The impact of plant stomatal control on mesoscale ... › 2009 › 09 › r-110.pdf · The impact of plant stomatal control on mesoscale atmospheric circulations. Agric. For. Meteorol.,

Agricultural and Forest Meteorology, 54 ( 1991 ) 353-372 353 Elsevier Science Publishers B.V., Amsterdam

The impact of plant stomatal control on mesoscale atmospheric circulations

R o n i Avissar a a n d R.A. Pie lke b aDepartrnent of Meteorology and Physical Oceanography, Rutgers University, New Brunswick,

NJ 08903, USA bDepartment of Atmospheric Science, Colorado State University, Fort Collins, CO 80523, USA

(Received 1 November 1989; revision accepted 22 October 1990)

ABSTRACT

Avissar, R. and Pielke, R.A., 1991. The impact of plant stomatal control on mesoscale atmospheric circulations. Agric. For. Meteorol., 54: 353-372.

Only a few numerical studies with mesoscale atmospheric models have been undertaken to explore the influence of vegetation on the generation and modification ofmesoscale atmospheric circulations. Nevertheless, these few studies have demonstrated the importance of a correct representation of vegetation in the simulation of atmospheric features. In the present study, a mesoscale atmospheric model with a sophisticated land surface parameterization is used to emphasize the major role of plant stomata on the control of the Bowen ratio and on mesoscale atmospheric circulations. This study demonstrates that refinement of the land surface parameterization and stomatal mechanism, including subgrid-scale processes, is required in order to simulate confidently land-atmosphere interactions. This is particularly important when plants experience relatively strong stress and a small change in the stomatal behavior generates large variations in sensible and latent heat fluxes at the Earth's surface.

I N T R O D U C T I O N

In atmospheric models, the Earth's surface is the only boundary that has a physical significance. Moreover, it is the differential gradient of the dependent variables along this surface that generates many mesoscale circulations (i.e. terrain induced mesoscale systems) such as sea and land breezes, moun- tain-valley winds, urban circulations and forced airflow over rough terrain. This gradient also has a pronounced influence on the remaining mesoscale flows (i.e. synoptically induced mesoscale systems) such as squall lines, hur- ricanes and traveling mesoscale cloud clusters. Pielke (1984) reviewed these circulations and flows in detail. Changes in this lower boundary over time (i.e. from anthropogenic activity or overgrazing by animals) can cause sub- stantial climatic changes such as desertification (Otterman, 1975; Idso, 1981 ). Thus, because of the crucial importance of this boundary for the atmospheric systems, it is essential to represent it as accurately as possible in models.

0168-1923/91/$03.50 © 1991 - - Elsevier Science Publishers B.V.

354 R. AVISSAR AND R.A. PIELKE

The representation of land surfaces as a bottom boundary requires different types of parameterizations than those required to represent the surface of water bodies (e.g. lakes, oceans, etc. ). This is because water is translucent to solar radiation and overturns easily, in contrast to the ground which is opaque and does not readily overturn. Also, owing to the complexity of the physical and biological processes involved in the representation of vegetation, it is much easier to represent bare soils than plant canopies in the parameterization of land surfaces.

Mesoscale models have become more sophisticated in the parameterization of bare soils. These models evolved from the prescription of potential temperature as a periodic heating function (e.g. Kuo, 1968; Neumann and Mahrer, 1971; Pielke, 1974; Mahrer and Pielke, 1976) to the solution of a surface energy budget (e.g. Physick, 1976; Mahrer and Pielke, 1977a, b; Es- toque and Gross, 1981; McCumber and Pielke, 1981; Ookouchi et al., 1984; Mahfouf et al., 1987; Avissar and Mahrer, 1988a; Segal et al., 1988; Avissar and Pielke, 1989; among many others). However, since much of the world is vegetated, it is necessary to represent correctly the contribution of plants as well as the soil surface to the surface processes.

Vegetation has been introduced in only a few mesoscale models and relatively little evaluation of the influence of vegetation on mesoscale circulations has been published in the research literature. An interactive atmosphere-vegetation-soil model was applied by McCumber (1980) who evaluated, among other aspects, the effect of vegetation on the development of the summer sea breeze over South Florida. This study also indicated that significant induced mesoscale circulations can be related to sharp horizontal changes in the char- acter and type of the vegetation cover. Garrett (1982) included a vegetation module in his model while studying the interactions between convective clouds, the convective boundary layer and forested surfaces. Yamada ( 1982 ) incorporated vegetation in a planetary boundary-layer model in order to study air circulations in the lower atmosphere.

However, these studies did not attempt to explore quantitatively and sys- tematically the impact of vegetation on the generation and modification of mesoscale systems. One of the first attempts to evaluate expected circulations between a vegetated area contrasted with bare soil was reported by Anthes (1984). That study provided an evaluation of the anticipated thermal contrast in subtropical latitudes. In addition, it provided scaling by means of a linear analytical model, illustrating the possible characteristics of mesoscale circulations generated by such contrasts.

Avissar and Mahrer ( 1988a, b) demonstrated the importance of different types of vegetations on the development and modification of small local circulations during radiative frost events. Mahfouf et al. ( 1987 ) studied the influence of soil and vegetation on the development of mesoscale circulations. They found that the juxtaposition of a well-transpiring vegetated area with a

IMPACT OF STOMATAL CONTROL ON MESOSCALE ATMOSPHERIC CIRCULATIONS 3 5 5

dry bare land can generate circulations as strong as sea breezes. Probably the most complete study on the interactions between vegetation and mesoscale circulations was recently presented by Segal et al. (1988 ) who analyzed the influence of various plant covers and densities on different mesoscale features such as thermally induced upslope flows and sea breezes, as well as the generation of circulations induced by the juxtaposition of vegetated areas with bare dry lands. Parts of their results were supported by observations and in- frared surface temperatures obtained from the Geostationary Operational Environmental Satellite (GOES).

