×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Mapping without the sun
Author
Zhang, Jixian

112
SOIL MOISTURE RETRIEVAL COMBINING OPTICAL AND RADAR DATA
DURING SMEX02
Chen Quan a-b ' *, Li Zhen a ,Tian Bangsen d,b
moistui
canopy
of rada
no em]
vegetat
1.3 SB
' State Key Laboratory of Remote Sensing Science, Jointly Sponsored by the Institute of Remote
Sensing Applications of Chinese Academy of Sciences and Beijing Normal University,
b Graduate School of Chinese Academy of Sciences
Commission VII, WG VII/6 & VII/7 & II/l
KEY WORDS: Soil Moisture, Optical. Radar, PALS, SMEX02, NDVI, NDWI
ABSTRACT:
Soil moisture is a key parameter in numerous environmental studies, including hydrology, meteorology, and agriculture. Microwave
remote sensing is a promising way for the estimation of surface soil moisture, which has been known by scientists for decades of
years. Unfortunately, soil moisture retrieval from dense vegetation area is still a troublesome nowadays. In this paper, the PALS
(Passive and Active L and S band System) radar data acquired during SMEX02 (The Soil Moisture Experiments in 2002) are used
for the estimation of soil moisture. Firstly, VWC (Vegetation Water Content, kg/m 2 ) estimation is compared by using NDVI (The
Normalized Difference Vegetation Index) and NDWI (The Normalised Difference Water Index), the result showed that VWC from
NDWI has high accuracy than from NDVI. Then, canopy transmittivity was retrieved using optical thickness, which is obtained by
VWC and a constant factor ‘b’. Then, under some assumptions, backscatter from underlayer soil is obtained after volume and
interaction scatter is removed from totally backscatter. Finally, soil moisture change between two acquisitions is retrieved using a
simple bare-soil scatter model, which is simplified from AIEM (Advanced Integral Equation Model) model. The retrievals are
compared with in-situ GSM (Gravimetric Soil Moisture) in SMEX02, the results showed that the algorithm can be used to estimate
soil moisture at vegetation area, but the retrieval is relatively poor when the vegetation is dense, such as in com field. Therefore,
some improvements need to be done in the future at high vegetation area.
1. INTRODUCTION
1.1 Importance of Soil Moisture
Soil moisture is an important parameter in atmospheric,
hydrological, and dynamic global vegetation and carbon
dynamics models. It plays an important role in the interactions
between the land surface and the atmosphere, as well as the
partitioning of precipitation into runoff and ground water
storage 11 1 Although objectives are different for each class of
model, soil moisture plays an important role in each discipline 121 .
In atmospheric models, the principle interest in soil water is its
impact on evaporation and sensible/latent heat partitioning at
scales that are appropriate for simulating the forcing of
atmospheric processes. In hydrological applications the focus is
generally on water balance components—infiltration, surface
runoff, evaporation, deep percolation, and changes in water
content of the vadose zone. In dynamic global vegetation
models the principal interest is in quantifying the distribution of
vegetation with implications for the terrestrial component of the
current global C0 2 budget. An analysis of end-user
requirements found that a volumetric water content accuracy of
0.04 m 3 m' 3 or 4% in absolute terms was realistic for
atmospheric modeling applications, though this may prove
inadequate for modeling soil profiles down to the root zone 121 .
Therefore, soil moisture provides a link between the earth’s
surface and atmosphere through its effect on surface energy and
moisture fluxes. The spatial and temporal dynamics of soil
* Corresponding author: cq801915@hotmail.com
moisture are important parameters for various processes in the
soil-vegetation-atmosphere-interface. 1 ' 1
1.2 Radar Soil Moisture Estimation
Radar is one of the newer instruments for studying Earth’s
environment. Although radar imaging was developed initially
for military applications, at present it has become a very
important tool in remote sensing. The reasons for this have been
their all-weather ability and their dependence on the
information of specific physical properties of the target.
The measurement of soil moisture using microwave sensor
(Radar and radiometer) is based on the large contrast between
the dielectric properties of dry soil («2.5) and water («80).
As the soil is moistened, its dielectric constant varies from
approximately 2.5 when dry to about 25-30 under saturated
conditions. This translates to an increase in power reflected by
almost 8dB for wet soil compared to dry soil. The range of
wavelength of the microwave bands is from 1mm (short) to lm
(long). Long wavelength microwave bands offer the greatest
potential for remote sensing of moisture due to their capability
to penetrate vegetation cover.
Currently, several theoretical and empirical models have been
set up to relate soil moisture to radar backscatter of bare soil.
Theoretical models are usually quite complex and difficult to be
used in inversion process. Empirical models are usually
statistical relationships between in-situ samples and truck-borne
scatterometer observation, and can be used in estimating soil
SMEXI
samplir
develop
from s<
vegetat:
spacebc
measun
these o
levels a
The stu
Iowa.
USD A
hydrolo
cover t
with co
were cc
which 1
vegetati
grew fr<
1.4 Th
The pc
nonscar
operatir
GHz (i
operates
VV, HF
The PA
June 25
(corresp
187, 18!
but scat
4th, 5th.
be obs
simultai
polarize
backsca
nadir-
tempera
(velocit;
angle c
configui
was 331
footprin
navigati
1.5 La:
As part
scenes i
Mapper
primary
resolutic
images 1
An exti