Efeito em cianobactérias filamentosas das condições hidráulicas da coagulação e floculação na presença ou não de pré-oxidação

• Ano de defesa: 2019
• Autor(a) principal: Souza, Allan Clemente de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Saneamento; Cianobactérias; Água - Tratamento; Fisher exact test; Odds ratio; Logistic binomial regression; Filamentous cyanobacteria; Cylindrospermopsis raciborskii; Dolichospermum circinale
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/40056

There is a lack of information in the literature on the possible harmful effects of the pre-oxidation, coagulation and flocculation steps in Dolichospermum circinale and Cylindrospermopsis raciborskii, cosmopolitan and potentially toxic cyanobacteria. Therefore, the present study evaluated, in the presence and absence of pre-oxidation, the impacts caused by coagulation and flocculation in D. circinale and C. raciborskii. For this purpose, bench scale experiments were performed to simulate the mechanical effects of coagulation (rapid mixing) and flocculation (slow mixing) steps in the presence or absence of the oxidants Cl2 or KMnO4 and using the recommendations of the velocity gradient and mixing time suggested by international technical standards. For statistical analysis, two methods were compared. In the traditional approach (Fisher exact test and odds ratio - OR), it was appropriated to identify the association between causes (experimental conditions) and effect (cell integrity), respecting the specific characteristics of each species. In the multivariate approach using logistic regression, the estimated ORs required a differentiated interpretation in relation to the traditional approach, since the estimates were adjusted for the effect of all variables of the model. Thus, the cell integrity results of one species influenced the odds of the other one. Additionally, the results demonstrated that only hydraulic conditions of rapid mixing (velocity gradient of 750 s-1 for 60 s) reduced the chances of D. circinale intact cells to less than 50% compared to baseline conditions. Already in the presence of 1 mg.L-1 of either of the two oxidants, these odds decay to less than 5%. The results showed that C. raciborskii was immune to the effects of hydraulic conditions of fast and slow mixing (velocity gradient of 70 s-1 for 14 min) even in the presence of 1 mg.L-1 of Cl2 or up to 4 mg .L-1 of KMnO4. However, with 2 mg.L-1 of Cl2, the chances of finding intact cells after rapid or slow mixing reduce to less than 6% of the initial experimental conditions. Hydraulic stresses can also significantly reduce the length of D. circinale trichomes in both the absence and presence of oxidants, a fact also observed in C. raciborskii only in the presence of the pre-oxidation stage. The results obtained indicated that the hydraulic conditions recommended by technical standards could cause lysis only in the strains of D. circinale, but not in C. raciborskii. This emphasizes that both design and operation of water treatment plants should be adjusted according to the dominant species in the reservoir, especially when the pre-oxidation stage is carried out.