prev

next

out of 52

View

215Download

0

Embed Size (px)

8/10/2019 Horton Calculations CH5

1/52

Copyright 2003 David G Tarboton, Utah State University

Chapter 5

At a Point Infiltration Models

for Calculating Runoff

8/10/2019 Horton Calculations CH5

2/52

Rainfall-Runoff Processes Chapter 5: 1

C

HAPTER

5:

A

T A POINT INFILTRATION MODELS

FOR CALCULATING RUNOFF

Infiltration is the movement of water into the soil under the driving

forces of gravity and capillarity, and limited by viscous forcesinvolved in the flow into soil pores as quantified in terms ofpermeability or hydraulic conductivity. The infiltration rate, f, is therate at which this process occurs. The infiltration rate actuallyexperienced in a given soil depends on the amount and distributionof soil moisture and on the availability of water at the surface. Thereis a maximum rate at which the soil in a given condition can absorbwater. This upper limit is called the infiltration capacity, fc. Note thatthis is a rate, not a depth quantity. It is a limitation on the rate atwhich water can move into the ground. If surface water input is lessthan infiltration capacity, the infiltration rate will be equal to the

surface water input rate, w. If rainfall intensity exceeds the ability ofthe soil to absorb moisture, infiltration occurs at the infiltrationcapacity rate. Therefore to calculate the actual infiltration rate, f, isthe lesser of fcor w. Water that does not infiltrate collects on theground surface and contributes to surface detention or runoff (Figure35). The surface overland flow runoff rate, R, is the excess surfacewater input that does not infiltrate.

R = w - f (28)

This is also often referred to asprecipitation excess.

The infiltration capacity declines rapidly during the early part of astorm and reaches an approximately constant steady state value aftera few hours (Figure 7). The focus of this section on at a pointinfiltration models for calculating runoff is on how to calculaterunoff accounting for the reduction of infiltration capacity. We useaccumulated infiltration depth, F, as an independent variable andwrite infiltration capacity as a decreasing function fc(F), then as Fincreases with time fcis reduced. fcmay be a gradually decreasingfunction, or a threshold function, as in the case of saturation excessrunoff where there is a finite soil moisture deficit that canaccommodate surface water input.

Several processes combine to reduce the infiltration capacity. Thefilling of fine pores with water reduces capillary forces drawing waterinto pores and fills the storage potential of the soil. Clay swells as itbecomes wetter and the size of pores is reduced. The impact of

8/10/2019 Horton Calculations CH5

3/52

Rainfall-Runoff Processes Chapter 5: 2

raindrops breaks up soil aggregates, splashing fine particles over thesurface and washing them into pores where they impede the entry ofwater. Coarse-textured soils such as sands have large pores downwhich water can easily drain, while the exceedingly fine pores in claysretard drainage. If the soil particles are held together in aggregates by

organic matter or a small amount of clay, the soil will have a loose,friable structure that will allow rapid infiltration and drainage.

Figure 35. Surface Runoff occurs when surface water input

exceeds infiltration capacity. (a) Infiltration rate = rainfall rate

which is less than infiltration capacity. (b) Runoff rate = Rainfall

intensity Infiltration capacity (from Water in Environmental

Planning, Dunne and Leopold, 1978)

The depth of the soil profile and its initial moisture content areimportant determinants of how much infiltrating water can be storedin the soil before saturation is reached. Deep, well-drained, coarse-textured soils with large organic matter content will tend to have highinfiltration capacities, whereas shallow soil profiles developed in clayswill accept only low rates and volumes of infiltration.

Vegetation cover and land use are very important controls ofinfiltration. Vegetation and litter protect soil from packing byraindrops and provide organic matter for binding soil particlestogether in open aggregates. Soil fauna that live on the organicmatter assist this process by churning together the mineral particlesand the organic material. The manipulation of vegetation during land

use causes large differences in infiltration capacity. In particular, thestripping of forests and their replacement by crops that do not coverthe ground efficiently and do not maintain a high organic content inthe soil often lower the infiltration capacity drastically. Soil surfacestrampled by livestock or compacted by vehicles also have reduced

8/10/2019 Horton Calculations CH5

4/52

Rainfall-Runoff Processes Chapter 5: 3

infiltration capacity. The most extreme reduction of infiltrationcapacity, of course, involves the replacement of vegetation by anasphalt or concrete cover in urban areas. In large rainstorms it is thefinal, steady state rate of infiltration that largely determines theamount of surface runoff that is generated.

