or given
iance of
xpressed
Stratified
it Based
ting the
>s and to
generally
illage is
Is treated
on study
selected
sponding
ors other
Sampling
Stratified
echnique
To. The
imate of
on of the
1g GIS is
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring", Hyderabad, India, 2002
Spatial Data Attribute Data
Ax U
UR Lr
E of E uns
ASS
Estimation
pm
+
»-.
ev,
Sample Selection
ac
neighbours
Calculation of Spatial
Fig 1 Implementation of Spatial Sampling Technique using GIS
The percent relative bias, relative efficiency and coefficient of
variation of different estimators are shown in table 2. All the
estimators seem to be almost unbiased. The proposed
estimators were found to be highly efficient as compared to the
estimators obtained by using traditional sampling techniques or
by DUST proposed by Arbia.
Table 2 The Relative Bias, Relative Efficiency and Coefficient
of Variation for different estimators
Esti Relative Relative Coefficient of
mato Bias (%) Efficiency | Variation
rs
Ti 0.030 8.79 2.63
T, 0.020 11.63 2.28
T3 0.114 - 727
T4 0.032 1.68 5.95
Ts 0.156 1.29 6.79
Te 0.014 1.51 6.29
5. EMPIRICAL STUDY USING REMOTE SENSING DATA
An empirical study has been carried out to estimate area under
wheat crop in the Rohtak district using remote sensing parameter,
Normalized Difference Vegetation Index (NDVI). The mean NDVI
of wheat pixels for each village was computed and was considered
asthe auxiliary character for estimating the area under wheat crop.
The estimates of area under wheat were obtained using the
estimators of Stratified CUBSS (T;) since its was found to be the
most efficient sampling technique. Area under wheat was also
calculated using usual remote sensing method of estimation. These
two estimates were compared with the actual area under wheat crop,
which was obtained from the patwari records. The entire
methodology followed for this part of study is explained with the
help of flow chart in Figure 2. The results emerging from this part
of study are highlighted in table 3.