It is widely accepted that these vegetation effects on mesoscale systems are the result of the fact that when the plants transpire effectively, the radiative energy received at the Earth's surface from the sun and the atmosphere is mostly redistributed into latent heat flux in contrast with the redistribution into sensible heat flux over a dry surface. Thus, the heating of the atmosphere during daytime hours above a dry surface is usually more intense than that above vegetated areas. This generates horizontal gradients of temperature (and, therefore, pressure) in the atmosphere which initiate circulations between the two areas. However, when the plants are under strong stress and do not transpire, the sensible heat flux produced over the vegetation is almost identical to that obtained over a bare surface and the vegetation does not contribute significantly to the mesoscale systems.

Transpiration is biophysically controlled by the plant stomata. It is, therefore, obvious that the contribution and realistic parameterization of the stomatal mechanism in mesoscale atmospheric models is essential.

This paper reviews the different parameterizations of plant stomata that have been adopted in mesoscale atmospheric models and demonstrates their contribution to regional atmospheric circulations.

PARAMETERIZATION OF PLANT STOMATA IN ATMOSPHERIC MODELS

In the parameterization of vegetation in mesoscale atmospheric models, the plant stomatal control is obtained by introducing a resistance (or conductance) in the equation which describes the latent heat flux from the vege- tative surface. For instance, in Avissar and Mahrer ( 1988a; also described in Segal et al. (1988) and Avissar and Pielke (1989) ) this equation is

,~Ev = a ' f p2u .q . ( 1 )

with

ty'f = 2Lty f / ( 1 + 2Ltyf) (2)

q. = k ( q~, - q~o)/0.74{ln [ (z , - d ) / zo] - -~H } (3)

qzo = tr~-qv~ + ( 1 -- tr~-)qG (4)

and

3 5 6 R. AVISSAR AND R.A. PIELKE

qva =gsrqv* + ( 1 --gsr )qzo (5)

where u., q. and ~'H are the friction velocity, the 'flux humidity' and the cor- rection parameter to the logarithmic profile resulting from the deviation from neutral stratification, respectively, calculated according to Businger et al. ( 1971 ), p is air density, 2 is the latent heat of evaporation and a~ is the relative contribution of the vegetation to the total latent heat flux between the surface and the atmosphere. It is important to mention that, in this parameterization, the surface consists of one or more layers, depending on the leaf area index (L) of the vegetation. When the soil is bare ( L = 0 ) , the surface consists of one layer which is the soil surface. However, when vegetation is present (L> 0), the surface consists of the soil surface and a number of vegetation layers equivalent to L. Thus, for instance, if L = 3 the surface consists of four layers (three vegetation layers and the soil surface). The global and thermal radiation are absorbed and emitted homogeneously within the canopy, and the leaf temperature and humidity are represented separately by one single value. The number of vegetation layers influences the total amount of global radiation absorbed by the canopy and, as a result, the amount of global radiation which reaches the soil surface. Sensible and latent heat fluxes are also distributed homogeneously within the canopy.

The shielding factor, af, represents the fractional coverage of the grid ele- ment (in the numerical model) by the canopy. This factor is 1 for a completely covered surface and 0 for bare soil. The remaining parameters in eqns. (1-5) are: k, von Karman's constant; qz,, the specific humidity at the first level in the atmosphere (Zl), usually a few meters above the vegetation; q~o, the specific humidity of the surface which has aerodynamic roughness z0 and zero plane displacement d; qva, the specific humidity at the air-vegetation interface, qG, the specific humidity at the air-soil interface; q*, the saturated specific humidity at the vegetation temperature. Finally, gsr is the dimensionless normalized stomatal conductance suggested by Avissar et al. ( 1985 ). Its value ranges between 0 when stomata are completely closed and 1 when they are completely opened.

It should be emphasized that because the surface consists of both vegetation and soil, the specific humidity of the surface (qzo) should account for sources of specific humidity provided by both vegetation and soil. Thus, when stomata are closed (gsr = 0 ), the soil only contributes to the specific humidity of the surface. As a result, the specific humidity of the surface (including the specific humidity at the air-vegetation interface described by eqn. (5) ) is equal to that at the air-soil interface (q~a = qzo = qG )-

The stomatal conductance of Avissar et al. ( 1985 ) is improved here to account for the leaf age

gsr = £2 [gsm -t- (gsM -gsm )fRfrfvfcf~]g~M (6)

IMPACT OF STOMATAL CONTROL ON MESOSCALE ATMOSPHERIC CIRCU LATIONS 3 5 7

where 12 is the leaf age function, gsm is the min imum stomatal conductance which occurs only through the leaf cuticle when the stomata are closed, g~M is the maximum stomatal conductance obtained when stomata are completely opened, and each of t h e f functions quantifies the influence of a specific environmental factor on the conductance (R stands for solar global radiation, T for leaf temperature, V for the vapor pressure difference between leaf and ambient air, C for the ambient atmospheric carbon dioxide concentration and 7-/for soil water potential in the root zone). The expression used for each of these functions is

1 f - (7)

1 +exp[ -S~(X, --Xb~) ]

where the subscript i refers to the environmental factor, X, is the intensity of the factor i, Xbi is the value of Xt at f = 1/2, and St is the slope of the curve at this point. If one of the environmental factors is in the critical range and should cause stomatal closure, then the value of its respective function is 0 and gsr =~dgsm/gsM. When, however, the given environmental factor does not inhibit stomatal opening, the value of its respective function is 1 and it does not influence gsr. When all t h e f functions are equal to 1 (i.e. when there is no environmental constraint on stomatal opening), gsr= g2 and transpiration is maximized for the given meteorological conditions. The constants gsm, g~M, Xbi and St are empirically determined. For example, Avissar et al. (1985) provided their value for a tobacco plant (Table 1 ). For this particular plant, they could not record any sensitivity to carbon dioxide, as indicated by the constants reported in Table 1. Two sets of constants are provided for the temperature. This is because both low and high temperatures affect stomatal conductance. Thus, the func t ionf r is in fact the product of two functions,fro and fr,, which quantify the influence of cold and hot temperatures, respectively, as illustrated in Fig. 1. Further studies are still required to establish whether

TABLE1

Values of Xb~ and S, for functions of stomatal response to environmental factors ( f rom Avissar et al., 1985)

