Precision agriculture is key to reduced chemical use.

The majority of crop prices vary depending on supply and demand on the market. The only way for farmers to increase profitability is to decrease production costs without compromising crop yield or increasing the risk of the activity. Precision agriculture is aimed at increasing the efficiency of using machines and inputs to improve yield and profitability and decrease environmental impact1. Precision agriculture allows an improvement in agricultural production management, using practices within the field, and requires an accurate assessment of spatial variations in short scales of resolution2.

Humanity requires ecological management practices with a decrease in the volume of pesticides used in farm fields. Accordingly, there are three possible ways to reduce the volume of pesticides used3: reducing the number of spray applications during the cropping period, which can compromise crop yield; decreasing the applied dose, which can decrease the level of control and promote the appearance of genetic resistance; and restricting the treated area. The third strategy is the basic principle of site-specific application of pesticides using precision agriculture techniques4.

This technology has been applied in several agricultural inputs (e.g., while using herbicides), as there is spatial variability in the growth of weeds in the field5. The management of weeds according to the spatial variability of their growth reduces the use of herbicides due to application in areas where the infestation is above the level of the economic threshold6. The misuse of herbicides is related to at least three main problems: the application of herbicides at an inadequate phenological stage, herbicide application without considering the infestation level, and application of herbicides in places free of weeds7. The first and second problems can be solved via application of technical knowledge by the farmer or consultant; however, strategic planning and tools are feasible for solving the third problem. Thus, one possible alternative would be weed management using precision spraying techniques, thereby potentially reducing the volume of herbicides applied to the soil and the environment8.

Currently, the site-specific application of pesticides focuses mainly on herbicides9; however, the same technique can be used for other pesticides, such as fungicides3,10. The site-specific fungicide application measures the dose across the field surface according to the canopy density regarding preventive control strategies. Denser canopies receive a higher dose of fungicide via this strategy, assuming that a dense canopy creates a microclimate that is more conducive to diseases and contributes significantly to crop yield3. Studies about site-specific insecticide application are less prevalent than those about herbicides because the population dynamics of flying insects are very intense11. However, this type of site-specific management is also possible for insects, as long as the treated mapped area is bordered as a safe area for proper control12.

Application of the theory of site-specific application of pesticides depends on mapping the spatial variability of the targeted crop problem (weed, disease, crop foliar mass, or insects). Site-specific weed control is more common and can be performed using two primary methodologies: weed mapping, followed by preparation of a prescription map containing volume rates of herbicide; and mapping and application in real-time9.

Weeds have an irregular distribution (irregular growth in terms of phenology and density), implying that some field areas are below the economic threshold level. However, the enormous challenge is to map the weed spatial variability in large areas in a short time (days) because they commonly proliferate rapidly in the field and can render the generated maps unusable4. The use of remote sensing tools can assist farmers in obtaining maps in a short time. There are several ways to map the spatial variability of weeds using remote sensing, which can be done either by satellite images, by unmanned or piloted aircraft, or by real-time detection sensors13. However, onboard satellite sensor platforms are often unsuitable for mapping weeds because of their low spatial and temporal resolution2. Studies with unmanned aircraft carrying cameras have been developed for weed mapping14; however, some processing time of the images in the laboratory makes the process impractical for its application in large areas.

Weed mapping using cameras still faces the problem of differentiation between weeds and crops because of the similarity between their spectral signatures15. Differentiating between plants and soil is relatively easy, as their spectral signatures are distinct; however, differentiating different plant species is not a simple process and often requires more expensive sensors, such as hyperspectral sensors16. Another strategy for performing the site-specific application of pesticides is through the use of sensors in real-time. The use of real-time sensor equipment for the instant detection of plants offers advantages in its use because of the spraying application at the same time as the target mapping, in addition to the advantage of performance in a single operation, without labor costs, data analysis, and the aid of a prescription map15. A study by Biller et al.17 was among the first studies with a focus on site-specific spraying of herbicides in real-time and proposed using electronic sensors to detect weeds. The sensors can distinguish the green color of the plants and the soil once plants have different reflectance characteristics. Dammer and Wartenberg5 achieved a reduction of 12.7% to 40.9% using site-specific weed control, compared to conventional spraying in the total area.

Methodologies for site-specific treatments have been developed to determine the spatial variability of plants and are mainly associated with weeds. However, using this equipment with sensors in real-time can detect and map weeds and the variability of planting patterns of commercial crops, i.e. crop rows and interrows. Thus, the focus of the current study was the evaluation of site-specific spraying application of pesticides other than herbicides. The current study aimed to quantify the reduction of pesticides applied to soybean and maize crops in several stages of the production cycle using a site-specific spraying application based on real-time sensors in the Brazilian Cerrado region.