Phytosociology and Behavior of Weeds in Maize as Influenced by Spatial Arrangements

Using the right spatial arrangement is a sustainable way to prevent or at least delay the emergence of weeds in the crop production. This study evaluated the influence of row spacing and plant density of maize on weed control based on the phytosociological survey. It was conducted on an Oxisol textured medium in a 400 m area under semiarid conditions. The hybrid maize 30F53YH was managed under a no-till cropping system with three types of row spacing (0.35 m, 0.50 m, and 0.70 m) and three plant densities (5.0 plants m, 6.5 plants m, and 8.0 plants m). The experimental design was randomized complete blocks with four replications in a factorial arrangement 3 × 3. The phytosociological survey of the weeds was randomly performed four times in each subplot, using the inventory square (0.5 m × 0.5 m). The collected data were analyzed using the R statistical program. Among the specimen’s families identified on the field, three of them need to be highlighted due to its high values of density, frequency, and dominance. These families were Fabaceae, Poaceae, and Amaranthaceae. Also, it was identified that the Leucaena leucocephala species may be classified as a weed, as it acted as an invasive species on maize. The weed control was greatly influenced by the interaction of both parameters rather than only row spacing or the plant density factor. The results showed that the reduced spacing and high crop population decreased the presence of weeds in the maize crop.


Introduction
Maize (Zea mays L.) is an annual grass crop belonging to the Poaceae family and represents one of the most cultivated cereals, especially in tropical, subtropical and temperate regions. The total production of maize in Brazil in the year of 2015-2016 reached 48.2 million tons and the state of Piauí represented 3.65% of that yield (CONAB, 2016). Increasing maize productivity depends directly on the technological level used in its cultivation, climate, fertilization and weed management (Barros, Faria, Tuffi Santos, Azevedo, & Governici, 2017). These factors may compromise crop development. Galon (2013) pointed out that density, frequency and dominance are the most significant parameters to describe the dynamics of weeds in agriculture. The most important species of herbs will be those with the highest number of individuals (density), wider distributed in the area (frequency), and those who are able to suppress the other species because of their growth and accumulation of faster mass (dominance).
In this context, it is fundamental to develop studies targeting the control mechanisms that are efficient in reducing weeds, since the research conducted for phytosociological studies in Brazil is still considered insufficient (Freitas & Magalhães, 2012). Therefore, the handling of the spatial arrangement of the crop of interest is fundamental in the control of weeds and constitutes a socially, economically and environmentally sustainable alternative. The present study had the objective of evaluating the influence of population spacing and density of maize plants on weed control based on the phytosociological survey.

Study Area and Experimental Design
This study was conducted on an Oxisol textured medium (Soil Survey Staff, 2014) in a 400 m 2 area under semiarid condition at the Federal University of Piauí located in Bom Jesus city, Piauí State, Brazil (09º04′47″ S; 44º19′37.60″ W; 285 m asl.). According to Köppen's climate classification (Alvares, Stape, Sentelhas, de Moraes Gonçalves, & Sparovek, 2013), the region is a Tropical zone with dry winter (Aw) and the average annual rainfall is about 880 mm. The maize hybrid 30F53YH was managed under a no-till cropping system from January 8, 2016 to April 17, 2016. The experimental design consisted of three types of row spacing (0.35 m, 0.50 m, and 0.70 m) and three plant densities (5.0 plants m -2 , 6.5 plants m -2 , and 8.0 plants m -2 ) in a factorial arrangement 3x3. Row spacing was the plot and plant densities were the subplot. This is a randomized complete block design and each subplot consisted of 4 rows with 3 m length. By vegetative stage (V2), the maize plants were thinned out to preserve the aimed plant density. Fertilization was performed according to soil analysis (Table 1), as well as the topdressing fertilizer in the initial phase of maize cultivation.    Note. OM = organic matter; AS = aluminium saturation; CS = cation saturation; BS = base saturation; ECEC = effective cation exchange capacity.

Field Collection
At the vegetative stage (V6), the phytosociological survey of the weeds was randomly performed in each subplot, four times, using the inventory square (0.5 m × 0.5 m) standard method, as explained by Braun-Blanquet (1979) and Erasmo, Pinheiro, & Costa (2004). Then, the species existent in the central point were counted and identified in field and laboratory, based on the Brazilian weed identification guide (Lorenzi, 2006). Additionally, the samples were air dried in the oven at 65 ºC and weighed in their total amount. The V6 vegetative stage of maize growth stage was chosen because of weed competition.

