Next Article in Journal
Inversion of Glycyrrhiza Chlorophyll Content Based on Hyperspectral Imagery
Next Article in Special Issue
Effect of Soil pH on the Uptake of Essential Elements by Tea Plant and Subsequent Impact on Growth and Leaf Quality
Previous Article in Journal
QTL Mapping for Bacterial Wilt Resistance in Eggplant via Bulked Segregant Analysis Using Genotyping by Sequencing
Previous Article in Special Issue
Continuous Intercropping Increases the Depletion of Soil Available and Non-Labile Phosphorus
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agronomy
Volume 14
Issue 6
10.3390/agronomy14061161
agronomy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Challenges in Mapping Soil Variability Using Apparent Soil Electrical Conductivity under Heterogeneous Topographic Conditions

by
István Mihály Kulmány
1,2,*,
László Bede
1,2,
Dávid Stencinger
1,2,
Sándor Zsebő
1,2,
Péter Csavajda
3,
Renátó Kalocsai
4,
Márton Vona
5,
Gergely Jakab
6,
Viktória Margit Vona
4 and
Ákos Bede-Fazekas
7,8
1
Agricultural and Food Research Centre, Széchenyi István University, 9200 Mosonmagyaróvár, Hungary
2
Department of Plant Sciences, Albert Kázmér Faculty of Agricultural and Food Sciences in Mosonmagyaróvár, Széchenyi István University, 9200 Mosonmagyaróvár, Hungary
3
Department of Logistics and Forwarding, Audi Hungaria Faculty of Vechicle Engineering, Széchenyi István University, 9026 Győr, Hungary
4
Department of Water Management and Natural Ecosystems, Albert Kázmér Faculty of Agricultural and Food Sciences in Mosonmagyaróvár, Széchenyi István University, 9200 Mosonmagyaróvár, Hungary
5
Csernozjom Ltd., 5065 Nagykörű, Hungary
6
HUN-REN Research Centre for Astronomy and Earth Sciences, Geographical Institute, 1052 Budapest, Hungary
7
HUN-REN Centre for Ecological Research, Institute of Ecology and Botany, 2163 Vácrátót, Hungary
8
Department of Environmental and Landscape Geography, Institute of Geography and Earth Sciences, ELTE Eötvös Loránd University, 1117 Budapest, Hungary
*
Author to whom correspondence should be addressed.
Agronomy 2024, 14(6), 1161; https://doi.org/10.3390/agronomy14061161
Submission received: 4 April 2024 / Revised: 26 May 2024 / Accepted: 27 May 2024 / Published: 29 May 2024
(This article belongs to the Special Issue Advances in Soil Fertility, Plant Nutrition and Nutrient Management)
Browse Figures
Versions Notes

Abstract

:
Site-specific management requires the identification of treatment areas based on homogeneous characteristics. This study aimed to determine whether soil mapping based on apparent soil electrical conductivity (ECa) is suitable for mapping soil properties of fields with topographic heterogeneity. Research was conducted on two neighbouring fields in Fejér county, Hungary, with contrasting topographic heterogeneity. To characterise the spatial variability of soil attributes, ECa was measured and supplemented by obtaining soil samples and performing soil profile analysis. The relationship between ECa and soil physical and chemical properties was analysed using correlation, principal component, and regression analyses. The research revealed that the quality and strength of the relationship between ECa and soil remarkably differed in the two studied fields. In homogeneous topographic conditions, ECa was weakly correlated with elevation as determined by soil physical texture and nutrient content in a strong (R2 = 0.72) linear model. On the other hand, ECa was significantly determined by elevation in heterogeneous topographic conditions in a moderate (R2 = 0.47) linear model. Consequently, ECa-based soil mapping can only be used to characterise the soil, thus delineating management zones under homogeneous topographic conditions.

1. Introduction

Hungary is situated in the Carpathian Basin. Arable land accounts for 59% (5.3 million ha) of the total territory of Hungary, with uniformly heterogeneous soil both spatially and temporally due to an array of soil-forming factors. These conditions were similar to our research fields, which are located on the Great Plain surrounded by hills in the southern part and highlands in the west and north. The surface elevation varies between 100 and 150 m and the loess layer thickness is between 15 and 20 m [1]. Typical soil-forming factors include slow land movement and water erosion [2].
Proximal soil sensing (PSS) technologies, mounted on diverse sensing platforms, are being developed to deliver cost-effective, high-density data for field variability. In recent decades, both passive and active PSS systems have significantly enhanced our understanding of soil–topography interactions and spatial variability in precision agriculture [3]. Geoelectrical and electromagnetic sensors are extensively used to determine soil dielectric properties and geospatial variability [4]. On-the-go PSS sensors powered by electromagnetic energy (such as DUALEM-21 and EM-38) have been instrumental in guiding soil management practices in precision farming. Despite this, most sensors record changes in parameters that indirectly influence agronomic conditions rather than direct crop growth indicators [5,6]. Research by Corwin and Lesch [7] and Friedman [8] revealed that the exact locations and depths of apparent soil electrical conductivity (ECa) measurements are strongly correlated with the physical properties of topsoil and subsoil, such as clay layer depth, soil salinity, and water content. Nevertheless, ECa data from varying depths require further processing for site-specific depth exploration before linking these measurements to specific soil constituents [9].
Soil properties can vary considerably among sampling sites due to the combined effects of soil physical, chemical and biological processes at different spatiotemporal scales [10,11]. Soil heterogeneity can be observed within a soil map unit and within a small area of seemingly uniform soil [12].
There are different methods for exploring soil heterogeneity. The first way is using traditional soil maps, typically drawn on a scale between 1:20,000 and 1:200,000. These maps were designed for a regional scale; therefore, they are not suitable for providing detailed information about within-field variability [13]. The second method is soil inventory based on intensive soil sampling, including soil profile description. Although this is an accurate method of exploring soil heterogeneity at a detailed scale, it is also extremely costly and labour-intensive [14]. The third method involves on-the-go sensor technologies to collect information by measuring apparent soil electrical conductivity (ECa) at numerous sites in the field. Management zone-based sampling strategies carried out by soil electrical conductivity sensors are widely used [15] to identify soil properties in precision agriculture [16]. According to Gonçalves et al., 2021 [17], soil sampling oriented by apparent soil electrical conductivity considers soil heterogeneity just as properly as elevation- or grid-based soil sampling methods. Thus, this method provides an adequate, environmentally and economically viable tool to map the distribution of soil properties.
Despite intensive soil inventory, directed soil sampling based on geo-referenced measurements of ECa is a proven and robust tool for characterising the spatial variability of edaphic and anthropogenic properties influencing ECa [18,19,20,21,22,23]. Accordingly, different strong relationships were observed between ECa and several soil attributes, such as salinity [24,25], water content [25,26], texture [27,28], bulk density [29], organic matter [7,30], CEC [20,31,32], CaCO3 content [33], heavy metals [34], pH [35] and soil map unit boundaries [36], and elevation [37].
Different relationships were found between ECa and soil properties in previous studies, but delineating homogeneous zones is still a challenging procedure. To facilitate the ECa-based homogenous zone delineation process, Corwin and Lesch (2003, 2005, 2013) [7,18,38] and Corwin and Scudiero (2016) [20] developed a guideline and protocol. However, the agricultural scientific literature is replete with studies that have failed to follow this guideline because of the inconsistent spatial and temporal relationship observed between ECa and soil properties [39]. This oversight has led to the application of different statistical and delineation methodologies [40,41,42,43,44]. Soil topography is important in the spatial distribution of soil properties and is well-studied. In heterogeneous terrain, however, there is limited information about the spatial relationship between ECa and soil properties that focuses on one or two specific areas of agricultural fields [45,46]. Therefore, the general aim was to fill this research gap by studying the usability of ECa for soil mapping in two neighbouring fields differing only in topographic heterogeneity. The objective of this study was to determine whether heterogeneous topographic conditions can affect or even hinder the usability of on-the-go ECa measurements for characterising soil and, thus, the establishment of management zones.

