Introduction

There are several studies that show the high spatial variability of soil properties (e.g., Peralta and Costa 2013). Soil formation mechanisms are the result of complex interactions between biological, physical, and chemical factors acting simultaneously with different intensities on a parent material, as influenced by topography (Moral et al. 2010). Consequently, uniform management of fields does not take into account the spatial variability, and is not the most effective management strategy. Site-specific management (SSM) is a form of precision agriculture whereby decisions on resource application and agronomic practices are improved to better match soil and crop requirements as they vary in the field (Peralta and Costa 2013). SSM enables the identification of regions homogeneous management zones (HMZ) within the area delimited by field boundaries. These subfield regions constitute areas that have similar permanent characteristics, such as topography and nutrient levels (Moral et al. 2010).

Traditional soil sampling and the necessary laboratory analysis are time-consuming, labour-intensive and cost prohibitive (Huang et al. 2014), not viable from a SSM perspective because they require a large number of soil samples in order to achieve a good representation of soil properties and fertility (Moral et al. 2010; Peralta and Costa 2013; Huang et al. 2014).

A major focus of SSM research has therefore been on the development of low cost technologies for the acquisition of spatially referenced data on soil and plant properties. Considering this perspective and although different techniques are available, the application of geospatial measurements of apparent soil electrical conductivity (ECa), combined with the use of global navigation satellite systems (GNSS) and geographical information systems (GIS), can be useful for characterizing the spatial pattern of soil properties within fields (Sudduth et al. 2003). As well as providing a correlation with soil properties, ECa is an efficient ground-based sensing technology that allows digital soil mapping (DSM) (Huang et al. 2014) and has the potential to identify sampling zones to implement SSM strategies (Corwin and Lesch 2003; Kühn et al. 2009; Moral et al. 2010; Peralta and Costa 2013), serving as a key tool in supporting decision making and management of soil nutrient balances (Serrano et al. 2014). ECa can be intensively recorded in an easy and inexpensive way, and is usually related to various physico-chemical properties across a wide range of soils (Sudduth et al. 2005). This methodology can improve the characterization of the spatial pattern of edaphic properties that influence the soil nutrient content, which in turn can be used to define SSM units (Moral et al. 2010). However, the ECa applications in HMZ showed weak and inconsistent relationships between ECa and soil characteristics (Corwin and Lesch 2003; Sudduth et al. 2005). These inconsistent relationships may be generated by the potentially complex interrelationships between ECa and soil characteristics (Moral et al. 2010; Peralta and Costa 2013), but also by the temporal variability of the ECa measurements carried out under diverse conditions (e.g., in terms of soil moisture content or vegetation cover) (Serrano et al. 2014).

The decision by farmers to adopt precision farming practices in terms of different management within a field is usually supported by maps of spatial variability and temporal stability. A number of methods have been used by different research teams to define the spatial and temporal trends found within a field (Blackmore 2000; Blackmore et al. 2003; Xu et al. 2006). Spatial variability and temporal stability are two conditions which may justify SSM strategies and are the basis for variable rate technology (VRT) implementation (Schellberg et al. 2008). A spatial and temporal trend map can help to differentiate areas of the field as a function of their soil characteristics (Blackmore et al. 2003) and to develop SSM units for each field in subsequent seasons (Xu et al. 2006). If a spatial pattern is temporally stable within a field, then it is reasonable to suppose that it should be a reasonably good predictor of the spatial patterns in the following years. This assumes that the prevailing conditions and limiting factors are consistent over the years (Blackmore et al. 2003). This method combines both spatial and temporal effects into a single map of management classes. Previously, Blackmore (2000) and Xu et al. (2006) carried out studies of the temporal stability of cereal crop yields (with a particular threshold value of the temporal coefficient of variation, CV of 30 %) and grassland crop yields (with two threshold values of 15 and 25 %), respectively. However, at present, little is known about the nature, extent or temporal stability of within-field variation in ECa in the complex Mediterranean grassland soil-animal-ecosystem and whether it would be feasible to manage such variation in a site-specific way.

