Utilizing Soil Electrical Characteristics for Precision Viticulture Practices

As we have recently discussed on The Academic Wino, soil quality and characteristics have a significant effect on the flavors, aromas, and quality of the wine that is created from grapes grown in that soil. Soil characteristics are just part of what constitutes the concept of terroir, together with other factors such as grape variety, environmental characteristics, and vineyard management practices. The soil is a major factor in terroir, and has been shown to affect growth, grape yield, and grape composition as well as the composition of the finished wine.

Studies have shown that variable soils throughout a vineyard will produce different wines depending upon where it is in the vineyard the grapevines were located. Traditional methods for testing soil variability are considered to be destructive and very time consuming. Additionally, testing soil variability by this method could be problematic, as it’s difficult to determine exactly where to take the soil samples as well as the accuracy of the subsequent map of soil variability. To address this, using technologies such as remote sensing could help identify the variation in the soil in order to help cue the soil tester where he or she should collect samples.

One new method for improving the identification of variability in the soil in the past several years is the use of electrical resistivity or electrical conductivity as a proxy for the physical and chemical characteristics of the soil. In essence, by

Photo By Megan Mallen (Flickr: Rhône Valley - Châteauneuf-du-Pape) [CC-BY-2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

Photo By Megan Mallen (Flickr: RhĂ´ne Valley – Châteauneuf-du-Pape) [CC-BY-2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

testing electrical resistivity or conductivity, one could determine the composition of that soil, which could be extremely useful in soil and vineyard management planning.

The two types of resistivity/conductivity sensors in the market today are classified as non-invasive electromagnetic induction system (EMI) sensors, and invasive electrode based direct current (DC) resistivity sensors. There are pros and cons to using either sensor, which depending upon your study system, one or the other may be more appropriate. For example, while the down side to using DC sensors is that they require a solid contact and do not function well under dry or frozen conditions, they do seem to be more “friendly” to use in the vineyard. EMI sensors require repeated calibrations throughout the day, since they are very sensitive to temperature and humidity changes, and they are also limited by their sensitivity to electric power fields (power lines), running motors, and interference cause by interactions with metal things like trellis wires, guide wires, and irrigation pipes. Overall, EMI sensors are not ideal for vineyard soil work, while DC sensors appear to be much more adaptable to the variable environment throughout the day.

One type of DC sensor that is currently available today is the Automatic Resistivity Profiler Soil Sensor On-The-Go (ARP©) device which was designed specifically for use in agriculture and allows for quick and reliable soil mapping by measuring the electric resistivity of the soil below. ARP has already been shown to function well in a vineyard setting, amidst the various wires and other stumbling blocks that plagued the EMI devices. Very briefly, the ARP works by measuring the electrical resistivity of the soil below it at different depths, and interacts with topographical software (specifically, a Digital Elevation Model) to help determine soil variability and identify definitive management areas within a vineyard.

The study presented today aimed to examine the use of the ARP DC sensor device in a vineyard setting and how the results of the zonal delineations identified by the ARP correlated with vine trunk circumference and crop yield.

Methods

The study vineyard was located in Logroño (La Rioja, Spain) at a 3.5 hectare dry-farmed commercial vineyard planted in 1990 with the Vitis vinifera variety, Tempranillo. Standard vineyard management practices for the area were practiced. The study vineyard area was defined to have a 7% linear concave SE-facing slope with Inceptisol soil further sub-categorized into Petrocalciccal cixerept, Typical cixerept, and Calcic cixerept soils.

Using GPS, 65 experimental blocks were defined in the vineyard. Within each block, three vines in two adjacent rows were chosen for measurement collection.

The DC recording unit used in this study was the Automatic Resitivity Profiling sensor unit (ARP© from Geocarta in Paris, France and was programmed with four settings: rolling electrodes to optimize signal quality; a resistivimeter to measure electrical resistivity; Doppler radar to identify location; and an absolute positioning system (DGPS) to provide further accuracy to locations for soil zone identifications.

Figure 1 from Rossi et al (2013) showing the experimental set up for the ARP On-The-Go sensor unit.

Figure 1 from Rossi et al (2013) showing the experimental set up for the ARP On-The-Go sensor unit.

To test soil electrical resistivity, the APR unit was towed behind a vehicle through the vineyard along transects that were 5.6 meters apart for a total of 7,887 meters of vineyard and 115,510 measurements. The total time elapsed from start to finish was 43 minutes, at an average speed of 9.96kg/hour (that’s 6.2mph). The electrical resistivity of 3 different soil depths was measured: 1) 0.5m; 2) 1.0m; and 3) 2.0m.

Prior to harvest, the designated vines in each experimental block was measured for trunk girth at a location right above the grafting insertion point of the vine. At harvest, the yield was measured for each vine and then averaged with the other vines in the experimental block (for a total of 65 values).

Linear regression and fuzzy cluster analysis were used for analyzing the data, with the assistance of the Management Zone Analyst software.