Environmental factor Units Xb~ Si

Global radiation W m - 2 3.5 0.034 CO2 concentration ppm ov - Temperature (cold range ) ° C 8.9 0.41 Temperature (hot range ) ° C 34.8 - 1.18 Vapor pressure difference Pa 2860 - 0.003 l Soil water potential Pa 8 X 105 0.25 × 10- 5


1.00 -

~0.75

o.5o m

03

0.25

0.00 0

1.00 -

~0.75 E

-

"~ 0.50 03

0.25

I 1 I I I I I I I 0.00 100 200 300 400 500 0

S o l a r Radiation (W/ /m 2)

1 . 0 0 , 1 . 0 0 - C

j0.75 p0.75 03 "~ 0.50 03 "~ 0.50 03 03

0.25 0.25

0.00 I I I I I I I I I 0.00

I I = I I 1 10 20 30

T e m p e r a t u r e (*C)

\

4O

0 1 2 3 4 5 VPD (KPo)

I

0 - 5 0 0 -1000 -1500 -2000 S o i l W a f e r P o t e n t i a l (KPo)

Fig, 1. Schematic representation of the stomatal mechanism (expressed dimensionless by the relative stomatal conductance, g,/gs,,~ ) as typically influenced by (a) global solar radiation, (b) leaf temperature, (c) vapor pressure difference (VPD) between the leaf and its environment, and (d ) soil water potential in the root zone. Adapted from Avissar et al. ( 1985 ).

or not these constants are specific for each type of plant and if they are constant within a single canopy.

The leaf age function (t2) varies between 0 (at the beginning and end of the growing season) and 1 (at the optimum growing stage). This type of op- t imum function is represented by the product of two functions, similarly to fTc and fTh.

It might be noticed that with this parameterization of the stomatal mechanism, the 'threshold concept' (schematically illustrated in Fig. 1 ), which is generally 'accepted as describing the stomatal response to the five environmental factors included in eqn. (6) (e.g. Hsiao, 1973; Szeicz et al., 1973; Takakura et al., 1975; Hall et al., 1976; Jarvis and Morison, 1981; among many others), is correctly represented.

Apparently, only five other parameterizations of stomatal mechanism have been used in atmospheric studies. They were suggested by Jarvis (1976) , Deardorff ( 1978 ), Federer ( 1979 ), Dickinson et al. ( 1986 ) and Noilhan and Planton (1989) .

IMPACT OF STOMATAL CONTROL ON MESOSCALE ATMOSPHERIC CIRCULATIONS 359

Jarvis' parameterization (1976) was adopted by Pinty et al. (1989) in mesoscale studies and by Sellers et al. (1986) in a general circulation model (GCM). Similar to Avissar et al. (1985 ), the equation used to describe the stomatal resistance (rs) consists of a product of the stress terms

r~ + c f~-l -1 -1 = f v f(T.,ea) (8)

where a, b and c are empirical constants specific for each type of plant and which are obtained from the measurement of max imum and min imum values of stomatal resistance. In this equation, fR, f L , f r and f(ra.ea) are the adjust- ment factors for the influence of photosynthetically active radiation (PAR), leaf water potential, leaf temperature and atmospheric water vapor deficit (e~-a - e a ) , respectively. The factors are also limited to the range 0-1, but use different types of equations to describe the variation of each factor within this range. The leaf water potential function accounts for the effects of soil moisture stress and excessive evaporation demand. Thus, although the number of equations and empirical parameters required to describe the stomatal mechanism with this formulation is slightly larger than with the parameterization of Avissar et al. (1985), the empirical constants of these two formulations can be set to express almost the same environmental stress on the vegetation. Avissar et al. ( 1985 ) found it more appropriate to relate the stomatal mechanism directly to the environmental factors which originate the stress, rather than incorporating additional empirical functions to 'translate' the stress into the stomatal closure. Therefore, they preferred to use a function which relates the soil water potential to the stomatal closure rather than a function which relates the soil water potential to the leaf water potential (through several equations) and then another function to relate the leaf water potential to the stomatal closure, as in Jarvis (1976).

Deardorff 's parameterization ( 1978 ) was adopted by McCumber ( 1980 ), Garrett ( 1982 ) and Mahfouf et al. ( 1987 ), and was used by Segal et al. ( 1988 ) for comparison with the parameterization of Avissar et al. ( 1985 ). This formulation is simpler than those of Avissar et al. (1985) and Jarvis (1976) since it accounts only for solar radiation and water content in the root zone

r Rmax (~wil, ~ 2 ] rs =re ~0.03Rmax + R - - - t-S+ (9) \?]root,/ ._l

where Rma x is the max imum noon incoming solar radiation which can be achieved, S is a seasonal dependence (somewhat equivalent to the leaf age function g2), ~]wilt is a wilting point value of soil moisture relative to its saturated value and qroot is a soil moisture value in the root zone. The coefficient re is a function of plant type and measures the surface resistance of the plant canopy. However, this formulation does not account for temperature and hu-

3 6 0 R. AVISSAR AND R.A. PIELKE

midity stresses which are known to affect considerably the plant water regime in various regions at various periods of the year.

Federer's parameterization (1979) has been adopted in the model of Wetzel and Chang (1988). The formulation of stomatal resistance, with this ap- proach, is

r~ =rso+f l v (V -Vo)+ f lT (T -To )+ f l , exp(fl2R)+fl~,/(~P-~Uc) (10)

where V, T, R and ~u are atmospheric vapor deficit at screen height, screen temperatures, mean short-wave radiation absorbed by the surface and soil water potential, respectively, and rso, fly, Vo, fiT, To, fl~, f12, fl~" and qsc are spe- cies-dependent empirical parameters. The value of rs is never allowed to ex- ceed fl~, which is the resistance of the leaf cell walls. Thus, this parameterization also accounts for most of the known stressing environmental conditions. It must be noticed, however, that the 'threshold concept' is not represented by this formulation since linear equations are used to describe the stresses. Since the number of empirical constants required to characterize a single plant type with this parameterization is equivalent to that needed to formulate the 'threshold concept' with exponential functions, as in Jarvis ( 1976 ) and Av- issar et al. ( 1985 ), it does not apparently offer any advantage.

