d crop
d crop
rming.
'Cision
on soil
>, Crop
1elp in
'cision
ack of
arming
ing or
| input
igation
ich is
valent.
d. The
y good
ain-fed
Zentral
ulture,
lies at
have
uld be
3 t ha!
uld be
lopting
(Singh
darsha
init) in
India,
) times
yields
maize/
ystem,
0.984 t
etation
em by
Assured irrigation (around 96%) with large fields and adequate
capital for various agricultural inputs (fertilizer consumption-
0.158 t ha"), and consequent higher productivity (nearly 4 t
ha”) is the characteristic features of agriculture in Punjab,
Haryana and Western Uttar Pradesh. Optimization of
agricultural inputs and minimizing the cost of production and
environmental impact are to be focussed. It is quite evident
from the foregoing that in order to improve the agricultural
production, to be competitive in the emerging seamless global
economy and to maintain environmental health, two
strategies, namely adoption of soil and water conservation
measures, and minimizing the cost of cultivation need to be
addressed. The implicit fact in the strategy is the applications
of agricultural inputs based on crop demand, and soil attributes
rather than applying at uniform rate across the field.
At national level, information on the nature, extent, spatial
distribution, potentials and limitations is available only at
regional level (1:500,000 scale), at meso level (1:50,000 scale)
only for part of the country, and at micro level no information
is available. With respect to soil fertility status, as pointed out
earlier, only regional / district-level recommendations based
on crop response trials in experimental plots is available,
which is used as a base for fertilizer applications. There is,
therefore, need to generate at least field-level information on
soil fertility. Similarly, for crop production, water resources is
equally, if not more, important. Optimal utilization of
irrigation water needs due focus. As witnessed in command
areas, if not managed properly, it may lead to waterlogging and
subsequent development of soil salinity and/or alkalinity.
A beginning towards adoption of precision farming could be
made in India by creating awareness amongst farmers about
consequences of applying imbalanced doses of farm inputs like
irrigation, fertilizers, insecticides and pesticides. The next step
would be the evaluation of soil fertility at individual field/plot
level and make it available to farmers for fertilizer
applications. Once it is achieved, in-field variability in soil
fertility need to be looked at and managed by judiciously
applying plant nutrients.
3.0 ROLE OF SPACE TECHNOLOGY
As evident from the foregoing, in order to pursue precision
farming, baseline information on nature, extent spatial
distribution, potentials and limitations of soils is a pre-
requisite. Since information on soils at meso-level is available
only for part of the country, such information needs to be made
available for entire country. Spaceborne multispectral
measurements have been operationally used for deriving
information on soils (Hilwig, and Karale,1973; Korolyuk and
Shcherbenko, 1994), and soil limitations like soil erosion
(Karale et al., 1989; Dwivedi et al.,1997), soil salinity and/or
alkalinity, waterlogging, etc. (Metterricht and Zinck, 1997;
Dwivedi et al,2001). The Department of Space, Government
of India has already taken initiative to generate soil resources
maps at 1:50,000 scale for entire country using the Indian
Remote Sensing Satellite (IRS-1C/-1D Linear Imaging Self-
scanning Sensor (LISS-III) data.
The next step would be to generate detailed-level information
on soil resources addressing potentials and limitations of
individual fields since except for states like Punjab, Haryana,
Madhya Pradesh and Maharastra where fields size is quite
large, practically individual field could be treated as a
homogenous management unit for the purpose of precision
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India, 2002
309
farming. Currently available high spatial resolution
multispectral data from IKONOS-II and Quick Bird-II, and
those from planned earth observation missions, namely
Resourcesat-1, Cartosat-1 and II would enable generating
desired information. Remote sensing has shown encouraging
results in providing information on soil fertility. Laboratory
and in situ spectral measurements have been directly related to
variability in soil organic matter (Baumgardner et al., 1970),
soil calcium carbonate content (Leone et al., 1995), iron oxide
content (Coleman and Montgomery, 1987), and soil nutrients
particularly those associated with soil texture and drainage
(Thompson and Robert, 1995).
Information on the potential yield that can be achieved from
a given piece of land and the likely yield of existing. crop is
required to bridge the gap by suitably adjusting the agricultural
inputs. Crop growth simulation model provide information on
potential yield while multispectral measurements made from
air and spaceborne platforms have shown immense potentials
in crop yield estimation and forecasting using spectral indices
(Tucker et al., 1980; Navalgund, 1991; Yang and Anderson,
1996).
Remote sensing also holds good promise in deriving
information on seasonally-variable soil and crop parameters,
namely soil moisture status, crop conditions like vigour,
infestation of weeds, pests and disease, required for farm
management. Spectral measurements in thermal regions have
been related with the variations in soil moisture content (Idso
et al, 1975). In fact, the combination of long and short
wavelengths e.g. Ku-band at 2 cm or x-band at 3 cm, have
been used for assessment of within -the- field soil moisture
conditions (Prevot et al., 1993).
Various stages of crop development i.e., grain filling in wheat
and anthesis of corn have been related to spectral
measurements (Railyan and Korobov,1993; Boissard et
al., 1993). Likewise, spectral measurements help measuring or
monitoring crop growth through empirical correlation of
Vegetation Index (VI) with such crop variables as leaf area
index (LAI), per cent vegetation cover, vegetation phytomass
and fraction of absorbed photosynthetically active radiance
(fAPAR) required for calibration and validation of crop growth
simulation models (Pinter, 1993). In addition, remotely sensed
data could be used for deriving crop co-efficients (the ratio of
actual crop evapo-transpiration and that of a reference crop)
for estimation of actual, site-specific crop evapo-transpiration
rate from readily available meteorological information
(Bausch, 1993; Ray and Dadhwal, 2001).
Reflectance measurements in the green (0.545 pm) spectral
band have been related to plant nitrogen content and canopy
nitrogen deficits (Fernandez et al., 1994). Besides, remote
sensing has some potential for detecting and identifying crop
diseases (Malthus and Madeira, 1993), weed infestation
(Brown et al., 1994) and insect infestation (Yang and Chang,
2001). Furthermore, remote sensing has a variety of roles in
determining the cause of spatial and temporal crop and soil
variability. The most obvious role is the use of remote sensing
information to improve the capacity and accuracy of decision
support system (DSS) and agronomic models by providing
accurate input information or as a means of model calibration
or validation. Another role is the use of hyperspectral images
for direct crop diagnosis.
The Geographic Information System (GIS) contributes
significantly to precision farming by allowing presentation of
