Welsh Government M4 Corridor around Newport Draft Hazel Dormouse Mitigation Strategy

Transcription

1 Draft Hazel Dormouse Mitigation Strategy Draft 4 May 2017 This report takes into account the particular instructions and requirements of our client. It is not intended for and should not be relied upon by any third party and no responsibility is undertaken to any third party. Job number CVJV/AAR 3 rd Floor Longross Court, 47 Newport Road, Cardiff CF24 0AD

2 Contents Contents A EXECUTIVE SUMMARY 1 B INTRODUCTION... 2 B.1 Background to activity/development... 2 B.2 Proposed Works on Site to be covered by the Licence... 4 B.3 Actions requiring licensing... 5 C SURVEY AND SITE ASSESSMENT... 7 C.1 Existing information on hazel dormice... 7 C.2 Statutory sites notified for the species within 10 km... 7 C.3 Objectives of surveys... 7 C.4 Scaled plan/map of survey area C.5 Site/habitat description C.6 Field surveys C.7 Survey results C.8 Interpretation / evaluation of survey results D.1 Short-term impacts: disturbance D.2 Long-term impacts: habitat modification D.3 Long-term impacts: habitat loss D.4 Long-term impacts: fragmentation and isolation D.5 Post-development impacts D.6 Predicted scale of impact E WORKS TO BE UNDERTAKEN E.1 Site clearance methods E.2 Dormouse habitat E.3 Mechanisms for ensuring delivery of mitigation E.4 Mitigation contingencies E.5 Biosecurity risk assessment F POST-DEVELOPMENT SITE SAFEGUARD F.1 Habitat/site management and maintenance F.2 Population and habitat monitoring F.3 Post-development mitigation contingencies F.4 Mechanism for ensuring delivery of post-development works G TIMETABLE OF PRIMARY WORKS H LAND OWNERSHIP OF THE MITIGATION SITE H.1 Mitigation/compensation site ownership - construction H.2 Mitigation site ownership post-construction I ASSESSMENT AGAINST THE FAVOURABLE CONSERVATION STATUS (FCS) TEST... 66

3 J REFERENCES Welsh Government K ANNEXES Annex 1: Survey schedule Tables Table 1: Description of the 2014 Survey Sites Table 2: Description of Additional 2015 Survey Sites Table 3 Dates of nest tube installation and survey visits Table 4: Habitat Grading Table 5: 2015 Hazel Nut Search Areas and Results Table 6 Timetable of Primary Works Figures Figure 1: Desk Study Figure 2: Survey Areas (including Coed Mawr) Figure 3: Survey Results (including Coed Mawr) Figure 4: Interpretation of survey results: distribution of hazel dormouse populations Figure 5: Coed Mawr Habitat Map Figure 6: Displacement, relocation and translocation proposal Figure 7: Long-term Mitigation

4 A EXECUTIVE SUMMARY A.1.1 TO BE ADDED IN FINAL VERSION. Page 1

5 B INTRODUCTION B.1 Background to activity/development Introduction B.1.1 B.1.2 B.1.3 This sets out the proposals for the mitigation of likely effects on hazel dormice (hereafter referred to more simply as dormice ) during the construction and operation of the new section of motorway proposed as part of the (M4CaN) Scheme (the Scheme). This report will form the basis of the Hazel Dormouse Method Statement that would be prepared and agreed with Natural Resources Wales (NRW) in advance of construction, and would be submitted to NRW in support of the Scheme s application to NRW for a European Protected Species licence for this species. For these reasons, it has specifically been structured to reflect NRW s online guidance on the preparation of a licence method statement: Template Method Statement to be used within a Dormice Development Licence Application. This strategy has been developed in consultation with NRW, and consultation would continue during the development of the Hazel Dormouse Method Statement. B.1.4 The strategy has been informed by the results of dormouse surveys undertaken in 2014 and 2015 to inform the Environmental Impact Assessment (EIA) of the Scheme, as well as surveys completed in The methodologies and results of these surveys are summarised in this document. The final Hazel Dormouse Method Statement to support the NRW licence application would be informed by further surveys to be undertaken in 2017 and pre-construction survey to be completed in 2018, as described in this report. B.1.5 B.1.6 In their response to the Draft Orders (dated 4th May 2016), NRW objected to the scheme pending the provision of a comprehensive dormouse mitigation strategy. Whilst they subsequently welcomed the provision of the (as Appendix SS10.4 of the December Environmental Statement Supplement (ESS)), in their subsequent letter of 31st January 2017, responding to the December ESS, they considered that the Strategy does not set out how the proposals will ensure no detriment to the maintenance of the favourable conservation status of dormice in its natural range. As a result, they concluded that on the basis of the current information, NRW would not be able to issue an EPS licence for this species. Their concerns were summarised as follows: Lack of demonstration that impacts on affected dormouse populations will be fully mitigated; Insufficient information that Coed Mawr has sufficient good quality habitat to support a translocation programme; Lack of detail on the re-planting strategy, including phasing and maturity of stock to be used; and Concerns with respect to captive breeding and the potential, exceptionally, to translocate dormice back into the wild years after capture. B.1.7 This version of the has thus been updated both in response to the detailed comments provided in the letter of 31st January 2017 and following detailed discussions with NRW (specifically Dr Liz Halliwell, who reviewed the previous version of the document and who is due to provide evidence on dormice to the M4CaN Public Local Inquiry). These discussions led to the inclusion of a number of dormouse commitments within an Ecology Statement of Common Ground, which is being Page 2

6 developed in consultation with NRW. B.1.8 It was agreed that this more detailed version of the Strategy would be provided to NRW with a view to satisfying their concerns (as raised in their letter of 31 st January and in the Proof of Evidence of Dr Halliwell), and that it would include the following modifications: Detailed illustrations of all areas known to support dormice will be provided, each providing the following details: o o o o o o A more accurate estimate of the total amount of habitat to be lost; An indication of any habitat within the red line that could potentially be retained; An evaluation of the habitat with regard to its likely suitability for dormice; An indication of whether displacement or translocation is proposed, and whether this is likely to be in the Spring or Autumn seasonal window (or both); An estimate (for both the Spring and Autumn windows) of the potential numbers of dormice likely to be encountered; and For displacement, the proposed direction of displacement, the area of habitat into which they would be displaced, and whether or not enhancement of these areas is possible. A review (with the Land Use Expert Witness, Julia Tindale) of any non-welsh Government parcels of land (including blight land) adjacent to the scheme where the potential for habitat enhancement by agreement could be sought, in order to increase the potential for displacement; A detailed assessment of the likely implications of the Scheme with regard to ensuring that it will not be detrimental to the maintenance of the Favourable Conservation Status (FCS) of the dormouse population, specifically in relation to the four parameters set out by the JNCC - i.e. population, range, habitat (extent and condition), and future prospects ; Detailed habitat mapping of Coed Mawr indicating the numbers of dormice that could be accommodated (both pre-breeding and post-breeding); A review of the situation at Berryhill Farm and, specifically, whether the proposed 18ha of woodland planting can be considered as part of the dormouse mitigation habitat. This review will need to consider the potential for natural colonisation of this area, the extent to which a culvert underneath the proposed motorway embankment (along the line of the existing watercourse) would facilitate this, and whether or not translocation of dormice into this area would be acceptable as an alternative to natural colonisation. Currently this area is considered not to contain dormice, and the repeat inspections of the nest tubes in this area in 2017 will confirm either presence or likely absence. B.1.9 B.1.10 All of these commitments are addressed in this updated version of the Mitigation Strategy. Background to the development The new section of motorway would be approximately 23 kilometres in length and would provide three lanes in both directions between Junction 29 of the M4 at Castleton and Junction 23 of the M4 at Magor. After leaving the existing M4 motorway at Junction 29, the new section of motorway would pass to the south of Duffryn before crossing the Rivers Ebbw and Usk to the south of the A48 at Newport Docks. The new section Page 3

7 of motorway would then continue to the south of the Solutia chemical works and the Tata Steel site at Llanwern before passing to the west of Magor and re-joining the existing M4. B.1.11 B.1.12 B.1.13 B.1.14 B.1.15 B.1.16 Dormice have been recorded at either end of the Scheme (at Junctions 29 and 23) as well as within the land south of Tata Steel, in the central section. In addition to the junctions at Castleton and Magor, two new junctions would be provided along the route of the new section of motorway at Newport Docks and at Glan Llyn. New or diverted lengths of highway, public rights of way and private means of access would be provided to replace those affected by the Scheme. The local highway network would also be realigned at ten locations and new road overbridges would be constructed at Church Lane, Lighthouse Road, New Dairy Farm, Nash Road and North Row. Road drainage would discharge into a series of water treatment areas comprising attenuation ponds and reed beds along the new section of motorway. These water treatment areas would attenuate and treat the collected surface water prior to discharging it into existing watercourses. Approximately two-thirds of the route for the proposed new section of motorway crosses the Gwent Levels. The Gwent Levels are areas of flat reclaimed coastal marshes adjoining the Severn Estuary and comprising the Wentlooge Levels and Caldicot Levels to the west and east of Newport respectively. The Gwent Levels are low lying with an elevation typically of between 5 6 metres (m) above ordnance datum (AOD). B.1.17 The Gwent Levels are dissected by an extensive network of tide-locked freshwater drains, locally known as reens. A number of designations apply to the Gwent Levels including a number of Sites of Special Scientific Interest (SSSIs). In addition, the River Usk is designated nationally and internationally for its nature conservation value. At the location of the proposed crossing, the river is designated as a SSSI and Special Area of Conservation (SAC). B.2 Proposed Works on Site to be covered by the Licence B.2.1 This relates to any works for the M4caN Scheme that could have an impact on dormice or habitat of value to the species. These works include those listed below. Site clearance. Establishment of temporary works compounds. Establishment of temporary storage areas (equipment, vehicles, materials including soil etc.). Excavation of temporary borrow pits. Construction of the new motorway with associated embankments. Construction of temporary access roads (construction). Construction of access/slip roads (permanent). Construction of Water Treatment Areas. Construction of drainage systems. Construction and operational lighting. B.2.2 The construction and operational phases of the Scheme are expected to result in the following impacts with regard to dormice: Page 4