The general objectives of this study were to evaluate (i) the specificity of the spatial and temporal dynamics of ECa measured by a EMI sensor, over a 7 year period, under variable conditions (of soil moisture content, soil vegetation cover and grazing management) and, consequently, (ii) the potential for implementing SSM.

Materials and methods

Site characteristics and management history

The experimental field, with an area of about 6 ha, is located at the Revilheira farm (38°27′51.6″N and 7°25′46.2″W) in Southern Portugal. The predominant soil of the field is classified as a Leptic Luvisol, common in Mediterranean region (FAO 2006). The Leptic Luvisol profile is characterized by a pedogenetic clay differentiation, with a lower clay content in the topsoil (0.15–0.50 m) and a higher clay content in the subsoil, without leaching of base cations or advanced weathering. These processes (especially clay migration) lead to a subsoil horizon with high-activity clays and a high base saturation at some depths (FAO 2006).

In Southern region of Portugal, intensive land use with cereal monoculture, subject to soil tillage operations with heavy implements just after the first autumn rains has prevailed for decades. The characteristic undulating relief, in association with this form of land use management, has originated mechanisms of erosion and soil transport, leading to degraded, shallow and stony soils, especially at upper slopes. These soils, characterized by loam texture, low levels of organic matter, slightly acid, poor in nitrogen and phosphorus are currently subject to recovery programs with extensive grazing or planted to tree crops (Serrano et al. 2014). A permanent bio-diverse pasture was established in this field in September 2000 in a flexible rotation system for grazing sheep throughout the year. Between June 2007 and March 2012 the field was not grazed; between April 2012 and February 2013 the field was permanently grazed by cattle under extensive conditions.

Digital elevation model

A topographic survey of the area was carried out using a Real Time Kinematic (RTK) GPS instrument (Trimble RTK/PP-4700 GPS, Trimble Navigation Limited, USA). The elevation data were sampled in the field with the GPS assembled on an all-terrain vehicle. The digital elevation model surface was created using the triangulated irregular network (TIN) interpolation tool from ArcGIS 9.3. TIN algorithm uses sample points to create a surface formed by triangles based on nearest neighbour point information. This vector information was converted into a grid surface with a 1 m resolution using the “Spatial Analyst” tool.

Soil and pasture sample collection and analysis

Soil spatial variability of the 6 ha experimental field was characterized by 76 samples geo-referenced with GPS taken from the whole parcel (one from each 28 × 28 m square). The soil samples were collected using a gouge auger and a hammer, from a depth of 0–0.30 m. The soil was characterized in terms of texture, moisture content, pH, organic matter content, nitrogen, phosphorus and potassium. Each composite sample was the result of five sub-samples, one taken from the center of the square, and the other four taken near the corners of the square. The soil samples were air-dried and analysed for particle-size distribution using a sedimentographer (Sedigraph 5100, manufactured by Micromeritics), after passing the fine components through a 2 mm sieve. These fine components were also analysed for pH in 1:2.5 (soil:water) suspension, using the potenciometric method. Organic matter was measured by combustion and CO2 measurement, using an infrared detection cell. The NO3 was measured using the selective ion method. P2O5 and K2O were extracted by the Egner-Riehm method, and P2O5 was measured using colorimetric method, while K2O content was measured with a flame photometer. The soil samples were weighed, dried at 70 °C for 48 h, and then weighed again to establish the SMC.

Seventy-six pasture samples of 1 m2 areas geo-referenced with GPS from each field square were collected in February of 2012. These samples were weighed to determine the green matter production per hectare, and subsamples in small paper bags were placed in a 65 °C oven for 48 h to determine the pasture moisture content, which was used to calculate pasture dry matter yield.

Apparent soil electrical conductivity surveys

EMI is a technique that measures ECa by inducing an electric current in the soil. The DUALEM 1S proximal sensor (Dualem, Inc., Milton, Ontario, Canada), equipped with a GPS antenna was used to measure the ECa in the experimental field on four occasions over a 7 year period (Fig. 1; Table 1).

