IPCC, 2001; L’Vovich & White, 1990; Myers,
1983; NRC/NAS, 1995; Postel, 1993; Postel, Daily, & Ehrlich, 1996; Rozanov,
Targulian, & Orlov, 1990; Shiklomanov, 1993; Westman, 1977).
The linkages between human activity and environmental degradation are myriad
but at the risk of some over-simplification one can usefully group the contributing
factors in three categories: human population size, the per-capita rate of
consumption of energy and materials that contribute to our affluence, and the
impacts stemming from the technologies used to provide that per-capita rate of
consumption. An ‘‘equation’’ is sometimes used to express this:
Environmental Impact ¼ðPopulation sizeÞ(per -capitaAffluence level)
(impact fromthe Technologies used to achieve that
level of per-capita affluence)
This ‘‘IPAT equation’’ (Ehrlich & Holdren, 1971) is a useful reminder that
population, affluence, and technology all play a role in determining environmental
impacts. But it is also misleading if taken too literally, for it conveys the notion that
population is a linear multiplier. In particular, it suggests that if per-capita affluence
is held constant, and the technologies and other means used to achieve that level of
per-capita affluence are also held constant, then the impacts simply grow in
proportion to population size....if the population doubles in size, then the impacts
double in magnitude.
In this article, I examine the more dynamic and complex role that population size
actually plays in shaping environmental quality. I argue that the notion that impacts
are roughly proportional to our numbers is hugely optimistic. In particular, it ignores
a host of threshold effects, synergies, feedbacks and other nonlinear phenomena that
tend to amplify environmental impacts and cause them to grow considerably faster
than linearly in population size, even if the per-capita level of affluence and the
types of technological systems deployed to achieve that affluence remain constant.
I first pointed out the misleading nature of the linearity implied by the IPAT
formulation ten years ago in the journal Biodiversity and Conservation (Harte,
1996). Since writing that article, I have become even more aware of and concerned
about the many non-linear multipliers that magnify the consequences of a growing
human population. This article reviews the basic mechanisms that induce non-
linearity, expanding on the examples introduced in 1996 and providing a rationale
for increasing concern about these issues.
Thresholds, synergies, feedbacks: an overview
Thresholds
There are many situations in which a small or intermediate-sized stress to a system
generates little or no impact, but when the stress exceeds a certain level (the
224 Popul Environ (2007) 28:223–236
123

threshold), then the impact increases dramatically. Then if the stress is simply
proportional to population size, the impact will behave as shown in Fig. 1.A
prominent example is the existence of threshold behavior in the response of surface
waters to acid deposition. In particular, the shape of the familiar acid buffer curve
describing pH as function of the amount of acid added to an initially alkaline
solution implies a threshold level of added acid. Below that threshold, alkalinity
prevents the lake water pH from falling significantly (that is, the lake water from
acidifying significantly), but above a certain level of added acid, corresponding to
the amount of alkalinity present originally in the water, the pH falls rapidly (Fig. 2).
Suppose we assume that the amount of acid rain falling to Earth each year is
simply proportional to the amount of coal being burned, and that the amount of coal
being burned is proportional to population size (because affluence and technology
are held constant). Then an increase in population size from a level below the
threshold to one above the level can generate a disproportionately large decrease in
the pH of the lake.
But the situation is even more non-linear than indicated by Fig. 2. Of the major
fossil fuels used today (coal, petroleum, and natural gas), coal is generally the one
that produces the most acidity per unit of energy. Thus, people interested in
reducing the threat of acid rain urge policy makers to replace coal with the cleanest
of these fuels, natural gas. Similarly, natural gas produces considerably less climate-
altering carbon dioxide than does coal for the same amount of produced energy. But
population size
impact
Fig. 1 A threshold relationship
cumulative amount of acidity in rain
(assume proportional to population)
lake pH
lethal level
for trout
Pop
1
2 x Pop
1
Fig. 2 The non-linear relationship between lake water pH and cumulative acid deposition
Popul Environ (2007) 28:223–236 225
123

unfortunately, natural gas supplies are far more limited than are coal supplies.
