First Conference on Proximal Sensing Supporting Precision Agriculture

The improvement in crop yield, both in quantity and quality, depends on the adoption of appropriate management strategies for the agronomic and irrigation practices. The adoption of site-specific (SS) management practices is fundamental, not only to improve crop yield, but also for a more efficient use of resources, increasing the environmental sustainability of the agricultural production The SS management requires the delineation of sub-regions with similar yield limiting factors or similar soil properties affecting yield (Site Specific Management Units – SSMU). It is a common practice in precision agriculture (PA) to characterize the spatial variability of soil properties, measuring the soil electrical conductivity through non-invasive electro-magnetic (EM) sensors to obtain high-resolution soil maps for the delineation of SSMUs.

Because of the expanding use in the future of the multi-frequency EM sensors in order to more effectively assess the soil variability, the objective of this work is to compare the measurements collected by Geonics EM38 (the most widely used EM sensor in PA) and GSSI Profiler-EMP400 (a multi-frequency EM sensor) in order to assess their reliability to delineate SSMUs. The data from 2-D electrical resistivity imaging were used to compare the response of the two different sensors to soil variability.

Electromagnetic induction (EMI) methods are widely used for mapping soil properties at the field scale, including soil moisture (θ), with benefits for precision agriculture. Recent studies showed ambiguous results on the influence of θ on the measured bulk electrical conductivity (ECa). At the Schäfertal hillslope site (Germany), spatial distribution of soil properties as well as the spatial and temporal dynamics of θ patterns were studied in detail within an intensive hydrological monitoring. At the same time, repeated EMI surveys were conducted. with the aim to asses the suitability of repeated EMI surveys for mapping θ, and to investigate the limitations of its applicability for soil mapping. Results suggest that i) stable soil properties have the major effect on the EMI signal, while θ plays a minor role; ii) soil moisture may strenghten the ECa pattern when local soil properties control the θ pattern; iii) when the contribution of other factors is considerable, ECa patterns related to stable soil properties may be partially hidden. In conclusion, EMI measurements carried out under unsuitable hydrological conditions may lead to misinterpretation of soil properties, therefore combined pedological and hydrological expertise is necessary to support proximal soil sensing campaigns, with benefits for precision agriculture practices.

We adopted electromagnetic induction (EMI) and self-potential measurements along irrigation canal bank in order to detect groundwater seepage. The EMI methods is sensitive to the electrical conductivity of the subsoil, that is affected by the salt and clay content and water content. The depth of investigation depends on the soil electrical conductivity, the geometry of the instrument and the frequency of the electromagnetic field.

The experimental data refers to anintegrated survey along a transect on an irrigation canals (400 m). We acquired broadband electromagnetic data using a Profiler EMP400 (GSSI), with coil separation of 1.22 m and a bandwidth ranging from 1 to 15 kHz.

The self-potential measurements were performed with a fixed electrode and a moving electrode along the same transect of EMI measurements. The interval spacing among self-potential measurements was 2 m. Visual inspection of the canal permitted to compare the evidence of seepage losses with the geophysical anomalies. The results of self-potential measurements show peaks of electrical potential at the same position of electrical conductivity anomalies. Integration of the EMI survey and SP measurements is useful to detect seepage losses from canals: a more detailed analysis of self potential anomalies is required to reduce “false” alarms.

The aim of this paper was to analyse the potential of laboratory Vis–NIR spectroscopy to determine organic carbon and nitrogen in a representative forest area of the Calabria region (south Italy). To do that, calibration models based on laboratory Vis–NIR spectroscopy and PLSR analysis were developed separately for soil organic carbon (SOC) and nitrogen (N). Soil samples (0–20 cm-depth) were collected at 216 locations, oven-dried and passed through a 2 mm sieve and and analyzed to estimate SOC and N concentrations. Subsequently the Vis-NIR reflectance of each soil sample was measured in laboratory, under artificial light, using an ASD FieldSpec IV 350 – 2500 nm spectroradiometer (Analytical Spectral Devices Inc., Boulder, Colorado, USA). In order, to develop models based on soil spectra and reference laboratory data of SOC and N, Partial least squares regression (PLSR) was used. Before applying the PLSR, spectra data were split into a calibration (144 samples) to develop the models and a validation set (72 samples) to assess the prediction accuracy of the calibration models. Results revealed a high level of agreement between measured and predicted values with high R2 and RMSE; model validation with independent data was satisfactory for both the studied soil properties.

