This document presents the results of a review of the performance and uses of Version I of SERCON (System for Evaluating Rivers for Conservation), and contains recommendations for revisions and enhancements to the system and software for incorporation into Version 2. The aspects which have been considered in this review include:

• a review of the current performance of SERCON as a tool for assessing the conservation value of rivers;

• a review of the field survey protocol, and in particular the links with River Habitat Survey (RHS); and the relationship between field survey recording and SERCON scoring;

• proposals for amendments to SERCON to improve its performance and ease of use, including the potential for upgrading the SERCON software and enhancing its database and analytical capabilities;

• identification of potential future applications for SERCON.

This Report presents the findings of Phase 1 of the SERCON Review Project, which was carried out under a SNIFFER contract by Young Associates and overseen by a Project Steering Group. More detailed input was provided by a small Project Development Group (PDG), and many other organisations (particularly the Environment Agency, EA) and individuals have contributed to the project. Particularly valuable contributions were provided by SERCON surveyors, assessors and other users at two workshops held in York in 1998, and the experiences and views of all these contributors have been taken into account in this review.

A number of individuals with in-depth knowledge of particular rivers were asked to review the SERCON scores and assessments. A variety of views were expressed by these consultees, but in general the responses indicated that the interpretation of SERCON outputs may present some difficulties for those who are not familiar with the system. To address this, it is recommended that expanded and clarified guidance on interpretation of outputs is included within the manual, and possibly reproduced as a stand-alone document which could be widely distributed.

A review of the time required to carry out the various elements of a SERCON evaluation found that field survey timings were fairly predictable. although there is some variation in different types and locations of rivers. The time needed for survey preparation and data collation/scoring, however. is much more variable and depends particularly on the level of co-operation received from the river agencies and other organisations. This means that it is important, before commissioning SERCON evaluations, to consider carefully the survey access and data supply requirements and to ensure that appropriate financial and time provision is made for these elements.

It is desirable for there to be some flexibility in the definition of the reaches of river for evaluation (ECSs) in order to take account of the findings of field survey and/or data analysis. For this reason it is proposed that approximate ECS boundaries are set prior to survey but that a facility to adjust the precise location by up to I km should be retained.

The effect of ECS length on SERCON scores has been tested. and no clear relationship emerged. It is therefore recommended that the rules for ECS length are not changed, but that the expanded score interpretation guidance should address the need to take account of ECS length.

Background Data

Most of the background information is relatively straightforward to obtain and collate. The main recommendations put forward concern amendments to the Water Quality bandings to accommodate changes in water quality classification, and to add a component relating to vulnerability to acidification. A list of soil categories for Northern Ireland is also recommended for inclusion.

A review of the scoring and performance of the three PDY attributes, which are all derived from River Habitat Survey (RHS) data, has been carried out. As a consequence, proposals are put forward for assessing substrates (PDY 1) and flow types (PDY 2 (a)) by applying diversity indices to RHS data. It is recommended that the remaining attributes (PDY 2 (b) and PDY 3) should be retained in their current form, although some changes to the score bandings will be needed. Expanded guidance on interpretation of PDY scores should be provided in the SERCON manual/software.

These two criteria have been thoroughly reviewed and restructured, as a consequence of general concern that in their current form there is a degree of overlap between them and that some of the scores do not adequately meet their objectives. In particular, most of the RE scores are not based on a robust typology, and the recommendation is that all attributes except the assessment of aquatic plants should be moved into the NA section, keeping macrophytes as the sole attribute in RE for the time being.

The new NA attributes which rely on RHS data will be changed so that the information can be obtained from the main RHS form, and in consequence the SERCON add-on component of the RHS form will be deleted. This will improve the recording of some aspects of NA which presented difficulties for many surveyors, but others will in future be scored from information gained through consultation. The recommended changes to the biotic NA attributes (NA 'B') are less far reaching, but the proposed new scores for invertebrates, fish and birds will incorporate the geographical elements from their RE equivalents to provide a more accurate reflection of naturalness than does the presence of alien species alone.

No major changes are proposed to this criterion, which presents few problems other than those associated with obtaining sufficient high quality data. Amendments to the species lists will be needed from time to time to reflect changes in legislation.

The scores in this criterion are generally straightforward to assess, but additional guidance on interpretation would be valuable. The only major changes proposed are the deletion of the fish and breeding birds scores, as these are covered by the relevant NA attributes.

Special Features (SF!