The parameterization suggested by Dickinson et al. (1986) is somewhat similar to that of Deardorff since it also accounts only for solar radiation and soil water content

r~ =rsminR, alM1 ( 11 )

where RI, S~, and M~ give the dependence of rs on solar radiation, seasonal temperature and the ability of plant roots to take water readily from the soil for a given level of root moisture, respectively.

Finally, the parameterization of Noilhan and Planton ( 1989 ) is an 'hybrid' of the other formulations. It is defined as

rs=rsminF1F21F~l F 4 ] (12)

where FI and F4 are identical to R~ and S~ of Dickinson et al. ( 1986 ), respectively, F3 is identical to the function f< Ta,ea) of Jarvis ( 1976 ), and 1:2 is a function of the soil volumetric moisture which varies linearly between a critical value (assumed to be 75% of the water content at saturation) and the wilting point volumetric moisture. It should be mentioned, however, that this parameterization is probably the only one in pre-operational use in an operational forecast atmospheric model.

This brief review of the parameterization of plant stomatal resistance (conductance) in mesoscale atmospheric models emphasizes the problem involved in obtaining the large number of empirical constants required to describe all the stresses which are known to have an important effect on the stomatal control of the Bowen ratio. Furthermore, these constants may have


to be redefined for each plant type. Also, considering that a single grid cell of a mesoscale model represents a region of up to 2500 km 2 and that many types of plants are usually found on this scale, it is understandable why modelers simplify their parameterization of the stomatal mechanism. Thus, in all the mesoscale studies which have been carried out so far, the problem of subgrid- scale land and plant heterogeneities has been ignored. Only recently (Wetzel and Chang, 1988; Avissar and Pielke, 1989 ), have first attempts to solve this problem been suggested.

In the next section, a mesoscale numerical model with a sophisticated land surface parameterization is presented. This model demonstrates the importance of the plant stomatal control on regional atmospheric circulations.

TABLE2

Soil properties and initial conditions of the sandy loam soil type which was adopted for the simulations, as provided by the United States Department of Agriculture textural classes

Soil property Value

Density 1250 kg m -3 Roughness 0.01 m qs 0 . 4 3 5 m 3 m -3 ~°cr - 0 . 2 1 8 m Kh, 2.95 m day-1 b 4.90 Albedo 0.20 Emissivity 0.95

Initial profiles

Soil depth (m) Temperature (K) Wetness ( m 3 m - 3 )

0.00 300 0.04 0.01 300 0.04 0.02 300 0.04 0.03 300 0.04 0.04 300 0.04 0.05 300 0.25 0.075 300 0.25 0.125 300 0.25 0.20 300 O.25 0.30 300 0.25 0.50 300 0.25 0.75 30O O.25 1.00 300 0.25

The hydraulic properties are soil water content at saturation (rh), soil water potential at which the water content (r/) starts to be lower than saturation (~¢r), hydraulic conductivity at saturation (Kh,), and b is an exponent in the function which relates soil water potential and water content. A complete description of the soil parameterization is provided in Avissar and Pielke ( 1989 ).

362

TABLE 3

Vegetation input parameters

R. AVISSAR AND R.A. PIELKE

Property Value

Roughness 0.1 m Leaf area index 3 Height 1.0 m Albedo 0.15 Transmissivity 0.125 Emissivity 0.98 Initial temperature 300 K

Root distribution

Soil depth (m) Root fraction

0.00 0.00 0.01 0.00 0.02 0.00 0.03 0.00 0.04 0.00 0.05 0.09 0.075 0.12 0.125 0.14 0.20 0.18 0.30 0.16 0.50 0.16 0.75 0.15 1.00 0.00

THE MODEL

The mesoscale numerical atmospheric model used in the present study consists of an atmospheric and a land surface module. The atmospheric module formulation is described comprehensively in Pielke (1974), Pielke and Mah- rer (1975), Mahrer and Pielke (1977a, 1978), and McNider and Pielke ( 1981 ). Validation of the model has indicated its ability to realistically re- solve mesoscale atmospheric features (e.g. studies by Pielkeand Mahrer, 1978; Segal et al., 1981; Abbs and Pielke, 1986; among others). The land surface module is described in detail in Avissar and Mahrer (1988a) and Avissar and Pielke (1989). It has been validated against a field experiment (Avissar et al., 1986 ) and has proven its ability to accurately describe the microclimate of homogeneous agricultural fields under various environmental conditions and at different plant growing periods. This module was used by Avissar and Mahrer (1988b), Segal et al. ( 1988 ) and Avissar and Pielke ( 1989 ) to study the effects of vegetation on different mesoscale features under various atmos-

IMPACT OF STOMATAL CONTROL ON MESOSCALE ATMOSPHERIC CIRCULATIONS

TABLE 4

Atmospheric input parameters and initial conditions

363

Input parameter Value

Number of vertical grid points Number of horizontal grid points (2D cases) Horizontal grid increment Synoptic flow for 2D simulations Synoptic flow for 1D simulations Initial surface temperature Initial potential temperature laPse

19 25

6 km 0 . 5 m s -~ 5 .0ms - l

300 K 3.5 Kper 1000m

Specific humidity profile

Height (m) Humidity (g kg -~ )

5 10.0 15 10.0 30 10.0 60 i0.0

100 10.0 200 10.0 400 10.0 70O 10.0

1000 8.0 1500 6.O 2000 2.5 2500 1.5 3OOO 1.0 3500 0.5 4000 O.5 5OO0 0.5 6000 0.5 8000 0.5

10000 0.5

TABLE 5

General model input parameters

Input parameter Value

Integration time step 60 s Latitude 37°N Day of the year August 5 Initialization time 02: 00 h LST


pheric stability conditions. For brevity, the formulation of these modules will not be described again here and the reader is referred to the aforementioned studies. The formulation of the latent heat flux and stomatal conductance in this model, however, is described in the section "Parameterization of plant stomata in atmospheric models", eqns. ( 1 ) - ( 7 ).

