Dilution effects on aggregates and filaments contents in automated image analysis methodologies O. Dias*, D. P. Mesquita*, A. L. Amaral* , **, E. C. Ferreira* * IBB - Institute for Biotechnology and Bioengineering, Centre of Biological Engineering Universidade do Minho, Campus de Gualtar, 4710-057, Braga, Portugal (E-mail: ecferreira@deb.uminho.pt) ** Instituto Superior de Engenharia de Coimbra, Instituto Politécnico de Coimbra, Rua Pedro Nunes, Quinta da Nora, 3030-199 Coimbra, Portugal Abstract Monitoring activated sludge processes by microscopic observations and image analysis is a well established technique with the utmost importance for microbial community characterization. Occasionally, there are biological systems operating with high biomass concentrations that need to be diluted, causing reproducibility problems for the sampling and image analysis methodologies. In the current work, image processing and analysis methodologies were used to determine the aggregates and filamentous bacteria contents and morphological characterization, of five dilutions from three activated sludge systems. The outcome of reducing the concentration of the samples was obtaining almost 3.5 times the pattern results for the total filamentous bacteria contents and almost the double for the total aggregates contents. Moreover, the differences were diverse for each aggregates class and, therefore, the dilution effects cannot be predicted and quantified for all of the aggregated biomass as a whole based on an overall correction factor. Keywords Activated sludge, image analysis, dilution, aggregates, filamentous bacteria INTRODUCTION Monitoring activated sludge processes by image analysis methodologies is, nowadays, a well established technique for the microbial community characterization. The balance between floc forming and filamentous bacteria is usually determined for the establishment of bulking events, which may induce biomass washout and foaming problems. Image analysis is particularly useful in determining bulking events (Jenne et al., 2007), although bulking problems can be looked upon as microbiological or engineering problems (Martins et al., 2004). Schuler and Jassby (2007) scrutinized the work of several authors and suggested a modified empirical model for filamentous contents and settling ability. According to Schuler and Jassby (2007) the previous belief of a threshold at approximately 150 mL/g sludge volume index (SVI) corresponding to 10000 mm/mL of filamentous bacteria is invalid and the notion of filamentous bulking above 150 mL/g of SVI is therefore, defied. However, these conclusions could be misled by incorrect evaluation of the work of the other researchers. In fact, some of the considered surveys could be conducted with diluted samples and, therefore, the end results would be inaccurate as it will be noticed further on. Image analysis is also used to determine settling ability and concentration properties through the categorization of the aggregated flocs (Grijspeerdt and Verstraete, 1997). Occasionally, there are biological systems that operate with high biomass concentration that needs to be diluted. When the dilution is not performed, aggregates and filaments may appear linked to each other leading to misleading information (da Motta, et al., 2002). Dilutions can be useful for microscopic inspection and especially for image analysis processing, leading to a more accurate evaluation of the activated sludge system. However, the dilution technique may also give raise to representative problems given the fact that the amount of screened biomass decreases. The aim of this study was to monitor the effect on the parameters provided by image analysis of increasing dilutions: aggregates morphology, total filamentous bacteria length (TL) and total aggregates area (TA). The image processing and analysis programs were developed in the