Hence, with every increase in the size of the human population, the day arrives
sooner when more coal, per capita, has to be used. In the IPAT formulation, one
would express this by stating that T, the technology, changed at the same time that P
increased. But the point is that this change in T is caused by a change in P, and thus
T should be considered as a function of P. And this means that P is not simply a
linear multiplier.
Another example of a threshold effect concerns the response of coral reefs to
global warming. The health of a coral reef is relatively insensitive to water
temperature provided temperatures remain below about 808F (or about 278C). Reef
productivity will exhibit temperature dependence, but the reef will remain viable
within a temperature of 58For108F. But above 808F a phenomenon called ‘‘coral
bleaching’’ occurs. In this process, the algae that combine with small animals to
create the symbiosis that comprises a living reef die, exhibiting what is called
‘‘coral bleaching’’ (Harte, 1993). Because many reefs occur in waters with
temperatures just a little below the critical value of 808F, as global warming raises
sea water temperatures around the world a time will come when coral reefs will
rather suddenly bleach and die.
For yet another example, we turn to the effect of habitat loss on the survival of
species. In places such as Amazonia, where roughly a quarter of the rain forest has
been destroyed, numerous species are driven to extinction because they depended on
intact habitat. Although we lack the knowledge needed to predict accurately how
many species are lost when a specified area of original habitat is lost, general
ecological principles do provide a way of characterizing the qualitative features of
the relationship between habitat area and species diversity. That understanding is
summarized in Fig. 3, which shows two alternative quantitative predictions that are
consistent with general ecological principles (Kinzig & Harte, 2000; NRC/NAS,
1995). Both exhibit a ‘‘soft-threshold’’ behavior, in which the loss of species
increases quite non-linearly as habitat loss increases. One of the curves in the figure
(based on species-area relationship) derives from the fact that the number of species
0
0.2
0.4
0.6
0.8
1
0
Based on species-area
relationship
Based on loss of endemic species
(species unique to an area)
Percent of Original Habitat Lost
Fraction of
species lost
When people destroy wild habitats, how many species are lost?
Under a wide
range of
assumptions,
species loss
increases
faster than
proportional to
habitat loss!
20 40 60 80 100
Fig. 3 The non-linear relationship between land degradation and species extinction
226 Popul Environ (2007) 28:223–236
123

that can be sustained in a region of intact habitat is a function of the area of that
intact habitat; the other (based on loss of endemic species) is derived from the fact
that the species that will go extinct when habitat is destroyed is determined by the
number of species that were endemic to (lived only in) the habitat area that was
destroyed. The inability of current ecological theory to pin down a unique response
function should not obscure the fact that virtually all ecological models predict that
the fraction of extinct species will increase faster than linearly with habitat loss.
Thresholds abound in nature. For example, threshold-like non-linearities
characterize the relationship between population density and the probability of an
infectious disease becoming epidemic in a region (Murray, 1989); a similar non-
linearity is likely to characterize the relation between the area of a monoculture crop
and the probability that a plague-like crop-pest outbreak will occur there. In some
instances, contingent and unpredictable factors blur the actual threshold level of a
stress, rendering it difficult to predict the level of an activity that causes a sudden
increase in impacts. For example, predicting the future time at which a lake will lose
its alkalinity and suffer a rapid drop in pH under increased acid loading is extremely
difficult because of uncertainties over rates and types of mineral weathering
processes and the dependence of those rates on water chemistry. In global warming
science, it is difficult to determine at what level of heating Greenland and West
Antarctic ice will rapidly melt, resulting in a rapid and very dangerous rise in sea
level. These difficulties point to a gap in predictive capability, however, and should
not obscure the fact that actual points of rapid response exist.
