Prerpints.org logo
Preprint
Article

Using a Non-contact Sensor to Delineate Management Zones in Vineyards and Validation with the Rasch Model

Altmetrics

Downloads

80

Views

21

Comments

0

A peer-reviewed article of this preprint also exists.

Submitted:

25 September 2023

Posted:

26 September 2023

You are already at the latest version

Alerts
Abstract
The production of high-quality wines is one of the primary goals of modern oenology. In this regard, it is known that the potential quality of a wine begins to be determined in the vineyard, where the quality of the grape, initially, and later that of the wine, will be influenced by the soil properties. Given the spatial variability of the fundamental soil properties related to the potential grape production, such as texture, soil organic matter content, or cation exchange capacity, it seems that a uniform management of a vineyard is not the most optimal way to achieve higher grape quality. In this sense, the delineation of zones with similar soil characteristics to implement site-specific management is essential, reinforcing the interest to incorporate technologies and methods to determine these homogeneous zones. A case study was conducted in a 3.3 ha vineyard located near Évora, South of Portugal. A non-contact sensor (DUALEM 1S) was used to measure soil apparent electrical conductivity (ECa) in the vineyard, and later, a kriged ECa map was generated. ECa and elevation maps were utilised to delineate homogeneous zones (management zones, MZs) in the field through a clustering process. MZs were validated using some soil properties (texture, pH, organic matter -OM, phosphorous -P2O5, potassium -K2O, sum of the exchange bases -SEB; and cation exchange capacity -CEC) which were determined from 20 soil samples taken in the different MZs. Additionally, validation was also performed using Rasch measures, which were defined based on the formula-tion of the objective and probabilistic Rasch model, integrating the information from the afore-mentioned soil properties at each sampling location. The comparison of the MZs was more evident with the use of the Rasch model, as only one value was to be employed in each MZ. The use of ECa constitutes an efficient technological tool for implementing site-specific man-agement in viticulture, which allows for the improvement of decision-making processes by con-sidering the inherent spatial variability of the soil.
Keywords: 
Subject: Environmental and Earth Sciences  -   Soil Science

