Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130

http://www.ijehse.com/content/12/1/130 JOURNAL OF
ENVIRONMENTAL HEALTH
SCIENCE & ENGINEERING

RESEARCH ARTICLE Open Access

The use of the GIS Kriging technique to determine

the spatial changes of natural radionuclide
concentrations in soil and forest cover
Turgay Dindarolu

Abstract
Background: The distribution of radionuclides occurring naturally in the earth depends on bedrock characteristics.
Therefore, the spatial distribution of radionuclides is not uniform. Consequently, radionuclide information is vitally
important in determining and monitoring the spatial variation of the radionuclide concentrations that are over the
limits for the sustainable environment and human health.
Methods: This research was carried out using GIS methods and geostatistical analysis as Kriging techniques to
reveal the spatial variation of the 226Ra, 232Th and 40 K concentrations of natural radionuclides in the oruh and
Aras Basin. The spatial variations of the detected radionuclides were correlated with soil groups and forest cover.
Results: In the study area, 43.17% of the concentration of 40 K, 26.67% of the concentration of 226Ra and 28.16%
of the 232Th concentration was determined to be over the average limits. Concentrations of radionuclides that are
over the average limits have been found to be on basalt and chestnut soils. Brown and reddish brown soils have a
low concentration of the spatial distribution of the radionuclides. Statistically positive correlations were detected
(0.865 **) between the 226Ra and 232Th. In addition, a positive relationship between forest cover and 226Ra and a
negative relationship between 232Th and 40 K were identified.
Conclusions: Excessive exposure to radiation may cause cancer and hereditary diseases. Ecological environments
include the soil and the plants. Hence, the periodical monitoring of the spatial variation in concentrations of
radionuclides is very important for the health of future generations.
Keywords: Natural radionuclide, Kriging technique, Soil, Forest cover

Background UNSCEAR. If the concentration of natural radioactive

Radioactive features have existed in our world since its elements goes over the average limits, there can be nega-
formation. High concentrations of natural radionuclides tive effects on human health [3]. Therefore, the spatial
are found in volcanic, phosphate, granite and salt rock. distribution of the concentrations of natural radionuclides
Rain and other water discharge crumble these rocks into in the soil has to be determined. It is possible to use CBS
very small pieces and mixes them into the groundwater. and geostatistical analysis to obtain these values.
Thus, rocks increase the natural radioactivity of the soil. Geostatistical techniques are a useful component of
The direction of the movement and the speed of the GIS applications that are frequently applied. Geosta-
radionuclides in the soil depends on natural processes tistics involves the analysis and estimation techniques
(e.g., soil structure, content of the plant species, irriga- used to obtain the value of a variable dispersed in time
tion conditions, weather conditions and accumulation) and location.
[1,2]. In some areas, the natural radionuclide concen- Kriging is one of the best and most widely-known
trations are above the established limits according to techniques used in spatial linear predictions. Kriging
methods have different flexible forms, according to the
Correspondence: [email protected]
survey area and data [4-7]. Kriging can also reveal the
Faculty of Forestry, Department of Forest Engineering, Kahramanmaras Sutcu reliability of the estimated surface [8,9].
imam University, 46100 Kahramanmaras, Turkey