In some instances, however, there is controversy regarding just the existence of
thresholds. For example, in the toxicology of radiation, debate exists over whether
there is a threshold level of radiation exposure, below which no harm (and possibly
even some benefit) is conferred but above which health effects arise and increase
with the dose (Harte, Holdren, Schneider, & Shirley, 1991).
Synergies
Synergy occurs when the combined effect of two causes is greater than the sum of
the effects of the two causes acting in isolation of each other. Clearly, if two
environmental stresses act synergistically, and each stress grows in proportion to
population size, then the combined effect of both stresses may grow faster than
linearly with increasing population size.
Synergies can either be ‘‘good’’ for us or ‘‘bad’’ for us, depending on whether
the effects are helpful or harmful. The combined benefit of eating a more healthy
diet and getting more exercise may exceed the sum of the benefits of each alone. On
the other hand, when anthropogenic stresses are acting on an ecosystem there can be
a harmful synergy, in which the harms from each are reinforcing. One possible
reason for this is that such stresses may not be ones to which ecosystems were
subjected when the links within and between these systems were forged over
evolutionary and geologic time scales. Under these unusual stresses, the links
that form life-sustaining synergies under natural conditions turn against the health
of the system as they amplify, rather than dampen, stress (Harte, 1993; Harte, Torn,
& Jensen, 1992).
Popul Environ (2007) 28:223–236 227
123

Consider the following example. Both climate warming and deforestation are
predicted to stress biodiversity in currently forested areas of the world. The former
is also projected to result in forest dieback because of increased frequency of
drought conditions while the latter releases carbon dioxide, a greenhouse gas, to the
atmosphere, thereby exacerbating climate warming. Thus the sum of the effects of
each of these two stresses, climate warming and deforestation, generates additional
stresses that reinforce the total response.
While greenhouse gases result in global warming at the Earth’s surface and in the
lower atmosphere, they result in a cooling of the stratosphere, thereby increasing
ice-cloud formation there (IPCC, 2001). These ice clouds speed up stratospheric
ozone depletion by freezing up nitrogen compounds that would otherwise inhibit the
ozone-depleting reactions. Thus the greenhouse effect will worsen the problem of
stratospheric ozone depletion. At the same time, stratospheric ozone depletion can
lead to more intense smog because the additional ultraviolet radiation penetrating a
thinned stratospheric ozone layer will speed up the chemical reactions that produce
smog. Greenhouse gas warming at Earth’s surface also makes the urban ozone
pollution problem worse by speeding up the chemical reactions that lead to smog
formation. Deforestation can contribute to the problem of acid rain, as has been
observed in Africa where nitric acid is formed from burning vegetation.
To make matters worse, plants and animals weakened by any one of these threats
will generally be more vulnerable to the others. There are numerous examples of
this. Fish weakened by radiation have been shown to be more easily damaged by
thermal pollution than are healthy fish, and trees subjected to some air pollutants
become more susceptible to insect damage.
Figure 4illustrates the dominant interactions between pairs of selected
environmental stresses. Here a plus sign indicates a synergistic effect that
exacerbates the activity toward which the arrow points and a minus sign indicates
a canceling effect, in which one activity reduces the threat from another. As an
example of the complexity of the interactions, consider the case of fish and
acid rain
ozone hole
global warming
deforestation
+,-
+
+
+,-
+
+
+
+
+
+,-
+
+: synergy
=> impacts
grow faster
than population
Fig. 4 Synergies among environmental stresses
228 Popul Environ (2007) 28:223–236
123

amphibians threatened by lake acidification. Damage from acidification is often
observed in spring time when a pulse of acid is released from snowmelt right at the
time when aquatic life is at its most vulnerable: when eggs are developing. Under
global warming, less snowfall and more rain is anticipated in many regions where
snow is currently a major component of precipitation, and with reduced snowpack
and earlier snowmelt, the acid pulse may weaken in strength, to the benefit of
aquatic life. On the other hand, global warming may result in either a drying up of
lakes, or if not a total drying up, at least decreased lake-water oxygen levels, either
of which can result in the death of aquatic animals.