2. Materials and Methods

2.1. Experimental Sites

The research was conducted in two neighbouring agricultural fields (15 ha and 48 ha) in Fejér county, Hungary. The surrounding territory of agricultural fields is under agricultural production and located in hilly terrain. According to Marosi and Somogyi (1990) [47], this region is characterized by a temperate climate with a mean annual temperature of 10.2–10.4 °C. The annual average rainfall is 530–560 m.
The T1 field (48 ha) is cultivated conventionally, whereas the T2 field (15 ha) has been cultivated under a no-tillage system since 2019. In the T2 field, maize (Zea mays L.) was sown in 2017 and 2018, and peas (Pisum sativum L.) and daikon (Raphanus sativus L. var. longipinnatus Bailey) in 2019. In the T1 field, wheat (Triticum aestivum L.) was sown in 2017 and 2019 and sunflower in 2018 (Helianthus annuus L.). The typical soil type was Luvic for the T1 field and Cambisol for the T2 field. The higher points were eroded due to elevation differences in the T1 field. Therefore, the humus layer was shallow (30–40 cm). The soil was accumulated in deeper parts of the field with a humus layer of 90 cm. There were minor elevation differences, but the soil was compacted in some parts of the T2 field. The soil moisture content was evenly distributed between 21.6% ± 1.7% and 22.1% ± 2.7% in the T1 and T2 fields at the time of ECa measurements. Since the moisture content was close to the field capacity, the soil’s apparent electrical conductivity was not significantly influenced by the soil moisture content [48,49]. Thus, this attribute did not reduce the reliability of the ECa in explaining physical and chemical soil attributes.

2.2. Field Measurements

2.2.1. In Situ Data Collection

The research framework, including measurements, sampling design, soil sampling and analyses, is summarized in Figure 1. Georeferenced soil ECa (ECa-field) and elevation mapping were carried out on 22 October 2019 with a Veris U3 sensor system (Veris Technologies, Inc., Salina, KS, USA). The sensor system uses an electrical array to map the soil rooting profile within 10 m of working width and 0–60 cm deep. The Veris U3 sensor was pulled across each field behind a utility terrain vehicle (UTV). The travel speed ranged between 7 and 11 km/h, corresponding to about 2–3 m spacing between measurements in the direction of travel.

2.2.2. Soil Sampling and Analysis

Soil sampling was based on the spatial distribution of three ECa classes supplemented by investigations into soil heterogeneity with a Pürckhauer type soil core sampler (Bürkle GmbH, Bad Bellingen, Germany). According to Peralta et al. (2013) [50], using more than three ECa classes does not increase available information. Soil sampling at 15 points with a Pürckhauer type soil core sampler [51] was used as a preliminary investigation of soil type, humus layer depth at different elevations, colour, physical characteristics, carbonate content, pH, and moisture. It was a relatively short, quick survey of up to 1 m depth to determine the field’s soil heterogeneity without significantly disturbing soil, which helped in selecting the optimum soil sampling points in the research field. The sampling protocol described by [50] was followed. Three or more soil samples were obtained from each field per zone for further laboratory analyses when justified by the spatial extent of the ECa zones and heterogeneity according to the Pürckhauer type soil core sampler. In other cases, two samples were selected. Accordingly, 17 georeferenced soil samples from depths of 0–30 cm were obtained (Figure 2A). Moreover, undisturbed soil samples were collected from depths of 0–10 cm to express the gravimetric water content of the soil in a mass of water per mass of dry soil (m/m%) (Table 1).
At each sampling point, the thickness of the humus layer was determined according to the elevation of the sampling point. The following parameters were measured by laboratory analyses based on Hungarian standards [52]:
  • Laboratory ECa (ECa-lab);
  • Nitrate, nitrite (KCl soluble);
  • P2O5 (ammonium–lactate soluble);
  • K2O (ammonium–lactate soluble);
  • Sulphate (KCl soluble);
  • Magnesium (KCl soluble);
  • Sodium (KCl soluble);
  • Copper EDTA (KCl-EDTA soluble);
  • Manganese (KCl-EDTA soluble);
  • Zinc (KCl-EDTA soluble);
  • CaCO3 (Scheibler volumetric method);
  • Soil organic matter—SOM (wet combustion, Turin method);
  • Salt (water-soluble);
  • pH (KCl, potentiometric method;
  • Arany-type texture;
  • Particle size distribution (Laser diffractometry, Fritsch Analysette 22 Microtech Plus).
At the representative points (P5, P6, P7, and P11), two soil profiles per field were excavated to precisely determine soil type in its natural state.

2.3. Database Setup

ECa-field data were pre-processed in ArcGIS 10.8 (ESRI). Data collected from measurement equipment were cleared of extreme outliers and interpolated using the inverse distance weighted interpolation technique. To establish the dataset, soil sampling data were matched to ECa values by averaging all ECa measurements from the transect portion to within a 10 m radius of the centre-point location at which the soil cores were collected.

2.4. Statistical Analyses

A cascading framework of statistical analyses was established for estimating the relationship between soil properties and apparent electrical conductivity (ECa) and the consistency of the found relationship (Figure 1). Beyond simple descriptive statistics, the framework consisted of three complementary statistical methods to capture different statistical features of the dataset: correlation analysis, ordination with principal component analysis, and linear regression. All statistical analyses were carried out using R statistical software (version 4.3.1) [53] and its packages ‘car’ [54], ‘corrplot’ [55], and ‘vegan’ [56].
In the first step, a decision was made whether the T1 and T2 fields should be analysed in subsequent steps together or separately. An informed decision was reached by following the four substeps: (1) topographic variation analysis, (2) descriptive statistics, (3) ordination and (4) analysis of variance on ordination. Accordingly, (1) topographic variation was measured [57] using the Topographic Ruggedness Index (TRI) according to Wilson et al. (2007) [58] in a moving window of 3 × 3 cells (10 × 10 m) on the fine resolution elevation raster. The resulting TRI raster was then averaged for the T1 and T2 fields. In the subsequent analyses, sampling points were studied instead of the fine resolution raster; (2) soil properties, including the level of nutrient concentration, were described with the following descriptive statistical indicators: arithmetic mean, median, standard deviation (SD), coefficient of variation (CV), maximum (max) and minimum (min) values, and range; (3) principal component analysis (PCA) was used to reduce the dimensions of the collinear variables and facilitate visual investigation of sample separation in the two-dimensional ordination space. PCA is an ordination method for orthogonal, or non-correlated, axes called principal components (PCs) to which the original independent variables are directly related. PCA was first introduced by Pearson (1901) [59] and developed independently by Hotelling (1933) [60]; (4) Type II multivariate analysis of variance (MANOVA) with Pillai test statistics was conducted on the first five PCs to provide statistical support for the visual investigation.
Based on the results of the first step, the T1 and T2 fields were treated separately in the second, third, and fourth steps of the analysis. In the second step, Pearson correlation analysis of the sampling points was used to determine the relationship between apparent electrical conductivity (ECa) and other continuous variables, i.e., elevation, humus layer thickness, and soil properties in each field. Moreover, the correlation analysis supported whether (1) the independent soil variables could be used later for regression analysis of ECa or, (2) prior to the regression analysis, the independent but correlated soil variables should be aggregated into new, non-correlated proxy variables (i.e., principal components). In the third step, PCA was applied due to the collinearity revealed by correlation analysis.
Two PCAs were run for the two fields separately (in contrast to the PCA step run on the full dataset in the first step). According to Sharma (1996) [61] and Peralta and Costa (2013) [50], PCs with an eigenvalue greater than 1 were selected for interpretation and further analysis since they explained a significant amount of the variance present in soil properties across each field.
In the fourth step, linear regression was used to determine which PCs of the ordination of soil properties were more associated with ECa-field (Figure 1).

