Tải bản đầy đủ - 0 (trang)
Chapter XII. Evaporation from Drying Ecosystems

Chapter XII. Evaporation from Drying Ecosystems

Tải bản đầy đủ - 0trang

B A R E SO IL S U R F A C E S



305



water out of the soil body as vapor depends therefore not only on the

energy supply, as has been true with the systems described in the two

preceding chapters, but also on how fast water can move from its

initial location in the soil body to the level where both water and

energy are available and vaporization can take place. This zone of

water and energy convergence is initially at the surface, but shifts as

the soil body dries. This shift is illustrated in the successive phases of

the evaporation process.



The Phases of Evaporation from Soil Bodies

Philip (1957, 1964) shows that evaporation from bare soil progresses

through three phases:

(1) While the soil surface is wet, evaporation proceeds as from a

saturated surface. It responds chiefly to the local supply of energy, that

is, the radiation surplus and such atmospheric factors as the profiles of

temperature and humidity. In this phase both energy and water are

available in the surface layer.

(2) As the moisture content of the top layers of the soil 8 declines,

the hydraulic conductivity or permeability of the soil K, which is a

complex function of moisture content, also declines. Water under the

same pressure head moves more slowly than before, because there is a

reduced cross section for it to move through. At the same time, the

tension forces holding the water in the micropores and in films on the

soil particles (capillary pressure or moisture suction $) are increased,

for they also are functions of soil-moisture content, as noted in the

chapter on soil-moisture intake. In this phase, evaporation from the

soil surface depends only on the vertical distribution of moisture in

the soil. If moisture lost from the top layers is not readily replaced

from below, evaporation slows down. Atmospheric conditions and

energy input have a diminished effect on evaporation, not being able

to affect the supply of moisture to the layer where energy can vaporize

it.

(3) With still smaller moisture content and hence less hydraulic

conductivity K, the movement of liquid water comes nearly to a

standstill. Outflow of vapor from the surface is supplied mainly by the

upward movement of vapor through the soil. Vapor diffuses in the soil

in response to the gradient of vapor pressure, which means that it

moves from warm levels to cooler ones. The temperature profile of the

soil is therefore of primary importance to evaporation in this phase.

In winter, when the upper layers of soil are cooler than the deeper

ones, vapor moves upward to the surface and out of the soil body. In



306



XII.



E V A P O R A T I O N FROM DRYING ECOSYSTEMS



summer, on the other hand, the upper layers are warm and the lower

ones relatively cool, so the flux of vapor is directed downward.

Evaporation from the surface is small or altogether lacking. This

counter-seasonal regime resembles the regime of evaporation from

deep bodies of water, described earlier.

In this third phase of evaporation from the soil, radiation and

atmospheric factors play a role that contrasts with their role in the case

of a saturated surface, phase (1).The energy that can support vaporization of water is distributed vertically like the distribution of temperature in the soi1 column. In any case, the rate of energy supply is small

because the zone where liquid water is located is far below the

irradiated surface where the suppIy of energy is greatest. Water is

effectually separated from a substantial source of energy, and evaporation is correspondingly small.

In speaking of these three phases in desiccation of a soil body,

Philip (1957) comments that it “appears futile to seek to relate

evaporation from soils with dry surfaces to the ‘evaporating power of

the air,’ as represented by evaporation from water surfaces of ’wet’

soils.” We saw earlier that the notion of “evaporating power of the

air” was a chimera in explaining evaporation from wet surfaces; it is

just as inadequate for drying soil bodies. It is clearly necessary to

examine the physics of the system as a whole.



Energy Fluxes and Temperature at the Soil Surface



Effects of Changes in the Energy fluxes Lysimeter measurements of

evaporation in early stages of a drying soil (Fritschen and Bavel, 1962)

and the associated heat fluxes are shown in Table I for the wet soil

described earlier, and on later days (Fig. XII-1). As the soil surface

dried, the increase in its reflectivity and temperature resulted in a net

surplus of radiant energy that was smaller by about 0.15 than when

the surface was wet. The drier surface lost less heat by evaporation,

and its higher temperature forced more heat to move from i t into

deeper soil layers S and into the air H. Associated with these three

shifts in heat flow, the rate of conversion of heat by vaporization

decreased by almost hall.

