Demographic Transition

Theory

Population and Life-extension

New Questions
 Demographic Transition Theory
Part One: What is it?

Demographic transition theory is a supposition (that sometimes really occurs and sometimes does not) that a population
undergoing a population explosion may somehow automatically and gradually correct its excessive growth by slowing
down its growth more and more until it returns to stability and equilibrium with its environment.

If a graph is drawn of such a population explosion as it gradually slows and returns to stability, the resulting graph is an
s-curve (logistic curve) as shown in this articles cover image. (Notice, however, that this articles cover also exhibits a
second image which importantly shows an s-curve supposition that is crumbling and falling apart.)

That is because demographic transition outcomes, despite the hopes, wishes, and suppositions (and/or deliberate decep-
tions) offered up by some authors, sectors of academe, etc., are not an automatic outcome, are not universal, do not re-
flect some universal or inviolate rule or law of nature, do not always occur, and do not have to occur at all.

So for the sake of humankind, rising generations of under-20s, and the ONLY planetary
biospheric life-support machinery so far known to exist anywhere in the universe,
let us clarify the picture and broaden our understanding as follows:

There are three population patterns that are commonly seen in natural systems. In the easiest of these, a population of
organisms may live for decades or even centuries at a time in relative equilibrium with their environment and with other
organisms in their surroundings. In such equilibrial conditions, the populations numbers remain relatively constant,
(sometimes with small or recurring seasonal oscillations, etc.) so that a graph of such a population exhibits ongoing sta-
bility over time.

Occasionally, however, some populations may escape such stability and undergo a sudden POPULATION EXPLOSION in
which their numbers begin to skyrocket upward along the y-axis of their population graph. Such population explosions
commonly exhibit one of two general outcomes: Sometimes the growth of an exploding population, for varying reasons,
begins to gradually slow down more and more, resulting in a gradual return to equilibrial stability as reflected in a class-
ical s-shaped or logistic curve (this first outcome, of course, would be the preferred and hoped-for result envisioned by
some who wishfully-contemplate a peaceful demographic transition for humankind).

There is, however, a second and exceedingly-dangerous classical outcome of real-world population explosions. And in
this second outcome, an exploding population does not slow-down at all, or does not slow-down enough, or does not
slow-down in time, and instead continues to rocket upwards in its surroundings until it overshoots the carrying capacity
limits of its environment and results in massive, calamitous, and catastrophic mass die-offs (e.g. 99%-plus die-offs)
that result in population-explosion graphs known as CLIMB-AND-COLLAPSE.

Therefore, what mechanisms and natural principles normally operate to keep populations stable? Populations do not
normally explode because whatever their fertility rates are, their numbers are held in check by outside factors such as
pathogens or disease, predation, competition with each other or with other species for available food and habitat, and/or
by limited supplies of habitat, food, and resources.

Thus, population explosions result from a phenomenon known to science as ECOLOGICAL RELEASE. Such events oc-
cur when one or more of the factors that originally held their numbers in check is suddenly diminished or removed. As
a quick example, wild coyotes eat jackrabbits, so when ranchers or dust storms killed large numbers of coyotes, jack-
rabbit populations suddenly exploded in numbers. Similarly, sea otters along the California coast commonly feed on sea
urchins, so when hunters began killing large numbers of sea otters for their pelts, sea urchin numbers suddenly exploded
and their grazing began to threaten the regions offshore kelp beds. Thus, sudden escape from disease, competition, or
predation can serve as an ECOLOGICAL RELEASE that results in a sudden population explosion.

Possible implications for us?

Among the foremost factors that have triggered humankinds dangerous J-curve population explosion (ecological re-
lease) have been our advances in medicine which have suppressed the pathogens and disease that once held our num-
bers in check. So we are all grateful, of course, because scientific research and medical advances save lives. A population
aspect develops, however, if such medical advances reduce average death rates while worldwide average fertility rates
do not diminish accordingly.

In another respect, humankinds population explosion also exhibits an unprecedented sequence never before seen in the
history of life on Earth, for we have undergone not just ONE ecological release, but a pattern of one ecological release
followed by another and another repeatedly (e.g., multiple ecological releases in medicine, for example, such as
antibiotics and transplants, etc., together with other multiple ecological releases in agriculture). And shortly we will ad-
dress recent research that has already achieved six-fold life-extensions in laboratory organisms.

