(introductory text...)

Abstract
Introduction
General circulation models
Response to variation in land surface properties
GCM Simulations of tropical rainfall
Recommendations for future research
Symbols and abbreviations
References
Assessment

P. R. Rowntree

Abstract

General circulation models (GCMs) are briefly described together with the parametrizations employed. Aspects especially relevant to changes in vegetation are emphasized. Factors such as surface albedo, ground hydrological processes, and aerodynamic surface roughness, which are affected by these changes, are considered in relation to their contribution to the sensitivity of the models. Results are summarized for experiments on both global and regional scales, simulating changes of surface albedo and surface water availability.

A removal of vegetation tends to increase albedo and reduce roughness and affects surface moisture availability by changing interception, runoff, and the depth of soil accessible to the roots. Increases in surface albedo decrease evaporation and also tend to reduce atmospheric moisture convergence and precipitation. Decreases in surface moisture availability reduce evaporation, and this generally leads to a decrease of rainfall, which helps to maintain the surface moisture anomaly. A decrease in surface roughness also affects the partitioning of upward energy flux between the sensible and latent (evaporation) forms, with evaporation increasing as roughness decreases for large stomata! resistances and decreasing for small.

Introduction

This paper reviews the atmospheric general circulation model as a tool for study of the climatic response to changes in vegetation in the tropics. It concentrates on (a) the models' treatments of land surface processes, which are crucial to estimations of the climatic response to variations in vegetation, and (b) the sensitivities of the models as revealed by experiments with perturbations of these properties, mainly the surface solar albedo and moisture availability. The quality of the model simulations in the tropics is briefly reviewed, with the emphasis on rainfall over continents, for which a good simulation is most important in sensitivity experiments.

It should be noted that for studies of global climatic change it is not appropriate to assume that sea surface temperatures are unaffected. However, because of the general inferiority to date of simulations of the atmosphere with coupled ocean-atmosphere models, such perturbation experiments have been made mostly with prescribed sea surface temperatures. The review of GCM parametrizations of land surface processes is necessarily brief compared with that of Carson (1982). Similarly, reference to Dickinson (1980) is recommended for a specific discussion of the likely effects of tropical deforestation on climate.

General circulation models

Basic Structure

Atmospheric GCMs solve the three-dimensional, time-dependent equations for the rates of change of surface pressure, wind components, temperature, and moisture content, taking account of sources and sinks of heat, moisture, and momentum. These equations are usually expressed either in finite-difference form for a three-dimensional array of points or in spectral form in each layer of the model. Whichever form is used, the source and sink terms, representing the physical parametrizations, are calculated at grid points. As the land surface processes, which are central to this review, are part of these, the distinction between spectral and finite-difference models need not concern us further. The basic equations with the source and sink terms on the right-hand side may be written:

(1)

Here t , H. E are the vertical fluxes (positive upward) of momentum, heat, and moisture due to vertical subgrid-scale eddies (turbulence and convection), R. is the net downward radiative flux, P is the precipitation or net sink of moisture, and a is the vertical coordinate, being p/r _*, or pressure normalized by surface pressure r _*. Other notation is conventional as given in the list of symbols.

Fluxes at the Land-Atmosphere Interface

The land surface affects the atmosphere through the fluxes R_N, t , H. and E at the bottom boundary (s = 1) of the atmospheric model. The treatment of these fluxes in GCMs has been reviewed by Carson (1982). Though an attempt to emulate Carson's review would be too lengthy for the present purpose and an unnecessary duplication, a summary of the present state of the art may be useful, especially as some information not available to Carson can be included. In this section, we shall first present and discuss the basic formulation for each of the fluxes and then summarize the parametrizations in the main GCMs in current use, as known to the reviewer. These models are listed in table 1. Some models for which no new information is available are omitted here. These include the Australian Numerical Meteorology Research Centre model, now adopted, with some changes, as the NCAR (National Centre for Atmospheric Research) "Community Climate Model," and the Oregon State University (OSU), Siberian Academy of Sciences Computing Centre, and Main Geophysical Observatory models. These have only low vertical resolution and cannot be expected to give a realistic representation of the near-surface processes important in assessing the response to surface perturbations. However, brief descriptions of the OSU model and also the ECMWF (European Centre for Medium Range Weather Forecasting) forecast model are at the end of this paper.

Radiative Fluxes. (A) Basic formulations. The net downward radiative flux at the land surface

(2)

where

is the emissivity or absorptivity of the surface for radiation of wavelength and

and B(l ) represent the downward and black body fluxes at wavelength lambda. However, in atmospheric radiative transfer, we are concerned mainly with two principal streams of radiation, solar radiation at l = 0.2µm-4µm and terrestrial radiation at l = 3µm-100µm (these wavebands contain over 99% of the radiation from each source). It is therefore convenient to make the twostream approximation and write:

(3)

where s _s is Stefan's constant and T₀ the surface temperature (degrees Kelvin). R_s(0) is the downward solar radiation and a _* R_s(0) is the reflected solar flux. Note that a _*, the albedo, is not a constant for a given surface but depends on the spectral distribution of the incoming solar radiation, which will depend on atmospheric constituents, including cloud, and on the zenith angle of the sun. Thus, formally

(4)

Similarly, the emissivities for terrestrial radiation are means over the appropriate range of wavelengths weighted by the intensity at each wavelength. For the downward flux, this is again dependent on the atmospheric structure (temperature, humidity, and cloud), whilst for the upward radiation it is, for given surface conditions, dependent only on the surface temperature because of the Planck function's temperature dependence.

TABLE 1. GCM models

Label	Centre	Vertical resolution	Horizontal representation	Reference
AES	Atmospheric Environment Service, Downsview, Ontario, Canada	5-layer	Spectral 20 wave rhomboidal	Boer and McFarlane 1979
EERM	l'Etablissement d'Etudeset de Recherches Météorolo- giques, Toulouse, France	10-layer	Spectral 10-13 waves trapezoidal	Royer et al. 1981
GFDL	Geophysical Fluid Dynamics Laboratory, Princeton, New Jersey, USA	9-layer (also 11 and 18)	Spectral 15, 21, or 30 waves rhomboidal (also gridpoint versions)	Manabe et al. 1979
GISS	Goddard Institute for Space Studies, New York, New York, USA	9-layer	Grid point 4° latitude x 5° longitude (also 8° x 10°, 12° x 15°)	Hansen et al. 1983
GLAS	Goddard Laboratory for Atmospheric Sciences, Greenbelt, Maryland, USA	9-layer	Grid point 4° latitude x 5° longitude	Randall 1983a
LMD	Laboratoire de Meteorologie Dynamique, Paris, France	11-layer	Grid point 5.6° east-west 50 point pole-pole (cost coordinate)	Sadourny 1983
MO	Meteorological Office, Bracknell, England	11-layer	Grid point 2.5° latitude x 3.75° longitude	Mitchell and Bolton 1983
NCAR	National Center for Atmospheric Research, Boulder, Colorado, USA	9-layer	Spectral 15 waves rhomboidal	Pitcher et al. 1983

^aRandall describes two versions of the GLAS model, referred to in this paper as GLAS 1980 and GLAS 1982.

