We have examined a variety of sampling strategies for detecting pathogens in turkey flocks undergoing infections with low pathogenicity avian influenza virus (LPAIV). We found that viral RNA was widely distributed in the barn environment of turkey flocks undergoing an active LPAIV infection and was in both water and drinker biofilm samples. Viral RNA was concentrated in drinker biofilm and sediment and was detectable using real-time reverse-transcription polymerase chain reaction (RRT-PCR) and by virus isolation. Drinker biofilm sample results correlated with concurrently collected oropharyngeal (OP) sample results from flocks on a farm with LPAI in which the two sampling strategies were directly compared. To evaluate the utility of biofilm sampling for the detection of highly pathogenic avian influenza virus (HPAIV), biofilm and OP swabs from mortality pools were collected daily from negative turkey flocks on an HPAI-positive premise. The biofilm swabs were positive 1-2 days prior to positives appearing in the OP sample pools. The drinker biofilm sampling strategy overcame the difficulty of finding a subclinical infectious bird in a population by collecting material from a large number of individuals and testing a sample in which a positive signal persists for several days to weeks. The sampling method is convenient for use in turkey barns and has been reliably used in both active and passive surveillance programs for LPAIV and HPAIV using RRT-PCR.