For the purpose of this study, the model was initialized with the parameters provided in Tables 2-5. The initial soil water content profile (Table 2 ) was set to the so-called 'field capacity' which represents the soil humidity a few hours after the end of a heavy rain. Thus, the root zone is wet and plants can use the soil water for transpiration. However, the soil surface was kept dry in order to avoid evaporation and emphasize the effects of stomatal conductance and transpiration. The vegetation characteristics (Table 3 ) are appropriate to many agricultural crops, and the initial atmospheric vertical profiles of potential temperature and specific humidity (Table 4) reflect summer conditions in the subtropical regions.

NUMERICAL EXPERIMENTS

Surface fluxes

In order to demonstrate the influence of stomatal control on the surface heat fluxes, the model was run in its one-dimensional version for several 18 h simulations, starting at 02: 00 h LST. For each simulation, a relative stomatal conductance between 0 and 1 was imposed and kept constant during the simulation. Thus, eqns. (6) and (7) were not applied into the model for these numerical experiments. The maximum values of the diurnal variations of sensible heat flux and latent heat flux predicted for each simulation are presented in Fig. 2. The strong non-linear relationship between the stomatal con-

500

~" 400 E

300

200

IO0

0

F i i I

. LATENT'-- ...... / -----SENSIBLE

O.O O 0.5 0.75 I.O RELATIVE STOMATAL CONDUCTANCE

Fig. 2. The influence of relative stomatal conductance on the daily max imum surface heat fluxes simulated with a b iosphere -a tmosphere model.


ductance and the partitioning of energy at the surface is evident. In particular, it is interesting to note that half of the max imum transpiration possible under the climatological conditions selected here is obtained with a relative conductance of 0.2, i.e. when the stomatal aperture is only 20%! The production of sensible heat flux above the canopy is relatively small and does not vary much when the relative conductance is between 0.50 and 1.0. However, at low relative conductance (below 0.25) a small modification of the stomatal aperture contributes significantly to the sensible heat flux.

The influence of various parameters (e.g. roughness, leaf area index, pho- tometric properties, air humidi ty and temperature, etc. ) on the relationship presented in Fig. 2 was also investigated. Although the absolute values of the fluxes might increase or decrease according to the characteristics of the simulation, the aforementioned remarks are also applicable to these simulations. The results illustrated in Fig. 2 can, therefore, be generalized.

The one-dimensional version of the model (including eqns. ( 6 ) - ( 7 ) ) was also used to demonstrate the effects of soil drying on the modification of the diurnal behavior of plant stomata and its resulting effects on the surface energy fluxes. For this purpose, a simulation of a 30 day period was executed. The results of the first 15 simulated days are illustrated in Fig. 3.

At the beginning of the simulation, owing to the combination of the initial values of soil water potential, air temperature and air humidi ty imposed on the biosphere-atmosphere system, the maximum relative stomatal conduc-

=~ ,ooL (a) ~ 1o3

1 co - 2 5 ~ - - S O I L WATER P O T E N T I A L 1 _ - , .- , . . . . . --.VEGETATION TE~,PE"ATU,RE 123

I 2 3 4. 5 6 7 8 9 I0 I I 12 13 [4

~ 1.0 (b) - - PARTIAL N I G H T CLOSURE r 0.7'5}'- f - - ' ~ - - - C O M P L E T E N I G H T CLOSURE

cou w ~ 0 5 > c : , ~ § °.23

O i i L I I I I I I I I I I , 2 ~ 4 3 6 7 8 9 ,0 ,, ,2 ,3 ,,~

I-~'~ 500F (c) ' A ' A ' A ' A ' a . . . . B LATENT HEAT FLUX '

~FA II II II /\ fl A ^ ---SENS,,,L~HE,,T~LUX ~23o~/ / / / / / / / / / / / /~ /t", ~',, ~" ," t, ;" ,,, ~,23~1..,/ /._.~ /../ /_~ M /../ /,~ /t i IAi I l i Ai / ! /~ /~

0 i i i i L I I I I I I L I " r 2 3 4 5 6 7 8 9 I 0 I I 12 13 14

D A Y O F S I M U L A T I O N

Fig. 3. Biosphere-atmosphere interactions obtained from the simulation of a vegetated sandy loam soil in a subtropical region, for summer conditions. (a) Soil water potential and vegetation temperature; (b) relative stomatal conductance; (c) latent andsensible heat fluxes.


tance is only 0.75. Carbon dioxide did not affect stomatal behavior in this numerical experiment and solar radiation was high enough during the daytime hours to allow a full opening of the stomata. It must be remembered that solar radiation intensity significantly affects the plant canopy energy balance and its temperature. Therefore, too much solar radiation can excessively warm the leaves and, as an indirect result, close the stomata.

Interestingly, in laboratory experiments carried out in order to validate their model of the stomatal mechanism, Avissar et al. ( 1985 ) could not observe a total closure of the stomata when their plants were kept in the dark. As a result, when none of the environmental parameters inhibit stomatal opening, the stomata remain partially opened during the night (Fig. 3 (b) ). Since these unexpected observations require more investigation in order to draw general conclusions, the simulation described in this section was re-executed assum- ing a total stomatal closure during the night (Fig. 3 (b)) . Although the resulting night-time stomatal behavior was very different during the first days of the two simulations, it did not affect at all the diurnal variations of the surface heat fluxes and of the surface temperature and is, therefore, not of relevance for the processes studied here.

Figure 3(a) shows the gradual decrease (almost linear) of the soil water potential in the root zone. This is the result of the combined effect of plant water uptake for transpiration and water drainage under the root zone resulting from gravity. After five simulated days, the soil water potential enters the stomatal 'stress range', resulting in a water shortage during midday hours when transpiration demand is high. As a result, the stomata close (this phenomenon is often referred to in the literature as 'mid-day closure' ), the leaf temperature increases significantly (Fig. 3 (a) ) and sensible heat flux dominates the latent heat flux (Fig. 3 (c)) . Progressively, the 'mid-day closure' extends to several hours a day as the water stress becomes stronger, with the continual depletion of water in the root zone. At the beginning of the second simulated week (Days 8 and 9), the stomata remain partially open only during the morning hours, depicting a half-day latent heat flux and half-day sensible heat flux (Fig. 3 (c)) . After the second simulated week, there is not enough water in the root zone to allow even a partial stomatal aperture. The surface heat fluxes resemble those obtained over a bare dry soil.