This criterion includes the assessment of a number of river features which are considered important to a conservation evaluation but which do not fit into the preceding categories. The range of attributes covered is quite wide, and consequently the recommended changes are varied. Two sets of attributes (SF 3/4 and SF 6/8) are to be amalgamated. giving single scores for floodplain habitats and 'other vertebrates' respectively. Changes to the score bands are recommended for on-line lakes, and to the score guidance for wintering birds. The marginal invertebrates attribute will be restricted to an assessment of suitable habitats. More radical changes are proposed to the riparian zone attribute (SF 2), although this needs further consideration and testing.

No satisfactory alternative to the 'expert subjective view' approach has been found for the scoring of impacts on rivers. The main recommendation made is that Impacts scores should use a simplified banding (0, 1, 3, 5) to equate to no impact and minor, moderate and major impacts which, whilst still subjective, may be easier to apply in a reasonably consistent manner.

No radical changes to AFIs are proposed, but as they are unscored, additional features may be added to the existing list without affecting SERCON scores or indices. It has been suggested that it would be useful to-include AFIs which relate to the presence of important assemblages of species in the river as a whole.

SERCON and RHS have evolved in parallel and, over the last three years, a number of changes have been made to the RHS form to improve its capacity for recording information required by SERCON. It is now intended that the field survey requirements for RHS and SERCON should become fully integrated, and to this end a number of changes to attribute scores are proposed to enable SERCON to use RHS data more effectively. Consequently, the SERCON add-on component will no longer be required. A small number of minor changes to the RHS form are still proposed, but these have been kept to a minimum and should not present RHS surveyors with any serious difficulties or require extensive testing.

Several analyses of survey sampling strategies have been carried during this review and in the earlier stages of SERCON's development. The results of these suggest that a random sampling strategy may not offer any significant benefits for SERCON assessments, and it is recommended that the current strategy of regular sampling (one RHS site every 2 km) is retained. The continued use of an inter-reach survey is recommended, but the existing format should be radically changed in favour of direct recording of a selected range of features. The form should be a single side of A4 and should include selected extracts from the main RHS form.

It is recognised that some additional training, beyond that provided by the RHS accreditation course. would be of benefit for SERCON surveyors and assessors. However, there is unlikely to be sufficient demand for a dedicated assessor training course and it is recommended that the emphasis is placed on providing as much assistance as possible through the manual and software.

There would be considerable benefit in developing a cut-down version of SERCON for use in situations where it is not feasible to carry out a full evaluation, such as in preliminary, wide-area or time-limited investigations. It is proposed that a carefully selected sub-set of physical and biotic attributes should be used in such ' SERCON-light' evaluations, and that the biotic elements (macrophyte assessments) should be weighted. A modified version of SERCON-light could be used for evaluating short river lengths, and the use of aerial video footage may have value for preliminary or wide-area assessments using SERCON-light.

SERCON Version I was developed to be used in both manual form and as a PC application, and the development of improved software is seen as a vital component of Version 2. Several alternatives for upgraded software are presented, but the favoured option is the development of a database which is closely linked to the RHS software but which can also be used independently. This should be capable of storing, interrogating and retrieving all the physical and biotic components of a SERCON evaluation. The RHS data would be shared with the RHS database, and the potential for creating electronic links with other databases (such as macrophytes and invertebrates data) could be explored. Whichever option is chosen, it is strongly recommended that the software automatically calculates attribute scores wherever possible.

The development of wider uses for the SERCON system will be the key to its continued use and future success. A number of potential uses have been considered; these include impact/risk assessments, Catchment Management/Local Environment Agency Plans, biodiversity plans and species/habitat management, planning and regulatory contexts and as a tool for use in environmental education or river rehabilitation assessments. There is also some potential for using SERCON beyond the UK, although this would require fairly major remodelling of the system.

Phase I of the review project has included a thorough assessment of the performance of SERCON Version 1, and as a consequence a number of recommendations for modifications and improvements are presented for discussion and decision. It is anticipated that Version 2 of SERCON will build on the experiences gained so far, but will be easier to use and interpret, more accurate in its assessments and will be supported by a comprehensive software package. It will also be more closely linked with RHS, but the opportunity should be taken to promote the additional benefits SERCON can offer in terms of holistic assessments incorporating both the physical and the biotic elements of rivers.

SERCON

RHS

River conservation

River habitats

River survey

Impact assessment

Conservation criteria

Diversity indices

Copies of the report are available from FWR, price £35.00, less 20% to FWR Members.