[web] Fix LaTeX formulae with escaped _.

Also some trailing whitespaces.

web/content/docs/benchmarks/hydro-component/HC_ogs6-vs-ogs5.pandoc

@@ -23,7 +23,7 @@ We extended the setup to show mass transport in the heterogeneous medium for tes
 
 ## Problem description
 
-The setups are steady-state for flow, with an extent of a $100$ m x $100$ m horizontal plane for the 2D setup and a $100$ m x $100$ m x $50$ m cube for the 3D setup. Mesh elements have side lengths of $1$ m. The initial conditions are hydrostatic and concentration $c=0$. The boundary conditions are translated into equivalent hydrostatic pressure values from hydraulic heads $h\_{left}=10$ m and $h\_{right}=9$ m and for concentration $c\_{left}=1$, $c\_{right}=0$ for left and right sides, respectively. All other sides are defined as no-flow (Zero-Neumann).
+The setups are steady-state for flow, with an extent of a $100$ m x $100$ m horizontal plane for the 2D setup and a $100$ m x $100$ m x $50$ m cube for the 3D setup. Mesh elements have side lengths of $1$ m. The initial conditions are hydrostatic and concentration $c=0$. The boundary conditions are translated into equivalent hydrostatic pressure values from hydraulic heads $h_{left}=10$ m and $h_{right}=9$ m and for concentration $c_{left}=1$, $c_{right}=0$ for left and right sides, respectively. All other sides are defined as no-flow (Zero-Neumann).
 
 Porosity is $0.01$, specific storage is $0$, fluid density is $1000$ kg$\cdot$m$^3$, dynamic viscosity is $10^{-3}$ Pa$\cdot$s, molecular diffusion coefficient is $2\cdot 10^{-9}$ m$\cdot$s$^{-2}$, dispersivities are longitudinal $1$ m and transverse $0.1$ m. The heterogeneous parameter fields of intrinsic permeability are shown in the figures below; creation of the tensor field is documented [here](https://github.com/ufz/ogs-utils/tree/master/post/merge-scalar-data-arrays).
 
@@ -42,7 +42,7 @@ The mass transport simulation results (figures below) show an expected heterogen
 {{< img src="../heterogeneous/concentration_2d.png" title="Concentration distribution at simulation time $1e8$ s for the 2D setup.">}}
 {{< img src="../heterogeneous/concentration_3d.png" title="Concentration distribution at simulation time $1e8$ s for the 3D setup.">}}
 
-[The project files for the 2D setup are here.]({{< data-url "Parabolic/ComponentTransport/heterogeneous/ogs5_H_2D/ogs5_H_2d.prj" >}})  
+[The project files for the 2D setup are here.]({{< data-url "Parabolic/ComponentTransport/heterogeneous/ogs5_H_2D/ogs5_H_2d.prj" >}})
 [The project files for the 3D setup are here.]({{< data-url "Parabolic/ComponentTransport/heterogeneous/ogs5_H_3D/ogs5_H_3d.prj" >}})
 
 

web/content/docs/benchmarks/python-bc/laplace-equation/python-laplace-eq.pandoc

@@ -40,22 +40,22 @@ The weak form is derived as usual by multiplying with a test function $v$ and
 integrating over the domain $\Omega$:
 $$
 \begin{equation}
-- \int\_{\Omega} v \mathop{\mathrm{div}} (a \mathop{\mathrm{grad}} u) \, \mathrm{d}\Omega = 0
+- \int_{\Omega} v \mathop{\mathrm{div}} (a \mathop{\mathrm{grad}} u) \, \mathrm{d}\Omega = 0
 \,,
 \end{equation}
 $$
 which can be transformed further to
 $$
 \begin{equation}
-\int\_{\Omega} a \mathop{\mathrm{grad}} v \cdot \mathop{\mathrm{grad}} u \, \mathrm{d}\Omega = \int\_{\Omega} \mathop{\mathrm{div}} (v a \mathop{\mathrm{grad}} u) \, \mathrm{d}\Omega = \int\_{\Gamma\_{\mathrm{N}}} v a \mathop{\mathrm{grad}} u \cdot n \, \mathrm{d}\Gamma \,,
+\int_{\Omega} a \mathop{\mathrm{grad}} v \cdot \mathop{\mathrm{grad}} u \, \mathrm{d}\Omega = \int_{\Omega} \mathop{\mathrm{div}} (v a \mathop{\mathrm{grad}} u) \, \mathrm{d}\Omega = \int_{\Gamma_{\mathrm{N}}} v a \mathop{\mathrm{grad}} u \cdot n \, \mathrm{d}\Gamma \,,
 \end{equation}
 $$
 where in the second equality Gauss's theorem has been applied.
-As usual, the domain boundary $\partial\Omega = \Gamma\_{\mathrm{D}} \cup \Gamma\_{\mathrm{N}}$ is subdivided
+As usual, the domain boundary $\partial\Omega = \Gamma_{\mathrm{D}} \cup \Gamma_{\mathrm{N}}$ is subdivided
 into the dirichlet and the Neumann boundary and $v$ vanishes on
-$\Gamma\_{\mathrm{D}}$.
+$\Gamma_{\mathrm{D}}$.
 The r.h.s. of the above equation is the total flux associated with $u$ flowing
-**into** the domain $\Omega$ through $\Gamma\_{\mathrm{N}}$:
+**into** the domain $\Omega$ through $\Gamma_{\mathrm{N}}$:
 $-a \mathop{\mathrm{grad}} u$ is the flux density and $-n$ is the inwards directed surface
 normal.
 
