Aquatic macrophytes can be used for herbicide remediation provided they exhibit tolerance to the contaminants. This research assessed the remediation potential of Salvinia minima, Echhornia crassipes, and Pistia stratiotes, some common aquatic macrophytes native to Brazil, and their tolerance to atrazine, an herbicide commonly detected in waterbodies. Plants were cultivated under controlled conditions with five atrazine concentrations (0, 2, 20, 200, and 1000 μg L^-1) for 15 days. S. minima and E. crassipes tolerated atrazine concentrations equal to or less than 20 μg L^-1 and died at 200 and 1000 μg L^-1, indicating the herbicide's potential toxicity and its selectivity against sensitive species. P. stratiotes tolerated the herbicide concentration up to 200 μg L^-1 and had its growth reduced at 1000 μg L^-1. All species demonstrated the ability to reduce atrazine concentrations in water at 20 μg L^-1 or lower, E. crassipes being the most efficient, reducing concentrations by 43% and 22% at 2 and 20 μg L^-1, respectively. Atrazine levels within Brazilian (2 μg L^-1 by CONAMA 2005) and European (0.1 μg L^-1 by Directive 2013/33) regulatory limits do not selectively affect these species. Thus, they show potential for use in arazine phytoremediation programs.

Limited chemical control options and high onion sensitivity to post-emergence herbicides hinder efficient weed management, especially under direct-seeding systems in semi-arid regions, increasing production costs. Reduced herbicide rates and split applications may improve crop selectivity while maintaining weed control. This study aimed to evaluate the selectivity of post-emergence herbicides applied at different doses and phenological stages in onion grown under Brazilian semi -arid conditions. Four field experiments were conducted in a randomized block design with three replications, each evaluating a single herbicide (oxyfluorfen, flumioxazin, pendimethalin, or oxadiazon). Fifteen treatments were evaluated, combining herbicide doses with three application timings (1-2, 3, and 5 fully expanded leaves), plus an untreated control. Crop injury was visually assessed at 7, 14, 21, and 28 days after application (DAA). Mild phytotoxicity symptoms (< 10%) were observed mainly at early growth stages, particularly when applications were performed at the 1- or 2-leaf stage, while treatments applied at more advanced stages (3 to 5 leaves) caused little or no injury. Symptoms were transient, with recovery observed by 21 DAA. Sequential applications occasionally prolonged low -intensity injury but did not exceed acceptable levels. No significant differences in commercial yield were detected among treatments. Bulb classification was predominantly concentrated in Class 3 (50-70 mm), with values generally above 60%, indicating high commercial quality and uniformity. The results demonstrate that the evaluated herbicides were selective for onion under semi-arid conditions when applied at appropriate doses and growth stages, with early stages being more sensitive and later stages more tolerant, without compromising yield or bulb quality.