Math 5110/6830 Homework 3.1 Solutions 1. (a) Mt+1 = Mt − g(Mt )Mt + S = Mt − Mtp Mt + S K p + Mtp with K = 2 and S = 1 gives Mt+1 = Mt − Mtp Mt + 1. 2p + Mtp A fixed point for this equation therefore satisfies M∗ = M∗ − M ∗p M ∗ + 1. 2p + M ∗ p With M ∗ = 2, ? 2 =2− ? 2 =3− 2p 2p ·2+1 + 2p 1 ·2 2 X 2 = 2. Because 2p2+2p = 21 is true independently of p, we know that M ∗ = 2 is an equilibrium point for any value of p. p (b) Recall that a fixed point M ∗ = f (M ∗ ) is stable when |f ′ (M ∗ )| < 1, where f (Mt ) = Mt − Mtp Mtp+1 M + S = M − + 1. t t K p + Mtp 2p + Mtp Differentiating, we obtain f ′ (Mt ) = 1 − (p + 1)Mtp (2p + Mtp ) − Mtp+1 (pMtp−1 ) . (2p + Mtp )2 For M ∗ = 2, f ′ (2) = 1 − =1− =1− =1− =1− =1− If 2−p 4 < 1, then (p + 1)2p (2p + 2p ) − 2p+1 (p · 2p−1 ) (2p + 2p )2 p p+1 (p + 1)2 · 2 − p · 22p (2p+1 )2 (p + 1)22p+1 − p · 22p 22p+2 2p+1 2 (p + 1 − p · 2−1 ) 22p+1 · 2 p + 1 − 2p 2 2−p 1 p+2 4 − (p + 2) = = 2 2 4 4 1 ⇒ −4 < 2−p < 4 2−p< ⇒ −6 < −p < ⇒ 6> −1 < 1 4 2 p > −2. M∗ Thus when −2 < p < 6, the fixed point = 2 is stable. It is unstable when p < −2 or p > 6. The solution oscillates toward the equilibrium when two conditions are satisfied: when −2 < p < 6 and when f ′ (2) < 0, i.e. when p > 2. Thus, stable oscillations occur when 2 < p < 6. (c) Choose p = 7 and M0 = 3. 4 Mt+1=f(Mt) Mt+1=Mt 3.5 cobweb 3 2.5 2 1.5 1 0.5 0 0 0.5 1 1.5 2 2.5 3 3.5 4 2. (a) Given an+1 = an + k(M − an )(an − m) with M = 5000, m = 100 and k = 0.0001, we have an+1 = an + 0.0001(5000 − an )(an − 100). A fixed point a∗ satisfies a∗ = a∗ + 0.0001(5000 − a∗ )(a∗ − 100) ⇒ 0 = 0.0001(5000 − a∗ )(a∗ − 100) So, a∗ = 5000 and a∗ = 100 are the fixed points of this model. Stability of a fixed point a∗ requires that |f ′ (a∗ )| < 1, where f (an ) = an + 0.0001(5000 − an )(an − 100). Differentiation gives f ′ (an ) = 1 + 0.0001[−1(an − 100) + (5000 − an )] = 1 + 0.0001(5100 − 2an ) = 1.51 − 0.0002an When a∗ = 5000, f ′ (5000) = 1.51−1 = 0.51. Since |f ′ (5000)| < 1, this fixed point is stable. 2 When a∗ = 100, f ′ (100) = 1.51 − 0.02 = 1.49. Since |f ′ (100)| > 1, this fixed point is unstable. Cobweb and sketch of solution for a0 = 50 250 50 an+1=f(an) 200 an+1=an cobweb 150 0 100 −50 n 50 a 0 −50 −100 −100 −150 −150 −200 −250 −100 −50 0 50 100 150 −200 200 0 0.5 1 1.5 2 n 2.5 3 3.5 4 For a0 = 50, the population of whales decreases without bound, which is not biologically possible. Cobweb and sketch of solution for a0 = 200 5000 6000 an+1=f(an) 4500 an+1=an 5000 cobweb 4000 3500 4000 n 3000 a 3000 2500 2000 2000 1500 1000 1000 500 0 0 1000 2000 3000 4000 5000 6000 0 0 5 10 15 20 25 n For a0 = 200, the population of whales will increase to the stable equilibrium at a∗ = 5000. (b) Observing the dynamics of the whale population over 20 years for varying a0 , notice that the model breaks down for a0 < 100, which gives rise to negative populations. With a0 > 100, however, the model stabilizes. As a0 increases, the time it takes to reach the equilibrium value of a∗ = 5000 gets shorter and shorter. 