The rate at which water in the soil body moved upward toward an

evaporating surface was measured (Rose, 1968) for movement in both

the liquid and the vapor state. Liquid-water flow in the soil was at

about 40 mm secc’ x

on the first day after the soil was wetted (or

about 4 rnm dayc1), and at half or less of this rate on the third and

fourth days The flux of water in vapor form was nearly zero at night,



307



BARE SOIL SURFACES



TABLE I



Exchanges

energy) "



of



Energy at a Drying



Surface, Phoenix, Arizona (24-ht average f i u x r i of



5011



Date

Parameter



29 April



30 April



2 May



Wet

30



Moist



Dry



State of surface

Surface temperature ("C)

Net surplus of energy in all radiation exchanges

R

Heat exchange with soil body S

Heat exchange with the air H

Heat of vaporization E

Equivalent evaporation (mm)

"



+I96

+4

-



+1

201

7.0



32



38



+ 193



+ 160



-5

- 15

- 173



6.0



-



16



-- 34



110

3.8



Units: W m-'. Source: Fritschen and Bavel (1962)



and by day proceeded at somewhat higher speeds than the flux of

liquid water-but it was directed downward, away from the heated

upper layer. The vapor flow out through the surface was obviously

supplied by the liquid flux, not the vapor. Evaporation at night

indicated a continued upward liquid-water flow in the soil body a t a

low rate.



840



L



L



~



-



tssn1



I



I



I



I



,



I



I



I



I



,



,



Fig. XII-1. Diurnal courbe of the energy fluxes (-net

all-wave r,idiation, latent-heat flux, ----sensible

heat flux, ... soil-heat flux) a t the dry surface o f soil five

days after i t was irrigated (Fritschen and Bavel, 1962).



308



XII.



E V A P O R A T I O N FROM D R Y I N G E C O S Y S T E M S



Because most of the water for evaporation must come from the top

layers of the soil mass, the amount of evaporation is not much greater

from soil deeply wetted by a big rain than from soil merely superficially dampened by a small one. Over a period of time, total evaporation depends less on infiltration amounts than on the number of times

per season that the surface is wetted and subsequently has access to

enough energy to evaporate the small depth of water within its reach.

As was true with water intercepted on litter and leaf surfaces (Chapter

VI), total evaporation is determined chiefly by the number of rains per

season or the duration of surface wetness. Bare soil behaves much like

city pavements and roofs, which draw upon no subsurface store of

water and evaporate only the 1mm or less that is caught in depression

storage and the film remaining after rain. In these cases of superficial

evaporation energy is assumed available and is evident in the

temperature of the surface.



Surface Temperature Estimates of evaporation from a drying soil, as

well as from the wet soil we discussed in a preceding chapter, can be

made if surface temperature is known. This implies that the effect on

evaporation of the progressive drying of the soil surface, to a moisture

content as low as 0.005, is taken into account in the increase in surface

temperature.

Fuchs and Tanner (19671, in an experiment on sandy soil in

Wisconsin, measured a maximum surface temperature of 44"C, which

was 16°C higher than air temperature and also much higher than the

surface temperature (32°C) at the wet soil surface in the Arizona

experiment described earlier. If we consider that some level around

32°C represents the upper temperature limit at an extensive, freely

evaporating surface, it is clear that drying out the tog 1-2 crn of a

sandy soil will be associated with a substantial rise in the temperature

of this layer and the temperature at its surface. Since the layer IL, -&I q ,

however, such a temperature rise does not produce more evaporation

but rather less.

The decrease of the evaporation rate is given by the expression

(Fuchs et a l. , 1969) in which p is the density of air (kg m-'); c,, specific

heat of air (J kg-' deg-'); h transfer coefficient for heat ana vapor from

the surface to height z in the air (rn sec-*); s slope of the curve that

relates saturation vapor pressure to surface temperature (mb deg-I); e

vapor pressure (mb); e," saturated vapor pressure that corresponds to

the temperature of the surface (mb); e , vapor pressure in the air at

height z (mb). As the surface heats, eo* Increases and actual evagora-



EVAPOTRANSPIRATION FROM A SOIL-VEGETATION SYSTEM



309



tion departs more and more from the potential rate. Thus, under

steady radiant heating, for example, the progressive drying of a

surface reduces evaporative cooling, the surface gets hotter, eo* rises,

and actual evaporation becomes a small fraction of potential evaporation.

Midday rates of evaporation from the dry, sandy top layer in

Wisconsin were only 0.06 mm hr-'. The daily total was about 0.3 mm,

an order less than the partly dried soil in the Arizona experiment.

