Research ArticleUsing Multivariate Geostatistics to Assess Patterns ofSpatial Dependence of Apparent Soil Electrical Conductivity andSelected Soil Properties

Gleacutecio Machado Siqueira1 Jorge Dafonte Dafonte2

Montserrat Valcaacutercel Armesto2 and Ecircnio Farias Franccedila e Silva3

1Centro de Ciencias Agrarias e Ambientais Universidade Federal do Maranhao BR-222 Km 04 snBoa Vista 65500-000 Chapadinha MA Brazil2Departamento de Ingenierıa Agroforestal Escuela Politecnica Superior Universidad de Santiago de CompostelaCampus Universitario sn 27002 Lugo Spain3Departamento de Engenharia Agrıcola Universidade Federal Rural de Pernambuco Dois Irmaos 52171-900 Recife PE Brazil

Correspondence should be addressed to Glecio Machado Siqueira gleciosiqueirahotmailcom

Received 24 June 2014 Revised 26 August 2014 Accepted 28 August 2014 Published 31 December 2014

Academic Editor Ana Maria Tarquis

Copyright copy 2014 Glecio Machado Siqueira et al This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

The apparent soil electrical conductivity (ECa) was continuously recorded in three successive dates using electromagnetic inductionin horizontal (ECa-H) and vertical (ECa-V) dipole modes at a 6 ha plot located in Northwestern Spain One of the ECa data setswas used to devise an optimized sampling scheme consisting of 40 points Soil was sampled at the 00ndash03m depth in these 40points and analyzed for sand silt and clay content gravimetric water content and electrical conductivity of saturated soil pasteCoefficients of correlation between ECa and gravimetric soil water content (0685 for ECa-V and 0649 for ECa-H) were higher thanthose between ECa and clay content (ranging from 0197 to 0495 when different ECa recording dates were taken into account)Ordinary and universal kriging have been used to assess the patterns of spatial variability of the ECa data sets recorded at successivedates and the analyzed soil properties Ordinary and universal cokriging methods have improved the estimation of gravimetric soilwater content using the data of ECa as secondary variable with respect to the use of ordinary kriging

1 Introduction

The quality of soil data collection for precision agriculturehas a very important influence since it has been found thatacquisition of exhaustive information in this phase supportsthe use of geospatial technologies for the estimation of soilspatial variability and later on assists in the determination ofldquomanagement unitsrdquo However for assessing the soil spatialvariability a large number of samples are generally neededwhich considerably increases costs of sampling and analysisNotwithstanding the sampling process can be improvedusing soil variables that can be recorded or measured quicklywhich can help in enhancing the estimation of other soilproperties more difficult to measure

The measurement of apparent soil electrical conductivity(ECa) allows the collection of information on the field and onthe spatial distribution of other properties that are correlatedIn accordance with Corwin and Rhoades [1] the mainmethods for the measurement of soil ECa are contact andelectromagnetic induction McNeill [2] Sudduth et al [3]Corwin and Lesch [4] and Kuhn et al [5] indicate that ECais mainly influenced by soil water content texture organicmatter content size and distribution of pores salinity cationexchange capacity concentration of electrolytes dissolved inthe soil solution temperature composition of soil colloidsand so on Thus the use of ECa for soil classificationallows recognition and delimitation of the physical chemicaland biological soil properties that play an important role

Hindawi Publishing Corporatione Scientific World JournalVolume 2014 Article ID 712403 11 pageshttpdxdoiorg1011552014712403

2 The Scientific World Journal

Castro de Ribeiras de LeaLugo (Galicia Spain)

(a)

622550 622650 622750

4779650

4779700

4779750

4779800

4779850

4779900

406

409

412

Dist

ance

Y(m

)

Distance X (m)

N

W

S

E

(b)

Figure 1 Geographical location of the study area (a) Field digital elevation model (b)

in agricultural production and environmental conservationThus these data are essential for monitoring the temporalcondition of the soil and application management processes[6] Therefore the ECa parameter is used as an aid inprecision agriculture to promote the evaluation of the spatialvariability of soil and the definition of management units

The use of geostatistics has great advantages because itallows the study of the spatial variability of soil propertiesKriging is a geostatistical method that can be used topredict the value of soil properties in unsampled locationsfavouring the application of differentiated soil managementin precision agriculture Several authors have devised soilsampling schemes directed by properties that directly orindirectly influence crop yield [7ndash10] and the success of thisapproach depends on the use of variables that are quickly andeasily measured such as ECa

Based on the above rationale the objectives of this workwere as follows (1) to analyze the spatial dependence ofsuccessive ECa data sets (2) to assess the spatial variability ofsoil texture attributes using a soil sampling scheme directedby soil ECa and (3) to improve the estimation of thespatial variability of soil variables such as soil water contentthrough multivariate geostatistics using ECa as secondaryinformation

2 Material and Methods

21 Study Site The experimental field is 6 ha in surface andit is located in Castro Ribeiras de Lea (Lugo NW Spain)Geographic coordinates are 43∘0910158404910158401015840N and 7∘2910158404710158401015840Waverage elevation is 410m and mean slope is 2 (Figure 1)

The area where the field is located is considered to be rep-resentative of both the topographic patterns and themain soiltype of the region ldquoTerra Chardquo which is characterized by anextensive livestock production on a landscape with seasonalconditions of hydromorphy due to impeded drainage

The crop succession of the experimental site was fallow-silage corn (Zea mays L) under no-till farming during the

Table 1 Soil texture data for a representative profile of the studyarea

Horizon Depth (m)Organicmatter Clay Silt Sand Gravel

g dmminus3 g kgminus1

Ap 00ndash035 5050 175 191 634 370Bw 035ndash070 720 192 207 591 448Btg gt070 260 479 280 241 mdash

study year Previously this site had been under pasture forsilage production Field data recording and soil samplingwere performed in spring 2008

The soil was classified as aGleyic Cambisol [11] and it wasdeveloped over Tertiary-Quaternary sediments the parentmaterial from the Quaternary has high gravel content and itis underlain by clayey Tertiary sediments with low saturatedhydraulic conductivity [12] According to Neira Seijo [13] thesoil profile of the studied field is represented by the sequenceAp-Bw-Btg developed on successive sedimentary layers withheterogeneous soil particle size distribution (Table 1) Thesoil texture of the fine earth (lt200mm) was sandy-loam atthe Ap horizon sandy-clay-loam in Bw horizon and clayeyin the horizon Btg and there was a general clay increasewith soil depth Moreover the Ap and Bw horizons werecharacterized by a high content of gravel attaining 37 and45 respectively The organic matter content was ratherhigh on the Ap horizon (505) contrasting with the lowercontents at the underlying horizons of the soil profile Theclimate of the Terra Cha region is classified as maritimetemperate climate (Cfb according to Koppen) characterizedby warm summers and no dry season average annual rainfallis as high as 930mm

22 Apparent Soil Electrical Conductivity Measurements andSampling Scheme Apparent soil electrical conductivity (ECa)was measured using electromagnetic induction equipment

The Scientific World Journal 3

N

EW

S

622500 622600 622700 622800

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

37

38

36

35

34

32

33

31

23

3024

25

2629

40

22

27

28

21

20

1918

8

6

14

9

39

7

1

2

3

4

5

11

12

1315

16

17

10

(a) (b)

Figure 2 Scheme showing apparent electrical conductivity (ECa) continuously recorded (line) and the location of 40 soil sampling points(circles) on 2362008 (a) and cart containing the EM38-DD equipment and GPS (b)

N

EW

S622550 622650 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

(a)

N

EW

S

622550 622650 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

(b)

Figure 3 Scheme showing apparent electrical conductivity (ECa) recorded during 1432008 (a) and 342008 (b)

EM38-DD [14] This device consists of two integrated EM38units oriented in the horizontal and vertical dipole positionsproviding simultaneous measurements for the two dipolesmodes in the vertical dipole mode the primary magneticfield is thought to effectively penetrate to a depth of about15m while in the horizontal dipole position EM38-DD isthought to be effective for a shallower soil depth estimatedat about 075m [14]

To complete continuous record of the apparent soilelectrical conductivity in horizontal dipole (ECa-H mSmminus1)and in vertical dipole (ECa-V mSmminus1) (Figures 2(a) and3) the EM38-DD was installed in a car built with plastic

materials (Figure 2(b)) In addition GPS RTK was used forgeoreferencing the recorded measures

The reference measurements of ECa-H and ECa-V wereperformed on 2362008 at 1859 sampling points followingthe scheme presented in Figure 2(a) The soil samplingscheme was devised using the software tool ESAP-RSSD(response surface sampling design) based on a multiplelinear regression model [7 9]This software aims to optimizethe position of new sampling points considering apparentsoil electrical conductivity (ECa) measured with horizontal(ECa-H) and vertical dipoles (ECa-V) (Figure 2(a)) Theoptimized soil sampling scheme consisted of 40 points In

4 The Scientific World Journal

Table 2 Statistical parameters of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Unit 119873 Min Max Mean plusmn SD Variance CV Skew Kurt 119863

