Development of image analysis techniques as a tool to detect and quantify morphological changes in anaerobic sludge: I. Application to a granulation process

Image analysis techniques were developed and applied to quantify the process of anaerobic granulation in an expanded granular sludge blanket reactor (EGSB) fed with a synthetic substrate based on glucose [60-30% COD (chemical oxygen demand)] and volatile fatty acids (40-70% COD) over 376 days. In a first operation period that lasted 177 days, the aggregation of dispersed sludge was quantitatively monitored through the recognition and quantification of aggregates and filaments. A parameter defined as the ratio between the filaments' length and the aggregates projected area (LfA) has proven to be sensitive to detect changes in the aggregation status of the anaerobic sludge. The aggregation time-defined as the moment when a balance between filaments' length and aggregates, size was established-was recognized through the LfA. The percentage of projected area of aggregates within three size ranges (0.01-0.1 mm, 0.1-1 trim, and > 1 mm, equivalent diameter) reflected the granular size spectrum during the aggregation process. When sudden increases on the upflow velocity and on the organic loading rate were applied to the previously formed granules, the developed image analysis techniques revealed to be good indicators of granular sludge stability, since they were sensitive to detected filaments release, fragmentation, and erosion that usually leads to washout. The specific methanogenic activities in the presence of acetate, propionate, butyrate, and H-2/CO2 increased along the operation, particularly relevant was the sudden increase in the specific hydrogenophilic activity, immediately after the moment recognized as aggregation time. (C) 2004 Wiley Periodicals, Inc.