(B) Surface albedo in GCMs. We shall discuss here only the snow-free albedo; snow cover is allowed to increase the albedo in all the GCMs except that the LMD model prescribes a geographical variation taking account of the climatological snow cover. For snow-free albedo, many models, including the AES, GFDL, GLAS, and NCAR models, use the geographical distribution specified by Posey and Clapp (1964), who specified a low albedo (0.07) for tropical forests. As discussed by Rowntree (1975) and Dickinson (1980), this is due to the use of a visible albedo, but there is a strong spectral variation with much larger values for infrared wavelengths, and the mean value for the whole spectrum is about 0.125. Other geographical distributions, used in the LMD model due to Bartman (1981) and the ECMWF model due to Preuss and Geleyn (1980), using satellite data, also have values below 0.1 over parts of the tropical continents. Probably the most satisfactory treatment is that in the GISS model, with a detailed specification of albedos for eight land surface types as classified by Matthews (1983). It allows for seasonal variation and separate values for the visible and nearinfrared. The desert albedo has a moisture dependence that can halve the albedo to 0.175 for a moist surface. Relatively simple formulations are used in the EERM model, with a _* = 0.31-0.17 h_u where h_u is a non-linear function of soil moisture content, and in the MO model, with a. = 0.2, though a vegetation-dependent variation due to M. F. Wilson (personal communication) has recently been tested in the latter.

(C) Surface emissivity in GCMs. Generally a value of 1.0 has been used for emissivity in GCMs. However, the model II of GISS (Hansen et al. 1983) uses realistic emissivities for deserts, snow, and ice. The NCAR model uses values less than unity at 812 µm; the EERM model uses 0.95.

Surface Turbulent Fluxes. (A) Basic formulations. The turbulent fluxes at the surface may be formally written

(5)

Here the formal definition of the flux of a quantity X is given first as

, where the primes (') indicate deviations from a time-mean, then the formulation commonly used in GCMs

, with C_x being a turbulent transfer coefficient depending on surface roughness and atmospheric stability; ro and

are near-surface density and windspeed and OX is an estimate of the excess of a surface value X₀ over the nearsurface atmospheric value

is generally taken as zero.

The right-hand terms in equation 5 for heat and moisture transfer are alternative expressions for the fluxes in resistance form. For heat this combines an atmospheric resistance r_H and the temperature gradient d q ; setting q ₀ at the surface temperature is observed to give reasonable estimates. With a wet surface such as a lake, the same is true of the moisture flux with the same, purely atmospheric, resistance r_E and with q₀ = q_sat (T₀), the saturation specific humidity at the surface temperature. However, for moisture transfer over land there is often an additional surface resistance due to the vegetation's stomata! resistance to transfer from moist surfaces within the leaf. This resistance is very large in arid regions and also at certain times in other regions (e.g. at night).

(B) GCM specifications of transfer coefficients. As discussed by Carson (1982), there is a wide range of complexity in the specifications of transfer coefficients. The GFDL and AES models use very simple forms (C_D = C_H = C_E = 2 and 3 x 10^-3 respectively) and do not distinguish between land and sea; similarly simple roughness and stability dependent formulations are used in the EERM, GISS, and MO models. Intermediate, relatively simple forms that depend only on surface type (land or sea) and windspeed are used in the NCAR and LMD models. The roughness lengths (z₀) used in the MO and EERM models over land are fixed (10 and 16 cm respectively), whilst in the GISS model they depend not only on the vegetation type but also on the orographic roughness.

(C) GCM specifications of near-surface variables. The models may be divided into two groups by their specifications of the near-surface values of V_s. q _s, q_s. The simplest approach of taking the values for the lowest model level is used in the GFDL, NCAR, and MO models; for all of these this level is at 70-100 metres above the surface (s = .987-.991).

Values for the EERM model's lowest level (s = .95) are also used directly, except that the wind is turned through an angle dependent on thermal stability, wind speed, and Coriolis parameter. Some such turning is probably beneficial to the accuracy of the surface stress computation even with the lowest level at nearly 100 m. Hansen et al. (1983) found that with realistic cross-isobar angles of surface flow, the ITCZ was sharpened, with decreased rainfall over the southern Sahara and increased Hadley cell mass flux.

Apart from the Randall (GLAS 1982) version of the GLAS model (see Randall 1982), the other models (GLAS 1980, AES, GISS) derive the near-surface values of q_s, and tetas by assuming that the surface flux (eqn. 5) equals a diffusive flux, which, for specific humidity q_s is of the form

, with K a stability dependent diffusion coefficient and

expressed as the gradient between q_s and the value for the lowest level. One might expect the form for temperature to be

, but for the AES model it is

with c = 5° C km^-1, whilst for the GISS model it is said to be

. In the AES model V_s is also obtained by flux continuity, while in the GLAS 1980 and earlier GISS models, downward extrapolation is used. In the latest GISS model, an Ekman formulation is used with explicit allowance for turning of the wind as discussed above. In middle latitudes this formulation gave increased eddy kinetic energies with warming of high latitudes.

(D) GCM specifications of surface variables. The surface temperature T₀ derived from the subsurface thermal parametrization discussed in the next section, is used for q _c in all the models. The limitation of evaporation in arid conditions is usually allowed for by calculating a potential evaporation E_p from equation (5c) with q₀ = q_sat(T₀) and obtaining the actual evaporation E from

(6)

with b a function of W the normalized soil moisture content W = (m/m_crit), where m is the soil moisture content for the top soil layer, and mcrit the lowest m for which E = E_p, and b = 1 for W>1. The calculation of m is discussed under (B) in Subsurface Processes. In the GFDL, MO, LMD, GISS, and AES models b = W, whilst in the EERM model, a weighted combination of two calculations of E is used, the weights depending on the vegetative cover such that with full vegetation cover b = h_u = 0.9 W² (3-2W). In the NCAR model, b is set to a constant value (0.25). The linear formulation (b = W) has been criticized by Mintz (personal communication) because in arid conditions it gives an excessive value for E_p. To appreciate the problem, it is instructive to consider an alternative formulation of evaporation, that using the PenmanMonteith equation (e.g. Monteith 1973). Equation is written in the resistance form

(7)

Here r_aE is the atmospheric resistance and r_s is the surface resistance (for water vapour).

Because of the difficulty of observing T₀, it is desirable to eliminate it from equation (7). In the Penman-Monteith approach, this is achieved by using the surface energy balance:

(8)

where d T = T₀ - T(z_e), G is downward heat flux into the soil, and r_aH is an atmospheric heat resistance analogous to r_aE.

By defining

so that

d T (and so T₀) can be eliminated to give the Penman-Monteith equation:

Priestley and Taylor (1972) analysed observations of drying surfaces using equation (9) and obtained a formulation like (6). However, it differs from (6) because their estimate of the potential evaporation in

(10)

depends only weakly on T₀ and hence on soil moisture. In contrast, in GCMs T₀, and hence qsat (T₀) and so also E_ps, all increase rapidly as the soil dries, so (6), as used in GCMs, is not consistent with observations. One solution to this problem, as proposed by Mintz and Serafini (1981) and used in the 1982 version of the GLAS model, is to use a separate wet surface energy balance to compute the surface temperature needed for the calculation of E_p. Randall (1983) used b = 1 - exp ( - 6.8 W) with this formulation. Randall reports that this formulation gave a considerable reduction in evaporation over subtropical deserts. In the GCM experiments with the 1982 GLAS model, the ground wetness data were based on climatology instead of depending on modelled precipitation and evaporation.

