Aquatic connectivity: challenges and solutions in a changing climate.

The challenge of managing aquatic connectivity in a changing climate is exacerbated in the presence of additional anthropogenic stressors, social factors, and economic drivers. Here we discuss these issues in the context of structural and functional connectivity for aquatic biodiversity, specifically fish, in both the freshwater and marine realms. We posit that adaptive management strategies that consider shifting baselines and the socio-ecological implications of climate change will be required to achieve management objectives. The role of renewable energy expansion, particularly hydropower, is critically examined for its impact on connectivity. We advocate for strategic spatial planning that incorporates nature-positive solutions, ensuring climate mitigation efforts are harmonized with biodiversity conservation. We underscore the urgency of integrating robust scientific modelling with stakeholder values to define clear, adaptive management objectives. Finally, we call for innovative monitoring and predictive decision-making tools to navigate the uncertainties inherent in a changing climate, with the goal of ensuring the resilience and sustainability of aquatic ecosystems.

Monitoring for the adaptive management of rivers.

Monitoring for adaptive management (AM) involves collection of data with the aim of reducing uncertainty about links between human pressures (e.g. water abstraction from rivers), consequent stressors (e.g. low river flows) and environmental state (e.g. biodiversity). 'Surveillance monitoring' involves documenting trends in state, without the aim of understanding relationships between state, stressors, and pressures. Critics have highlighted that surveillance monitoring dominates monitoring investments but is not supporting AM. Decision-makers continue to be disappointed by monitoring data that are unsuitable for AM, yet designers of monitoring programs tend to make decisions that reinforce rather than reimagine the status quo. We argue that a structured, collaborative approach to objective-setting is required to break the status quo. We collaborated with regional management authorities to develop monitoring objectives and implementation strategies to support AM of New Zealand's rivers. Our collaborative approach discouraged 'failure fearing' and encouraged reimagining 'what could be' as opposed to 'what is.' Seventeen monitoring objectives were identified based on the AM requirements of national policy and regional authorities. Several objectives-particularly those arising from national policy-stretch the limits of what environmental science can currently provide. There were also strong trade-offs among objectives. We offer practical implementation strategies for overcoming the technical challenges of, and reducing trade-offs among, monitoring objectives. These strategies point to a monitoring program that contrasts strongly with one aimed at surveillance. Monitoring for AM is more complex than monitoring for surveillance, so strong leadership is required for successful implementation.

Estimation of policy-relevant reference conditions throughout national river networks.

A method for objectively estimating reference states for suspended fine sediment (turbidity) is presented. To be fit for water policy development and implementation the method had to satisfy four requirements: (1) the method must not be dependent on data from minimally-disturbed reference sites; (2) the method must facilitate characterization of reference states throughout heterogeneous river networks, given patchy data; (3) the classification of reference states must be relevant and legitimate to end-users; (4) the method should provide several classifications of reference states at different spatial resolutions allowing selection of the resolution yielding the most parsimonious classification of reference states throughout the network. Implementing the method involves two stages: (1) Development of a river classification based on sediment supply and retention regimes (defining 'turbidity classes') at multiple spatial resolutions. (2) At each resolution, for each turbidity class, estimation of a reference state based on relationships between turbidity and anthropogenic stressors, then objective selection of the resolution yielding the most parsimonious classification of reference states throughout the network. Implementing the method requires a river network GIS and turbidity data within classes, preferably from monitoring sites spanning the domains of the anthropogenic stressor variables used for model-based estimation of reference states.•A method is presented for estimating reference states for suspended fine sediment (turbidity) throughout spatially heterogeneous river networks.•Development of the method was guided by the requirements of policy analysts during reform of water policy in New Zealand.•The method presented was used to develop fine sediment regulatory thresholds of national water policy.

A comparison of passage efficiency for native and exotic fish species over an artificial baffled ramp.

This study used an experimental approach to compare the passage success of native and exotic fish species from the temperate Southern Hemisphere over an artificial baffled fish ramp designed for overcoming low-head (≤1.0 m) fish migration barriers. Passage efficiency was, on average, lower for the exotic species [koi carp (Cyprinus carpio), rudd (Scardinius erythrophthalmus) and rainbow trout (Oncorhynchus mykiss)] compared to the native species [inanga (Galaxias maculatus), redfin bully (Gobiomorphus huttoni) and common bully (Gobiomorphus cotidianus)]. Nonetheless, there was considerable variation between individual species, with rainbow trout outperforming common bully and juvenile inanga, but koi carp and rudd failing to pass any of the ramps. The differences in predicted probability of passage success between the native and exotic fish species in this study were sufficient in some cases to indicate the potential for the baffled fish ramps to operate as a selective migration barrier. Nonetheless, further testing is required to validate these results across a broader range of conditions before deployment.