1. Introduction

The use of precision agriculture techniques in vineyards, known as precision viticulture, is necessary for effectively managing soil spatial variability. This becomes particularly crucial in the current context of climate change, where ensuring sustainability and the quality of wine grape production is paramount. Therefore, the adoption of methods and technologies to gather information on the key variables influencing soil spatial variability throughout vineyards is essential.
It is known that soil properties are key factors for grapevine and grape phenolic development [1,2]. The amount of available water and nutrients in soil depends on their particular characteristics, including texture, depth and fertility, biology and organic matter content [3]. The fertility of the topsoil determines the availability of nutrients to vines. Since soil fertility is usually variable, understanding its spatial pattern across a particular vineyard is crucial, as fertiliser requirements and vine growth are related to the variability of fertility.
Soil texture is a key factor to determine the amount of water a soil can hold, which is particularly important in arid regions. Soil pH is another important property as it directly impacts the availability of nutrients to grapevines. The combination of a soil (soil texture) and its acidity (pH) determines the extent to which nutrients are available to plants. Cation exchange capacity (CEC) determines the long-term productivity of a soil. Soil texture and CEC are often related; usually, high CEC is found in soils with high clay content. Therefore, if soil texture, pH, CEC, and primary nutrients availability are considered, a rational indication about the soil fertility potential in a soil can be obtained.
Differences in the key soil properties result in variations in soil fertility. Thus, the implementation of strategies for a suitable vineyard management requires the delineation of zones with similar permanent soil properties. These subfields, with similar permanent soil properties and where similar production can be expected, are usually referred to as management zones (MZs) [4]. Implementing site-specific management strategies in a field is based on the MZs, but the accurate delineation of MZs in a field is difficult because many factors can affect crop yield and quality. However, the main physical and chemical properties of soil have to be considered by any approach to determine MZs because productivity is highly related to the spatial patterns of soil nutrients [5,6].
Accurately understanding the spatial patterns of soil properties necessitates the collection and analysis of data across the vineyard. Traditional soil sampling methods are often impractical due to the significant time and cost required to collect numerous samples, followed by laboratory analysis. A more efficient approach involves the use of sensors, which reduces the need for extensive on-site soil sampling to describe spatial variability. Soil apparent electrical conductivity (ECa) is a valuable variable for characterizing different patterns in fields and has been widely used by researchers to map soil properties, particularly those related to ECa, such as clay content, cation exchange capacity, and various soil nutrient levels [7,8,9].
Delineating homogeneous zones in a field is difficult because many factors can interact in a complex way. But the fact that some variables responsible for variability remain stable over time, such as texture and topography, provides a better understanding of the main causes of the existing spatial patterns and, in consequence, a better delimitation of management zones. This is the first stage in the application of site-specific crop management [10] and it can allow grape growers to consider a smart methodology based on zones of uniform vine properties for sampling [11].
Nowadays, spatial variability across soils can be studied with high spatial resolution by means of new technologies. These new tools provide intensive georeferenced information which can be analysed according to different areas within a field [12]. In this context, the use of proximal sensing constitutes an important source of data about some soil properties [11]. There are many available sensors for the proximal monitoring of soil patterns [8]. Some of the most common sensors used in precision viticulture include the ECa sensing devices, which are coupled with GNSS receivers and mounted on mobile platforms [11]. Depending on the method of electrical conductivity measurement, there are two types of electrical conductivity sensors: non-contact or contact. The non-contact sensors (e.g., EM38 and DUALEM) work on the principle of electromagnetic induction. They are composed of a transmitter and a receiver coil, usually at opposite ends of a nonconductive bar. A magnetic field is generated, inducing an electric current in the soil which, in turn, generates a second magnetic field proportional to its conductivity [8]. The contact sensors (e.g., Veris) utilises electrodes in the shape of coulters that make contact with the soil, generating electrical currents in the soil with an electrode and measuring the potential difference with a second receiver electrode. These sensors measure the electrical resistivity, which is the inverse of the electrical conductivity [10]. ECa is affected by many soil physical and chemical properties, including texture, soluble salts, water content, organic matter content, cation exchange capacity, and many soil nutrients [7,8].
The information obtained from any sensor has to allow management decisions to vary in different locations within the field. The correct delineation of MZs is an important task to properly manage spatial variability. There are different methods to determine MZs [13,14,15]. However, cluster analysis algorithm is an optimum technique to properly divide a field using the information stored in a geographical information system [16]. When MZs are delineated, the final step is to validate them. For this purpose, it is necessary to have information from various soil samples collected in each of the MZs and to consider in those samples the most significant variables that may influence crop productivity. In this sense, the Rasch model serves as an appropriate tool to integrate all the variables in each of the samples and perform an overall validation of the MZs based on the various soil variables being considered.
The objectives of this study were to: (i) assess the soil spatial variability of a vineyard through ECa survey with a non-contact sensor; (ii) delineate MZs based on ECa and topography data, and (iii) validate the MZs using some soil properties and the Rasch measures.

2. Materials and Methods

2.1. Site Description

The field was an experimental vineyard located at the Mitra farm (University of Evora), in Southern Portugal (38.532° N; 8.015° W). The vineyard is divided in two areas (Figure 1): the smaller one has 1.3 ha and is planted with white grape varieties (“AntãoVaz”, “Arinto” and “Síria”); the larger one has 2 ha and is planted with red grape varieties (“Trincadeira”, “Aragonez” and “Castelão”).
This Mediterranean region has a climate type which is classified as Csa, according to the Köppen-Geiger classification [17]. Its main characteristics are a low precipitation, with a mean value lower than 600 mm, a high interannual variability and practically no precipitation during summer; and a wide temperature range, with a minimum around 0ºC in winter and more than 40°C, maximum, in summer.
In the experimental field, the predominant soil type is classified as a Cambisol, derived from granite [18]. These soils are characterised by slight or moderate weathering of parent material and by absence of appreciable quantities of organic matter, aluminium and/or iron compounds and illuviated clay. Usually, acid Cambisols are not very fertile, being mainly used for mixed arable farming or grazing and forest land. There are also hydromorphic soils (Leptosols and Luvisols) in the field. While Luvisols show a pedogenetic differentiation of clay content, that is, lower clay content in the topsoil and higher clay content in the subsoil, Leptosols have a are very shallow horizon, limiting the root growth.