2014 Dindarolu; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 2 of 11
http://www.ijehse.com/content/12/1/130

Geostatistical methods also allow for examining a rela- using this technique. Concentrations of 226Ra, 232Th and
40
tionship with spatial variations of radionuclides between K were determined for the distribution area. The
forest cover and soil groups. Radioactive elements in the experimental variogram model was constructed using the
soil do not indicate a uniform distribution in the earth. Kriging method, with data obtained from the research
Therefore, the concentrations of radionuclides should be area. The spatial transformation was performed to
checked regularly as an important step in protection determine the most appropriate model to use with the
from the negative effects of radioactivity [10-12]. parameters of the generated maps.
Radionuclide concentrations may be caused by the The ordinary Kriging formula is as follows [8,21]:
high amount of organic matter in the soil. As such,
radionuclides can be absorbed by the forest soil [13,14]. X
N
Z S0 i Z Si
Some radionuclide compounds build up in the humic i1
acids in the soil organic layer [15,16].
Measuring the radioactivity concentration in the soil, where:
as well as the concentrations in the plants and the water, Z(si) is the measured value at the location (ith),
is necessary to estimate the concentrations of radionu- i is the unknown weight for the measured value at
clides [17,18]. This study used 226Ra, 232Th and 40 K the location (ith) and
concentration values measured periodically by TAEK s0 is the estimation location.
[19]. Spatial distributions were determined according The unknown weight (p) depends on the distance to
to UNSCEAR (2000) [3], who used a kriging technique the location of the prediction and the spatial relationships
in the oruh and Aras Basin and the surrounding among the measured values.
areas. The statistical relationships between the spatial The statistical model estimates the unmeasured values
variations, forest cover and soil groups were analyzed. using known values. A small difference occurs between
the true value Z(s0) and the predicted value, i Z(si).
Materials and methods Therefore, the statistical prediction is minimized using
This research was carried out in 10 provinces in eastern the following formula:
Turkey: Rize, Bayburt, Erzurum, Artvin, Ardahan, Kars, " #2
Idr, Mu, Bingl and Erzincan. The bounding geograph- X
N
Z S0 i Z Si
ical coordinates of the study area are 395007 to 3946 i1
17 north latitudes and 382652 to 443546 east lon-
gitudes (Figure 1). The study area is 9.815.000 ha in size. The Kriging interpolation technique is made pos-
Forest areas were detected using local Forest Management sible by transferring data into the GIS environment.
Plan data [20]. In this way, analysis in areas that have no data can be
conducted. The following criteria were used to evalu-
Sampling methods and analysis ate the model: the average error (ME) must be close
Concentrations of radioactive elements (226Ra, 232Th and to 0 and the square root of the estimated error of the
40
K) for determining soil sampling were carried out by mean standardized (RMSS) must be close to 1 [22].
the provincial offices of the Ministry of Environment and While implementing the models, the anisotropy effect
Urbanization. Surface soil sampling was conducted at 46 was surveyed.
points. The concentration of radioactivity in the surface
soil samples was determined by the Atomic Research Results and discussion
Council of Turkey [19]. Anisotropic variogram models were preferred. The
226
Ra, 232Th and 40 K concentration values showed a
Risk assessment directional change. The spatial dependencies (Nugget/Sill
The average natural radioactivity rate, the concentration ratio) were found to be related to the degree of auto-
range and the average values are presented in Table 1 correlation between the sampling points. If the spatial
[3]. To determinate the concentration of the natural ra- dependence was higher between the sample points, the
dionuclides, soil samples from 10 cm thickness in 1 km2 spatial correlation was very high. The spatially dependent
land were collected in the research area. variables were classified as: strongly spatially dependent if
the ratio was 25%, mid-spatial-dependent if the ratio was
Geostatistical analysis 25% - 75% and weakly spatially dependent if the ration
To conduct the geostatistical analysis, the "Kriging" was 75% [4,23-26].
interpolation technique was used within the spatial Because the spatial dependence was strong, the vari-
analyst extension module in ArcGIS 9.3 software. The ables did not differ over short distances. In this research
spatial analyses were carried out with prepared maps area, spatial dependence was too high for 232Th (16.91%).
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 3 of 11
http://www.ijehse.com/content/12/1/130

Figure 1 Location of study area.

Spatial dependence was identified as normal for 226Ra and to be closer to a normal distribution. After the log
40
K (25.88% and 70.60%), respectively. An effective spatial transformation was conducted, the 226Ra concentrations
dependence distance was found between 254369.7 and were found to be approximately normally distributed.
375276 meters (Table 2). The histogram of 226Ra and log transformations data is
The sample point data will involve converting the presented in Figure 2. To check the data, the best model
spatial data of the 226Ra, 232Th and 40 K concentrations was applied to the cross-validation of the spatial correl-
used in the interpolation for kriging. The lowest error ation of the 226Ra concentration of the study area. A
rate models were chosen; they were the Exponential comparison of the ME and the RMSSE for the log 226Ra
and Stable models. The maps were produced and illustrates that the exponential model and its parameters
field data were obtained in accordance with this kriging were best for the 226Ra concentration. The exponential
model. model has the best fit with the nugget effect (Co); it is
equal to 48.35, a sill (Co + C) equal to 186.83 and a range
Spatial distribution of 226Ra: of influence equal to 254369.7. The ratio of the nugget
The prediction map, according to the optimized model, variance to the sill is expressed in percentages equal to
was determined during the cross-validation process. The 25.88% (Table 2). This value is greater than 25% and less
226
Ra concentration prediction map shows the log 226Ra. than 75%. Thus, the 226Ra distribution has a moderate
The dataset for the 226Ra concentration has a high spatial dependence in the study area. A spatial prediction
kurtosis and is positively skewed, so it is not a normal and distribution map for the ordinary kriging interpolation
distribution. The data log transformation was applied of 226Ra is presented in Figure 3.

