Submission on Publicly Notified District Plan

Clause 6 of Schedule 1, Resource Management Act 1991

To: Hamilton City Council Date: 29/3/2013 Submission method: on-line Submission number: 305

Submitter Details: Company Email id: [email protected] Contact Name: The Waikato Tree Trust Name: The Waikato Tree Trust Address: c/- 26 Eton Drive, Hillcrest, Hamilton, 3216, New Zealand Phone daytime: 64 07 8566563 Mobile:

I wish to be heard in support of my submission. If others make a similar submission, I will consider presenting a joint case with them at a hearing. I could not gain an advantage in trade competition through this submission.

This is a submission on the Hamilton City Council Proposed District Plan (the Proposal):

The specific Support, My submission is: I seek the following decision from the local authority: provision(s) of Oppose the proposal or that my Support submission in part relates to is: 20.1 Purpose Support Note; We would appreciate for "Purpose", that it would c) The sites are identified on the Planning Maps and are listed in Schedule 9C: in part be desirable to comment specifically identifying the Long Significant Natural Areas in Volume 2, Appendix 9. The significance of a site is tailed bat (Chalinolobus tuberculata) population within the based on the ecological significance of its indigenous vegetation. The sites were city as this is a unique fauna characteristic of the city. We assessed using the criteria for determining significant indigenous vegetation and also seek that where possilbe when there is inclusion in significant habitats of indigenous fauna contained in the Regional Policy Statement (2000). Also included is a map of the identified long tailed bat (Chalinolobus commnets in relation specifically to Natural Environments tuberculata) roost sites within the city and surrounding rural area. that proetction of this species is mentioned. A survey of the city for long tailed bats in 2012 identifed locations that e) Significant Natural Areas include: the bats frequently roost, an inclusion of this map would be useful to include it the Significant Natural Areas. i. Identified areas of the Waikato River corridor and gully systems. ii. Peat lakes and wetlands.

iii. Remnant indigenous vegetation or trees, specifically in relation to areas identified that long tailed bats (Chalinolobus tuberculata)) roost.

iv. Other areas that contribute to indigenous biodiversity.