It must be emphasized that the soil type adopted for this simulation (Table 2 ) has a relatively high hydraulic conductivity which caused a rapid drainage of the water from the root zone and, consequently, accelerated plant stress. When a 'heavy' soil (i.e. a soil with low hydraulic conductivity) is simulated, it takes much more time to drain the soil and water is available in the plant root zone for a longer period.

Mesoscale atmospheric circulations

The two-dimensional version of the model was used to demonstrate the impact of stomatal control on mesoscale atmospheric circulations. The do-


main simulated in the numerical experiments presented in this section consists of 25 horizontal grid cells, each representing a subdomain of 6 km width. Thus, a region 150 km wide is simulated. The vegetated part of this domain (vegetation is assumed to completely cover the soil surface) is 60 km wide ( 10 grid cells) and is juxtaposed to a 90 km wide bare soil. Additional characteristics for the simulations are provided in Tables 2-5.

The meteorological fields presented in Fig. 4 (time is 14:00 h LST) are derived from a simulation in which the plants did not experience any stress and, therefore, the stomata remained open all day (i.e. gsr = 1 ). By this hour, the circulation is strongly developed, as indicated by the intensity of the horizontal component of the wind parallel to the domain (u), the vertical component of the wind (w) and by the thermal contrast implied by the field of potential temperature (0). The component of the wind perpendicular to the domain is about 0.15u at this hour and thus is not of significance for illustra- tion purposes. Only later in the day does its intensity increase significantly as a result of the Coriolis effect. A shallow planetary boundary layer (PBL) de- velops above the vegetated area, as can be seen from the 0 field. Transpiration of the vegetation provides a supply of moisture which significantly increases the amount of water vapor present in the shallow PBL. This moisture is ad- vected by the thermally induced flow toward the dry region where it is well mixed within the relatively deep convective PBL. As already discussed by Mahfouf et al. (1987), Segal et al. (1988) and Avissar and Pielke (1989), this 'vegetation breeze' is similar in intensity and characteristics to a sea breeze.

The max imum values of u, w, surface temperature and height of the PBL above the vegetation, obtained from different simulations for which a relative stomatal conductance between 0 and 1 was imposed and kept constant during the simulation, are presented in Fig. 5.

When the relative stomatal conductance is 0.0 (i.e. stomata are completely closed), the flow over the domain is very similar to the weak flow prescribed at the initialization of the simulation. The small vertical component of the wind which has been generated during the simulation is related to the contrast of roughness between the vegetated and bare parts of the domain. The height of the PBL above the vegetation (Fig. 5 (c ) ) is similar to that above a bare dry soil surface, and the canopy temperature (Fig. 5 (d ) ) is almost as high as that of a bare dry soil surface. Increasing the relative stomatal conductance to a value of 0.25 considerably modifies the atmospheric fields in the simulated domain. The horizontal and vertical components of the wind are, respectively, almost 80 and 85% of those of the max imum values obtained with a relative stomatal conductance of 1.0. The height of the PBL is about 1400 m for a relative stomatal conductance of 0.25, as compared to a height of about 800 m obtained with a value of 1.0 and a height of about 2400 m obtained with a value of 0.0. Canopy temperature decreases from a high of 53 to 41 and 33°C for relative stomatal conductances of 0.0, 0.25 and 1.0, respec-

3ooo

~ ...

......

.. ;,

~,,.

~,.

......

,

sooo

, .

....

~', o

'~;

-''

-':

-" ~

....

aooo

- o

,' (/

////

~ J

2000

1 J~

-~

F lc

) )

A /~

.°._

---~

q • --

'

~ O

1 ;,'

-' ..

....

J'

-/'/-"

----

'-"f/

....

....

I

~oo[

...

.. P'

' '4'

~'/'

/,;-;~

',gA'

~ ...

...

l 30

00

~ ...

......

. ->

"--",'

~--'-

XW;

....

2 t

ll

ll

^

_

N .oo

ol I I,!

,?",'/,'

l/lllAk

tlt/t

I ~

~ooo

~__.~ ~/~,.,-~

' ~/,

";,,

,,?,,'/

f/~./1

i11~/

t

~,oo

o i

," ~,,

o

o ...

......

1,

2:,I,

,l ...

. I

o ...

...

Q~,

,t.

,,,

....

3Okm

Fi

g. 4

. Ver

tical

cro

ss-s

ecti

on o

f the

sim

ulat

ed d

omai

n at

14:

00 h

LST

for:

(a)

the

hor

izon

tal w

ind

com

pone

nt p

aral

lel

to t

he d

omai

n (u

) in

m

~-',

posi

tive

fro

m le

ft t

o ri

ght;

(b)

the

ver

tica

l w

ind

com

pone

nt (

w)

in c

m s

-',

posi

tive

upw

ard;

(c)

the

pot

enti

al t

empe

ratu

re (

0) i

n K

; (d

) he

spe

cifi

c hu

mid

ity

(q)

in g

kg-

', r

esul

ting

fro

m t

he ju

xtap

osit

ion

of 6

0 km

of d

ense

uns

tres

sed

vege

tati

on (

indi

cate

d by

the

dar

k un

derb

ar)

lnd

90 k

m o

f bar

e dr

y la

nd. T

he v

eget

atio

n co

mpl

etel

y co

vers

the

left

-han

d so

il su

rfac

e. S

olid

con

tour

s in

dica

te p

osit

ive

valu

es;

dash

ed c

onto

urs

ndic

ate

nega

tive

val

ues.

