1. introduction - wit press€¦ · as soil moisture evolution involves interactions between the...

A sustainable land use information system

R.K. Munro\ W.F. Lyons', M.S. Wood*, Y. Shao\ L.M. Leslie^

Bureau of Resource Sciences, PO Box Ell, Kingston, ACT 2604,

Australia

Centre for Advanced Numerical Computation in Engineering and

Science, University of New South Wales, Kensington, NSW 2052,

Australia

School of Mathematics, University of New South Wales,

Kensington, NSW2052, AustraliaCorresponding author e-mail: [email protected]

Abstract

Soil moisture plays a major role in atmospheric and hydrological processes, since itinfluences the partitioning of both surface available energy into sensible and latent heat fluxes,and of precipitation into evapotranspiration and runoff. Soil moisture is also an importantparameter for ecological processes, for instance, the availability of soil water directly affectsplant growth. For studies of plant growth and land use sustainability, there is an increasingdemand for detailed predictions of soil moisture over large areas.

The Sustainable Land Use Information System (SLUIS) is an application of coupledhigh resolution atmospheric and land surface models to support ecologically sustainable landmanagement. It draws environmental parameters from Geographic Information Systemdatabases developed by BRS and other institutions. A major output of SLUIS is an estimate ofAustralian soil moisture with regional or continental coverage and horizontal resolution downto 1 kilometre. The soil moisture model solves the Richards' equation and the temperaturediffusion equations for multiple soil layers. It includes parameterisations of energy and massexchanges between the atmosphere, soil and vegetation and parameterisations for soil waterrunoff and drainage. The soil moisture model has the flexibility of using either Broadbridge andWhite [1], Clapp and Hornberger [2] or Von Genuchten [3] soil water retention curves, as wellas the flexibility of using any number of soil horizons.

Work to date has shown that predicted soil moisture patterns over the Australiancontinent in summer are closely related to the distribution of soil types. Water stress iswidespread in Australia in summer and the resultant soil moisture patterns are dominated bythe ability of the soil type to hold moisture in the drying conditions.

Transactions on Ecology and the Environment vol 16, © 1997 WIT Press, www.witpress.com, ISSN 1743-3541

636 Ecosystems and Sustainable Development

1. Introduction

Soil moisture is a major resistor variable in the global energy cyclesince it influences the partitioning of both surface available energy intosensible and latent heat fluxes, and of precipitation into evapotranspiration andrunoff. Consequently, all of the major physically based models of the biosphereattempt to simulate land surface conditions by including parameterisations forsoil moisture. From biophysical and agricultural perspectives, the measurementof soil moisture and estimation of soil water content (depth-integrated soilmoisture) are important activities for determining the status of agriculturalproduction since water in soil represents that component of the hydrologiccycle that is available to plants. Soil moisture is therefore also important inecological processes and most biomass production models will includeestimates of soil water availability.

Soil moisture simulation for a continental coverage poses three majorchallenges. Firstly, soil moisture predictions with land surface schemes will belimited by the empirical or semi-empirical nature of the parameterisations.Opinions differ on how reliable soil moisture predictions with land surfaceschemes are. An assessment of various schemes for soil moisture simulationwith prescribed atmospheric forcing data and prescribed land surfaceparameters for soil hydraulic properties, aerodynamic properties and vegetationcharacteristics for a single point has been examined in Shao et al [4] andrelated studies (Shao and Henderson-Sellers [5]). The differences for each ofthe schemes that was demonstrated in these studies can mainly be attributed tothe different treatment of soil hydrological processes in the schemes. Secondly,as soil moisture evolution involves interactions between the atmosphere, soil,and vegetation, land surface schemes are usually complex. The prediction ofsoil moisture depends critically on the input parameters that describe soilhydrological properties, surface aerodynamic properties and vegetation features(e.g. leaf area index). Finally, the interactions between the land surface and theatmosphere involve complex feedback processes which are not yet wellunderstood, but are known to have a significant impact on the climatevariability.

In the case of soil moisture simulation, it appears that the uncertaintiesin the choice of land surface parameters and in the lower boundary condition ofthe soil layer exceeds those arising from the atmospheric data. In currentgeneral circulation models, the land surface parameters contain significantuncertainties and the lower boundary is crudely treated. Therefore, it is likelythat soil moisture predictions from current general circulation models are notsufficiently accurate, to facilitate meaningful analysis of land surface processes.