Fig. 1
figure 1

Measurements of the ECa in the experimental field in a June 2007 (DSDG dry soil, small and dry vegetation, grazed by sheep); b March 2010 (WSGN wet soil, small and green vegetation, not grazed); c February 2012 (DHGN dry soil, high and green vegetation, not grazed); and d February 2013 (WSGG wet soil, small and green vegetation, grazed by cattle)

Full size image
Table 1 Conditions of the measurements of the ECa in the experimental field, between 2007 and 2013
Full size table

The DUALEM 1S sensor has a 1-m separation between its transmitter (Tx) and two receivers (Rx), so its dual depths of conductivity measurement are 0–0.50 and 0–1.50 m, with 70 % of the cumulative influence of the ECa evaluated by the sensor coming from the top 0–0.50 and 0–1.50 m soil layers, respectively (McNeill 1980). Given the height of the vehicle (0.20 m) and assuming that the ECa of the air is zero (Kühn et al. 2009), the ECa readings were effectively made from the top 0–0.30 and 0–1.30 m pseudo-dephts (PD). Due to the sensitivity of this sensor to metallic structures, a four-wheel vehicle equipped with a 3 m long tow arm was built from polyvinyl chloride (PVC), and pulled by a pickup truck, a conventional tractor or a all-terrain vehicle at an average speed of 5 km h−1. Each 28 m by 28 m square was covered twice in opposite directions, with a spacing of about 14 m (Serrano et al. 2014). The electrical conductivity sensor was programmed to register the measurements each second. In processing the data, using geographic co-ordinates as the basis, average electrical conductivity values were obtained using the values registered in each square. A safety filter was used to exclude the values registered in the first and last 4 m of each square (Serrano et al. 2010). In practice, the average value of electrical conductivity of each square was obtained using the arithmetic average of 28–30 registered values.

Since ECa was obtained under four very different conditions (of soil moisture content, vegetation cover and grazing management), for each of the four test situations, the value of ECa of each sampling point was divided by the average value of ECa of the set of 76 sampling points considered in the experimental field. The purpose of this procedure was to transform the absolute values of ECa into relative values, allowing to identify the areas of the field where the ECa showed a consistent spatial pattern over the 7 years of evaluation.

Classical statistics, spatial interpolation and regression analyses

Descriptive statistics including mean, standard deviation (SD), CV and range were determined for each dataset.

The linear correlation analysis between the ECa data from the 0–0.30 and 0–1.30 m PD in the four evaluations (June 2007; March 2010; February 2012 and February 2013) and between the ECa data from the 0–0.30 m PD and soil parameters was carried out in 20.0 SPSS statistical package for Windows (SPSS Inc., Chicago, IL, USA) to obtain the Pearson correlation coefficients (r) using the method of least square (p < 0.05).

The surface maps of soil parameters and relative ECa were produced with the ArcMap/Spatial Analyst module of the ArcGIS (ESRI 2009; version 9.3, ESRI Inc, Redlands, CA, USA), using a 5 m grid inverse squared-distance interpolator which was subsequently average filtered with a 5 × 5 m mesh grid.

Spatial variability analysis

The spatial variability in ECa measurements was calculated as the mean value (\(\overline{{y_{i} }}\)) at the i th sampling point over the years of evaluation (2007, 2010, 2012 and 2013) (Eq. 1) (Blackmore 2000; Xu et al. 2006):

$$\overline{{y_{i} }} = \frac{{\sum\limits_{t = 1}^{n} {y_{{i_{t} }} } }}{n}$$
(1)

where \(y_{{i_{t} }}\) is the ECa value at the sampling point i at time t and n is the number of sampling years.

