The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008
1342
3.2.2 Getting the data of crops area: Using GPS to get
spatial information attribute of polygons. The proportion of
surveying crop such as rice based on sampling frames is easily
calculated in GIS, which is used as sampling unit data. RSAC
used two continuous years’ data of GRS to calculate the
variation rate of crop area based on statistical rules at a certain
surveying region such as administrative district.
3.2.3 Sampling accuracy and samples amount: It is
obvious that increasing samples amount can improve the
sampling accuracy. However, increasing samples amount will
definitely increase the cost of surveying. Thus it is necessary to
balance the cost and the accuracy of surveying. The formula
used to express the relation between samples amount and
sampling accuracy is shown as below (Wu, 2004):
(9)
Where (7 = standard deviation
n = samples amount
A = sampling accuracy
Z = the parameter of standard normal distribution
1 — a = confidence level
In this formula, the standard deviation can be replaced with the
variable error of samples via testing in advance. When the
sampling accuracy is fixed on a certain value such as 0.95, the
least samples amount can be calculated.
4. CONCLUSION
Using RS and GIS, stratified sampling method has been
successfully applied to main crop area monitoring for many
years at national scale in China. RSAC uses two kinds of
methods to stratify: one is called Frequency Accumulation
Means (FAM) and the other is System Cluster Means (SCM).
The two methods have different characters. With GPS, Ground
Random Sampling (GRS) has also been adopted as complement
to RS when estimating the variation rate of main crop’ area in
China.
ACKNOWLEDGEMENTS
The data used for stratified sampling and calculating crops area
is produced by RSAC. With deep gratitude we want to give our
thanks to all the members of RSAC.
REFERENCES AND/OR SELECTED BIBLIOGRAPHY
Chen, Z., etc., J.,2000. Sampling and scaling scheme for
monitoring the change of winter wheat acreage in China.
Transactions of the Chinese Society of Agricultural Enginerring,
16(5), pp. 126-129.(in China).
Duda T., etc., J.,2002. Unsupervised classification of satellite
imagery: choosing a good algorithm. Remote Sensing, 23(11),
pp. 2193-2212.
Jiao, X., etc., J.,2002. Design of sampling method for cotton
field area estimation using remote sensing at a national level.
Transactions of the Chinese Society of Agricultural Enginerring,
18(4), pp. 159-162.(in China).
Liu, H.,J.,1999. MARS project of European Union and plan of
China using RS. Chinese Journal of Agricultural Resources
and Regional Planning, 20(3), pp. 55-57.(in China)
Panigrahy S., etc., J.,1997. Mapping of crop rotation using
multidate Indian remote sensing satellite digital data.
Photogrammetry & Remote Sensing, 52, pp. 85-91.
Wu, Q., J.,2004. Influence of small features on crop area
estimation at a nation scale using remote sensing and a double
sampling method. Chinese Journal of Agricultural Resources
and Regional Planning, 20(3), pp. 130-133.(in China)
Yang X., etc., J.,2002. Using a time series of satellite imagery
to detect land use and land cover changes in the Atlanta,
Georgia metropolitan area. Remote Sensing, 23(9), pp. 1175-
1798.
Chen, S.,etc, M.,1990. Geography Analysis of Remote Sensing.
Beijing, Mapping publishing, pp. 147-167. (in China)
Hu, R., etc, M.,2002. Application of RS and yield estimation of
Agricultural crops in European Union. Beijing, Weather
publishing, pp. 34-59. (in China)
Thomas M. Lillesand, etc.,2002. Remote sensing and imagery
interpretation. New York, John Wiley & Sons, Inc, pp. 576-586.