In natural systems, one ecological release and its subsequent population explosion is dangerous enough to invite 99%-
plus die-offs and/or climb-and-collapse outcomes. But multiple ecological releases, one after another and another and
another and followed over and over again on a global and worldwide scale is unheard of until now

Multiple Ecological Releases: One after Another?

Below is a graph of human births and deaths on the island of Sri Lanka in the mid-twentieth century.

Births/1000

Deaths/1000

Sri Lanka: Falling birth rates and falling death rates

Upper line: Births/1000 Lower line: Deaths/1000

What on Earth does the above graph of the births and deaths on the island of Sri Lanka in the mid-twentieth century
have to do with demographic transition theory and the future of our planet? As the section below will demonstrate,
many of today's population projections may be .underestimating. the worldwide population numbers that may emerge
in this century.
(And where research into life-extension may be about to send us.)

If explosive population growth, for example, poses potentially-dangerous biospheric, humanitarian, and civilizational chal-
lenges, then falling birth rates might seem to indicate some easing of the crisis. During the past three decades, for instance,
birth rates have fallen in many parts of the world, so that, if taken on its own, such data might initially be interpreted as a
critically-important step in the right direction (and, in an environmental sense, as potentially good news for our planet).

Births/1000

Deaths/1000

Sri Lanka: Falling birth rates and falling death rates

Upper line: Births/1000 Lower line: Deaths/1000

There is just one catch, however, to the above interpretation and it can be seen in the graph above: Quite often when
birth rates are falling (top line), death rates are falling even faster (lower line). Clearly, given our impacts upon our
planet, falling birth rates can be considered good news. And falling death rates also constitute good news because we
are saving more lives and people are living longer. Ironically, however, when taken together, they add up to biospheric
bad news.

Sri Lanka
In the decades following World War II the nation of Sri Lanka generated a set of data (as depicted in the..
above graph) that illustrates the interactive role of birth rates and death rates in a human population.
Births Deaths Extra
Year per 1000 per 1000 per 1000

1939 35 21 .14 .
1940 34 20 14
1945 38 20 18
1947 36 19 17 This chart outlines the Sri
1950 39 11 . 28 Lanka data set for the
1955 35 0 9 26 years 1939-1984 which
1960 32 08 24 was used to generate the
1965 33 09 24 graph.
1970 30 08 22
1975 28 06 22
1980 25 06 19
1984 27 0 6 .21.

Examine the numbers for 1939, and notice that for every thousand residents of Sri Lanka that year, there were 35
births and 21 deaths. Thus, by the end of 1939, every person who had died had been physically replaced,
and then fourteen additional babies were born per 1000.
 Next, examine the numbers for 1984. Notice that birth rates fell substantially from 35/1000 in 1939 to just 27/1000
forty-five years later. Initially, this seems like a step toward reducing Sri Lanka's rapid rate of population growth,
for with a fairly substantial decline in its birth rate, we might expect Sri Lankas population to grow less rapidly.

Notice that in the previous page chart, however, that in 1939 there were 14 extra babies per thousand
residents, while in 1984 there were now 21 extra babies born per thousand residents.

How could this be? What had happened?

Even WITH generally FALLING birth rates (35/1000 down to 27/1000 is about 25% reduction) Sri Lankas population
after more than four decades of falling birth rates found its RATE of population growth had INCREASED by fifty percent.
(It was NOW growing larger by 21 extra per thousand instead of the earlier 14 extra per thousand.)

Births/1000

Deaths/1000

In the graph, the top line depicts birth rates over the study period. The bottom line tracks death rates recorded during the
same years. If we consider the gap that separates the two lines, we note that in 1939 the gap between births and deaths
is fairly narrow. But when we check for its value forty-five years later (1984), the gap had appreciably widened.

What had happened? Even though there was a 25% reduction in birth rates, death rates had fallen EVEN MORE (50%).
In other words, improvements in health and medicine had completely cancelled out the effects of forty years of fertility
reduction and after 45 years, Sri Lankas rate of population growth was EVEN FASTER than it was at the outset.

What we see in Sri Lanka's data above mirrors events around the world over the past two centuries (and may foreshadow
worldwide events in the decades just ahead). For example, in the 100 years between 1850 and 1950 the worlds most
developed countries doubled their populations, "largely due to a decline in the death rate" (Mader, 1996).