This "combination equation is limited to surfaces such as bare

soils" (Fuchs et ai., 19691, at which the turbulent exchanges of heat

with the air are made at a well-defined interface. It then shows how

evaporation from soil becomes a self-limited process.

In vegetation, in contrast, the zone filled with leaf surfaces at which

the exchanges of sensible and latent heat take place is thick, and in it

the idea of a "surface temperature is not well-defined" (Fuchs et a l . ,

1969). Let us now look at this more complicated system.



EVAPOTRANSPIRATION FROM A DRYING SOIL-VEGETATION

SYSTEM



Evaporation is altered when plants occupy a soil body. Leaves grow

over the soil surface, and roots occupy the soil body. Energy and water

now come together at a new convergence zone, the leaf surfaces. These

are supplied with energy from the sun and with water brought up

from deep in the soil body.

Exposed to radiation and the atmosphere, the cells of plant leaves

must contain,. and also leak, water. Therefore they must be supplied

with water, which is actively searched out in the soil by roots. Because

the forces holding water in the smali pores of the soil increase as soil

moisture declines, roots find it increasingly hard to extract water

during periods between rainstorms.

Whiie movement of water in bare soil is difficult to formulate and

measure physically, the location and movement of water in the soilplant-atmosphere system is far rnore complex. Even within the plant,

water movement forms one of the ciassicai problems of plant physioiogy (Bonner, 1959). Dependent on the varying forces in drying soil

and the regular and irreguiar regimes of heat suppiy and vapor

removal at foliage surfaces, evapotranspiration from drying vegetation

is a highly complicated process, although a common one. Bare soil

tends to be quickly colonized by plants. A special example is the solidwaste disposal site, or landfill, which yields a leachate percolation



310



XII.



E V A P O R A T I O N FROM DRYING ECOSYSTEMS



SOIL MOISTURE ( P E R C E N T BY VOLUME)



Fig. XII-2. Relative rate of transpiration in tenths as a function of volumeti-ic soil

moisture on days of different energy levels (potential evapotranspiration, but Libeled in

energy terms W m-'). Field capacity of the soil wa5 3 h % , the 15-bar content 22"/0

(Denmead and Shaw, 1962).



stream of complex and often harmful composition. Even if a landfill is

sealed from the native groundwater body, the volume of leachate

should be minimized, and this can be done by promoting evapotranspiration. Molz et a l . (1974) reduced leachate volume by 300 mm yr-',

or 0.7, by planting the landfill surface. Similar efforts are made on

septic-tank effluent fields.

Different plants behave differently in the same conditions of soil

a n d atmospheric forces. Behavior patterns such as closing stomata,

offering resistance to diffusion of water from leaves, extending the

absorbing roots, and tending to grow roots at different depths are

species dependent, as we saw earlier.



Effects of Drying Soil on Ecosystem Evapotranspiration

Evapotranspiration is reduced by slow movement of soil water to

root hairs as the soil dries. When this happens the tension gradients

along which water moves must increase in order to keep water moving

across the space between the soil particles and the root hairs. For this

increase to develop, the diffusion pressure deficit in the leaves

increases; there is less leaf turgor and the guard cells close the stomata.

Transpiration declines.

If the energy supply is high, this change in the water stream moving

from soil to root to leaf to air might take place even when the moisture



E V A P O T R A N S P I R A T I O N FROM A SOIL-VEGETATION SYSTEM



311



content of the soil is high (Denmead and Sha147, 1962). ”We expect

transpiration rates to decline with decreasing soil moisture content

and we expect that this decline will be evident at higher and higher

soil moisture contents as the potential transpiration rate increases,”

i.e., on days of a large supply of energy. On days of great energy

falls short of the potential

supply-say 20Cl-300 W m-’-transpiration

rate even when plants are rooted in a soil of moisture content dose to

iield capacity (0.36 moisture content) (Fig. XIi-2).

On days of low energy supply (about 100 W m-’);, however, Fig. XII2 shows that transpiration proceeds at- the potential rate even fror,

plants rooted in soil having a small moisture content. In this case,

moisture content is much iess than field capacity. Only in very dry soil

does transpiration on such low-energy days fall short of the potential

rate.

