This research is a comparative study on the diversity of halophilic bacteria with hydrolytic activities in three significant hypersaline lakes; Urmia in the northwest and Howz-Soltan and Aran-Bidgol in the central desert in Iran. Isolated strains from these saline lakes were found to be halotolerant, moderately and extremely halophilic bacteria. The bacteria in each saline lake were able to produce different hydrolytic enzymes including amylase, protease, lipase, DNase, inulinase, xylanase, carboxy methyl cellulase, pectinase and pullulanase. 188, 302, 91 halophilic strains were isolated from Urmia Lake, Howz-Soltan and Aran-Bidgol playa, respectively. The numbers of Grampositive strains were more than Gram-negatives, and among Gram-positive bacteria; spore-forming bacilli were most abundant. Due to the unique physico-chemical conditions of the lake environments, the hydrolytic activities of isolated strains were significantly different. For instance, isolated strains from Howz-Soltan playa did not produce pectinase, DNase, amylase, lipase and inulinase, while the isolates from Aran-Bidgol playa had a great ability to produce pectinase and DNase. The strains from Urmia Lake were also good producers of DNase but failed to show any chitinase activity. The diversity of halophilic bacteria from the mentioned three saline lakes was also determined using PCR-amplified 16S rRNA followed by phylogenetic analysis of the partial 16S rRNA sequences.

No abstract

The physical level of interaction between fluid and structure can be either one-way or two-way depending on the direction of information exchange at the interface of fluid and solid. The former can be solved by a partitioned approach and weak coupling. In problems involving two-way fluid-structure interaction, using a partitioned approach and strong coupling, sometimes stability restriction is encountered. This is an artificial added mass effect, which is independent of the numerical time step. Unfortunately an accurate and efficient method to deal with all the different levels of interaction is scarce. Conventionally, relaxation is applied to remedy this problem. The computational cost is directly related to number of sub-iterations between fluid and structural solver at each time step. In this study, the source of this instability is investigated. A discrete representation of a basic added mass operator is given and instability conditions are assessed. A new method is proposed to relax this restriction, the idea essentially is to remove the instability source from the structure and move it to the fluid and solve it monolithically with the fluid. We call this an interaction law. An estimate of the structural response is derived from structural mode shapes. As a test case, a 2D dam break problem interacting with an elastic vertical flexible beam is selected. The interaction of fluid with the beam undergoes several stages. The breaking waves on the beam can increase the added mass drastically, therefore the added mass ratio increases as well. In such a cases, the asset of interaction law is better elaborated, while the stability condition requires very high relaxation without interaction law, but the relaxation can be lowered by only using first five beam mode shapes. As a consequence, the number of sub-iterations reduces by one order. The numerical observations confirm the reduction in computational time due to utilization of the interaction law.

No abstract