3 5000 4000 an 3000 2000 1000 0 −1000 0 2 4 6 8 10 n (years) 12 14 16 18 20 Matlab Code % k M m Define parameters in the model. = .0001; = 5000; = 100; N = 20; % total number of years % Define the given equation as a function with x as the independent variable. f = @(x) x + k.*(M-x).*(x-m); year = 0:N; % This for-loop will set up the initial value of a. for j=20:40:1060 a(1) = j; % This for-loop generates the solution for the given a0. for n=1:N % Plug the previous value of a_n into the function to get the next % value. a(n+1) = f(a(n)); end % Open figure 1 figure(1) % Plot the solution and don’t erase it from the figure. "year" will be % the x-values, "a" will be the y-values. plot(year,a) hold all end 4 % Specify the ranges for x- and y-values in the form [xmin xmax ymin ymax]. axis([0 N -1000 M]) %Label axes xlabel(’n (years)’) ylabel(’a_n’) Note: Any single quotations in the code will not copy correctly into Matlab (a LA TE X issue), so you will need to retype them after pasting the code into the Editor. Also, in order to erase the figure after you run this code, but before running another code that requires plotting, type ‘clf’ (without quotations) in the command window. 5 Homework 3.2 Solutions 1. (a) As can be seen in the pictures below, the form for g(xn ) used in the Beverton-Holt model is quite different from that used in the logistic model. For most values of r, g(xn ) > 0, and it decreases nonlinearly to zero for increasing xn . When r < 1, however, three things happen: K K ; (2) g is undefined for xn = 1−r ; and (3) g is (1) g is positive and increasing for xn > 1−r K negative and increasing for xn < 1−r . K determines how quickly g decreases (for r > 1) or increases (for r < 1); smaller values of K produce faster changes in g. r = 0.5 4 K = 1000 3 2 g(xn) 1 0 −1 −2 −3 −4 0 500 1000 1500 2000 2500 xn 3000 3500 4000 4500 5000 r = 10 r = 100 4 4 K=1 K = 10 K = 1000 K = 5000 3 3 2.5 2.5 2 1.5 2 1.5 1 1 0.5 0.5 0 K=1 K = 10 K = 1000 K = 5000 3.5 g(xn) n g(x ) 3.5 0 500 1000 1500 2000 2500 x 3000 3500 4000 4500 0 5000 0 n 500 1000 1500 2000 2500 x 3000 3500 4000 4500 5000 n (b) Fixed points satisfy the equation x∗ = x∗ r , ∗ 1 + r−1 K x to which x∗ = 0 and x∗ = K are solutions. Stability: Let f (xn ) = g(xn )xn . Stability of x∗ is guaranteed when |f ′ (x∗ )| < 1. Differentiating f (xn ) gives r−1 r 1 + r−1 r ′ K xn − rxn K = f (xn ) = 2 . 2 r−1 1 + r−1 x 1 + x n n K K For x˜ = 0: f ′ (0) = r, and so this fixed point is stable for |r| < 1. Since r must be positive, it follows that x∗ = 0 is stable for 0 < r < 1 and unstable for r > 1. 