Using Precision Agriculture to Manage Nematodes in Brazil
As promised in my first article, this month I am writing about some technical aspects related to the use of precision farming tools for the site specific management of nematodes, with a focus on cotton.
Phytopathogenic nematodes are among the main factors limiting the achievement of higher yields in the Brazilian Cerrado. Failure to adopt adequate management practices allows nematode populations to reach very high levels, causing yield losses that can exceed 50%. In the soybeans-cotton crop sequence, the more important nematodes are the soybean cyst nematode (Heterodera glycines) and the root-knot nematode (Meloidogyne incognita). The identification of these species can be done by observing the symptoms in the roots, but the laboratory analysis is important to confirm the diagnosis and to estimate the population levels of each species.
Once introduced in an area, it becomes very difficult to eradicate a nematode problem, and it is necessary to adopt practices of population reduction and minimization of potential damages. The most important practice is the use of resistant or tolerant crop cultivars. In the case of tolerant cultivars, it is important that the populations of the nematode are below some threshold, otherwise it must be associated with other control techniques, such as the use of nematicides in the seeding furrow.
The spatial distribution of nematodes is characterized by the presence of circular or oval-shaped infected spots, with marked spatial and temporal variability. The existence of these spots is related to the low mobility capacity of nematodes, which is a few meters in a year, unless they are transported by some disseminating agent, such as agricultural machines. However, temporal variability occurs because plants submitted to a high initial population may have reduced root growth, reducing the supply of food to the nematodes in that place and thus reducing the final population. Places with an intermediate population allow better establishment of the crop, greater supply of food and with that greater nematode population. This makes the spots with larger populations vary from one place to another in each season.
Typically, the problem starts as few and small spots in a field, but the use of susceptible cultivars can result in high nematode populations in a few years. Because of that, the early diagnosis of the problem and the adoption of practices to avoid greater losses is of great importance. This can mean adopting a particular practice across the whole area when only 5 or 10% of the field actually has problems.
Due to these facts, the above mentioned control tools — genetics and nematicides — represent opportunities for the use of precision agriculture, either as localized practices or through variable rate application.
The main limitation for adopting precision farming tools for nematode management is the difficulty of accurately mapping the spatial distribution of nematodes. In general the nugget effect is high and the range of spatial dependence is short, which requires large numbers of samples to perform a good characterization of the spatial distribution. In addition, the observation of temporal variability makes it necessary to identify the population of nematodes present in the area at a time close to sowing, since the most important information for decision making regarding the control methods to be used is the population of nematodes that may affect the crop in its initial development.
The collection of samples of roots and soil in high density for sending to the laboratory can be cost prohibitive, and alternative methods are necessary to represent the distribution of nematodes with the appropriate resolution level. Remote sensing, mainly with the use of multispectral images of high resolution, is one of the techniques that has been used. For this, it it is necessary to assume that high populations of the nematode cause a significant reduction of the biomass of the crop, which allows the use of vegetation indexes that present a good correlation with the biomass to indicate places with greater probability of occurrence of nematodes. The main limitations of this tool are related to the fact that other operational or phytosanitary problems can also cause the reduction of plant development in a similar way, and it is difficult to differentiate the cause of the variability. There is also the possibility that the nematodes are causing yield loss without biomass reduction. Thus, it is important the use of images is associated with field scouting to verify the relationship of low development areas with high nematode populations.
In this sense, the use of hyperspectral sensors associated with novel techniques of data analysis can allow the differentiation of the causes of variability and to isolate the sites with nematode problems with greater accuracy. While this does not become reality, the strategy that has been used is the use of images associated with dense field verification and the collection of some samples for laboratory confirmation. Field scouting uses a scale of symptom notes and rely on good people training so that the visual interpretation is well calibrated.
Once the spatial distribution of the nematodes is mapped, one can optimize the available control tools according to the risk of damage to the next crop. From a genetic point of view, varieties with resistance or greater tolerance to a certain species of nematode may have lower yield potential than another similar cultivar that is susceptible. Thus, it would be possible to use a cultivar with greater tolerance or resistance only in nematode-infested sites, and to use susceptible varieties in areas without the presence of the nematode. This practice, however, requires the availability of specialized equipment, such as a sowing machine capable of changing varieties in real-time.
The control with the use of chemical nematicides in the seeding furrow can be done locally and also at variable rates. Localized application will be done only in areas where the nematode population exceeds a certain limit. Variable rate application is based on the fact that the protection time provided by the nematicides is proportional to the application rate, and the tolerance of the plant to the nematodes increases as the size of the root system increases, by a dilution effect of the nematodes. Thus, higher doses are used in sites with larger populations of nematodes to protect plants for a longer period of time until they are able to support nematodes that are not controlled by nematicides. For this type of application, the necessary equipment is readily available in most farms.
The cost of a nematicide application in the seeding furrow for cotton crops is around US$150 per hectare, which represents around 6% of the production cost of the crop. With variable rate application, reductions of 40% in the volume of nematicide applied have been observed, mainly in the areas of a field where the nematodes are not yet present and the product does not need to be applied. The yield of cotton in the areas where this technology has been adopted has been equivalent or greater than in areas with flat rate application.
After two years developing this technology, we conclude that the tools for surveying variability need to evolve, but there are great possibilities for expanding its use, since the technique has proven to be viable. Considering the typical case, the investment of US$10 per hectare for the mapping of the spatial distribution of the nematodes can provide US$60 per hectare of savings in the amount of nematicide used, obtaining the same cotton yield. In addition to the US$50 benefit per hectare, the use of the site-specific application brings an environmental gain not yet quantified, since the nematicides used are highly toxic and its application in places where it is not necessary can contribute to the elimination of other organisms and cause biological imbalances.