Results

  • Trunk girths of vines were quite variable, ranging from 10.67 cm to 18.50 cm in diameter.
  • Crop yield was also quite variable, ranging from 1.22 kg per plant to 4.27 kg per plant.
  • The electrical resistivity of the top soil layer (down to 0.5m) was negatively correlated with trunk girth and crop yield.
    • In other words, the higher the electrical resistivity of the top soil, the smaller the trunk girth and crop yield and the lower the electrical resistivity of the top soil, the larger the trunk girth and crop yield.
    • To think of this another way, as electrical resistivity goes up, electrical conductivity goes down, so less electrically conductive soils result in smaller vines and lower crop yields than soils with higher electrical conductivities.
  • The lowest electrical resistivity values were found in the shallowest layer of soil, while the highest electrical resistivity values were found in the deepest shallow layer of the soil that was tested.
  • The highest electrical resistivity was found in soils that had many rock fragments at the surface, while the lowest electrical resistivity was found in soils that were finer in texture and richer in organic matter.
    • These findings are supported by other studies which found that soils near hills and ridges (more well-drained, coarser texture) had lower electrical conductivity (i.e. higher electrical resistivity) while soils lower in the landscape (lower sediment size and greater organic matter content) had higher electrical conductivity (i.e. lower electrical resistivity).
  • Trunk girth was found to be negatively correlated with slope, and was found to fit the following linear regression model (p<0.01):

Trunk girth = 16.843 – 0.049 x Electrical Resistivity – 0.307 x Slope

  • What the above equation means is that if you know the electrical resistivity of the soil in a particular area as well as the slope, you can plug those two values into the equation to get the predicted value of the vine trunk girth grown in that spot (at least under these experimental conditions).
  • The pattern of trunk variability throughout the vineyard was predictable by electrical resistivity and slope nearly exclusively.
  • Similar to trunk girth, crop yield was also found to be negatively correlated with slope and electrical resistivity, and fit the following linear regression model (p<0.05):

log(Crop Yield) = 1.276 – 0.008 x Electrical Resistivity – 0.322 x Slope

  • Areas found to be high in electrical resistivity corresponded to soils that were more highly concentrated with rock fragments.
  • Average trunk girth values were significantly different between soil zones, with the thinnest vines found in areas with the highest electrical resistivity.
  • There was a trend that trunk girth increased with decreasing electrical resistivity, though the variability was high enough to consider that perhaps other factors were at work as well, including things such as weather and the presence of pests (this was speculation).

Conclusions

The results of this study showed that those spatial/location patterns based on electrical resistivity was significantly correlated with vine size (trunk girth) and crop yield in Tempranillo vines. Similarly, trunk girth and crop yield were also highly correlated with slope, which has been shown in previous studies. Therefore, the plant characteristics of trunk girth and crop yield seem to be dependent upon the permanent physical characteristics of the soil.

This study showed that using DC sensor technologies, such as the ARP© (Automatic Resistivity Profiler Soil Sensor On-The-Go), one could reliably and accurately predict the soil variability in the vineyard, which is critical in determining certain quality features of the plant, including trunk size and crop yields. With very little information (i.e. electrical resistivity and slope), this study was able to accurately predict the trunk size and crop yields of the Tempranillo vines in the experimental vineyard, which if you ask me, is pretty cool (in a nerdy sort of way…).

I think using this type of DC sensor technology could be a great investment for vineyard managers, as it appears to be simple and quick to use, and also seems to provide accurate information in terms of identifying different soil zones within

Photo taken at Fox Meadow Winery in Linden, Virginia (copyright RYEAMANS 2011)

Photo taken at Fox Meadow Winery in Linden, Virginia (copyright RYEAMANS 2011)

the vineyard, which could help vineyard managers determine what varieties to plant in certain places in the vineyard, and also could potentially help in the restructuring of existing vineyards that aren’t performing to their full potentials. Before I turn this into a full blown advertisement of the product, one should note that certainly more research should be done before making sweeping conclusions. Yes, the results of this study were certainly telling, however, how does this technique hold up under different climate conditions? As the weather changes, does this technique become less reliable at predicting soil zones? A more long-term study is needed to determine if this is a technique that can be used reliably in variable climatic conditions, or if it only functions well under specific meteorological conditions. Finally, testing other grape varieties is essential, as it’s possible that we may see varied results from one grape variety to the next, due to the great variability in plant physiology between different grape varieties.

I think this technology could be extremely useful in the field of precision viticulture, and could potentially change how vineyard selection and management is done. Maybe some of you have already played around with this type of technology?…please feel free to leave your comments!

Source: Rossi, R., Pollice, A., Diago, M-P., Oliveira, M., Millan, B., Bitella, G., Amato, M., and Tardaguila, J. 2013. Using an Automatic Resistivity Profiler Soil Sensor On-The-Go in Precision Viticulture. Sensors 13: 1121-1136. (NOTE: This article is open access, so click on the link and download the PDF if you’d like to read the original paper!)

4 comments for “Utilizing Soil Electrical Characteristics for Precision Viticulture Practices

  1. July 2, 2013 at 9:33 am

    This is so interesting! Soil can make a huge difference in grown foods and I’ve heard that some wineries are offering dirt tastings in conjunction with wine tastings to showcase how the flavor of wine is affected. I also like the idea of using DC sensor technologies. Excellent post!

    • Becca
      July 7, 2013 at 8:39 pm

      Dirt tastings??? As in, you eat dirt? ;) I’m intrigued! ;)

      I think DC sensor technology could be really helpful based on the research so far–could make analyzing soils a lot faster, for sure!

      Thank you for reading and commenting!

  2. Bubba Beasley
    October 22, 2013 at 9:29 am

    Hey, I’m a geologist currently doing similar research funded by the Virginia Wine Board and just stumbled upon this post… Cool! I’m investigating the use of EM and GPR as vineyard prospecting and management tools and our results have been extremely positive. I’ve got data from a bunch of different VA vineyards and what we’ve seen so far has been incredible… I’d love to talk about my work sometime if you’re interested!

    • Becca
      October 28, 2013 at 6:57 am

      I am definitely interested!! Feel free to shoot me a message: becca@academicwino.com

Comments are closed.