1432008 ECa-V mSmminus1 1887 575 1838 1048 plusmn 119 142 1135 0527 0124 0045Ln1432008 ECa-H mSmminus1 1887 925 1900 141 plusmn 077 060 546 0065 1810 0040Ln342008 ECa-V mSmminus1 1871 963 2050 1404 plusmn 215 464 1531 0662 0083 0073Ln342008 ECa-H mSmminus1 1871 663 1950 1459 plusmn 077 060 528 0160 1051 0095Ln2362008 ECa-V mSmminus1 1886 413 2013 1121 plusmn 247lowast 612 2203 0485 minus0243 0071Ln2362008 ECa-H mSmminus1 1886 663 2000 1212 plusmn 179lowast 322 1477 0839 1285 0092Ln2362008 CEe mSmminus1 40 700 2800 1382 plusmn 509 2594 3683 1200 1008 0159n2362008 Clay g kgminus1 40 11900 22000 16837 plusmn 3092 95654 1836 minus0190 minus1321 0153n2362008 Silt g kgminus1 40 23300 35700 29625 plusmn 3506 122973 1183 0149 minus1041 0098n2362008 Sand g kgminus1 40 48700 58600 53537 plusmn 2253 50772 421 0055 minus0270 0069n2362008 120579119892 38 1341 4567 2674 plusmn 681 4950 2547 minus2904 6510 0085n119873 number of measurements Min minimum value Max maximum value Mean plusmn SD mean plusmn standard deviation CV coefficient of variation () Skewskewness Kurt kurtosis and119863 normality of the data for test of Kolmogorov-Smirnov (119875 lt 001 n normality and Ln log normality) lowastNonsignificant at 5level of ANOVA (SNK)

addition continuous measurements of ECa-H and ECa-Vwere previously taken in the experimental field on 1432008and 342008 as shown in Figure 3 Note however that theschemes of the continuously recorded ECa data sets taken inthe three successive dates were different as shown in Figures2(a) 3(a) and 3(b)

In the 40 points selected during the ECa campaign of2362008 soil samples were taken at the 00ndash03m depthwith a manual soil probe Soil texture soil water contentand electrical conductivity of saturated paste extracts weredetermined using standard methods Soil texture (clay siltand sand in g kgminus1) was determined by the sieve-pipettemethod following Camargo et al [15] in this method a mix-ture of sodium hydroxide and sodium hexametaphosphatewas used as chemical dispersant The gravimetric soil watercontent (120579119892 ) was obtained after weighing the mass ofthe wet and dry sample according to Camargo et al [15]To determine the electrical conductivity of soil saturatedextracts (ECe) a mixture of soil and distilled water of 1 1 wasprepared as proposed by USDA [16] electrical conductivitymeasurements were performed using a conductivity meterORIONModel 122

23 Statistical and Geostatistical Analysis All the values werestatistically analyzed using SPSS package 115 at 5 level ofSNK (Student-Newman-Keuls) method ANOVA The testof normality Kolmogorov-Smirnov was used to test thenormality of data with probability of error 1 (119875 lt 001)Thecorrelation was calculated with the correlation coefficient ofPearson

The analysis of the spatial variability of soil physicalproperties was conducted using the experimental variogramthe fitting of variogram model was performed using themethod described by Vieira [17] based on cross-validationInitial analysis showed that the variogram of any studiedproperties showed a trend so the universal kriging was usedin these cases in which the residual variogram is required[18] For those variables that showed no trend ordinary

kriging was used The degree of spatial dependence (SD) wasdetermined according to the following

SD () =1198620

1198620+ 1198621

lowast 100 (1)

where 1198620is nugget effect and (119862

0+ 1198621) is the sill (119862

0+ 1198621)

according to Cambardella et al [19] which is considered ashigh (SD le 25) moderate (SD = 25ndash75) and low (SD ge75)

Cross-variogramwas used to study the spatial correlationbetween soil variables when there was a trend in some ofthese variables universal cokriging was used [18] instead ofordinary cokriging The software used to perform ordinarykriging universal kriging and universal cokriging was Gstat[20] In cokriging the covariance matrix must be positive anddefinite [18 21ndash23] The use of cokriging was used only for acouple of attributes that showed correlation coefficient values(|119903|) greater than 05

3 Results and Discussion

Statistical analysis of the data (Table 2) indicates that thereis great variation between samples in accordance with low(CV le 12) and middle (CV = 12ndash60) variation coefficientvalues by classification of Warrick and Nielsen [24] It isverified that the apparent electrical conductivity of the soil(ECa) measurement with the horizontal dipole (ECa-H) haslower CV than the measurements with the vertical dipole(ECa-V)This fact can be explained because the vertical dipolemode explores a larger volume of soil than the horizontaldipole and there is greater heterogeneity in those variablesthat affect the ECa values mainly clay content organicmaterial water content in the soil porosity salinity and soforth [2ndash5]

Only data ECa-V and ECa-H sampling in 2362008did not show differentiation by the average test (ANOVA)between the different sampling dates

The Scientific World Journal 5

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)Log ECa-V2368Log ECa-V1438Log ECa-V348

(a)

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)

Log ECa-H 348

Log ECa-H 2368Log ECa-H 1438

(b)

Figure 4 Standardized sample semivariogram for log ECa-V (a) and log ECa-H (b) recorded during three successive dates

Table 3 Precipitation and reference evapotranspiration betweensuccessive dates in which apparent soil electrical conductivity (CEa-V and CEa-H) was recorded

Period Precipitation(mm)

Referenceevapotranspiration (mm)

1522008ndash1432008 526 3801432008ndash342008 804 371342008ndash2362008 3978 2146

The values of the electrical conductivity of the saturationpaste extract of the soil (ECe) are higher than the values ofECa-V and ECa-H this fact is because ECe is a parameterthat depends on the content of anions and cations in the soilsolution the water content is homogeneous in all samplesbecause the sample is saturatedwithwater and the soil appar-ent electric conductivity valuesmeasuredwith the equipmentEM38-DD (ECa-V and ECa-H) are very influenced by the soilwater content [2 3 5] and the water content in the soil isvariable along the field

ECa-V and ECa-H measured on several sampling dates(1432008 342008 and 2362008) showed lognormal dis-tribution (Table 2) Other attributes studied showed normalfrequency distribution (ECe clay silt sand and soil watercontent)

In the geostatistical analysis lognormal transformationwas used for properties that showed lognormal distributionThe highest values of coefficient of correlation betweenECa variables and clay and silt content are on the firstmeasurement date (1432008) on this date the soil moistureis lower coincided with data of precipitation and evapotran-spiration (Table 3) Grandjean et al [25] describe that soilwith moisture lower is ideal for characterization of soil bulkdensity and of soil texture using measurements of electricalconductivity

The coefficient of correlation values between the apparentsoil electrical conductivity (ECa-V and ECa-H) measure-ment on several sampling dates (1432008 342008 and2362008) presented moderate positive correlation coeffi-cient values (05 le 119903 lt 08)

The coefficient of correlation between ECa-V2362008 times120579119892 (119903 = 0685) and ECa-H2362008 times 120579119892 (119903 = 0648)was moderately positive (05 le 119903 lt 08) confirming thecorrelation between the values of ECa-V and ECa-H andthe water content in soil because according to Grandjeanet al [25] under wet conditions electrical conductivitymeasurements are dominated by the effect of water contentwhich tends to hide the influence of the other factors

The values of log ECa-V are affected by the groundwaterlevel so the variogram follows the trend in the ground waterlevel (Figure 4) As can be seen on standardized variogramswith the value of the sample variance data from ECa-Vmeasurement on 1432008 and 342008 present a trendfollowing the same pattern of the digital elevation map ofthe area (Figure 1(b)) Analyzing standardized variogramsfor log ECa-H data can be seen that only shows trend forthe variogram of 1432008 and 342008 but not for thevariogram of 2362008 on this date the water table wasprobably located below the depth of soil investigated with thehorizontal dipole mode

Corwin and Lesch [4] found higher values of correlationbetween data from log ECa-V and log ECa-H and electricalconductivity of the saturation extract (ECe) and clay contentbut lower than those found for log ECa-V and log ECa-Hand the water content in soil Martınez and Vanderlinden[26] described a higher correlation between ECa and watercontent in loamy soils while in clay soils the correlation waslower The correlation coefficients for ECe silt and clay withthe log ECa-H are greater than with the log-ECa-V

In order to improve the correlation between the values ofECa-V andECa-Hwith clay content soil water content shouldbe as homogeneous as possible within the study area better if

6 The Scientific World Journal

its value is closer to field capacity and unlike the water tableis as low as possible so the best time to take measurementsunder these conditions would be in the autumn when therewas heavy rainfall although under these conditions the watertable probably would not have ascended enough to be close tothe surface

The initial geostatistical data analysis showed that thephysical properties of the soil (clay silt sand and gravimetricwater content) showed no trend then being possible theestimate of the variable using the original data with ordinarykriging Moreover ECe data and apparent electrical conduc-tivity of the soil (ECa-V and ECa-H) on several samplingdates show trend in Figure 3 the semivariance value is notstabilized around variance value of data and the universalkriging was used to construct the maps of spatial variabilityof these variables

The fitted variogram parameters (Table 5) show that thespherical model was the fitted model to the properties understudy according to Cambardella et al [19] Goovaerts [18]Vieira [17] and Siqueira et al [27] describing this modelit is usually best fitted to the properties of soil and plantAll attributes had low values of nugget effect (119862

0) Range

values (a) varied from 4000m (log ECa-H Residual342008

)to 13000m (clay silt and soil water content) The degreeof spatial dependence between samples was high (SD le2500) across the study the exception being log ECa-H Residual

1432008which presented a moderate degree of

spatial dependence (SD = 3167)The spatial variability maps obtained with universal

kriging (Figure 5) show that there is a similarity between themaps ECa-V (Figures 5(a) 5(c) and 5(e)) and ECa-H (Figures5(b) 5(d) and 5(f)) on several sampling dates with furtherdifferentiation of maps of ECa-V and ECa-H on themeasureddata on 2362008 (Figures 5(e) and 5(f))when thewater tablelevel was lower

It is observed that the maps of ECa-V and ECa-H(Figure 5) and the map of the water content in soil (Fig-ure 6(e)) obtained with ordinary kriging (Figure 6(e)) looksimilar following the same pattern of digital elevation model(Figure 1(b))