Our intention in this study is to provide a simulation of soil moisture forthe Australian continent. To this end there are three major tasks; the first onebeing the development of a new land surface scheme with an improvedtreatment of surface soil hydrology. The second task is to establish a set of up-


Ecosystems and Sustainable Development 637

to-date parameters for the land surface, including soil and vegetation over theAustralian continent on the basis of GIS data, and the third task is to couple theland surface scheme with an atmospheric model for the four-dimensionalassimilation of soil moisture.

The Sustainable Land Use Information System (SLUIS) is anapplication of coupled high resolution atmospheric and land surface models tosupport ecologically sustainable land management. It draws environmentalparameters from Geographic Information System databases developed by BRSand other institutions. The soil moisture is calculated with the AtmosphericLand Surface Interaction Scheme (ALSIS) driven by the output of the newUniversity of New South Wales (UNSW) High-resolution Limited-areaAtmospheric Model (HLAM). This modelling approach and results fromSLUIS are described in detail in the following sections of this paper. The aimof this paper is to describe preliminary results from an evolving system whichis used to assist in the development of Australian Government policiesconcerning sustainable land-use management.

2. Simulation of soil moisture

Within the context of the foregoing issues, the last decade has seen arapid development of sophisticated land surface schemes for atmospheric,hydrologic, and ecologic modelling (e.g., Dickinson et al. [6], Sellers et al. [7],Noilhan and Planton [8], Liang et al [9], Wetzel and Boone [10]). A landsurface scheme is composed of three major components: bare soil transferprocesses, vegetation canopy transfer processes, and soil thermal andhydrologic processes. Almost all land surface schemes are based on the one-dimensional conservation equations for temperature and soil moisture

ar __j_aG

6f C9z (1)

dt dz (2)

where T is soil temperature, C is volumetric soil heat capacity, 0 is volumetricsoil water content, S is a sink term which includes runoff and transpiration (ithas been assumed that the temperature sink term is zero), G is soil heat flux,and q is soil water flux. A land surface scheme is the algorithm required tosolve this system for a particular soil layer configuration. Theparameterisations occur in the boundary conditions, in soil hydraulic andthermal properties, and in the treatment of the sink terms. The upper boundaryconditions at the atmosphere and land surface interfaces include sensible andlatent heat fluxes.



In most land surface schemes, there is little conceptual difference in theformulation for atmospheric transfer, such as the calculation of sensible andlatent heat fluxes. There is also little difference in the treatment of canopy (the'big leaf assumption), with a possible exception of the Simple BiosphereModel of Sellers et al [7]. However, the treatment of soil hydrologic processes,which can reflected in the number of soil layers, may be significantly different.Depending on the number of computational soil layers, the schemes can begrouped into bucket-type single-layer schemes (Manabe [11]), force-restoretwo-layer schemes (Noilhan and Planton [8]) and diffusion-type multi-layerschemes (Wetzel and Boone [10]). Most schemes have less than three soillayers, as they are designed mainly for use in general circulation models wherethe demand on computational efficiency is important. Clearly, land surfaceschemes with a small number of soil layers may represent soil moisturedistribution poorly, and have shortcomings in the treatment of the functioningof plant roots and evaporation from bare surfaces. Although these schemesmight be adequate for global climate models, for ecologic modelling more soillayers are required.

3. The Sustainable Landuse Information System

One of the major new ingredients of SLUIS is the improved treatmentof soil surface hydrology through the use of the ALSIS landsurface schema. InALSIS, the one-dimensional (vertical) Richards' equation is used directly todescribe the evolution of soil moisture. Soil temperature and soil moisture aresimulated with finite difference solutions of equations (1) and (2). Theparameters suggested by Broadbridge and White [1] are used to characterise thehydraulic properties of various soils. The forms of their hydraulic functionshave the unique property of making solutions of both differential and finitedifference from Equation (2) determinate under all conditions, includingsaturated soil and completely dry soil. This eliminates the numerical failuresthat have previously made routine numerical solution impracticable. Numericalspeed is greater than that of a generalisation of the Green and Ampt [12] modelwith infinitely sharp wetting fronts. This is due to a combination ofdeterminacy and numerical strategy, whereby mixed dependent variables areused directly in a Newton-Raphson solution scheme. This feature allowsALSIS to incorporate as many soil layers as required to provide a bettervertical resolution of soil moisture and better treatment of heterogeneity (in thevertical) of soil hydraulic properties. This flexibility in choosing the number ofsoil layers also facilitates a more effective treatment of root activities.

