ABSTRACT1. There is a need to relate changing river flows to ecological response, particularly using methods which do not require extensive new data for water bodies without historical data. This paper describes how local-scale habitat features and changing discharge appear to influence a macroinvertebrate-based biotic index.2. The study used 87 time-series of river biomonitoring data from upland, wadeable streams with quasi-natural flow regimes across England and Wales. Twenty-seven of the sites were matched to a nearby flow gauging station, and historical, natural flows using a generalized rainfall-runoff model were synthesized for 60 sites. All sites were matched to a River Habitat Survey (RHS) within 1 km.3. The data were analysed using multilevel linear regression, combining sample-and site-level characteristics as predictors. Common responses were assessed across sites, using the biotic index LIFE (Lotic Invertebrate index for Flow Evaluation), an average of abundance-weighted Flow Groups which indicate the relative preference among taxa for higher velocities with gravel/cobble substrates or slow velocities with finer substrates. The aim was to understand the influence of physical habitat on macroinvertebrate response to antecedent high and low flow magnitude.4. There was a positive relationship between LIFE score calculated from spring and autumn samples and antecedent high (Q10) and low flows (Q95). The relationship between summer Q10 and autumn LIFE score was steeper than the relationship between winter Q10 and spring LIFE score. Bed and bank resectioning reduced overall LIFE and increased the steepness of the response of LIFE to low (Q95) flow.5. The models derived may be used to guide environmental flow allocations and to quantify the relative influence of flow and physical habitat change on macroinvertebrate responses. The interaction between resectioning and low flow has particular implications for the conservation of macroinvertebrate taxa with requirements for faster flowing water.

River flow regime is believed to have a fundamental effect on riverine biota. It influences key aquatic 10 processes, including levels of dissolved oxygen, sediment transport and deposition, water quality and 11 habitat type and distribution. We review the impact of flow on the abundance and distribution of 12 salmonid fishes in the context of developing approaches to regulating, setting and restoring river flow 13 regimes as a means of conserving and managing populations. Flow can have direct impacts on salmonids, 14 both through peak flow resulting in the washout of juveniles, and stranding of all life stages under low 15 flow conditions. Salmonids can also be adversely affected through indirect effects of flow, from impacts 16 on water temperature, dissolved oxygen condition, sediment deposition, and habitat availability. Early life 17 stages, particularly eggs and larvae, appear particularly susceptible to the adverse impacts of flow, since 18 they have a limited capacity for behavioral responses to altered flow conditions. A constraint to 19 conservation and management efforts for salmonids is in selecting river flow targets at the catchment 20 scale with confidence. Most studies linking flow with salmonid population processes are site specific, and 21 may not be readily transferable to other sites. Despite this uncertainty, the requirement for catchment 22 level flow targets has become critical as pressure on water resources has intensified, at the same time that 23 salmonid populations have declined. Our proposal is that hypothesis-led analyses of broad scale long-24 term datasets are key to quantifying variability in fish abundance with respect to flow and informing flow 25 modification field experiments. The water industry, conservation organizations, and environmental 26 regulators are charged with collaboratively tackling the question of how to set, manage and restore river 27 flow parameters, within the framework of the emerging science of hydroecology.

The concept of defining environmental flow regimes to balance the provision of water resources for both human and environmental needs has gained wide recognition. As the authority responsible for water resource management within England, the Environment Agency (EA) uses the Environmental Flow Indicator (EFI), which represents an allowable percentage deviation from the natural flow to determine where water may be available for new abstractions. In a simplified form, the EFI has been used as the hydrological supporting component of Water Framework Directive classification, to flag where hydrological alteration may be contributing to failure to achieve good ecological status, and to guide further ecological investigation. As the primary information source for the EFI was expert opinion, the EA aims to improve the evidence base linking flow alteration and ecological response, and to use this evidence to develop improved environmental flow criteria and implementation tools. Such tools will be required to make predictions at locations with no or limited ecological monitoring data. Hence empirical statistical models are required that provide a means to describe observed variation in ecological sensitivity to flow change. Models must also strike a balance between generic and local relationships. Multilevel (mixed effects) regression models provide a rich set of capabilities suitable for this purpose. Three brief examples of the application of these techniques in defining empirical relationships between flow alteration and ecological response are provided. Establishment of testable hydrological–ecological relationships provides the framework for improving data collection, analysis, and ultimately water resources management models. This article is categorized under: Water and Life > Conservation, Management, and Awareness

In some taxa, males perform multiple ejaculations, which may function in sperm competition or in maintaining a baseline density of spermatozoa in the female reproductive tract to ensure fertilization, a process that has been termed 'topping up'. We investigated the function of multiple ejaculations in two species of bitterling, the European bitterling (Rhodeus amarus) and Chinese rose bitterling (Rhodeus ocellatus). Bitterling oviposit in living freshwater mussels, with fertilization taking place within the mussel gill cavity. Thus, although fertilization is external, the mussel is analogous to the female reproductive tract in an internally fertilizing species. We measured the frequency of ejaculations and mussel inspections by individual males of two bitterling species in 28 replicated mesocosms and examined focal male responses to rival ejaculations and the presence of females in spawning condition. We used a model of ejaculatory behaviour to simulate the temporal abundance of spermatozoa in mussels. Male R. amarus exhibited high rates of ejaculation and inspection of the siphons of mussels and increased their ejaculation rate in response to the presence of females in spawning condition. Rhodeus ocellatus showed lower overall rates of ejaculation, but significantly elevated ejaculation rate in response to rival ejaculations. The ejaculatory strategy of R. amarus is one that maintains a minimum level of spermatozoa in mussels, which is elevated when the probability of oviposition increases. In contrast, R. ocellatus engages more directly in sperm competition with rivals. We discuss these results in the context of the function of multiple ejaculations and male mating tactics.