The soil-moisture content at which actual transpiration begins to

depart from the potential rate is identified by Denmead and Shaw

(1962) as the point at which leaf turgor is iost. In a given soil type this

point varies with energy supply, as Fig. XI1-3 shows. In the soil

illustrated, the following critical levels of soil moisture are found:



22



-/

I



/I



I



30



I



1



go

12;

151

in6

210

POTENTIAL EVAPOTRANSPIRATION ( W 6’’



60



Fig. XII-3. Estimated soil moisture at the point of loss of leaf turgor on days of

different energy levels or potential evapotranspiration On days of high energy, leaf

turgor is lost at relatively high levels of soil moisture (percent by volume) (Denmead and

Shaw, 1962)



312



XII.



E V A P O R A T I O N FROM DRYING ECOSYSTEMS



Soil moisture 0.36 (Field capacity). Soil suction 0.3 atm.

Soil moisture 0.34 Turgor-loss point on days when the potential

evapotranspiration rate is 6 mm day-' (energy

conversion rate approximately 300 W m-').

Soil moisture 0.27 Turgor-loss point on days when potential evapotranspiration is 4 mm day-'.

Soil moisture 0.23 Turgor-loss point on days when potential evapotranspiration is 1.4 mm day-' (approximately

80 W m-'). This is also the level of a soil suction

of 15 bars.



As we saw in a preceding chapter, soil moisture in nature is not

evenly distributed, and this affects transpiration. Calculations of

evapotranspiration at 25 sites in a field of grass in Ontario revealed a

large dispersion; over a 3-month period in which the average evapotranspiration at all sites was 252 mm, the standard deviation among

sites was 33 mm, giving a coefficient of variation of 0.13. These

differences resulted from differences not in energy supply or vapor

removal but in "large soil moisture differences at the individual sites"

(Rouse, 1970). Uncovering these differences required both the careful

measurements of soil moisture itself, by two methods, and also the

calculation of downward movement of water F beyond 1.8-m depth by

use of a flux gradient relation:



F



=



-K, d$Idz,



where K is a coefficient of transport related to the gradient of potential

4 (Rouse and Wilson, 1972). Evaporation rates determined by this

water budget of the soil were not acceptable over one-day periods, but

were adequate for periods of several days if no rain fell and if at least

six soil-moisture sites were measured.

Forests approach the potential evapotranspiration ideal more closely

than does lower vegetation, considering the course of the whole growing

season. Compared to annual crops, they are active longer in late

summer and autumn; compared with most low ecosystems, they root

more deeply and can keep on transpiring during dry spells in summer.

Both their darker appearance (they absorb more solar radiation) and

roughness (extracting more heat from the air) give forests a larger

supply of available energy to support transpiration. As Rauner demonstrates (1972, p . 146), actual evapotranspiration from forests tends to

approach potential evapotranspiration, except where dry seasons occur. In Douglas fir in southwestern Oregon differences between actual



313



E V A P O T R A N S P I R A T I O N FROM A SOIL-VEGETATION SYSTEM



and potential transpiration indicate strong stomatal control in summer

(Reed and Waring, 1974). Growth of the fir in spring is limited only by

energy input, but in late summer by plant resistance. Over monthly

periods it is related to the ratio of actual to potential transpiration.



Effects of Stornatal Resistance

When loss in leaf turgor brings about a narrowing of the stomata,

the stomatal resistance to diffusion of water Y~ increases. For example,

when stomata narrow from a width of 5-1 pm, stomatal resistance

increased in one experiment from 0.8 to 2.8 sec cm-', as shown in

Table 11. With no change in the external resistance to vapor diffusion

Y ~ ,the "total resistance nearly doubles, but transpiration decreases by

only one fifth because the leaf becomes warmer and the water

concentration difference increases by about half" (Waggoner and

Zelitch, 1965), i.e., from 24 to 35 mb. In other words, the greater

heating of the leaf counteracts some of the throttling effect of closing

stomata by increasing the vapor-pressure difference between leaf and

air.

In the usual flux versus gradient formula, the transpiration stream

depends both on the difference between the concentration of water

vapor at the cell walls and in the atmosphere, and on the resistance to

TABLE I1



Leaf Transpiration in Quiet Air"

Width of stomata ( p m )

Parameter

Resistance (sec cm-')

Stornatal r I

Atmospheric r,



Total

Temperature of leaf ("C)

Vapor pressure in leaf (mb)

Leaf-air temperature difference ("C)

Leaf-air vapor-pressure difference (mb)



0



1



5



x



1.7



2.8

1.?



0.8

1.7



cc



4.5



2.5



44

91

14

70



35

56

5

35



31

45

1

24



a Wind speed 5 cm secc', radiation surplus +410 W mc2, air

temperature 3 0 T , vapor pressure 21 mb, leaf width 5 cm. Data from

Waggoner and Zelitch (1965).