2.2. Soil apparent electrical conductivity (ECa) survey and delineation of management zones (MZ)

A DUALEM 1S non-contact sensor (Dualem, Inc., Milton, Ontario, Canada) equipped with a GPS antenna was used to measure ECa in the experimental vineyard. This sensor has a 1 m separation between its transmitter and two receivers, enabling measurements at depths of 0–0.50 m and 0–1.50 m with successive passes across the field. A four-wheel vehicle equipped with a 3 m long tow arm was built from polyvinyl chloride (PVC), and pulled by an all-terrain vehicle at an average speed of 5 km h-1. ECa readings were effectively obtained from depths of 0–0.30 m and 0–1.30 m, taking into account that the ECa of the air is zero and considering a vehicle height of 0.20 m. The sensor was programmed to record measurements every second. ECa survey was conducted on the 24th of October, 2022. This day, the soil water content had a mean value of 17.7 % in the white grapes vineyard and 16.2 % in the red grapes vineyard. The ECa measurements have a within-field temporal stability, that is, the ECa spatial pattern is maintained over time, so the ECa survey should be conducted once in the medium term [19].
After conducting the ECa surveys, the database was revised and erroneous values, such as negative or excessively high values, were removed. Finally, 4 937 ECa values were considered, 1596 in the subparcel of the white grapes and 3 341 in the subparcel of the red grapes (Figure 1). According to Serrano et al. [20], the use of the DUALEM sensor is recommended for soils with sparse vegetation cover, similar to the conditions in the experimental vineyard.
From the ECa data, kriged ECa maps were generated in each subfield utilising the extension Geostatistical Analyst of ArcGIS (version 10.5, ESRI, Inc., Redlands, California, USA). ECa at unsampled locations was estimated with the ordinary point kriging algorithm, taking into account the spatial correlation structure described with the variograms. Subsequently, the spatial patterns of ECa were visualised with the ArcMap module of the ArcGIS.
Using the GPS instrument, a topographic survey of the subparcels was also performed. Consequently, elevation data were sampled with the GPS assembled on the all-terrain vehicle and, then, the digital elevation model surface (Figure 1) was generated with the triangulated irregular network (TIN) interpolation tool within ArcGIS. This vector information was later converted into a grid surface with the Spatial Analyst Tools in ArcGIS.
Homogeneous zones (management zones) were delineated using a classification technique. As previous studies have shown (for instance, [21]), topography is a significant factor influencing the delineation of these zones. Therefore, both topography and ECa data were considered in the zone delineation process. The final classified map was generated using an unsupervised classification technique, the ISO Cluster algorithm in ArcGIS. Two sets of input data, ECa and elevation, were used for this classification. To ensure practicality and simplicity, the number of groups (MZ) was set to three, as only a few homogeneous zones needed to be delineated.

2.3. Soil sample collection and analysis

A total of twenty samples, each georeferenced with a GPS, were collected from both subparcels. These sample locations were chosen based on the previously defined MZ. Composite soil samples were created, consisting of nine subsamples gathered around each main sampling point. All samples were collected in October 2022. The soil samples were taken using a gouge auger and a hammer, from a depth of 0–0.30 m.
The soil samples were kept in plastic bags, air-dried and analysed for their particle-size distributions (using a Sedigraph 5100, Micromeritics, Norcross, GA 30093-2901, USA), after passing the fine components through a 2 mm sieve. Afterwards, the fine components were analysed using standard methods [22]: pH in a 1:2.5 (soil:water) suspension, organic matter (OM), phosphorous (P2O5), potassium (K2O), sum of the exchange bases (SEB; calcium –Ca2+, magnesium – Mg2+, sodium – Na+ and potassium – K+) and cation exchange capacity (CEC).