8 permanent loss of habitat of value to hazel dormice; temporary disturbance of hazel dormice; temporary small-scale disturbance of habitat of value to hazel dormice during construction; temporary and potential permanent disruption to hazel dormice movement due to construction and/or operation; temporary short-term disturbance of habitats by run-off of pollutants and dust deposition during construction; temporary and permanent displacement of hazel dormice due to construction and/or operation; and potential trapping and translocation/relocation of hazel dormice during construction to a pre-prepared receptor site. B.2.3 B.2.4 B.2.5 B.2.6 The Public Local Inquiry (PLI) for the Scheme is expected to run between February and July Should the Orders be confirmed following the PLI it is expected that construction would commence in July 2018, and that the new road would be operational by the autumn of Early works to commence in July 2018 would include site clearance, the construction of the haul road, and enabling works, including the construction of works compounds, temporary Water Treatment Areas and replacement watercourses. Advance mitigation works on Welsh Government-owned land could commence in May It is expected that the main construction activities would commence in late Following construction, there would be a five-year maintenance period under the contract until autumn Prior to the commencement of construction, pre-construction surveys would be undertaken in 2017 and 2018 (as described in Section C.6 below). B.3 Actions requiring licensing B.3.1 The dormouse is listed in Schedule 2 of the Conservation of Habitats and Species Regulations Its resting/breeding sites are protected in Wales under Regulation 41. Anyone who: deliberately captures, injures or kills any wild animal of a European protected species; deliberately disturbs wild animals of any such species; deliberately takes or destroys the eggs of such an animal; or damages or destroys a breeding site or resting place of such an animal; is guilty of an offence. B.3.2 B.3.3 B.3.4 Under Regulation 53, it is possible to obtain a derogation (or licence) to carry out activities that would otherwise be considered likely to result in an offence being committed. The construction of the M4CaN Scheme would require the clearance of vegetation of value to dormice within the Scheme boundary and, therefore, could result in injuries or fatalities of dormice during construction. In order to help prevent injuries or fatalities, dormice would be displaced or translocated from working areas prior to the commencement of habitat clearance and construction in an area. Page 5

9 B.3.5 Therefore, a licence under the Conversation of Habitats and Species Regulations 2010 would be required for the following measures: capture of dormice and translocation from working areas to NRW-approved receptor sites (temporary or permanent) prior to construction in an area; displacement or local relocation of dormice from working areas to adjacent favourable habitat prior to construction in an area; and damage to or destruction of habitat of value to dormice, including breeding and resting sites. B.3.6 B.3.7 This Mitigation Strategy sets out how these licensable activities would be achieved whilst at the same time ensuring that there would be no detriment to the maintenance of the favourable conservation status (FCS) of the local dormouse population in its natural range. This is the test that must be satisfied to enable NRW to grant a licence. The strategy would then be updated and submitted as a Hazel Dormouse Method Statement as part of the licence application. For this reason, the Strategy, which now includes additional detail to allow NRW to reconsider their objection to the Scheme on the grounds of dormouse conservation, is structured in the same way as required by NRW of an EPS licence method statement. However, in order specifically to address the above FCS question, an additional section has been added at the end of this document to assess the impacts with the mitigation measures in place and to thus summarise how FCS would be maintained (Section I Assessment against the FCS test). Page 6

10 C SURVEY AND SITE ASSESSMENT C.1 Existing information on hazel dormice C.1.1 Historic dormouse records are shown on Figure 1. C.1.2 C.1.3 Evidence of previous dormouse presence within the vicinity of the proposed M4CaN scheme was identified from records held by the local records centre (SEWBReC) and from results of previous studies in the area (including surveys undertaken by Jacobs for the M4 Widening Project between Junctions 29 and 32 in 2005/2006 and by Arup for earlier stages of the M4CaN Scheme in 2007/8). To the west of the Scheme, the majority of dormouse activity was found to be concentrated between the central and eastern areas of the M4 Widening Project and mostly to the north of the existing M4 corridor. The subsequent surveys in 2007/2008 revealed that dormice were also present further east, in hedgerows around Penylan Farm and Gwaunshonbrown Farm in Castleton. C.1.4 Dormouse monitoring surveys were also subsequently undertaken by Jacobs in 2014 and Atkins in 2015 as part of the M4 Widening project. Survey areas included parts of New Park Farm. Results of the surveys reported the presence of dormice along the M4 road verge at New Park Farm - four dormice were recorded in a single box in October (one adult female and three juveniles (one female and 2 male)). No other evidence of dormouse was reported. C.1.5 In addition, in 2015, NRW reported to RPS historic records of dormice in a number of woodlands in the surrounding area, namely: Cefn Mably woods, Ruperra Castle and woodlands, Park Wood, Coed Cefn Pwll-du, Plas Machen Wood, Penhow Woods and Wentwood. All of these locations are also shown on Figure 1. C.1.6 No historic records of dormice on the Gwent Levels were identified through the desk study process. C.2 Statutory sites notified for the species within 10 km C.2.1 There are no statutory designated sites that have been designated for the presence of dormice within 10 km of the Scheme. C.3 Objectives of surveys 2014 survey (Appendix 10.9 to the M4CaN March 2016 ES) C.3.1 The objectives of the 2014 field surveys were to: identify the presence of dormice within the study area; identify areas where mitigation may be required; and identify further studies that may be required to ensure that dormice are fully considered within the development of the mitigation design for the Scheme. Page 7

11 2015 survey (Appendix to the M4CaN March 2016 ES) C.3.2 C.3.3 C.3.4 C.3.5 C.3.6 C.3.7 C.3.8 C.3.9 C.3.10 C.3.11 The objectives of the 2015 field surveys were to continue monitoring nest tubes installed in 2014, both in areas where dormice had previously been recorded and where sufficient survey effort had not been achieved in In addition, further survey effort was focused at the far eastern end of the new section of motorway near Magor/Knollbury, near the proposed Ifton Quarry haul road and other areas not previously surveyed during The survey in the area of the Castleton Junction was also extended into connected hedgerows in order to further investigate the distribution of dormouse in that area survey (December 2016 ES Supplement) The objective of the 2016 field surveys was to continue monitoring nest tubes to the east of the River Usk in order to collate additional data to inform a population size estimate. In addition, surveys were also undertaken on land along the M4 embankment and between the M4 and M48 at the eastern end of the Scheme, where no access had been available to survey in off-site survey Due to the potential need for an off-site receptor site for dormice translocated from the Scheme, consideration was given in to the identification of suitable woodlands in the surrounding area. Favourable receptor sites would require long-term management security and, therefore, woodlands owned and/or managed by Welsh Government or NRW were specifically considered. Results of consultation with NRW confirmed that most woodlands in the nearby surrounding area already supported dormice; however, no records of dormice were reported for Coed Mawr (Figure 1). Coed Mawr is a woodland over 90 hectares in size, which is owned by Welsh Government and managed by NRW. Chanin (2014) suggests dormouse receptor sites should form part of a cluster of sites, that are: 4-6 woodlands each at least 20 hectares and totalling >150 hectares (may be <20 hectares if well connected to other woods); located within a radius of no more than 5 km; and connected by parcels of woodland, scrub and/or hedgerows, excluding barriers (e.g. roads >12 m wide; or open rivers unless with connecting canopies or <10 m wide with bridges). Figure 1b illustrates the suitability of Coed Mawr with regard to the above criteria. In addition to the M4 corridor, dormice have been recorded in a considerable number of large parcels of woodland within 5 km of Coed Mawr, and there is a good network of interconnecting hedgerows and linear strips of woodland across the area. The A468 to the south/south-west of Coed Mawr (Figure 1b) is on average approximately 10 metres in width, increasing to approximately 15 metres in width at junctions, and is crossed by bridges and, therefore, should not present a significant barrier to the movement of dormice. Bright et al. (2006) and Chanin (2014) suggest suitable receptor sites should contain: a diverse, unshaded and productive understory, preferably dominated by hazel; Page 8