Poverty traps
We see, therefore, that something very, very important has happened. When todays less developed countries imported
modern medicines following World War II, mortality declines followed quickly, while birth rates remained high and
changed little.

In contrast to the experience of the developed world, however, todays Least-Developed Countries do not have the luxury
of one hundred years to build infrastructure and adjust to their exploding numbers, because many of todays poorest and
least-developed nations currently exhibit population doubling times as short as 25 to 30 years, and many of them are
doubling and doubling repeatedly.
The population of sub-Saharan Africa, for example, has quadrupled since 1950 (e.g., Sachs, 2008). As a result, many
such populations are growing so fast that it is impossible for their infrastructure, jobs, educational, and service sectors
to keep up. As Mader points out (ibid), the population of the LDCs began to increase dramatically and many people
in these countries are underfed, ill-housed, unschooled, and living in abject poverty. Jeffrey Sachs at Columbia
University characterizes similar suites of conditions as poverty traps (2008).

As author Sachs notes, the decline of fertility lagged behind the decline of mortality, and a massive
bulge of population ensued, and now, as the worlds population has continued to increase, the threats
to human well-being from rising populations have also multiplied (Sachs, 2008).
In effect, humanity today has painted itself into a demographic
and environmental corner with few, if any, painless exit options.

Good News and Bad News

The good news in Sri Lanka was that: (a) Its birth rates fell, and (b) Its death rates also dropped, saving many
lives and allowing its citizens to live longer and healthier lives. Thus, on two fronts, there were significant
advances in bettering human lives. Of course, something else was also happening, that, in retrospect, might have
been expected: Major advances in medicine, agriculture, and technology lowered Sri Lanka's death rate even
more than its birth rate had fallen, partly due to significant progress in the war against malaria.

This last fact is the lesson that Sri Lanka holds for the world today: Even if we succeed in lowering birth rates
around the world, progress in medical research and biotechnologies may well end up lowering our death rates
even more. Thus, while both trends each constitute one sort of good news, at the end of the day, when taken
together, our populations could end up growing even faster instead of more slowly.

Sri Lanka illustrates the need to look again at our demographic projections for the decades ahead. In those cases
where dramatically lower death rates have not been factored in, current projections, which are often presented
with an unduly optimistic cast, may turn out to be vastly incorrect. Just as Sri Lanka grew FASTER after three
decades of falling birth rates, something similar may be about to happen on a worldwide scale.

If a similar set of events takes place worldwide and affects generations now living, world population by 2100
could end up closer to 16 billion than to the 11 billion imagined by recent U.N. projections. And, if we are al-
ready close to or beyond Earth's long-term limits, each of these extra and unexpected billions increases the pos-
sibility of overshoot and collapse.

Life-extension
In this section we will see how medicine, life-extension, and molecular genetics may dramatically affect the
demographics of humanity's near future -- beginning with a tiny little roundworm known as Caenorhabditis ele-
gans. Recent studies of C. elegans (and other species) are generating provocative insights into the biology and
genetics of longevity and aging (e.g., see Kenyon, 2005; Asencio, et al., 2003, and Friedman and Johnson,
1988).

Some experiments, for example, have shown that dietary change in C. elegans can extend lifespan up to 60%
(Larsen and Clarke, 2002). Other recent laboratory studies have succeeded in multiplying lifespan six-fold (for
example, see Kenyon, 2005).
In a significant review article, Kenyon reports on six-fold extensions of lifespans in Caenorhabditis that have
already occurred, noting that in human terms, an equivalent extension would result in healthy, active 500-year-
olds (ibid).
Although the thought of human approaches certainly comes to mind, pending further
research, of course, those possibilities remain, at least for the time being, hypothetical.

It is interesting to envision, however, a human being born in the year 1600 who, at age 33, was alive to witness
the trial of Galileo in the Court of the Inquisition. If what has already been accomplished in Caenorhabditis had
occurred back then in humans, a person alive to witness such events in 1633 could still be alive today - and with
many, many decades of active and healthy life expectancy still remaining.

In addition, if anything like such a degree of life-extension in humans were to actually occur in humans, then
our populations might be expected to explode at obliterating rates - (unless of course, humankind were to reduce
its worldwide fertility rates to replacement-levels of just four-tenths of a child per woman.... per century).