In the previous subsection ‘‘Thresholds’’ showed that if habitat loss is linearly
proportional to the size of the human population, the loss of species would likely
increase faster than linearly with population growth because of an intrinsic
nonlinearity in the relationship between habitat loss and species loss. But the
situation is even worse than this because habitat loss is likely to increase faster than
linearly with population size. To see this, consider Fig. 5, which illustrates how a
doubling of population can lead to a greater-than-doubling loss of habitat. Imagine
that initially a nation has three cities indicated by the three dots in the left panel.
Between each of the cities are highways and transmission line rights-of-way, which
are barriers to wildlife movement, death traps for some species, and opportunities
for invasive species (that may displace native species) to spread. The right-hand
panel shows the situation when the population has doubled and there become six
cities. Now there are 12 connecting lines, implying a fourfold increase in these
harmful stresses on plants and animals. While Fig. 5is only a schematic and not to
be taken too literally, it does indicate how the web of infrastructural interconnec-
tions in human society, which becomes increasingly dense as population grows, can
result in synergistic land use effects that are damaging to biodiversity.
Feedback
When the global climate system, an ecosystem, an organism, or any other complex
entity is disturbed by some perturbation, the effect on the entity (expressed, for
example, as a temperature change, an alteration of species diversity, or a change in
lifespan) is the sum of two terms. The first is the direct response. In the case of
climate change from the buildup of greenhouse gases in the atmosphere, the direct
response is the change in surface temperature due to the extra heat-absorbing
capacity of these gases. As the general circulation models (GCMs) consistently
show, this direct effect is roughly 18C for a doubling of atmospheric carbon dioxide
(IPCC, 2001).
population
doubling
Fig. 5 Non-linear habitat implications of population growth
Popul Environ (2007) 28:223–236 229
123

But this direct warming can also trigger secondary or indirect effects (Lashof,
DeAngelo, Saleska, & Harte, 1997). In the case of climate change, the area of Earth
that is covered by ice or snow decreases under warming. This induces a feedback to
the warming because ice and snow reflect more sunlight back to space than does the
surface exposed when the ice or snow melts, and hence melting results in an
additional warming due to the greater amount of sunlight absorbed at Earth’s
surface. This is called a positive feedback effect because it causes an even greater
warming than would have occurred in the absence of the process. Had warming led
to a shinier Earth surface, and thus caused some cooling, the feedback would be
negative.
The mathematical description of feedback is understood: if the direct effect of
warming is denoted by DT
0
, then the full effect with all the feedbacks is given by
DT=DT
0
/(1 g). The quantity gis called the feedback factor and a mathematical
procedure (Harte, 1988; Torn & Harte, 2006) exists for calculating its numerical
value provided the mechanisms that induce feedback are sufficiently understood. If
gis positive but less than 1, then the feedback results in DT>DT
0
, which
corresponds to positive feedback. If gis negative, DT<DT
0
and the feedback is
negative. If g> 1 the system is unstable and the analysis fails.
Torn and Harte (2006) explored the consequences of the fact that when gis near
1, say 0.8, then a relatively small increase in g can cause a proportionally large
increase in the factor 1/(1 g). For example, a 10% increase in gfrom 0.80 to 0.88
results in a 67% increase in 1/(1 g), from 5 to 8.33. If the magnitude of gis
linearly proportional to population size then it is easy to see that the full impact will
grow faster than linearly in population size because the function 1/(1 g) grows
faster than linearly in gas ggets larger.
