Library of analytic solutions (under construction)

From Aemwiki

Jump to: navigation, search

Categorization is loosely based upon scheme used by Bruggeman (1999)

Contents

[edit] Steady-state Confined/Unconfined Potential Flow

Governing Equation (1D or 2D):

\nabla \cdot \left(k\phi \nabla h\right)=-N

Where φ is the saturated thickness of the aquifer [L], N is the vertical influx to the aquifer (recharge/leakage) [LT − 1], and h [L] is the head measured with the respect to a flat aquifer base.

A common transformation (the Girinskii potential) can be used to simplified the governing equation, and make it equally valid for both confined flow (φ = H) and unconfined flow (φ = h):

\Phi=\frac{1}{2}k h^2

if h < H, and

\Phi=kHh-\frac{1}{2}kH^2

if h > H. This results in the Laplace/Poisson equation:

\nabla ^2\Phi=-N

The integrated discharge vector may be obtained from the gradient of the potential.

[edit] One-dimensional horizontal flow

Governing Equation:

\frac{\partial^2 \Phi}{\partial x^2}=-N

Solution #1: Steady-state / Two Dirichlet Boundaries

Boundary conditions: 

\Phi(0)=\Phi_1; \frac{}{} \Phi(L)=\Phi_2

Solution:

\Phi(x)=-\frac{1}{2}Nx^2+\left(\frac{\Phi_2-\Phi_1}{L}+\frac{1}{2}NL\right)x+\Phi_1

Solution #2: Steady-state / One Dirichlet, One Neumann Boundary

Boundary conditions:

\Phi(0)=\Phi_1; \frac{}{} Q_x(L)=\beta

Solution:

\Phi(x)=-\frac{1}{2}Nx^2+\left(-\beta+NL\right)x+\Phi_1

[edit] Two-dimensional radially-symmetric horizontal flow

Governing Equation:

\frac{\partial}{\partial r}\left(r\frac{\partial \Phi}{\partial r}\right)=-N

Solution #1 (Steady State Thiem Solution):

Boundary conditions: 

\underset{r\rightarrow 0}{\lim}\int_0^{2\pi} rkh\frac{\partial h}{\partial r}d\theta=\underset{r\rightarrow 0}{\lim}\int_0^{2\pi} rQ_rd\theta=Q 
(i.e., the total flow out at r=0 is equal to the total flow from the well, Q)

\Phi(R)=\Phi_0\frac{}{}

Solution:

\Phi(r)=-\frac{Q}{2\pi}\ln\left(\frac{r}{R}\right)+\Phi_0

[edit] Two-dimensional general horizontal flow

Governing Equation:

\frac{\partial^2 \Phi}{\partial x^2}+\frac{\partial^2 \Phi}{\partial y^2}=-N

[edit] Transient Phreatic (Unconfined) groundwater

Governing Equation (1D or 2D):

\nabla \cdot \left(kh \nabla h\right)=S_y\frac{\partial h}{\partial t}-N

Where Sy is the specific yield [-], N is the vertical influx to the aquifer (recharge/leakage) [LT − 1], and h [L] is both the head and the saturated thickness (i.e., the head is measured with the respect to a flat aquifer base).

[edit] Exact solutions

[edit] General two-dimensional horizontal flow

[edit] Transient Confined groundwater

Governing Equation (1D or 2D):

\nabla \cdot \left(kH \nabla h\right)=S_sH\frac{\partial h}{\partial t}-N

Where Ss is the specific storage [-], N is the vertical influx to the aquifer (recharge/leakage) [LT − 1], H is the saturated thickness of the aquifer and h [L] is the head, measured with respect to an arbitrary datum.

For piecewise constant properties, we can substitute in the confined Girinskii potential (Φ = kHh) to obtain the following governing equation:

\nabla ^2\Phi=\frac{1}{\alpha}\frac{\partial \Phi}{\partial t}-N

Where \alpha=\frac{k}{S_s} is the hydraulic diffusivity [LT − 1]of the aquifer.

[edit] One-dimensional flow

[edit] Two-dimensional flow

[edit] Two-dimensional radially-symmetric flow

Solution #1 (Transient Theis Solution):

Boundary/initial conditions:

\Phi(r,0)=\Phi_0\frac{}{}

\underset{r\rightarrow 0}{\lim}\int_0^{2\pi} rkH\frac{\partial h}{\partial r}d\theta=\underset{r\rightarrow 0}{\lim}\int_0^{2\pi} rQ_rd\theta=QH(t) 
(i.e., the total flow out at r=0 is equal to the total flow from the well, Q). 
Here, H(t) is the heaviside function, equal to 1 for t > 0, zero otherwise.
 
\underset{r\rightarrow \infty}{\lim}\Phi(r)=\Phi_0\frac{}{}

Solution: 

\Phi(r,t) = \Phi_0-\frac{Q}{4\pi}E_1\left(\frac{r^2 S_s}{4kt}\right)

Where E1() is the exponential integral (a.k.a. the "well function")

[edit] Three-dimensional flow

[edit] Three-dimensional spherically-symmetric flow

[edit] Three-dimensional axially-symmetric flow

[edit] Multi-layer systems

[edit] Solute Transport

Governing equation for single aqueous solute, with sorption:

\frac{ \partial \theta C}{\partial t}=-v(\mathbf{x})\theta\nabla C+\nabla \cdot (\mathbf{\theta D(\mathbf{x})} \nabla C) - R(C) -\frac{\rho_b}{\theta}\frac{\partial q}{\partial t}

Where C and q are the aqueous and sorbed concentrations of the solute, respectively, v(\mathbf{x}) is the velocity vector, \mathbf{D(\mathbf{x})} is the dispersion tensor, R(C) is a general reaction term, ρb is the bulk dry density of the porous media, and θ is the porosity of the media.

For development of analytical solutions, it is typically assumed that the flow is uniform in the x-direction, the porosity is uniform and constant, sorption may be modeled using a linear retardation factor, R=1+\frac{\rho_b}{\theta}\frac{\partial q}{\partial C}, where \frac{\partial q}{\partial C}=K_d, and the reaction terms are limited to simple first or zeroth order decay and growth. This gives us:


R\frac{\partial C}{\partial t}=-v_x\frac{\partial C}{\partial x}+D_{xx}\frac{\partial^2 C}{\partial x^2}+D_{yy}\frac{\partial^2 C}{\partial y^2} +D_{zz}\frac{\partial^2 C}{\partial z^2}- \lambda C + \gamma

Where λ is the first-order decay constant and γ is a zeroth-order growth term. This equation is linear with constant coefficients, and thus amenable to analytical solution using a wide variety of methods.

[edit] One-dimensional Transport

Solution #1: Ogata-Banks Solution

Boundary/initial conditions:

C(x,0)=C_i\frac{}{}
C(0,t)=C_0\frac{}{}
\frac{\partial C}{\partial x}(\infty,t)=0\frac{}{}
λ = γ = 0

Solution: 

C(x,t) = C_i+(C_0-C_i)A(x,t) \frac{}{}  

Where A(x,t)=\frac{1}{2}\mathbf{erfc}\left(\frac{Rx-vt}{2\sqrt{DRt}}\right)+\frac{1}{2}\exp\left(\frac{vx}{D}\right)\mathbf{erfc}\left(\frac{Rx+vt}{2\sqrt{DRt}}\right)

note that if D = 0, then 

A(x,t)=H(Rx-vt)\frac{}{}

Where H(x) is the Heaviside step function.

[edit] Two-dimensional Transport

Solution #1: Cleary and Ungs Solution

[edit] Density flow (interface)