2.4. The Rasch model. Validation of MZ

In order to validate the MZs, information obtained from the sampling locations was utilised. As multiple soil properties were measured in each sample, each with different units of measurement, they were integrated into a unified measure using the Rasch model. The objective of this study is to transcend the individual measurements related to various soil properties taken at different locations and condense them into a single, comprehensive variable. By utilising an adimensional common referent known as the latent variable, a synthesised representation of the various items can be generated. In this study, the latent variable was defined as “soil fertility”.
The categorization of data is the first stage to achieve an adimensional measure. In consequence, five categories were established for each soil property, following a methodology similar to that used in other studies [23,24]. Category 1 represents the lowest contribution to soil fertility and, while category 5 represents the highest contribution to soil fertility. The sum of the ratings for each soil property at each sampling location results in an initial classification of all samples.
If n are the locations in the vineyard where measurements of each soil property, i, were performed, Xni, is defined as the latent variable, soil fertility. Data are arranged in matrix form: the rows are the sampling locations and the columns are the soil properties. Each cell has a value of Xni which corresponds to the category. The sum of the item ratings is used by the Rasch model as a starting point for estimating the Rasch measures. The simple idea that some items, soil properties in this study, are more important to subjects, sampling locations in this case, is the foundation of the Rasch model. A line of measurement with the items placed hierarchically on this line according to their importance to subjects is constructed in this model. Rasch measurement construction applies a stochastic Guttman model, converting rating scale observations into linear measures. In consequence, linear statistics can be usefully applied and, additionally, tests for goodness-of-fit to validate its item calibrations and subject measures can also be used. It is important to denote that, when the Rasch model is employed, the model is given and then the data should fit the model.
To assess the contribution of each soil property to the measurement of soil fertility in the experimental vineyard, chi-square fit statistics, the Infit and Outfit Mean-Square (Infit and Outfit MNSQ), should be calculated. It is important to note that Infit and Outfit MNSQ values within the range of 0.6 to 1.5, as recommended by Bond and Fox [25], indicate an acceptable fit for an item.
The Rasch model was implemented by the Winsteps v. 4.0 computer program [26]. The mathematical formulation of this model can be revised, for example, in Ferrari and Salini [27] and Edwards and Alcock [28].
In this study, the Infit and Outfit MNSQ values (not shown) were found to fall within the acceptable range for these statistics. Therefore, all the soil properties considered were included and integrated into the characterisation of the latent variable, soil fertility. Consequently, Rasch measures for soil fertility were obtained for all sampling locations where soil samples were collected, taking into account information from nine soil properties. The procedure used in this study to determine soil fertility through the Rasch model formulation aligns with methodologies employed in previous studies [6,24,29].
The validation of the MZs was performed by computing the differences in the values for some soil properties within each zone. Later, the Rasch measures were also used to validate the MZs. To do this, the Kruskal–Wallis nonparametric test and the Dunn test as a post hoc analysis were conducted using the IBM SPSS statistical package (version 24, IBM Corp., Armonk, NY, USA).

3. Results

3.1. Delineation of management zones based on ECa and topography

The spatial distribution of ECa in both subparcels was visualised from the kriged map (Figure 2). In the subparcel with red grape varieties, a greater degree of ECa variability was observed, particularly in the central section of the vineyard where high ECa values predominated. In contrast, the subparcel with white grape varieties exhibited low ECa values in one part of the vineyard, characterised by a uniform pattern, while the other part showed higher ECa values with greater spatial variability.
It is well-established that ECa serves as a reliable indicator of soil variability [5], given its association with certain permanent soil properties such as clay and sand content, as well as soil depth [10,15]. Furthermore, soil moisture and nutrient availability are closely linked to the spatial variability of these permanent soil properties, and consequently, to ECa.
Vineyard responses are influenced by the permanent spatial variability of the soil, which can be delineated using ECa maps and topography (the latter influencing solar exposure and microclimate), in addition to seasonal variations in temperature, precipitation, and the vine cultivars. It is worth noting that in the experimental vineyard, the ECa patterns exhibit temporal stability (as reported by Serrano et al. [5]), with values changing throughout the year in response to soil moisture conditions. When delineating the spatial variability of a particular soil, it is important to conduct the ECa survey when the soil is not excessively dry, which often occurs during the summer in the region where this case study's vineyard is located.
The MZ map (Figure 3) was generated using both ECa and elevation maps. In the white grapes vineyard, two MZs were delineated, while in the red grapes vineyard, three MZs were established based on the differences in the spatial variability of ECa.