@@ -77,7 +77,7 @@ solves Laplace's equation inside $\Omega$ for any $b$.
 In this example we set $b = \tfrac 23 \pi$.
 
 As boundary conditions we apply Dirichlet BCs at the top, left and bottom of the
-domain with values from $u(x,y)|\_{\Gamma\_{\mathrm{D}}}$.
+domain with values from $u(x,y)|_{\Gamma_{\mathrm{D}}}$.
 On the right boundary of the domain a Neumann BC is applied.
 There $n = (1, 0)$, which implies that $a \mathop{\mathrm{grad}} u \cdot n
 = a \, \partial u / \partial x$.

web/content/docs/benchmarks/richards-flow/richards-component-transport.pandoc

@@ -28,13 +28,13 @@ We use the Padilla et al. (1999) problem which features a series of saturated an
 
 ### Model setup
 
-We use a 1D domain with $0 < x < 0.25 m$, and a spatial resolution of $1.25 mm$ with a total of 200 elements. Top boundary conditions are concentration $c = 1$ (as Dirichlet) for both setups, and a specific flux $q\_{NaCl1} = 2.12789E-05$ and $q\_{NaCl6} = 6.46004E-06$ (as Neumann) for saturated and unsaturated cases, respectively. Bottom boundary conditions are a free exit boundary for mass transport, and Dirichlet pressure values are chosen so that the saturation in the column follows the respective values according to table 1 in Padilla et al. (1999) as $p\_{NaCl1} = 0 Pa$ and $p\_{NaCl6} = -4800 Pa$.
+We use a 1D domain with $0 < x < 0.25 m$, and a spatial resolution of $1.25 mm$ with a total of 200 elements. Top boundary conditions are concentration $c = 1$ (as Dirichlet) for both setups, and a specific flux $q_{NaCl1} = 2.12789E-05$ and $q_{NaCl6} = 6.46004E-06$ (as Neumann) for saturated and unsaturated cases, respectively. Bottom boundary conditions are a free exit boundary for mass transport, and Dirichlet pressure values are chosen so that the saturation in the column follows the respective values according to table 1 in Padilla et al. (1999) as $p_{NaCl1} = 0 Pa$ and $p_{NaCl6} = -4800 Pa$.
 
-Saturated intrinsic permeability is calculated from given flux and pressure gradient information as $\kappa = 1.174E-10 m^2 $; total porosity is $\theta = 0.45$; van-Genuchten-values are $m = 0.789$ (from $n = 4.74$ via $m = 1-1/n$), bubbling pressure $p_d = 3633.33 Pa$ (from $\alpha = 0.027 cm^{-1}$ via $p_c = \rho \cdot g / \alpha$), residual saturation $s_r = 0.1689$. Molecular diffusion coefficient was set to $D_m = 1e-9 m^2/s$; dispersivities were chosen from table 2 in Padilla et al. (1999) and set to $\alpha\_{NaCl1} = 3.4173E-04 m$ and $\alpha\_{NaCl6} = 4.6916E-03 m$ for saturated and unsaturated conditions, respectively.
+Saturated intrinsic permeability is calculated from given flux and pressure gradient information as $\kappa = 1.174E-10 m^2 $; total porosity is $\theta = 0.45$; van-Genuchten-values are $m = 0.789$ (from $n = 4.74$ via $m = 1-1/n$), bubbling pressure $p_d = 3633.33 Pa$ (from $\alpha = 0.027 cm^{-1}$ via $p_c = \rho \cdot g / \alpha$), residual saturation $s_r = 0.1689$. Molecular diffusion coefficient was set to $D_m = 1e-9 m^2/s$; dispersivities were chosen from table 2 in Padilla et al. (1999) and set to $\alpha_{NaCl1} = 3.4173E-04 m$ and $\alpha_{NaCl6} = 4.6916E-03 m$ for saturated and unsaturated conditions, respectively.
 
 Initial conditions are $c = 0$ and hydrostatic pressure conditions with steady state flow for both scenarios.
 
-{{< data-link "The NaCl1 project file" "Parabolic/RichardsComponentTransport/Padilla/Padilla_NaCl1/Padilla_NaCl1.prj" >}}  
+{{< data-link "The NaCl1 project file" "Parabolic/RichardsComponentTransport/Padilla/Padilla_NaCl1/Padilla_NaCl1.prj" >}}
 {{< data-link "The NaCl6 project file" "Parabolic/RichardsComponentTransport/Padilla/Padilla_NaCl6/Padilla_NaCl6.prj" >}}