Let $q$ be a prime power and define the bilinear form $$((x_1, x_2, x_3),(y_1, y_2, y_3)) \mapsto x_1 y_3^q + x_2 y_2^q +x_3 y_1^q$$ on the three-dimensional vector space $V$ over the finite field $F_{q^2}$ of $q^2$ elements.
Direct computation shows that the set of 1-dimensional isotropic subspaces of $V$ is $$\Omega = \{ \langle(1,0,0) \rangle \} \cup \{ \langle(\alpha,\beta,1) \rangle : \alpha+\alpha^q + \beta \beta^q=0\}.$$
It is known that $|\Omega|=q^3+1$ but I do not see how this is so.
Clearly, the problem reduces to showing that there are $q^3$ solutions $(\alpha,\beta)$ to $$\alpha+\alpha^q+\beta\beta^q =0$$ but I am unsure how to proceed.
For context, this claim is from the Unitary groups section of the seventh chapter in the book `Permutation groups' by Dixon and Mortimer. It provided no proof that the size of $\Omega$ is $q^3+1$ and the citations therein for further information on the unitary groups either do not have the proof or are not available to me.
Basic properties of the relative trace $$T:\Bbb{F}_{q^2}\to\Bbb{F}_q, T(x)=x+x^q$$ and the relative norm $$N:\Bbb{F}_{q^2}\to\Bbb{F}_q, N(x)=x\cdot x^q$$ give this quickly.
We can equally well use the fact $N(y)=0$ if and only if $y=0$, and otherwise $N$ is a $(q+1)$-to-$1$ mapping onto $\Bbb{F}_q^*$. Therefore for $q$ choices of $\alpha$ (namely those in the kernel of $T$) there is a single solution $\beta=0$ such that $N(\beta)=0=T(\alpha)$. On the other hand for any of the other $q^2-q$ choices of $\alpha$ there are $q+1$ solutions $y$ to $N(y)=-T(\alpha)$. Hence a total of $$q+(q^2-q)(q+1)=q^3$$ solutions.