Significant Support 20.2.1 Significant Natural Areas are protected, Significant Natural Areas 20.2.1 Objective and Policies Natural Areas in part maintained, restored and enhanced. Agree, Yes like the 20.2.1 inclusion of 'enhanced' 20.2.1a The values and Objective Policies Objective and characteristics that define the City’s Significant Natural 20.2.1 20.2.1a Policies Areas shall be identified. Has this not already been Significant Natural Areas are The values and characteristics that define the completed? If so why not? How can a plan be written to protected, maintained, restored City’s Significant Natural Areas shall be identified. protect areas that haven't been identified or characteristics and enhanced. not yet defined!!!. This is essential for the District Paln, 20.2.1b without this what are we writing objectives and policies Areas of indigenous vegetation, biodiversity and for! 20.2.1c i. The character and degree of modification, habitats of indigenous fauna shall be scheduled as damage, loss or destruction that will result from the Significant Natural Areas. activity. ii. The duration and frequency of effect (e.g. long- 20.2.1c term or recurring effects). iii. The magnitude or scale of The particular values and characteristics that effect, including effects on ecological processes supporting make an area a Significant Natural Area shall be or provided by the Significant Natural Area. iv. The protected from adverse effects by having regard irreversibility of effect. v. The resilience of the area to to: assimilate change. Who determines the resilience of the i. The character and degree of modification, area to assimilate change? If the area is deemed a damage, loss or destruction that will result from 'Significant Natural Area' surely it should be enhanced not the activity. compromised by a determination of resilence to assimulate ii. The duration and frequency of effect (e.g. change. Only include if change is proactive i.e. enhancing long-term or recurring effects). an existing remnant that has extensive pest plant vegetation iii. The magnitude or scale of effect, including presnt that requires removal etc, may require methods to effects on ecological processes supporting or remove that could be deterimental to other flora in short provided by the Significant Natural Area. iv. The irreversibility of effect. term. vi. The opportunities to minimise pre-existing or v. The resilience of the area to assimilate potential adverse effects (e.g. restoration or enhancement), change. where avoidance is not practicable. As above, include if proactive enhancement. However could be seen as a vi. The opportunities to minimise pre- indication that it is OK to transition areas vii. The existing or potential adverse effects (e.g. probability of effect. How is this measured? include only restoration or enhancement), where avoidance is due to exclusion may warrent no consideration to not practicable. probability of effects and therefore no refence can be vii. The probability of effect. made. viii. Cumulative effects. Surely cumulative effects MUST be proven. In the area of new develoments the viii. Cumulative effects. cumulative efffects may not be known for some time or evidence may take years to gather i.e. therefore research. ix. Need for, or purpose of, the works. This poses the question who makes the assessment and 20.2.1d how is the information provided? e.g. a major inland port Adverse effects of development on the City’s development may have a signifcant effect on "Signifcant Significant Natural Areas shall be avoided. Natural Areas" due to the increased risk of a biosecurity 20.2.1e breach (opening many containers that could contain The reduction, fragmentation and isolation of biological threats to native flora a & fauna). Is it therefore indigenous ecosystems and habitats shall be a compromise to offset economic gains for the avoided. environmental degradation? ix. Need for, or purpose of, 20.2.1f the works. The greater good of the environment?, society? The loss or disruption of corridors or connections or the bank? Need to consider however could be seen to linking indigenous ecosystems and habitat be used in an argument to justify removal of significant fragments shall be avoided. vegetation or habitat distruction. 20.2.1d Adverse effects 20.2.1g of development on the City’s Significant Natural Areas The loss or disruption to migratory pathways in shall be avoided. 20.2.1e The reduction, fragmentation water, land or air shall be avoided. and isolation of indigenous ecosystems and habitats shall 20.2.1h be avoided. 20.2.1f The loss or disruption of corridors or Adverse effects on ecosystems resulting from connections linking indigenous ecosystems and habitat changes to hydrological flows, water levels and fragments shall be avoided. Agree, specific to long tailed water quality shall be avoided. bats and their movements around the city. 20.2.1g The 20.2.1i loss or disruption to migratory pathways in water, land or The loss or disruption of protective buffering of air shall be avoided. Agree 20.2.1h Adverse effects on indigenous ecosystems shall be avoided. ecosystems resulting from changes to hydrological flows, 20.2.1j water levels and water quality shall be avoided. Agree, The loss of ecosystem services shall be avoided. water quality in the city streams and aquatic ecosystems needs to improve significantly, require filters in stormwater 20.2.1k pipes. (Kevin Collier paper) 20.2.1i The loss or disruption The loss, damage or disruption to ecological of protective buffering of indigenous ecosystems shall be processes, functions and ecological integrity shall avoided. 20.2.1j The loss of ecosystem services shall be be avoided. avoided. 20.2.1k The loss, damage or disruption to 20.2.1l ecological processes, functions and ecological integrity The loss or reduction of the cultural and spiritual shall be avoided. 20.2.1l The loss or reduction of the association with indigenous biodiversity which are cultural and spiritual association with indigenous held by tangata whenua shall be avoided. biodiversity which are held by tangata whenua shall be 20.2.1m avoided. 20.2.1m Non-native pest species within Non-native pest species within Significant Natural Significant Natural Areas shall be eradicated. 20.2.1n The Areas shall be eradicated. loss of habitat that supports indigenous species under 20.2.1n threat of extinction shall be avoided. 20.2.1o Significant The loss of habitat that supports indigenous Natural Areas shall be restored and enhanced to meet at species under threat of extinction shall be least the 10% threshold for habitat sustainability. Definitely avoided. agree! This is a minimum requirement! 20.2.1o Significant Natural Areas shall be restored and enhanced to meet a minimum of10% of the cities total land area for habitat sustainability. 20.3 Rules – Support Support rules, Remove "Trimming" add "Pruning" instead 20.3 Rules – Activity Status Table Activity Status in part in all text. Add require a professional Arboriculture report Table on any the status of all " Signifciant Trees" on listed on the Activity Class the register 'Volume 2, Appendix 9, Schedule 9D: Significant Trees', biannually and for catergory 1 trees, Activities within a Significant Natural Area, Schedule 9C (Volume 2, Appendix annually. 9) a) Pruning and maintenance of existing indigenous vegetation or trees where: P i. Necessitated by disease or age ii. It affects the operation of existing network utilities iii. Necessitated to maintain existing tracks and fencing and this will not result in the death, destruction, irreparable damage of indigenous vegetation and trees, or a reduction in the particular values and characteristics that make an area a Significant Natural Area b) Removal of dead or damaged indigenous vegetation or trees where this is P necessary to: i. Maintain or enhance the particular values and characteristics that make an area a Significant Natural Area ii. Ensure the operation of existing network utilities iii. Maintain existing tracks and fencing c) Pest control P d) Planting and management of eco-sourced indigenous vegetation or trees P e) Emergency works to, or removal of, an indigenous tree where: P i. There is an imminent threat to life, property or a network utility ii. The tree carries a fatal disease f) Removal of exotic trees or pest plants P g) The following activities located within any Significant Natural Area NC Schedule 9C (Volume 2, Appendix 9) i. Earthworks ii. The laying or forming of any impervious surface iii. Additions to, or the replacement of, any existing building or structure that is proposed to exceed the envelope or footprint of the existing building(s) or structure(s) iv. The placement and/or construction of any building or structure v. Directional drilling or boring vi. The storage of chemicals or other toxic substances vii. Removal of, or transplanting indigenous vegetation Volume 2, Appendix 9, Schedule 9D: Significant Trees h) Emergency works to, or removal of, a scheduled tree where: P i. There is an imminent threat to life, property or a network utility ii. The scheduled tree carries a fatal disease i) Minor pruning and maintenance of a scheduled tree P j) The following activities located within the root protection zone of any RD scheduled tree i. Earthworks ii. The laying or forming of any impervious surface iii. Additions to, or the replacement of, any existing building or structure that is proposed to exceed the envelope or footprint of the existing building(s) or structure(s) iv. The placement and/or construction of any building or structure v. Directional drilling or boring vi. The storage of chemicals or other toxic substances k) Non-emergency works to, removal of, or transplanting of a scheduled D tree Note

