How to improve the quality of seed distribution

​Study highlights the need to monitor the operation to identify sources of variability intrinsic to the process

13.11.2020 | 20:59 (UTC -3)

Study shows that although seed distribution is an operation carried out largely in a satisfactory manner, there is still a need to monitor the sowing operation to identify sources of variability intrinsic to the process, in order to reduce faulty spacing and improve seed quality. distribution.

Brazil is expected to harvest a production of 234,4 million tons of grains in the current harvest, an increase of 25,6% compared to the last harvest. Soybeans should have a production of 113,9 million tons. The crop's production gains reflect a 19,4% increase in total production (Conab, 2017).

This growth is due to soil management techniques such as direct planting, which consists of homogeneous management that aims to improve soil quality and, consequently, maintain or increase crop productivity to satisfactory levels. However, this improvement does not manifest itself homogeneously across the entire area, as within the same crop there may be sub-areas with different soil qualities and productive potential, which implies variability in crop productivity (Amado et al.

Also read

Spatial variability or heterogeneity in grain productivity can be associated with a series of factors that interact in a complex way and condition the expression of the crop. The study of the spatial variability of soil and plant attributes and crop productivity is fundamental for understanding the factors that determine the expression of the crop's productive potential and its variability in an agricultural area, which can lead to the development of crop management practices. site-specific management, aiming to maximize the productive potential in different crop zones (Vian et al.

In this context, precision agriculture tools must be used to search for information in the field, either through the use of agronomic sensors or through the use of soil sampling meshes and/or agronomic components of the crop, in order to optimize agricultural processes and increase agricultural income (Soares Filho & Cunha, 2015).

In addition to this, the use of process quality tools, such as statistical quality control, has been of great value in agricultural processes to verify the stability of the process according to pre-specified limits, helping to monitor and control sources. of variability (Voltarelli et al, 2015; Green arch et al.

Therefore, the objective was to map the quality of soybean sowing in a commercial area, analyzing the plant stand and longitudinal distribution, using descriptive statistics and geostatistics as tools.

The work was conducted in a commercial area of ​​Fazenda São Bento in the municipality of Itaporã (MS), whose shape and altitude spatialization are shown in Figure 1. The location is located at latitude W 54 07´ 01" and longitude S 22 01´ 06". The climate is type (CWa), according to the Köppen classification, and the soil in the area is a distroferric red Oxisol and is evaluated at stage v2.

Figure 1 - Plot used and altitude of the area (m)
Figure 1 - Plot used and altitude of the area (m)

The variety sown was a Monsoy 6410 soybean seed, with 98% purity and 92% minimum germination. In the area in question, direct planting has been used for 12 years, the seeder used is from the Semeato brand, a FastFil model with 30 rows spaced 60cm apart, which uses a pneumatic system and distributes 16 seeds per linear meter.

The area analyzed has 132 hectares and was divided into 71 plots (points), called sampling point (PA), as characterized in Figure 2.

Figure 2 - Map of the area and sampling points
Figure 2 - Map of the area and sampling points

Georeferencing of sample points for seeding uniformity analysis was carried out using the GNSS application available for the Android operating system, using metric coordinates (UTM WGS84).

The distribution of sampling points for sowing uniformity analyzes and soil sample collection was done in a regular grid with 200 meters between cells. This regular grid with the respective coordinates was inserted into the receiver, the points were located in the area to be studied and the sowing lines were identified. At each point, three sowing lines were analyzed, each measuring two consecutive meters.

Samples of the number of soybean seedlings emerged at stage V2 were collected, two consecutive meters and in three lines at each sampling point.

In the evaluation of longitudinal distribution or uniformity of spacing between seedlings, a measuring tape was used. The percentage of normal, faulty and double spacing was obtained in accordance with ABNT (1984) and Kurachi standards. et al (1989), considering percentages of spacing: "double" (D): <0,5 times the Xref. reference spacing; “normal" (A): 0,51,5 o Xref.

Initially, the data were analyzed using descriptive statistics and then, to verify spatial dependence, data interpolation and construction of maps, geostatistical analysis was used.

Result

For the plant stand, it is observed that the mean and median are close, as well as low values ​​of the coefficient of variation (CV) (Table 1). However, despite the low variation, this component affects crop productivity, due to the uniform distribution of plants in the area (Sangoi et al, 2012; Vian et al

For the variables related to longitudinal distribution (Table 1), it was observed that the spacing, flawed and double, had a relatively close mean and median, however, with high CV. However, for normal spacing, low CV rates were observed. Low variability between spacings is common for soybean sowing, which is the opposite in values ​​obtained for the longitudinal distribution of plants in corn crops (Santos et al, 2011; Green arch et al.

The variability of the stand was verified (Figure 3), with a predominance of bands of 12 and 13 plants and 13 and 14 plants/m, which occupied just over 70% of the area, with the desired 15 plants.

Figure 3 - Mapping of plant stands per meter
Figure 3 - Mapping of plant stands per meter

Regarding the distribution of normal spacings, there is a greater predominance of bands between 73% and 83% (Figure 4), in more than 85% of the area, which for a pneumatic seeder-fertilizer represents acceptable performance.

Figure 4 - Spatialization of the longitudinal distribution for normal spacing (%)
Figure 4 - Spatialization of the longitudinal distribution for normal spacing (%)

It is worth highlighting the predominance of percentage bands of failed spacings between 10% and 20% in more than 85% of the area (Figure 5A) and, less representatively, the distribution of double spacings with a predominance of bands below 10% in 85% of the area (Figure 5B). Therefore, the operational points related to the increase in faulty spacings must be checked, in order to increase the plant stand and the percentage of normal spacings corresponding to the regulation of seed distribution by the seeder-fertilizer.

Figure 5 - Longitudinal distribution maps (A – failed; B - double)
Figure 5 - Longitudinal distribution maps (A – failed; B - double)

This result highlights the need for monitoring and monitoring the operation, aiming to identify and control points that reduce the quality of the agricultural process, such as an inadequate disc and/or ring for the hybrid sieve, improper pressure in the pneumatic system, lack of or excess graphite, seed treatment with high abrasiveness, positioning of seeds within the furrow, pest attack, soil-seed contact made difficult by the amount of straw in the direct seeding system, inadequate soil moisture for sowing, opening and closing of the furrow ( Weirich Neto et al.

Conclusions

The plant stand and normal spacing distribution were considered good, but not excellent for a pneumatic seeder.

The plant stand and the longitudinal distribution of normal, faulty and double spacing for soybeans were spatially dependent.

There is a need to monitor the sowing operation to identify sources of variability intrinsic to the process, in order to reduce faulty spacing and improve distribution quality.


Mateus Augusto Estevão, Jorge Wilson Cortez, Sálvio Napoleão Soares Arcoverde, Anamari Viegas de Araújo Motomyia, Maiara Pusch, UFGD


Article published in issue 175 of Cultivar Máquina

Mosaic Biosciences March 2024