108 water to suspended matter, k1s describes the transfer from water to bottom sediments and k2 is the kinetic transfer coe?cient which describes radionuclide release from suspended matter or bottom sediments to water. Finally, ? is a correction factor which takes into account that some of the sediment particle surface may be hidden by other particles. Radioactive decay is described by the following equation: ?(? + ??)????? = ?(?)[1 ? exp (???] (VI-7) where C is the radionuclide concentration and ? is the radioactive decay rate. Radionuclide concentrations in seawater (Cw), suspended matter (Cs) and bed sediments (Cb) are calculated in the domain of interest by counting the number of particles as follows: ?? = ? ? ?? ?????? (VI-8) ?? = ? ? ?? ??????? (VI-9) ?? = ? ? ?? ??????? (VI-10) Here I = Q/NP, where Q is the source term and NP is the number of particles used in the simulation. ?x?y?z is the volume of each cell, m is suspended matter concentration, H is the mixing depth in the bottom sediment and ?b is sediment bulk density. Finally, Nw, Ns and Nb are the number of particles in each phase. Parallel techniques on Linux OS are used to reduce the simulation time for emergency response against a nuclear accident. Fastest processing times are achieved when the problem is divided into equally-sized chunks onto the available computer cores. However, splitting the mesh implies some e?orts which have to be considered. Each subdivision would have to pass particle information (ghosts) to each other because each particle exerts forces on all other particles. Also, particles that move out of a node boundary will have to be sent to the corresponding node. In LORAS, the interaction between particles can be ignored (between radionuclides) and Figure VI-1 shows a scheme on the particle distribution method through the masking.