Table 1 Natural radioactivity the concentration range and Spatial distribution of 232Th
average values The dataset of the 232Th concentration has a high kurtosis
Natural radionuclides (Bq/kg) and is positively skewed, so it is not a normal distribution.
226
Ra 232
Th 40
K The data log transformation was applied, so could be
Concentration range 17-60 11-64 140-850
closer to a normal distribution. After the log trans-
formation was conducted, the 232Th concentrations
Average values 35 45 400
were approximately normally distributed. The histogram
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 4 of 11
http://www.ijehse.com/content/12/1/130

Table 2 Model parameters

Model Regression function Nugget, (Co) Range, A Sill, (Co + C) Nugget/Sill, (%) ME1 RMSSE2
226
Ra Exponential 0.2695 x +18.32 48.35 254369.7 186.83 25.88 0.004 1.06
232
Th Exponential 0.5063 x +15.33 50.74 304696.1 300.03 16.91 0.008 0.89
40
K Stable 0.1596 x +381.99 41767.52 375276.0 59158.77 70.60 0.007 0.997
ME1: mean standard error.
RMSSE2: estimated standardized mean of error of mean square root.

of 232Th and log-histogram transformation data is pre- distribution map for the ordinary kriging interpolation
232
sented in Figure 4. To check that the best model was Th is presented in Figure 5.
applied, a cross-validation of the spatial correlation of the
232
Th concentration of the study area was conducted. A Spatial distribution of 40 K
comparison of ME and the RMSSE for log 232Th shows An optimized model determined from the cross-validation
that the exponential model and its parameters are the process and the 40 K concentration prediction map shows
more appropriate for the 232Th concentration. The expo- the log 40 K. The dataset for the 40 K concentration has a
nential model has the best fit, with the nugget effect (Co) high kurtosis and is positively skewed, so it is not a normal
being equal to 50.74, a sill (Co + C) equal to 300.03 and distribution. The data log transformation was applied, so
a range of influence equal to 304696.1. The ratio of the the data would be closer to a normal distribution. After the
nugget variance to the sill expressed in percentages is log transformation, the 40 K concentrations were found
equal to 16.91% (Table 2). This value is smaller than to be approximately normally distributed. The histogram
25; thus, the 232Th distribution has a powerful spatial of 40 K and the log transformation data is presented in
dependence in the study area. A spatial prediction and Figure 6.

Figure 2 Frequency distribution of the 226Ra and Log 226Ra.

Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 5 of 11
http://www.ijehse.com/content/12/1/130

226
Figure 3 Spatial prediction map for the ordinary kriging interpolation of Ra.

Figure 4 Frequency distribution of the 232Th and Log 232Th.

Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 6 of 11
http://www.ijehse.com/content/12/1/130

232
Figure 5 Spatial prediction map for the ordinary kriging interpolation of Th.

Figure 6 Frequency distribution of the 40 K and Log 40 K.

Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 7 of 11
http://www.ijehse.com/content/12/1/130