The map of the electrical conductivity of the saturationextract (ECe Figure 6(a)) shows inverse behaviour to mapsECa-V and ECa-H (Figure 5) Moreover the map of spatialvariability of clay content in the study area (Figure 6(b))shows no similarity to maps ECa-V and ECa-H (Figure 5)this fact is also repeated with silt (Figure 6(c)) and sand(Figure 6(d))

Table 6 presents the fitting parameters cross-variogrambetween 120579119892 times ECa-V (119903 = 0685) and 120579119892 times ECa-H (119903 = 0648)The cross-variograms were fitted to a spherical model withthe same range compared to single variograms to obtain alinear model coregionalization (Table 6)

The spatial variability maps constructed using ordinaryand universal cokriging (Figure 7) demonstrate that theuse of the soil apparent electrical conductivity measured byelectromagnetic induction (ECa-V and ECa-H) on 2362008was a secondary variable that improves the estimation of thesoil water content using cokriging This improvement in theestimation of 120579119892 can be observed in Table 7 where it showed

an increase in the value of the correlation coefficient betweenthe measured and the estimated values from cross-validationusing ordinary cokriging with log-ECa-V (0746) and withlog-ECa-H (0756) as secondary variables with respect to useof ordinary kriging (0637) Moreover in the case of the soilwater content map obtained with ordinary cokriging using assecondary data ECa-H (Figure 7(b)) is less smooth than themap obtained with ordinary kriging (Figure 6(e))

4 Conclusions

When taking into account all the soil properties studied ECaand gravimetric soil water content measured at the samedate that is 2362008 showed the highest coefficients ofcorrelationTable 4Moreover ECa showedhigher coefficientsof correlation to clay and silt content than to silt contentand the strength of the correlation was higher for the firstECa recording date that is 1432008 when the soil moisturewas lower Thus coefficient of correlation of ECa with siltand clay content showed a trend to increase the soil moisturedecreased this result suggests the usefulness of recording ECaon successive dates with different soil water contents

The spatial patterns of spatial variability of the logarith-mic values of apparent soil electrical conductivity (ECa) andthe electrical conductivity of the soil saturated paste (ECe)were modeled by universal kriging whereas those of sandclay silt and gravimetric water content were modeled byordinary kriging The use of cokriging with ECa data as sec-ondary variable improved the estimation of the gravimetricsoil water content with respect to the use of kriging

Abbreviations

ECa Apparent soil electrical conductivityECa-V CEa in vertical dipoleECa-H CEa in horizontal dipoleECe Electrical conductivity of soil saturation extract

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper The referencingbrands of commercial products in the paper do not imply thatthe authors recommend the equipment utilized in this studyThe support funds presented by the Development Agenciesdo not show conflict of interests regarding the publication ofthis paper

Acknowledgments

The authors are grateful to the Ministerial de AsuntosExteriors y de Cooperation (MAEC-AECID) from Spain forthe granting of scholarships for PhD studies This workhas been funded by Ministerial de Education y Cienciawithin the framework of research Project CGL2005-08219-C02-02 and cofunded by Xunta de Galicia within theframework of research Project PGIDIT06PXIC291062PNand by the European Regional Development Fund (ERDF)The authors acknowledge the provincial farm Gayoso Castro

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

The Scientific World Journal 3

N

EW

S

622500 622600 622700 622800

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

37

38

36

35

34

32

33

31

23

3024

25

2629

40

22

27

28

21

20

1918

8

6

14

9

39

7

1

2

3

4

5

11

12

1315

16

17

10

(a) (b)

Figure 2 Scheme showing apparent electrical conductivity (ECa) continuously recorded (line) and the location of 40 soil sampling points(circles) on 2362008 (a) and cart containing the EM38-DD equipment and GPS (b)

N

EW

S622550 622650 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

(a)

N

EW

S

622550 622650 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

(b)

Figure 3 Scheme showing apparent electrical conductivity (ECa) recorded during 1432008 (a) and 342008 (b)

EM38-DD [14] This device consists of two integrated EM38units oriented in the horizontal and vertical dipole positionsproviding simultaneous measurements for the two dipolesmodes in the vertical dipole mode the primary magneticfield is thought to effectively penetrate to a depth of about15m while in the horizontal dipole position EM38-DD isthought to be effective for a shallower soil depth estimatedat about 075m [14]

To complete continuous record of the apparent soilelectrical conductivity in horizontal dipole (ECa-H mSmminus1)and in vertical dipole (ECa-V mSmminus1) (Figures 2(a) and3) the EM38-DD was installed in a car built with plastic

materials (Figure 2(b)) In addition GPS RTK was used forgeoreferencing the recorded measures

The reference measurements of ECa-H and ECa-V wereperformed on 2362008 at 1859 sampling points followingthe scheme presented in Figure 2(a) The soil samplingscheme was devised using the software tool ESAP-RSSD(response surface sampling design) based on a multiplelinear regression model [7 9]This software aims to optimizethe position of new sampling points considering apparentsoil electrical conductivity (ECa) measured with horizontal(ECa-H) and vertical dipoles (ECa-V) (Figure 2(a)) Theoptimized soil sampling scheme consisted of 40 points In

4 The Scientific World Journal

Table 2 Statistical parameters of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Unit 119873 Min Max Mean plusmn SD Variance CV Skew Kurt 119863

1432008 ECa-V mSmminus1 1887 575 1838 1048 plusmn 119 142 1135 0527 0124 0045Ln1432008 ECa-H mSmminus1 1887 925 1900 141 plusmn 077 060 546 0065 1810 0040Ln342008 ECa-V mSmminus1 1871 963 2050 1404 plusmn 215 464 1531 0662 0083 0073Ln342008 ECa-H mSmminus1 1871 663 1950 1459 plusmn 077 060 528 0160 1051 0095Ln2362008 ECa-V mSmminus1 1886 413 2013 1121 plusmn 247lowast 612 2203 0485 minus0243 0071Ln2362008 ECa-H mSmminus1 1886 663 2000 1212 plusmn 179lowast 322 1477 0839 1285 0092Ln2362008 CEe mSmminus1 40 700 2800 1382 plusmn 509 2594 3683 1200 1008 0159n2362008 Clay g kgminus1 40 11900 22000 16837 plusmn 3092 95654 1836 minus0190 minus1321 0153n2362008 Silt g kgminus1 40 23300 35700 29625 plusmn 3506 122973 1183 0149 minus1041 0098n2362008 Sand g kgminus1 40 48700 58600 53537 plusmn 2253 50772 421 0055 minus0270 0069n2362008 120579119892 38 1341 4567 2674 plusmn 681 4950 2547 minus2904 6510 0085n119873 number of measurements Min minimum value Max maximum value Mean plusmn SD mean plusmn standard deviation CV coefficient of variation () Skewskewness Kurt kurtosis and119863 normality of the data for test of Kolmogorov-Smirnov (119875 lt 001 n normality and Ln log normality) lowastNonsignificant at 5level of ANOVA (SNK)

addition continuous measurements of ECa-H and ECa-Vwere previously taken in the experimental field on 1432008and 342008 as shown in Figure 3 Note however that theschemes of the continuously recorded ECa data sets taken inthe three successive dates were different as shown in Figures2(a) 3(a) and 3(b)

In the 40 points selected during the ECa campaign of2362008 soil samples were taken at the 00ndash03m depthwith a manual soil probe Soil texture soil water contentand electrical conductivity of saturated paste extracts weredetermined using standard methods Soil texture (clay siltand sand in g kgminus1) was determined by the sieve-pipettemethod following Camargo et al [15] in this method a mix-ture of sodium hydroxide and sodium hexametaphosphatewas used as chemical dispersant The gravimetric soil watercontent (120579119892 ) was obtained after weighing the mass ofthe wet and dry sample according to Camargo et al [15]To determine the electrical conductivity of soil saturatedextracts (ECe) a mixture of soil and distilled water of 1 1 wasprepared as proposed by USDA [16] electrical conductivitymeasurements were performed using a conductivity meterORIONModel 122

23 Statistical and Geostatistical Analysis All the values werestatistically analyzed using SPSS package 115 at 5 level ofSNK (Student-Newman-Keuls) method ANOVA The testof normality Kolmogorov-Smirnov was used to test thenormality of data with probability of error 1 (119875 lt 001)Thecorrelation was calculated with the correlation coefficient ofPearson

The analysis of the spatial variability of soil physicalproperties was conducted using the experimental variogramthe fitting of variogram model was performed using themethod described by Vieira [17] based on cross-validationInitial analysis showed that the variogram of any studiedproperties showed a trend so the universal kriging was usedin these cases in which the residual variogram is required[18] For those variables that showed no trend ordinary

kriging was used The degree of spatial dependence (SD) wasdetermined according to the following

SD () =1198620

1198620+ 1198621

lowast 100 (1)

where 1198620is nugget effect and (119862

0+ 1198621) is the sill (119862

0+ 1198621)

according to Cambardella et al [19] which is considered ashigh (SD le 25) moderate (SD = 25ndash75) and low (SD ge75)

Cross-variogramwas used to study the spatial correlationbetween soil variables when there was a trend in some ofthese variables universal cokriging was used [18] instead ofordinary cokriging The software used to perform ordinarykriging universal kriging and universal cokriging was Gstat[20] In cokriging the covariance matrix must be positive anddefinite [18 21ndash23] The use of cokriging was used only for acouple of attributes that showed correlation coefficient values(|119903|) greater than 05