An alternative solution to Mintz and Serafini's may be to introduce the surface resistance r_s explicitly in equation (5c). In one practical application of this approach with observed data, Thompson, Barrie, and Ayles (1982) allow r_s to depend on the minimum stomatal resistance, leaf area index, and available water capacity of different types of vegetation as well as on soil moisture. Evaporation of precipitation intercepted by vegetation is included. An approach of this kind could be used in a GCM. It would allow incorporation of geographical distributions of vegetation characteristics and soil types and the use of multilayer soil models.

Subsurface Processes. (A) Basic formulations. To specify the surface temperature and soil moisture variables needed for calculating the surface turbulent fluxes, some representation of subsurface processes is needed. For a layer of ground between depths z and z + d z, neglecting horizontal subsurface transfers of heat and water, and defining z, G, and the water flux M as positive downward

(11)

(12)

where m is the soil water content, Q_g and N are source and sink terms. For the heat budget (11), the only significant heat sources are due to moisture phase changes. In the moisture budget (12), provided we consider m (and M) to refer to the sum of water vapour and liquid water contents (and flux), there are no sources and sinks except those associated with melting and freezing. With this definition, the "surface" is strictly the interface between vegetation and air and it is there that the surface boundary condition

(13)

applies. The terms in parentheses represent the net contribution of surface and atmospheric processes (rainfall P_R and snowmelt M_S less surface runoff Y(0)) to the downward water flux in the soil, whilst E(0) is the evapotranspiration, which may be partly a sink at the soil surface but in the presence of vegetation also takes water out of the soil throughout the root zone and transfers it to the atmosphere throughout the canopy. For GCMs, with a lowest layer of order 100 m in depth, it is probably unnecessary to apply the boundary condition for the atmospheric model so as to allow for this. However, the distribution of the sink in the soil needs to be taken into account by allowing direct transfer of water to the atmosphere from layers throughout the root zone.

With heat fluxes the heat capacity of the vegetation is probably small enough to justify use of

(14)

as the surface boundary condition.

The treatment of the subsurface moisture fluxes is discussed in the paper by Dooge chap 7). The subsurface heat flux G(z) can be represented by a diffusive term of form

(15)

Typical values of the heat capacity C and conductivity lambdag are given by Geiger (1965); both depend markedly on moisture content so that changes in moisture content due to changes in vegetation or climate can affect G(0). In the tropics G(0) is generally small on seasonal time scales but its diurnal variation can be large for dry surfaces.

(B) GCM treatments of subsurface hydrological processes. Most models represent subsurface water content, generally called soil moisture, by a single variable that is updated according to equation (13), with surface runoff generally represented by limiting the soil moisture to a maximum value (field capacity), generally 100 to 200 kg m^-2. Exceptions to this are the NCAR model (no surface hydrology) and the GLAS and GISS models, with more elaborate parametrizations.

The GISS model has two layers whose water capacities depend on the vegetation characteristics, with diffusion between the layers. During development of the model the upward diffusion coefficient was made infinite during the mid-latitude growing season and all the year in lower latitudes (except deserts) to represent the ready accessibility of water in the deeper soil to vegetation. Together with more rapid downward diffusion of water to the lower layer, this change delayed the limitation of evaporation due to soil dryness by about two months. Runoff is allowed before field capacity is reached, with runoff taken as 0.5 WP_R. The model was quite sensitive to variations in the coefficient; doubling it to unity reduced summer continental temperatures by 7-8 K and increased global mean runoff by 70 mm y^-1.

The 1980 version of the GLAS model treats soil moisture in a way similar to other models but allows runoff of a fraction of the precipitation, this fraction being a nonlinear function of the soil moisture deficit (see Carson 1982).

(C) GCM treatments of subsurface thermal processes. The treatment of subsurface thermal processes in GCMs may be conveniently described in terms of a thermal capacity C. equal to p_gCD as obtained by vertical integration of equation (11) through a depth D. For the GFDL and NCAR models, C_* = 0, so that the substantial diurnal variation of heat storage cannot be represented. Most other models (see Carson 1982 for more details) use a value of C_* roughly appropriate for the diurnal variation of surface temperature by taking

, where

is the e-folding depth of the diurnal temperature wave (v ₀ = 2p per day). In the MO model, D = d is used, which, with a typical value of C_* gives C_* ~ 21 J cm^-2 K^-1. The LMD model uses C_* ~ 18 J cm^-2 K^-1, whilst in the AES model, l _g and C are specified functions of soil moisture.

Two of the models (EERM and GISS) use two layer treatments with thermal capacity (and also conductivity in the GISS model) dependent on soil moisture. These two-layer treatments allow representation of the seasonal cycle of soil heat flux, though this is unlikely to affect simulations significantly in the tropics.

Response to variation in land surface properties

Radiative

Surface Solar Atbedo (a *). The surface albedo is of major importance in determining the absorption of solar energy. Large variations are possible due to vegetation. Generally albedo decreases for a given vegetation type as the height of the vegetation increases because of internal reflections, though for short, sparse vegetation this may not be true if the albedo of the soil is low relative to that of the vegetation. Albedo also decreases generally for soil and vegetation as the surface wetness increases. The range of albedo for snow-free conditions is from about 0.1 for tropical forest to about 0.4 for some dry, sandy surfaces. With incident mean daily solar fluxes typical of the tropics (300-400 Wm^-2 in cloudless conditions), a variation of about 100 Wm^-2 is possible. However, this would be an upper limit to spatial variations and is only conceivable at one place with an extreme climatic change or extensive human intervention such as deforestation or irrigation on highly reflective soil. Actual changes on a large scale (1,000 km and greater) seem unlikely to exceed about half this magnitude, or 50 Wm^-2. These could be due to variations in surface wetness (Idso et al. 1975, Norton, Mosher, and Hinton 1979) or replacement of forests by grassland or dry soil.

Much larger variations in albedo can occur due to snow cover, which may have an albedo of over 0.9. This is of significance in the context of deforestation in middle and high latitudes where, because snow seldom covers trees for long, the effects of snow on the heat budget are less in forested regions. In lower latitudes, because of the lack of a large seasonal variation, forests are unlikely to exist above the snow-line.

Variations in surface albedo may be expected to affect local climate in two ways through their modification of the net radiation. Firstly, by equation (14), a reduction in R_N(0) must decrease the energy available for upward transfers of sensible and latent heat and for downward transfer into the soil. The Penman-Monteith equation (9) may be used to estimate the variations of the components of the turbulent fluxes as (R_N - G) varies. For typical tropical temperatures of 26°C,

and

so that

(16)

For given d q/r_aE, the change in LE will be a fraction x = 3/(4 + r_s/r_aE) of the change in (R_N - G). The value of x varies from about 0.75 for a moist surface to zero for a dry surface. Thus, as (R_N - G) decreases, the evaporation will decrease: for incoming solar radiation of 250 W m^-2, a change in albedo of 0.1 will decrease (Rx - C) by 25 W m^-2 if C is unchanged, reducing evaporation by 0.65 mm d^-1 if r_s = 0. Whether relative humidity will be decreased or increased will depend on both temperature and r_s/r_aE. If it is decreased, precipitation is likely to be reduced (see discussion below in Surface Moisture Availability).

