Springer International Publishing Switzerland 2014
Mimmo Iannelli and Andrea Pugliese An Introduction to Mathematical Population Dynamics UNITEXT 10.1007/978-3-319-03026-5_1
1. Malthus, Verhulst and all that
Apparently, no book has been so extensively discussed by people who seem not to have read it, as it happens with the Essay on the Principle of Population by Mr. Malthus.
Such a harsh note by the editor, against the comments received by the Essay , takes us back to the atmosphere of those times and to the discussions that the Principle of population sustained by Thomas Robert Malthus, caused and fed since the first edition of the book in 1798. Actually this Principle can be stated in a few words:
population, when unchecked, increases in a geometrical ratio; subsistence increases only in an arithmetical ratio.
Projected to the future, such a mechanism produces catastrophic scenarios that, rightly or wrongly, when the Essay was published, were the object of great discussions and quarrels (see and the introduction to that volume, in Italian, for an interesting presentation). In fact, the thesis presented by Malthus in his famous book is part of a wide discussion between the conservative views (supported by Malthus) and the optimistic ones of the Enlightenment philosophers (especially Godwin and Condorcet) who supported a Theory of Progress . Right or wrong in his conclusions, Malthus however deserves the merit of having pointed out, about two hundred years ago, the problems related to demographic expansion.
Since the publication of the Essay , the name of Malthus is associated to exponential growth and in fact the model we will present at the beginning of this chapter is named after him. Actually, in the context of the Essay , as in other demographic studies, the use of an exponential curve to fit experimental data is essentially empirical, while the modeling viewpoint that we introduce in aims at obtaining exponential growth as the consequence of a priori constitutive assumptions. This approach will be maintained throughout the chapter where different mechanisms regulating intraspecific competition and exogenous effect will be introduced and discussed.
1.1 1.1 A look at exemplary data
The empirical approach to data description can be conveniently illustrated by exemplary data from the demographic context. In fact in we present classical numbers of a human population, provided by census institutions. Actually, demographic data have received much attention for thousands of years since, as it is obvious, man has initially paid much attention to most important events like births, deaths, diseases Thus we can take advantage of rather accurate observations.
Table 1.1
U.S.A. population
year | individuals a | rate |
---|
1790 | 3,929 |
1800 | 5,308 | 0,035 |
1810 | 7,240 | 0,036 |
1820 | 9,638 | 0,033 |
1830 | 12,861 | 0,033 |
1840 | 17,064 | 0,033 |
1850 | 23,192 | 0,036 |
1860 | 31,443 | 0,036 |
1870 | 38,558 | 0,023 |
1880 | 50,189 | 0,030 |
1890 | 62,980 | 0,025 |
1900 | 76,212 | 0,021 |
1910 | 92,228 | 0,021 |
1920 | 106,021 | 0,015 |
1930 | 123,203 | 0,016 |
1940 | 132,165 | 0,007 |
1950 | 151,326 | 0,015 |
1960 | 179,323 | 0,019 |
1970 | 203,302 | 0,013 |
1980 | 226,456 | 0,011 |
1990 | 255,712 | 0,010 |
2000 | 285,003 | 0,013 |
a millions
In data are reported by 10-years steps and the third column shows the growth rate of the population, i.e. the yearly relative change of the population. Namely, if N i denotes the number of individuals in the year i and h the length of the time interval (in the table h = 10 years ), then the growth rate can be defined through the formula
When the growth rate is approximately constant (as in the first part of , but also for the years 19001930) data can be represented through an exponential curve. In fact, we have the equation
and, iterating from i = 0 we get
and we see that the population undergoes geometrical increase or decrease if > 0 or < 0 respectively.
If now we suppose that holds for h small enough, the previous formula can be written using the exponential function as
and, adopting a time-continuous framework, for a given fixed time t the number of individuals N ( t ) is approximately given by
where we have approximated
and the approximation is better as h is smaller.
A procedure to fit data with an exponential curve, estimating the best value of the parameter r , is to apply the least square method , fitting a line to the data of is transformed into the line
with slope r and intercept y o = ln N 0. Then, the least square procedure applied to the data of provides the following values