In environmental science, are there feedback processes in which the feedback
factor, g, grows with the size of the human population? Some very important
feedback processes have feedback factors that are essentially independent of
population size. Consider, for example, the snow- and ice-melt effect discussed
above. The direct temperature change that triggers the feedback will depend on
population (because people emit greenhouse gases), but the feedback factor, g,
depends only on the physical properties of ice and snow, and on the relative
response of temperature to a unit change in surface reflectance; it will not depend on
population size.
On the other hand, there are classes of feedback mechanisms for which gdoes
grow with population size—mechanisms that directly involve human response to
the direct effect. For example, in a warmer climate people may rely more on air-
conditioning, thereby burning more fossil fuel, augmenting emissions of carbon
dioxide, and causing a positive feedback. A decreased use of heating fuels during
warmer winters associated with global warming would have the opposite effect and
result in a negative feedback. More significant feedback effects involving
population size and climate involve the interaction of climate-independent human
activities and biophysical responses to warming. For example, carbon dioxide
release from soils is likely to increase with soil temperature and population size. In
particular, climate warming is likely to accelerate the decomposition of soil organic
matter, particularly in tilled, fertilized, and irrigated soils (Schlesinger, 1991) and
230 Popul Environ (2007) 28:223–236
123

this will lead to the release to the atmosphere of carbon dioxide. Because the total
area of such lands depends on population size, the gain factor, g, associated with this
positive feedback is dependent on population size.
Another example concerns wildfire, which will become more frequent and more
intense under the drier conditions anticipated under global warming. The positive
feedback arises because forest fires release significant quantities of carbon dioxide to
the atmosphere, and thus contribute to further climate warming. The value of the
feedback factor is population-dependent because even though fires are exacerbated
by drought they are often inadvertently, or sometimes purposely, triggered by people.
Additional reasons why impacts grow disproportionally faster than population
In addition to thresholds, synergies, and feedbacks, there are an assortment of other
reasons why the it is overly optimistic to assume that impacts merely increase in
proportion to the size of the human population. One has to do with the exhaustion of
the natural processes that result in ‘‘sinks’’ for our pollutants. For example, each
year the oceans and forests remove from the atmosphere a net quantity of carbon
dioxide, thus alleviating somewhat the problem of global warming. Currently a
quantity of carbon dioxide equal to about a third of what we emit each year is
removed each year, with somewhat more than half of that going into the oceans and
much of the remainder going into net growth of forests. But these natural sinks have
only a limited capacity to take up carbon, and like kitchen sinks they can partially
clog if too much is put down them. Thus the rate at which carbon dioxide can be
removed from the atmosphere is a declining function of the amount that has already
been removed, and so, in effect, is a declining function of population size at a fixed
per capita affluence level and fixed technology.
The natural carbon sinks can be thought of as a non-renewable resource. As with
natural gas, which is a cleaner alternative fuel to coal but one which we are rapidly
exhausting, environmental problems increase when we use up the carbon sink. With
both natural gas and the carbon sink, the faster our population grows, the quicker we
use up these desirable resources and thus the greater the impacts on the
environment. Similarly, when natural supplies of clean water, either in aquifers or
in surface streams, are exhausted, then for substitutes we have to turn to new
technologies to provide water, and many of these, such as dams and desalination,
bring about environmental costs.
For another example of this, consider the effect of climate change on water
supplies. As hotter temperatures and more frequent and intense drought increase
water demand in agriculture, with both the magnitude of climate warming and the
amount of agricultural land and water needed to grow food increasing with
population size, we will be forced into more energy-intensive means of obtaining
water and thus will accelerate the warming in order to sustain a fixed level of per-
capita affluence. Hence we can think of the term T in IPAT, the impacts from the
technologies needed to sustain a given per capita level of affluence, as inevitably a
function of population, P. These two drivers cannot be disentangled and expressed
as a product.