Research like that mentioned by Kenyon in her review paper is also being conducted by major universities and the
National Institutes of Health (for instance, papers by Sinclair and Guarente, 2006; Larsen and Clarke, 2002; Jazwinski,
2000; and H.Y. Cohen, 2004). Other research labs are targeting a gene (sir2) and its protein (Sir2) that suggests that, the
higher the level of this protein, the longer the life-span for the organisms under study" (e.g., Guarente, et al., 2006, 2004,
2000). "We believe that these studies could lead to... a drug that... provides the benefits of calorie restriction without the
extreme difficulty of the regimen itself" (ibid).

Other Recent Advances with Far-reaching Implications

What might be the longevity, demographic, and mortality implications of todays genomics, genetically-modi-
fied organisms (GMOs), and similar medical and molecular advances? Recently, scientists have begun testing
sets of DNA-based BioBricks that, upon insertion into a living cell, may direct cells to carry out various active-
ties that are spelled-out by the scientists themselves. Though in its infancy, this capability already exists and is
under study at labs like those at Caltech, Lawrence Berkeley National Laboratory, Scripps Research Institute,
M.I.T., and Columbia (Gibbs, 2004).

What if other technologies also lower mortality rates in ways that are not currently incorporated in most popu-
lation projections? Researchers, for instance, have already generated genes that alternate between two stable
states, producing a modified bacterium that exhibits a rudimentary digital memory. As just one example, in
early 2003, sixteen top students at M.I.T. were able in one month to specify four genetic programs to make
groups of E. coli cells flash in unison as a demonstration of the new technologys potential (ibid).

Other insights arise from important research at the synthetic biology department at Lawrence Berkeley National
Laboratory where Jay Keasling and his team have engineered a large network of wormwood and yeast genes
into E. coli, enabling the bacterium to synthesize a chemical precursor to artemisinin an anti-malarial drug
that is currently too expensive for parts of the developing world that need it most (ibid).

In just three years, Keaslings lab increased yields by a factor of one million. And by amplifying the yields
another 25-to-50 times, Keasling anticipates that artemisinin-based medicines might be supplied to the Third
World for about one-tenth the current price. In a similar way, bioengineered bacteria may be used to produce
the cancer drug taxol and other medicines (ibid).
Thus, as our Sri Lanka discussion has shown, falling mortality has repeatedly undermined population project-
tions in the past, and current research suggests that the same thing may be about to happen once again
this time on a worldwide basis.

Aeronautics, DNA, and Caenorhabditis

Over the past one hundred years, humankind has followed a repeated pattern following new discoveries and tech-
nical advances. First there is an initial achievement or discovery that is quickly followed by rapid advances, pro-
liferation, and wide and novel applications. A good example of this is illustrated by the development of aeronaut-
tics: At Kitty Hawk, North Carolina (1903), Orville and Wilbur Wright flew a heavier-than-air vehicle for twelve
seconds and a distance of 120 feet.
Less than seven decades later, U.S. astronauts traveled to the moon, landed
on its surface, and returned safely to Earth again in just over one week.
Similar patterns have also characterized the development of computers, DNA technologies, communications,
and molecular biology each beginning with technical advances, followed by quick proliferation and progress-
ion to today's capabilities with breathtaking speed.

All of this suggests that today's advances in medicine, molecular genetics, and biotechnologies (including early
work like that seen with Caenorhabditis) may have far-reaching impacts on death rates and demographics in the
half-century just ahead. Some writers note that genetic engineers will certainly aim for greater longevity" and,
if such engineering proves even "...partly successful, it will create vast social and economic dislocations (E.O.
Wilson, 1998). And, as this article has argued, such advances may well cause humanity to further overshoot cal-
amitous demographic and ecological thresholds.

For now, however, think of the Wright brothers, who in 1903 flew, for the first time, a heavier-than-air vehicle for twelve
seconds and 120 feet. Consider next that just sixty-six years later the astronauts of Apollo 11 traveled to the moon and back in
a little over one week.

Which brings us back to 1993 and the initial doubling of lifespan in individual members of C. elegans. If we envision, beginning
in 1993, a life-extension progression similar to that seen in aeronautics then extending 1993 sixty-six years into the future
would bring us to 2059.

Even if a real possibility of a six-fold / 500 year life-extension for humans were to (or might actually) exist, the exact details and
the exact amount of lengthening are of less importance than the fact that: (1) such research is occurring, and (2) that some por-
tion of dramatic success may not just be possible, but may well be probable.