The second mechanism by which surface radiative characteristics can modify tropical climate is that discussed by Charney (1975), whereby heating of a vertical column (surface and atmosphere) relative to adjacent regions can increase the ascent of air masses and vice versa. Thus the geographical distribution of heat sources and sinks leads to vertical circulations that are mainly responsible for limiting the atmospheric temperature gradients to values in line with the dynamical constraint of the low rotation rates found in the tropics. Thus an increase in surface albedo, by weakening the heat source, tends to reduce the ascent of air masses and the associated rainfall. This mechanism can be enhanced by the drying of the atmosphere (which increases upward long wave radiative flux) and drying of the land surface, which can cause surface heating and so also increase upward long wave radiation. These may be counterbalanced by a decrease in cloudiness, reducing reflected solar radiation more than increasing upward long wave radiation.

An impression of the possible impact of albedo changes is gained by noting that typically in the tropics the net radiative heating of the surface of ~ 150 Wm^-2 is offset by an atmospheric radiative cooling of ~ 100 Wm^-2. With clear-sky solar radiation of 400 Wm^-2 a 0.1 increase in albedo will thus eliminate most of the net heating.

GCM Experiments with Surface Albedo Changes. For a more quantitative estimate of the effects of albedo changes we must consider the results of GCM experiments. However, in doing so it must be remembered that the complexity of the interactions, between radiation, cloud, and atmospheric and surface moisture content, requires a degree of realism in modelling surface and cloud processes that is probably not yet attainable. The GCM experiments made to date, therefore, cannot be expected to give more than an indication of the likely impacts.

Studies with GCMs on the effects of albedo variations, some of which include albedo changes in more than one area, are listed in table 2. All are for northern summer. Except as indicated the models included interactive surface hydrology and radiative transfer dependent on modelling water vapour and cloud.

Although the global-scale change in albedo studied by Carson and Sangster is not a realistic change, their results are instructive in showing clearly the major features of a model's response. In their first experiment (fig. 1a) albedos of all snow-free land were increased from 0.1 to 0.3; the land was kept wet to avoid confusing soil moisture-albedo interactions. The major effects were:

1. higher surface pressure over land (by up to 12 mb in the inner northern continents);
2. decreased average evaporation over land (from 3.6 to 2.7 mm d^-1);
3. decreased average precipitation (and ascent) over land (from 4.6 to 3.4 mm d^-1) with compensating increases over oceans (from 2.9 to 3.2 mm d^-1).

Note that the decrease in precipitation exceeded that in evaporation, showing it was partly due to a decrease in moisture convergence. These results are consistent with the above discussion.

Similar results were obtained in the second experiment (fig. 1b), in which soil moisture was interactive. In both experiments —but especially the second—some land areas, mostly on the western edge of continents (western North Africa, Europe), were wetter with the higher albedo, probably because the higher pressures over the land mass to the east decreased the equatorward advection of dry air.

The other experiments in table 2 all involved albedo increases over relatively small areas. The principal results are:

1. evaporation was decreased in all cases;
2. precipitation was decreased in all experiments except two, where changes were small: both were near mountain ranges (Rockies, Himalayas) and poleward of latitude 20°;
3. moisture convergence was decreased in all but three cases, they being poleward of latitude 20°.

Although for Chervin's (1979) experiment there is no direct evidence concerning (i) and (iii), indirect evidence is provided by the maps, showing decreased ascent and reduced soil moisture with increased albedo, especially over the Sahara.

For the areas equatorward of 20° latitude that should be most relevant to tropical deforestation, the ratios of the fractional change in rainfall to the change in albedo (last column of table 3) average - 2.1, suggesting a 21% decrease in rainfall for a 0.1 increase in albedo. On average, for the limited area anomalies more than half the change is due to moisture convergence, though there is considerable variability in this, particularly between Charney et al.'s (1977) results and those of Sud and Fennessy (1982).

TABLE 2. Experiments on effect of surface albedo changes

Reference	Area	Latitude	Averaging period	Albedo		Comments
				Control	Modified
Carson and Sangster 1981	Global	—	90 days	0.1	0.3	1, 2, 4
	Global	—	90 days	0.2	0.3	1, 2
Charney et al. 1977	Sahel	12°-16°N	31 days	0.14	0.35	4
	NW India	24°-32°N
	Great Plains	32°-48°N
	Central Africa	8°-12°N
	Bangladesh	20°-28°N
	Mississippi	32°-48°N
Sud and Fennessy 1982	Sahel	12°-20°N	31 days	0.183	0.3
	NW India	24° 32°N	31 days	0.15	0.3
	NE Brazil	4°-24°S	31 days	0.091	0.3
	Great Plains	32°-48°N	31 days	0.129	0.3
Chervin 1979	Sahara	7.5°- 37.5°N	60 days	.08-.35	.45
	W. USA	27.5°-52.5°N	60 days	.07-.17	.45
Henderson- Sellers and Gornitz 1984	Brazil	15.6°S-7.8°N	5 years	.11	.17	3, 5

Comments:
1. No moisture feedback in radiation
2. No cloud feedback in radiation
3. Roughness length and ground water capacity also changed
4. Evaporation set equal to potential evaporation
5. Ocean surface temperatures and sea ice not prescribed

An important aspect of the results is the effect of albedo on cloud amount. Here also Sud and Fennessy's and Charney et al.'s results differ, the latter finding decreases in cloudiness of 7 to 24% in five of their six cases, whereas Sud and Fennessy obtained decreases of at most 4%. In consequence, radiative impacts were greater in Sud and Fennessy's experiment (30 W m^-2 in net surface radiation compared with 17), yet rainfall was less affected.

FIG. 1a. Excess of precipitation with an albedo of 0.1 relative to that with an albedo of 0.3 at Days 21 to 110 (soil moisture fixed at 15 cm).

FIG. 1b. Excess of precipitation with an albedo of 0.2 relative to that with an albedo of 0.3 (interactive initially zero soil moisture).