3. Results and Discussion

3.1. Comparison of the Two Fields

Soil mapping managed by the Pürckhauer type soil core sampler has proven suitable for rough separation of different soil types and layer depths [62]. The soil was classified according to Hungarian soil classifications, a genetic system based on the appearance and strength of soil formation processes in the profile [63,64]. The soils in the fields are long cultivated, and erosion and compaction are visible. In the T1 field, based on the profile description of the excavated pit, the typical soil type was Luvic. These soils have high base saturation and thick, dark, mollic horizons. They commonly form from loess or loess-like parent material. The parent material was sandy loess. The soil in the higher part of the field was eroded (Figure 3). At sampling point P6, the dark humus layer was shallower (25 cm) than at sampling point P5, where the dark fertile layer was almost 80 cm deep.
In the T2 field, the typical soil type was Cambisol. These soils represent transitions with Chernozems and correlate with Luvic Chernozems [65]. The parent material was alluvial sediment mixed with loess. The higher part of this field was also eroded due to elevation differences. Differences were observed in the thickness of the fertile dark humus layer and the soil texture of this field. On the west side of the field, the soil was more clayey and compact than the loamy east side. Figure 4 shows the two excavated soil profiles in the T2 field. Sampling point P7 had a deep humus layer (80 cm), whereas sampling point P11 had a shallower fertile humus layer (45 cm). The differences in soil properties are also visible in the soil colour. While the soil colour in the T1 field is darker, the T2 field is yellower due to the eroded surface mixed with the humus layer.
The mean topographic ruggedness indices of the T1 and T2 fields were 0.084 and 0.070, respectively, indicating that the T1 field is remarkably more heterogeneous in topography than the T2 field. Table 1 shows the descriptive statistics results. Elevation and ECa-field values differed between the two fields. ECa-field values varied between 147.09 and 552.15 mS/cm in the T2 field. By contrast, ECa-field values in the T1 field ranged between 114.38 and 321.35 mS/cm. Consequently, the coefficient of variation (CV) in the T2 field (0.43) was also higher than that in the T1 field (0.31). The T2 field (mean: 176 m) is situated higher than the T1 field (mean: 169 m). A west-to-east downward slope is found in the T2 field while the T1 field has heterogeneous hilly terrain (Figure 2B).
The following parameters showed high variability in the T1 field: sand, salt, CaCO3, Mg, Na, P2O5, Cu, Mn, and Zn. In the case of the T2 field, ECa, sand, CaCO3, N, Na, P2O5, Cu, Mn, and Zn had high variability.
CVs for soil properties indicated high spatial variability. High spatial variability in soil properties is due to the interaction of (1) soil formation processes, (2) meteorological processes, and (3) anthropogenic influences. Soil formation processes result from complex interactions between biological, physical, and chemical mechanisms acting on parent material over time and influenced by topography [66].
PCA conducted on the full dataset (i.e., both fields) revealed a clear separation of samples in the ordination space (Figure 5). The first two PCs explained 86.29% of the variance, while the first five PCs used in the MANOVA explained almost all variance (99.18%). The MANOVA found significant differences (p < 0.01) between the two fields (approximately F(5, 10) = 10.09). Hence, an informed decision was made to carry out all subsequent statistical analyses separately for the two fields.

3.2. The Pearson Correlation Analysis—Relationships among ECa-field and Soil Properties

Pearson’s correlation analyses on the sampling points showed different results between the two fields (Figure 6). The correlation between soil properties and on-the-go measured ECa (ECa-field) was weak in the T1 field. A moderate positive correlation was found between the ECa-field and phosphorus pentoxide (r = 0.56). A moderate negative correlation was attributed to ECa-lab (r = −0.54) and a moderate positive correlation to salt (r = 0.51), which is in agreement with previous studies [50,67,68]. The correlation between ECa-field and elevation was strong (r = −0.75) in the T1 field, whereas a negative weak correlation (r = −0.06) was observed between ECa-field and elevation in the T2 field. The analyses revealed considerable correlations between ECa-field and soil parameters in the T2 field. A strong positive correlation was observed between ECa-field and silt (r = 0.74), strong negative with sand (r = −0.72), moderately positive with clay (r = 0.53), and moderately negative with nitrogen (r = −0.62). Rhoades et al. (1976) [25] reported a strong correlation between ECa and soil texture, while the effect of nutrient content on soil ECa was demonstrated previously [32,69]. Moore et al. (1993) [70] found a close relationship between topography and some nutrient content (SOM, pH, and P content). The relationship between ECa and physical soil types is well known [25], but their correlation varies from field to field [28,39,45]. Similar to our experiment, Carroll and Oliver (2005) [71] also found a moderate to strong correlation between ECa and soil textural fractions (sand, silt, and clay). Heil and Schmidhalter [72] and Nyéki et al. [73] further emphasised the role of ECa measurement in estimating clay content in the field. These findings were also confirmed by Broge et al. [74], who measured apparent soil electrical conductivity with an EM38 sensor in Denmark. At the same time, Rezaei et al. [75] found a weak correlation (r = 0.16) between ECa measured by DUALEM-21S and clay content in Canada. In terms of soil nutrients, the correlation analysis did not confirm whether nitrogen, phosphorus, or several meso- (Mg) and microelements (Mn, Zn, and Cu) in the soil were correlated with apparent soil electrical conductivity [50,69,76,77]. However, Serrano et al. [78] revealed that the correlation between ECa measured by the DUALEM sensor and soil macronutrient content was weak or moderate, depending on the rainfall season in Portugal. The weak correlation between humus layer thickness and ECa-field for both experimental fields contradict the findings of previous studies [79,80].