1. For activities and buildings in the Electricity Transmission Corridors see Chapter 25.7: Citywide – Network Utilities and Electricity Transmission Corridors.

2. All trees listed under 'Volume 2, Appendix 9, Schedule 9D: Significant Trees', require an arboriculturalist report annually and for catergory 1 trees, and biannually for allotehrs listed. 20.4.1 Support 20.4.1 Trimming and Maintenance a) Maximum amount of 20.4.1 Pruning and Maintenance Trimming and foliage to be removed per calendar year 15% b) Maintenance Maximum thickness (cross-section) of any branch or root a) Maximum amount of foliage to be removed per calendar year 15% that may be cut 50mm It is not TRIMMING but b) Maximum thickness (cross-section) of any branch or root that may 50mm PRUNING is the terminology used in arboriculture , be cut woudl be best to reiiterae this in this document. NOTE: A qualified arboriculturalist MUST only be engaged in any trimming and maintenance. A report must be made prior to any pruning or removal of branches and roots of Catergory 1, Signifcant Trees to indicate the works to be undertaken and extent. Thsi will be checked and commented on by a qualified arborist engaged by council. 20.4.1 Support 20.4.1 Trimming and Maintenance a) Maximum amount of 20.4.1 Pruning and Maintenance Trimming and foliage to be removed per calendar year 15% b) Maintenance Maximum thickness (cross-section) of any branch or root a) Maximum amount of foliage to be removed per calendar year 15% that may be cut 50mm It is not TRIMMING but b) Maximum thickness (cross-section) of any branch or root that may 50mm PRUNING is the terminology used in arboriculture , be cut woudl be best to reiiterae this in this document. NOTE: A qualified arboriculturalist MUST only be engaged in any trimming and maintenance. A report must be made prior to any pruning or removal of branches and roots of Catergory 1, Signifcant Trees to indicate the works to be undertaken and extent. Thsi will be checked and commented on by a qualified arborist engaged by council. 20.4.2 Support 20.4.2 Emergency Works to, or Removal of, an 20.4.2 Emergency Works to, or Removal of, an Indigenous Tree in a Significant Emergency in part Indigenous Tree in a Significant Natural Area or a Natural Area or a Scheduled Tree1 Works to, or Scheduled Tree1 a) Confirmation of the necessity for the Removal of, an works or removal shall be provided to Council: i. Before a) Confirmation of the necessity for the works or removal shall be provided to Indigenous any works are undertaken, and ii. The works shall be Council: Tree in a carried out by an appropriately qualified person (e.g. an Significant arborist). Requires a standard to measure this against or i. Before any works are undertaken, and Natural Area we will have anyone calling themselves an arborist 1 ii. The works shall be carried out by an appropriately qualified person (e.g. an or a Scheduled Immediate legal effect only applies to an indigenous tree in arborist with either a qualification to level of NZQA or extensive experience and Tree1 a Significant Natural Area. All trees identified as long tailed proof of understanding of arboricutlrual processes). bat (Chalinolobus tuberculatus) roost sites are portected 1 under Wildlife Act 1953. The Department of Conservation Immediate legal effect only applies to an indigenous tree in a Significant Natural Area. All trees identifed as long talied bat (Chalinolobus tuberculatus) roost sites are protected under WIldlife Act is responsible for administering this Act on all land in New 1953. The Department of Conservation is responsible for administering this Act on all land in New Zealand. Zealand. 20.6 Support Agree 20.6 Notification Rule Notification Rule a) Except as provided for by Section 95A(2)(b) and (c), 95B(2) and (3) and 95C(1) to (4) of the Act applications for any Restricted Discretionary Activity identified with an asterisk (*) in the table above will be considered without notification or the need to obtain approval from affected persons.