%

10

I M P A C T O F S T O M A T A L C O N T R O L O N M E S O S C A L E A T M O S P H E R I C C I R C U L A T I O N S 369

f i n

E

8.0

7.0

6.0

5.0

4.0

3.0

2.0

1.0

0.0

i i i i i

(a)

I I I I ! 0.0 025 0.5 0.75 1.0

RELATIVE STOMATAL CONDUCTANCE

0.4 }- (b)

0.3F

" o2[ E

O.II

o.oi- J I I I

o.o 0.5 o.75 ,.b RELATIVE STOMATAL CONDUCTANCE

2 5 0 0

2000 E

1500

I looo

500

i i i i

(c)

o.o o15 o% ,.b RELATIVE STOMATAL CONDUCTANCE

55 (d)

~ 5O

35[ 3O

I ~ I L I 0.0 O. 5 0.5 075 1.0

RELATIVE STOMATAL CONDUCTANCE

Fig. 5. The influence of relative stomatal conductance on the development of mesoscale circulation generated by the juxtaposition of 60 km of dense vegetation which completely covers the soil surface and 90 km of bare dry land. The values reported here are the maximum values, at 14:00 h LST, of (a) the horizontal wind component parallel to the domain, (b) the vertical wind component, (c) the height of the planetary boundary layer above the vegetation and (d) the surface temperature.

tively. Thus, as expected, based on the results of the one-dimensional simulations described in "Surface fluxes", stomatal behavior does not contribute significantly to mesoscale circulations as long as the normalized stomatal conductance does not drop below 0.50. However, a small modification of the stomatal aperture below this critical value considerably modifies the intensity and characteristics of mesoscale circulations.

CONCLUSIONS

Plant stomata control the Bowen ratio of vegetated surfaces. Since contrasts of sensible heat flux at the Earth's surface can generate relatively strong atmospheric circulations as well as directly influence lower troposphere ther- modynamic stability, it is important to understand the stomatal mechanism


and to represent it adequately in models that are aimed at simulating biosphere-atmosphere interactions.

The various simulations executed under the present study suggest that as long as the plants are under partial stress, it is probably not necessary to describe the stomatal mechanism with very sophisticated mathematical models. That also implies that the evaluation of the min imum resistance (maximum conductance) is not a major parameter for mesoscale atmospheric simulations. However, it is very important to identify conditions of strong stresses and to describe correctly the stomatal behavior when the plants experience such stresses.

Finally, it must be noticed that the two-dimensional simulations described in this paper were initialized with nearly calm atmospheric conditions (i.e. 0.5 m s- ~ ) in order to emphasize the local circulations which may develop as a result of extensive contrasts of surface sensible heat flux. As discussed by Mahrer and Pielke (1978) and recently by Avissar and Pielke (1989), ini- tializing the atmosphere with synoptic-scale winds which advect cool near- surface air towards a warmer land surface will result in a weaker mesoscale circulation as a result of the spreading of the horizontal mesoscale pressure gradient force. In contrast, a synoptic wind from the warmer land towards the cooler surface will tighten the mesoscale horizontal pressure gradient force and result in a stronger mesoscale circulation.

ACKNOWLEDGMENTS

This research was supported by the National Science Foundation (NSF) under grant No. ATM-8616662, by the New Jersey Agricultural Experiment Station project No. 13105 and by the Research Council of Rutgers Univer- sity. Computat ions were performed using the National Center for Atmos- pheric Research (NCAR) CRAY Computer (NCAR is supported by NSF). The authors wish to thank D.G. Saltz for making valuable comments on the manuscript.

This paper was written as a direct result of a workshop on stomatal resistance held at Pennsylvania State University, 10-13 April 1989. The workshop was made possible principally through a grant from the National Science Foundation (BSR-8822164).

REFERENCES

Abbs, D.J. and Pielke, R.A., 1986. Thermally forced surface flow and convergence patterns over northeast Colorado. Monthly Weather Rev., 114:2281-2296.

Anthes, R.A., 1984. Enhancement of convective precipitation by mesoscale variations in vege- tative covering in semiarid regions. J. Clim. Appl. Meteorol., 23: 541-554.


Avissar, R. and Mahrer, Y., 1988a. Mapping frost-sensitive areas with a three-dimensional local scale model. Part I: Physical and numerical aspects. J. Appl. Meteorol., 27:400-413.

Avissar, R. and Mahrer, Y., 1988b. Mapping frost-sensitive areas with a three-dimensional local scale model. Part II: Comparison with observations. J. Appl. Meteorol., 27:414-426.

Avissar, R. and Pielke, R.A., 1989. A parameterization of heterogeneous land surface for atmospheric numerical models and its impact on regional meteorology. Monthly Weather Rev., 117:2113-2136.

Avissar, R., Avissar, P., Mahrer, Y. and Bravdo, B.A., 1985. A model to simulate response of plant stomata to environmental conditions. Agric. For. Meteorol., 34:21-29.

Avissar, R., Dagan, N. and Mahrer, Y., 1986. Evaluation in real time of the actual evapotranspiration using a numerical model. Agrotique 86, Automation and Robots for Agriculture, Bordeaux, pp. 255 -263 (available from the first author ).

Businger, J.A., Wyngaard, J.C., Izumi, Y. and Bradley, E.F., 1971. Flux-profile relationships in the atmospheric surface layer. J. Atmos. Sci., 28:181-189.

Deardorff, J.W., 1978. Efficient prediction of ground surface temperature and moisture, with inclusion of layer of vegetation. J. Geophys. Res., 83: 1889-1903.

Dickinson, R.E., Henderson-Sellers, A., Kennedy, P.J and Wilson, M.F., 1986. Biosphere-At- mosphere Transfer Scheme (BATS) for the NCAR Community Climate Model. National Center for Atmospheric Research Note, NCAR/TN-275 + STR, Boulder, CO, 69 pp.

Estoque, M.A. and Gross, J.M., 1981. Further studies of a lake breeze. Part II: Theoretical study. Monthly Weather Rev., 109: 619-634.

Federer, C.A., 1979. A soil-plant-atmosphere model for transpiration and availability of soil water. Water Resour. Res., 15: 555-562.

Garrett, A.J., 1982. A parameter study of interactions between convective clouds the convective boundary-layer, and forested surface. Monthly Weather Rev., 110: 1041-1059.