3.3. Principal Component and Regression Analyses

PC loading values, eigenvalues, and cumulative variance for PCs in the T1 and T2 fields are shown separately in Table 2. In both fields, the absolute value of factor loadings for soil properties higher than 0.3 was used to evaluate their role and influence on variability.
After performing PCA, the first three components with eigenvalues greater than 1 were retained in the T1 field, explaining 93% of the total variance. The first PC (PC1) had a positive factor loading for pH (KCl) and CaCO3, but was negative for Mg, Cu, and Mn. The second PC (PC2) was repeatedly related to pH (KCl), but here it had a negative factor loading. In PC2, binding capacity had a negative loading, while salt had a positive loading. The factor loadings most associated with the third component were SOM and Clay. SOM had the highest negative factor loading of all components.
Four components (PCs) with an eigenvalue greater than 1 were selected in the T2 field, explaining 92% of the total variance. The first PC (PC1) had negative loading for CaCO3 and positive loadings for Cu, Mn, and Zn. The second PC (PC2) had positive factor loading for clay and silt and negative factor loading for sand. In Peralta and Costa’s study (2013) [50], sand content was also a factor loading, but it was negative and the low ECa-field was associated with soil lighter texture. In this study, ECa was mostly observed in eroded areas where basic bedrock, fluvial deposits, and a bit of sandy loess came closer to the surface. Humus layer thickness, CaCO3 N, Mg, S, and Na contributed to the third component (PC3). The fourth PC was strongly and positively influenced by binding capacity, humus layer thickness, and CaCO3. Carbonate content had a strong factor loading in all three PCs, consistent with results by other researchers [33,50].
Generally, the PCA results were consistent with the correlation analyses. However, the variables revealed different patterns. According to Reyes et al. (2019) [81] variables contribute to particular processes and their correlations vary based on different field and soil conditions. The highest factor loadings for the first component were reported for different variables in each study: pH, slope [5], and ECa [82] in Northwestern Europe; elevation and soil depth in Argentina [83]; and SOM and available N [84] in Eastern China.
Based on the principal component analysis, linear regression models were carried out to determine the overall impact of PCs on ECa-field measurements. The first three PCs explained 47.0% of the variation based on linear regression analysis of the T1 field (Table 3). The first four PCs in the T2 field explained 72.1% of the variation. The difference in the explanatory power of the two models is similar if adjusted R2 is studied: −0.059 and 0.442, respectively.
According to Heiniger et al. (2003) [69], the regression model that explains the greater portion of ECa variability indicates situations where ECa could be used to estimate soil properties and nutrient levels. Only the topographically homogeneous T2 field model met this finding, in which the ECa-field was strongly linked to soil physical properties (sand, silt, and clay content), indicating that the ECa measurements in this field were driven primarily by soil texture. Although it was correlated with some chemical properties (N, Cu, and Mn), it was only weakly correlated with SOM. The relationship between ECa and soil physical properties was consistent with findings in other studies [50,85,86].
In the T1 field, where topographic heterogeneity was greater than in the T2 field, ECa measurement could not be used to map soil properties and nutrient levels. According to the regression model, ECa was strongly influenced by SOM and clay content, whereas it was weakly linked to soil chemical properties (N, P2O5, and K2O). These findings are consistent with the results of Martinez et al. (2010) [80].

4. Conclusions

  • Our research indicated that the relationships between the soil’s physical and chemical properties and soil apparent electrical conductivity (ECa) can vary even between two neighbouring fields. While strong correlations were found in the T2 field, weak correlations were observed in the T1 field between soil properties and ECa-field measurements. The only difference between the experimental fields was topographic variability, which was found to be a key factor in determining the relationship between ECa and soil properties, and therefore, how we interpret and use ECa for soil studies;
  • Based on our results, in the absence of the relationship between the ECa-field and the soil’s physical and chemical properties, measurements based on apparent soil electrical conductivity are not suitable for characterising soil or establishing management zones. Accordingly, ECa might replace labour and time-consuming field measurements in practice only if this relationship is strong, but further experiments are required to support these findings [27,66,84];
  • According to our hypothesis, ECa-based soil mapping can only be used to characterise soil and delineate management zones under homogeneous topographic conditions. To support this finding, implementing further targeted field-plot experiments with a fine-resolution grid soil sampling strategy is recommended.

Author Contributions

Conceptualization, I.M.K. and Á.B.-F.; methodology, Á.B.-F.; software, Á.B.-F. and I.M.K.; validation, I.M.K. and L.B.; formal analysis, D.S.; investigation, S.Z.; resources, P.C.; data curation, V.M.V.; writing—original draft preparation, I.M.K. and Á.B.-F.; writing—review and editing, I.M.K. and Á.B.-F.; visualization, G.J.; supervision, I.M.K. and Á.B.-F.; project administration, M.V.; funding acquisition, R.K. All authors have read and agreed to the published version of the manuscript.

Funding

The research was funded by project no. RRF-2.3.1-21-2022-00001 in the support provided by the Recovery and Resilience Facility (RRF), financed under the National Recovery Fund budget estimate, RRF-2.3.1-21 funding scheme.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author. The data are not publicly available due to their use in subsequent studies.

Acknowledgments

This article and the research behind it would not have been possible without the exceptional support of ENVI-AGRO Consulting Ltd. and Csernozjom Ltd. Moreover, the competencies and infrastructure of Zalaegerszeg Innovation Park contributed to the research and preparation of the article.

