River regulation and recruitment in a protracted-spawning riverine fish.

We present and test an extension of the "match/mismatch" hypothesis that attempts to explain the persistence, under conditions of flow alteration, of small, short-lived, native, riverine, fish species. The premise is that flow alteration typically changes environmental conditions, such as temperature and prey abundance, which may affect survival during the larval period of fishes. This "window-of-opportunity hypothesis" states that, if optimal conditions for recruitment vary temporally within a year, the probability that a proportion of the larvae of protracted-spawning species will encounter a period of optimal conditions is greater than for larvae with only a brief spawning period, and so the former will have a recruitment advantage. We determined whether all hatching events contributed equally to juvenile recruitment of the protracted-spawning Australian smelt (Retropinna semoni) during one breeding season in three pairs of heavily regulated and largely free-flowing unregulated rivers in the Murray-Darling Basin, Australia, and related patterns in the hatch dates of recruits to temperature or prey biomass for one pair. For all rivers, heavily regulated or not, recruits present at the end of the breeding season most commonly hatched in the latter part of the breeding season. Mortality of those fish hatched in the first part of the season likely explains this trend. Furthermore, while hatching times were similar for all rivers, each river showed a distinct pattern of hatching and recruitment, which may relate to the temperature range within which epigenetic processes are aligned. Patterns of zooplankton biomass differed between the largely free-flowing ovens and regulated Goulburn rivers and likely had different sources: within the channel and within the storage lake, respectively. For the Ovens River, recruits hatched subsequent to the period when the first significant increase in zooplankton biomass occurred. We hypothesize that temperature may largely influence the "window" during which recruitment can take place but that prey density, responding to river-specific interactions between temperature and discharge, plays a role in the timing and magnitude of recruitment of Australian smelt. We conclude that the match/mismatch hypothesis may be applicable to rivers, that the window-of-opportunity hypothesis has some currency and deserves further investigation, and that river regulation may have significant impacts on fish recruitment.