A comparison of the ME and the RMSSE for log 40 K forest ecosystems effect the spatial variation of radionu-
shows that the stable model and its parameters is the clides in the soil.
best model that illustrates for 40 K concentration. The
stable model has the best fit with the nugget effect (Co) Spatial changes of great soil groups and radionuclides in
being equal to 41767.52, a sill (Co + C) equal to 59158.77 the soil
and a range of influence equal to 375276. The ratio of The soil groups map of the study area is presented in
nugget variance to sill is expressed in percentages equal Figure 11. It was determined that 43.17% of the concen-
to 70.60% (Table 2). This value is greater than 25 and tration of 40 K, 28.16% of the concentration of 226Ra and
less than 75%; thus, the 40 K distribution has a moderate 26.67% of the concentration of 232Th was above the
spatial dependence in the study area. The spatial prediction average UNSCEAR (2000) concentrations.
and distribution map for the ordinary kriging interpolation Within the study area, 17.49% of the Basaltic soils and
40
K is presented in Figure 7. 11.51% of the Chestnut soils had above average concen-
trations of the 40 K radionuclides. In addition, 10.8% of
The relationship between forest cover and natural the Basaltic soils and 9.59% of the Chestnut soils had
radionuclides above average concentrations of 226Ra radionuclides. It
In the study area, the spatial distribution was analyzed in was also determined that 10.47% of the Basaltic soils and
relation to the forest ecosystems and the radionuclides 9.32% of the Chestnut soils had above average concen-
(Figures 8, 9 and 10). Between the 226Ra and 232Th, a posi- trations of 232Th radionuclides (Table 4). Consequently, all
tive increase in the 0.01 significance level (0.865**) was de- basalt and chestnut soil field locations had above average
tected. Between the 40 K and 232Th, a positive increase in radionuclide concentrations.
the 0.05 significance level (0.718*) was detected. Between Brown soils, High Mountain soils, Reddish Brown soils
the forest cover and 226Ra, a negative relationship was and Reddish Yellow Podsol soils in areas containing high
identified. Between the 232Th and 40 K, a positive relation- concentrations of 226Ra and 232Th radionuclides had no
ship was identified (Table 3). The radionuclide concentra- spatial distribution pattern. In areas with Brown soils
tions were found to depend on the soil humus content and Reddish Brown soils, high concentrations of 40 K ra-
[27]. Different contents and amounts of organic matter in dionuclides had no spatial distribution pattern (Table 4).

40
Figure 7 Spatial prediction map for the ordinary kriging interpolation of K.
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 8 of 11
http://www.ijehse.com/content/12/1/130

Figure 8 226Ra and spatial distribution of forest cover.

Figure 9 232Th and spatial distribution of forest cover.

Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 9 of 11
http://www.ijehse.com/content/12/1/130

Figure 10 40 K and spatial distribution of forest cover.

In summary, changes in the concentrations of radio- distribution values can vary greatly according to chan-
nuclides in the soil depend on the formation of iron ging areas [31,32].
oxide and other elements. Some of the acids formed
in the soil, through calcium carbonate found in the
environment, prevent the retention of the radionu- Conclusions
clides. Therefore, radionuclide concentrations in the Over time, the infiltration of radionuclides has resulted in
rocks can be reduced with calcium carbonate; this, in high radionuclide concentrations in the lower soil layers.
turn, reduces the level of external radiation [28,29]. The radionuclides in these lower soil layers can move
According to the Anonymous [30], some rocks and a upwards through the roots of plants and be transferred to
soil type typical of the specific radioactivity values the plant during the growth process. Since radionuclides
was identified in the follow: For 40 K; Granite (1000 Bq/kg), can have detrimental health effects on humans, it is
clay stone (700 Bq/kg), Sandstone (350 Bq/kg) Basalt important to determine the spatial variation of concen-
(250 Bq/kg) and limestone (90 Bq/kg). For 232Th; Granite trations of radioactive elements.
(80 Bq/kg), clay stone (50 Bq/kg), Sandstone (10 Bq/kg), This research was conducted to examine to spatial
Basalt (10 Bq/kg) and Limestone (7 Bq/kg). Local distribution of natural radioactive element (226Ra, 232Th
and 40 K) concentrations and their relationship with soil
groups and forest cover using the Kriging method.
Table 3 Correlations test result between radionuclides
and forest cover
According to the statistical analyses, positive correlations
were detected between the 226Ra and the 232Th (0.865**),
Parameters Radionuclides Forest cover
226 232 40
as well as between the 40 K and 232Th (0.718*). Negative
Ra Th K
226
correlations between forest cover and 226Ra were found,
Ra 1 while positive correlations between 232Th and 40 K were
232
Th 0,865** 1 detected.
40
K 0,592 0,718* 1 The basalt and chestnut soils in the study area
Forest cover 0,008 0,096 0,232 1 were found to have above average concentrations of
**Correlation is significant at the 0.01 level (2-tailed).
radionuclides. The Brown soils, High Mountain soils,
*Correlation is significant at the 0.05 level (2-tailed). Reddish Brown soils and Reddish Yellow Podsol soils
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 10 of 11
http://www.ijehse.com/content/12/1/130

Figure 11 Map of great soil groups.

did not have high concentrations of 226Ra and 232Th. basic food chain. Excessive exposure to radiation can lead
The Brown soils and the Reddish Brown soils also to cancer; it is also the cause of hereditary diseases. There-
did not have high concentrations of 40 K. fore, spatial variations of radioactive element concentra-
Radionuclides are present in different concentrations tions need to be monitoring for the sustainability of a
in the soil, plants and water, which comprise parts of the healthy environment.