Conflicts of Interest

Author Márton Vona was employed by the company Csernozjom Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Horváth, A. A Mezőföldi Löszvegetáció Términtázati Szerveződése; Scientia Kiadó: Budapest, Hungary, 2002; pp. 1–175. (In Hungarian) [Google Scholar]
  2. Ádám, L.; Marosi, S.; Szilárd, J. A Mezőföld Természeti Földrajza; Akadémiai Kiadó: Budapest, Hungary, 1959; pp. 1–514. (In Hungarian) [Google Scholar]
  3. Hutengs, C.; Seidel, M.; Oertel, F.; Ludwig, B.; Vohland, M. In situ and laboratory soil spectroscopy with portable visible-to-near-infrared and mid-infrared instruments for the assessment of organic carbon in soils. Geoderma 2019, 355, 113900. [Google Scholar] [CrossRef]
  4. Watson, H.D.; Neely, H.L.; Morgan, C.L.S.; McInnes, K.J.; Molling, C.C. Identifying subsoil variation associated with gilgai using electromagnetic induction. Geoderma 2017, 295, 34–40. [Google Scholar] [CrossRef]
  5. Vitharana, U.; Van-Meirvenne, M.; Simpson, D.; Cocks, L.; De-Baerdemaeker, J. Key soil and topographic properties to delineate potential management classes for precision agriculture in the European loess area. Geoderma 2008, 143, 206–215. [Google Scholar] [CrossRef]
  6. Adamchuk, V.; Lund, E.D.; Sethuramasamyraja, B.; Morgan, M.; Dobermann, A.; Marx, D. Direct measurement of soil chemical properties on-the-go using ion-selective electrodes. Comput. Electron. Agric. 2005, 48, 272–294. [Google Scholar] [CrossRef]
  7. Corwin, D.L.; Lesch, S.M. Characterizing soil spatial variability with apparent soil electrical conductivity: Part II. Case study. Comput. Electron. Agric. 2005, 46, 135–152. [Google Scholar] [CrossRef]
  8. Friedman, S.P. Soil properties influencing apparent electrical conductivity: A review. Comput. Electron. Agric. 2005, 46, 45–70. [Google Scholar] [CrossRef]
  9. Nocco, M.A.; Zipper, S.C.; Booth, E.G.; Cummings, C.R.; Loheide, S.P.I.I.; Kucharik, C.J. Combining Evapotranspiration and Soil Apparent Electrical Conductivity Mapping to Identify Potential Precision Irrigation Benefits. Remote Sens. 2019, 11, 2460. [Google Scholar] [CrossRef]
  10. Goovaerts, P. Geostatistical tools for characterizing the spatial variability of microbiological and physico-chemical soil properties. Biol. Fertil. Soils 1998, 27, 315–334. [Google Scholar] [CrossRef]
  11. Jenny, H. Factors of Soil Formation: A System of Quantitative Pedology; McGraw-Hill Book Company Inc.: New York, NY, USA, 1944; pp. 1–320. [Google Scholar]
  12. Tsegaye, T.; Hill, R.L. Intensive tillage effects on spatial variability of soil test, plant growth, and nutrient uptake measurements. Soil Sci. 1998, 163, 155–165. [Google Scholar] [CrossRef]
  13. Robert, P. Characterization of soil conditions at the field level for soil specific management. Geoderma 1993, 60, 57–72. [Google Scholar] [CrossRef]
  14. Cook, S.E.; Corner, R.J.; Grealish, G.; Gessler, P.E.; Chartres, C.J. A rule-based system to map soil properties. Soil Sci. Soc. Am. J. 1996, 60, 1893–1900. [Google Scholar] [CrossRef]
  15. Mallarino, A.P.; Wittry, D.J. Efficacy of grid and zone soil sampling approaches for site-specific assessment of phosphorus, potassium, pH, and organic matter. Precis. Agric. 2004, 5, 131–144. [Google Scholar] [CrossRef]
  16. Prado, E.V.; Machado, T.A.; Prado, F.M.T. Geração e correlação de zonas de manejo usando sensor spad e condutividade elétrica aparente do solo para a cafeicultura irrigada na zona da mata mineira. Rev. Científica Eletrônica Agron. 2015, 1, 1–7. (In Spanish) [Google Scholar]
  17. Gonçalves, J.R.M.R.; Ferraz, G.A.E.S.; Reynaldo, É.F.; Marin, D.B.; Ferraz, P.F.P.; Pérez-Ruiz, M.; Rossi, G.; Vieri, M.; Sarri, D. Comparative analysis of soil-sampling methods used in precision agriculture. J. Agric. Eng. 2021, 52, 1117. [Google Scholar] [CrossRef]
  18. Corwin, D.L.; Lesch, S. Application of soil electrical conductivity to precision agriculture: Theory, principles and guidelines. Agron. J. 2003, 95, 455–471. [Google Scholar] [CrossRef]
  19. Corwin, D.L.; Lesch, S.M.; Shouse, P.J.; Soppe, R.; Ayars, J.E. Identifying soil properties that influence cotton yield using soil sampling directed by apparent soil electrical conductivity. Agron. J. 2003, 95, 352–364. [Google Scholar] [CrossRef]
  20. Corwin, D.L.; Scudiero, E. Field-scale apparent soil electrical conductivity of soil analysis field-scale apparent soil electrical conductivity. In Methods of Soil Analysis; Logsdon, S., Ed.; Soil Science Society of America: Madison, WI, USA, 2016; pp. 1–29. [Google Scholar]
  21. Doolittle, J.; Stuebe, A.; Price, A.; Kelly, E. Mapping bedrock depths with electromagnetic induction in Costilla County, Colorado. Soil Surv. Horiz. 2015, 43, 14–21. [Google Scholar] [CrossRef]
  22. Lesch, S.M.; Corwin, D.L.; Robinson, D.A. Apparent soil electrical conductivity mapping as an agricultural management tool in arid zone soils. Comput. Electron. Agric. 2005, 46, 351–378. [Google Scholar] [CrossRef]
  23. Lesch, S.M.; Rhoades, J.D.; Corwin, D.L. ESAP-95 Version 2.10R: User Manual and Tutorial Guide; Research Report; USDA-ARS, US Salinity Laboratory: Riverside, CA, USA, 2000; p. 146.
  24. Paz, A.M.; Castanheira, N.; Farzamian, M.; Paz, M.C.; Gonçalves, M.C.; Santos, F.A.M.; Triantafilis, J. Prediction of soil salinity and sodicity using electromagnetic conductivity imaging. Geoderma 2020, 361, 114086. [Google Scholar] [CrossRef]
  25. Rhoades, J.D.; Raats, P.A.C.; Prather, R.J. Effects of liquid-phase electrical conductivity, water content, and surface conductivity on bulk soil electrical conductivity. Soil Sci. Soc. Am. J. 1976, 40, 651–655. [Google Scholar] [CrossRef]
  26. Fortes, R.; Millán, S.; Prieto, M.H.; Torres, C.C. A methodology based on apparent electrical conductivity and guided soil samples to improve irrigation zoning. Precis. Agric. 2015, 16, 441–454. [Google Scholar] [CrossRef]
  27. Moral, F.J.; Terrón, J.M.; da Silva, J.R.M. Delineation of management zones using mobile measurements of soil apparent electrical conductivity and multivariate geostatistical techniques. Soil Tillage Res. 2010, 106, 335–343. [Google Scholar] [CrossRef]
  28. Williams, B.G.; Hoey, D. The use of electromagnetic induction to detect the spatial variability of the salt and clay contents of soils. Aust. J. Soil Res. 1987, 25, 21–27. [Google Scholar] [CrossRef]
  29. Rhoades, J.D.; Corwin, D.L.; Lesch, S.M. Geospatial measurements of soil electrical conductivity to determine soil salinity and diffuse salt loading from irrigation. In Assessment of Non-Point Source Pollution in the Vadose Zone; Corwin, D.L., Loague, K., Ellsworth, T.R., Eds.; American Geophysical Union: Washington, DC, USA, 1999; pp. 197–215. [Google Scholar] [CrossRef]
  30. Greenhouse, J.P.; Slaine, D.D. The use of reconnaissance electromagnetic methods to map contaminant migration. Groundw. Monit. Rev. 1983, 3, 47–59. [Google Scholar] [CrossRef]
