Algebraic length-scale with two master scales


INTERFACE:

   subroutine potentialml(nlev,z0b,z0s,h,depth,NN)
DESCRIPTION:

Computes the length scale by defining two master length scales $ l_u$ and $ l_d$

\begin{displaymath}\begin{array}{l} \int_{z_0}^{z_0+l_u(z_0)} (b(z_0)-b(z)) dz =...
... \int_{z_0-l_d(z_0)}^{z_0} (b(z)-b(z_0)) dz =k(z_0) \end{array}\end{displaymath} (184)

From $ l_u$ and $ l_d$ two length-scales are defined: $ l_k$, a characteristic mixing length, and $ l_\epsilon$, a characteristic dissipation length. They are computed according to

\begin{displaymath}\begin{array}{l} l_k(z_0)= \text{Min} ( l_d(z_0),l_u(z_0)) \;...
...=\left( l_d(z_0)l_u(z_0)\right)^\frac{1}{2} \quad . \end{array}\end{displaymath} (185)

$ l_k$ is used in kolpran() to compute eddy viscosity/difussivity. $ l_{\epsilon}$ is used to compute the dissipation rate, $ \epsilon$ according to

$\displaystyle \epsilon=C_{\epsilon} k^{3/2} l_{\epsilon}^{-1} \; , \quad C_{\epsilon}=0.7 \quad .$ (186)


USES:

   use turbulence, only: L,eps,tke,k_min,eps_min
   use turbulence, only: cde,galp,kappa,length_lim
 
   IMPLICIT NONE
INPUT PARAMETERS:
 
   number of vertical layers
   integer,  intent(in)                :: nlev
 
   bottom and surface roughness (m)
   REALTYPE, intent(in)                :: z0b,z0s
 
   layer thickness (m)
   REALTYPE, intent(in)                :: h(0:nlev)
 
   local depth (m)
   REALTYPE, intent(in)                :: depth
 
   buoyancy frequency (1/s^2)
   REALTYPE, intent(in)                :: NN(0:nlev)
REVISION HISTORY:
   Original author(s):  Manuel Ruiz Villarreal, Hans Burchard

Karsten Bolding 2012-12-28