Popul Environ (2007) 28:223–236 231
123

It is not just the magnitude of environmental impacts, I, that increase
disproportionately with P when A and T are fixed. In addition, our efforts to solve
social problems, such as environmental injustice arising from inequitable distribu-
tion of impacts and of resources across income, cultural, and racial groups, are
hindered by rapidly growing population. Institutions cannot maintain a sufficiently
fast enough growth to provide universal access to education and health care for a
rapidly growing population. And again there is a pernicious feedback at work here:
the existence of these social problems makes it more difficult to solve
environmental problems. For example, in an equitable society with only small
income disparities, a carbon tax to discourage fossil fuel consumption would make a
great deal of sense. While it is a sales tax, its burden would not fall on a particular
group (the poor, who spend a larger fraction of their income on fuel than do the
rich), and thus it could replace an income tax without exacerbating inequality. Thus
the population trap catches us twice. Growing population size exacerbates social
problems and growing social problems exacerbate efforts to confront environmental
problems.
As discussed previously (Harte, 1996) it cannot even be claimed that the IPAT
equation is a units identity, the way we can write J=WT, where Jis joules of energy
consumed, Wis watts of power consumed, and Tis time over which the power is
utilized. J=WT is a valid equation because it simply expresses a conversion
between units. It works because the first joule consumed is no different from the last
joule consumed—all joules, all watts, and all time intervals are equivalent here and
our time scale does not depend on how much energy we have used. Unlike the
situation for fungible quantities such as joules, seconds, or watts, a unit of ‘‘impact
per unit of economic activity’’ does not exist. A marginal increment in activity
often leads to impacts that are different from the impacts from the average
increment in activity.
Implications for confronting global warming
Insight into the policy options for dealing with global warming has been provided
by Pacala and Socolow (2004), who have introduced the idea of policy wedges.
They argue that no single technology is going to lead to a substantial reduction in
the future level of carbon dioxide emissions, and so a collection of small
contributions from many technologies will be needed. Figure 6a, modified from
their paper, illustrates this idea and the reason for calling the contributions
‘‘wedges’’. The wedges that are potentially available include improving the
efficiency with which we use energy, switching from coal to natural gas,
sequestering carbon underground, and relying more on solar photovoltaics, wind,
nuclear and biomass. Although not included in the Socolow/Pacala conceptualiza-
tion, reduction of human population size could be considered to be an additional
wedge.
In Figure 6a I have slightly modified their original figure to include not just a
future stabilization of emissions, but rather a stabilization of the climate. The point
is that if emissions could be held in the future to 2000 emissions levels, global
232 Popul Environ (2007) 28:223–236
123

warming will still intensify in the future because a steady level of emissions at the
year 2000 rate will lead to a continuing buildup of carbon dioxide in the atmosphere.
To eventually stabilize the human contribution to climate change, we actually have
to reduce emissions down to a level at which the removal of carbon dioxide from the
2000 2050
14
Climate
Stabilization
7
Gigatons
Carbon Emitted
per Year
1950
Doubled
emissions from
“business as
usual”
Current
emissions
The Stabilization Wedges
0
Emissions
Stabilization
2000 2050
14
7
Gigatons
Carbon Emitted
per Year
1950
Current
emissions
0
Destabilization
Wedges!
Effect of
feedbacks, loss
of carbon sinks
The size of these
wedges, and thus the
burden on future
generations,
will depend upon which
energy strategy
we choose
a
b
Fig. 6 Future carbon dioxide emissions: a., The Socolow stabilization wedges for mitigating climate
change; b. Adding destabilization wedges, exacerbated by population growth, makes the task of climate
mitigation more difficult
Popul Environ (2007) 28:223–236 233
123

atmosphere by natural sinks will result in a constant concentration of this
greenhouse gas. The figure sketches a possible future trajectory of emissions that
would be needed to eventually accomplish the goal of climate stabilization.