ALSIS is essentially a surface hydrologic model for prediction ofevapotranspiration, surface and subsurface runoff and deep soil drainage, byparameterisation and solving the Richards' equation and the temperaturediffusion equation for multi-soil layers. The Richards' equation requiresclosure relationships between hydraulic conductivity, soil water diffusivity,



matric potential and soil water content. These relationships depend on themorphology of soil pores, with the average pore size being an importantindicator for soil types. The closure relationships, as solved in ALSIS, arebased on the Broadbridge-White soil model, although more recent versions ofALSIS allow for hydraulic processes to be described using the water retentionfunctions of Clapp and Hornberger [2] and other workers (e.g., Von Genuchten

[3]).In ALSIS, the land surface is divided into areas of bare soil and areas

covered by different types of vegetation. The energy transfer processes overbare soil surfaces are described using aerodynamic resistance laws, while thedescription of the canopy transfer processes is based on studies as summarisedin Raupach [13].

A more detailed description of ALSIS can be found in Irannejad andShao [14]. Shao et al. [15] have also described the treatment of the surfacehydrology and the functioning of vegetation roots by ALSIS and the reader isreferred to those papers for a further discussion of ALSIS.

4. Soil moisture simulations over the Australian continent

Shao et al [15] have presented verification studies of ALSIS using theHAPEX-MOBILHY dataset as atmospheric and biophysical forcing data.HAPEX was conducted in 1986 in southern France and has been welldocumented by a series of papers (e.g., Goutrobe et al. [16]). The ALSIS modelwas demonstrated to have excellent agreement with the observations obtainedfrom the intensive field experimentation phase of HAPEX. Further validationfor other locations has also been undertaken by Lyons et al. [17]. Preliminaryvalidation results for western NSW are also presented in a subsequent sectionof this paper. Sensitivity analyses were also undertaken by Shao et al. [15] anddemonstrated the importance of selecting appropriate land surface parameters,particularly the parameterisation of soil hydraulic properties for different soiltypes. This paper makes use of these soil hydrologic parameterisations of Shaoetal. [15].

4.1. Soil moisture distributions over the Australian continent

A continuous simulation of soil moisture over the Australian continentover a two month period from 1 January to 1 March 1996 has been performed.The depth of the soil layer is 2m and is assumed to be vertically homogeneous(this will be modified in a future study). The 2m soil layer is divided into fivelayers with depth of 0.05, 0.15, 0.3, 0.5 and 1.0m, respectively. Atmosphericforcing data was supplied using the HIRES model of Leslie and Purser [18].This model was run over the Australian continent with a 20 x 20 km horizontalresolution and 30 vertical layers. The atmospheric data are stored every 30minutes and averaged to a 50x50 km grid for soil moisture simulation. The



model has been tested extensively in both research and operational modes(Leslie and Purser [18]). Standard statistical evaluation, averaged over 30stations in the Murray-Darling Basin, has shown that the model performance isvery good. For near-surface air temperature predictions, the RMS error is 2.1 Kwith a mean absolute error of 1.7 K; for near-surface wind speed the RMS erroris approximately 3ms".

An example of the predicted soil moisture pattern is shown in Figure 1,where soil moisture content of layer 1 (0-0.05m), layer 2 (0.05-0.2m) and layer4 (0.5-1m) are illustrated together with the total soil water in the top 1m, for10:00 UTC 15 Feb 1996 (day 46 of the simulation). The basic soil moisturepattern is typical for the Australian continent in summer. As expected, in largeareas in the north-western part of Australia, including the Great Sandy Desert,Gibson Desert, Great Victoria Desert and Nullarbor Plain, the soil moisturecontent is very low for the time of the year (late summer in the SouthernHemisphere). Although, there is a slight increase in soil moisture towarddeeper layers, for a considerable soil depth, the soil moisture content falls inthe range between 0.05 to 0.15 m"* rrf\ with the total soil water in the top 1m ofsoil being around 100 mm. Away from the desert areas, there is a gradualincrease in soil moisture both toward the east and west coasts. In large areas ofthe 'Channel Country' of western Queensland and the Murray-Darling Basin,typical values of soil moisture are around 0.25 m^ m"^ in layer 1 and 2, and 0.3m m~^ in layer 4. Total soil water in the top 1m of soil is around 250 mm.