314



XII.



E V A P O R A T I O N F R O M D R Y I N G E C O SY ST E MS



diffusion of vapor from cell walls into the free atmosphere. The

reduction in transpiration by a fifth puts more of the heat-removal

burden on the sensible flux and long-wave radiation, which can

handle the load if the leaf gets warmer (as is shown in Table 11).

In stronger air movement (Table 111), narrowing of stomata from 5 to

1 pm increases the stomatal resistance r l , as in the previous example.

The effect is enhanced on total resistance, which now is chiefly

composed of the stomatal component and contains only a small

contribution r, due to atmospheric resistance in the moving air.

Increase in stomatal resistance results in a large decrease in transpiration, which falls to half its former rate.



Energy Considerations

As a transpiring ecosystem runs through its store of available water,

as soil moisture gets harder to extract from the tiny pores, and as

stomatal resistance to vapor diffusion increases, the vaporization of

water at cell walls in the leaves decreases. Given the same supply of

energy to the leaves in the sunny weather between rainstorms, there

must be a shift in its division; less can be converted into the latent

heat of vaporization, more must be converted into sensible heat and

transferred into the air El.* This is represented as an increase in the

Bowen ratio ( p = HIE), as H grows and E diminishes.

Some of the methods described in the preceding chapter to determine rates of evapotranspiration from well-watered ecosystems can

also be utilized in the circumstances of drying soil. One of these is the

Bowen ratio method, in which the ratio is approximated by calculation

from the gradients of temperature T and specific humidity q in the air

over the ecosystem, as



p



=



c,(AT



+ r Az)lL 4,



where c, (specific heat of air at constant pressure



=



1J g-' deg-I), and



L (latent heat of vaporization of water = approximately 2500 J g-')



serve as coefficients that convert temperature and humidity readings

into units of heat and hence make the ratio nondimensional. The term

r Az is a minor adjustment that corrects for the adiabatic cooling of

ascending air with change of height z .



' A small increase will also occur in the subsurface movement of heat S, as seen in the

Arizona energy budget for drying soil (Table l), although less than where soil is bare.

Warmer foliage also radiates away more energy; this shift is also of minor significance.



315



E V A P O T R A N S P I R A T I O N F R O M A S O I L - V E G E T A T I O N SYSTEM



TABLE 111



Leaf Transpiration in M o v i n g A i r a

Width of stomata (pm)

Parameter

Resistance (sec cm-')

Stomata1

Atmospheric

Total

Temperature of leaf ( " C )

Vapor pressure in leaf (mb)

Leaf-air temperature difference ("C)

Seat-air vapor-pressure difference (mb)



1



5



0.2



2.8

0.2



0.8

0.2



J



3.0



1.0



0



3c



34

53

+4

32



31



45

f!



24



28

38



-2

17



" Wind speed 223 cm sec-I, radiation surplus +410 b m-'? air

temperature 30"C, vapor pressure 21 mb, leaf width 5 cm Data frorr

Waggoner and Zelitch (1965)



This ratio is then applied, as noted earlier, to divide the net

available heat at the surface, i.e., radiation surplus less subsurface

heat intake. In dry conditions Aq is small, since both crop and air are

d r y , but AT is large, because the crop surface gets very hot. lnstead of

a few tenths oS a unit as over well-watered surfaces, the ratio reaches

values an order of magnitude greater; a value of 2, for instance,

indicates that twice as much heat is removed from the ecosystem in

sensible form as in latent. Values of the Bowen ratio were determined

from micrometeorological measurements in a four-year study (Denmead and McIlroy, 1370) of a lysimeter in which wheat was growing,

and agreed satisfactorily with those found from measurements of R , S,

and E in the lysimeter. This Australian experiment confirms the

validity of the meteorologically determined ratio when it is applied to

divide available heat at the surface of a crop often cultivated in dry

climates.

In a Canadian study (Wilson and Rouse, 1972) a variation of the

combination model for evapotranspiration mentioned earlier ( E = (X

- S)[s/(s + y ) ] ) was found also to work well on moderately dry days.

The partition of available energy ( R - S) is still represented by the

term sNs + r), until the soil gets too dry. Budyko (1971, p. 94) notes

that this formula is useful during periods when the major components



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Chapter XII. Evaporation from Drying Ecosystems

Tải bản đầy đủ ngay(0 tr)

×