3 Results and Discussion

Statistical analysis of the data (Table 2) indicates that thereis great variation between samples in accordance with low(CV le 12) and middle (CV = 12ndash60) variation coefficientvalues by classification of Warrick and Nielsen [24] It isverified that the apparent electrical conductivity of the soil(ECa) measurement with the horizontal dipole (ECa-H) haslower CV than the measurements with the vertical dipole(ECa-V)This fact can be explained because the vertical dipolemode explores a larger volume of soil than the horizontaldipole and there is greater heterogeneity in those variablesthat affect the ECa values mainly clay content organicmaterial water content in the soil porosity salinity and soforth [2ndash5]

Only data ECa-V and ECa-H sampling in 2362008did not show differentiation by the average test (ANOVA)between the different sampling dates

The Scientific World Journal 5

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)Log ECa-V2368Log ECa-V1438Log ECa-V348

(a)

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)

Log ECa-H 348

Log ECa-H 2368Log ECa-H 1438

(b)

Figure 4 Standardized sample semivariogram for log ECa-V (a) and log ECa-H (b) recorded during three successive dates

Table 3 Precipitation and reference evapotranspiration betweensuccessive dates in which apparent soil electrical conductivity (CEa-V and CEa-H) was recorded

Period Precipitation(mm)

Referenceevapotranspiration (mm)

1522008ndash1432008 526 3801432008ndash342008 804 371342008ndash2362008 3978 2146

The values of the electrical conductivity of the saturationpaste extract of the soil (ECe) are higher than the values ofECa-V and ECa-H this fact is because ECe is a parameterthat depends on the content of anions and cations in the soilsolution the water content is homogeneous in all samplesbecause the sample is saturatedwithwater and the soil appar-ent electric conductivity valuesmeasuredwith the equipmentEM38-DD (ECa-V and ECa-H) are very influenced by the soilwater content [2 3 5] and the water content in the soil isvariable along the field

ECa-V and ECa-H measured on several sampling dates(1432008 342008 and 2362008) showed lognormal dis-tribution (Table 2) Other attributes studied showed normalfrequency distribution (ECe clay silt sand and soil watercontent)

In the geostatistical analysis lognormal transformationwas used for properties that showed lognormal distributionThe highest values of coefficient of correlation betweenECa variables and clay and silt content are on the firstmeasurement date (1432008) on this date the soil moistureis lower coincided with data of precipitation and evapotran-spiration (Table 3) Grandjean et al [25] describe that soilwith moisture lower is ideal for characterization of soil bulkdensity and of soil texture using measurements of electricalconductivity

The coefficient of correlation values between the apparentsoil electrical conductivity (ECa-V and ECa-H) measure-ment on several sampling dates (1432008 342008 and2362008) presented moderate positive correlation coeffi-cient values (05 le 119903 lt 08)

The coefficient of correlation between ECa-V2362008 times120579119892 (119903 = 0685) and ECa-H2362008 times 120579119892 (119903 = 0648)was moderately positive (05 le 119903 lt 08) confirming thecorrelation between the values of ECa-V and ECa-H andthe water content in soil because according to Grandjeanet al [25] under wet conditions electrical conductivitymeasurements are dominated by the effect of water contentwhich tends to hide the influence of the other factors

The values of log ECa-V are affected by the groundwaterlevel so the variogram follows the trend in the ground waterlevel (Figure 4) As can be seen on standardized variogramswith the value of the sample variance data from ECa-Vmeasurement on 1432008 and 342008 present a trendfollowing the same pattern of the digital elevation map ofthe area (Figure 1(b)) Analyzing standardized variogramsfor log ECa-H data can be seen that only shows trend forthe variogram of 1432008 and 342008 but not for thevariogram of 2362008 on this date the water table wasprobably located below the depth of soil investigated with thehorizontal dipole mode

Corwin and Lesch [4] found higher values of correlationbetween data from log ECa-V and log ECa-H and electricalconductivity of the saturation extract (ECe) and clay contentbut lower than those found for log ECa-V and log ECa-Hand the water content in soil Martınez and Vanderlinden[26] described a higher correlation between ECa and watercontent in loamy soils while in clay soils the correlation waslower The correlation coefficients for ECe silt and clay withthe log ECa-H are greater than with the log-ECa-V

In order to improve the correlation between the values ofECa-V andECa-Hwith clay content soil water content shouldbe as homogeneous as possible within the study area better if

6 The Scientific World Journal

its value is closer to field capacity and unlike the water tableis as low as possible so the best time to take measurementsunder these conditions would be in the autumn when therewas heavy rainfall although under these conditions the watertable probably would not have ascended enough to be close tothe surface

The initial geostatistical data analysis showed that thephysical properties of the soil (clay silt sand and gravimetricwater content) showed no trend then being possible theestimate of the variable using the original data with ordinarykriging Moreover ECe data and apparent electrical conduc-tivity of the soil (ECa-V and ECa-H) on several samplingdates show trend in Figure 3 the semivariance value is notstabilized around variance value of data and the universalkriging was used to construct the maps of spatial variabilityof these variables

The fitted variogram parameters (Table 5) show that thespherical model was the fitted model to the properties understudy according to Cambardella et al [19] Goovaerts [18]Vieira [17] and Siqueira et al [27] describing this modelit is usually best fitted to the properties of soil and plantAll attributes had low values of nugget effect (119862

0) Range

values (a) varied from 4000m (log ECa-H Residual342008

)to 13000m (clay silt and soil water content) The degreeof spatial dependence between samples was high (SD le2500) across the study the exception being log ECa-H Residual

1432008which presented a moderate degree of

spatial dependence (SD = 3167)The spatial variability maps obtained with universal

kriging (Figure 5) show that there is a similarity between themaps ECa-V (Figures 5(a) 5(c) and 5(e)) and ECa-H (Figures5(b) 5(d) and 5(f)) on several sampling dates with furtherdifferentiation of maps of ECa-V and ECa-H on themeasureddata on 2362008 (Figures 5(e) and 5(f))when thewater tablelevel was lower

It is observed that the maps of ECa-V and ECa-H(Figure 5) and the map of the water content in soil (Fig-ure 6(e)) obtained with ordinary kriging (Figure 6(e)) looksimilar following the same pattern of digital elevation model(Figure 1(b))

The map of the electrical conductivity of the saturationextract (ECe Figure 6(a)) shows inverse behaviour to mapsECa-V and ECa-H (Figure 5) Moreover the map of spatialvariability of clay content in the study area (Figure 6(b))shows no similarity to maps ECa-V and ECa-H (Figure 5)this fact is also repeated with silt (Figure 6(c)) and sand(Figure 6(d))

Table 6 presents the fitting parameters cross-variogrambetween 120579119892 times ECa-V (119903 = 0685) and 120579119892 times ECa-H (119903 = 0648)The cross-variograms were fitted to a spherical model withthe same range compared to single variograms to obtain alinear model coregionalization (Table 6)

The spatial variability maps constructed using ordinaryand universal cokriging (Figure 7) demonstrate that theuse of the soil apparent electrical conductivity measured byelectromagnetic induction (ECa-V and ECa-H) on 2362008was a secondary variable that improves the estimation of thesoil water content using cokriging This improvement in theestimation of 120579119892 can be observed in Table 7 where it showed

an increase in the value of the correlation coefficient betweenthe measured and the estimated values from cross-validationusing ordinary cokriging with log-ECa-V (0746) and withlog-ECa-H (0756) as secondary variables with respect to useof ordinary kriging (0637) Moreover in the case of the soilwater content map obtained with ordinary cokriging using assecondary data ECa-H (Figure 7(b)) is less smooth than themap obtained with ordinary kriging (Figure 6(e))

4 Conclusions

When taking into account all the soil properties studied ECaand gravimetric soil water content measured at the samedate that is 2362008 showed the highest coefficients ofcorrelationTable 4Moreover ECa showedhigher coefficientsof correlation to clay and silt content than to silt contentand the strength of the correlation was higher for the firstECa recording date that is 1432008 when the soil moisturewas lower Thus coefficient of correlation of ECa with siltand clay content showed a trend to increase the soil moisturedecreased this result suggests the usefulness of recording ECaon successive dates with different soil water contents

The spatial patterns of spatial variability of the logarith-mic values of apparent soil electrical conductivity (ECa) andthe electrical conductivity of the soil saturated paste (ECe)were modeled by universal kriging whereas those of sandclay silt and gravimetric water content were modeled byordinary kriging The use of cokriging with ECa data as sec-ondary variable improved the estimation of the gravimetricsoil water content with respect to the use of kriging

Abbreviations

ECa Apparent soil electrical conductivityECa-V CEa in vertical dipoleECa-H CEa in horizontal dipoleECe Electrical conductivity of soil saturation extract

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper The referencingbrands of commercial products in the paper do not imply thatthe authors recommend the equipment utilized in this studyThe support funds presented by the Development Agenciesdo not show conflict of interests regarding the publication ofthis paper

Acknowledgments

The authors are grateful to the Ministerial de AsuntosExteriors y de Cooperation (MAEC-AECID) from Spain forthe granting of scholarships for PhD studies This workhas been funded by Ministerial de Education y Cienciawithin the framework of research Project CGL2005-08219-C02-02 and cofunded by Xunta de Galicia within theframework of research Project PGIDIT06PXIC291062PNand by the European Regional Development Fund (ERDF)The authors acknowledge the provincial farm Gayoso Castro

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

4 The Scientific World Journal

Table 2 Statistical parameters of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Unit 119873 Min Max Mean plusmn SD Variance CV Skew Kurt 119863

