spatial data in the form of a map. In addition, GIS forms an
ideal platform for the storage and management of model input
data and the presentation of model results which process model
provides. The Global Positioning System (GPS) technology
provides accurate positioning system necessary for field
implementation of variable rate technology (VRT). The
Internet makes possible the development of a mechanism for
effective farm management using remote sensing. The
potentials of remote sensing in providing information required
for precision farming, in general, have been reviewed by
Moran et al.(1997), and for Indian conditions by Ray et
al.(2001).
4.0 THE INDIAN INITIATIVE
Realising the potential of space technology in precision
farming, the Department of Space, Government of India has
initiated eight pilot studies in well-managed agricultural farms
of the ICRISAT, the Indian Council of Agricultural Research
and the Agricultural Universities, as well as in farmers' fields.
The pilot studies aim at delineating homogeneous zones with
respect to soil fertility and crop yield, estimation of potential
yield, yield gap analysis, monitoring seasonally-variable soil
and crop conditions using optical and microwave sensor data,
and matching the farm inputs to bridge the gap between
potential and actual yield through spatial decision support
systems (SDSS). The test sites are spread over a fairly large
area across a cross section of agro-climatic zones of the
Indian sub-continent, and cover some of the important crops
like wheat, rice, sorghum, pigeon pea, chickpea, soybean and
groundnut.
5.1 A Case Study
The study was taken up (i) to analyze the gap between
potential and existing crop yields using crop growth simulation
models, and (ii) to develop a spatial decision support system
(SDSS) at ICRISAT farm bound by geo-co-ordinates17.6° to
7.33? N and 78.1° to 78.4° E, and located in Patancheru,
Medak district of Andhra Pradesh, southern India (Fig-1). The
test site forms part of the pediplain developed over granite-
gneiss complex. Both red soils (Alfisols) and black soils
(Vertisols) and their intergrades are encountered in the farm.
The climate is semi-arid and sub-tropical with around 800 mm
of mean annual rainfall, which is received mostly from
southwest monsoon. Within the farm, two nano-watersheds-
one in the red soils (RW2) and another in black soils (BW7)
have been selected. Whereas sorghum and groundnut were
taken in RW2, BW7 had two types of cropping system viz., (i)
Soybean var. PK72- a 90 to 100 days crop, during kharif
followed by chickpea during rabi, and (ii) Soybean-pigeonpea
(ICPL87119- a 210 to 240 days crop) intercrop. In the
following section, the work on BW7 nano-watershed will be
briefly discussed.
Soybean and pigeonpea were sown on June 21,2002 with a
row to row spacing of 30cm in case of soybean sole, and
22.5cm for soybean and pigeon pea intercrop. A plant-to-plant
distance of 7-10cm was maintained for soybean whereas it was
25cm for pigeonpea. In order to demonstrate the utility of
raised-bed system, crops were sown in two types of land
configuration- the conventional flat bed, and broad bed and
furrow (BBF) system which facilitates free movement of farm
machinery apart from maintaining good drainage especially
during rainy season. Whereas the distance between the
furrows was maintained at 150cm, the furrow width and bed
height were kept at 30cm and 20cm, respectively. The raised
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India, 2002
310
beds were tending to taper on either side towards the furrow
while leaving an effective bed width of 110cm for sowing
crops. Due care was taken to keep the crop free from weeds,
insects, pests and diseases.
Owing to persistent cloud cover during monsoon (kharif
season), the microwave data from synthetic aperture radar
(SAR) onboard Radarsat-1 in fine resolution beam mode with
a spatial resolution of 8m and acquired from three overpasses-
one on Augustl4, another on September7 and the third on
October1,2002 were utilized. The cloud-free IRS-1D LISS-III
and PAN data were also collected. Apart from satellite
overpass-synchronous ground truth for soil moisture estimation
and for deriving biophysical parameters crops, namely leaf
area index (LAT) and phytomass, such observations were also
routinely made at biweekly intervals throughout the crop
growing season. Soybean crop was harvested on October] and
2,2002. A 3x3m sample size was selected for harvesting and
ultimately for yield estimation and mapping. Samples were
taken both from broad bed and furrow (BBF) and flat bed (FB)
plots. The locations of each segment was identified with the
help of Nikon Total Station model 801.
An attempt was made to correlate the back scattering
coefficient as measured by Radarsat-1 SAR and soil
moisture, and crop parameters including LAI and phytomass.
An analysis of simulated yield using Agriculture Production
Systems Simulator (APSIM) model and observed yield data
(Table-1) for soybean/pigeonpea crop system for 1999-2000
reveals a seed yield gap of 783kg ha'! (Singh et al.,2002).
Table-1 Simulated and observed yields (t ha!) of soybean/
pigeonpea intercrop system on a shallow soil during 1999 -
2000 season
Flat shallow BBF shallow
Soybean Pigeonpea Soybean Pigeonpea
Simulated yield
Total biomass 7162 1985 7377. 1980
Seed yield 2083 325 2145 334
Observed yield
Total biomass 3286 1861 3781 2107
Seed yield 1300 . 603 1497 721
Singh et al., 2002 (Unpublished).
Furthermore, a close look at historical yield data for a field
within BW7 nanao-watershed, which is based on random
sample of 10x4.5m segments, indicates a fairly large in-field
heterogeneity (3,756 to 4,556kg/ha) in soybean grain yield
(Table-2).
Table-2 Soybean (variety PK472) yield at ICRISAT farm in
1999,
S.No Weight(kg)* Yield(kg/ha)
Grain Fodder Grain Fodder
with pods with pods
1 6.911 16.9 1,536 3,756
2 8.173 19.8 1,816 4,400
3 8.556 20.5 1,901 4,556
4 7.66 18.7 1,702 4,156
5 8.344 19.4 1,854 4,311
*Se;
Cou
Gro
Prec
in |
stak
How
be «
redi
witl
pote
con
esti:
to f
Croj
maj
mul
low
not
mul
det
ord
pre
nee
ben
Bai
Zac
pro
Bai
bas
227
Bo
gre
ant
Br
19
COI