Rapid, precise and quantitative assessment of soil quality is crucial for sustainable evaluation and monitoring of the effects of management on soil resource under agricultural systems and for the characterization and monitoring of land degradation processes. Over the past three decades, Visible (VIS) and near-infrared (NIR) spectroscopy have been shown to be an effective alternative to conventional laboratory analysis, and can provide time and cost effective approaches for the prediction of several soil properties related to soil quality indicators. For this study, VIS-NIR spectroscopic and chemometric analysis were employed for the assessment of soil quality indicators in three degraded areas (two survey sites depleted in organic carbon and one polluted by organic and inorganic compounds) located in Southern Italy (Apulia Region). The soil reflectance properties in the wavelengths range between 350 – 2500 nm were measured in three experimental sites(fields) selected for the project, before and after a recovery treatment by using compost (organic fertilizer). The objectives was to evaluate the efficiency of soil VIS-NIR spectra for prediction of selected soil indicators closely related to soil quality in Mediterranean areas, such as those investigated in this study, affected by land degradation processes (contamination and/or organic carbon impoverishment).

The influence of fertilizer on apparent electrical conductivity (ECa) was investigated in laboratory and in field. Long-term experiments are well suitable to study the changes in soil organic matter and its influence on ECa, but also on-farm experiments over one season cause a variability in biomass and yield and consequently in electrical values.

The glyphosate is an herbicide which is widely used in Argentina. The impact of its application on soil and groundwater is still unknown. The objective of the paper was to identify, anomalies of apparent electrical conductivity ECa associated with the presence of glyphosate in the soil or groundwater.

Electrical resistivity tomography (ERT) and electromagnetic (EMI) soundings were performed at a rural zone: a background site without any application of the herbicide (Chac), another plot with application of glyphosate (Ran), and another plot (Scho) at a zone with frequent application of the herbicide.

The results of ECa in the soil obtained with both methods do not show great contrasts between Chac and Ran. At Scho, the values of ECa were greater than other sites, 70.5 mS/m (EMI) and 60 mS/m (ERT). Glyphosate reached 52 ng/g in soil and 0.94 ng/l in groundwater.

The increase of the apparent electrical conductivity was mainly associated with the site that has been exposed to glyphosate application steadily over time, through the sprays that are made around it.

The study has presented an integrated approach to viticultural soil characterization using geophysical and geochemical methods. The purpose of this work is to evaluate the presence of geophysical and geochemical anomalies in soil within a vineyard located in well-known Cinque Terre Zone, North-Western Italy. To this scope soil and bedrock features have been studied with electromagnetic induction (EMI) and spettrophotometry method: both the methods have allowed to realize maps of conductivity anomalies values and major chemical elements distribution respectively for the investigated area. Although still at a preliminar stage, the results presented here are encouraging, indicating a good agreement between EMI and EDXRF data. The geological heterogeneity of the study area is clearly reflected both in the conductivity data and in the geochemical composition of the near-surface soils examined. The approach applied here can be extended to larger agricultural areas used by the agrofood industry, either alone or with the aim of integrating data from standard approaches based on pedological analysis. Correlation among data of different nature can nowadays be simply examined through the use of GIS software

Soil Electrical conductivity (ECa) are not directly related to the soil attributes used to make management decisions. Nevertheless, this information provide an opportunity to oriented soil sampling based on its pattern, rapidly and at a relatively low cost. This paper discuss how Kriging with external drift (KED), using low-density orientated soil sampling, based in ECa, can be used to provide reasonable soil attribute maps. The results are compared with other 4 different strategies. An example of lime recommendation at variable rate is presented. The results show that KED had clear advantages over the other methods when using fewer sample points.

Farmers adopting precision viticulture usually subdivide the vineyards through mapping the vigour of the grapevines. However, the reasons of the low vigour of the grapevines are often unclear and not associated to soil functionality. The spatial variabily of soil functionality within vineyards can be mapped at high detail by proximal soil sensing methods.

This case study want to test the use of different type of sensors, such as gamma-ray spectrometer, georesistivimeter and electromagnetic induction sensor, to map two large homogeneous zones (HZs) within two vineyards.

The quality of the wines produced in each HZ was tested for three consecutive years (2012-‘13-‘14).