TABLE 3. Effects of albedo

Experiment	Change in albedo (d a.)	Change in evap. (d E) (mm/d)	Change in rainfall (d R) (mm/d)	Change in moisture convergence (d R-d E) (mm/d)	d R/R	d R/R d a .
Charney et al. 1977
Sahel	.21	- 0.9	- 3.4	- 2.5	- .46	- 2.2
NW India	.21	- 0.5	- 2.6	- 2.1	- .53	- 2.5
Great Plains	.21	- 1.0	- 1.5	-0.5	- .41	-2.0
Central Africa	.21	- 0.7	- 3.1	- 2.4	- .62	- 3.0
Bangladesh	.21	- 0.2	0	0.2	0	0
Mississippi	.21	- 1.6	- 1.1	0.5	- .25	- 1.2
Sud and Fennessy 1982
Sahel	.12	- 0.7	- 1.5	-0.75	- .26	- 2.1
NW India	.15	-0.2	-0.5	-0.3	-.13	-0.9
NE Brazil	.21	- 0.3	- 0.5	- 0.25	- .24	- 1.1
Great Plains	.17	-0.2	0	0.25	.01	0.1
Chervin 1979
Sahara	- .22	-	~ - 2.5	-	~ - .4	~ - 1.8
W. USA	- .33	-	~ - 1	-	~ - .15	- 0.5
H.-S. and Gornitz 1984
Brazil	.06	-0.45	-0.6	- 0.15	~ - .12	- 2.0
Carson and Sangster 1981
0.1 (r) 0.3	.2	- 0.95	- 1.22	-0.27	- .27	- 1.3
0.2 (r) 0.3	.1	-	- 0.4	-	- .14	- 1.4

It is to be expected that an increase in surface albedo will tend to reduce the mean global temperature. However, of the available experiments only that of HendersonSellers and Gornitz can show this since ocean surface temperatures were prescribed in the others. Although no details of the geographical distribution are available, a mean cooling of 0.10 K was obtained over the last year of their experiment (HendersonSellers, personal communication). This is a little greater than expected with the prescribed mean albedo change of 8 x 10^-4 from results of experiments with one dimensional models (e.g. Sagan, Toon, and Pollack 1979), which suggest cooling of about 1 K for a surface albedo increase of 0.01. A rather larger response is to be expected from three-dimensional models (e.g. Manabe and Wetherald 1975) because of the possibility of snow/ice albedo feedback.

Carson and Sangster (as reported by Rowntree 1982) made experiments with albedo dependent on soil moisture through a simple linear relation from 0.15 for a moist to 0.30 for a dry surface. This provides a positive feedback to any tendency for change in surface moisture and may be expected to enhance contrasts between wet and dry regions. This expectation was confirmed with sharpened contrasts between southern India and north-western India/Pakistan and across the Sahel by days 21 to 50; later, however, there was a northward expansion of rainfall over western North Africa, probably associated with the higher albedo over much of Africa and Asia for the reasons suggested above in connection with Carson and Sangster's albedo experiments. Laval (1983) has also recently run experiments with soil moisture dependent albedos in the LMD model, as well as with increased prescribed Sahel albedos; with the latter she noted a tendency for weakening of the upper tropospheric easterlies in the Sahel region.

Surface Long Wave Emissivity. The effect of a change in surface long wave emissivity on the net surface radiation may be written:

(17)

Typically in the tropics

(e.g. Viswanadham 1972), so that if

and

then

A survey of data on surface emissivities was presented by Kondratyev et al. (1982), who commented on the inadequate spectral detail of the available observations. Their limited data suggest that the emissivity integrated over the long wave spectrum is about 0.92 for soils and 0.94 for vegetation, while for the 9-12µm "water vapour window," the mean for dry soils is 0.96 and for vegetation 0.98. Whilst the relative uncertainties in these figures must be substantial, the suggested value of 0.02 for d e _* for both the integrated emissivity appropriate for upward radiation and the water vapour window, which is more appropriate for the net long wave radiation, is so small that one may have sufficient confidence in the consequent estimate of

to assert that the effects of deforestation on the net radiation through emissivity changes can be neglected in comparison to the changes due to surface albedo. The only relevant GCM experiment known to the writer is that reported briefly by Hansen et al. (1983), who found little effect on the general circulation or atmospheric temperatures" from the introduction of spectrally dependent emissivities for deserts, snow, and ice.

Continue

GCM Simulations of tropical rainfall

The aspect of GCM simulations of most relevance to studies of the impact of tropical vegetation changes is rainfall. An assessment has therefore been made of the GCM simulations of rainfall. The observed precipitation according to Jaeger (1976) (from Randall 1982) is shown in figure 3a, c.

In addition to the models listed in table 1, the long established OSU model and the ECMWF medium-range forecast model are included here. The OSU model (Schlesinger and Gates 1980) is a two-layer model with 4° latitude by 5° longitude grid. The boundary-layer parametrization uses constant flux layer assumptions like the GISS model. The model albedos are from Posey and Clapp (1964) and there is a one-layer ground temperature simulation. The hydrology has recently been changed from that described by Carson to one like the GISS model's (Ghan et al. 1982). The version of the ECMWF model used by Tiedtke (1983) for two 50-day experiments is a 1 7/8° by 1 7/8° model with 15 layers. The physics include a one-layer soil treatment but with diffusive linking of both moisture and temperatures to prescribed deep soil values (Tiedtke et al. 1979). The surface and boundary layer fluxes are based on Monin-Oboukhov similarity theory and are dependent on thermal stability and mixing lengths.

In order to relate them to tropical deforestation, the following summary of each model's success in predicting the observed rainfall pattern is restricted to northern South America, Africa, and southern Asia.

AES:	(January only) wet areas too far north over all three continents with excessive rain north of the equator over west and east Africa and parts of Asia
ECMWF:	good agreement in shape both in February and July. Too wet over north east South America and southern India in February and over all India in July, when the African rainbelt is also too intense and a few degrees too far north
EERM:	generally too dry with most of the tropical land below 3 mm/day rainfall
GISS:	too wet over North Africa and southern Asia in January. In July (8° by 10° model) rain too far north over Africa and Arabia with eastern South America and Africa also too wet
GLAS 1980:	(see fig. 3) good generally; rain too far north over Africa in February
GLAS 1982:	African rains too extensive to north and south in January. July African and Indian rainfall about 10° too extensive to the north
LMD:	rain belts too narrow and intense but generally well-centred, except dry area over north India in July
MO:	good agreement in January except Somalia and south India too wet; generally right July pattern over South America and Africa but too dry, northern India also too dry
NCAR:	northern South America and southern Asia too wet both in January and July; ahara too wet in July
OSU:	patterns generally good (except western Sahara wet in July); rain too intense in January over South America and especially Africa
GFDL:	wet over South-East Asia in January and Sahara in July; most maxima Rhomboidal 30) too intense
GFDL:	July pattern good except northern India too dry; India rather wet in (250 km grid point) January; South American maxima too intense January and July.

Overall, the spectral models, especially those of lower resolution, perform less well than the grid point models. There is less evidence of a dependence of quality on resolution for grid point models, with the 4° by 5° GLAS 1980 model probably the best. However, some aspects of that model's simulation in the tropics are less realistic: for example, the upper troposphere is too warm and the flow tends to be too westerly, especially in the upper troposphere near, and south, of the equator in July (Randall 1983).

FIG. 3a. February observed precipitation, taken from Jaeger (1976). The contour interval is 1 mm d^-1

FIG. 3b. February simulated precipitation (GLAS). The contour interval is 1 mm d^-1 except in congested regions.

FIG. 3c. July observed precipitation, taken from Jaeger (1976). The contour interval is 1 mm d^-1.

FIG. 3d. July simulated precipitation (GLAS). The contour interval is 1 mm d1 except in congested regions.