12 a high density and diversity of plant species for food throughout the year; and a commitment to appropriate long-term site management. C.3.12 C.3.13 C.3.14 Therefore, surveys were undertaken of Coed Mawr with the objective of determining whether or not dormice are present, and to assess the favourability of habitats for dormice survey Nest tube surveys are being undertaken in 2017 in order to further inform the Hazel Dormouse Method Statement, should the Orders for the Scheme be granted. Surveys to be completed in 2017 would include a continuation of previous nest tube surveys in order to update previous survey results. Nest boxes will also be installed in survey areas to increase survey effort. Additional nest tube and nest box surveys would also be undertaken in order to: take into account the development of habitat within the footprint of the Scheme that previously had been recorded to be unfavourable to dormice but since has developed to a more favourable condition; and identify where dormice are present in areas immediately surrounding the Scheme. C.3.15 The aims of the additional 2017 surveys are to: confirm whether or not dormice have moved into new areas that would be affected by the Scheme; and determine presence or likely absence in surrounding areas that would not be affected by the Scheme but could be receptor sites for displaced/relocated dormice. C.3.16 In addition, a more detailed assessment of the suitability for dormice of habitat areas to be lost is also being undertaken. The objective of this work is to allocate grades of dormouse-suitability to these areas in order to better estimate the number of animals that might be affected by the works (and thus need to be either displaced, relocated or translocated). These habitat grades are as follows: Grade 1 optimal habitat Grade 2 potentially optimal habitat Grade 3 sub-optimal habitat Grade 4 very sub-optimal habitat C.3.17 C.3.18 C.3.19 C.3.20 A description of each habitat grade is provided in Table 4, Section C.6 (2017 Survey Methods) below. For each of these grades, a likely density of dormice will be provided, and agreed with NRW. This will allow the predicted number of dormice to be calculated using the area of the habitat and the likely density of dormice within it off-site survey Further habitat and nest tube surveys are being undertaken in 2017 at Coed Mawr. Nest boxes will also be used to increase survey effort. The objective of the nest tube/box survey is to extend the area covered and to continue monitoring those tubes installed previously, in order to determine where dormice are present or likely absent in the wood and, where present, to assess population size. The objective of the habitat survey is to map in detail the various habitats at Coed Mawr and to allocate grades of suitability for dormice to these areas. This will use the same grading Page 9

13 system applied to the on-site habitats (above), so that a similar assessment of habitat suitability can be made. C.3.21 C.3.22 Survey results would inform an assessment of the amount of suitable receptor habitat that already exists and thus the number of dormice from the Scheme that the woodland might be able to accommodate. Results would inform the Hazel Dormouse Method Statement for a translocation of dormice from the Scheme to Coed Mawr pre-construction survey The nest tube and nest box surveys undertaken in 2017 would be continued in 2018 in order to update the baseline and further inform the Dormouse Method Statement and measures to be commenced in 2018, should the Orders for the Scheme be granted. C.4 Scaled plan/map of survey area C.4.1 Figure 2 shows the locations of the survey areas, both on-site and at Coed Mawr. C.5 Site/habitat description C.5.1 C.5.2 The following section sets out the detailed locations of the surveys on a year-by-year basis. In the previous version of this Strategy, survey locations were illustrated on Figure 2 with different survey years denoted by different colours. However, in order to make this version of Figure 2 easier to read, survey years for each survey area are now provided in the schedule at Annex 1, which thus shows the evolution of the survey baseline. There have been some minor discrepancies between the names given to survey areas over the years; therefore, these have been combined in the following paragraphs and on the revised Figure 2 so as to prevent confusion if referring back to original survey reports survey areas 2014 nest tube survey C.5.3 During 2014, dormouse nest tubes were installed in five areas, and these were monitored during four to seven survey visits in order to record dormouse presence. Survey areas were selected taking into account habitat preferences of hazel dormice and records of where dormice had previously been encountered. Survey sites were grouped into the following broad areas: New Park Farm area - (North and South), Figure 2a Castleton area (North and South), Figure 2a Berryhill Farm, Figure 2a Gwent Levels Pye Corner and Tata Steelworks, Figure 2c/d Magor/Knollbury area, Figure 2e/f C.5.4 Descriptions of the 2014 survey locations within these 5 broad survey areas are provided in Table 1 below. Table 1: Description of the 2014 Survey Sites Survey site Description New Park Farm survey area New Park Farm (North and South), Areas C and D, Figure 2a M4 and A48(M) verges (excluding areas of motorway planting between the M4 and A48(M)). Consisted of broadleaved plantation with hazel and dense bramble scrub. Page 10

14 Survey site Description Castleton survey area Castleton (North and South), Area K, Figure 2a Surveys of Penylan Farm, Pound Hill and Spring Court Woodland to the north of Lower House. Penylan Farm - survey area consisted of broadleaved plantation with bramble scrub, connected to well established species-poor hedgerows. Pound Hill survey area - comprised broadleaved plantation with sparse bramble understory, connected to well established species-poor hedgerows. Spring Court Woodland to the south of the survey area, north of Lower House - semi-natural broadleaved woodland with bramble scrub understorey and connectivity to species-poor hedgerows with trees. Berryhill Farm survey area Berryhill Farm, Area O, Figure 2a Church Lane Woodland, part of Area O, Figure 2a Gwent Levels survey area Gwent Levels - Pye Corner, Area M, Figure 2c/d Gwent Levels areas Tata Steelworks, part of Area N, Figure 2c/d Magor / Knollbury survey area Areas of ancient broadleaved woodland, broadleaved and mixed plantation, and hedgerows, including species-rich hedgerows with trees. Plantation broadleaved woodland with areas of bramble and a speciespoor hedgerow located along the northern verge of Church Lane and to the south of Church Lane. A derelict laboratory site which had become overgrown with seminatural broadleaved woodland, planted coniferous trees and bramble scrub. A species-poor native intact hedgerow with wet ditch along the southern boundary of the Tata Steelworks site. Magor Vicarage, part of Area P, Figure 2e Upper Grange Farm hedgerows, part of Area P, Figure 2e/f. Junction 23a Services, part of Area P, Figure 2e/f Magor/Knollbury, Area I, Figure 2e/f An area of semi-natural broadleaved woodland with an understorey of bramble at Magor Vicarage to the north of the Newport Road into Magor, in the southern part of Area P. Hedgerows and road verges to the south of Upper Grange Farm, to the east of Junction 23a. Hedgerows were species-poor. Verges comprised broadleaved mixed plantation. Areas of broadleaved plantation with coppiced hazel to the south of the M4 at Junction 23a. M4 road embankments comprising broadleaved plantation and areas of dense bramble scrub hazel nut search C.5.5 In areas of habitat of potential value to dormice, where fruiting hazel was present, searches were undertaken for hazel nuts that had been characteristically opened by dormouse. Areas surveyed were: New Park Farm North (Area i, Figure 2a); A48(M) verges (Area iii, Figure 2a); Spring Court Woodland (Area iii, Figure 2a); and M4 verge to the south of J23a services (Areas vii-ix, Figure 2e) Page 11

15 2015 survey areas 2015 nest tube survey C.5.6 C.5.7 C.5.8 C.5.9 Monitoring of nest tubes installed in 2014 in the following locations was continued in 2015: New Park Farm (North and South) survey areas (Areas C and D, Figure 2a); and Castleton (North and South) survey areas (Area K, excluding Spring Court Woodland which was relatively isolated from the known dormouse populations and no dormice had been recorded in 2014, Figure 2a). In addition, due to the known presence of dormice in the area, in 2015 nest tubes were installed in Areas G and J, Figure 2a, so as to further investigate the distribution of dormouse in the area. The surveys which had been partially undertaken in the Gwent Levels in 2014 (Areas M Figure 2c, and Area N Figure 2d) were completed in Due to access and health and safety issues at Pye Corner (Area M, Figure 2c), some additional nest tubes were installed in the area to enable a continued survey. In addition, due to the identification of a dormouse nest in a nest tube alongside the southern boundary of the Tata Steelworks survey area (Area N, Figure 2d), additional nest tubes were installed in areas of scrub and hedgerows in the survey area. No other areas in the Gwent Levels were surveyed due to the conclusions of Arup s previous consultation with NRW and results of the extended Phase 1 habitat surveys undertaken in 2014 (ES Appendix 10.2) and 2015 (ES Appendix 10.19), which reported the area to be unsuitable for dormice due to the Levels being relatively wet, characterised by a network of drainage ditches and historic floodplain grassland, with no significant parcels of woodland or dense scrub. C.5.10 However, new survey locations were identified outside the Gwent Levels in These primarily focused on suitable habitat that had not been previously surveyed due to access restrictions. These new survey locations were located in: the Magor/Knollbury area (Area I - additional M4 embankment habitat, and Area L - hedgerows and woodland adjacent to the Scheme), Figure 2f); small parcels of woodland/scrub and hedgerow along Minnet s Lane (Area H, Figure 2f); and woodland adjacent to the proposed Ifton Quarry haul road (Areas E and F, Figure 2f). C.5.11 Descriptions of the habitats recorded in the additional 2015 survey areas are provided in Table 2 below. Table 2: Description of Additional 2015 Survey Sites Survey site New Park Ffarm (North and South), Areas G and J, Figure 2a Castleton (North and South), Areas Description Plantation including n woodland. Species include hazel, dogwood, hawthorn, sycamore, oak and ash. Species-rich hedgerow along Pen-y-land Farm access track (including bramble, hawthorn, elder, holly, ash, hazel and goat willow. Semi-natural broadleaved woodland and connecting hedgerow beside Pwll Diwaelod, between the M4 and A48(M). Woodland species included hazel, holly, hawthorn and willow. Hedgerow species included elder, hawthorn, bramble, hazel and alder. Broadleaved semi-natural woodland dissected by the A48 (M) and M4. Species included hazel, hawthorn, alder, sycamore, hazel, oak, ash and goat willow. Species rich hedgerows on Gwaunshonbrown Farm. Species included hazel, hawthorn, ivy, holly, field maple, willow and bramble. Page 12