The concept of stabilization wedges, whether it be emissions or climate
stabilization, is appealing because it delineates the challenge in a manner that
provides an antidote to those who say ‘‘nuclear power is the answer’’ or to those
who say ‘‘we can never build enough windmills to solve this problem’’. But it is
helpful to understand the full dimensions of a problem, and so in Figure 6b I have
included in the Socolow wedge diagram a needed modification. It reminds us that
there are also destabilization wedges...that is, Earth-system processes such as
feedbacks and clogging of the carbon sinks that will operate in the future to actually
increase the magnitude of global warming above the level currently projected.
These destabilization wedges are additional increments of greenhouse gases that
will be emitted to the atmosphere as a feedback response to climate warming, and
their magnitude will be exacerbated by population growth. Additional sources of
greenhouse gas imply that additional remedial wedges will be needed to bring about
climate stabilization.
Conclusion
There are five underlying reasons why the effect of population size on environ-
mental problems is much greater than is suggested by a simple linear multiplier
equation such as IPAT....five general principles that in various combinations apply
to virtually all instances of resource use and environmental degradation:
1. Thresholds lurk throughout nature: The ‘‘dose-response’’ relation between
environmental stresses (such as acidity of precipitation, or area of drained
wetlands, or amount of carbon dioxide loaded to the atmosphere) and the
response of the natural environment to that stress (such as acidification of soil
and lake water, or useful lifetime of an aquifer, or amount of warming induced
by carbon dioxide) are often nonlinear, with threshold-like responses increasing
faster than stress over a range of stresses relevant to current conditions. Thus,
even if stresses are merely proportional to population, the responses increase
faster than proportionally.
2. The whole is often more than the sum of the parts: Numerous synergies exist
among different kinds of environmental responses; these synergies are such that
the impacts from deforestation, acidification, ozone depletion, climate change,
erosion, water impoundment, pesticide use, etc., tend to mutually reinforce each
other, making the whole environmental impact from many stresses much more
damaging than the sum of the component impacts.
3. Scientific feedbacks amplify environmental degradation: Positive feedback
processes in which the gain factor is proportional to population size generate
impacts that can increase considerably faster than linear population growth.
4. Low-hanging fruit gets picked first: Because we would like to use the most
fertile soil, the cleanest water, the least polluting fuels now, the effect of future
234 Popul Environ (2007) 28:223–236
123

drawdown of these desirable resources implies that over time our options will
be increasingly limited to second-rate resources. While this tendency may be
partially remedied by putative technological innovations (such as carbon
sequestration), the effect of increasing population size on the rapidity with
which we use up or degrade the most desirable resources implies yet another
non-linear role of population growth in exacerbating our environmental
problems.
5. Rising numbers impede governance and problem-solving: High rates of
population growth make it more difficult to ensure adequate schooling,
material resources, and civic order, thereby worsening social conditions. That,
in turn, makes implementing solutions to environmental problems more
difficult. Unsolved environmental problems further exacerbate injustice and
inequity, again weakening the social order. The overall effect is an intensifying
downward spiral.
The relation between population size and the sustainability of the human
enterprise is complex, indeed. For a specified means of achieving a constant per-
capita level of well-being, impacts can grow far faster than linearly in population
size. The general structure of that relation is such that considerably greater concern
over population growth is warranted than has generally been shown by public policy
makers. The synergies, feedbacks, and thresholds discussed above force us to
reassess the traditional viewpoint that impacts are simply a product of population
size times some population-independent measure of the impacts of achieving a
given per-capita level of well-being.
To conclude, humanity is degrading environmental goods and services such as
clean water, air, soil, and biodiversity, and simultaneously reducing the capacity of
natural processes to replenish these contributors to the quality of life. As abhorrent
as is the current inequity in the distribution of resources between north and south,
rich and poor, it pales in comparison with the impending inequity between us, living
today, and those who will be born tomorrow and who, under current trends, will
inherit a rapidly deteriorating planetary life support system. The future habitability
of our planet will be shaped by highly nonlinear dynamical mechanisms that have
the potential to generate unintended and undesirable consequences for future
generations. As the stewards of our descendents, it is our moral obligation to better
understand that landscape and seek pathways to the future that avoid such rapidly
escalating damages. Implementation of family planning policies throughout the
world that give people greater control over reproduction, in under-developed and
over-developed nations alike, is a critical step toward that end.