Table 4 Spatial changes between radionuclides and great soil groups

Spatial changes of the radionuclides concentration (Hectare)
40 226 232
Great soil goups K Ra Th
Under optimal Upper optimal Under optimal Upper optimal Under optimal Upper optimal
concentrations concentrations concentrations concentrations concentrations concentrations
(<400 Bg/kg) (>400 Bg/kg) (<35 Bg/kg) (>35 Bg/kg) (<45 Bg/kg) (>45 Bg/kg)
Bazaltic Soils 1187656 1717242 1915127 989771 1876826 1028072
Chesnut Soils 1101871 1129844 1290559 941156 1316923 914792
Limeless Brownish Soils 428955 582383 620178 391160 762213 249125
Limeless Brown Forest Soils 94789 228346 119664 203471 135446 187689
Vertisol Soils 0 76254 0 76254 0 76254
Brown Forest Soils 464221 287303 684116 67408 684116 67408
Brown Soils 1505651 0 1505651 0 1505651 0
High Mountain Soils 262129 159430 421559 0 421559 0
Reddish Yellow Podsol Soils 438377 56680 495057 0 495057 0
Reddish Brown Soils 94684 0 0 94684 0 94684
Dindarolu Journal of Environmental Health Science & Engineering 2014, 12:130 Page 11 of 11
http://www.ijehse.com/content/12/1/130