Temporal stability analysis

The temporal stability of ECa measurements was determined by calculating the coefficient of variation at each sampling point over the years of evaluation (Eq. 2) (Blackmore 2000; Xu et al. 2006).

$$CV_{i} = \frac{{\left( {\frac{{n\sum\limits_{t = 1}^{t = n} {y_{{i_{t} }}^{2} -\left( \sum\limits_{t = 1}^{t = n} y_{{i_{t} }} \right)^{2} } }}{n(n - 1)}} \right)}}{{\overline{{y_{i} }} }}^{0.5} \times \;100$$
(2)

where CV i is the CV over time at sampling point i.

The average CV (\(\overline{CV}\)) for each year for all sampling points was calculated as follows (Eq. 3) (Xu et al. 2006):

$$\overline{{CV}} = \frac{{\left( {\sum\limits_{{i = 1}}^{m} {CV_{i} } } \right)}}{m}$$
(3)

where m is the number of soil sampling points.

The annual \(\overline{CV}\) for all sampling points was calculated to show the relative magnitudes of temporal variation for ECa measurements; large values of \(\overline{CV}\) indicate considerable temporal variation.

Maps of management classes

Although the two techniques described above quantify the spatial and temporal variation, they can be combined further into a single map of management classes, which can be used for future decision making. These maps distinguish between different areas of the field based on their spatial and temporal characteristics. The following five classes were established: (1) greater than field mean value of ECa and stable; (2) greater than field mean value of ECa and moderately stable; (3) smaller than field mean value of ECa and stable; (4) smaller than field mean value of ECa and moderately stable; and (5) unstable (Xu et al. 2006). Each sampling point was represented by a coded value. The sampling points were classified by applying combinational logic statements to the spatial variation and temporal stability data sets, considering the following conditions: condition 1 (relative value) identifies whether the point is above or below the average of all points for all the years; condition 2 (temporal stability) identifies the stability of ECa at a particular point by comparing the CV to an arbitrary threshold (15 and 25 %, stable and moderately stable respectively, as used by Xu et al. 2006). A point was considered to belong to a particular class if both conditions were true, and it was then assigned an arbitrary class code shown in brackets.

Results and discussion

Soil properties of the experimental field

Table 2 presents the mean, SD and the range of relative field elevation (RFE) and soil parameters in the experimental field in sampling years.

Table 2 Range, mean and standard deviation of relative field elevation, apparent soil electrical conductivity and others soil parameters of the experimental field
Full size table

The general characteristics of the soil are: clay loam texture; low organic matter content (<2 %); slightly acid; relatively rich in potassium; poor in nitrogen and phosphorus.

The digital elevation model surface (Fig. 2) shows that the field is composed of rolling hills, with a 20 m difference in elevation between the highest and the lowest sampled points. This contributes decisively to a high spatial variability of the soil and pasture characteristics (Marques da Silva et al. 2008).

Fig. 2
figure 2

Relative field elevation (RFE) of the experimental field

Full size image

One of the clear differences between the four test situations is related to SMC, which varied from 10–12 % (in 2007 and 2012) to 22–24 % (in 2010 and 2013). These SMC reflect the moment at which the tests were conducted: in the summer of 2007 and the end of winter in 2012, both after a relatively long period without rainfall; at the end of winter of 2010 and 2013, both after heavy rains, with high levels of SMC, very close to its field capacity.

On the other hand, the ECa showed similar mean values under conditions of low SMC (2007 and 2012), with mean values between 76 and 78 mS m−1 at the 0–1.30 m PD and between 58 and 66 mS m−1 at the 0–0.30 m PD. In 2010, the highest values of SMC corresponded to the highest ECa (mean value of 86.7 mS m−1 at 0–1.30 m PD and 75.7 mS m−1 at 0–0.30 m PD). According to Brevik et al. (2006), ECa techniques may prove to be a more effective soil-mapping tool when the soil profile is moist and less effective during dry periods. Huang et al. (2016) also have shown that, when other soil properties are uniform, the ECa measured by EMI sensors is a function of SMC. In this study the increase in the ECa from dry (2007 and 2012) to wet (2010) conditions varied between 10 and 30 % (Table 1), which is in agreement with the results of Islam et al. (2011). Although some of the measurements were carried out under conditions of low humidity (SMC of 10–12 %, Table 1), this was enough to ensure the flow of electrons in the aqueous solution, the only conducting phase of the three-phase system soil–water-air (Friedman 2005).