References
Daily, G. (Ed.). (1997). Nature’s services. Washington, DC: Island Press.
Ehrlich, P., & Holdren, J. (1971). The impact of population growth. Science, 171, 1212–1217.
Harte, J. (1988). Consider a spherical cow: A course in environmental problem solving. Mill Valley, CA:
University Science Books.
Harte, J. (1993). The green fuse: An ecological Odyssey. Berkeley, CA: University of California Press.
Popul Environ (2007) 28:223–236 235
123

Harte, J. (1996). Feedbacks, thresholds, and synergies in global change: Population as dynamic factor.
Biodiversity and Conservation, 5, 1069–1083.
Harte, J., Holdren, C., Schneider, R., & Shirley, C. (1991). Toxics A to Z: A guide to everyday pollution
hazards. Berkeley: University of California Press.
Harte, J., Torn, M., & Jensen, D. (1992). The nature and consequences of indirect linkages between
climate change and biological diversity. In R. Peters & T. Lovejoy (Eds.), Global warming and
biological diversity (pp. 325–343). New Haven: Yale University Press.
IPCC (2001). Climate change 2001: The scientific basis. Contribution of working group I to the third
assessment report of the intergovernmental panel on climate change. Cambridge, UK: Cambridge
University Press.
Kinzig, A., & Harte, J. (2000). Implications of endemics-area relationships for estimates of species
extinctions. Ecology, 81(12), 3305–3311.
Lashof, D., DeAngelo, B., Saleska, S., & Harte, J. (1997). Terrestrial ecosystem feedbacks to global
climate change. Annual Review of Energy and the Environment, 22, 75–118.
L’Vovich, M., & White, G. (1990). Uses and transformations of terrestrial water systems. In B. Turner
(Ed.), The Earth as transformed by human action (pp. 235–252). Cambridge, UK: Cambridge
University Press.
Murray, J. D. (1989). Mathematical biology. Berlin: Springer-Verlag.
Myers, N. (1983). A wealth of wild species. Boulder: Westview.
NRC/NAS (1995). Science and the Endangered Species Act. Committee on Scientific Issues in the
Endangered Species Act, Commission on Life Sciences, National Research Council. Washington
DC: National Academy Press.
Pacala, S., & Socolow, R. (2004). Stabilization wedges: Solving the climate problem for the next 50 years
with current technologies. Scientific American, 305, 968–972.
Postel, S. (1993). Water and agriculture. In P. Gleick (Ed.), Water in crisis: A guide to the world’s fresh
water resources (pp. 56–66). Oxford: Oxford University Press.
Postel, S., Daily, G., & Ehrlich, P. (1996). Human appropriation of renewable fresh water. Science, 271,
785–788.
Rozanov, B., Targulian, V., & Orlov, D. (1990). Soils. In B. Turner (Ed.), The Earth as transformed by
human action (pp. 203–214). Cambridge, UK: Cambridge University Press.
Schlesinger, W. (1991). Biogeochemistry: An analysis of global change. San Diego: Academic Press.
Shiklomanov, I. A. (1993). World fresh water resources. In P. Gleick (Ed.), Water in crisis: A guide to the
world’s fresh water resources (pp. 13–24). Oxford: Oxford University Press.
Torn, M., & Harte, J. (2006). Missing feedbacks, asymmetric uncertainties, and the underestimation of
future warming. Geophysical Reserch Letters, 33, L10703. doi: 10.1029/2005GL025540M.
Westman, W. (1977). How much are natures services worth? Science, 197, 960–964.
236 Popul Environ (2007) 28:223–236
123