Competing interests 21. ESR: The principels of geostatistical analysis. 2013, 54. http://maps.
The author declares that he has no competing interests. unomaha.edu/Peterson/gisII/ESRImanuals/Ch3_Principles.pdf 22.02.2014.
22. Johnston K, Hoef M, Krivoruchko K, Lucas N: Using ArcGIS Geostatistical
Analyst. New York: ESRI; 2001.
Authors contribution
23. Clark I: Practical Geostatistics. London: Applied Science Publishers Ltd.;
TD participated in the design of the study and has collected the data and
1979:129.
drafted the manuscript used the GIS and makes statistical analysis related
24. Trangmar BB, Yost RS, Uehara G: Application of geostatic to spatial studies
data with soil groups and forest area.
of soil properties. Advances in Argon 1985, 38:4594.
25. Cambardella CA, Moorman TB, Novak JM, Parkin TB, Karlen DL, Turco RF,
Acknowledgment Konopka AE: Field-scale variability soil properties in Central Iowa soils.
Authors present their great thanks to Turkey Atomic Energy Agency for Soil Sci Soc Am J 1994, 58:15011511.
data supply. 26. Erahin S: Alluvial soil in a field, some physical and chemical properties
of the spatial variability of the determination. SU Journal of the Faculty of
Received: 27 March 2014 Accepted: 16 October 2014 Agriculture 1999, 13:3441.
27. Chelmicki V, Mietelski JW, Macharski P, Swicchowicz J: Natural Factors
of Cs-137 Rcdistribution in Soil. Poland: Institute of Nuclear Physics
Cracow, Report No. 1615lD; 1993:19.
References 28. NCRP: National Council on Radiation Protection and Measurements, No.45,
1. Theodorsson P: Measurements of Weak Radioactivity. Singapore: World Natural Background Radiation in the U.S., Soil Radioactivity. 1975:54.
Scientific Pub. Co; 1997. 29. zger AG: District of Ceyhan, Yumurtalk and Pozant Determination of
2. TAEK: Resource of natural radiation. http://www.taek.gov.tr/bilgi-kosesi/ Natural Radioactivity Levels. In M.Sc. Thesis. ukurova University: Graduate
184-radyasyon-insan-ve-cevre/radyasyonla-birlikte-yasiyoruz/501-dogal- School of Science; 2005.
radyasyon-kaynaklari.html Accessed on: 11.03.2014. 30. Anonymous: Federal Ministry for the Environment, Nature Conservation and
3. UNSCEAR: Report on sources and effects of ionizing radiation to the Nuclear Safety, environmental radioactivity and radiation exposure, Annual
general assembly, United Nations, Vienna. 2000, http://www.unscear.org/ Report, Bonn. ; 1995:4345.
unscear/en/publications/2000_1.html Accessed on: 25.02.2014. 31. Atakan Y: Natural radioactivity, natural and human created in the radiation
4. Robertson GP: Geostatistics in ecology: interpolation with known doses, International Symposium on Medical Geology Participatory Books.
variance. Ecology 1987, 68:744748. 2008:5163.
5. Cressie NAC: Statistics for Spatial Data. Canada: John Wiley and Sons, Inc; 32. Vukainovi I, orevi A, Rajkovi M, Todorovi D, Pavlovi V: Distribution
1993:900. of natural radionuclides in anthrosol-type soil. Turk J Agric 2010,
6. Goovaerts P: Geostatistics for Natural Resources Evaluation. New York: Oxford 34:539546.
University Press; 1997:483.
7. Ecker MD: Geostatistics: Past, Present and Future, Encyclopedia of Life Support
doi:10.1186/s40201-014-0130-6
Systems (EOLSS), Developed under the Auspices of the UNESCO. Oxford, UK:
Cite this article as: Dindarolu: The use of the GIS Kriging technique to
Eolss Publishers; 2004. determine the spatial changes of natural radionuclide concentrations in
8. Isaaks EH, Srivastava RM: An Introduction to Applied Geostatistics. New York: soil and forest cover. Journal of Environmental Health Science & Engineering
Oxford University Press; 1989. 2014 12:130.
9. Brown RB, Huddleston JH: Presentation of statistics data on Map units to
the user. In spatial variabilities of soil and landform. SSSA Special 1991,
28:127147.
10. Khan HM, Khan K, Atta MA, Jan F: Measurement of gamma activity of soil
samples of Charsadda district of Pakistan. J Chem Soc Pakistan 1994,
16:183188.
11. Rhm W, Kammerer L, Hiersche L, Wirth E: Migration of 137Cs and 134Cs
in different forest soil layers. J Environ Radioact 1996, 33:6375.
12. Rosn K, born I, Lnsj H: Migration of radiocaesium in Swedish soil
profiles after the Chernobyl accident, 19871995. J Environ Radioact 1999,
46:4566.
13. Bergman R: The distribution of radioactive caesium in boreal forest
ecosystems. In Nordic Radioecology, the Transfer of Radionuclides
through Nordic Ecosystems to Man. Studies in Environmental Science 1994,
62:335379. (ed. H. Dahlgaard) Elsevier, New York, NY.
14. Gorham E: A comparison of natural and fallout radioactivity in Ontario
soils under pine. Can J Bot 1963, 4:1130911319. USA.
15. Degering D, Schlenker S, Unterricker S: Radionuclide Behaviour in Natural
Organic Matter (Peat, Cola and Forest Soil Surfaces), NRC 5, 5th International
Conference on Nuclear and Radiochemistry. Pontresina, Switzerland: Paul
Scherrer Institut; 2000:449452.
16. Sokolik GA, Ivanova TG, Leinova SL, Ovsiannikova SV, Kimlenko IM: Submit your next manuscript to BioMed Central
Migration ability of radionuclides in soil-vegetative cover of Belarus and take full advantage of:
after Chernobyl accident. Environ Int 2001, 26:183187.
17. UNSCEAR: Sources and Effects of Ionizing Radiation, Report of the General
Convenient online submission
Assembly With Scientific Annex B. New York: United Nations Publicaitons;
1993. Thorough peer review
18. Hlgye Z, Mal M: Sources, vertical distribution, and migration rates of No space constraints or color gure charges
239, 240
Pu, 238Pu, and 137Cs in grassland soil in three localities of central
Immediate publication on acceptance
Bohemia. J Environ Radioact 2000, 47:135147.
19. TAEK: Turkey Atomic Energy Agency: Turkey Atlas of Environmental Inclusion in PubMed, CAS, Scopus and Google Scholar
Radioactivity. Ankara: 2014. http://www.taek.gov.tr/radyasyon-izleme/turkiye- Research which is freely available for redistribution
cevresel-radyasyon-atlasi.html Accessed on: 26.03.2014.
20. OGM: Forest Cover Map, General Directorate of Forestry. Ankara: OGM
Publications; 2012. Submit your manuscript at
www.biomedcentral.com/submit