Recommendations for future research

For detailed surveys of the availability of relevant data sets and requirements for future observational, theoretical, and modelling work, reference should be made to the appropriate sections of Eagleson (1981) and ICSU/WMO (1983, sect. 5 and annexes). Space precludes more than a brief statement of the more important needs. In the following, these needs are classified into three types: (a) observational studies; (b) improvements in GCM parametrizations; and (c) GCM experiments.

Albedo

1. Data sets on actual surface albedo are needed both for GCM experiments and to provide a basis for future assessments of changes (a).
2. The dependence of vegetation albedo on moistness of foliage and of soil should be studied to allow specification of these dependences in GCMs (a).
3. The dependence of albedos on soil moisture as well as on vegetation should be included in GCMs on the basis of the above studies and used in vegetation change experiments (b, c).

Hydrology

1. Research is needed to overcome the fundamental parametrization problems of transfer of theory and models based on microscale observations to the GCM grid scale, allowing for inhomogeneities of soils, slopes, vegetation, precipitation, etc. (a,b).
2. Particular efforts are needed on the parametrization of runoff, which is generally very crudely represented in GCMs (b).
3. The evaporation of intercepted rainfall and the effects of root depth variations need to be included in GCMs to allow proper assessments of the sensitivities to vegetation changes (b).
4. GCM experiments are needed to assess the sensitivities to variations in runoff and root depths associated with vegetation changes (c).

Roughness

1. Assessments of GCM formulations in relation to the surface stress are desirable in view of the large sensitivity to its formulation found in the GISS model by Hansen et al. (1983) (b).
2. Sensitivity experiments are needed to estimate the GCM response to perturbations of surface roughness (z₀) (c).
3. the actual variations of z₀ are poorly known. Observational studies of the relation of z₀ to vegetation are required if GCMs are found to be sensitive to it (a).

General

The simulation of cloudiness in GCMs is generally poor. Special attention should continue to be given to this aspect of GCMs to represent realistically the feedbacks between clouds and surface parameters (b).

Symbols and abbreviations

C	soil thermal capacity
C0	soil thermal capacity = pgCD
Cs	turbulent coefficient depending on surface roughness and atmosphericstability
cp	specific heat of air
D	soil depth
E	vertical moisture flux or evaporation
Ep	potential evaporation
G	downward heat flux into soil
G(z)	subsurface heat flux
H	vertical heat flux
K	stability dependent diffusion coefficient
	unit vertical vector
L	latent heat of condensation
M	water flux in soil
Ms	snowmelt
N	soil source/sink water term (melt/freeze)
P	precipitation or net sink of moisture
PR	rainfall
Qg	soil source/sink heat term (moisture phase changes)
R	gas constant of air
	downward radiative flux at wavelength A
	downward longwave radiative flux
Rs	net vertical radiative flux
RN(0)	net vertical radiative flux at land surface
Rs0)	downward solar radiation flux at land surface
T	temperature
To	surface temperature
	horizontal wind vector
	near-surface windspeed
W	soil moisture content
Y(0)	surface runoff
Z	depth from surface in air/soil
V	vector gradient operator
( )	mean
( )'	deviation from mean
e *­	emissivity
e *¯	absorptivity
f	Coriolis parameter
g	force due to gravity
hu	non-linear function of soil moisture content
m	soil moisture content for top soil layer
p	pressure
p*	surface pressure
q	specific humidity
raE	atmospheric resistance to water vapour transfer
raH	atmospheric resistance to heat transfer
rs	surface resistance
t	time
w	vertical velocity (dz/dt)
ze	height of lowest model layer
z0	roughness length
a *	reflectivity of surface
b	function of normalized soil moisture
s	vertical co-ordinate value p/p*
d q	atmosphere near surface saturation deficit
d T	temperature gradient
X	excess value of surface value X0, over mean-surface value Xs
q	potential temperature
x	ratio of gas constant to specific heat at constant pressure
l	wavelength of radiation
l g	soil thermal conductivity
p	density
pg	soil density
s s	Stefan's constant
t	upward momentum flux
f	geopotential
v 0	diurnal frequency
*AES	Atmospheric Environment Service
ANMRC	Australian Numerical Meteorology Research Centre
CCSAS	Siberian Academy of Science Computing Centre
ECMWF	European Centre for Medium Range Weather Forecasting
*EERM	l'Etablissement d'Etudes et de Recherches Météorologiques
*GFDL	Geophysical Fluid Dynamics Laboratory
*GISS	Goddard Institute for Space Studies
*GLAS	Goddard Laboratory for Atmospheric Sciences
*LMD	Laboratoire de Météorologie Dynamique
*MO	Meteorological Office
*NCAR	National Centre for Atmospheric Research
OSU	Oregon State University
UCLA	University of California, Los Angeles

References

Bartman, F. L. 1981. Time-variable earth's albedo mode/ characteristics and applications to satellite sampling errors. NASA Contr. Rep. 165781.

Boer, G. J., and N. A. McFarlane. 1979. "The AES atmospheric general circulation model." Report of the JOC study conference on climate mode/e, GARP No. 22, 1: 409-460.

Carson, D. J. 1982. "Current parametrizations of land surface processes in atmospheric general circulation models." In P. S. Eagleson, ea., Lond-surface processes in atmospheric general circulation mode/e, pp. 67-108. Cambridge University Press, Cambridge, UK.

Carson, D. J., and A. B. Sangster, 1981. The influence of land-surface albedo and soil moisture on general circulation model simulations. Numerical Experimentation Programme Report No. 2, 5.14-5.21.

Charney, J. G. 1975. "Dynamics of deserts and droughts in the Sahel." Quart. J. R. Met. Soc., 101: 193-202.

Charney, J. G., W. J. Quirk, S. H. Chow, and J. Kornfield. 1977. "A comparative study of the effects of albedo change on drought in semi-arid regions." J. Atmos. Sci., 34: 1366-1385.

Chervin, R. M. 1979. "Response of the NCAR general circulation model to changed land surface albedo." Report of the JOC study conference on climate models Performance, intercomparison and sensitivity studies, 1: 563-581.

Cunningtion, W. M., and P. R. Rowntree. 1986. "Simulations of the Saharan atmosphere: Dependence on moisture and albedo." Quart. J. R. Met. Soc., 112: 971-999.

Dickinson, R. E. 1980. "Effects of tropical deforestation on climate." In Plowing in the wind. Deforestation and long-range implications, pp. 411-441. Studies in Third World Societies, no. 14, College of William and Mary, Dept. of Anthrop., Williamsburg, Va., USA.

Eagleson, P. S., ed. 1981. Report of the JSC study conference on land surface processes in atmospheric general circulation models, Greenbelt, Maryland, USA, 5-10 January, 1981. ICSU/WMO Publication, WCP-46, Geneva.

Geiger, R. 1965. The climate near the ground. Harvard University Press, Cambridge, Mass., USA.

Ghan, S. J., J. W. Lingaas, M. E. Schlesinger, R. L. Mobley, and W. L. Gates. 1982. A documentation of the OSU two-level atmospheric general circulation model Report No. 35, Climate Research Institute, Oregon State University, Corvallis, Oregon, USA.