  31. Kitchen, N.R.; Sudduth, K.A.; Drummond, S.T. Characterizing Soil Physical and Chemical Properties Influencing Crop Yield Using Soil Electrical Conductivity; ERIM International: Ann Arbor, MI, USA, 2000; pp. 122–131. [Google Scholar]
  32. McBride, R.A.; Shrive, S.C.; Gordon, A.M. Estimating Forest Soil Quality from Terrain Measurements of Apparent Electrical Conductivity. Soil Sci. Soc. Am. J. 1990, 54, 290–293. [Google Scholar] [CrossRef]
  33. Kühn, J.; Brenning, A.; Wehrhan, M.; Koszinski, S.; Sommer, M. Interpretation of electrical conductivity patterns by soil properties and geological maps for precision agriculture. Precis. Agric. 2009, 10, 490–507. [Google Scholar] [CrossRef]
  34. Corwin, D.L.; Ahmad, H.R. Spatio-temporal impacts of dairy lagoon water reuse on soil: Heavy metals and salinity. Environ. Sci. Process. Impacts 2015, 17, 1731–1748. [Google Scholar] [CrossRef] [PubMed]
  35. Bekele, A.; Hudnall, W.H.; Daigle, J.J.; Prudente, J.A.; Wolcott, M. Scale dependent variability of soil electrical conductivity by indirect measures of soil properties. J. Terramech. 2005, 42, 339–351. [Google Scholar] [CrossRef]
  36. Fenton, T.E.; Lauterbach, M.A. Soil map unit composition and scale of mapping related to interpretations for precision soil and crop management in Iowa. In Proceedings of the Fourth International Conference on Precision Agriculture, St. Paul, MN, USA, 19–22 July 1998; Robert, P.C., Rust, R.H., Larson, W.E., Eds.; ASA, CSSA, and SSSA Books: Madison, WI, USA, 1999; pp. 295–341. [Google Scholar]
  37. Peralta, N.R.; Costa, J.L.; Balzarini, M.; Angelini, H. Delineation of management zones with measurements of soil apparent electrical conductivity in the southeastern pampas. Can. J. Soil Sci. 2013, 93, 205–218. [Google Scholar] [CrossRef]
  38. Corwin, D.L.; Lesch, S.M. Protocols and guidelines for field-scale measurement of soil salinity distribution with ECa–directed soil sampling. J. Environ. Eng. Geophys. 2013, 18, 1–25. [Google Scholar] [CrossRef]
  39. Sudduth, K.A.; Kitchen, N.R.; Wiebold, W.J.; Batchelor, W.D.; Bollero, G.A.; Bullock, D.G.; Clay, D.E.; Palm, H.L.; Pierce, F.J.; Schuler, R.T.; et al. Relating apparent electrical conductivity top soil properties across the North-Central USA. Comput. Electron. Agric. 2005, 46, 263–283. [Google Scholar] [CrossRef]
  40. Córdoba, M.A.; Bruno, C.I.; Costa, J.L.; Peralta, N.R.; Balzarini, M.G. Protocol for multivariate homogeneous zone delineation in precision agriculture. Biosyst. Eng. 2016, 143, 95–107. [Google Scholar] [CrossRef]
  41. Fridgen, J.J.; Kitchen, N.R.; Sudduth, K.A.; Drummond, S.T.; Wiebold, W.J.; Fraisse, C.W. Management zone analyst (MZA): Software for subfield management zone delineation. Agron. J. 2004, 96, 100–108. [Google Scholar] [CrossRef]
  42. Kitchen, N.R.; Sudduth, K.A.; Myersb, D.B.; Drummonda, S.T.; Hongc, S.Y. Delineating productivity zones on claypan soil fields using apparent soil electrical conductivity. Comput. Electron. Agric. 2005, 46, 285–308. [Google Scholar] [CrossRef]
  43. Ortega, R.A.; Santibáňez, O.A. Determination of management zones in corn (Zea mays L.) based on soil fertility. Comput. Electron. Agric. 2007, 58, 49–59. [Google Scholar] [CrossRef]
  44. Xin-Zhoung, W.; Guo-Shun, L.; Hong-Chao, H.; Zhen-Hai, W.; Qing-Hua, L.; Xu-Feng, L.; Wei-Hong, H.; Yan-Tao, L. Determination of management zones for a tobacco field based on soil fertility. Comput. Electron. Agric. 2009, 65, 168–175. [Google Scholar] [CrossRef]
  45. Singh, G.; Williard, K.W.J.; Schoonover, J.E. Spatial relation of Apparent soil electrical conductivity with crop yield and soil properties at different topographic positions in a small agricultural watershed. Agronomy 2016, 6, 57. [Google Scholar] [CrossRef]
  46. Zhu, Q.; Lin, H. Repeated electromagnetic induction surveys for determining subsurface hydrologic dynamics in an agricultural landscape. Soil Sci. Soc. Am. J. 2010, 74, 1750–1762. [Google Scholar] [CrossRef]
  47. Marosi, S.; Somogyi, S. Magyarország Kistájainak Katasztere, I–II; MTA-Földrajztudományi Kutató Intézet: Budapest, Hungary, 1990; pp. 1–1023. (In Hungarian) [Google Scholar]
  48. Rhoades, J.D.; Ingvalson, R.D. Determining Salinity in Field Soils with Soil Resistance Measurements. Soil Chem. 1971, 35, 54–60. [Google Scholar] [CrossRef]
  49. Rhoades, J.D.; Manteghi, N.A.; Shouse, P.J.; Alves, W.J. Soil Electrical Conductivity and Soil Salinity: New Formulations and Calibrations. Soil Sci. Soc. Am. J. 1989, 53, 433–439. [Google Scholar] [CrossRef]
  50. Peralta, N.R.; Costa, J.L. Delineation of management zones with soil apparent electrical conductivity to improve nutrient management. Comput. Electron. Agric. 2013, 99, 218–226. [Google Scholar] [CrossRef]
  51. Finnern, H.; Grottenthaler, W.; Kühn, D.; Palchen, W.; Schraps, W.G.; Sponagel, H. Kartieranleitung. 4. Verbesserte und Erweiterte Auflage; AG Boden: Hannover, Germany, 1994. (In German) [Google Scholar]
  52. Buzás, I.; Daróczi, S.; Dódony, I.; Kálmán, A.; Kocsis, I.; Pártay, G.; Rajkai, K. Talaj-És Agrokémiai Vizsgálati Módszerkönyv 1; Inda 4231 Kiadó: Budapest, Hungary, 1993. (In Hungarian) [Google Scholar]
  53. Team, R.C. R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing; R-Project: Vienna, Austria, 2018.
  54. Fox, J.; Weisberg, S. An R Companion to Applied Regression, 3rd ed.; SAGE Publications, Inc.: Thousand Oaks, CA, USA, 2019. [Google Scholar]
  55. Wei, T.; Simko, V. R Package “Corrplot”: Visualization of a Correlation Matrix, Version 0.84; R-Project: Vienna, Austria, 2017.
  56. Oksanen, J.; Blanchet, F.G.; Kindt, R.; Legendre, P.; Minchin, P.R.; O’Hara, R.B.; Simpson, G.L.; Solymos, P.; Stevens, M.H.M.; Wagner, H. Vegan: Community Ecology Package, R package Version 2.3-4; R-Project: Vienna, Austria, 2019.
  57. Riley, S.J.; DeGloria, S.D.; Elliot, R. A terrain ruggedness index that quantifies topographic heterogeneity. Intermt. J. Sci. 1999, 5, 23–27. [Google Scholar]
  58. Wilson, M.F.J.; O’Connell, B.; Brown, C.; Guinan, J.C.; Grehan, A.J. Multiscale terrain analysis of multibeam bathymetry data for habitat mapping on the continental slope. Mar. Geod. 2007, 30, 3–35. [Google Scholar] [CrossRef]
  59. Pearson, K. On lines and planes of closest fit to systems of points in space. Lond. Edinb. Dublin Philos. Mag. J. Sci. 1901, 2, 559–572. [Google Scholar] [CrossRef]
  60. Hotelling, H. Analysis of a complex of statistical variables into principal components. J. Educ. Psychol. 1933, 24, 417–441. [Google Scholar] [CrossRef]
  61. Sharma, S. Applied Multivariate Techniques; John Wiley and Sons: New York, NY, USA, 1996. [Google Scholar]
  62. Boecker, D.; Centeri, C.; Welp, G.; Möseler, B.M. Parallels of secondary grassland succession and soil regeneration in a chronosequence of central-Hungarian old fields. Folia Geobot. 2015, 50, 91–106. [Google Scholar] [CrossRef]