1432008 ECa-V mSmminus1 1887 575 1838 1048 plusmn 119 142 1135 0527 0124 0045Ln1432008 ECa-H mSmminus1 1887 925 1900 141 plusmn 077 060 546 0065 1810 0040Ln342008 ECa-V mSmminus1 1871 963 2050 1404 plusmn 215 464 1531 0662 0083 0073Ln342008 ECa-H mSmminus1 1871 663 1950 1459 plusmn 077 060 528 0160 1051 0095Ln2362008 ECa-V mSmminus1 1886 413 2013 1121 plusmn 247lowast 612 2203 0485 minus0243 0071Ln2362008 ECa-H mSmminus1 1886 663 2000 1212 plusmn 179lowast 322 1477 0839 1285 0092Ln2362008 CEe mSmminus1 40 700 2800 1382 plusmn 509 2594 3683 1200 1008 0159n2362008 Clay g kgminus1 40 11900 22000 16837 plusmn 3092 95654 1836 minus0190 minus1321 0153n2362008 Silt g kgminus1 40 23300 35700 29625 plusmn 3506 122973 1183 0149 minus1041 0098n2362008 Sand g kgminus1 40 48700 58600 53537 plusmn 2253 50772 421 0055 minus0270 0069n2362008 120579119892 38 1341 4567 2674 plusmn 681 4950 2547 minus2904 6510 0085n119873 number of measurements Min minimum value Max maximum value Mean plusmn SD mean plusmn standard deviation CV coefficient of variation () Skewskewness Kurt kurtosis and119863 normality of the data for test of Kolmogorov-Smirnov (119875 lt 001 n normality and Ln log normality) lowastNonsignificant at 5level of ANOVA (SNK)

addition continuous measurements of ECa-H and ECa-Vwere previously taken in the experimental field on 1432008and 342008 as shown in Figure 3 Note however that theschemes of the continuously recorded ECa data sets taken inthe three successive dates were different as shown in Figures2(a) 3(a) and 3(b)

In the 40 points selected during the ECa campaign of2362008 soil samples were taken at the 00ndash03m depthwith a manual soil probe Soil texture soil water contentand electrical conductivity of saturated paste extracts weredetermined using standard methods Soil texture (clay siltand sand in g kgminus1) was determined by the sieve-pipettemethod following Camargo et al [15] in this method a mix-ture of sodium hydroxide and sodium hexametaphosphatewas used as chemical dispersant The gravimetric soil watercontent (120579119892 ) was obtained after weighing the mass ofthe wet and dry sample according to Camargo et al [15]To determine the electrical conductivity of soil saturatedextracts (ECe) a mixture of soil and distilled water of 1 1 wasprepared as proposed by USDA [16] electrical conductivitymeasurements were performed using a conductivity meterORIONModel 122

23 Statistical and Geostatistical Analysis All the values werestatistically analyzed using SPSS package 115 at 5 level ofSNK (Student-Newman-Keuls) method ANOVA The testof normality Kolmogorov-Smirnov was used to test thenormality of data with probability of error 1 (119875 lt 001)Thecorrelation was calculated with the correlation coefficient ofPearson

The analysis of the spatial variability of soil physicalproperties was conducted using the experimental variogramthe fitting of variogram model was performed using themethod described by Vieira [17] based on cross-validationInitial analysis showed that the variogram of any studiedproperties showed a trend so the universal kriging was usedin these cases in which the residual variogram is required[18] For those variables that showed no trend ordinary

kriging was used The degree of spatial dependence (SD) wasdetermined according to the following

SD () =1198620

1198620+ 1198621

lowast 100 (1)

where 1198620is nugget effect and (119862

0+ 1198621) is the sill (119862

0+ 1198621)

according to Cambardella et al [19] which is considered ashigh (SD le 25) moderate (SD = 25ndash75) and low (SD ge75)

Cross-variogramwas used to study the spatial correlationbetween soil variables when there was a trend in some ofthese variables universal cokriging was used [18] instead ofordinary cokriging The software used to perform ordinarykriging universal kriging and universal cokriging was Gstat[20] In cokriging the covariance matrix must be positive anddefinite [18 21ndash23] The use of cokriging was used only for acouple of attributes that showed correlation coefficient values(|119903|) greater than 05

3 Results and Discussion

Statistical analysis of the data (Table 2) indicates that thereis great variation between samples in accordance with low(CV le 12) and middle (CV = 12ndash60) variation coefficientvalues by classification of Warrick and Nielsen [24] It isverified that the apparent electrical conductivity of the soil(ECa) measurement with the horizontal dipole (ECa-H) haslower CV than the measurements with the vertical dipole(ECa-V)This fact can be explained because the vertical dipolemode explores a larger volume of soil than the horizontaldipole and there is greater heterogeneity in those variablesthat affect the ECa values mainly clay content organicmaterial water content in the soil porosity salinity and soforth [2ndash5]

Only data ECa-V and ECa-H sampling in 2362008did not show differentiation by the average test (ANOVA)between the different sampling dates

The Scientific World Journal 5

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)Log ECa-V2368Log ECa-V1438Log ECa-V348

(a)

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)

Log ECa-H 348

Log ECa-H 2368Log ECa-H 1438

(b)

Figure 4 Standardized sample semivariogram for log ECa-V (a) and log ECa-H (b) recorded during three successive dates

Table 3 Precipitation and reference evapotranspiration betweensuccessive dates in which apparent soil electrical conductivity (CEa-V and CEa-H) was recorded

Period Precipitation(mm)

Referenceevapotranspiration (mm)

1522008ndash1432008 526 3801432008ndash342008 804 371342008ndash2362008 3978 2146

The values of the electrical conductivity of the saturationpaste extract of the soil (ECe) are higher than the values ofECa-V and ECa-H this fact is because ECe is a parameterthat depends on the content of anions and cations in the soilsolution the water content is homogeneous in all samplesbecause the sample is saturatedwithwater and the soil appar-ent electric conductivity valuesmeasuredwith the equipmentEM38-DD (ECa-V and ECa-H) are very influenced by the soilwater content [2 3 5] and the water content in the soil isvariable along the field

ECa-V and ECa-H measured on several sampling dates(1432008 342008 and 2362008) showed lognormal dis-tribution (Table 2) Other attributes studied showed normalfrequency distribution (ECe clay silt sand and soil watercontent)

In the geostatistical analysis lognormal transformationwas used for properties that showed lognormal distributionThe highest values of coefficient of correlation betweenECa variables and clay and silt content are on the firstmeasurement date (1432008) on this date the soil moistureis lower coincided with data of precipitation and evapotran-spiration (Table 3) Grandjean et al [25] describe that soilwith moisture lower is ideal for characterization of soil bulkdensity and of soil texture using measurements of electricalconductivity

The coefficient of correlation values between the apparentsoil electrical conductivity (ECa-V and ECa-H) measure-ment on several sampling dates (1432008 342008 and2362008) presented moderate positive correlation coeffi-cient values (05 le 119903 lt 08)

The coefficient of correlation between ECa-V2362008 times120579119892 (119903 = 0685) and ECa-H2362008 times 120579119892 (119903 = 0648)was moderately positive (05 le 119903 lt 08) confirming thecorrelation between the values of ECa-V and ECa-H andthe water content in soil because according to Grandjeanet al [25] under wet conditions electrical conductivitymeasurements are dominated by the effect of water contentwhich tends to hide the influence of the other factors

The values of log ECa-V are affected by the groundwaterlevel so the variogram follows the trend in the ground waterlevel (Figure 4) As can be seen on standardized variogramswith the value of the sample variance data from ECa-Vmeasurement on 1432008 and 342008 present a trendfollowing the same pattern of the digital elevation map ofthe area (Figure 1(b)) Analyzing standardized variogramsfor log ECa-H data can be seen that only shows trend forthe variogram of 1432008 and 342008 but not for thevariogram of 2362008 on this date the water table wasprobably located below the depth of soil investigated with thehorizontal dipole mode

Corwin and Lesch [4] found higher values of correlationbetween data from log ECa-V and log ECa-H and electricalconductivity of the saturation extract (ECe) and clay contentbut lower than those found for log ECa-V and log ECa-Hand the water content in soil Martınez and Vanderlinden[26] described a higher correlation between ECa and watercontent in loamy soils while in clay soils the correlation waslower The correlation coefficients for ECe silt and clay withthe log ECa-H are greater than with the log-ECa-V

In order to improve the correlation between the values ofECa-V andECa-Hwith clay content soil water content shouldbe as homogeneous as possible within the study area better if

6 The Scientific World Journal

its value is closer to field capacity and unlike the water tableis as low as possible so the best time to take measurementsunder these conditions would be in the autumn when therewas heavy rainfall although under these conditions the watertable probably would not have ascended enough to be close tothe surface

The initial geostatistical data analysis showed that thephysical properties of the soil (clay silt sand and gravimetricwater content) showed no trend then being possible theestimate of the variable using the original data with ordinarykriging Moreover ECe data and apparent electrical conduc-tivity of the soil (ECa-V and ECa-H) on several samplingdates show trend in Figure 3 the semivariance value is notstabilized around variance value of data and the universalkriging was used to construct the maps of spatial variabilityof these variables

The fitted variogram parameters (Table 5) show that thespherical model was the fitted model to the properties understudy according to Cambardella et al [19] Goovaerts [18]Vieira [17] and Siqueira et al [27] describing this modelit is usually best fitted to the properties of soil and plantAll attributes had low values of nugget effect (119862

0) Range

values (a) varied from 4000m (log ECa-H Residual342008

)to 13000m (clay silt and soil water content) The degreeof spatial dependence between samples was high (SD le2500) across the study the exception being log ECa-H Residual

1432008which presented a moderate degree of

spatial dependence (SD = 3167)The spatial variability maps obtained with universal

kriging (Figure 5) show that there is a similarity between themaps ECa-V (Figures 5(a) 5(c) and 5(e)) and ECa-H (Figures5(b) 5(d) and 5(f)) on several sampling dates with furtherdifferentiation of maps of ECa-V and ECa-H on themeasureddata on 2362008 (Figures 5(e) and 5(f))when thewater tablelevel was lower