16 B and K, Figure 2a Castleton - Cefn Llogelll, Area A, Figure 2a Gwent Levels - Pye Corner, Area M, Figure 2c Gwent Levels - Tata Steelworks, Area N, Figure 2d Magor/Knollbury, Areas I and L, Figure 2f Minnetts Lane, Area H, Figure 2f Haul road to Ifton Quarry, Areas E and F, Figure 2f Plantation along Poundhill Road. Specie included dogwood, hawthorn, field maple and oak. Species-rich hedgerows lining a small road, plantation along the A48(M) and dense scrub lining a bridleway. Species included holly, bramble, elder, hawthorn, dogwood, hazel, oak, goat willow, field maple, ash, rose, alder and buddleia. Scattered scrub and hedgerow. Species included willow, bramble, hawthorn and ash. Species-poor hedgerow and scattered scrub. Species included bramble, hawthorn, willow, sycamore, blackthorn and buddleia. Species-poor hedgerows (Area I). Species included holly, hazel, hawthorn, rose and elder. Two areas of semi-natural broadleaved woodland and species-rich hedgerows (Area L). Species included hazel, hawthorn, sycamore, holly, privet, elder, field maple, blackthorn and elm. Species-poor hedgerow and two small areas of semi-natural broadleaved woodland along Minnet s Lane. Species included elder, blackthorn, field maple, hazel, hawthorn and holly. Rogiet Brake Wood: semi-natural broadleaved woodland and pine plantation. Species included field maple, hazel, elm, ivy, sycamore and holly. Quarry Wood: semi-natural broadleaved woodland. Species included hawthorn, field maple, rose, elm, hazel, bramble, oak, ivy and ash hazel nut search C.5.12 C.5.13 Hazel nut searches were undertaken at several locations along the proposed route between 2 November and 20 November 2015 as shown on Figure 2 (Areas i-v, Figure 2a; areas vi and vii, Figure 2e; and ix-xviii, Figure 2f). In addition, a hazel nut search of Coed Mawr (Figure 2g) was undertaken using the same methodology as described above for the on-site nut searches survey areas 2016 nest tube survey C.5.14 C.5.15 C.5.16 C.5.17 In 2016, the survey of Tata Steelworks land was continued in order to gain a better understanding of dormouse activity in the area. Nest tubes were also installed in a small parcel of woodland to the north of the M4 at the eastern end of the scheme (Woodview woodland along the northern boundary of Area L, Figure 2f). The woodland contained abundant ash with understory of occasional hazel, field maple and hawthorn. Semi-mature silver birch was present throughout and had substantial bark damage inflicted by grey squirrels. Approximately ten old apple trees were present in the north of the woodland. In the east of the wood areas of dense bramble had been cleared and a mix of native and non-native tree species had been planted, including young common oak, turkey oak, hornbeam and pine. This area had not been surveyed in 2015 due to access issues. In addition, areas of young to semi-mature broadleaved plantation with areas of dense bramble scrub and occasional hazel located along the embankments of the M4 and M48 at the eastern end of the scheme (Area Q, Figure 2f) were surveyed. These areas had not been surveyed in 2015 due to access issues. Nest tubes were also installed in Coed Mawr (Figure 2g) in order to inform the assessment of its potential value as a receptor site. Page 13

17 2017 survey areas 2017 nest tube survey C.5.18 Taking into account results of the dormouse surveys in , nest tube surveys are to be continued at the following locations in 2017: Areas C, D (road embankments only), G, J, K (excluding Spring Court Woodland), A and O, Figure 2a; Area M, Figure 2c; Area N (nest tubes surveyed in ), Figure 2d; Areas I, L, Q, H, E and F, Figure 2f; and Coed Mawr (Figure 2g) (additional nest tubes have been installed in order to increase survey effort). C.5.19 In addition, taking into account the potential for dormouse populations to move into adjacent suitable habitat, nest tubes have been installed in the following locations for monitoring between May and November Area O, Figure 2a: in addition to replacing any lost or damaged nest tubes from the 2014 survey locations, additional nest tubes have been installed in suitable scrub, woodland and hedgerows across the farm. Area R, Figure 2b: Fox Covert survey area comprises areas of scrub, woodland and hedgerows. Although the woodland and scrub are limited in extent and isolated from other substantial areas of scrub or woodland, as a precautionary measure, nest tubes have been installed in suitable habitat. Area S, Figure 2c: areas of dense scrub and woodland are located throughout the area and although no dormice have been recorded at Pye Corner (Area M, Figure 2c as detailed in the survey results section below) and the area is a considerable distance from the Tata Steelworks population (Area N, Figure 2d), as a precautionary measure, nest tubes have been installed in suitable habitat in the area. Area N, Figure 2d: additional nest tubes have been installed in the Tata Steelworks survey area in order to develop a better understanding of the size of the population. Area T, Figure 2d/e: in order to determine where dormice on the Tata Steelworks survey area (Area N, Figure 2d) are dispersing from, nest tubes have been installed in a wide strip of woodland and scrub to the north of the A4810, which is of potential value to dormice and has good connectivity to railway embankment scrub and parcels of woodland further to the north that are also of potential value to dormice. Area U, Figure 2e: as a precautionary measure, nest tubes have been installed in suitable hedgerows to the east of the Tata Steelworks survey area (Area N, Figure 2d) in order to determine whether or not dormice have moved into adjacent habitat from the Tata Steelworks site. Area Y, Figure 2e/f: in response to changes to the proposed provision of an east bound off slip-road at junction 23a of the existing M4, which would result in loss of some woodland/scrub along the western verge of the A4810, as a precautionary measure, nest tubes have been installed in the area. Area W, Figure 2e/f: taking into account the development of embankment scrub since 2014, as precautionary measure, nest tubes have been installed in suitable Page 14

18 areas of scrub to the south of the M4 and east of Junction 23a. Area Q, Figure 2f: taking into account results of the 2016 surveys in the surrounding area, nest tubes have been installed in embankment scrub along the M4 and suitable adjacent hedgerows to the south of Area Q. Areas X and Z, Figure 2f: taking into account the development of embankment scrub since 2014, additional nest tubes have been installed in Areas X and Z. C.6 Field survey methods C survey methods The 2014 dormouse survey comprised a combination of a nest tube survey and a hazel nut search as recommended in the Design Manual for Roads and Bridges (DMRB) (Highways Agency, 2001) and in accordance with guidelines set out in the Dormouse Conservation Handbook (Bright et al., 2006). C.6.2 A schedule of nest tube surveys for each survey area is provided at Annex nest tube survey C.6.2 Survey locations were: New Park Farm area (North and South), Figure 2a Castleton area (North and South), Figure 2a Berryhill Farm, Figure 2a Gwent Levels Pye Corner and Tata Steelworks, Figure 2c/d Magor/Knollbury area, Figure 2e/f C.6.3 C.6.4 C.6.5 C.6.6 A total of 310 dormouse nest tubes were installed in areas of woodland and suitable hedgerows in April and May 2014, with the exception of nest tubes at the Magor/Knollbury area (Area I, Figure 2e/f) and at Pye Corner (Area M, Figure 2c) where they were installed in July The numbers and distribution of tubes installed in each survey site depended on the site-specific conditions; however, in woodland or areas of scrub, nest tubes were generally installed in a grid pattern of 20 m squares and along hedgerows at approximately m intervals. Nest tubes were inspected on a monthly basis until November 2014 in order to locate any signs that could confirm the presence of dormice, including dormice nests. Each survey visit was carried out by a licensed surveyor accompanied by an accredited agent or assistant. All evidence of dormice and their activity was recorded and mapped. Where individual dormice were found, the sex of animals was determined and their weights measured where possible. Evidence of other small mammal activity was also recorded where this occurred. During the November check where torpid dormice were found, the sex of the animal was not determined so as to avoid undue disturbance survey effort C.6.7 In accordance with the Dormouse Conservation Handbook (Bright et al., 2006) the scores for survey effort were as presented below. New Park Farm - North and South (Areas C and D, Figure 2a): 42 tubes surveyed for three months and 36 tubes surveyed for seven months giving a score of 24. Castleton - North and South (Area K, Figure 2a): 59 tubes surveyed for seven months and 14 tubes surveyed for four months giving a score of 33. Page 15

19 Berryhill Farm (Area O, Figure 2a): 106 tubes surveyed for seven months giving a score of 51. Gwent Levels: 19 tubes (Pye Corner, Area M, Figure 2c) plus 10 tubes (Tata Steelworks, Area N, Figure 2d) surveyed between August and November giving a total score of 9 (6 for Pye Corner and 3.2 for Tata Steelworks). Magor/Knollbury area (Area I, Figure 2e/f): 67 tubes surveyed for seven months giving a score of 32. C.6.8 Bright et al. (2006) consider 20 to be the minimum acceptable score to confirm sufficient survey effort to determine presence/likely absence of dormice. Therefore, it is considered that the survey effort was sufficient for the following survey locations: New Park Farm - North and South (Areas C and D, Figure 2a); Castleton - North and South (Area K, Figure 2a); and Magor/Knollbury (Area I, Figure 2e/f). C.6.9 However, the survey effort score indicated the survey effort insufficient at: Gwent Levels Pye Corner (Area M, Figure 2c); and Gwent Levels Tata Steelworks (Area N, Figure 2d) hazel nut search C.6.9 Hazel nut searches were carried out at each survey location within five 10 m x 10 m quadrats for a period of twenty minutes each, or until 100 nuts (not opened by birds or squirrels) had been collected survey limitations C.6.10 C.6.11 C.6.12 C.6.13 Due to the delayed start to the survey of the Gwent Levels Pye Corner and Tata Steelworks sites, insufficient survey effort was undertaken in order to confirm presence/likely absence. No access was permitted to Gwaunshonbrown Farm (Area B, Figure 2a) for the entire survey season and so no survey of this area was undertaken. However, surrounding areas were surveyed and, therefore, it is considered that this limitation did not adversely affect the conclusions of the survey. At the site to the south of Penylan Farm (Area K to the north of the M4, Figure 2a), a number of the nest tubes were damaged by cattle. These were replaced once but further damage occurred. Therefore, to minimise the risk to dormice the decision was made to remove the affected tubes from the survey. This is not thought to have adversely affected the conclusions of the survey due to the number of tubes placed within the wider connected habitat and the historic presence of dormice within the area. As the Gwent Levels survey areas (Areas M, Figure 2c; and Area N, Figure 2d) were only surveyed for the latter part of the survey period and only achieved a score of 9, results from this area are insufficient to consider a complete survey. Therefore, surveys were continued in 2015 to ensure sufficient survey effort survey methods C.6.14 As in 2014, the 2015 surveys comprised a hazel nut search and nest tube survey. The surveys followed the same methodologies undertaken in 2014 and set out above and in the Dormouse Conservation Handbook (Bright, et al. 2006). Page 16