Further towards the east coast, soil moisture is over 0.3 m m" , underthe influence of rainfall that occurred during the simulation period. Oneobvious feature of the spatial distribution of soil moisture is that the spatialpatterns are closely related to soil hydraulic properties. For instance, the lowsoil moisture content in the desert areas is characteristic of the predominantsandy soils in the region. It is well-recognised that the region has littleprecipitation, and as the sandy soils have a high (saturated) hydraulicconductivity and low air dry soil moisture content, soil moisture is low formost of the time. The exception is immediately after rainfall. In the desertareas, soil moisture is rapidly lost through drainage or evapotranspiration. Soilmoisture content is significantly higher toward the east coast, apart frompatches of very dry areas in Queensland. For a very large area in the easternparts of Australia, the volumetric soil moisture content is around 0.3 m m" , asthe soils in this region are predominantly sandy clay or silty clay with highvalues of 9. For some of these areas, however, the absolute soil moisturecontent is quite high when compared with that of sandy soils. The strongsimilarity between soil moisture patterns and soil type patterns supports thenotion that the Australian continent in summer is predominantly under waterstress.

The simulation also provides detailed information in space and time ofsoil moisture distribution. As shown in Figure 2, for example, across largeareas of North Queensland, available soil moisture is very high for 17 Feb



Soil Moisture Layer 4

Soil Moisture Layer 1 (m3/m3)

0.0150.03 0.06 0.09 0.1B 0.18 0.18 O.E1 0.24 0.27 0.3

Figure 1. Soil moisture distribution over the Australian continent for 15 Feb1996. (a) Predicted soil moisture in m m" for layer 0-0.05m, (b) for layer0.05-0.20m, (c) for layer 0.5-1.Om and (d) total soil water in mm for the top1m of soil.



Total Soil Water in Top 1m (mm)

40 80 120 160 200 240 280 320 380 400 440

Soil Moisture Layer 2



1996, due to the rainfall influence of a tropical rotational system. On 18 Feb1996, rainfall in Western Australia also resulted in extensive areas of high soilmoisture regions. These changes in soil moisture are also found to be consistentwith the atmospheric quantities of the forcing data.

4.2. Point based simulations using SLUIS

It is difficult to produce spatially consistent and accurate observationsof soil moisture in order to verify the model outputs described in the precedingsection. Scaling issues and the question of the feasibility of translating frompoint locations to grided estimates will tend to dominate discussions of thevalidation. Instead, offline simulations of soil moisture from 1991 to 1995 forsites located at White Cliffs, NSW (30.85°S, 143.08%) have been undertaken.Other validation work using ALSIS is presently underway for additionallocations in Australia and overseas.

Australian Bureau of Meteorology meteorological observations wereused as atmospheric forcing data for the ALSIS model. The data was initiallyobtained as 6 hourly observations of rainfall, temperature, dew pointtemperature, atmospheric pressure, cloud amount and type at three levels, andsurface wind speed and direction. Daily sunshine hours were also obtained forthe two sites. Using a simple spline function, these data were increased infrequency to a three hour interval. Table 1 outlines the meteorologicalparameters required by ALSIS. It was therefore necessary to convert some ofobservations to more meaningful values for assimilation by ALSIS. That is,wind speed and direction was converted to u and v components of wind, anddew point and pressure were used to calculate specific humidity. Radiationparameters were somewhat more problematical. For the two locations a spectralradiation model, and empirical corrections (based on cloud type and amount) asreported in Oke [19] were used to calculate the total incoming solar radiationand downward longwave emission. These were calculated as instantaneousmeasurements and integrated over a three hour time interval.

As ALSIS is a land surface schema, specific vegetation parameters arerequired in order to accurately describe land-atmosphere energy fluxes. Sinceboth sites are dominated by perennial grasses the principal vegetationcharacteristics required for the simulations were Leaf Area Index (LAI),fraction of vegetation cover (/) and vegetation height (/zj. Other landsurfaceand vegetative cover parameters, such as surface albedo, plant root distributionand stomatal resistance are also required by ALSIS and these are discussed inIrannejad and Shao [22].

Using the input parameters outlined above, and the soil hydraulicparameterisations of Shao et al. [15], it was possible to construct a full set ofinput parameters for the ALSIS model to enable comparisons with soilmoisture observations for the White Cliffs site. These are presented as timeseries of simulated and observed soil moisture values in Figure 3. The



Day 46

AvSM: 0-5cm

AvSM: Layer 3

AvSM: Depth Ave

Figure 2. Predicted available soil moisture for 15 Feb 1996 (left column); 17Feb 1996 (middle column) and 18 Feb 1996 (right column). In each column,available soil moisture for soil layer 0-0.05m, layer 0.20-0.50 and the depthaverage over the top 1m soil are shown.