It is observed that the maps of ECa-V and ECa-H(Figure 5) and the map of the water content in soil (Fig-ure 6(e)) obtained with ordinary kriging (Figure 6(e)) looksimilar following the same pattern of digital elevation model(Figure 1(b))

The map of the electrical conductivity of the saturationextract (ECe Figure 6(a)) shows inverse behaviour to mapsECa-V and ECa-H (Figure 5) Moreover the map of spatialvariability of clay content in the study area (Figure 6(b))shows no similarity to maps ECa-V and ECa-H (Figure 5)this fact is also repeated with silt (Figure 6(c)) and sand(Figure 6(d))

Table 6 presents the fitting parameters cross-variogrambetween 120579119892 times ECa-V (119903 = 0685) and 120579119892 times ECa-H (119903 = 0648)The cross-variograms were fitted to a spherical model withthe same range compared to single variograms to obtain alinear model coregionalization (Table 6)

The spatial variability maps constructed using ordinaryand universal cokriging (Figure 7) demonstrate that theuse of the soil apparent electrical conductivity measured byelectromagnetic induction (ECa-V and ECa-H) on 2362008was a secondary variable that improves the estimation of thesoil water content using cokriging This improvement in theestimation of 120579119892 can be observed in Table 7 where it showed

an increase in the value of the correlation coefficient betweenthe measured and the estimated values from cross-validationusing ordinary cokriging with log-ECa-V (0746) and withlog-ECa-H (0756) as secondary variables with respect to useof ordinary kriging (0637) Moreover in the case of the soilwater content map obtained with ordinary cokriging using assecondary data ECa-H (Figure 7(b)) is less smooth than themap obtained with ordinary kriging (Figure 6(e))

4 Conclusions

When taking into account all the soil properties studied ECaand gravimetric soil water content measured at the samedate that is 2362008 showed the highest coefficients ofcorrelationTable 4Moreover ECa showedhigher coefficientsof correlation to clay and silt content than to silt contentand the strength of the correlation was higher for the firstECa recording date that is 1432008 when the soil moisturewas lower Thus coefficient of correlation of ECa with siltand clay content showed a trend to increase the soil moisturedecreased this result suggests the usefulness of recording ECaon successive dates with different soil water contents

The spatial patterns of spatial variability of the logarith-mic values of apparent soil electrical conductivity (ECa) andthe electrical conductivity of the soil saturated paste (ECe)were modeled by universal kriging whereas those of sandclay silt and gravimetric water content were modeled byordinary kriging The use of cokriging with ECa data as sec-ondary variable improved the estimation of the gravimetricsoil water content with respect to the use of kriging

Abbreviations

ECa Apparent soil electrical conductivityECa-V CEa in vertical dipoleECa-H CEa in horizontal dipoleECe Electrical conductivity of soil saturation extract

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper The referencingbrands of commercial products in the paper do not imply thatthe authors recommend the equipment utilized in this studyThe support funds presented by the Development Agenciesdo not show conflict of interests regarding the publication ofthis paper

Acknowledgments

The authors are grateful to the Ministerial de AsuntosExteriors y de Cooperation (MAEC-AECID) from Spain forthe granting of scholarships for PhD studies This workhas been funded by Ministerial de Education y Cienciawithin the framework of research Project CGL2005-08219-C02-02 and cofunded by Xunta de Galicia within theframework of research Project PGIDIT06PXIC291062PNand by the European Regional Development Fund (ERDF)The authors acknowledge the provincial farm Gayoso Castro

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

The Scientific World Journal 5

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)Log ECa-V2368Log ECa-V1438Log ECa-V348

(a)

0

02

04

06

08

1

12

14

0 50 100 150

Sem

ivar

ianc

e

Distance (m)

Log ECa-H 348

Log ECa-H 2368Log ECa-H 1438

(b)

Figure 4 Standardized sample semivariogram for log ECa-V (a) and log ECa-H (b) recorded during three successive dates

Table 3 Precipitation and reference evapotranspiration betweensuccessive dates in which apparent soil electrical conductivity (CEa-V and CEa-H) was recorded

Period Precipitation(mm)

Referenceevapotranspiration (mm)

1522008ndash1432008 526 3801432008ndash342008 804 371342008ndash2362008 3978 2146

The values of the electrical conductivity of the saturationpaste extract of the soil (ECe) are higher than the values ofECa-V and ECa-H this fact is because ECe is a parameterthat depends on the content of anions and cations in the soilsolution the water content is homogeneous in all samplesbecause the sample is saturatedwithwater and the soil appar-ent electric conductivity valuesmeasuredwith the equipmentEM38-DD (ECa-V and ECa-H) are very influenced by the soilwater content [2 3 5] and the water content in the soil isvariable along the field

ECa-V and ECa-H measured on several sampling dates(1432008 342008 and 2362008) showed lognormal dis-tribution (Table 2) Other attributes studied showed normalfrequency distribution (ECe clay silt sand and soil watercontent)

In the geostatistical analysis lognormal transformationwas used for properties that showed lognormal distributionThe highest values of coefficient of correlation betweenECa variables and clay and silt content are on the firstmeasurement date (1432008) on this date the soil moistureis lower coincided with data of precipitation and evapotran-spiration (Table 3) Grandjean et al [25] describe that soilwith moisture lower is ideal for characterization of soil bulkdensity and of soil texture using measurements of electricalconductivity

The coefficient of correlation values between the apparentsoil electrical conductivity (ECa-V and ECa-H) measure-ment on several sampling dates (1432008 342008 and2362008) presented moderate positive correlation coeffi-cient values (05 le 119903 lt 08)

The coefficient of correlation between ECa-V2362008 times120579119892 (119903 = 0685) and ECa-H2362008 times 120579119892 (119903 = 0648)was moderately positive (05 le 119903 lt 08) confirming thecorrelation between the values of ECa-V and ECa-H andthe water content in soil because according to Grandjeanet al [25] under wet conditions electrical conductivitymeasurements are dominated by the effect of water contentwhich tends to hide the influence of the other factors

The values of log ECa-V are affected by the groundwaterlevel so the variogram follows the trend in the ground waterlevel (Figure 4) As can be seen on standardized variogramswith the value of the sample variance data from ECa-Vmeasurement on 1432008 and 342008 present a trendfollowing the same pattern of the digital elevation map ofthe area (Figure 1(b)) Analyzing standardized variogramsfor log ECa-H data can be seen that only shows trend forthe variogram of 1432008 and 342008 but not for thevariogram of 2362008 on this date the water table wasprobably located below the depth of soil investigated with thehorizontal dipole mode

Corwin and Lesch [4] found higher values of correlationbetween data from log ECa-V and log ECa-H and electricalconductivity of the saturation extract (ECe) and clay contentbut lower than those found for log ECa-V and log ECa-Hand the water content in soil Martınez and Vanderlinden[26] described a higher correlation between ECa and watercontent in loamy soils while in clay soils the correlation waslower The correlation coefficients for ECe silt and clay withthe log ECa-H are greater than with the log-ECa-V

In order to improve the correlation between the values ofECa-V andECa-Hwith clay content soil water content shouldbe as homogeneous as possible within the study area better if

6 The Scientific World Journal

its value is closer to field capacity and unlike the water tableis as low as possible so the best time to take measurementsunder these conditions would be in the autumn when therewas heavy rainfall although under these conditions the watertable probably would not have ascended enough to be close tothe surface

The initial geostatistical data analysis showed that thephysical properties of the soil (clay silt sand and gravimetricwater content) showed no trend then being possible theestimate of the variable using the original data with ordinarykriging Moreover ECe data and apparent electrical conduc-tivity of the soil (ECa-V and ECa-H) on several samplingdates show trend in Figure 3 the semivariance value is notstabilized around variance value of data and the universalkriging was used to construct the maps of spatial variabilityof these variables

The fitted variogram parameters (Table 5) show that thespherical model was the fitted model to the properties understudy according to Cambardella et al [19] Goovaerts [18]Vieira [17] and Siqueira et al [27] describing this modelit is usually best fitted to the properties of soil and plantAll attributes had low values of nugget effect (119862

0) Range

values (a) varied from 4000m (log ECa-H Residual342008

)to 13000m (clay silt and soil water content) The degreeof spatial dependence between samples was high (SD le2500) across the study the exception being log ECa-H Residual

1432008which presented a moderate degree of

spatial dependence (SD = 3167)The spatial variability maps obtained with universal

kriging (Figure 5) show that there is a similarity between themaps ECa-V (Figures 5(a) 5(c) and 5(e)) and ECa-H (Figures5(b) 5(d) and 5(f)) on several sampling dates with furtherdifferentiation of maps of ECa-V and ECa-H on themeasureddata on 2362008 (Figures 5(e) and 5(f))when thewater tablelevel was lower

It is observed that the maps of ECa-V and ECa-H(Figure 5) and the map of the water content in soil (Fig-ure 6(e)) obtained with ordinary kriging (Figure 6(e)) looksimilar following the same pattern of digital elevation model(Figure 1(b))

The map of the electrical conductivity of the saturationextract (ECe Figure 6(a)) shows inverse behaviour to mapsECa-V and ECa-H (Figure 5) Moreover the map of spatialvariability of clay content in the study area (Figure 6(b))shows no similarity to maps ECa-V and ECa-H (Figure 5)this fact is also repeated with silt (Figure 6(c)) and sand(Figure 6(d))

Table 6 presents the fitting parameters cross-variogrambetween 120579119892 times ECa-V (119903 = 0685) and 120579119892 times ECa-H (119903 = 0648)The cross-variograms were fitted to a spherical model withthe same range compared to single variograms to obtain alinear model coregionalization (Table 6)