6 For x˜ = K: f ′ (K) = r 1+ 2 r−1 K K r (1 + r − 1)2 r = 2 r 1 = . r = This fixed point is stable for 1r < 1, and since r > 0, this means that x∗ = K is stable for r > 1 and is unstable for 0 < r < 1. (c) Cobwebbing for r > 1 (left) and 0 < r < 1 (right) r = 2.5 r = 0.5 xn+1=f(xn) 5000 xn+1=xn 4500 xn+1=f(xn) 5000 xn+1=xn 4500 cobweb cobweb 4000 4000 3500 3500 3000 3000 2500 2500 2000 2000 1500 1500 1000 1000 500 0 500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 5000 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 (d) Solutions for r > 1 (left) and 0 < r < 1 (right) 2500 4500 4000 2000 3500 n 1500 2500 x x n 3000 2000 1000 1500 1000 500 500 0 0 2 4 6 8 n 10 12 14 16 0 0 2 4 6 8 n 10 12 14 16 As predicted in part (b), r > 1 leads to stability of x∗ = K and instability of x∗ = 0, while 0 < r < 1 leads to the opposite result. 2. (a) Comparison to the logistic model: The Ricker model doesn’t become negative for any value of xn and decreases nonlinearly. Comparison to the Beverton-Holt model: As in the BH model, the expression for g(xn ) in the Ricker model is nonlinear. Also, for r > 1, the model decays nonlinearly to zero and is 7 always positive. Larger values of K cause g to decrease more slowly, as in the BH model. In contrast to the BH model, the Ricker model decreases to zero for any value of r, not just those greater than 1. Another difference is that g is extremely sensitive to changes in r, as smaller increases in r lead to significantly greater changes in g. r = 0.5 1.8 K = 10 K = 100 K = 1000 K = 5000 1.6 1.4 g(xn) 1.2 1 0.8 0.6 0.4 0.2 0 0 500 1000 1500 2000 2500 xn 3000 3500 4000 4500 5000 4 r=5 150 2.5 K = 10 K = 100 K = 1000 K = 5000 r = 10 x 10 K = 10 K = 100 K = 1000 K = 5000 2 100 g(xn) g(xn) 1.5 1 50 0.5 0 0 500 1000 1500 2000 2500 x 3000 3500 4000 4500 5000 0 0 500 1000 1500 n 2000 2500 x 3000 3500 4000 4500 5000 n (b) Fixed points satisfy the equation x∗ x = x exp r 1 − . K ∗ ∗ This leads to the following: x∗ 0 = x∗ 1 − exp r 1 − K x∗ = 0 x∗ and 1 = exp r 1 − K ⇒ ⇒ ⇒ x∗ 0=r 1− K x∗ = K . (c) Let f (xn ) = g(xn )xn . Stability of x∗ is guaranteed when |f ′ (x∗ )| < 1. Differentiating f (xn ) gives h h xn i xn i r f ′ (xn ) = exp r(1 − ) + xn exp r(1 − ) ·− K K K h r xn i ) · 1 − xn = exp r(1 − K K ˜ ′ r r For x = 0: f (0) = e , which is stable when |e | < 1. Since er > 0 for all r, |er | = er , and so x∗ = 0 is stable when r < 0. Since we’re given that r > 0, this fixed point is always unstable. 8 For x˜ = K: f ′ (K) = 1 − r, so we require |1 − r| < 1 for stability. This means ⇒ −1 < 1 − r < 1 −2 < 0 ⇒ −r < 2> r > 0. Therefore, x∗ = K is stable provided 0 < r < 2 and is unstable whenever r > 2. (d) Cobweb and sketch of solution for r = 1.8 6000 xn+1=f(xn) 6000 xn+1=xn cobweb 5000 5000 4000 x n 4000 3000 3000 2000 2000 1000 1000 0 0 1000 2000 3000 4000 5000 0 6000 0 2 4 6 8 10 12 14 16 18 20 12 14 16 18 20 n Cobweb and sketch of solution for r = 0.4 7000 xn+1=f(xn) 4500 xn+1=xn 6000 cobweb 4000 5000 3500 3000 x n 4000 3000 2500 2000 1500 2000 1000 1000 500 0 0 1000 2000 3000 4000 5000 6000 7000 0 0 2 4 6 8 10 n 9