|
|
|
Remote Sensing and GIS based Sampling Methodology for Estimation of Crop Acreage in North-Eastern Hilly Region
Dr. (Mrs)Prachi Misra Sahoo
Scientist(Senior Scale)
Indian Agricultural Statistics Research Institute,
India
Email: rprachi@iasri.res.in
Dr. Anil Rai
Senior Scientist
Indian Agricultural Statistics Research Institute
Email: anilrai@iasri.res.in
Dr. Randhir Singh
Principal Scientist
Indian Agricultural Statistics Research Institute, New Delhi
Email: rsingh@iasri.res.in
J. S. Parihar
Group Director, RSAM
Space Application Centre
Availability of reliable and timely agricultural statistics has paramount importance for policy decisions regarding production, pricing, procurement, marketing, export/import, public distribution etc in the country like India predominantly influence by agrarian economy. The Directorate of Economics & Statistics in the Department of Agriculture & Cooperation, Ministry of Agriculture is the nodal official agency for collection, compilation and publication of major agricultural statistics like area and production of principal crops in temporary settled states and permanently settled states. But crop statistics in the North-Eastern Region, Sikkim, Goa, UTs of Andaman Islands, Daman & Diu and Lakshwadeep are collected on the basis of ad-hoc methods based on impressionistic approach from the village headman which are quite subjective and unreliable. Satellite Remote sensing has been successfully used for crop acreage estimation in the country except North East regions because of specific problems pertaining to this region such as undulating topography, non-accessibility to vast area, less percentage of area under the crops, thick forest cover and cloud coverage over the area most of the time during the year. As there are no cadastral maps and well-defined village boundary maps for these regions, reliable information regarding total number of villages in each district/block is not available. Also information on total number of farmers in a village, number of fields owned by each farmer, crops grown by the framer etc. are not available in the records. Therefore, use of remote sensing satellite data alone may not be able to provide reliable information regarding crop acreages. Keeping all this in view, a sampling methodology was proposed in the present study for retrieving reliable acreage estimates using satellite data along with the well designed sample survey in GIS environment.
The study has been carried in the district Ri Bhoi of Meghalaya which is the rice bowl of the state. Paddy is major crop of the district growing in three seasons namely Autumn (Ahu), Winter (Sali) and Spring (Boro). The winter paddy crop accounts for 90 per cent of the total area under paddy in a year. The present study was confined to estimate area under winter rice only.
The problems in use of satellite remote sensing in the hilly area for acreage estimation was handled through following approaches. A composite image was developed using mulitdate satellite data (i.e IRS 1D, LISS-III) acquired during rabi season to obtain a cloud free image of the area. The supervised maximum likelihood classification procedure was followed to transform the composite multi-spectral data into land use/ land cover map and rice growing area were delineated with the help ground truths. Due to undulating topography, topographic geometry and misclassification, there was large difference of area under crop in the image and actual area under crop. In order to rectify the estimate of area under paddy crop affected by these factors, a relationship between area under paddy in the classified image and actual area under paddy on the ground has been developed. The area under paddy crop falling under hill shades or deep valleys could not be estimated through conventional classification. Further, smaller paddy fields were also not detectable due to lower spatial resolution of the sensors. The extent of errors in the estimate due to the loss of area under paddy in hill shades and limitations of spatial resolution of the sensor, has been rectified by a suitable sample survey conducted in a buffer zone of 250m created along the National Highway and 14 state roads of the district in GIS environment. The roads were conceptually divided into segments of 500 meters. A sample of segments have been selected randomly on each of the road and a grid of 500X 500 m2 has been observed for recording the area under paddy by GPS as well as eye estimate. Suitable estimates for area under paddy were developed.
This study provides estimates of crop acreage on scientific basis involving remote sensing and GIS. Further, the relationships first time developed are likely to stabilize within few years say three to five years after which field survey may not be required for estimation of crop area and the crop acreage estimates would be obtained only through digital classification of remote sensing data. The accuracy of estimation by this method depends on the spatial resolution of the satellite sensor and area coverage of the crop under consideration.
|
|
|