Day 48 Day 49



observational data comprises once daily estimations of soil water content attwo depths (0-0.1m and 0.3-0.5m). Soil water content was determined byconverting soil water potential measurements to soil water volume using theBroadbridge and White [1] water retention function.

Parameter Dimension2. -\

Saturated hydraulic conductivity m sSaturated volumetric water content m m"Air dry volumetric water content m m"'*Macroscopic capillary length scale mSoil hydraulic characteristic parameterVolumetric water content at field capacity m m"Volumetric water content at wilting point m^m"Heat diffusivity for dry soil m s"Soil surface albedoFraction of vegetation coverHeight of vegetation mLeaf area indexMinimum vegetation stomatal resistance mRoot fraction in different soilsVegetation albedo -

Table 1. List of parameters required by the ALSIS land surface schema.

This figure indicates excellent agreement with observed fluctuations insoil moisture. Where there are departures from the observational data, they arechiefly in terms of the magnitude of the simulated values. Essentially, thesimulations suggest that ALSIS successfully translates the precipitation signalinto soil moisture fluctuations and can propagate those fluctuations down thesoil profile. This is particularly the case with near surface soil moisture (0-O.lm). There does, however, appear to be a short-period lag in the timing ofsimulated events and this is probably an artefact of the different temporalresolutions of the observations (24 hours) and the simulations (3 hours).



Soil Moisture for White Cliffs, 1991 to 1995Observed and Predicted

45Predicted - Layer 1 (8-.In)Observed Layer 1 (0.1*)Predicted- Layer 4 (.3-.5m)Observed - Layer 4 (.3-.Sm)

8 288.8 488.8 688.8 888.8 1888.8 1288.8 1488.8 1608.8 1888.8Days since 1 January, 1991

Figure 3. Time series of predicted and available soil moisture for WhiteCliffs, NSW for Jan. 1, 1991 to Dec. 31, 1995

In the case of soil moisture at greater depth (0.3-0.5m), thecorrespondence between observed and simulated soil moisture occasionallybreaks down, largely in terms of the magnitude of soil water contentfluctuations. Nevertheless, there is still good overall agreement between thecurves. It is suggested that the results of this simulation may have beeninfluence by the choice of soil hydraulic parameters. While a two horizonsimulation was undertaken, the depth to the B horizon was assumed to be 0.2m,which may have been unrealistic for this site. In addition, as soil moisture isstrongly influence by pedology and how these translate into hydraulicparameterisations becomes crucial for accurate simulation. Further work, usingother parameterisations is currently underway and will form the basis of asensitivity analysis.

In essence, this simulation has indicated good agreement with observeddata at point locations. They suggest some confidence for the results from thespatially based simulations of ALSIS.

5. Discussion and conclusions

This paper describes an system to model soil moisture patterns and theirevolution over the Australian continent. The land surface scheme, ALSIS,



differs from many other schemes in the treatment of surface hydrology andthe numerical formulation of the scheme. The non-linear relationships betweensoil hydraulic conductivity, matrix potential and soil moisture content are basedon the Broadbridge and White [1] soil model. The soil hydraulic parametersused to represent these relationships differ considerably from those of Clappand Hornberger [2], which are widely used in current land surface schemes.The scheme can accommodate as many soil layers as are required and thealgorithms used in the scheme are numerically efficient. Nevertheless, it cannotbe overstated that all land surface schemes are sensitive to the choice of soilhydrological parameters. Consequently, an additional emphasis of SLUIS is thedevelopment of a suitable set of land surface parameters based on the best GISdata currently available. In comparison to GCMs, the land surface informationused in this study is more detailed.

This study provided a prediction of soil moisture evolution anddistribution over the Australian continent in summer. Although there is not yeta comparison of the predictions through independent studies, apart from singlepoint evaluations with observational data such as, the results are in goodagreement with expectations. The simulations showed that over the Australiancontinent in summer, the soil moisture pattern is closely related to thedistribution of soil types. This implies that, apart from isolated areas and timesunder the influence of precipitation, the drying factors such asevapotranspiration and drainage are primarily responsible for soil moisturestatus.