3.2. Validation of management zones using the Rasch measures

Usually, the validation of MZ involves comparing soil samples taken from different areas within the experimental field. Soil properties analysed in each sample are compared to assess whether the delineated zones exhibit significant differences. Soil properties analysed in each sample are compared to assess whether the delineated zones exhibit significant differences. This process requires making numerous comparisons between MZs, equal to the number of soil properties considered. Essentially, this analysis determines in which MZs there are differences, and if these differences are observed across a significant number of soil properties, it can be concluded that the zones are distinct. However, when using the Rasch measures determined at each sampling location, only one comparison between MZs needs to be conducted.
To compare both validations, a total of 20 georeferenced soil samples were collected in this case study. The distribution of samples across MZs was proportional to their respective areas: 7 in the less productive, 9 in the intermediate and 4 in the more productive (Figure 3).
Table 1 shows the differences in the mean values of the soil properties among the MZs. It is apparent that most of the soil properties exhibited significant differences across the MZs. In the more productive zones, lower values of sand and higher values of clay, silt, organic matter (OM), pH, sum of exchange bases (SEB), and cation exchange capacity (CEC) were observed. With respect to the assimilable phosphorous and potassium, there was a notable contrast between the MZs. A recent study in the same vineyards found a high variability in both soil properties, with coefficients of variations between 38% and 62% [5]. However, despite this, the MZs displayed significant differences in terms of soil texture and other fundamental soil properties, thus confirming the homogeneity of the delineated MZs.
When the soil data was processed by the Winsteps program, the initial analysis focused on assessing how well the data fit the model. Since the mean infit and outfit MNSQ values were very close to 1, the expected value, items (soil properties) properly fit the model [30]. In consequence, each soil property fits the general pattern of the model and, additionally, contributes to support the underlying latent variable, soil fertility.
The sum of points of all categories (raw score) and the measured values, obtained with the Winsteps programme from the raw scores, for each soil property at each sampling location, are shown in Table 2. The sampling locations are displayed in this table from the site with highest soil fertility (highest measure) to the one with lowest soil fertility (lowest measure). Figure 3 visually shows that soil samples with higher Rasch measures are located within the MZs where higher productivity is expected. When the mean values of these samples were analysed based on their location within each MZ, significant differences were observed between the three MZs (Table 2). Soil fertility in the most productive zone was higher than in the intermediate zone, and the least productive zone exhibited the lowest soil fertility.

4. Discussion

According to Ammoniaci et al. [8], precision viticulture currently focuses on four key areas: the evaluation of spatial variability, the delineation of MZs, the development of variable rate technologies and the implementation of site-specific management. The first two areas, the analysis of soil spatial variability and the delineation and validation of MZs, were addressed in this study.
Mobile soil ECa measurements are the most efficient method for mapping soil spatial variability, saving time and resources not only during the planning and establishment of new vineyards but also in the management of established ones. In this study, the high spatial variability of ECa was initially evidenced by its high coefficient of variation, 59.95%. This suggested the existing soil spatial variability in the vineyard and the convenience of delineating MZs.
As in similar studies [5,19], soil elevation was also incorporated into the process of delineating MZs in this vineyard. It is known the important influence of elevation on crop development; water accumulation and flow, and, in turn, redistribution of organic matter and soil nutrients, is closely related to elevation [31]. However, given the absence of substantial elevation differences and excessively steep areas within the experimental vineyard, topography had a limited impact on the delineation of the MZs. Consequently, the ECa and MZ maps exhibited similar patterns in this study, suggesting that the ECa map alone could effectively define the MZs.
The MZs in each subparcel had different shape. In the red grape vineyard, the MZs were arranged concentrically. Moreover, the more productive zone was subdivided. From a practical perspective for site-specific management, it would be advisable to define a continuous zone in this area. Conversely, in the white grape vineyard, both MZs were more compact. Considering a site-specific management, this vineyard can be divided into two areas, with the division aligning along the north-south limit of both MZs.
Although the validation of the MZs was initially conducted using soil variables measured in field-collected samples, located through targeted sampling considering the MZs, a second validation was performed using Rasch measures. Both validations confirmed that the MZs were distinct. However, the Rasch model formulation provided a single value for each MZ, simplifying the comparison process. Consequently, the MZ with more productive potential have a soil profile with higher Rasch measure (indicating higher soil fertility). This profile includes higher clay content, pH, OM, SEB and CEC, along with lower sand content.
There are numerous studies where relationships between ECa and soil texture have been established [7,8,10]. These studies often reveal a postive correlation between ECa and clay content and a negative correlation with sand content. Moreover, other soil properties are also related to ECa, such as CEC, pH, organic matter and different soil nutrients [5,7,8,32]. In this case study, the proposed zoning explained variations in some key soil properties, related to soil fertility, such as clay content, CEC, or OM. However, the deviation of assimilable phosphorus and potassium from the expected trends, with higher values in the MZ with lower productivity potential, may be attributed to local factors, concretely, the proximity of a pigsty near the vineyard. Nevertheless, the low values of assimilable phosphorus and potassium in the most productive MZs, around 30–40 mg kg-1, indicate higher nutrient extractions and, additionally, highlight the importance of considering spatial variability in vineyard management.

