Initial Growth and Roots Development of Soybean as Function of Water Availability and Soil Bulk Density

Soil compaction is preponderant in soil physical-hydric relationships, which in turn, exert direct effect on plant development. In this context, this work aimed to evaluate the initial development of shoot and roots of soybean plants (Glycine max (L.) Merril), cv. BMX Ícone, cultivated in different combinations of soil bulk densities and water availability. A greenhouse experiment was carried out at the EMBRAPA Lowland Experimental Station, Rio Grande do Sul, Brazil. Soybean plants were grown in seven levels of soil bulk density (1.4, 1.5, 1.6, 1.7, 1.8, 1.9 and 2.0 kg dm) coupled to two soil water tensions (10 and 50 kPa). Plant height and leaf area, as well as root volume, decreased when soybean was cultivated at 50 kPa, associated to soil bulk densities above 1.8 kg dm. Soybean crop showed to be most sensitive to water deficit than to soil compaction, and soil water tension around the field capacity (10 kPa) should be associated to soil bulk density lower than 1.8 kg dm to allow adequate soybean crop development.

majorly determines photosynthesis rate and water use by plants; its productivity potential is severely inhibited when exposed to water deficit.
Another response to soil compaction occurs in plant root system. Ramos et al. (2018) evaluated soybean roots in areas with no compaction and with low and high compaction levels; according to the authors, three different forms of soil exploration were identified in soybean roots, according to the increasing soil compaction. The first one shows a pivotal behavior (control plants); the second is characterized by shorter primary roots and the presence of numerous secondary roots; and in the third type roots only explore a shallow zone of soil.
On the other hand, the proper development of the root system, especially in depth, promotes greater water capture ability. According to Gliṅski and Lipiec (2018), when the upper roots are water stressed, plants tend to maintain the transpiration rate by compensatory increase in water uptake from lower unstressed roots.
In this context, this work aimed to evaluate the initial development of shoot and roots of soybean plants, cultivated in different combinations of soil bulk densities and water availability.

Experimental area and design
The experiment was carried out in a greenhouse at Terras Baixas Experimental Station (ETB), Embrapa Clima Temperado, located in Rio Grande do Sul, Brazil. Soybean plants cv. BMX Ícone were cultivated in PVC pots (15 cm diameter, 25 cm height), being sowed in October 2017, with seven seeds per pot at a depth of 2 cm. Six days after emergence, thinning was done leaving the four most homogeneous plants.
Experimental design was randomized blocks with four replications in factorial scheme 7 × 2. Factor A corresponded to seven levels of soil bulk density (1.4, 1.5, 1.6, 1.7, 1.8, 1.9 and 2.0 kg dm -3 ) and factor B to two levels of soil water tension (10 and 50 kPa).
The procedure to obtain soil bulk density (BD) was to fill the pot with sieved soil in layers of 4 cm, placing in the volume corresponding to that layer the soil mass corresponding to the soil bulk density established, by physical compaction to obtain the correct mass:volume ratio. BD of 1.4 kg dm-3 corresponded to placing soil in pot without compaction. Soil was collected at ETB experimental field and classified, according to Santos et al. (2018), as Planosol. On field, at 10 cm depth, natural soil bulk density was 1.6 kg dm -3 .
In each pot, soil water tension (SWT) was monitored by using Watermark ® sensors, installed at 0.10 m depth (vertical sensor center). Watermark ® readings were performed every day, early in the morning; when necessary, water was added to restore predefined SWT for the treatment.

Measures
Plants height was measured from the soil surface to the youngest leaf point of insertion once every two days between 1 and 56 days after plant emergence (DAE). From 4 to 56 DAE, leaf area measurements were estimated twice a week by measuring the length and width of the central trifolium of each leaf and multiplying it by a correction factor, according to the method modeled by Richter et al. (2014).
At the end of the experiment (56 DAE), soil penetration resistance (PR) was determined by using an impact penetrometer Stolf type with three measurements per pot. After determination of PR, the soil of each pot was divided in 4 layers for the evaluation of root development, corresponding to depths of 0-5 cm, 5-10 cm, 10-15 cm and 15-20 cm. To separate roots, each soil layer was washed on a 2 mm sieve. After the separation, the root volume was evaluated by inserting them into a graduated cylinder with water and recording the change in volume, where 1 mL of water corresponded to 1 cm3 of roots.

Statistics
To evaluate the relationship between plant height and leaf area with soil bulk density and water tension, the locally weighted regression (LOESS) model was adjusted to each variable (Cleveland & Devlin, 1988). The effect of soil bulk density and water tension on roots volume was evaluated by comparing regressions based on their respective 95% confidence intervals (95% CI); in sections when they overlapped, there was no difference between regressions; where they did not touch each other, treatments differed. Similarly, for root distribution in soil layers, means were considered distinct when the difference between means was superior to 2 × [95% CI] (Reinhart, 2015). The soil resistance to penetration as a function of soil depth for different soil bulk densities was evaluated by 1 st or 2 nd degree linear regressions. All analysis were performed into the R statistical environment (R Development Core Team, 2016). jas.ccsenet.

Results
The growt plants cult this differe water defi kPa (Figur This decre the most expansion water avai availability   Figure 1); from than those gro the height of p lower than tho nd nutrient def er, 1995), and y is also a con l. (2013)    The highe lateral roo is in the zo In soil wat was simila higher in availability believed th by our data