20 2015 nest tube survey C.6.15 A schedule of nest tube surveys for each survey area is provided at Annex 1. C.6.16 Nest tubes were inspected by appropriately experienced ecologists in accordance with best practice guidelines. Any evidence of dormouse activity identified in the nest tubes was recorded, including the animals themselves, nests or gnawed nuts survey effort C.6.17 In accordance with Bright et al. (2006) the score for survey effort at each survey area was as set out below. The survey effort for all survey locations was sufficient to enable a conclusion to be drawn regarding dormouse presence/likely absence in accordance with the guidance, as described below. Minnet s Lane/Haul Road to Ifton Quarry, Areas H, E and F, Figure 2f C.6.18 The nest tube locations are considered as one survey area due to the proximity of each of the groups of nest tubes and the presence of suitable connecting habitat between them. The total number of nest tubes was 188, surveyed from July to November. Survey effort = Magor/Knollbury, Areas I and L, Figure 2f C.6.19 The nest tube locations are considered as one survey area due to their proximity to each other and the presence of suitable connecting habitat. The total number of nest tubes was 135, surveyed from July to November. Survey effort = Gwent Levels Pye Corner, Area M, Figure 2c C.6.20 A total of 9 additional nest tubes were installed due to the loss of the 2014 nest tubes. These were surveyed from August to November. Survey effort = Gwent Levels - Tata Steelworks land, Area N, Figure 2d C.6.21 In addition to the 10 remaining 2014 nest tubes, an additional 80 nest tubes were installed on the Tata Steelworks land. Tubes were surveyed from August to November. Survey effort = Castleton - North and South, Areas B and K (excluding Spring Court Woodland); and Castleton - Cefn Llogelll, Area A, Figure 2a C.6.22 C.6.23 C.6.24 The area can be divided into two separate areas, due to the distance between each survey area and the limited availability of suitable connecting habitat. The total number of nest tubes in one survey area was 83, surveyed from June to November. Survey effort = There were 42 nest tubes within the second survey area. 28 were surveyed from June to November (Survey effort = 11.2), whilst 14 were surveyed from July to November (Survey effort = 5.04). In addition, the total number of retained 2014 nest tubes was 34 which were surveyed between May and November (Survey effort = 16.32). Therefore, the total survey effort for the area = New Park Farm - North, Areas C and J, Figure 2a C.6.25 The nest tube locations are considered as one survey area due to their proximity to each other and the presence of suitable connecting habitat. The total number of nest tubes was 65, surveyed from July to November. Survey effort = Page 17

21 New Park Farm - South, Areas D and G, Figure 2a C.6.26 C.6.27 The nest tube locations are considered as one survey area due to their proximity to each other and the presence of suitable connecting habitat. The total number of nest tubes was were surveyed from June to November (Survey effort = 11.6), whilst the remaining 66 were surveyed from July to November (Survey effort = 23.76). Therefore, the total survey effort for the area = In addition, monitoring of the Arup 2014 nest tubes was continued in the area, where accessible hazel nut search C.6.28 C.6.29 Hazel nut searches were undertaken at several locations along the proposed route between the 2 November and 20 November At each location, due to the fact survey areas did not contain a large number of heavily fruiting hazel, the following method of survey was undertaken: 100 hazel nuts that had been opened by small rodents (avoiding caches and nuts opened by squirrels) were collected. In areas where there was a very small number of hazel plants present, all hazel nuts located were collected. Collected hazel nuts were then inspected off-site, using a magnifying glass/microscope as required in order to determine whether or not they had been opened by dormice. The hazel nut search of Coed Mawr (Figure 2) was undertaken using the same methodology as described above. Limitations to the 2015 survey C.6.30 Access for nest tube installation and checking of tubes previously installed by Arup was available from mid-may C.6.31 Access arrangements and subsequent inductions onto Tata Steelworks land (Area N, Figure 2d) were not completed until August 2015 and, therefore, following the discovery of a dormouse nest in a nest tube on the boundary of the Tata Steelworks site, additional nest tubes were not installed on the Tata Steelworks site until September C.6.32 C.6.33 C.6.34 In survey areas I and L, Figure 2f, a number of the nest tubes were damaged by cattle during the survey. The tubes were not replaced to minimise the risk to dormice, if present. However, this is not thought to have adversely affected the conclusions of the survey due to the number of tubes placed within the wider connected habitat. A few other nest tubes elsewhere on site were damaged during the survey period, by horses or other factors, but again it is thought that this did not adversely affect the conclusions drawn from the survey due to the number of unaffected tubes. In some of the proposed nut search areas there was little or no hazel so minimal or no hazel nuts were collected survey methods 2016 nest tube survey C.6.35 C.6.36 Nest tube surveys were undertaken following the same methodology as previous years, i.e. in accordance with the methodology set out in the Dormouse Conservation Handbook (Bright, et al., 2006). A schedule of nest tube surveys for each survey area is provided at Annex 1. Table 3 below shows the dates and number of dormouse nest tubes installed and the dates when each site was surveyed. Page 18

22 Table 3 Dates of Nest Tube Installation and Survey Visits Site DATE April 2016 May 2016 June 2016 July 2016 Aug 2016 Sept 2016 Oct 2016 Nov 2016 Coed Mawr woodland (Figure 2g) 500 tubes Installed 1 st - 2 nd 1 st 30 th 1 st 31 st 1 st Gwent Levels - Tata Steelworks (Area N, Figure 2d) 50 tubes reinstalled 16 th 30 th 31 st Magor/Knollbury - Woodview woodland (north of Area L, Figure 2e/f) M4/M48 motorway islands (Area Q, Figure 2f) 50 tubes installed 30 th 29 th 31 st tubes installed (50 north, 25 south of M4) 7 th 1 st 2016 survey effort C.6.37 As described above, in accordance with Bright et al., 2006, the survey effort in 2016 was as described below. Coed Mawr (Figure 2g) C nest tubes surveyed from May to October. Survey effort = 180. Tata Steelworks land (Area N, Figure 2d) C nest tubes surveyed from May to October. Survey effort = 22. Woodview woodland north of Magor/Knollbury survey area (Area L, Figure 2e/f) C nest tubes surveyed from May to October. Survey effort = 22. M4 / M48 motorway islands (Area Q, Figure 2f) C.6.41 C.6.42 Although dormice have only been recorded on the southern island, the two embankments are considered together as, due to the limited extent of the islands, any dormouse located here is likely to have originated from a population located on the adjacent motorway embankments and individuals on one island could move to the adjacent island in an attempt to locate more resources. 50 nest tubes between the M4/M48, 25 nest tubes on the M4 embankment, surveyed from July to October. Survey effort = = 24. Coed Mawr habitat mapping C.6.43 Initial habitat mapping and assessment of Coed Mawr (Figure 2g) was also undertaken in 2016 during the nest tube survey period in order to identify broad habitat types and to assess the potential suitability of the site for dormice. Limitations to the 2016 survey C.6.44 At the end of October 2016, nest tubes along North Row, to the south east of Tata Steelworks land (area N, Figure 2d) were found to have been damaged during hedgerow management. A few other nest tubes elsewhere in the Tata Steelworks survey area were Page 19

23 also damaged by horses or other factors. However, the majority of the nest tubes in the survey area remained intact; therefore, the loss of this small number of nest tubes was not considered to significantly affect the overall survey. The survey was additional to the survey in 2014/2015, when in accordance with English Nature 2003, sufficient survey effort was undertaken. C.6.45 C.6.46 Access to the M4 embankment at the eastern end of the Scheme (southern part of Area Q, Figure 2f), where 25 dormouse tubes were installed, was limited due to the density of the vegetation (scrub and trees) and health and safety reasons with regard to the proximity to the M4 carriageway in places. The survey effort achieved did not meet the target score of 20; however, if combined with the survey area between the M4/M48 Island (northern part of Area Q, Figure 2f), the survey effort was greater than 20. (The area will continue to be surveyed in 2017.) The initial habitat mapping of Coed Mawr (Figure 2g) was undertaken during other site visits and comprised a survey of broad habitat types only. A detailed habitat survey and assessment of the existing and potential value of the site is being undertaken in survey methods 2017 nest tube survey C.6.47 A schedule of nest tube surveys for each survey area is provided at Annex 1. C.6.48 C.6.49 C.6.50 Monitoring of dormouse nest tubes will continue until November 2017 in order to develop a greater understanding of the size and extent of the existing dormouse population and inform the NRW licence application. During the 2017 survey period, in all survey areas located within the Scheme boundary, nest boxes will also be installed so as to increase the survey effort and provide as much time as possible for dormice to locate the nest boxes that would later be used in any preconstruction trapping and translocation surveys (as described under Translocation of dormice, below). Where possible, the aim would be to install nest boxes at a density of 30 boxes per hectare of suitable habitat. All dormouse nest boxes to be used would be of standard specification as described in section of the Dormouse Conservation Handbook (Bright et al., 2006) habitat mapping C.6.51 C.6.52 Habitats in each survey area where dormice were recorded have been graded in accordance with the definitions in Table 4 below. This assessment has initially been undertaken on the basis of aerial photographs and previous habitat surveys, and will be ground-truthed and refined following the 2017 survey work, when a targeted assessment of each of the areas will be made in parallel with the nest tube surveys. Any amendments will be included in a revised set of figures presented in the Dormouse Method Statement presented with the licence application. Table 4 Habitat Grading Habitat Grade Grade 1 optimal habitat Description of Habitat Species-rich scrubby woodland habitat with a good three-dimensional structure and connectivity within the scrub canopy, and significant amounts of favoured dormouse food plants such as hazel, honeysuckle and bramble. Already suitable for displaced/relocated dormice and for softrelease of translocated dormice. Likely adult dormouse density/carrying capacity of up to 10 per hectare. Page 20