5. Conclusions

The delineation of MZs in a vineyard can be effectively achieved using a ECa sensor, as the spatial variability of the main permanent soil properties (such as texture, OM, pH, sum of the exchange bases and cation exchange capacity) are related to ECa pattern.
After determining the MZs, it is necessary to validate them, because this is the first step to implement a site-specific management in a vineyard. To perform the validation, soil information has to be obtained from some locations throughout the field. In this sense, a useful tool to integrate data from different soil variables is the formulation of the Rasch model, obtaining estimates of soil fertility at each sampling location. Utilising Rasch measures simplifies the validation process.
When new vineyards have to be designed, an accurate delineation of MZs can be relevant. They can also be useful for guiding viticultural practices and a more informed decision-making process in site-specific input management. Moreover, grape growers can implement a soil smart sampling, that is, an optimised technique to take soil samples in the vineyard. Consequently, a more cost-effective field management, besides energetic and environmental benefits, can be expected using precision viticulture techniques.

Author Contributions

Conceptualization, F.J.M; methodology, F.J.M., F.J.R. and J.S.; software, F.J.M and F.J.R.; validation, F.J.M. and F.J.R.; formal analysis, F.J.M. and J.S.; investigation, F.J.M., F.J.R. and J.S.; resources, J.S.; data curation, F.J.R.; writing—original draft preparation, F.J.M; writing—review and editing, F.J.M; visualization, F.J.M. and F.J.R.; supervision, F.J.M. and J.S.; project administration, J.S.; funding acquisition, J.S. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by National Funds through FCT (Foundation for Science and Technology) under the Project UIDB/05183/2020.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Van Leeuwen, C.; Friant, P.; Chone, X.; Tregoat, O.; Koundouras, S.; Dubourdieu, D. Influence of climate, soil and cultivar on terrior. American Journal of Enology and Viticulture 2004, 55, 207–217. [Google Scholar] [CrossRef]
  2. White, R.E. Understanding vineyard soils; Oxford University Press: New York, USA, 2015. [Google Scholar]
  3. Managing grapevine nutrition and vineyard soil health. Available online: http://sustainableagriculture.perthregionnrm.com/sites/default/files/Managin_Grapevine%20_Nutrition_Soil_Healt h.pdf (accessed on 15 May 2021).
  4. Peralta, N.R.; Cicore, P.L.; Marino, M.A.; Marques da Silva, J.R.; Costa, J.L. Use of geophysical survey as a predictor of the edaphic properties variability in soils used for livestock production. Spanish Journal of Agricultural Research 2015, 13, e1103. [Google Scholar] [CrossRef]
  5. Serrano, J.; Mau, V.; Rodrigues, R.; Paixão, L.; Shahidian, S.; Marques da Silva, J.; Paniagua, L.L.; Moral, F.J. Definition and Validation of Vineyard Management Zones Based on Soil Apparent Electrical Conductivity and Altimetric Survey. Environments 2023, 10, 117. [Google Scholar] [CrossRef]
  6. Moral, F.J.; Serrano, J.M.; Rebollo, F.J.; Teixeira, R.F.M.; Domingos, T. Comparative Analysis of Pasture Soil Fertility in Semiarid Agro-Silvo-Pastoral Systems. Agronomy 2022, 12, 2711. [Google Scholar] [CrossRef]
  7. Esteves, C.; Fangueiro, D.; Braga, R.P.; Martins, M.; Botelho, M.; Ribeiro, H. Assessing the contribution of ECa and NDVI in the 791 delineation of management zones in a vineyard. Agronomy 2022, 12, 1331. [Google Scholar] [CrossRef]
  8. Ammoniaci, M.; Kartsiotis, S.-P.; Perria, R.; Storchi, P. State of the art of monitoring technologies and data processing for Preci- sion Viticulture. Agriculture 2021, 11, 201. [Google Scholar] [CrossRef]
  9. 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. Agronomy Journal 2003, 95, 472–482. [Google Scholar] [CrossRef]
