ity and
tograph
rom the
in how
rom 7-7
riterion
, In 3-4
first and
4). The
rors are
iction of
relative
e. This
iple size.
and 80
b) With the decrease of sample size, the differences between
the relative variances of estimations, corresponding to separate
sampling plans, do also decrease. The reason for this is that the
total sample, which was used for the calculations of the error of
separate sampling plans, does contain the data selected for the
testing of the sampling plans, and in case of large sample sizes
the overlap between the two data sets and samples is larger.
c) There is not significant difference between the weighted and
non weighted sampling plans.
d) The estimation errors decrease from sampling stratum ] to
stratum 4, but the errors of strata 3 and 4 are practically the
same. The reason for the decrease of error is the decrease in
GR. .
Estimation error
non-weighted sample
2.5
t o
9 20 o g 2
o LT
g = 15 A4 strata
E = 1.0 02 strata
2
2 ost —
o |
0.0 + } i } 4
0 20 4 60 80 100
Sample size
Estimation error
4-strata, non-weighted sample
4 el ising
5 AA Astratum 1
£ 3 A A |Ostratum 2
y 8 2 Eat © stratum 3
E T. Ostratum 4
fud
o
0 + + + +
0 20 4 60 80 100
Sample size
Estimation error
2-strata, non-weighted error
gal 4
£3
EZ A
25 AA A | Astrata 1-2
& 3 2 O—OÓ Ó
99 Ostrata 3-4
= |
8
0 } } } }
0 20 40 60 80 100
| Sample size
Figure 5. Relative estimation variances and estimation errors
of applying the field sampling plans
CONCLUSIONS
Our criteria for the field sampling plan are the following:
a) The conclusions drawn from the sampling plan should be
right.
b) The sample size should be minimal.
c) The sampling plan should be the simplest.
Criterion a) excludes all sampling plans that result in
'erroneous conclusion based on Table 3, and criterion 2
excludes the sample size of 80. The third criterion is the
weakest, since the difficulty of obtaining and evaluating an
aerial photograph is much larger than that of its stratification,
and weighting by intensity. Based on these, we suggest the use
of 2-strata sampling plan with sample size of 20. There was not
significant difference between the weighted and non weighted
sampling plans. In that case when there are few samples falling
into the sampling strata with large intensity, based on spatial
statistical considerations, we suggest the use of weighted
sampling plan.
The suggested technique for the preparation of allocating plans
of reclaiming materials is applicable not only for gypsum, but
for other materials as well, if there is linear relationship
between the intensity of aerial photograph and the required
amount of reclaiming matenal.
REFERENCES
Csillag F., L. Pasztor, and L. L. Biehl. 1993. Spectral band
selection for the characterization of salinity status of soils.
Remote Sensing of Environment. 43 pp. 231-242.
Kertész, M. and T. Toth. 1994. Soil survey based on sampling
scheme adjusted to local heterogeneity. Agrokémia és Talajtan.
43 pp. 113-132.
Mc Bride, M. 1994. Environmental chemistry of soils. Oxford
University Press. New York.
Metternicht, G. I. and Zinck, J. A. 1996. Modelling salinity-
alkalinity classes for mapping salt-affected topsoils in the
semiarid valleys of Cochabamba (Bolivia). ITC Journal 1996
(2) pp. 125-135.
MSZ 9693/1-77 (National standard for the quality of soil
reclaiming materials) (in Hungarian).
Tôth, T and M. Kertész. 1996. Application of soil-vegetation
correlation to optimal resolution mapping of solonetzic
rangeland. Arid Soil Research and Rehabilitation. 10 pp. 1-12.
Tóth, T. and L. Pásztor. 1996. Field reflectance measurements
as a means of distinguishing different grades of salinity and
alkalinity. pp. 23-34. in: Soil salinization and alkalization in
Europe. Thessaloniki.
Tóth, T., F. Csillag, L. L. Biehl and E. Michéh. 1991a.
Characterization of semi-vegetated salt-affected soils by means
of field remote sensing. Remote Sensing of Environment. 37
pp. 167-180.
Tóth, T., F. Csillag and Gy. Büttner. 1991b. Satellite remote
sensing of salinity-alkalinity in the Great Hungarian Plain.
Proceedings of International Symposium Impacts of
Salinization and Acidification on Terrestrial Ecosystem and its
Rehabilitation. September 26-28, Fuchu, Tokyo, Japan. pp.
100-107.
van Meirvenne, M., P. de Groote, M. Kertész, T. Tóth and G.
Hofman. 1996. Multivariate geostatistical inventory of sodicity
hazard in the Hungarian puszta. in: Monitoring soils in the
environment with remote sensing and GIS. Proceedings of the
ISSS International Symposium (working group RS and DM)
Ouagadougou, from 6 to 10 Feb. 1995). pp. 293-305.
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 181
