How to estimate the future global population

[kthouz]

Hello everybody!
I have been reading about the evolution of population growth since 1950 till now. It had many factors such as development of technology, of medecine, ...
Now i want to calculate the future population growth let's say for example for 2050. In those readings i found also those estimations but my problem is what type of models to they use to predict it? For example i know the IPAT model that describe the impact of population on environment. As there exist that one, do there exist one that can predict the future population?

 

[denisv]

The population in 2050 will likely be somewhere between 0 and 50 billion. IMHO if you want to be more precise with any certainty you'll simply have to wait until 2050.

 

[kthouz]

Please, i just need mathematical proofs! Being there should be one of them but not a scientific one.

 

[jim mcnamara]

One of the reasons it is hard to model is because K - carrying capacity - can and has been changed by humans, mostly through technology.

google for Verhulst model or logistic difference model. Modern models are generally based on some sort of logistic growth curve.

 

[kthouz]

google for Verhulst model or logistic difference model. Modern models are generally based on some sort of logistic growth curve.

Thanks a lot! This replied to my question. Is it possible to combine this last model, i.e Verhulst model, with the IPAT model to predict the population that should live on Earth such that it is in equivalent with the environment!? Of course it is another question but if there are some pre-made models for this calculation, please let me know.

 

[jim mcnamara]

This http://www.arcytech.org/java/population/facts_math.html
seems to have a decent discussion of age-class distributions, predator-prey (or morbidity and death rates), emigration & immigration, density-dependant inhibition, fecundity - and so on.

age-class distributions and the age of onset of menses (controls fecundity) are the biggest determinants for modern human population growth in developed countries - period.

Example: a female having one child at age 15 has more impact on pop. growth rate than does a female who starts having children at age 28 and has three offspring. A population (all things being equal which they never are) made of females of type #1 has a higher Malthusian r than does a population of type #2.

 

[Moonbear]

I'm not sure if you're using fecundity correctly...unless your example wasn't meant to illustrate fecundity at all. Terms such as fertility and fecundity are frequently misused.

Fecundity is basically live birth rate per insemination attempt. You can't really compare the fecundity of the 15 y.o. and 28 y.o. in your example without knowing how many times each has had unprotected intercourse.

Fertility refers to the number of conceptions per insemination attempt, whether or not they lead to a live birth.

These terms both get tricky with humans, though, because we don't just mate once per ovulatory cycle. So, sometimes you can determine fertility rate as a ratio of conceptions to ovulatory cycles in which insemination was attempted, and fecundity as the ratio of live births to those same ovulatory cycles.

These aren't terribly useful numbers in predicting human population changes, though, as they refer more to potential for offspring production. More useful would be generation intervals (as your example illustrates between the 15 y.o. vs 28 y.o. having a first child) and actual numbers of offspring over a woman's reproductive lifespan.

The biggest difficulty, as has been pointed out in this thread already, is that humans can tinker with their own variables, making most models pretty useless. Will there be another medical breakthrough in the next 40 years that increases lifespan significantly? Will people's views on things like euthanasia change significantly enough to shorten the average lifespan? In talking about global population, what political influences will change in the next 40 years? Will infant, child and maternal mortality be significantly reduced in developing nations? It's all imprecise at best.

At least I can predict that immigration and emigration are not likely to be significant within the next 40 years...otherwise we'll have some really interesting S&D discussions.

 

[jim mcnamara]

I'm not sure if you're using fecundity correctly...unless your example wasn't meant to illustrate fecundity at all. Terms such as fertility and fecundity are frequently misused.

I disagree. Not that the terms are misused - but what fecundity means.

Please Reference Chapter 52 -
Campbell, N.A., Reece, J.B. et al. (2008). Biology.
8th edition. Benjamin/Cummings Publishing Company, Inc.

Fecundity, as defined by most texts in the field of population ecology,
is the live birth rate for females, period. Not corrected for any other factors,
sort of a max possible r as in N1=N0*e^rt. We take a very mammal-centric view.

It is not a measure of the potential number of oocytes or copluations or anyhting else.
It is the observed value.

Let's use humans.

Assume menses starts at age 15, ends at age 50; menstrual cycle
duration is 28 days; and our perfectly average woman never becomes pregnant;
one follicle matures per period. This gives:
eggs produced == (365.25 * 35)/28 ~ 456 follicles shed. Human do not have
450 offspring - so this measure is pointless. Rates of intercourse resulting
in fertilization are affected by all sorts of things - nutrition, for example.

This is the basic reason why Population Ecologists use fecundity to mean
birth rate as it was observed. Not follicle production, copulation rates or anything else.

Sorry for the late reposnse to Moonbear's post but this little post took me
30 minutes -- time which I don't have in large quantities anymore. :)