  10. Moral, F.; Terrón, J.; Silva, J.M. Delineation of management zones using mobile measurements of soil apparent electrical conductivity and multivariate geostatistical techniques. Soil and Tillage Research 2010, 106, 335–343. [Google Scholar] [CrossRef]
  11. Tardaguila, J.; Stoll, M.; Gutiérrez, S.; Proffitt, T.; Diago, M.P. Smart applications and digital technologies in viticulture: A review. Smart Agricultural Technology 2021, 1, 100005. [Google Scholar] [CrossRef]
  12. Córdoba, M.; Bruno, C.; Costa, J.; Balzarini, M. Subfield management class delineation using cluster analysis from spatial prin- cipal components of soil variables. Computers and Electronics in Agriculture 2013, 97, 6–14. [Google Scholar] [CrossRef]
  13. Ortega, R.A.; Santibáñez, O.A. Determination of management zones in corn (Zea mays L.) based on soil fertility. Computers and Electronics in Agriculture 2007, 58, 49–59. [Google Scholar] [CrossRef]
  14. Xin-Zhong, 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. Computers and Electronics in Agriculture 2009, 65, 168–175. [Google Scholar] [CrossRef]
  15. Morari, F.; Castrignanò, A.; Pagliarin, C. Application of multivariate geostatistics in delineating management zones within a gravelly vineyard using geo-electrical sensors. Computers and Electronics in Agriculture 2009, 68, 97–107. [Google Scholar] [CrossRef]
  16. 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. Agronomy Journal 2004, 96, 100–108. [Google Scholar] [CrossRef]
  17. Peel, M.C.; Finlayson, B.L.; McMahon, T.A. Updated world map of the Köppen–Geiger climate classification. Hydrology and Earth System Sciences 2007, 11, 1633–1644. [Google Scholar] [CrossRef]
  18. FAO. World Reference Base for Soil Resources; Food and Agriculture Organization of the United Nations, World Soil Resources Reports N 103; FAO: Rome, Italy, 2006. [Google Scholar]
  19. Serrano, J.; Shahidian, S.; Paixão, L.; Marques da Silva, J.; Moral, F. Management zones in pastures based on soil apparent electrical conductivity and altitude: NDVI, soil and biomass sampling validation. Agronomy 2022, 12, 778. [Google Scholar] [CrossRef]
  20. Serrano, J.; Shahidian, S.; Marques da Silva, J. Spatial and Temporal Patterns of Apparent Electrical Conductivity: DUALEM vs. Veris Sensors for Monitoring Soil Properties. Sensors 2014, 14, 10024–10041. [Google Scholar] [CrossRef]
  21. Moral, F.J.; Serrano, J.M. Using low-cost geophysical survey to map soil properties and delineate management zones on grazed permanent pastures. Precision Agriculture 2019, 20, 1000–1014. [Google Scholar] [CrossRef]
  22. AOAC. Official Methods of Analysis of AOAC International, 18th ed.; AOAC International: Arlington, VA, USA, 2005. [Google Scholar]
  23. Moral, F.J.; Rebollo, F.J. Characterization of soil fertility using the Rasch model. Journal of Soil Science and Plant Nutrition 2017, 17, 486–498. [Google Scholar] [CrossRef]
  24. Moral, F.J.; Rebollo, F.J.; Serrano, J. Estimating and mapping pasture soil fertility in a portuguese montado based on a objective model and geostatistical techniques. Computers and Electronics in Agriculture 2019, 157, 500–508. [Google Scholar] [CrossRef]
  25. Bond, T.G.; Fox, C.M. Applying the Rasch Model: Fundamental Measurement in the Human Sciences, 2nd ed.; Lawrence Erlbaum Associates Inc.: Mahwah, NJ, USA, 2007. [Google Scholar]
  26. Linacre, J.M. Winsteps (Computer program and manual); MESA Press: Chicago, USA, 2009. [Google Scholar]
  27. Ferrari, P.A.; Salini, S. Complementary use of rasch models and nonlinear principal component analysis in the assessment of the opinion of european about utilities. Journal of Classification 2011, 28, 53–69. [Google Scholar] [CrossRef]
  28. Edwards, A.; Alcock, L. Using Rasch analysis to identify uncharacteristic responses to undergraduate assessments. Teaching Mathematics and Its Applications 2010, 29, 165–175. [Google Scholar] [CrossRef]
