The population $P(t)$ of a biological species can be modelled by $$ \frac{dP}{dt}=P(\beta - \delta P) $$ subject to $P(0)=P_0$ where $\beta$ is the birth rate and $\delta$ is the death rate. Assuming $P>0$, find the population $P(t)$ for three cases:
- $P_0<\hat{P}$
- $P_0=\hat{P}$
- $P_0>\hat{P}$
where $\hat{P}=^{\beta}/_{\delta}$
So far, I have:
We see the equations is separable and so $$ \int\frac{dP}{P(\beta - \delta P)}=\int dt $$ then in partial fractions we have $$ \int\frac{dP}{\beta P}+\int\frac{\delta}{\beta (\beta - \delta P)}dP=\int dt $$ integrating both sides $$ \frac{1}{\beta}\ln P+\frac{\delta}{\beta}\ln(\beta - \delta P)=t + k $$ $$ \ln P+\delta\ln(\beta - \delta P)=\beta t + k $$ $$ P\cdot(\beta - \delta P)^\delta=ke^{\beta t} $$ so at $t=0$ we have $$ P_0\cdot(\beta - \delta P_0)^\delta=k $$ hence $$ P\cdot(\beta - \delta P)^\delta=P_0\cdot(\beta - \delta P_0)^\delta\cdot e^{\beta t} $$
I don't really see how to apply the three cases to this equation; I mean it seems that $P_0=\hat{P}$ would mean $P(t)$ is identically 0 but I don't think this makes any sense in terms of the original problem. So my reasoning must be wrong..? And for the other two cases I don't see how I would express $P(t)$ generally.