Hansen, J., et al. 1983. "Efficient three-dimensional global models for climate studies: Models I and 11." Mon. Weath. Rev., 111: 609-662.

Henderson-Sellers, A., and V. Gornitz. 1984. "Possible climatic impacts of land cover transformations, with particular emphasis on tropical deforestation." Climatic Change, 6: 231-257.

ICSU/WMO. 1983. Report of the fourth session of the Joint Scientific Committee, Venice, I-8 March, 1983. World Climate Research Programme, Geneva.

Idso, S. B., R. D. Jackson, R. J. Reginato, B. A. Kimball, and F. S. Nakayama. 1975. "The dependence of bare soil albedo on soil water content." J. Appl. Met., 14: 109-113.

Jaeger, L. 1976. Monatskarten des Niederschlags für die ganze Erde. Berichte des Deutschen Wetterdienstes, 18, No. 139.

Kondratyev, K. Ya., V. I. Korzov, V. V. Mukhenberg, and L. N. Dyachenko. 1982. "The short wave albedo and the surface emissivity." In P. S. Eagleson, ea., Land surface processes in atmospheric general circulation models, pp. 463-514. Cambridge University Press, Cambridge, UK.

Kurbatkin, G. P., S. Manabe, and D. G. Hahn. 1979. The moisture content of the continents and the intensity of summer monsoon circulation. Soviet Meteorology and Hydrology No. 11, pp. 16.

Laval, K. 1983. "GCM experiments with surface albedo changes." Paper presented at Third International School of Climatology, Erice, Italy, October 1983.

Lockwood, J. G., and P. J. Sellers. 1982. "Comparison of interception loss from tropical and temperate vegetation canopies." J. Appl. Met., 21: 1405-1412.

Manabe, S. 1975. "A study of the interaction between the hydrological cycle and climate using a mathematical model of the atmosphere." Summary of presentation at meeting on weather-food interactions, Massachusetts Institute of Technology, Cambridge, Mass., USA, 4-11 May, 1975.

Manabe, S., D. G. Hahn, and J. L. Holloway. 1979. "Climate simulations with GFDL spectral models of the atmosphere: Effect of spectral truncation." Report of the JOC study conference on climate models, GARP No. 22, 1: 41-94.

Manabe, S., and R. T. Wetherald. 1975. "The effects of changing the solar constant on the climate of a general circulation model." J. Atmos. Sci, 32: 2044-2059.

Matthews, E. 1983. "Global vegetation and land use: New high-resolution data bases for climate studies." J. Appl. Met., 22: 474-487.

Mintz, Y. 1984. "The sensitivity of numerically simulated climates to land-surface boundary conditions." In J. T. Houghton, ea., Global climate, pp. 79-105. Cambridge University Press, Cambridge, UK.

Mintz, Y., and V. Serafini. 1981."Monthly normal global fields of soil moisture and land-surface evapotranspiration." Symposium on Variations in the Global Water Budget, Oxford, UK., 10-15 August, 1981.

Mitchell, J. E B., and J. A. Bolton. 1983. "Some differences between the Met. 0.20 5- and 11-layer model annual cycle integrations." Proceedings of ECMWF Workshop on Intercomparison of Large-Scale Models Used for Extended Range Forecasts, Reading, UK., pp. 193-223.

Monteith, J. L. 1973. Principles of environmental physics E. Arnold, London.

Norton, C. C., E R. Mosher, and B. Hinton. 1979. "An investigation of surface albedo variations during the recent Sahel drought." J. Appl. Met., 18: 1252-1262.

Pitcher, E. J., et al. 1983. "January and July simulations with a spectral general circulation model." J. Atmos. Sci. 40: 580-604.

Posey, J. W., and P. E Clapp. 1964. "Global distribution of normal surface albedo." Geofisica Intl., 4: 33-48.

Preuss, H. J., and J. E Geleyn. 1980. "Surface albedos derived from satellite data and their impact on forecast models." Arch. Met. Geoph. Biokl., Ser. A., 29: 345-356.

Priestley, C. H. B., and R. J. Taylor. 1972. "On the assessment of surface heat flux and evaporation using large-scale parameters." Mon. Weath. Rev., 100: 81-92.

Randall, D. A. 1982. "Performance of the PBL parametrizations in the GLAS and UCLA models." Proceedings of the ECMWF Workshop on the Planetary Boundary Layer, Reading, UK., pp. 81-118.

———. 1983. "Monthly and seasonal simulations with the GLAS climate model." Proceedings of the ECMWF Workshop on Intercomparison of Large-Scale Models Used for Extended Range Forecasts, Reading, UK., pp. 107-166.

Rowntree, P. R. 1975. The representation of radiation and surface heat exchange in a general circulation model. Met. 0.20 Tech. Note 11/58, Meteorological Office, Bracknell, UK.

———. 1982. Sensitivity of general circulation models to fond surface processes. Met. 0.20 Tech. Note 11/192, Meteorological Office, Bracknell, UK.

Rowntree, P. R., and J. A. Bolton. 1983. "Simulation of the atmospheric response to soil moisture anomalies over Europe." Quart J. R. Met. Soc., 109: 501-526.

Royer, J. F., M. Deque, H. Canetti, and M. Boulanger. 1981. Présentation d'un modèle spectral de circulation générale a faible résolution. Note de travail de l'Etablissement d'Etudes et de Recherches Météorologiques, Direction de la Météorologie No. 16, 124 pp.

Sadourny, R. 1983. "Some aspects of the performance of the LMD general circulation model in January and July simulations." Proceedings of the ECMWF workshop on Intercomparison of Large-Scale Models Used for Extended Range Forecasts, Reading, UK., pp. 167-192.

Sagan, C., O. B. Toon, and J. B. Pollack. 1979. "Anthropogenic albedo changes and the earth's climate." Science, 206: 1363-1368.

Schlesinger, M. E., and W. L. Gates. 1980. "The January and July performance of the OSU two-level atmospheric general circulation model." J. Atmos Sci, 37: 1914-1943.

Shukla, J., and Y. Mintz. 1982. "Influence of land-surface evapotranspiration on the earth's climate." Science, 215: 1498-1501.

Sud, Y. C., and M. Fennessy. 1982. "A study of the influence of surface albedo on July circulation in semi-arid regions using the GLAS GCM." J. Clim., 2: 105-125.

Thompson, N., I. A. Barrie, and M. Ayles. 1982. The Meteorological Office rainfall and evaporation calculation system: MORECS (July 1981). Hydrological Memorandum No. 45, Meteorological Office, Bracknell, UK.

Tiedike, M. 1983. "Winter and summer simulations with the ECMWF model." Proceedings of the ECMWF Workshop on Intercomparison of Large-Scale Models Used for Extended Range Forecasts, Reading, UK., pp. 263-313.

Tiedtke, M., J. F. Geleyn, A. Hollingsworth, and J. E Louis. 1979. ECMWF model: Parametrization of sub-grid scale processes. ECMWF Technical Report No. 10., Reading, UK.

Viswanadham, Y. 1972. "Studies on radiation balance at a tropical station." Pure and Appl. Geoph., 97: 183-213.