  29. Moral, F.J.; Rebollo, F.J.; Serrano, J.; Carvajal, F. Mapping management zones in a sandy pasture soil using an objective model and multivariate techniques. Precision Agriculture 2021, 22, 800–817. [Google Scholar] [CrossRef]
  30. Bode, R.K.; Wright, B.D. Rasch measurement in higher education. In Higher Education: Handbook of Theory and Research; Smart, J.C., Tierney, W.G., Eds.; Agathon Press: New York, NY, USA, 1999. [Google Scholar]
  31. Verdugo-Vásquez, N.; Acevedo-Opazo, C.; Valdés-Gómez, H.; Pañitrur-De la Fuente, C.; Ingram, B.; García de Cortázar-Atauri, I.; Tisseyre, B. Identification of main factors affecting the within-field spatial variability of grapevine phenology and total sol- uble solids accumulation: towards the vineyard zoning using auxiliary information. Precision Agriculture 2022, 23, 253–277. [Google Scholar] [CrossRef]
  32. Rodríguez-Pérez, J.R.; Plant, R.E.; Lambert, J.-J.; Smart, D.R. Using apparent soil electrical conductivity (ECa) to characterize vineyard soils of high clay content. Precision Agriculture 2011, 12, 775–794. [Google Scholar] [CrossRef]
Preprints 86062 g001
Figure 1. Study site: (a) Blue dots indicate the locations where soil apparent electrical conductivity (ECa) was measured; (b) digital elevation model.
Figure 1. Study site: (a) Blue dots indicate the locations where soil apparent electrical conductivity (ECa) was measured; (b) digital elevation model.
Preprints 86062 g001
Preprints 86062 g002
Figure 2. Kriged map of soil apparent electrical conductivity (ECa).
Figure 2. Kriged map of soil apparent electrical conductivity (ECa).
Preprints 86062 g002
Preprints 86062 g003
Figure 3. Management zones based on soil apparent electrical conductivity (ECa) and elevation. Sampling locations used for validation, with their Rasch measure, are also shown as dots.
Figure 3. Management zones based on soil apparent electrical conductivity (ECa) and elevation. Sampling locations used for validation, with their Rasch measure, are also shown as dots.
Preprints 86062 g003
Table 1. Mean values of each soil property for samples in the less productive zone (Zone_less), the intermediate (Zone_medium) and the most productive zone (Zone_more) according to the delineation of the homogeneous zones.
Table 1. Mean values of each soil property for samples in the less productive zone (Zone_less), the intermediate (Zone_medium) and the most productive zone (Zone_more) according to the delineation of the homogeneous zones.
Sand
(%)		 Clay
(%)		 Silt
(%)		 OM (%) pH P2O5
(mg kg-1)		 K2O
(mg kg-1)		 SEB
(cmol kg-1)		 CEC
(cmol kg-1)
Zone_less 77.64 a 10.75 a 11.61 a 1.13 a 5.96 a 146.55 a 104.94 a 3.30 a 14.51 a
Zone_medium 70.29 b 14.05 b 15.66 b 1.77 b 6.62 b 370.48 b 97.49 b 9.52 b 15.29 b
Zone_more 68.48 c 13.08 c 18.44 c 1.19 a 6.60 b 19.85 c 32.95 c 14.98 c 21.28 c
OM – Organic matter; P2O5 – Assimilable phosphorous; K2O – Assimilable potassium; SEB – Sum of the exchange bases; CEC – Cation exchange capacity (CEC); Different letters indicate significant differences (p < 0.01) according to the Dunn test.
Table 2. Mean values of each soil property for samples in the less productive zone (Zone_less), the intermediate (Zone_medium) and the most productive zone (Zone_more) according to the delineation of the homogeneous zones.
Table 2. Mean values of each soil property for samples in the less productive zone (Zone_less), the intermediate (Zone_medium) and the most productive zone (Zone_more) according to the delineation of the homogeneous zones.
Rasch measure
Zone_less -1.89 a
Zone_medium -0.50 b
Zone_more -0.59 c
Different letters indicate significant differences (p < 0.01) according to the Dunn test.
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.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated