Guidance on Environmental Flow Releases from Impoundments to Implement the Water Framework Directive
Project Extension 2 – Validation of Managed Flow Standards
WFD82b
January 2008

EXECUTIVE SUMMARY

Background to research
Under the Water Framework Directive (WFD), member states of the European Union should aim to (a) achieve the objective of at least "Good Status", comprising Good Chemical Status and Good Ecological Status, in all bodies of surface water and groundwater, and (b) to prevent deterioration in the status of those water bodies.

SNIFFER research project WFD82 was completed in May 2007, and produced guidance on environmental flow releases from impoundments for the Water Framework Directive. It recommended thresholds of alteration (condition limits) from the natural flow regime which determined the risk of failure of Good Ecological Status. The condition limits determine the deviation, expressed as a percentage of the natural condition, of a set of flow regime descriptors. The condition limits recommended in WFD82 were chosen using expert judgement. WFD82 identified a clear need to improve the scientific basis of these condition limits using hydrological and ecological data. This study has used river flow and macroinvertebrate biology data to try to validate these proposed condition limits.

Objectives of research
  1. Agree an approach for the project, specifically the target biological group and the data that would be available to indicate generic ecological impacts of managed flow regimes.
  2. Collate from national agencies (Environment Agency – England and Wales, SEPA - Scotland, DOE NI – Northern Ireland) a matched hydrology-biology dataset suitable for analysis to answer the project aims and process the dataset to a form suitable for subsequent analysis. This includes
    1. hydrological data, including gauged flows and long term natural flow statistics estimated using Low Flows 2000 for downstream gauging stations and biology sites.
    2. Biology data as raw taxonomic data and indices, plus expected values for indices (Average Score per Taxon – ASPT and Lotic Invertebrate Indicator for Flow Evaluation) calculated using RIVPACS. 3.    
  3. Determine the degree to which the impact of a dam on a watercourse can be identified through differences between indices calculated from downstream and reference biology samples
  4. Determine the relationship between the magnitude of any impact to the biotic indices, and the magnitude of deviation from estimated natural values for the hydrological parameters identified as part of the WFD82 project
  5. Validate the WFD 82 flow standards against biology status classes for at least one quality element (overall project aim)
  6. Assess the suitability of the WFD82 flow parameters.
Key findings
  1. The project assembled a dataset from national monitoring sources. For selected sites downstream of impoundments, It paired observed (impacted) and expected (natural) hydrological data with observed biology data and expected biotic scores. Where possible downstream sites were also paired with reference sites upstream of the impoundment or on an unregulated tributary. The analysis was split into two parts; firstly considering upstream / downstream differences, secondly considering change in hydrological impact as one moves downstream.
  2. From the initial analysis of Environment Agency data it appeared that there was a consistent difference in O/E (observed/expected) biotic scores between the potentially impacted biology sites downstream of dams and associated reference sites. Although statistically significant, the reference / downstream difference may not be of practical significance given other sources of variation in the biology scores. The LIFE O/E score was a more precise indicator than APST O/E. For status class intervals based on ASPT, out of 16 impoundments considered, there were only three cases where the downstream site(s) failed to achieve good status but the reference sites did achieve good status.
  3. For the SEPA data alone, there was no evidence of a consistent upstream / downstream effect on O/E ASPT scores, although it should be noted that on average, the SEPA sites were less hydrologically impacted than those of the Environment Agency.
  4. For both SEPA and Environment Agency data there was clear evidence for significantly reduced variability in biotic scores downstream of impoundments. There could well be consistent trends in the raw biological data which lead to this reduced variability.
  5. Considering degree of hydrological alteration which varies between impoundments and reduces as one moves downstream, there were relatively few downstream biology sites with a greater than 40% alteration to the flow regime. There was no overall clear pattern between degree of hydrological alteration and the biotic scores. For the situations with multiple sampling sites downstream of a single impoundment there were some weak patterns, but there were not sufficient examples to make further conclusions. Hence overall, it was not possible to validate the WFD82 hydrological standards.
  6. There were relatively few sites with elevated flows. There did not seem to be a continuum of response of the biota from reduced flows, through minimally altered flows to elevated flows. However there is some weak evidence that sites with elevated flows did show lower variability in biotic scores.
Recommendations
  1. Further progress to define hydrological standards for impoundments will need an alteration of the current monitoring network, or additional monitoring sites.
  2. Additional biological data are required to improve representation of downstream attenuation of hydrological impact, in order to characterise reference sites and to improve spatial representation.
  3. Further analysis would also benefit from more complete characterisation of the releases from the impoundments; this would also enable more detailed temporal analysis of the biological data and better characterisation of changes in flow variability that may be driving the observed changes in the variability of the biological indices.
  4. This project has focused on using biotic indices as the response variable, and in particular ASPT. An alternative tool would be multivariate analysis. If this technique is applied to spatial and temporal biological data, it can be hampered by the natural differences in community composition between catchments. However the community response needs to be addressed and further work is definitely needed to define indicator taxa.
  5. Further work is needed to assess the suitability of the RIVPACS approach to defining expected values for indices. This is primarily because the at-site physical variables used in RIVPACS may themselves be impacted by impoundment. Because of the complexities in impacts of dam regulation, accompanying hydromorphological data are particularly valuable for assessing ecological impacts.
  6. Collation and analysis of data for algae, macrophytes and fish will undoubtedly improve our understanding of the impacts of impoundments.
Key words: water framework directive, compensation flow, dam, impoundment, flow regime, flow release, hydrological regime, validation, standards, rivers, biotic scores, monitoring, APST, LIFE.

Copies of this report are available from the Foundation, in electronic format on CDRom at £20.00 + VAT or hard copy at £25.00, less 20% to FWR members.
N.B. The report is available for download from the SNIFFER Website