Thus, if even a fraction of that 500-year human equivalent were to emerge in the decades ahead (imagine, perhaps, a twenty or
a thirty year extension), then today's optimistic U.N. population projections that envision a stabilization of world population
or that envision a total medium-fertility world population of 11 or 12 billion around 2100 would go right out the window.

Historically, demographic projections often adopt an assumption of business as usual when modeling alternative
demographic futures. Today, such sweeping assumptions are no longer justified. Over and over again, we under-
estimate the impacts of dramatic and unexpected medical and technical advances and rapid implementation that
characterize modern science, research, and genomics. The benefits of each such advance tend to be realized rela-
tively quickly by immediately saving lives and reducing mortality (imagine anti-malarial applications, for exam-
ple).
At the same time, however, lengthy delays or lag-times occur when traditions, social customs, and fertility adjustments
do not occur for decades, if at all. If our early steps in understanding longevity, lifespan, and aging develop and prolif-
erate in a way analogous to the advances we have already seen in biotechnology, aeronautics, DNA, computers, and
communications, then precipitous declines in mortality could be on the horizon. If so, then today's demographic project-
tions may constitute serious underestimates of Earth's future populations and may lull us into an exceedingly dangerous
state of complacency and inaction.

Past projections have generated underestimates that have occurred again and again exactly as seen in our data from Sri
Lanka. We should appreciate, therefore, that dramatic or unexpected declines in mortality are capable of offsetting or
cancelling-out anticipated gains that we otherwise expect based on declining fertility.

Problematic Aspects of Transition Theory

Suppose that unexpected medical advances bring about a sudden reduction in mortality. In this circumstance, demo-
graphic theory envisions a period of demographic transition during which there is a time-delay before reductions in
fertility occur to reflect the reduced mortality and during this lag-time, populations skyrocket as births greatly exceed
the lowered death rates. Finally, however (perhaps after one or more generations), transition theory postulates (imagines,
hopes-for, wishes-for) a gradual (and sufficient) decline in fertility rates that slowly reduces them to levels commensurate
with mortality rates, and a population stabilizes.

Thus, too-many demographers commonly envision our time of soaring populations as a transition period during which
fertility rates have not yet caught up to our falling mortality rates. And they hope, imagine, and suppose that the transition
will complete itself any decade now. One worrisome problem is, however, that such hopes and expectations may well
be subverted by a nuanced and unexpected limitation within the theory.

How? Why? Because science, medicine, and technology repeatedly lower mortality rates not just once, but over and
over and over again so that we live in a perpetual state of transition. In other words, we repeatedly extend and postpone
the period of transition (with its skyrocketing populations) so that its completion never occurs or is repeatedly postponed.

In effect, each of our breakthroughs in medicine and longevity re-initiate the transition period, delaying its com-
pletion and extending its duration more and more- so that our falling fertility rates are never allowed to catch up.

As fertility rates slowly and gradually adjust to an initial mortality reduction, todays genetics, technologies, and medical
advances institute a second, third, fourth, and fifth mortality reduction in increasingly quick succession. As a result, fall-
ing fertility never catches up to the multiple new reductions in mortality and the interim stage of the transition (with its
period of soaring population) is never completed.
(It may be completed eventually, of course, but with each delay in the
transition, the completion is increasingly likely to occur as a collapse.)

What current theory does not fully articulate, therefore, is the role of science, technology, and medicine that are cur-
rently making reductions in death rates so quickly and so repeatedly that offsetting fertility reductions do not (or can-
not) occur in the short times available.

In effect, falling fertility is never able to completely catch-up and conclude the transition, because science and medi-
cine keep perpetually extending the transition over and over again. In the meantime, of course, our already exploding
populations continue to rocket to heights and totals that invite collapse.

And finally, the coup de grace of all this is that the emerging advances in longevity that we have already seen in
Caenorhabditis and other research seem set to perhaps amplify and worsen our current overshoot and carry us past
multiple thresholds and tipping points that should not be transgressed. Most strikingly, our ingenious advances in gen-
etics, molecular biology, and medicine are repeatedly and systematically reducing mortality rates (which we should pur-
sue, of course, and for which their discoverers should be greatly honored and rewarded).
The hoped-for transition, however, becomes perpetual and never ends because constant advances keep extending it be-
fore it can proceed to completion. Each such advance, therefore, acts to cancel out or neutralize a key expectation of the
demographic transition theory and repeatedly postpones completion of the expected transition so that it never occurs or
it never ends until our degree of overshoot is so great that complete collapse can no longer be avoided.