High resolution atmospheric predictions and geographic data were usedin the simulation, and as a result a detailed spatial distribution and timeevolution of soil moisture was obtained. This level of detail is useful for manypractical purposes, such as plant growth modelling and soil erosion prediction.This paper represents the first attempt of an ongoing effort in 4-dimensionalsimulation of soil moisture. In addition to the feedback process between theatmosphere and the land surface, there is considerable scope for furtherimprovement of the database used in this study. Notably, the addition oftopography, soil depth, the lower boundary for soil moisture prediction and thetemporal changes in vegetation cover. These necessary geographic data arecurrently being collected and will lead to further improvement of soil moisturesimulation.

References

1. Broadbridge, P., White, I., Constant rate infiltration: a versatile nonlinearmodel: 1. Analytic solution, Water Res. Res., 1988, 24, 145-154.

2. Clapp, R.B., Hornberger, G.M., Empirical equations for some soilhydraulic properties, Water Re sour. Res., 14, 1978, 601-604.

3. Von Genuchten MT, A closed-form equation for predicting the hydraulicconductivity of unsaturated soils, Soil Sci. Soc Am. J., 1980, 44, 892-898.



4. Shao, Y., Anne, R.D., Henderson-Sellers, A., Irannejad, P., Thornton, P.,Liang, X., Chen, T.H., Ciret, C., Desborough, C., Balachova, O., Soilmoisture simulation, A report of the RICE and PILPS Workshop, 1994,IGPO Publication Series, No 14.

5. Shao, Y., Henderson-Sellers, A., 1996: Validation of soil moisturesimulation in landsurface parameterisation schemes with HAPEX data.Global and Planetary Change, 13, 11-46.

6. Dickinson, R.E., Henderson-Sellers, A., Kennedy, P.J., Giorgi, F.,Biosphere-Atmosphere Transfer Scheme (BATS) Version le as coupled tothe NCAR Community Climate Model, 1992, NCAR Tech. Note, Natl.Cent. Atmos. Res.

7. Sellers, P.J., Mintz, Y., Sud, Y.C., Dalcher, A., A simple biosphere model(SiB) for use within general circulation models, J. Atmos. Sci., 1986, 43,501-531.

8. Noilhan, J., Planton, S., A simple parameterisation of land surfaceprocesses for meteorological models, Mon. Wea. Rev., 1989, 117, 536-549.

9. Liang, X., Lettenmaier, D.P., Wood, E.F., Burgess, S.J., A simplehydrologically based model of land surface water and energy fluxes forgeneral circulation models, J. Geophys. Res., 1994, 99, 14415-14428.

10. Wetzel, P., Boone, A., 1995: A parameterization for land-atmosphere-cloudexchange: Testing a detailed process model of the partly cloudy boundarylayer over heterogeneous land. J. of Climate, 8, 1810-1837.

11. Manabe, S., Climate and Ocean circulation, 1, The atmospheric circulationand the hydrology of the earth's surface, Mon. Wea. Rev., 1969, 97, 739-805.

12. Green, W.H., Ampt, G.A., Studies in soil physics, I, The flow of air andwater through soils, J. Agric. Sci., 1911,4, 1-24.

13. Raupach, M.R., Canopy Transport Processes. In Flow and Transport in theNatural Environment: Advances and Applications, W.L. Steffen and O.T.Denmead (Eds.), Springer-Verlag, 1988, 95-127.

14. Irannejad, P., Shao, Y., The Atmosphere-Landsurface Interaction Scheme(ALSIS): Description and Validation. CANCES Technical Report, No. 2,1996, University of New South Wales.

15. Shao, Y., Leslie, L.M., Munro, R.K., Lyons, W.F., Irannejad, P., Morison,R., and Wood, M.S., Soil Moisture Prediction over the Australiancontinent, Meteorology and Atmospheric Physics, 1997. In Press.

16. Goutorbe, J.P., Noilhan, J., Cuenca, R., Valancogne, C., Soil moisturevariations during HAPEX-MOBILHY, Ann. Geophys., 1989, 7, 415-426

17. Lyons, W.F., Munro, R,K., Shao, Y., Leslie, L.M., Wood, M.S. and Hood,L.M., Soil moisture modelling and prediction over the Australian continentusing a land surface schema, Proceedings of Climate Variability forAgricultural and Resource Management, 6-8 May, 1997, AustralianAcademy of Science, In Press.



18. Leslie, L.M., Purser, R.J., Three-Dimensional Mass-Conserving Semi-Lagrangian Scheme Employing Forward Trajectories, Man. Wea. Rev.,1995, 123,2551-2566.

19. Oke, T.R., Boundary Layer Climates, 2nd Edition, 1992, Routledge, 435pp.


1. introduction - wit press€¦ · as soil moisture evolution involves interactions between the...

Documents