In 2013, however, despite the high levels of SMC, the ECa presented a major fall (mean value of 30.6 mS m−1 at 0–1.30 m PD and 20.0 mS m−1 at 0–0.30 m PD). The grazing with cows between March 2012 and February 2013 was the only change in the field management, which can have some long-term effect on the soil compaction (particularly in wet season), bulk density or soil organic matter content (by animal faecal return), all parameters that affect the ECa (Corwin and Lesch 2005), but this does not justify these important differences. Since, according to Corwin and Lesch (2005) the application of commercial fertilizer can influence ECa to the point where texture- dominated systems can be transformed into salt-dominated systems, another small contribution to the decrease of ECa in 2013 can be related to the interruption in the annual fertilization of the experimental field between 2007 and 2013 (Serrano et al. 2014).

Apparent soil electrical conductivity-correlation with soil properties

Table 3 shows the correlation coefficients between the ECa data measured by DUALEM sensor at both pseudo-depths (0–1.30 and 0–0.30 m), in the sampling years. Despite such different test conditions (relative to SMC, vegetation cover and grazing management), there were significant correlations in all comparisons (between PD and between years for each considered PD), which is a good indicator of reliability and temporal stability of the measurements taken by the EMI sensor. The measurements made in 2012 systematically present weaker correlations, since that measurement had a particularity that clearly distinguishes it from others: a highly developed vegetation cover (dry matter yield generally between 1000 and 3000 kg ha−1; Fig. 3a). This is important in view of the working principle of the sensor, placed about 0.20 m above the ground, and is influenced by the moisture content of the pasture (generally >65 %; Fig. 3b) (Serrano et al. 2014). In this case, the sensor measures the ECa of the soil layer 0–0.30 m, but also the ECa of 0.20 m of pasture between the soil and the sensor. The behavior of the sensor under these conditions should be considered in the selection of ECa sensing systems (contact vs. non-contact) for a particular application (Serrano et al. 2014).

Table 3 Correlation coefficients (r) between the apparent soil electrical conductivity data measured by DUALEM sensor at both pseudo-depths (0–1.30 and 0–0.30 m) and the four evaluation times
Full size table
Fig. 3
figure 3

Pasture dry matter yield (a) and pasture moisture content (b) in February 2012 in the experimental field

Full size image

Table 4 shows the correlation coefficients (r) between the ECa data (measured in the 0–0.30 m PD by DUALEM sensor) and topographic and soil parameters. Only the ECa data of 0–0.30 m PD were used because the soil samples were also collected from that layer.

Table 4 Correlation coefficients (r) between the apparent soil electrical conductivity data (measured in the 0–0.30 m PD by DUALEM sensor) and topographic and soil parameters
Full size table

The significant and negative values of the coefficients of correlation between ECa and the RFE in the four sampling years demonstrate the expected inverse relationship of these two attributes, shown previously by Tarr et al. (2005), Marques da Silva et al. (2008), Peralta and Costa (2013) and Huang et al. (2014).

Regarding the correlation of ECa data with the soil parameters, without considering the ECa measured in 2012 for the reasons already indicated (great development of vegetation cover) only pH, SMC and silt present significant correlations with the ECa measured in 2007, 2010 and 2013 (DSDG, WSGN and WSGG conditions, respectively). Figure 4 shows the example of the maps of these parameters (pH, SMC and silt) with significant correlations with ECa measured in 2013 (condition WSGG, wet soil, small and green vegetation, grazed by cattle).

