457
IMPACT OF MICROWAVE DERIVED SOIL MOISTURE ON HYDROLOGIC
SIMULATIONS USING A SPATIALLY DISTRIBUTED WATER BALANCE MODEL
D.-S. LIN, E. F. WOOD, J. S. FAMIGLIETTI
Department of Civil Engineering and Operations Research
Princeton University, Princeton, NJ 08544, U.S.A.
M. MANCINI
DILAR, Politecnico di Milano, Milano, Italy
ABSTRACT:
Spatial distributions of soil moisture over an agricultural watershed with a drainage area of 60 ha
were derived from two NASA microwave remote sensors, and then used as a feedback to determine
the initial condition for a distributed water balance model. Simulated hydrologic fluxes over a period
of 12 days were compared with field observations and with model predictions based on a streamflowderived
streamflowderived initial condition. The results indicated that even the low-resolution remotely sensed data
can improve the hydrologic model’s performance in simulating the dynamics of unsaturated zone soil
moisture. For the particular watershed under study, the simulated water budget was not sensitive to
the resolutions of the microwave sensors.
KEY WORDS: Remote sensing, soil moisture, distributed water balance model.
1 INTRODUCTION
The spatial distribution of soil moisture is a critical hydrologic variable which exerts a major
control on the land surface water and energy balance. Despite its importance, direct usage of this
information in climatic, hydrologic, or agricultural simulations has not gained widespread applica
tions. This is in part due to the fact that most models do not consider soil moisture a measurable
state variable. Also, it is difficult to acquire this information on a consistent and spatially compre
hensive basis using the conventional techniques. Recent advances in remote sensing technology have
shown that surface soil moisture distribution can be measured at various resolutions by microwave
sensors (see Engman, 1990). Since microwave remote sensing techniques can be implemented on high
altitude platforms, it is likely that such measurements could be obtained routinely over large regions
in the near future. However, how to incorporate the remotely sensed soil moisture data into the
modeling framework remains to be a big challenge. On what level of temporal and spatial resolutions
that will be required by different applications also needs to be addressed. This paper presents a
spatially distributed water balance model that is capable of using the microwave derived soil mois
tures to facilitate simulations. Using the model as a tool and data collected during a multi-sensor
aircraft campaign over an agricultural catchment, we examine the impact of remotely sensed data
on hydrologic simulations, and the sensitivity of simulated water budget to the resolutions at which
microwave remote sensing methods can currently provide spatial patterns of soil moisture.
2 MODEL FORMULATION
The building block of the distributed hydrologic model is a process-based one-dimensional wa
ter balance model that accounts for the soil moisture movements in the unsaturated zone of a soil
column. This one-dimensional model is based on the formulation described in Famiglietti (1992),
but improves upon the treatment of storage by using a steady-state approximation to represent the
moisture profile. The lower boundary of this model is at the water table where the steady percolation
or exfiltration is added to, or extracted from, the regional groundwater system. Lateral subsurface