24 Habitat Grade Description of Habitat Grade 2 potentially optimal habitat Grade 3 suboptimal habitat Grade 4 very sub-optimal habitat Area of largely suitable dormouse habitat, with a mix of broadleaved woody species, a good existing understory (though with less connectivity within the scrub canopy) and with the potential to become optimal habitat with relatively limited habitat management intervention in the short term. Potentially suitable for displaced/relocated dormice or for soft-release of translocated dormice in the short term. Likely adult dormouse density/carrying capacity of up to 5 per hectare (increasing to 10 with enhancement). Area contains some scrub/broadleaf habitat but only small amounts of the favoured food plants such as hazel, honeysuckle and bramble, and will therefore require extensive clearance, re-planting and management to create optimal habitat. Unlikely to provide potential for release of translocated dormice except in the long term. Likely adult dormouse density/carrying capacity of no more than 2 per hectare. Whilst dormice would be able to move across this area (e.g. grassland, recent clear-fell, dense conifers, etc.) they would not be expected to derive any other use from it, either as foraging or nesting habitat. Would not provide potential for release of translocated dormice except in the very long term and with extensive habitat creation. Likely dormouse density/carrying capacity of no more than 1 per hectare Coed Mawr habitat mapping C.6.53 C.6.54 Detailed habitat mapping and assessment of Coed Mawr is being undertaken in 2017 in order to assess the existing value of the woodland as a receptor site for dormice. The aim is to determine the carrying capacity of the wood, and to identify habitat enhancement and management requirements for the benefit of dormice. Results of the survey will inform the detailed habitat management plan for the wood that would form part of the NRW licence application to be agreed with NRW. C.7 Survey results 2014 survey results C.7.1 C.7.2 C.7.3 Dormice were confirmed to be present at the western end of the survey area, within the New Park Farm (North and South) areas (Areas C, D and J, Figure 3a) and the Castleton (north and south) survey areas (Areas G and K, Figure 3a). The survey results provided evidence that dormice are present to the north and south of the M4 and also to the north and south of the A48(M) (Figure 3a) indicating connectivity of the habitat along the verges to the west, where known populations of dormice exist and possibility to the north and south of these major roads. No dormice were recorded within the Berryhill Farm area (Area O, Figure 3b). This may be due to the limited number of potential habitat connections between known populations and these areas and also the limited understorey and sheep grazing having an adverse impact on the food availability in these woodlands. No dormice were recorded across the Gwent Levels (Areas M (Pye Corner - eastern area) and N (Tata Steelworks - southern hedgerow), Figure 3e-f). C.7.4 No dormice were recorded within the Magor/Knollbury Survey Area (Areas P and I (M4 embankment only), Figure 3h) despite having suitable habitat situated further to the north of the M4. This may be due to the lack of habitat linkages such as hedgerows between potentially suitable habitat and known existing populations of dormice. Page 21

25 2015 survey results 2015 nest tube survey results New Park Farm North and South (Figure 3a) C.7.5 C.7.6 C.7.7 C.7.8 A single dormouse was recorded to the north of the M4 during the June 2015 visit (Area C, Figure 3a). Dormouse nests were recorded during further visits to this area in July and September An old dormouse nest and two dormice were recorded during the November 2015 visit. Along the M4 road verge close to New Park Farmhouse (Area J, Figure 3a), one dormouse nest was recorded with a dormouse present in July Five dormice and five dormouse nests were also recorded in this area in September 2015 and thirteen dormouse nests were recorded in the area November 2015, including one with one adult and one juvenile present. One dormouse nest was recorded along a hedge off Penylan Road (Area C, Figure 3a), at the far west of the survey area, during the survey visit in July This was recorded again in September 2015 and two different dormouse nests were recorded in the hedge in November Between the M4 and A48(M) one dormouse was recorded in a wood mouse nest (Area D, Figure 3a) during the September 2015 visit. Castleton North and South (Figure 3a) C.7.9 A single dormouse was recorded in June 2015 in the area south of Penylan Farm (Area B, Figure 3a) and dormouse nests were recorded during further visits undertaken in July, September and November C.7.10 No evidence of dormouse was recorded in the Pound Hill area (Area K, Figure 3a) during the June, July and November 2015 survey visits; however, one dormouse and three dormouse nests were recorded during the September 2015 visit. C.7.11 C.7.12 On Gwaunshonbrown Farm, two dormouse nests were recorded in July 2015 (Area B, Figure 3a); five dormouse nests were recorded each in September and November Along the verges of Pound Hill Road (Area K, Figure 3a), three dormouse nests were recorded in November Castleton Cefn Llogell (Figure 3b) C.7.13 Four dormouse nests were recorded in the Castleton Cefn Llogell area (Area A, Figure 3b) in July Five live dormice and nine nests were recorded in September and 17 potential dormouse nests were recorded in November Gwent Levels - Pye Corner (Figure 3e) C.7.14 The tubes around Pye Corner (Area M, Figure 3e) were not present/easily found so additional nest tubes were installed in September No signs of dormouse were recorded during the survey period. Gwent Levels - Tata Steelworks (Figure 3f/g) C.7.15 Access to the Tata Steelworks site was only granted in August 2015 following which a dormouse nest was recorded in a hedgerow along the southern boundary of the survey area (Area N, Figure 3f/g). An additional possible nest was later recorded in a hedgerow on site (Area N, Figure 3f). Page 22

26 Magor (Figure 3h) C.7.16 C.7.17 A single dormouse nest was recorded in a hedge along Bencroft Lane (Area I, Figure 3h) during the September 2015 visit and again in November A single dormouse nest was recorded in a hedge to the north of a small parcel of broadleaved woodland (Area L, Figure 3h) during the October 2015 visit and again during the November 2015 visit, although a wood mouse was present in the nest in October Minnet s Lane (Figure 3i) C.7.18 A possible nest of old man s beard was located in one nest tube along Minnet s Land (Area H, Figure 3i) during the November 2015 survey visit. No other evidence of dormouse has been recorded in this area. Haul Road to Ifton Quarry (Figure 3i) C.7.19 No evidence of dormouse was recorded at Roggiett Brake Wood (Area E, Figure 3i) or Quarry Wood (Area F, Figure 3i) hazel nut search results C.7.20 The locations, number of nuts collected and number of nuts that had been opened by dormice are listed in Table 5 below. The survey locations are shown on Figure 2. Table 5: 2015 Hazel Nut Search Areas and Results Search area ref. Figure 2 i ii iii Description of search area Abundant small and mid-size hazel, few hazel nuts found. Small to mid-size hazel abundant/dominant at western end. Hazel trees much sparser further east where woodland has a distinct canopy and sprawling bramble and ivy ground flora. Frequent small and mid-size hazel, frequent fresh hazel nuts. Number of collected nuts opened by mammals Number of collected nuts opened by dormice iv No hazel. 0 0 v No hazel. 0 0 vi No hazel. 0 0 ix Occasional small hazel on roundabout, 20 0 occasionally fresh hazel nuts which were mostly uneaten. No hazel in verge to west or east. x Occasional small to mid-hazel trees with fresh, 33 0 uneaten hazel nuts frequently found. xi Few mid-size hazel, hazel nuts mostly old viii No hazel. 0 0 xii No hazel to the east of St Brides road, small 7 0 hazel trees to the west with frequent uneaten fresh hazel nuts. vii Occasional/frequent hazel to the north, few 70 0 uneaten or fresh hazel nuts found. xiii Occasional small to mid-size hazel, fresh uneaten or squirrel eaten hazel nuts frequent xiv Not accessed. 0 0 xv Occasional mid-size hazel, hazel nuts mostly old Page 23

