SANTOS, E. B.; SANTOS, Emanuelle Bezerra dos.
Resumo:
Population growth, climate change, and the degradation of water and soil resources have negatively impacted global food security. To address this challenge, it is crucial to adopt sustainable practices in food production and seek alternative sources of protein. In this context, duckweed emerges as a promising solution. This versatile aquatic plant can be cultivated hydroponically, minimizing water consumption and reducing the need for chemical fertilizers. Furthermore, duckweed is rich in essential nutrients, making it a healthy food option. By using arribadas seaweed flour as a bio-stimulant in its cultivation, it is possible to further enhance the benefits of this crop, contributing to sustainable and nutritious production. For this study, the species Wolffia brasiliensis was identified, and the healthiest plants were selected for monitoring the cultivation, aiming to investigate the impact of the variable concentration of the bio-stimulant on the relative growth rate and doubling time calculated from the mass and number of fronds of duckweed. Four concentration levels (0%, 2%, 4%, and 6%) were analyzed, with cultivation and analyses performed in triplicate over eight days. The behavior of temperature (°C), pH, electrical conductivity (EC), salinity (PSU), dissolved oxygen (%DO), turbidity (FNU), and total dissolved solids (TDS) present in the nutrient solution was monitored. Based on the obtained data, a kinetic analysis of the parameters was performed. The Shapiro-Wilk normality test was conducted for the dependent variables, confirming that the relative growth rate and doubling time follow a normal distribution, allowing the use of parametric statistical methods. It was found that the relative growth rates increased with higher concentrations of the bio-stimulant, while the doubling rates decreased, indicating an influence of the factor on the responses. Linear, quadratic, and cubic statistical models were proposed to represent the relationship between the variables, and the results for all responses showed that the cubic model had the best coefficient of determination (R²), indicating a good fit quality. The obtained F-values were significantly higher than the tabulated values in the rejection region with a confidence level of 95%, validating the use of these models to predict the results of RGR and DT within the studied concentration range.