Statistical Analysis
Data were analysed using the R-package ExpDes (Ferreira, Cavalganti, & Nogueira, 2014) for better visualisation of the results. Moreover, the phytosociological parameters were expressed using the Mueller-Dombois and Ellenberg (1974) equations: Eleusine indic la (IVI = 52.1 at sequence wi Vol. 10,No. 9; pecies j occurs um of the abs Ad j is the abs sampled, Rd j i Aa j is the abs a k is the sum o e importance v ukey (p < 0.05) nsity and spac cks through bo formation to o four species w e were four sp y. One specim es ( Figure 1a   Note. Rf = Relative Frequency; Rd = Relative Density; Ra = Relative Abundance; IVI = Importance Value Index (IVI).
The specimens identified in row spacing of 0.70 m are summarized in Table 4. The foremost point is that A. tenella Colla showed higher importance value index into the three-plant density compared with others. The second relevant species was E. indica in the plant densities of 5.0 plants m -2 and 6.5 plants m -2 . It was only above L. leucocephala in the plant density of 8.0 plants m -2 . Thus, the total number of specimens for this treatment was ten.   The dry m inspection in all treatm

Discussion
As described on the results section, three specimen's families had high observations. These are Amaranthaceae, Poaceae and Fabaceae. The first one represents plants, predominantly herbs, that are common in temperate and tropical regions. These plants have a branched tap root. Representing Poaceae family is grass, usually herbaceous, that presents fibrous root. Last but not least, Fabaceae is also called Leguminosae and they are distributed mostly tropical and subtropical regions. During maize development, there may be an interspecific competition between weeds and maize, and intraspecific competition among maize. The interspecific competition occurs because of the low plant density and high row spacing. Just as weeds compete for water and nutrients, the main crop (maize) has its development affected resulting in yield losses. However, farmers can use this competition in favour of maize when they sow at high plant density and reduce row spacing achieving yield gains, and performing a physical weed control. One of the causes must be related to the rapid canopy closure between the rows by maize plants, in reduced spacing and in high population, struggling the germination and growth of weeds between the rows of the maize crop. The results described here are in agreement with those obtained by Kolb, Gallandt, and Molloy (2010). They studied physical weed control versus interspecific competition in organic spring barley and found improvements with respect to weed control at wide row strategy. Moreover, the narrower row and higher plant density approach of maize can reduce transpiration and photosynthesis of weeds. Barros et al. (2017) described in their study the physiological response of maize and weeds living under different densities and their results further support the idea of reducing row spacing and increasing plant density on weed control. The results of our study are in agreement with those of previous and recent research (Chauhan & Johnson, 2011;Johnson & Hoverstad, 2002;Mhlanga, Chauhan, & Thierfelder, 2016;Mohammadi, Ghobadi, & Shaekheh-Poor, 2012). The effectiveness of maize arrangement was tested by Mendes, Drews, Medeiros, Rosa, Gualberto, and Mielezrski (2017) in the same study area. Therefore, they achieved better yield gains with narrower row spacing (0.35 m) combining 8.0 plants m -2 . Statistically, the interaction of both parameters had more impact on weed control rather than only row spacing or the plant density factor. This is because of canopy closure. High plant densities can accelerate canopy development and consequently weed suppression resulting in low competition between weed and maize for water and nutrients, which are vital for maize growth under semiarid or any environmental conditions. Additionally, physical plant attributes can lead to a viable competition against weeds, such as early canopy closure (Bajwa, Walsk, & Chauhan, 2017;Fahad et al., 2015;Weerarathne, Marambe, & Chauhan, 2017). Another interesting point is, according to the results observed, we can assume that economic and environmental trends could be solved because of reducing herbicides applications. Hence, this can increase farmers' incomes and yield results contributing positively to the environment. Colbach et al. (2017) simulated different cropping practices using glyphosate-resistant maize. The authors concluded that combining weed control methods with different cropping practices can reach farmers and food production needs. Furthermore, selecting and analysing weed species abundance in the field may help to decide which combined approaches are sustainable, economic, environment, and social. Phytosociological survey not only help to identify weed specimen importance aiming to control them, but also have a crucial role identifying exotic species that were not consider weed, but can become one. Our survey, for example, allowed us to classify L. leucocephala as weed in this experiment. Even though Leucaena is a species applied in reforestation and/or feeding cattle under semiarid regions, they can be troublesome for farmers during maize development. Costa and Durigan (2010) conducted a study evaluating if L. leucocephala acts as a weed disturbing environmental equilibrium or not. They concluded that L. leucocephala was a weed. Our results support their research presenting Leucaena as an invasive species on maize crop. The Brazilian weed guide does not have this species on its catalogue and it certainly should be included in the next update.

Conclusion
Spatial arrangements with reduced spacing and high population decrease the presence of weed in maize. Moreover, the best weed control was reached with plant density of 8.0 plants m -2 in row spacing of 0.35 m. The recommended maize population that is 5.0 plants m -2 in a row spacing of 0.50 m had a good weed control, however, not as efficient as the treatment using 8.0 plants m -2 at the row spacing of 0.35 m. The statistics showed that the weed control suffers more influence from the plant density factor than the row spacing, or the interaction of both parameters. Among the specimen's families identified on the field, Fabaceae, Poaceae and Amaranthaceae deserve to be highlighted due to its high values of density, frequency and dominance. Additionally, Leucaena leucocephala was classified as a weed in the maize field.