Hall, A.E., Schulze, E.D. and Lange, O.L., 1976. Current perspectives of steady-state stomatal response to environment. In: O.L. Lange, L. Kappen, and E.D. Schulze (Editors), Water and Plant Life. Springer Verlag, Berlin, pp. 169-188.

Hsiao, T.C., 1973. Plant response to water stress. Annu. Rev. Plant Physiol., 24:519-570. Idso, S.B., 1981. Surface energy balance and the genesis of deserts. Arch. Meteorol. Geophys.

Bioklimatol. Ser. A, 30: 253-260. Jarvis, P.G., 1976. The interpretation of the variations in leaf water potential and stomatal

conductance found in canopies in the field. Philos. Trans. R. Soc. London Ser. B, 273: 593- 610.

Jarvis, P.G. and Morison, J.I.L., 1981. The control of transpiration and photosynthesis by stomata. In: P.G. Jarvis and T.A. Mansfield (Editors), Stomatal Physiology. Cambridge Uni- versity Press, Cambridge, pp. 248-279.

Kuo, H.L., 1968. The thermal interaction between the atmosphere and the earth and progaga- tion of diurnal temperature waves. J. Atmos. Sci., 25: 682-706.

Mahfouf, J.F., Richard, E. and Mascart, P., 1987. The influence of soil and vegetation on the development ofmesoscale circulation. J. Clim. Appl. Meteorol., 26: 1483-1495,

Mahrer, Y. and Pielke, R.A., 1976. Numerical simulation of the air flow over Barbados. Monthly Weather Rev., 104: 1392-1402.

Mahrer, Y. and Pielke, R.A., 1977a. A numerical study of the air flow over irregular terrain. Contrib. Atmos. Phys., 50:93-113.

Mahrer, Y. and Pielke, R.A., 1977b. The effects of topography of the sea and land breezes in a two-dimensional numerical model. Monthly Weather Rev., 105:1151-1162.

Mahrer, Y. and Pielke, R.A., 1978. A test of an upstream spline interpolation technique for the advective terms in a numerical mesoscale model. Monthly Weather Rev., 106: 818-830.

McCumber, M.C., 1980. A Numerical Simulation of the Influence of Heat and Moisture Fluxes upon Mesoscale Circulations. Ph.D. Dissertation, University of Virginia, Charlottesville, VA.


McCumber, M.C. and Pielke, R.A., 1981. Simulation of the effects of surface fluxes of heat and moisture in a mesoscale numerical model. J. Geophys. Res., 86: 9929-9938.

McNider, R.T. and Pielke, R.A., 1981. Diurnal boundary-layer development over sloping terrain. J. Atmos. Sci., 38: 2198-2212.

Neumann, J. and Mahrer, Y., 1971. A theoretical study of the land and sea breeze circulations. J. Atmos. Sci., 28: 532-542.

Noilhan, J. and Planton, S., 1989. A simple parameterization of land surface processes for meteorological models. Monthly Weather Rev., 117: 536-549.

Ookouchi, Y., Segal, M., Kessler, R.C. and Pielke, R.A., 1984. Evaluation of soil moisture effects on generation and modification of mesoscale circulations. Monthly Weather Rev., 11: 2281-2292.

Otterman, J., 1975. Possible Rainfall Reduction through Reduced Surface Temperature due to Overgrazing. NASA Rep. Tech. Inf. Div. Code 250, Goddard Space Flight Center, Green- belt, MD.

Physick, W., 1976. A numerical model of the sea-breeze phenomenon over a lake or gulf. J. Atmos. Sci., 33: 2107-2135.

Pielke, R.A., 1974. A three-dimensional numerical model of the sea breezes over south Florida. Monthly Weather Rev., 102:115-139.

Pielke, R.A., 1984. Mesoscale Meteorological Modeling. Academic Press, New York City, 612 PP.

Pielke, R.A. and Mahrer, Y., 1975. Technique to represent the heated-planetary boundary layer in mesoscale models with coarse vertical resolution. J. Atmos. Sci., 32: 2288-2308.

Pielke, R.A. and Mahrer, Y., 1978. Vertification analysis of the University of Virginia three- dimensional mesoscale model prediction over south Florida for July 1, 1973. Monthly Weather Rev., 106: 1568-1589.

Pinty, J.P., Mascart, P., Richard, E. and Rosset, R., 1989. An investigation of mesoscale flows induced by vegetation inhomogeneities using an evapotranspiration model calibrated against HAPEX-MOBILHY data. J. Appl. Meteorol., 28: 976-992.

Segal, M., Mahrer, Y. and Pielke, R.A., 1981. Numerical study of wind energy characteristics over heterogeneous terrain - central Israel case study. Boundary-Layer Meteorol., 22: 373- 392.

Segal, M., Avissar, R., McCumber, M. and Pielke, R.A., 1988. Evaluation of vegetation effects on the generation and modification ofmesoscale circulations. J. Atmos. Sci., 45: 2268-2292.

Sellers, P.J., Mintz, Y., Sud, Y.C. and Dalcher, A., 1986. A simple biosphere (SiB) for use within general circulation models. J. Atmos. Sci., 43:505-531.

Szeicz, G., van Bavel, C.H.M. and Takami, S., 1973. Stomatal factor in water use and dry mat- ter production by sorghum. Agric. Meteorol., 12:361-389.

Takakura, T., Goudriaan, J. and Louwerse, W., 1975. A behavior model to simulate stomatal resistance. Agric. Meteorol., 15: 393-404.

Wetzel, P.J. and Chang, J.T., 1988. Evapotranspiration from nonuniform surfaces: A first ap- proach for short-term numerical weather prediction. Monthly Weather Rev., 116:600-621.

Yamada, T., 1982. A numerical model simulation of turbulent airflow in and above canopy. J. Meteorol. Soc. Jpn., 60: 439-454.

Documents

The impact of plant stomatal control on mesoscale ... › 2009 › 09 › r-110.pdf · The impact of plant stomatal control on mesoscale atmospheric circulations. Agric. For. Meteorol.,