  63. Jassó, F.; Darab, K.; Fórizsa, J.-N.; Földvári, G.; Várallyay, G. A Genetikus Üzemi Talajtérképezés Módszerkönyve; Országos Mezőgazdasági Minőségvizsgáló Intézet: Budapest, Hungary, 1966; pp. 509–510. (In Hungarian) [Google Scholar]
  64. Stefanovits, P.; Filep, G.; Füleky, G. Talajtan; Mezőgazda Kiadó: Budapest, Hungary, 2010; pp. 1–433. (In Hungarian) [Google Scholar]
  65. Michéli, E.; Fuchs, M.; Hegymegi, P.; Stefanovits, P. Classification of the major soils of Hungary and their correlation with the world reference base for soil resources (WRB). Agrokémia És Talajt. 2006, 55, 19–28. [Google Scholar] [CrossRef]
  66. Fraisse, C.W.; Sudduth, K.A.; Kitchen, N.R. Delineation of site-specific management zones by unsupervised classification of topographic attributes and soil electrical conductivity. Am. Soc. Agric. Biol. Eng. 2001, 44, 155–166. [Google Scholar] [CrossRef]
  67. Ayele, G.T.; Demissie, S.S.; Jemberrie, M.A.; Jeong, J.; Hamilton, D.P. Terrain effects on the spatial variability of soil physical and chemical properties. Soil Syst. 2019, 4, 1. [Google Scholar] [CrossRef]
  68. Bakr, N.; Ali, R.R. Statistical relationship between land surface altitude and soil salinity in the enclosed desert depression of arid regions. Arab. J. Geosci. 2019, 12, 715. [Google Scholar] [CrossRef]
  69. Heiniger, R.W.; McBride, R.G.; Clay, D.E. Using soil electrical conductivity to improve nutrient management. Agron. J. 2003, 95, 508–519. [Google Scholar] [CrossRef]
  70. Moore, I.D.; Gessler, P.; Nielsen, G.; Peterson, G. Soil attribute prediction using terrain analysis. Soil Sci. Soc. Am. J. 1993, 57, 443–452. [Google Scholar] [CrossRef]
  71. Heil, K.; Schmidhalter, U. Characterisation of soil texture variability using the apparent soil electrical conductivity at a highly variable site. Comput. Geosci. 2012, 39, 98–110. [Google Scholar] [CrossRef]
  72. Carroll, Z.; Oliver, M. Exploring the spatial relations between soil physical properties and apparent electrical conductivity. Geoderma 2005, 128, 354–374. [Google Scholar] [CrossRef]
  73. Nyéki, A.; Milics, G.; Kovács, A.J.; Neményi, M. Effects of soil compaction on cereal yield. Cereal Res. Commun. 2017, 45, 1–22. [Google Scholar] [CrossRef]
  74. Broge, N.H.; Thomsen, A.G.; Greve, M.H. Prediction of topsoil organic matter and clay content from measurements of spectral reflectance and electrical conductivity. Acta Agric. Scand. Sect. B Soil Plant Sci. 2006, 54, 232–240. [Google Scholar] [CrossRef]
  75. Rezaei, M.; Saey, T.; Seuntjens, P.; Joris, I.; Boënne, W.; Meirvenne, M.V.; Cornelis, W. Predicting saturated hydraulic conductivity in a sandy grassland using proximally sensed apparent electrical conductivity. J. Appl. Geophys. 2016, 126, 35–41. [Google Scholar] [CrossRef]
  76. Bronson, K.F.; Booker, J.D.; Officer, S.J.; Lascano, R.J.; Maas, S.J.; Searcy, S.W. Apparent electrical conductivity, soil properties and spatial covariance in the U.S. Southern High Plains. Precis. Agric. 2005, 6, 297–311. [Google Scholar] [CrossRef]
  77. Omonode, R.A.; Vyn, T.J. Spatial dependence and relationships of electrical conductivity to soil organic matter, phosphorus, and potassium. Soil Sci. 2006, 171, 223–238. [Google Scholar] [CrossRef]
  78. Serrano, J.; Shahidian, S.; Silva, J.M.D. Spatial and Temporal Patterns of Apparent Electrical Conductivity: DUALEM vs. Veris Sensors for Monitoring Soil Properties. Sensors 2014, 14, 10024–10041. [Google Scholar] [CrossRef]
  79. Lal, R. Soil carbon sequestration impacts on global climate change and food security. Science 2004, 304, 1623–1627. [Google Scholar] [CrossRef] [PubMed]
  80. Martinez, G.; Vanderlinden, K.; Giráldez, J.V.; Espejo, A.J.; Muriel, J.L. Field-scale soil moisture pattern mapping using electromagnetic induction. Vadose Zone J. 2010, 9, 871–881. [Google Scholar] [CrossRef]
  81. Reyes, J.; Wendroth, O.; Matocha, C.; Zhu, J. Delineating site-specific management zones and evaluating soil water temporal dynamics in a farmer’s field in Kentucky. Vadose Zone J. 2019, 18, 1–19. [Google Scholar] [CrossRef]
  82. Van Meirvenne, M.; Islam, M.M.; De Smedt, P.; Meerschman, E.; Van De Vijver, E.; Saey, T. Key variables for the identification of soil management classes in the Aeolian landscapes of north–west Europe. Geoderma 2013, 199, 99–105. [Google Scholar] [CrossRef]
  83. Peralta, N.R.; Costa, J.L.; Balzarini, M.; Franco, M.C.; Córdoba, M.; Bullock, D. Delineation of management zones to improve nitrogen management of wheat. Comput. Electron. Agric. 2015, 110, 103–113. [Google Scholar] [CrossRef]
  84. Li, Y.; Shi, Z.; Li, F. Delineation of site-specific management zones based on temporal and spatial variability of soil electrical conductivity. Pedosphere 2007, 17, 156–164. [Google Scholar] [CrossRef]
  85. Johnson, C.K.; Doran, J.W.; Duke, H.R.; Wienhold, B.; Eskridge, K.; Shanahan, J. Field-scale electrical conductivity mapping for delineating soil condition. Soil Sci. Soc. Am. J. 2001, 65, 1829–1837. [Google Scholar] [CrossRef]
  86. Sudduth, K.A.; Kitchen, N.R.; Bollero, G.A.; Bullock, D.G.; Wiebold, W.J. Comparison of electromagnetic induction and direct sensing of soil electrical conductivity. Agron. J. 2003, 95, 472–482. [Google Scholar] [CrossRef]
Agronomy 14 01161 g001
Figure 1. Flowchart of the research framework. ECa stands for apparent electrical conductivity. *: Although Pürckhauer type soil core sampling was conducted once it is indicated twice in the flowchart, reflecting its dual target: determining the sampling design and humus layer thickness.
Figure 1. Flowchart of the research framework. ECa stands for apparent electrical conductivity. *: Although Pürckhauer type soil core sampling was conducted once it is indicated twice in the flowchart, reflecting its dual target: determining the sampling design and humus layer thickness.
Agronomy 14 01161 g001
Agronomy 14 01161 g002aAgronomy 14 01161 g002b
Figure 2. Soil sampling points for describing soil profiles and laboratory analysis according to ECa classes (A) and elevation differences (B).
Figure 2. Soil sampling points for describing soil profiles and laboratory analysis according to ECa classes (A) and elevation differences (B).
Agronomy 14 01161 g002aAgronomy 14 01161 g002b
Agronomy 14 01161 g003
Figure 3. Soil profiles in excavation pits (0–150 cm depth) and with Pürckhauer type soil core samplers (0–100 cm depth) at sampling points 5 and 6.
Figure 3. Soil profiles in excavation pits (0–150 cm depth) and with Pürckhauer type soil core samplers (0–100 cm depth) at sampling points 5 and 6.
Agronomy 14 01161 g003
Agronomy 14 01161 g004
Figure 4. Soil profiles in excavation pits (0–150 cm depth) and with Pürckhauer type soil core samplers (0–100 cm depth) at the sampling points 7 and 11.
Figure 4. Soil profiles in excavation pits (0–150 cm depth) and with Pürckhauer type soil core samplers (0–100 cm depth) at the sampling points 7 and 11.
