The 1(St) Step To Healthy Ecosystems: Application Of A New Integrated Assessment Framework Informs Stream Management In The Tukituki Catchment, New Zealand

There have been increasing calls for the holistic assessment of river condition, moving beyond the traditional focus on water quality and biology alone. Legislation in New Zealand now requires integrated assessments of freshwater ecosystem health. In this paper we report on the application of a framework for New Zealand freshwater assessment, which includes 5 core components-water quality, water quantity, aquatic biota, physical habitat, and ecological processes-in the Tukituki River catchment. Indicators of ecological processes included ecosystem metabolism and cellulose decomposition potential, which were measured and reported alongside other river-health metrics to provide an integrated picture of river ecological integrity at site, stream-type, and catchment scales. Every site in the Tukituki catchment failed to meet bottom-line benchmarks for at least 1 assessment metric. When aggregated by stream type, warm-wet lowland streams in the Tukituki River catchment had the poorest ecosystem health because of poor water quality and quantity. At the catchment scale, the Tukituki scored highly in overall ecosystem health despite 20% of the targeted network failing bottom-line benchmarks for 11 out of 22 metrics. This study demonstrates the value of: 1) a probabilistic sampling design for predicting the spatial extent of stream condition, 2) accounting for environmental variation through a process of data harmonization informed by site-specific reference conditions and bottom-line benchmarks, and 3) involving resource-management practitioners in the development of the assessment framework to ensure that the level of detail is achievable in practice and to provide examples of how the results can inform management actions. Applications could include directing land-use mitigations and stream restoration towards warm-wet lowland areas, where excess phosphorus and aquatic plants proliferate. Finally, we demonstrate how a report-card approach can simultaneously provide information across multiple components and avoid the loss of information. The framework is adaptable and can be used across contrasting spatial scales, it is consistent and representative, and it is easily understood.