The spatial variability maps constructed using ordinaryand universal cokriging (Figure 7) demonstrate that theuse of the soil apparent electrical conductivity measured byelectromagnetic induction (ECa-V and ECa-H) on 2362008was a secondary variable that improves the estimation of thesoil water content using cokriging This improvement in theestimation of 120579119892 can be observed in Table 7 where it showed

an increase in the value of the correlation coefficient betweenthe measured and the estimated values from cross-validationusing ordinary cokriging with log-ECa-V (0746) and withlog-ECa-H (0756) as secondary variables with respect to useof ordinary kriging (0637) Moreover in the case of the soilwater content map obtained with ordinary cokriging using assecondary data ECa-H (Figure 7(b)) is less smooth than themap obtained with ordinary kriging (Figure 6(e))

4 Conclusions

When taking into account all the soil properties studied ECaand gravimetric soil water content measured at the samedate that is 2362008 showed the highest coefficients ofcorrelationTable 4Moreover ECa showedhigher coefficientsof correlation to clay and silt content than to silt contentand the strength of the correlation was higher for the firstECa recording date that is 1432008 when the soil moisturewas lower Thus coefficient of correlation of ECa with siltand clay content showed a trend to increase the soil moisturedecreased this result suggests the usefulness of recording ECaon successive dates with different soil water contents

The spatial patterns of spatial variability of the logarith-mic values of apparent soil electrical conductivity (ECa) andthe electrical conductivity of the soil saturated paste (ECe)were modeled by universal kriging whereas those of sandclay silt and gravimetric water content were modeled byordinary kriging The use of cokriging with ECa data as sec-ondary variable improved the estimation of the gravimetricsoil water content with respect to the use of kriging

Abbreviations

ECa Apparent soil electrical conductivityECa-V CEa in vertical dipoleECa-H CEa in horizontal dipoleECe Electrical conductivity of soil saturation extract

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper The referencingbrands of commercial products in the paper do not imply thatthe authors recommend the equipment utilized in this studyThe support funds presented by the Development Agenciesdo not show conflict of interests regarding the publication ofthis paper

Acknowledgments

The authors are grateful to the Ministerial de AsuntosExteriors y de Cooperation (MAEC-AECID) from Spain forthe granting of scholarships for PhD studies This workhas been funded by Ministerial de Education y Cienciawithin the framework of research Project CGL2005-08219-C02-02 and cofunded by Xunta de Galicia within theframework of research Project PGIDIT06PXIC291062PNand by the European Regional Development Fund (ERDF)The authors acknowledge the provincial farm Gayoso Castro

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

6 The Scientific World Journal

its value is closer to field capacity and unlike the water tableis as low as possible so the best time to take measurementsunder these conditions would be in the autumn when therewas heavy rainfall although under these conditions the watertable probably would not have ascended enough to be close tothe surface

The initial geostatistical data analysis showed that thephysical properties of the soil (clay silt sand and gravimetricwater content) showed no trend then being possible theestimate of the variable using the original data with ordinarykriging Moreover ECe data and apparent electrical conduc-tivity of the soil (ECa-V and ECa-H) on several samplingdates show trend in Figure 3 the semivariance value is notstabilized around variance value of data and the universalkriging was used to construct the maps of spatial variabilityof these variables

The fitted variogram parameters (Table 5) show that thespherical model was the fitted model to the properties understudy according to Cambardella et al [19] Goovaerts [18]Vieira [17] and Siqueira et al [27] describing this modelit is usually best fitted to the properties of soil and plantAll attributes had low values of nugget effect (119862

0) Range

values (a) varied from 4000m (log ECa-H Residual342008

)to 13000m (clay silt and soil water content) The degreeof spatial dependence between samples was high (SD le2500) across the study the exception being log ECa-H Residual

1432008which presented a moderate degree of

spatial dependence (SD = 3167)The spatial variability maps obtained with universal

kriging (Figure 5) show that there is a similarity between themaps ECa-V (Figures 5(a) 5(c) and 5(e)) and ECa-H (Figures5(b) 5(d) and 5(f)) on several sampling dates with furtherdifferentiation of maps of ECa-V and ECa-H on themeasureddata on 2362008 (Figures 5(e) and 5(f))when thewater tablelevel was lower

It is observed that the maps of ECa-V and ECa-H(Figure 5) and the map of the water content in soil (Fig-ure 6(e)) obtained with ordinary kriging (Figure 6(e)) looksimilar following the same pattern of digital elevation model(Figure 1(b))

The map of the electrical conductivity of the saturationextract (ECe Figure 6(a)) shows inverse behaviour to mapsECa-V and ECa-H (Figure 5) Moreover the map of spatialvariability of clay content in the study area (Figure 6(b))shows no similarity to maps ECa-V and ECa-H (Figure 5)this fact is also repeated with silt (Figure 6(c)) and sand(Figure 6(d))

Table 6 presents the fitting parameters cross-variogrambetween 120579119892 times ECa-V (119903 = 0685) and 120579119892 times ECa-H (119903 = 0648)The cross-variograms were fitted to a spherical model withthe same range compared to single variograms to obtain alinear model coregionalization (Table 6)

The spatial variability maps constructed using ordinaryand universal cokriging (Figure 7) demonstrate that theuse of the soil apparent electrical conductivity measured byelectromagnetic induction (ECa-V and ECa-H) on 2362008was a secondary variable that improves the estimation of thesoil water content using cokriging This improvement in theestimation of 120579119892 can be observed in Table 7 where it showed

an increase in the value of the correlation coefficient betweenthe measured and the estimated values from cross-validationusing ordinary cokriging with log-ECa-V (0746) and withlog-ECa-H (0756) as secondary variables with respect to useof ordinary kriging (0637) Moreover in the case of the soilwater content map obtained with ordinary cokriging using assecondary data ECa-H (Figure 7(b)) is less smooth than themap obtained with ordinary kriging (Figure 6(e))

4 Conclusions

When taking into account all the soil properties studied ECaand gravimetric soil water content measured at the samedate that is 2362008 showed the highest coefficients ofcorrelationTable 4Moreover ECa showedhigher coefficientsof correlation to clay and silt content than to silt contentand the strength of the correlation was higher for the firstECa recording date that is 1432008 when the soil moisturewas lower Thus coefficient of correlation of ECa with siltand clay content showed a trend to increase the soil moisturedecreased this result suggests the usefulness of recording ECaon successive dates with different soil water contents

The spatial patterns of spatial variability of the logarith-mic values of apparent soil electrical conductivity (ECa) andthe electrical conductivity of the soil saturated paste (ECe)were modeled by universal kriging whereas those of sandclay silt and gravimetric water content were modeled byordinary kriging The use of cokriging with ECa data as sec-ondary variable improved the estimation of the gravimetricsoil water content with respect to the use of kriging

Abbreviations

ECa Apparent soil electrical conductivityECa-V CEa in vertical dipoleECa-H CEa in horizontal dipoleECe Electrical conductivity of soil saturation extract

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper The referencingbrands of commercial products in the paper do not imply thatthe authors recommend the equipment utilized in this studyThe support funds presented by the Development Agenciesdo not show conflict of interests regarding the publication ofthis paper

Acknowledgments

The authors are grateful to the Ministerial de AsuntosExteriors y de Cooperation (MAEC-AECID) from Spain forthe granting of scholarships for PhD studies This workhas been funded by Ministerial de Education y Cienciawithin the framework of research Project CGL2005-08219-C02-02 and cofunded by Xunta de Galicia within theframework of research Project PGIDIT06PXIC291062PNand by the European Regional Development Fund (ERDF)The authors acknowledge the provincial farm Gayoso Castro

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

The Scientific World Journal 7

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)1432008

(a)

NEW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-H (mS mminus1)1432008

(b)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

NEW

S

Log ECa-V (mS mminus1)342008

(c)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

NEW

S 6

9

12

15

Log ECa-H (mS mminus1)342008

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

Log ECa-V (mS mminus1)2362008

(e)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

6

9

12

15

2362008Log ECa-H (mS mminus1)

(f)

Figure 5 Krigingmaps of apparent soil electrical conductivity (ECa-V and ECa-H) from the continuous recordsmade in 1432008 342008and 2362008

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

8 The Scientific World Journal

N

EW

S622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

7

13

19

25

ECe (mS mminus1)

(a)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

120

150

180

210

Distance X (m)

Dist

ance

Y(m

)

Clay (g kgminus1)

(b)

N

EWS

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

230

265

300

335

Distance X (m)

Dist

ance

Y(m

)

Silt (g kgminus1)

(c)

N

EW

S622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

485

515

545

575

Distance X (m)

Dist

ance

Y(m

)Sand (g kgminus1)

(d)

622550 622600 622650 622700 622750 622800

4779650

4779700

4779750

4779800

4779850

4779900

Distance X (m)

Dist

ance

Y(m

)

Soil water content (120579g )

10

20

30

40

(e)

