AND 
JATA 
Synthetic 
effect of 
cated the 
ow beam 
It has an 
S6) with 
5 III data 
gronomic 
at the S6 
fields in 
scatter of 
could be 
ng water, 
moisture 
arge area 
90). Soil 
/3-degree 
' ground- 
rved that 
scattering 
onditions 
unity has 
es to soil 
les. The 
zation it 
59? with 
am mode 
iited data 
n is also 
er studies 
oisture is 
yw beam 
Based on 
igate the 
ds at an 
) to study 
ice fields 
data of 
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India, 2002 
  
2. RESEARCH OBJECTIVES 
Objective of the study was to analyze the effect of steep and 
shallow incidence angle on backscatter from rice fields and to 
explore the possibility of retrieval of moisture status of rice 
fields using RADARSAT-1 Extended low beam data for soil 
moisture status of rice fields. 
3. STUDY AREA AND DATA USED 
The present study was conducted for Bardhaman district, West 
Bengal, India. The area was predominantly agricultural having 
more than 75 per cent area under agriculture. Rice was the 
dominant crop of the area. It is grown under rice-rice crop 
rotation. The main crop was grown during July-October and 
known as kharif. The second crop is grown during January to 
March and known as the summer crop. The summer crop was 
mainly irrigated by the canal irrigation of Damodar command 
and partially by ground water use. Our study addresses the 
summer crop, where the water use is controlled and there was 
no possibility of additional effect of rain on soil moisture. 
Semi-dwarf plant varieties of 100-110 day cycle were 
transplanted during January and harvested during April. 
Wetland practice was adopted in the area with 5-10 cm water 
level maintained upto 50 days after transplanting. Hence, the 
soil moisture effect could only be attempted only after grain 
filling stage, which is attained by late March. 
Extended low beam mode data of RADARSAT SAR at center 
incidence angle of 16? acquired on 23'? March 2000 and 
Standard beam mode 6 having center incidence angle of 40? 
acquired on 28" March 2000 were used for soil moisture study. 
Indian Remote Sensing (IRS) LISS -III data acquired on 31“ 
March 2000 was used to delineate crop area and assess growth. 
IRS-LISS-III has three bands in visible and near infrared (0.52- 
0.59, 0.62-0.68, 0.77-0.86 micron) and one band in short-wave 
infrared region (1.55-1.70 micron). The spatial resolution is 
23.6m for VNIR and 70.8m for SWIR. 
Ground truth data was collected on each day of satellite pass, 
which included soil samples and crop information. The soil 
moisture values for each of the sampling locations were 
measured by oven drying the soil samples for 24 hours. GPS 
and 1:50,000 scale Survey of India (SOI) topographic maps 
have been used to locate the sampling location during the 
ground truth collection. Soil and physiographic map of the area 
of 1:50,000 scale was used as base information to interpret the 
backscatter variation of rice fields. 
4. METHODOLOGY 
4.1 Selection of sampling location 
Selection of sampling locations for ground truth data collection 
for in-situ soil moisture measurements was carried out after 
reconnaissance field survey. Sampling locations were selected 
with the help of GPS. Enough care was taken regarding size 
(minimum 100meter X 100meter) and homogeneity of 
sampling location. Average of five samples was used to 
represent the each sample site. The fields selected were also 
nearly homogeneous with respect to crop cover. 
775 
4.2 Ground truth collection 
A ground truth data collection was done in synchronous with 
both the RADARSAT passes. The information gathered during 
ground truth data collection consisting of soil samples for 
gravimetric moisture measurements, crop variety, crop height, 
percent plant cover and associated field conditions. 
4.3 Radiometric calibration 
For the RADARSAT data supplied from CCRS, output scaling 
gain and offset are applied to the data to ensure optimum 
utilization of the available dynamic range. The scaling used can 
vary for each scene, making it difficult to directly relate 
information between scenes. Hence for any quantitative 
analysis, it is necessary to convert the image data to calibrated 
radar backscatter (Sigma naught). Both of extended low as well 
as Standard beam mode data were radio metrically calibrated 
using the Equation 1. 
o° = 10*log((DN? + offset)/gain)+10*log(sin (a) (1) 
where DN = Digital number of SAR image 
a = Local incidence angle 
For the calculation of a, the local incidence angle at each pixel, 
the necessary information required and the information of gain 
and offset applied were supplied in the header of the 
RADARSAT image. These conversions yielded a 32-bit real 
image, which is radio metrically, calibrated. These operations 
were carried out with the help of EASI/PACE image processing 
software. 
4.4 Speckle suppression and data compression 
  
LE Wu 
After conversion of DN to c?, Speckle suppression was carried 
out using Enhanced Lee filtering algorithm (Lee, 1986). Owing 
to the large disk space requirement (about 2 GB) and data 
handling, the 32 bit real o? images were converted to 8 bit 
unsigned images. A user defined linear relation was used for 
this purpose, so that once 8 bit DN values were extracted from 
image, later on it can be inverted to get backscattering 
coefficient values in order to carry out quantitative analysis.