1
$\begingroup$

What is the area of the triangle having $z_1$, $z_2$ and $z_3$ as vertices in Argand plane? Is it $$\frac{-1}{4i}[z_1(z_2^* - z_3^*)-z_1^*(z_2-z_3)+{z_2(z_3^*)-z_3(z_2^*)}]$$ where $w^*$ denotes the complex conjugate?

$\endgroup$
1
$\begingroup$

Let $z_j = x_j + iy_j$, $j = 1, 2, 3$. The area of the triangle is given by \begin{align*} \frac{1}{2} \begin{vmatrix} 1 & x_1 & y_1 \\ 1 & x_2 & y_2 \\ 1 & x_3 & y_3 \end{vmatrix}&= \frac{1}{2} \begin{vmatrix} 1 & x_1+iy_1 & y_1 \\ 1 & x_2+iy_2 & y_2 \\ 1 & x_3+iy_3 & y_3 \end{vmatrix}\\ &= \frac{1}{4i} \begin{vmatrix} 1 & z_1 & z_1-z_1^*\\ 1 & z_2 & z_2-z_2^* \\ 1 & z_3 & z_3-z_3^* \end{vmatrix}\\ \end{align*} Now expand via first column to get the required expression.

$\endgroup$
1
$\begingroup$

If ${\bf 0}=(0,0)$, ${\bf z}_1=(x_1,y_1)$, and ${\bf z}_2=(x_2,y_2)$ are the vertices of a triangle $\triangle$ in the $(x,y)$-plane then the signed area of $\triangle$ is given by $$\alpha(\triangle)={1\over2}(x_1y_2-x_2y_1)\ ,\tag{1}$$ as learned in high school analytic geometry. In the complex world we write $0$, $z_1$, $z_2$ for the three vertices, whereby the real coordinates $(x,y)$ are related to the corresponding $z$ via $$z:=x+iy,\quad x={z+\bar z\over 2},\quad y={z-\bar z\over 2i}\ .$$ Plugging this into $(1)$ we obtain $$\alpha(\triangle)={1\over8i}\bigl((z_1+\bar z_1)(z_2-\bar z_2)-(z_2+\bar z_2)(z_1-\bar z_1)\bigr)={1\over4i}(\bar z_1z_2-z_1\bar z_2)\ ,$$ which then can be rewritten as $$\alpha(\triangle)={1\over2}{\rm Im}(\bar z_1z_2)\ .$$ I leave it to you to get the formula for a triangle with vertices $z_0$, $z_1$, $z_2$ from this.

$\endgroup$
0
$\begingroup$

You can find the side lengths as $|z_1- z_2|, |z_2- z_3|, |z_3- z_1|$ and then use Heron's formula.

$\endgroup$
0
$\begingroup$

I assume you mean the oriented area of that triangle (since the number changes sign when you swap $z_2$ and $z_3$).

In order to verify it, allow me to write your formula as $$g(z_1,z_2,z_3)=\mathfrak{Im}\frac{\overline{z_1}(z_2-z_3)+z_3\overline{z_2}}{2}$$ It holds $$g(z_1+t,z_2+t,z_3+t)-g(z_1,z_2,z_3)= \mathfrak{Im}\frac{2\mathfrak{Re}(z_2\bar t)+\lvert t\rvert^2}{2}=0$$

So it is translation-invariant (hence, we need verify it only when $z_1=0$).

But that case is much easier.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.