BIO 357 Evolutionary Ecology Fall 2000 Handout 1 Population Modeling First, we would like to know how many individuals are in each age class of a population. The simplest model assumes non-overlapping generations (and therefore, no age class structure). 1) 2) Nt+1 = Nt Number at time t+1 is equal to number at time t multiplied by the finite rate of increase, , a constant. Nt+2 = Nt+1 = ( Nt) = 2 Nt We can use this to find the number of individuals at any time (and eventually we will be able to use this for age classes). Nt+x = x Nt Generalization of the above. To make this age class specific, we need to add the probability of surviving from the previous age class as well as the new young being produced. Nt+1 = p Nt + Nt where p = probability of survival *A Leslie Matrix can simplify all of this! Use this diagram to help you visualize the age structure of a population. P0 P1 m3 m2 m1 P2 m1 P3 = 0 m2 m3 N0 N1 N2 N3 age class 0 age class 1 age class 2 age class 3 Px = lx+1/lx We must assume that mortality precedes reproduction. N0(t+1) = N0(t) P0 m1 + N1(t) P1 m2 + N2(t) P2 m3 + N3(t) P3 m4 N1(t+1) = N0(t) P0 N2(t+1) = N3(t+1) = N1(t) P1 N2(t) P2 The above equations are taken directly from the diagram. These equations make up the basis of the Leslie Matrix when rewritten: N0(t+1) P0m1 N1(t+1) P1m2 P2m3 0 N0(t) P0 0 0 0 N2(t+1) 0 P1 0 0 N2(t) N3(t+1) 0 0 P2 0 N3(t) = Vector = N1(t) Vector of Variables Leslie Matrix (constant) To make the notation a little less cumbersome, let: N0(t+1) P0m1 N1(t+1) P0 P1m2 P2m3 0 0 0 N0(t) 0 = Nt+1 N1(t) =L = Nt N2(t+1) 0 P1 0 0 N2(t) N3(t+1) 0 0 P2 0 N3(t) Now, as before, let’s look at time t + 1 and time t + 2 to find a useful generalization. 3) = L N N t+1 N t+2 = Note the similarity to equation 1 t L N t+1 = L ( L N t) 4) N = L2 N t Lx N t t+x = Generalization – note the similarity to equation 2 Using this information, we can predict the number of individuals in all age classes at any time in the future. Equation 4 can be used to find the population trajectory for all age classes. Note, the size of n the whole population is the sum of all of the age classes ( Nx (t ) ). x 0 If the population is not already at its stable age structure, it will converge on this stable structure through a series of damped oscillations. The stable age structure is reached when the proportion of the population in each of the age classes stays constant over time. This will occur whether the population size is increasing (a), decreasing (b), or not changing (c). N0 N1 N2 N0 N1 ln Nt N0 N2 N1 N2 (a) t (b) (c) Matrix Algebra Here are some basic rules of matrix algebra to aid you in using the Leslie Matrix. Matrix – vector multiplication a b c d e f g h i j k l 3 3 matrix aj + bk + cl dj + ek + fl gj + hk + il = 1 3 vector 1 3 vector Scalar –vector multiplication a d b c = da db dc Matrix – matrix multiplication a b c d e f g h i j k l m n o p q r = (aj + bm + cp) (ak + bn + cq) (al + bo + cr) (dj + em + fp) (dk + en + fq) (dl + eo + fr) (gj + hm + ip) (gk + hn + iq) (gl + ho + ir) Notice that with matrix-matrix multiplication: A B B A , unless A = B Relationship between r and Both r (the intrinsic rate of increase) and ( the finite rate of increase) correspond to the growth rate of a population. The primary difference between them is that r is used in differential equations whereas is used in discrete equations. Nt+x = x Nt dN/dt = rN solve Nt+x = N0 er(t+x) set t = 0 and x = 1 N1 = N0 N1 = N0 er So, N1/N0 = er = r = ln So, the intrinsic rate of increase of a Leslie Matrix (r) is the natural log of the finite rate of increase () which is, in turn, the dominant eigenvalue of the Leslie Matrix. An important side note The proportion of a population in age class j is given as cj = l j e rj n n l x e rx and cj = 1 j 1 x 0 The Euler Equation Assuming that we have a stable age distribution, we can use the Euler equation to find r. This process requires tedious iteration and therefore, you will not be required to do it. Derivation of the Euler equation n N0(t) = N x (t ) m x Observation 2) Nx(t) = N0(t-x)lx Observation 3) N0(t) = N 0(t x ) l x m x Combine equations 1 and 2 4) N0(t) = N0(0)ert Observation 5) N0(t-x) = N0(t)e-rx Observation 6) N0(t) = N 0(t ) e rxl x mx 1) x 0 n x 0 n Combine equations 3 and 5 x 0 7) n N0(t) = N0(t) e rxl x mx Remove constant from summation x 0 n 8) 1 = e rxl x mx x 0 Divide both sides by (N0(t)), to get the Euler equation