27 Search area ref. Figure 2 Description of search area Number of collected nuts opened by mammals Number of collected nuts opened by dormice xvi Very few mid-size hazel, hazel nuts mostly old xvii Very small few hazel, hazel nuts mostly old. 8 0 xviii Occasional mid-size hazel, hazel nuts mostly old C.7.21 Dormouse-opened hazelnuts were only located at search area reference iii, Figure 3a, to the south of the A48 and west of Pound Hill in Castleton. C.7.22 Several of the areas had either no hazel or no mammal-opened hazelnuts; therefore, no nuts were collected during the survey period. Access could not be gained to Location Q, Figure 2f due to the road layout and barriers that excluded access. Coed Mawr hazel nut search results C.7.23 C.7.24 C.7.25 C.7.26 C.7.27 In Area A of the Coed Mawr search area (Figure 2g), only five hazel plants were located, three of which were young plants located along the eastern edge of the wood, one was a semi-mature plant located in the north-east corner of the area, and one was a young plant located in the centre of the area. Hazel nuts were only located around the semi-mature hazel plants. All nuts were either intact or had been opened by squirrels. No hazelnuts opened by dormice were recorded. In Area B (Figure 2g), three mature hazel plants were recorded at the north of the search area. Hazel nuts collected were largely still intact; however, a small number had been opened by squirrels and wood mice. No hazelnuts had been opened by dormice. Occasional young hazel plants were recorded adjacent to the footpath in the eastern section of Area C (Figure 2g). Most of the hazel nuts collected had been opened by squirrels or were still intact, and only a handful had been opened by wood mice. No hazelnuts had been opened by dormice. Scattered hazel shrubs were recorded throughout Area D of the search area (Figure 2g), including young and semi-mature plants. The majority of hazel nuts collected were located beneath a group of semi-mature hazel plants in the eastern half of Area D. Most had been opened by squirrels and the occasional nut had been opened by wood mouse. No nuts had been opened by dormice. Three mature hazel plants were located in the centre of Area E of the search area (Figure 2g). Hazel nuts were relatively abundant; however, the majority had been opened by squirrels and the occasional nut had been opened by wood mice. No hazel nuts had been opened by dormice survey results 2016 nest tube survey results C.7.28 C.7.29 C.7.30 C.7.31 During the September 2016 survey visit, one possible dormouse nest was located in Woodview woodland in the north of the Magor/Knollbury survey area (Area L, Figure 3h). In addition, during the October 2016 survey visit, a possible dormouse nest was recorded in the motorway island to the south of the M4 (Area Q, Figure 3h/i). No other signs of dormouse activity were recorded in the areas surveyed along the route of the Scheme. With regards to Coed Mawr, one probable dormouse nest was located in a nest tube in November 2016 (Figure 3j). The nest tube was located amongst naturally- regenerating scrub (dominated by birch and bramble) in the southern tip of the woodland, where conifers had been felled in recent years. No other signs of dormouse presence were Page 24

28 recorded survey results 2017 nest tube survey results C.7.32 C.7.33 C.7.34 C.7.35 C.7.36 The results of the 2017 nest tube survey will be provided either in the next version of this Draft Dormouse Mitigation Strategy and/or within the Dormouse Method Statement for the licence application. Similarly, with regards to the Coed Mawr nest tubes, these results will also need to be presented subsequently. Summary of nest tube survey and nut search results, Taking into account the results of all nest tube surveys and nut searches undertaken since 2014, the locations of habitat known to support, or likely to support, dormouse populations and adjacent habitat of potential value to dormice are shown on Figure 4 (and discussed in Section C.8, below). Habitat assessment results Results of the 2016 and 2017 initial mapping of broad habitat types in Coed Mawr, and an assessment of potential value to dormice are shown on Figure 5. Results confirm the presence of a diversity of age structures and habitat types across the wood, including habitats of potential value to dormice. This assessment will be further updated with a more detailed habitat mapping exercise undertaken in the summer of Results of the assessment of habitats of known or likely value to dormice along the route of the Scheme that would be directly affected by construction (i.e. would need to be cleared for construction) are indicated by a full yellow line on Figure 6. The total areas of each habitat grade assessed are as follows. New Park Farm (North and South) and Castleton (North, South and Cefn Llogell) areas (Figure 6a) - Grade 2 potentially optimal habitat: hectares - Grade 2/3 potentially optimal to sub-optimal habitat: 2.13 hectares - Grade 3 sub-optimal habitat: 0.73 hectares Tata Steelworks area (Figure 6b) - Grade 2 potentially optimal habitat: 2.65 hectares - Grade 3 sub-optimal habitat: 2.04 hectares Magor/Knollbury areas (Figure 6c) - Grade 2 potentially optimal habitat: 0.86 hectares - Grade 3 sub-optimal habitat: 2.55 hectares M4/M48 islands and embankments (Figure 6c) - Grade 2 potentially optimal habitat: 2.33 hectares - Grade 3 sub-optimal habitat: 0.67 hectares Minnet s Lane (Figure 6d) - Grade 2 potentially optimal habitat: 0.13 hectares Page 25

29 - Grade 2/3 potentially optimal to sub-optimal habitat: 0.06 hectares - Grade 3 sub-optimal habitat: 0.1 hectares C.7.37 In addition, the availability of different grades of dormouse habitat within the adjacent land that could receive displaced/relocated dormice has also been assessed; these are shown as areas within the dashed yellow line on Figure 6. The areas of these parcels have not been calculated, as it is not possible to define the limits of the potential habitat that will be available to any displaced dormice. C.8 Interpretation / evaluation of survey results On-site surveys Nest tube surveys C.8.1 C.8.2 C.8.3 C.8.4 C.8.5 Results of the surveys confirm the presence of a considerable population of hazel dormice in woodlands and hedgerows in the New Park Farm and Castleton areas (Figure 3a/b). Results have also confirmed the presence of a limited dormouse population on and adjacent to the Tata Steelworks land (Figure 3f/g). Dormice in this area appear to be utilising areas of dense scrub and well-developed and connected hedgerows, which are largely confined to the Tata Steelworks survey area. This favourable habitat, especially areas of dense scrub, is uncommon in the more open parts of the Gwent Levels to the south. However, areas of dense scrub and woodland are located immediately to the north of the survey location (i.e. to the east and north of the main Tata Steelworks site, along railway lines to the north of the main Tata Steelworks site and in areas of woodland further to the north). Taking this into account, it is considered that the survey area on or around the Tata Steelworks site is likely to be at the southern extent of dormouse habitat in the area. The population recorded to the north of Magor at Knollbury (Figure 3h) and along Minnet s Lane, the proposed haul road to Ifton Quarry (Figure 3i), is also considered to be limited in size due to the limited availability of significant parcels of woodland in the survey area. It is considered that the main population from which the small number of individuals recorded is likely to inhabit the larger parcels of woodland further to the north. The dormouse nest recorded in the M4/M48 island at the eastern end of the Scheme (Area Q, Figure 3i) is considered likely to represent a small number of animals in the motorway islands/embankments, as there has so far only been one positive record. However, results of the 2017 surveys will confirm whether or not dormice have moved into the motorway islands from embankment habitat along the Scheme to the south. Should this be the case, results of the surveys would inform any potential revisions to the estimated population size with regard to this area. For the purposes of this Strategy, it has been assumed that populations in the motorway islands are linked by the existing motorway verges to the north and west of Area Q (Figure 3i), and that the potential is for much of the habitat in this area to support dormice. Hazel nut search C.8.6 All hazel nut search locations primarily overlapped with nest tube survey areas, with the exception of Fox Covert (Area R, Figure 3b). This area will be subject to a nest tube survey in Results of the survey will inform the final mitigation strategy and method statement for this area. Dormouse presence C.8.7 On the basis of the results of the nest tube surveys and hazel nut searches, as well as Page 26

30 the desk study information and habitat assessments, it is currently considered that there are three discrete areas of the Scheme where dormice are present and, therefore, where construction activities would require a derogation. These are: the New Park Farm/Castleton interchange; the area south of Tata Steelworks; and the area around Magor/Knollbury/Junction 23/Minnet s Lane. C.8.8 C.8.9 C.8.10 C.8.11 C.8.12 The likely geographical limits of these areas have been determined on the basis of a combination of survey results and the suitability of habitat beyond the positive records of presence, and are illustrated on Figure 4. The solid line denotes those areas of habitat that are either known (from the survey results), or considered highly likely (on the basis of the type of habitat present and the proximity of positive dormouse records) to contain dormice, whilst the dashed line indicates areas where there is potential but where the nature of the habitat and/or the quality of links to known dormouse habitat suggest that presence is less likely. The dotted lines indicate likely habitat linkages to the wider landscape and wider dormouse population. This exercise has been important as it demonstrates that each of the three areas where there would be direct effects on dormice is linked to adjacent habitat where there is scope for enhancement. In particular, those areas within the purple dashed line (on Figure 6) that have been assessed as Grade 2 (potentially optimal) habitat could readily be enhanced through a combination of coppicing (to increase the availability of foraging resources by opening the canopy to encourage fruiting and flowering), the planting of scrub or hedgerow plants (to increase the resources and habitat connectivity, where necessary), and/or the installation of dormouse boxes (to increase the amount of nesting and/or breeding habitat). This has informed the displacement/relocation strategy (also illustrated on Figure 6), as it has helped to identify those areas where enhancement is specifically required. This is discussed in more detail in Sections E.1 and E.2.2, below. As agreed with NRW in the Ecology SoCG (see Section B.1.8, above), the ownership of these adjacent areas of land has been established in consultation with Julia Tindale, and is discussed below. Much of the adjacent land is owned by Welsh Government (Figure 6); therefore, enhancement measures for the benefit of dormice would be undertaken as soon as practicable after the Orders have been confirmed. In addition, some land is owned by Tata Steelworks and Monmouthshire County Council (Figure 6b and 6c). Consultation with both parties has commenced with regard to carrying out enhancement works after the Orders have been confirmed: Tata have confirmed agreement in principle to enhancement of areas to the south of the Scheme; no response has yet been received from MCC. Figure 4 also indicates that the dormouse populations in the areas adjacent (and/or linked) to the locations where dormouse habitat is to be lost could accommodate additional animals, and that there is scope in all areas to implement habitat enhancement such that additional dormice could survive in nearby areas if displaced or relocated into them. Habitat assessments of immediately surrounding areas have been undertaken, and the results are shown on Figure 6. Taking the results of the habitat assessment into account, it is thus considered that adjacent habitats shown on Figures 4 and 6 would be sufficiently large to support displaced/locally relocated dormice for the relatively short period whilst newly-planted areas created as part of the Scheme develop into suitable habitat for dormice. This was one of the areas of concern of NRW in their response to the previous draft of the Strategy. Should the results of the 2017 (and indeed 2018) surveys indicate that the boundaries of population shown on Figure 4 need to be amended, these will be modified within any revisions to the Strategy and/or in the Dormouse Method Statement to be submitted with the licence application. This is important, because these lines mark the licensable area, and thus those parts of the Scheme where special mitigation measures will need to be implemented. Page 27