Walker, J., and P. R. Rowntree. 1977. "The effect of soil moisture on circulation and rainfall in a tropical model. " Quart. J. R. Met. Soc., 103: 29-46.

Webb, E. K. 1975."Evaporation from catchments." In T. G. Chapman and F. X. Dunin, eds., Prediction in catchment hydrology, pp. 203-236. Australian Academy of Science.

Assessment

The current indications from simulations using general circulation models are that the climatic effects of any changes by man's management of the surface vegetation appear likely to be significant mainly on a regional scale. The effects are difficult to quantify, but there are likely to be secondary effects of considerable local significance in terms of water resources, runoff, and erosion. General circulation models offer a way of objectively quantifying these global and regional climatic responses, so that investment in improving their accuracy and reliability is essential.

Current models are able to predict the weather for short periods (about ten days at most). For longer periods the more realistic models generate a simulated climate with similar properties to the real atmosphere. The most realistic models should run indefinitely without producing infeasible conditions, but a great deal more development is needed before they will provide useful predictions over long periods. Besides those parameters actually being investigated, it is instructive to check values of other parameters generated over time by the model for comparisons with real values.

In general circulation models feedback loops are very important. This is the way in which variation of a parameter, due to, say, a change in vegetation, can affect the behaviour of the atmospheric model. The two types of feedback loop, positive and negative, are important since they affect the stability of the system. Positive feedback tends to reinforce the initial process. This leads to even greater effects and so destabilizes or even destroys the system. Negative feedback opposes the initial process, tending to damp down its effects and so stabilizes the system. With so complex a system as the global circulation of the atmosphere the numerous positive and negative feedback loops may be expected generally to balance each other. Changes in surface vegetation alter several surface parameters and so affect feedback loops. Important examples of land surface parameters are the surface albedo or reflectivity, which determines the fraction of solar radiation absorbed, and the surface moisture availability, which affects the partitioning of energy between thermal and latent heat.

There are very real dangers in modelling the atmospheric processes that the simplifications used may emphasize one type of feedback more than another. The query was raised that many of the effects shown are positive feedbacks, not negative, yet the atmosphere appears to be a very stable system. It was agreed that there are many negative feedback systems leading to stability but that, for instance, over the last 20 years anomalies of rainfall have been occurring over the Sahelian regions.

Some model simulations have demonstrated that a relatively small change in evaporation due to vegetation change has resulted in a knock-on effect causing a relatively large change in rainfall. Equilibrium at these new levels was apparently established within ten years (Henderson-Sellers and Gornitz 1984). However, negative feedbacks were undoubtedly omitted, such as small but significant changes in sea temperatures, a slight change in the Walker circulation, a decrease in cloudiness, feedback from surrounding terrestrial areas, and biosphere responses resulting from the inevitable change in vegetation. Similarly, some positive feedbacks were omitted, such as that caused by changes in runoff. This example demonstrates the importance of improving GCMs until they include and mimic all the features that significantly affect the output.

The grid scale used for GCMs precludes the incorporation of the fine scale pattern of the land surface, although this may have a significant effect upon the simulation. This deficiency is most pronounced in the treatment of the runoff process where such important parameters as slope, aspect, elevation, vegetation, soil type, canalization, as well as rainfall inhomogeneity, are omitted. A reduction of grid size might lead to a marginal improvement; similarly an increase in the number of vertical layers might improve the incorporation of inversions in the lower atmosphere. However, resolution is intimately linked with the time step of the model and in the final analysis to the capacity of the computer.

At the moment GCMs ought not to be regarded as predictive, since further refinement is needed before reliable predictive outputs are obtained. However, they already enable us to rank the parameters used in the models. To a degree, the recommendations for further investigations of parameters specified by Dr. Rowntree are based on such tests of sensitivity. The models not only indicate the level of accuracy needed for the parameters but also distinguish the precision required for different regions. They also lend support to the International Satellite Land Surface Climatology Project (ISLSCP).

Perhaps the most suspect data is that on grid square vegetation type. Even within one international agency or between national data sets there are serious disagreements often due to uncorrelated (independent) definitions. Thus although published atlases of vegetation and soil may be used as data files, the information available is often contradictory. In fact, it would be valuable to see how sensitive GCMs are to these differences. Since the surveys and classifications used are for other purposes, it may be necessary to collect GCM orientated data sets on vegetation types (correlated with albedo and aerodynamic roughness, perhaps) using satellite remote sensing.

Land surface topography that can generate gravity waves in the atmosphere is grossly simplified in most models, resulting in a generally "smooth" grid scale topography. Similarly, with the size of grid scales currently in use, the effect on momentum transfer of the aerodynamic roughness of different vegetation types is at sub grid scale. The problem also arises of averaging the values for a mosaic of vegetation types over the large grid areas of, say, 2° by 2°. When it becomes necessary to incorporate the aerodynamic roughness of vegetation in GCMs, correlation with canopy height and vegetation type will probably be sufficient.

Actual measured values are probably more vital for the albedo of land surfaces, as these may be poorly correlated with vegetation type due to the effects of soil moisture and foliage moisture status. An accuracy of between 1% and 5% is required by GCMs. This suggests that data are needed as a function of time of day, wavelength, cloud cover, and season. One study showing that a tall yellow-brown grass cover had half the albedo of short green grass illustrates the significance of the density and depth of foliage in trapping radiation rather than merely its colour. Measurements need to be taken from large homogeneous areas while "grid square vegetation" needs to be given a characteristic rather than average albedo. Long wave (>0.7 m) should be separated from short wave radiation, but at this stage further refinement is unnecessary.

It is evident that for the successful development of GCM techniques for better simulation of the effects of the manipulation of the land surface by man, more infor mation on surface parameters is needed. However, unless they are by-products of other hydrological or meterological projects conducted in their own right, it is difficult to define possible sponsors or, indeed, even find scientists interested in simply collecting the data for the modellers. A solution may be found through closer integration of the mesoscale experiment (mentioned in chap. 8) with satellite remote sensing. The former would provide the ground truth for the latter, while the satellite data might allow the micrometeorological and hydrological observations to be extended beyond the 50 km x 50 km experimental area. This approach would bring the mesoscale experiments significantly closer to the objectiveness of the ISLSCP programme. Such an experiment is intimately bound to the fortunes of GCMs, since these models are of increasing interest to the climatological and meteorological community. Previously the emphasis has been on two- to five-day forecasts and there has been little incentive to improve GCMs for this purpose. However, it is now evident that the response time in GCM simulations of land surface changes can be much faster than was suspected. Improving GCMs and obtaining more realistic values of parameters take time and must be justified by improved performance.

Several reasons can be given for the need for faster computers to allow the use of finer grids and better vertical resolution of the atmosphere. For instance, convectional atmospheric events are difficult to model despite their significant interaction with land surface processes, partly because their dimensions are much smaller than the grid sizes currently in use. However, a finer grid means that the length of time that is simulated is shorter, unless a faster computer is used. While for many transfer processes a finer grid scale also needs a finer time scale, the current trend is towards the use of coarser grid scales with a simulated time of 1,000 days or more rather than 30 to 60 days. A realistic aim would be for a 1° x 1° grid with a simulation for up to 1,000 or even 10,000 days.