] RC, 
index 
mote 
ndian 
a on 
r 11- 
ntre: 
]ture. 
(Eds: 
i, pp 
on of 
pping 
, and 
lation 
igital 
169- 
NVK 
yield 
at. J. 
KP, 
ndian 
crop 
9. 
idhari 
for a 
adesh 
lation 
ature. 
nt of 
d are 
2 and 
o the 
ribed 
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India, 2002 
AN INTEGRATED REMOTE SENSING AND GIS APPROACH FOR 
SPATIAL SAMPLING TECHNIQUE 
Prachi Misra Sahoo, Randhir Singh and Anil Rai 
Indian Agricultural Statistics Research Institute, Library Avenue, New Delhi, India (rprachi, rsingh, anilrai) @iasri.delhi.nic.in 
KEY WORDS: Spatial autocorrelation, Spatial sampling, GIS, Remote sensing, NDVI, Contiguity based neighbour 
ABSTRACT : 
The aim of this study was to explore the advantage of integrating GIS and remote sensing technologies in sampling of spatial data. The 
traditional sampling techniques does not account for the dependency present in the spatial data. As a result of which they fails to provide 
stable and reliable estimates. Therefore there is need to modify the existing sampling techniques if applied to spatial data. The proposed 
sampling techniques were developed by studying the correlation structure present in the data. The estimators were found to be more 
efficient and stable than the one obtained from traditional sampling techniques. The handling of enormous spatial data was eased by the 
use of the capabilities of GIS. The integration of Remote sensing with GIS further enhanced the performance of the estimators. 
1. INTRODUCTION 
In agricultural surveys the parameter of interest is often 
geographical in nature. This implies that the observations are not 
independent in nature. Such data is termed as spatial data. The 
existence of dependence in the spatial data violates the basic 
assumption of independence of the traditional sampling 
procedures raising questions of estimator sufficiency, bias, 
efficiency and consistency. Therefore for sampling spatial data 
classical methods are inappropriate unless modified in some way. 
The recent technological developments in the computer and space 
technology have shifted the emphasis of survey research work 
display of spatial data in the form of Geographical Information 
System (GIS). Also Remote Sensing in the form of satellite data 
emerged as a powerful tool for agricultural surveys and can be put 
to best use if they are incorporated in GIS. Capabilities of GIS, 
therefore, when combined with an up-to-date remote sensing 
system, become manifold. The present study aims to improve the 
conventional survey methodology for agricultural surveys 
particularly for area estimation by exploring the potential of GIS 
and Remote sensing techniques in handling spatial data. 
2. STUDY AREA AND DATA USED IN THE STUDY 
The study has been carried out for the Rohtak district of Haryana. 
The data which were used for this study along with their source 
  
towards newer emerging areas. A major development has been ale mentioned in table L 
that of integrated software for the capture, storage, analysis and 
Table 1. Data used in the study 
Source 
Data Type Data Form 
  
Spatial Data Village boundary maps 
District Handbook of Census (DHC) 1991 
  
Attribute data Village-wise data of irrigated and 
cultivated area 
District Handbook of Census (DHC) 1991 
  
Sensed data Row 47, Date of pass - 17 Feb., 1996 
Remotely Satellite data of IRS-1B ,LISS-II, Path 30, 
National Remote Sensing Agency (NRSA), Hyderabad 
  
Collateral data 1) Toposheets (1: 50, 000 scale) 
2) Figure for area under wheat in the 
district 
  
  
1) Source: Survey of India (1: 50, 000 scale) 
2) From Patwari records 
  
  
  
3. METHODOLOGY 
Consider a population of N areal units out of which a sample of 
size n was selected. Let Y be the character under study and X be 
an auxiliary character highly correlated with Y. Let D be the first 
lag spatial correlation for the auxiliary character. Contiguous 
neighbours were identified for calculating the spatial correlation. 
The sampling technique proposed for the estimation of population 
mean was thus termed as Contiguous 
Unit Based Spatial Sampling (CUBSS) technique is described 
below: 
3.1 Contiguous Unit Based Spatial Sampling (CUBSS)