More problematic than we currently imagine

Realistic population projections need to incorporate at least one variant that contemplates the possibility that widespread
and nearly-instantaneous mortality declines may cancel out gains otherwise expected from falling fertility. And these
models should also incorporate assorted lag-times and delayed feedbacks that arise from decades of cultural fertility tra-
ditions and social customs.

If advances in medicine, molecular genetics, and longevity (e.g., Kenyon, 2005) develop and proliferate in ways
approximating those we have seen in computers, DNA technologies, communications, and the progress of aeronautics
from zero to moon landings in a span of less than seven decades, we could be in more severe trouble than we currently
imagine. In short, with maturation of advances that are currently underway, we may see falling death rates in the decades
ahead that were unimaginable just a few years ago.
Continuation of todays demographic tidal wave increasingly constitutes
the greatest single risk that our species has ever undertaken.

Excerpted from
What Every Citizen Should Know About Our Planet
Used with permission

Footnotes:
1. This graph, which is known as an S-curve,
depicts stabilizing outcomes imagined by
Demographic Transition Theory but such
outcomes are not automatic and do not
constitute some inviolate law of nature.

As one example, ecological release in a

population can occur, resulting in a pop-
ulation explosion, overshoot, and climb-and-collapse outcomes that result in massive die-offs, and/or 99% and/or
even-worse mass mortalities. Examples: (a) Population explosions of jackrabbits when too many coyotes are
killed; (b) Population explosions of sea urchins when too many sea otters are removed (and then the offshore
kelp beds begin to collapse in the face exploding pressures of feeding urchins); (c) Similarly, advances in health,
medicine, and antibiotics removed diseases that once held humankinds populations in check (and that was good,
of course), but traditional fertility rates did not adjust, birth rates remained unchanged or high, and humankinds
worldwide population explosion today has unfolded as a classical example of ecological release.

Note: In recent years, formal demographic presentations and journal articles often include graphs that depict worldwide population
growth as an s-curve. We note here that authors of such s-curve graphs typically do not begin their graphs until the 1960s or so, then
(having omitted data from 8000 BC to the present), they bring their graphs up to the present, and project several decades into the future
based upon their own hopes, wishes, guesstimates, and suppositions. Readers should consider, however, that in SCIENCE it is not
generally permissible to OMIT 9,900 years of population data in order to force-fit ones graph into a more wishful theoretical config-
uration. (When, however, all 10,000 years of world population numbers are included in such
a graph, the resulting graph is an extreme J-curve as shown in footnote 3 below.)
2. Note that sigmoid curves are not universal in real-
world conditions. Instead, multiple real-world populations
like this study of a reindeer population in Alaska (after
Scheffer, 1951) continue their exponential trajectories
and OVERSHOOT the Carrying Capacity of their envir--
onment, resulting in another classical population out-
come known as Climb-and-collapse.

In the Scheffer study summarized in this graph, more than

99% of the herd died in the herds collapse.

In addition, the peak reindeer population in the collapse depicted above occurred when the combined bodies of ALL the
reindeer on the island physically-occupied less than 2/1000ths of 1% of the island on which they lived.
in other words, in surroundings that visually
appeared to be ALMOST COMPLETELY EMPTY.

3. Lastly, note that the J-curve exhibited by our own graph (shown left)
is far more pronounced and far more extreme than the J-curve that
preceded reindeer herd collapse and die-off depicted above.

Previously, the two most famous J-curves in all of human history

were the two atomic detonations that destroyed Hiroshima and
Nagasaki at the close of World War II. Now, humankind, civili-
zation, and Earths razor-thin surface films of atmosphere, oceans,
and seas are living in an obliterating and global-scale J-curve that
is unfolding in the middle of the only planetary life-support ma-
chinery so far known to exist anywhere in the universe.

References, etc.

Bongaarts, 1994. Campbell, Reece, and Mitchell, 1999. Chapin, et al., 1997.
Cohen, 2002, 1995. Farrell, et al., 1994. Gallagher, et al., 1995. Hudson, 1989.
Myer, 1995. Revelle, 1976, 1974. UNDESA, 2004. Vitousek, et al., 1997.
Waggoner, 1994. Wilson, 20