The calculation of infiltration at a point combines the physicalconservation of mass (water) principle expressed through thecontinuity equation with quantification of unsaturated flow throughsoils, expressed by Darcy's equation. Here we will derive thecontinuity equation then substitute in Darcy's equation to obtain as aresult Richard's equation which describes the vertical movement ofwater through unsaturated soil. Figure 36 shows a control volume inan unsaturated porous medium. Consider flow only in the verticaldirection. The specific discharge across the bottom surface into thevolume is denoted as q, and the outflow across the top surface is

denoted as q+q. The volumetric flux is specific discharge timescross sectional area, yxA = . The volume of water in thecontrol volume is the moisture content times the total volume(equation 7), here zyxV = . Therefore we can write

Change in Storage = (Inflow rate Outflow rate) x (timeinterval)

x y z = (q x y (q+q) x y) t (29)

Dividing by x y z t and simplifying results in

z

q

t =

(30)

Now letting z and t get smaller and approach 0, as is usual incalculus, we get

z

q

t

=

(31)

This is the continuity equation in one direction (the vertical direction

z). In a more general case where flow can be three dimensional, thecontinuity equation is obtained in a similar fashion as

qz

q

y

q

x

q

t

zyx =

+

+

=

(32)

8/10/2019 Horton Calculations CH5

5/52

Rainfall-Runoff Processes Chapter 5: 4

where the operator is used as shorthand notation for

z,

y,

xand q denotes the specific discharge vector (qx, qy, qz)

with component in each coordinate direction.

Figure 36. Control volume for development of the continuity

equation in an unsaturated porous medium (from Chow et al.,

1988).

Substituting Darcy's equation (16) into (31) gives

z

hK

zt

=

(33)

In this equation h = +z (equation 18) resulting in

+

=

Kz

Kzt

(34)

This equation is known as Richard's equation and it describes thevertical movement of water through unsaturated soil. Althoughsimple appearing, its solution is complicated by the soil moisturecharacteristic relationships relating moisture content and pressure

head () and Hydraulic conductivity and pressure head or moisture

content K() or K() discussed above (equations 25, 26, 27 andFigures 29-32). Richard's equation may be written in one of two

forms depending on whether we take moisture content, , orpressure head, , as the independent variable.

8/10/2019 Horton Calculations CH5

6/52

Rainfall-Runoff Processes Chapter 5: 5

In terms of moisture content, Richard's equation is written

+

=

+

=

+

=

)(Kz

)(Dz

)(Kzd

d)(Kz

)(Kz

)()(K

zt

(35)

In this equation the explicit functional dependence on moisture

content, , has been shown. The quantity D()=

d

d)(K is called

the soil water diffusivity, because the term involving it is similar to adiffusion term in the diffusion equation. For specific

parameterizations of the soil moisture characteristic curves () andK(), such as equations (25-27), D() can be derived.

In terms of pressure head, Richard's equation is written

+

=

+

=

=

)(Kz

)(Kzt

)(C

)(Kz

)(Kztd

d

t

)(

(36)

As above, in this equation the explicit functional dependence onpressure head, , has been shown. The quantity C() = d/discalled the specific moisture capacity.

Analytic solutions for Richard's equation are known for specific

parameterizations of the functions K() and D() or K() and C()and for specific boundary conditions (see e.g. Philip, 1969; Parlangeet al., 1999; Smith et al., 2002). There are also computer codes thatimplement numerical solutions to Richard's equation. Hydrus 1-D isone such code available from the USDA-ARS SalinityLaboratory (http://www.ussl.ars.usda.gov/models/hydr1d1.HTM)

Computational codes based on the moisture content form tend to bebetter at conserving moisture and dealing with dryer soil conditions.These have problems as saturation is increased because moisture

content becomes capped at the porosity and d/dtends to infinity.Computational codes based on the pressure head form are able to

8/10/2019 Horton Calculations CH5

7/52

Rainfa