Fig. 4
figure 4

Maps of pH (a), silt (b) and soil moisture content (c) of the experimental field in 2013 in WSGG conditions (SMC soil moisture content; WSGG wet soil, small and green vegetation, grazed by cattle)

Full size image

These results confirm that surface topography, due to its effect on the SMC, plays a significant role in influencing spatial ECa variation. Under these conditions, in shallow soils, SMC is the main factor influencing the value of ECa (Serrano et al. 2014). Various authors have observed a linear relationship between SMC and ECa for each level of soil salinity over a certain range of measured soil moistures (King et al. 2005). Soil moisture estimated from ECa reflects changes in landscape and soil characteristics.

The significant correlation of ECa with pH is consistent with findings in previous studies and is expected because it reflects the influence of salts on the ECa readings (Peralta and Costa 2013).

According to Corwin and Lesch (2005) and Friedman (2005) ECa is frequently used for the spatio-temporal characterization of soil texture. The results of this study, however, showed only significant correlation between ECa data measured by DUALEM sensor and soil silt content. The works of Sudduth et al. (2003) and Bronson et al. (2005) also showed that the ECa can be a reasonable estimator of this soil property. An important aspect that might have contributed to the lack of significant correlations between soil clay content and ECa is the presence of shallow soils at the hill tops, where the depth is quite reduced (Serrano et al. 2014). At these locations, the sensor was measuring the ECa of some of the bedrock, and the results are strongly influenced by the degree of weathering of the bedrock, with an adverse effect on the values of this parameter. Similar results were obtained by Bronson et al. (2005), possibly due, according to these authors, to low bulk density of the shallow calcic horizon at that site.

The main objective of this study was to evaluate the stability of ECa patterns, independently of the conditions under which the survey is carried out. Therefore, more important than determining absolute values of ECa, this study sought to assess the reliability of the measurements through the use of relative ECa. Figures 5 and 6 show that relative ECa patterns remain stable over time, with higher values in the South and East of this experimental field. This behavior is an excellent indicator of the reliability of this tool to survey the spatial soil variability and to delineate potential management zones. This finding of temporal stability is essential to successfull use of ECa mapping as a basis for identifying soil-sampling points (Johnson et al. 2005) and as a soil-mapping tool in precision agriculture (Brevik et al. 2006).

Fig. 5
figure 5

Relative apparent soil electrical conductivity patterns from the top 0–0.30 m pseudo-depth in a June 2007 (DSDG dry soil, small and dry vegetation, grazed by sheep); b March 2010 (WSGN wet soil, small and green vegetation, not grazed); c February 2012 (DHGN dry soil, high and green vegetation, not grazed); and d February 2013 (WSGG wet soil, small and green vegetation, grazed by cattle)

Full size image
Fig. 6
figure 6

Relative apparent soil electrical conductivity patterns from the top 0–1.30 m pseudo-depth in a June 2007 (DSDG dry soil, small and dry vegetation, grazed by sheep); b March 2010 (WSGN wet soil, small and green vegetation, not grazed); c February 2012 (DHGN dry soil, high and green vegetation, not grazed); and d February 2013 (WSGG wet soil, small and green vegetation, grazed by cattle)

Full size image

Apparent soil electrical conductivity-management classes

Understanding relationships between sensor-based measurements and soil properties related to soil quality may help to develop soil management strategies (Mertens et al. 2008).

Table 5 presents the management classes in percentage of the area of experimental field and the \(\overline{CV}\) of ECa. Temporal stability can be very important, because if the spatial patterns of soil vary significantly from year to year, SSM may not be practicable (Xu et al. 2006). These results show that ECa is a parameter with greater potential for implementing differential management; about 95 % of the area has temporal stability and the \(\overline{CV}\) of ECa is 10.48 ± 7.74 and 13.68 ± 9.92 %, respectively for 0–1.30 m PD and 0–0.30 m PD.

