ESTIMATION OF INDIAN AGRICULTURAL PRODUCTIVITY
BASED ON PRODUCTION EFFICIENCY MODEL
a go Hooda, 52.6, Dye and R°. Shibasaki
lüaryana State Remote Sensing Application Centre, CCS HAU Campus,
2 Hisar 125 004, India.
Department of Geography, Boston University, 675 Commonwealth
3 Avenue, Boston, MA, 02215 - 1401, USA.
Institute of Industrial. Science, University of Tokyo, 7-22-1
Roppongi, Minato-ku, Tokyo-106, Japan.
KEY WORDS: Remote Sensing, Agriculture, Productivity, Estimation,
Modeling, Photosynthetically Active Radiations
ABSTRACT:
Production Efficiency Model (PEM) being used to evaluate Net
Primary Productivity (NPP) requires decomposition of productivity
into independent parameters involved in the production built up
process. PEM has been used for the estimation of NPP of the
natural vegetation but in a first attempt of its kind it was used
to estimate agricultural productivity for Indian territory. The
study involved mainly three steps, (i) identification and map-
ping of=agricultural areas (ii) estimation of, agricultural pro-
duction and (iii) analyses of annual and interannual variations
in agricultural productivity.
The agricultural areas were identified and mapped using NDVI-
Climatological modeling technique. NASA/NOAA Pathfinder AVHRR
Land (PAL) 10 day composited NDVI data with a spatial resolution
of 8 km was used for the study. The agricultural pixels were
identified as outliers in the NDVI-rainfall relationship devel-
oped using annual integrated NDVI and annual rainfall data for
the year 1989. An irrigated agricultural areas map was generated
using the value of these pixels.
The NDVI data for the years 1987, 1988, 1989 was used to estimate
fraction of PAR absorbed (fAPAR) based on the relationship fAPAR
= -0.31+1.39*NDVI provided by the SAIL model. Incident PAR (IPAR)
data set for India was extracted from the monthly global IPAR
data set already generated using UV reflectivity data from Nimbus
Total Ozone Mapping Spectrometer (TOMS). The IPAR data when com-
bined with the fAPAR data, provided absorbed PAR (APAR). Assuming
the irrigated agricultural, areas mapped above as constant over
the three years period, the agricultural APAR was extracted using
the irrigated agricultural areas mask. Agricultural APAR was
subsequently converted to agricultural NPP using the mean conver-
sion efficiency (c) value of 2.07 calculated for cultivations
based on literature survey. The agricultural NPP was finally
converted to economic yield based on the area weighted average
harvesting index of various crops grown in India. The annual and
interannual variation in agricultural productivity of India have
been discussed vis-a-vis reliability of the model for these
studies.
1. INTRODUCTION Agricultural productivity with
its. fundamental role in food
Intenational Geosphere Bio- supply : has been the obvious
Sphere Program (IGBP, 1992) focus. Due to its dependence on
envisaged creation of improved various external factors and
global data sets to properly the attendant uncertainties,
evaluate the environmental large regional disparities
changes occuring on regional exist "in the agricultural pro:
and global scales. Frequent duction which needs to be
availability of: remote sensing properly evaluated for global
data through various satellites planning. Studies on: products
have now made it possible to ivity usuai?y focus on two
get better estimates of carbon aspects:
fixation and terrestrial pro- (a) To predict the crop produt:
ductivity on earth. Various tion of a Certain year "before
estimates for global. net prim- harvest using simple statisti:
ary productivity has, been made cal models. Models used for
with rather large discripancies this purpose, specially Spec
between the estimates (Ruimy tral Indices-Yield Regression
et.al: 1994). Better tech- models (Dubey et.al. 1994), are
niques, therefore, needs -to be usually developed for a certain
developed for making reliable kind of crop in à j/small region
estimates of productivity. and have strong dependence OR
298
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996