Agronomy 14 01161 g004
Agronomy 14 01161 g005
Figure 5. Samples arranged in two-dimensional ordination space according to principal component analyses of the full dataset (T1: red, T2: blue).
Figure 5. Samples arranged in two-dimensional ordination space according to principal component analyses of the full dataset (T1: red, T2: blue).
Agronomy 14 01161 g005
Agronomy 14 01161 g006aAgronomy 14 01161 g006b
Figure 6. Matrix of Pearson’s correlation coefficients (×100) between measured variables in each field. For the full names and measurement units of the variables, please refer to Table 1. Grey crosses indicate non-significant correlations at level 0.05.
Figure 6. Matrix of Pearson’s correlation coefficients (×100) between measured variables in each field. For the full names and measurement units of the variables, please refer to Table 1. Grey crosses indicate non-significant correlations at level 0.05.
Agronomy 14 01161 g006aAgronomy 14 01161 g006b
Table 1. Descriptive statistics of apparent electrical conductivity (ECa) and elevation and soil properties. CV stands for coefficient of variation. CVs greater than 0.3 are bold.
Table 1. Descriptive statistics of apparent electrical conductivity (ECa) and elevation and soil properties. CV stands for coefficient of variation. CVs greater than 0.3 are bold.
FieldVariablesMeanMedianSDCVMinMaxRange
T1ECa-field (mS/cm)223.75241.3769.370.31114.38321.35206.97
ECa-lab (mS/cm)476.57465.0036.980.08417.00532.00115.00
Elevation (m)169.00169.181.750.01166.49170.804.31
pH (KCl)7.007.190.440.066.157.371.22
Binding capacity (KA)49.8650.002.670.0545.0053.008.00
Humus layer thickness (cm)39.2935.007.870.2035.0055.0020.00
SOM (% by weight)2.132.000.390.181.602.701.10
Sand (%)16.6616.316.370.389.7726.0516.29
Silt (%)77.4278.345.760.0768.7083.8015.10
Clay (%)5.925.720.810.145.257.502.25
Salt (% by weight)0.030.030.010.380.010.050.04
CaCO3 (% by weight)5.637.405.580.990.0012.3012.30
N (mg/kg)10.1410.002.040.207.0013.006.00
Mg (mg/kg)295.00241.00132.560.45191.00557.00366.00
S (mg/kg)2.872.900.420.152.403.401.00
K2O (mg/kg)253.57266.0047.350.19159.00305.00146.00
Na (mg/kg)29.0031.0011.800.4114.0044.0030.00
P2O5 (mg/kg)56.2953.0022.510.4029.0093.0064.00
Cu (mg/kg)1.511.200.740.490.902.801.90
Mn (mg/kg)159.71128.0095.570.6071.00315.00244.00
Zn (mg/kg)0.590.600.210.360.300.900.60
T2ECa-field (mS/cm)314.78320.91136.040.43147.09552.15405.06
ECa-lab (mS/cm)646.89602.0088.560.14548.00806.00258.00
Elevation (m)176.05176.642.280.01172.61179.506.89
pH (KCl)7.117.120.150.026.887.300.42
Binding capacity (KA)50.6750.002.870.0648.0057.009.00
Humus layer thickness (cm)58.3360.0014.790.2535.0090.0055.00
SOM (% by weight) 1.861.800.300.161.502.501.00
Sand (%)15.7718.737.060.452.2824.3822.10
Silt (%)77.9375.696.160.0870.9490.3219.38
Clay (%)6.306.191.020.164.687.522.84
Salt (% by weight)0.070.060.010.210.050.090.04
CaCO3 (% by weight)4.993.506.551.310.0019.7019.70
N (mg/kg)19.7815.009.110.4610.0034.0024.00
Mg (mg/kg)302.11320.00100.340.33154.00444.00290.00
S (mg/kg)4.424.201.210.273.006.503.50
K2O (mg/kg)321.56301.0063.210.20255.00460.00205.00
Na (mg/kg)36.1132.0014.950.4123.0071.0048.00
P2O5 (mg/kg)184.00183.0085.340.4665.00365.00300.00
Cu (mg/kg)1.941.500.890.461.003.202.20
Mn (mg/kg)182.00133.00101.380.5665.00321.00256.00
Zn (mg/kg)0.930.700.470.510.501.901.40
Table 2. Eigenvalues of, and cumulative variance explained by three and four principal components (PCs) in the T1 and T2 fields, respectively, and the factor loadings of the soil properties. Loadings higher than 0.3 are bold.
Table 2. Eigenvalues of, and cumulative variance explained by three and four principal components (PCs) in the T1 and T2 fields, respectively, and the factor loadings of the soil properties. Loadings higher than 0.3 are bold.
FieldsT1T2
PCs PC1PC2PC3PC1PC2PC3PC4
Eigenvalue12.783.752.119.3134.0033.1181.978
Cumulative σ263.8982.6493.1846.56566.58082.17192.060
Parameter scores
ECa-lab (mS/cm)−0.0380.1880.0710.1170.205−0.194−0.312
Elevation (m)0.0120.0170.239−0.019−0.109−0.230−0.136
pH (KCl)0.359−0.324−0.267−0.078−0.070−0.1550.025
Binding capacity (KA)0.150−0.359−0.119−0.0250.147−0.2030.460
Humus layer thickness (cm)−0.005−0.2470.228−0.0470.0970.3840.330
SOM (% by weight)0.2270.208−0.6640.1910.057−0.1570.282
Sand (%)−0.274−0.289−0.0390.161−0.4530.0060.177
Silt (%)0.2720.286−0.022−0.1590.421−0.017−0.209
Clay (%)0.2170.2410.488−0.1310.5860.0780.113
Salt (% by weight)−0.0400.3260.0840.1220.125−0.159−0.218
CaCO3 (% by weight)0.3190.1500.024−0.3120.010−0.3120.305
N (mg/kg)0.025−0.0300.0300.2800.017−0.4050.253
Mg (mg/kg)−0.3370.272−0.1240.1380.2640.2500.080
S (mg/kg)0.0190.078−0.0830.2600.135−0.325−0.163
K2O (mg/kg)0.0330.283−0.1410.1560.144−0.088−0.178
Na (mg/kg)0.2230.1760.043−0.1870.149−0.3300.232
P2O5 (mg/kg)0.0570.010−0.1150.245−0.063−0.126−0.142
Cu (mg/kg)−0.3450.266−0.1730.4110.1280.2390.177
Mn (mg/kg)−0.436−0.009−0.0770.4460.0720.1200.086
Zn (mg/kg)−0.133−0.102−0.1380.3270.091−0.0840.120
Table 3. Linear regression results of the ECa-field using key principal components as independent proxy variables. Coefficient of determination (R2) adjusted coefficient of determination (adjusted R2) and root mean squared error (RMSE) are provided by the linear equation.
Table 3. Linear regression results of the ECa-field using key principal components as independent proxy variables. Coefficient of determination (R2) adjusted coefficient of determination (adjusted R2) and root mean squared error (RMSE) are provided by the linear equation.
FieldsModelR2Adjusted R2RMSE
T10.148 + 0.013 × PC1 − 0.131 × PC2 − 0.262 × PC30.4700.0510.107
T20.631 − 0.137 × PC1 + 0.306 × PC2 + 0.241 × PC3 − 0.281 × PC40.7210.4410.155
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kulmány, I.M.; Bede, L.; Stencinger, D.; Zsebő, S.; Csavajda, P.; Kalocsai, R.; Vona, M.; Jakab, G.; Vona, V.M.; Bede-Fazekas, Á. Challenges in Mapping Soil Variability Using Apparent Soil Electrical Conductivity under Heterogeneous Topographic Conditions. Agronomy 2024, 14, 1161. https://doi.org/10.3390/agronomy14061161

AMA Style

Kulmány IM, Bede L, Stencinger D, Zsebő S, Csavajda P, Kalocsai R, Vona M, Jakab G, Vona VM, Bede-Fazekas Á. Challenges in Mapping Soil Variability Using Apparent Soil Electrical Conductivity under Heterogeneous Topographic Conditions. Agronomy. 2024; 14(6):1161. https://doi.org/10.3390/agronomy14061161

Chicago/Turabian Style

Kulmány, István Mihály, László Bede, Dávid Stencinger, Sándor Zsebő, Péter Csavajda, Renátó Kalocsai, Márton Vona, Gergely Jakab, Viktória Margit Vona, and Ákos Bede-Fazekas. 2024. "Challenges in Mapping Soil Variability Using Apparent Soil Electrical Conductivity under Heterogeneous Topographic Conditions" Agronomy 14, no. 6: 1161. https://doi.org/10.3390/agronomy14061161

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop