Natural seawater and freshwater microcosms inoculated with pig manure were set up to determine the persistence of pig faecal microbial and chemical markers in these two types of surface water. The concentrations of Lactobacillus amylovorus, the Bacteroidales Pig-2-Bac 165 rRNA genetic marker, five stanols and the evolution of two ratios of stanols, R(1) (coprostanol to the sum of coprostanol and 24-ethylcoprostanol) and R(2) (sitostanol to coprostanol) were analyzed during two months along with the concentration of Faecal Indicator Bacteria (FIB). Pig manure was inoculated to unfiltered water microcosms incubated aerobically at 18 degrees C in the dark. The faecal contamination load represented by the concentrations of culturable Escherichia coli and/or enterococci remained for two months in the freshwater and seawater microcosms water column. These concentrations followed a biphasic decay pattern with a 97% reduction of the initial amount during a first rapid phase (<6 days) and a remaining proportion undergoing a slower or null second decline. The L. amylovorus marker and five stanols persisted as long as the indicators in both treatments. The Pig-2-Bac marker persisted 20 and 27 days in seawater and freshwater, respectively. The ratios R(1) and R(2) were in the range specific to pig manure until day 6 in both types of water. These results indicate that Pig-2-Bac, L. amylovorus and stanol ratios might be used in combination to complement FIB testing to determine the pig source of fecal pollution. However, stanol ratios are to be used when the time point of the discharge is known. (C) 2011 Elsevier Ltd. All rights reserved.