I'd convert your positions to Carstesian coordinates, and then scale these positions down to the unit sphere. This can be achieved using a function
$$f(\lambda,\phi,h)=
\begin{pmatrix}
\tfrac1a\cdot\bigl(N(\phi)+h\bigr)\cos\phi\cos\lambda \\
\tfrac1a\cdot\bigl(N(\phi)+h\bigr)\cos\phi\sin\lambda \\
\tfrac1b\cdot\bigl(\frac{b^2}{a^2}N(\phi)+h\bigr)\sin\phi \\
\end{pmatrix}
\quad\text{with }
N(\phi)=\frac{a^2}{\sqrt{a^2\cos^2\phi+b^2\sin^2\phi}}$$
where $a$ and $b$ are the equatorial resp. polar radius of the earth. So plugging in your numbers you'd get
\begin{align*}
h=f(49.836573°, 24.083819°, 27\,\mathrm m)&=\begin{pmatrix}
0.58915599 \\
0.69807533 \\
0.40693307\end{pmatrix}
\\
s=f(-0.04°, 100.51°, 35779.65\,\mathrm{km})&=\begin{pmatrix}
-1.206 \\
0.001 \\
6.517\end{pmatrix}
\end{align*}
Now you want to know whether the line connecting $h$ to $s$ is above or below the horizon. To do that, compute the dot product between $h$ and $s-h$. That dot product is proportional to the sine of the elevation.
$$e=\arcsin
\frac{\langle h,s-h\rangle}{\lVert h\rVert\cdot\lVert s-h\rVert}
= 8.5°$$
So yes, the sattelite should be a bit over $8°$ above the horizon. The measurement was taken out on the sphere, and correcting for the ellipsoid shape of the earth may change that slightly, but I doubt that effect will be noticable compared to other inaccuracies.