Figure 6 Kriging maps of the soil properties analyzed

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

The Scientific World Journal 9

Table4Linear

correla

tionmatrix

betweenthec

ontin

uouslyrecorded

ECadatasetsandthes

oilpropertiesa

nalyzed

14032008

03042008

23062008

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

LogEC

a-VLo

gEC

a-H

ECe

Clay

Silt

Sand

120579119892

14032008

LogEC

a-V10

0014032008

LogEC

a-H

0780

1000

03042008

LogEC

a-V0972

0759

1000

03042008

LogEC

a-H

0724

0796

0792

1000

23062008

LogEC

a-V0861

0729

0855

0688

1000

23062008

LogEC

a-H

0541

064

40515

0569

0751

1000

23062008

ECe

0127

0185

minus0012

minus0141

0156

0224

1000

23062008

Clay

0344

0495

0252

0197

0253

0346

0221

1000

23062008

Silt

minus0247

minus0423

minus0172

minus0216

minus0137

minus0228

minus0145minus0773

1000

23062008

Sand

minus0086

minus0012

minus0076

0065

minus0133

minus0119

minus0076minus0170minus0494

1000

23062008

120579119892lowast

lowastlowast

lowast0685

064

80005

0221minus0145minus0075

1000

lowast

Values

weree

xcludedin

linearc

orrelationanalysisbecausethe

soilwater

contentw

ason

lymeasuredon

236200

6andforthisreasonthec

oefficientsof

correlationwe

reno

tcalculatedbetweensoilwater

content

andapparent

electric

alcond

uctiv

ityof

thes

oil(EC

a-VandEC

a-H)for

othersam

plingdates(1432008and0342008)

lowastlowast

Tocorrela

tethemeasuremento

flog

ECa-V

andlogE

C a-H

forthe

measurementd

ates

(1432008and342008)w

iththemeasurementsmadeon

2362008itwas

necessaryto

estim

atethemeasurementsof

logE

C aE

C eclay

siltand

sand

contentu

singkrigingin

locatio

nsmeasuredon

2362008

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

10 The Scientific World Journal

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-V

(a)

622550 622600 622650 622800622700 622750

Distance X (m)

Dist

ance

Y(m

)

4779900

4779850

4779800

4779750

4779700

4779650

10

20

30

40

120579g times log ECa-H

(b)

Figure 7 Map of gravimetric water content 120579119892 obtained by the universal cokriging

Table 5 Fitted semivariogram parameters and respective models of the continuously recorded ECa data sets and the soil properties analyzed

Date Variable Geostatistical method Model 1198620

1198621

119886 SD

14032009 Log ECa-V residual UK Spherical 00001 314 10500 000Log ECa-H residual UK Spherical 014 0302 4400 3167

03042008 Log ECa-V residual UK Spherical 000 510 14500 000Log ECa-H residual UK Spherical 010 032 4000 2380

23062008

Log ECa-V residual UK Spherical 0001 001 13000 909Log ECa-H residual UK Spherical 0001 0005 13000 196

ECe residual UK Spherical 000025 00018 10000 909Clay OK Spherical 0001 106000 13000 000Silt OK Spherical 000 140000 13000 000Sand OK Spherical 000 51000 7000 000120579119892 OK Spherical 0001 6000 13000 000

UK universal kriging OK ordinary kriging 1198620 nugget effect 119862

1 structural variance 119886 range (m) and SD spatial dependence ()

Table 6 Fitted cross-semivariogram models and respective parameters between gravimetric water content (principal variable) and log ECa(secondary variable)

Variable Geostatistical method Model 1198620

1198621

119886 (m)2362008 120579119892 times log ECa-V Universal cokriging Spherical 100 2000 130002362008 120579119892 times log ECa-H Ordinary cokriging Spherical 000 1500 130001198620 nugget effect 119862

1 structural variance and 119886 range (m)

Table 7 Correlation coefficients between measured gravimetricwater content and data estimated by kriging and cokriging

120579119892 (ordinary kriging) 0637

120579119892 times log ECa-V(universal cokriging) 0746

120579119892 times log ECa-H(ordinary cokriging) 0756

of the Deputation of Lugo for allowing the use of theirfacilities to carry out this work The authors thank theCNPq (National Council for Scientific and TechnologicalDevelopment (Brazil)) and FACEPE (Fundacao de Amparo a

Ciencia e Tecnologia do Estado de Pernambuco (Brazil)) andthey also thank CNPq for the scholarship DCRmdashRegionalScientific Development awarded to the first author Alsothanks are given to FAPEMA MA Brazil for funding thepublication of this paperThe authors would like to thank twoanonymous reviewers for the comments that undoubtedlyimproved the quality of this paper

References

[1] D L Corwin and J D Rhoades ldquoMeasurement of inverted elec-trical conductivity profiles using electromagnetic inductionrdquo

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)

The Scientific World Journal 11

Soil Science Society of America Journal vol 48 no 2 pp 288ndash291 1984

[2] J D McNeill ldquoElectrical conductivity of soils and rocksrdquoTechnical Note TN-5 Geonics Limited Ontario Canada 1980httpwwwgeomatrixcouktoolsapplication-notestn5pdf

[3] K A Sudduth N R Kitchen W J Wiebold et al ldquoRelatingapparent electrical conductivity to soil properties across thenorth-central USArdquo Computers and Electronics in Agriculturevol 46 no 1 pp 263ndash283 2005

[4] D L Corwin and S M Lesch ldquoCharacterizing soil spatialvariability with apparent soil electrical conductivity I SurveyprotocolsrdquoComputers and Electronics in Agriculture vol 46 no1ndash3 pp 103ndash133 2005

[5] J Kuhn A Brenning M Wehrhan S Koszinski and MSommer ldquoInterpretation of electrical conductivity patterns bysoil properties and geological maps for precision agriculturerdquoPrecision Agriculture vol 10 no 6 pp 490ndash507 2009

[6] C K Johnson J W Doran H R Duke B J Wienhold K MEskridge and J F Shanahan ldquoField-scale electrical conductivitymapping for delineating soil conditionrdquo Soil Science Society ofAmerica Journal vol 65 no 6 pp 1829ndash1837 2001

[7] S M Lesch D J Strauss and J D Rhoades ldquoSpatial predictionof soil salinity using electromagnetic induction techniques 1Statistical prediction models a comparison of multiple linearregression and cokrigingrdquoWater Resources Research vol 31 no2 pp 373ndash386 1995

[8] J W van Groenigen W Siderius and A Stein ldquoConstrainedoptimisation of soil sampling for minimisation of the krigingvariancerdquo Geoderma vol 87 no 3-4 pp 239ndash259 1999

[9] SM Lesch J D Rhoades andD L Corwin ldquoTheESAP version201r user manual and tutorial guiderdquo Research Report SalinityLaboratory Riverside Calif USA 2000

[10] B Minasny A B McBratney and D J J Walvoort ldquoThevariance quadtree algorithm use for spatial sampling designrdquoComputers and Geosciences vol 33 no 3 pp 383ndash392 2007

[11] Fao-Isric ldquoWorld reference base for soil resourcesrdquo Roma yWageningen p 161 2014 httpwwwfaoorgdocrepw8594ew8594e00HTM

[12] A C Gegunde and F Diaz-FierrosOs solos da Terra Cha Tiposxenese e aproveitamento Publicacion Diputacion Provincial deLugo Lugo Spain 1992

[13] X X Neira Seijo Desenrolo de tecnicas de manexo de augaaxeitadas a umuso racional de regadıos [PhD thesis] USCEPS1993

[14] Geonics EMD38-DD Ground Conductivity Meter-Dual DipoleVersion Geonics Ontario Canada 2005

[15] O A Camargo A C Moniz J A Jorge and J M A SValadares ldquoMethods of chemical mineralogical and physicalanalyses of soilsrdquo Technical Bulletim of the Agronomic Instituteof Campinas Campinas no 106 p 94 1986 (Portuguese)

[16] USDA Soil Quality Test Kit Guide USDA Washington DCUSA 1999 (Spanish) httpwwwnrcsusdagovInternetFSEDOCUMENTSstelprdb1044786pdf

[17] S R Vieira ldquoGeoestatıstica em estudos de variabilidade espacialdo solordquo in Topicos em Ciencia do solo R F Novais V HAlvarez and G R Schaefer Eds vol 1 pp 1ndash54 SociedadeBrasileira de Ciencia do Solo Vicosa Brazil 2000

[18] P Goovaerts Geostatistics for Natural Resources EvaluationOxford University Press New York NY USA 1997

[19] C A Cambardella T B Moorman J M Novak et al ldquoField-scale variability of soil properties in central Iowa soilsrdquo Soil

Science Society of America Journal vol 58 no 5 pp 1501ndash15111994

[20] E J Pebesma ldquoGstat userrsquos manualrdquo Department of Physi-cal Geography Utrecht University 2014 httpwwwgstatorggstatpdf

[21] P Goovaerts ldquoGeostatistics in soil science state-of-the-art andperspectivesrdquo Geoderma vol 89 no 1-2 pp 1ndash45 1999

[22] C V Deutsch and A G Journel GSLIB-Geostatistical SoftwareLibrary and Userrsquos Guide Oxford University Press New YorkNY USA 2nd edition 1998

[23] J-P Chiles and P Delfiner Geostatistics Modeling SpatialUncertainty Wiley Series in Probability and Statistics AppliedProbability and Statistics Wiley-Interscience New York NYUSA 1999

[24] A W Warrick and R R Nielsen ldquoSpatial variability of soilphysical properties in the fieldrdquo in Applications of Soil PhysicsD Hillel Ed Academic Press New York NY USA 1980

[25] D Grandjean I Cousin M Seger et al From GeophysicalParameters to Soil Characteristics FP7mdashDIGISOIL ProjectDeliverable D21 2009 httpeusoilsjrceceuropaeuprojectsDigisoilDocumentsDigisoil-D21pdf

[26] G Martınez and K Vanderlinden ldquoAnalisis de la relacion espa-cial entre la humedad gravimetrica del suelo y la conductividadelectrica aparenterdquo in Tendencias Actuales de la Ciencia delSuelo N Bellinfante andA Jordan Eds pp 29ndash36Universidadde Sevilla Sevilla Spain 2007

[27] GM Siqueira S R Vieira andM B Ceddia ldquoVariability of soilphysical attributes determined by differentmethodsrdquoBragantiavol 67 no 1 pp 203ndash211 2008 (Portuguese)