Chapter 15
Remote Sensing Parameterization of
Meso-scale Land Surface Evaporation
W.G.M. Bastiaanssen, M. Menenti, A.J. Dolman, R.A. Feddes and H. Pelgrum
DLO Winand Staring Centre for Integrated Land, Soil and Water Research
P.O. Box 25, NL-6700 AC Wageningen
The Netherlands
15.1 Introduction
Land surface fluxes vary spatially as a result of the spatial heterogeneity of soil physical properties, fractional soil cover, land use, rainfall and hydrological processes. The scale at which this
variability of surface fluxes occurs may be as low as a fraction of a meter, and time variations
over a period of hours may be significant. The spatial average land surface flux at regional
scale is because of length scale and sampling size difficult to quantify from field measurements.
Although considerable progress has been made in this area over the last decade (e.g. Michand
and Shuttleworth, 1996), there is no general consensus at this moment how to extrapolate
surface fluxes from patch scale to meso-scale.
Atmospheric Models are the only available tools to forecast climate trends and to assess the
impact of reduced emissions strategies on global warming. The grid-lengths of these models
at which the land surface fluxes need to be described are of the order of 100 km to 300 km;
at this scale the land surface processes exhibit a considerable spatial variability. Hence, a
serious problem exists in the application of large scale models: How do we test the prediction of
area-averaged land surface fluxes at meso-scale? The number of land surface parameterization
schemes which simulate meso- scale land surface fluxes by continuity of energy, momentum,
heat and water is still growing (e.g. Sellers et al., 1986; Noilhan and Planton, 1989; Dolman,
1993) while research on techniques to quantify model input parameters and verify model output
parameters at the scale of a grid-square is still lagging behind. Henderson-Sellers et al. (1995)
for the PILPS project intercompared the performance of 24 land surface parameterization
schemes and plans to use observational data of relative homogeneous grasslands to validate the
fluxes. Tower-based flux measurements apply at scales smaller than the correlation length of
the fluxes, and a direct link between model output at a horizontal scale manyfold larger than
the correlation length of field measurements is not straightforward. In this respect, remote
sensing measurements may be useful, since the land surface characteristics can be measured by
satellite platforms with different spatial (as low as meters) and temporal resolutions (as low as
half hours) over large areas.
NATO ASI Series, Vol. I 45
Radiation and Water in the Climate System:
Remote Measurements
Edited by Ehrhard Raschke
Ii:> Springer.Verlag Berlin Heidelberg 1996
Remote Sensing Parameterization of
Meso-scale Land Surface Evaporation
W.G.M. Bastiaanssen, M. Menenti, A.J. Dolman, R.A. Feddes and H. Pelgrum
DLO Winand Staring Centre for Integrated Land, Soil and Water Research
P.O. Box 25, NL-6700 AC Wageningen
The Netherlands
15.1 Introduction
Land surface fluxes vary spatially as a result of the spatial heterogeneity of soil physical properties, fractional soil cover, land use, rainfall and hydrological processes. The scale at which this
variability of surface fluxes occurs may be as low as a fraction of a meter, and time variations
over a period of hours may be significant. The spatial average land surface flux at regional
scale is because of length scale and sampling size difficult to quantify from field measurements.
Although considerable progress has been made in this area over the last decade (e.g. Michand
and Shuttleworth, 1996), there is no general consensus at this moment how to extrapolate
surface fluxes from patch scale to meso-scale.
Atmospheric Models are the only available tools to forecast climate trends and to assess the
impact of reduced emissions strategies on global warming. The grid-lengths of these models
at which the land surface fluxes need to be described are of the order of 100 km to 300 km;
at this scale the land surface processes exhibit a considerable spatial variability. Hence, a
serious problem exists in the application of large scale models: How do we test the prediction of
area-averaged land surface fluxes at meso-scale? The number of land surface parameterization
schemes which simulate meso- scale land surface fluxes by continuity of energy, momentum,
heat and water is still growing (e.g. Sellers et al., 1986; Noilhan and Planton, 1989; Dolman,
1993) while research on techniques to quantify model input parameters and verify model output
parameters at the scale of a grid-square is still lagging behind. Henderson-Sellers et al. (1995)
for the PILPS project intercompared the performance of 24 land surface parameterization
schemes and plans to use observational data of relative homogeneous grasslands to validate the
fluxes. Tower-based flux measurements apply at scales smaller than the correlation length of
the fluxes, and a direct link between model output at a horizontal scale manyfold larger than
the correlation length of field measurements is not straightforward. In this respect, remote
sensing measurements may be useful, since the land surface characteristics can be measured by
satellite platforms with different spatial (as low as meters) and temporal resolutions (as low as
half hours) over large areas.
NATO ASI Series, Vol. I 45
Radiation and Water in the Climate System:
Remote Measurements
Edited by Ehrhard Raschke
Ii:> Springer.Verlag Berlin Heidelberg 1996