Table 5 Management classes in percentage of the area of experimental field and average coefficient of variation (\(\overline{CV}\)) of the apparent soil electrical conductivity
Full size table

Figure 7 shows that the area of the field where the ECa is unstable is relatively small and corresponds to the effect already reported in the case of dynamic nutrient transfer resulting from grazing animals (Page et al. 2005). The combined effects of an undulated landscape, with sparse trees and animals that selectively graze the plant species and make a heterogeneous deposition of dung, provide a notable spatial and temporal variability of soil nutrient concentration (Serrano et al. 2014).

Fig. 7
figure 7

Maps of apparent soil electrical conductivity management classes from the top 0–0.30 m pseudo-depth (a) and 0–1.30 m pseudo-depth (b) (1—greater than field mean value and stable; 2—greater than field mean value and moderately stable; 3—less than field mean value and stable; 4—less than field mean value and moderately stable; 5—unstable)

Full size image

As a practical application of this study, based on ECa measured in the experimental field during the 7 years of testing, as suggested by Peralta and Costa (2013), apart from the unstable area, four zones can be identified, corresponding to four ECa classes: one of systematically lower values (Eq. 4); another of consistently high value (Eq. 5); and two of intermediate values (Eq. 6 and 7); these zones are the basis for selection of soil-sampling points.

$$REC_{a} \le \overline{{REC_{a} }} - SD$$
(4)
$$REC_{a} \ge \overline{{REC_{a} }} + SD$$
(5)
$$1 > REC_{a} \ge \overline{{REC_{a} }} - SD$$
(6)
$$1 \le REC_{a} < \overline{{REC_{a} }} + SD$$
(7)

where \(\overline{{REC_{a} }}\) is the mean and SD the standard deviation of relative values of the ECa (RECa) at the ith sampling point over the years of evaluation (2007, 2010, 2012 and 2013).

In the specific case of this experimental field, it is possible to confirm (Table 6) that soil parameters that showed significant correlations with ECa measured in 2007, 2010 and 2013 (pH, silt and SMC) have a tendency to follow the behavior of the relative ECa in the four areas, validating the patterns that the DUALEM sensor captured through systematic surveys.

Table 6 Validation of relative apparent soil electrical conductivity classes based in the soil parameters obtained in 2013 from 76 samples points of experimental field
Full size table

These results suggest that this geospatial information of the ECa has a great potential, for example, to guide and optimize soil sampling to better characterize the spatial pattern of edaphic properties influencing crop yield (Moral et al. 2010) and, where appropriate (based on the results of the laboratory analysis), considering the site-specific nutrient management (Corwin and Lesch 2003; King et al. 2005; Peralta and Costa 2013), more specifically, the differentiated application of fertilizers (King et al. 2005).

Conclusions

Decision making is one of the critical steps in the scenario of precision agriculture. The reliability of this stage depends on the consistent survey of the variability. The decision affects the steps related to the closing of the cycle, with differentiated management of soil sampling or the use of inputs (water, fertilizers, pesticides). The use of tools to expedite survey of soil variability finds its most common expression in the measurement of ECa. The practical interest of monitoring the ECa consists in identifying areas with similar characteristics, for which the same type of management (site specific management zones) can be recommended.

This study shows a significant temporal stability of the ECa patterns under several conditions (e.g., in terms of soil moisture content, soil vegetation cover or grazing management), behavior that is an excellent indicator of the reliability of this tool to survey spatial soil variability and to delineate potential SSMZ. However, these results are based on four measurements in one field located in Southern Portugal, with shallow and stony soil, and used as a permanent pasture, and thus this work represents a exploratory study and opens perspectives for other works that allow the generalization of these conclusions.

Considering the significant correlations obtained in this work between the ECa and some soil properties, the results open perspectives for using the electromagnetic induction sensor as an indicator of SMC content in irrigation management and of needs of limestone correction in Mediterranean pastures. It would be interesting to extend the findings to other types of soils, in order to verify the origin of the lack of correlation between the ECa data measured by DUALEM sensor and soil properties such as clay, organic matter or phosphorus soil content, fundamental parameters and support tools for establishment of pasture SSM projects.