31 Population size estimates C.8.13 C.8.14 C.8.15 Bright et al. (2006) suggested a typical dormouse home range could cover approximately hectares of woodland or 300 m of hedgerow, whilst in optimal habitat the mean population density in spring could be up to 10 adults per hectare. The National Dormouse Monitoring Programme (NDMP) suggested an average population density of between 1.75 and 2.5 adults per hectare of favourable habitat. The People s Trust for Endangered Species (PTES) suggest the home range of a female dormouse will generally be up to 1 hectare in size and that males will have larger home territories that overlap several female home ranges. PTES also suggest that, under optimal habitat conditions, the population density could be between 3 and 5 adults per hectare. NRW have indicated that they would expect that densities could be as high as 10/hectare in optimal habitat (although this would be post-breeding, when young are present). Taking the above into account, a habitat grading system was developed for this Scheme to assess likely habitat suitability for dormice (and thus dormouse density) within those areas expected to be affected, either directly or indirectly. This has therefore been applied not only to the areas due to be lost, but also to retained adjacent habitat (into which dormice might be displaced) and also to the proposed receptor habitat at Coed Mawr (see below). The definitions of the four grades are provided in Table 4 above. Section C.7, also above, discusses the results of the habitat assessments. On the basis of the grading applied to those areas where the desk study and survey results have indicated dormouse presence (illustrated on Figure 4), it is estimated that the following numbers of dormice would need to be displaced, relocated or translocated. The areas considered for these calculations are shown on Figure 6. It should be noted that a precautionary approach has been taken to calculate estimates so as to help minimize the potential for underestimates the assumption has been made that dormice would be present in areas of likely value to dormice as well as known value to dormice (as shown on Figure 4). New Park Farm (North and South) and Castleton (North, South and Cefn Llogell) areas (Figure 6a): - Grade 2 habitat: hectares x 5 adults = 60 adults - Grade 2/3 habitat: 2.13 hectares x 3 adults = 7 adults - Grade 3 habitat: 0.73 hectares x 2 adults = 2 adults Therefore, a total of 69 adult dormice might need to be displaced, relocated or translocated from the Castleton interchange. Of these, at least 40 adults of an appropriate female:male ratio would be translocated, primarily from the 5.4 hectares of habitat on the island between the M4 and A48(M) and the land immediately to the north of this (see Figure 6a). It should be noted that this is a precautionary estimate to date dormice have not been recorded in all areas considered for this estimate and shown on Figure 6a, including the island between the M4/A(48)M, and to the south of the M4/east of Poundhill Lane. Tata Steelworks area (Figure 6b) - Grade 2 habitat: 2.65 hectares x 5 adults = 14 adults - Grade 3 habitat: 2.04 hectares x 2 adults = 4 adults Therefore, a total of 18 adult dormice might need to be displaced, relocated or translocated from the Tata Steelworks site. It should be noted that this is considered a precautionary estimate as only two Page 28

32 dormice nests have been recorded in the Tata Steelworks survey area to date. Taking into account the surrounding habitats, it is considered likely that the Tata Steelworks survey area represents the southern-most extent of the local dormouse population s range that predominantly lies to the north of the A4810, where there are well-connected parcels of scrub and woodland of potential value to dormice. This assumption is being tested in 2017, with additional nest tube surveys to the north of the A4810 as well as to the east and south of the Tata Steelworks survey area. Magor/Knollbury areas (Figure 6c) - Grade 2 habitat: 0.86 hectares x 5 adults = 5 adults - Grade 3 habitat: 2.55 hectares x 2 adults = 6 adults Therefore, a total of 11 adult dormice might need to be displaced, relocated or translocated from the Magor/Knollbury area. It should be noted that to date no dormice have been recorded in the habitat in this survey area that would be cleared for construction (the three dormouse nests recorded to date were located to the north of the Scheme). M4/M48 islands and embankments (Figure 6c) - Grade 2 habitat: 2.33 hectares x 5 adults = 12 adults - Grade 3 habitat: 0.67 hectares x 2 adults = 2 adults Therefore, a total of 14 adult dormice might need to be displaced, relocated or translocated from the M4/M48 islands and embankments. It should be noted that the above is considered to be an over-estimate due the limited size of the islands and the fact that dormice would need to cross a motorway in order to reach them. Only one possible nest has been recorded to date. Minnet s Lane (Figure 6d) - Grade 2 habitat: 0.13 hectares x 5 adults = 1 adult - Grade 2/3 habitat: 0.06 hectares x 3 adults = 1 adult - Grade 3 habitat: 0.1 hectares x 2 adults = 1 adult Therefore, a total of 3 adult dormice might need to be displaced, relocated or translocated from Minnet s Lane area. The above is considered to be a likely over-estimate due to the limited extent and low quality of the habitat, as well as the fact just one possible dormouse nest has been recorded to date. C.8.16 C.8.17 It can therefore be estimated that a total of approximately 115 adult dormice might need to be either displaced/relocated (to adjacent habitats) or translocated. In order to help minimise the number of dormice that would be affected by construction, an intensive displacement, relocation and trapping exercise would be undertaken, in accordance with an NRW approved Hazel Dormouse Method Statement, as soon as possible during the dormouse active season (i.e. from April/May 2018) and prior to the main dormouse breeding season, where permission to access has been granted. With regard to Welsh Government-owned motorway embankments and islands, and Monmouthshire County Council and Tata Steelworks-owned land, permission to access is expected to be obtained after the confirmation of the Orders (March 2018) and, Page 29

33 therefore, dormouse displacement, relocation and trapping in these areas would commence in April/May Off-site survey - Coed Mawr C.8.18 C.8.19 C.8.20 It is clear from the results of the initial habitat mapping surveys undertaken at Coed Mawr in 2016 and 2017 (Figure 5) that although the site remains a plantation dominated by conifers, there are significant areas of habitat suitable for dormice present. These include not only areas of relatively mature broadleaved woodland and woodland edge, but also areas of relatively young scrub (developing on recent clear fell) with large amounts of bramble. As discussed in Section C.7, above, whilst the nest tube surveys at Coed Mawr have revealed the presence of only one dormouse nest, in the far south-east of the site, the initial habitat mapping (see Figure 5b) has indicated that the site could currently accommodate in the order of 58 adult dormice (i.e. up to 5 adults over hectares scattered across the northern part of the wood). This is a precautionary estimate, as it excludes the potential value of the southern half of the woodland, which is separated from the northern half of the wood by areas of conifer, although these would not present a barrier to dormouse movement. The natural spread of dormouse populations is a relatively slow process (Chanin, 2014). Therefore, the distance between Coed Mawr and the closest woodland where dormice have been recorded (Figure 1b), as well as the relatively recent nature of changes to the woodland habitat (i.e. felling of sub-optimal conifer plantation), would explain why only one probable dormouse nest has been recorded to date. Further surveys are being carried out in 2017 to confirm the status of the population here. Plan A. Proposed forestry operations at Coed Mawr. [Kindly provided by Andrew Shinton, NRW]. C.8.21 The amount of suitable habitat for dormice at Coed Mawr is likely to increase significantly over the coming years. A large percentage of the larch currently on site is due to be felled in 2018 (shaded brown in Plan A, above), with the remaining areas of larch (shaded orange in the figure) scheduled for felling in Page 30

34 C.8.22 In addition, the existing areas of beech plantation in the west of the site (shaded green), which also represent sub-optimal habitat for dormice, are due to be thinned in the early 2020s. All of this activity, which is part of NRW s plan to convert the site back to broadleaved woodland, will further increase the carrying capacity of Coed Mawr for dormice, such that it is likely to be able to accommodate a very large population within the next 10 to 20 years (e.g. 450 if it is considered that the 90 hectares of woodland can support a dormouse population of 5 adults/hectare). Status of the dormouse population C.8.23 The third UK report into the implementation of the Habitats Directive (JNCC, 2013) concluded that the status of the dormouse population was bad declining, future prospects were inadequate and the overall assessment of conservation status was bad declining. Plan B. Dormouse distribution in Wales ( ) based on National Biodiversity Network. [Reproduced from the M4CaN Proof of Evidence of Dr Liz Halliwell, NRW]. C.8.24 C.8.25 In this context, it is clearly very important to ensure that the Scheme does not contribute further to this decline. Indeed, the aim of the mitigation measures set out in Section E, below, is to reverse this trend, at least in the area local to Newport. Plan B, above, illustrates that this area is within probably the most densely-populated part of Wales for dormice. It is evident from the map in Plan B and from the desk study (see Figure 1) that dormice are relatively widespread in this part of south-east Wales, with significant populations occurring within roadside landscape planting as well as within the more traditional habitat of coppiced broadleaf woodland. The majority of this habitat lies within and to the north of the existing M4 corridor (where the landscape comprises a much greater density of Page 31