9-1.16 Method Support Costs have increased signifcantly over past 15 years 9-1.16 Method of Applying a Monetary Value of Applying a in part increase cost to $45 per unit. Monetary A monetary value is obtained by multiplying the evaluation score by the unit value. A Value value of $45 per unit has been adopted, based on the 2013 cost of purchase and establishment of a tree scoring 1.

The value of an individual tree, in a group or stand of more than 10 trees, shall be the unit value multiplied by the average score for 10 typical trees as established in the registration procedure. The stand value will be the individual value multiplied by the number of trees. Schedule 9C Support To achieve a minimum of 10% of indigenous flor and fauna Schedule 9C: Significant Natural Areas Significant in the city it is critical to protect these areas as "Signifcant Site Number Name Map Number (refer to Natural Areas Natural Areas" refer Key ecological sites of Hamilton City: Planning Volume 1 Cornes, Toni S. ; Thomson, Rachel Elizabeth ; Maps) Clarkson, Bruce D. SNA 1 Te Rapa North kahikatea I 7B SNA 2 Te Rapa North kahikatea II 7B http://researchcommons.waikato.ac.nz/handle/10289/6565 SNA 3 River Road North Gully 7B, 8B Include Whatukoruru Reserve FItzroy area, Peacocks SNA 4 Riverside alder with tree ferns 7B, 8B Area Some designations may require renaming i.e. SNA SNA 5 Pukete kanuka Gully I 16B, 8B SNA 6 Riverbank mahoe scrub, Pukete 16B, 8B 49 Hammond Bush there is also another area by SNA 7 Pukete Riverside mamaku-mahoe Forest 16B, 17B Tuikaramea Road referred to locally as Hammond Bush. SNA 8 Pukete kanuka Gully II 16B SNA 31 Claudelands Bush is Jubilee Park (Te Papa Nui) SNA 9 Pukete Riverside kanuka 17B SNA 10 Puketaha astelia Gully 19B SNA 11 Burbush Road Forest/Perkins Bush 14B SNA 12 Horseshoe Lake (Waiwhakareke Natural Heritage Park) 33B SNA 13 Riverbank north of Pukete Bridge 17B SNA 14 Kirikiriroa Gully, Harrowfield 17B Site Number Name Map Number (refer to Planning Maps) SNA 15 Totara Park 17B SNA 16 Mooney Street kahikatea 25B SNA 17 Kirikiriroa Gully adjacent to Gordonton Road 19B SNA 18 Kirikiriroa Gully, Chartwell 19B SNA 19 Riverbank opposite St Andrew’s Golf Course 18B, 27B SNA 20 St Andrew’s kanuka 27B SNA 21 Donny Park raupo 27B SNA 22 Riverbank opposite Anne Street 27B SNA 23 Ranfurly Park, Fairfield 36B, 37B SNA 24 Kirikiriroa Gully, Mangaiti 18B, 19B SNA 25 Nawton Wetland 33B SNA 26 Brymer Park 33B SNA 27 Lake Rotokaeo (Forest Lake) 35B SNA 28 Grove Park kahikatea 42B SNA 29 Waitawhiriwhiri Gully, Edgecumbe Park 36B SNA 30 Waitawhiriwhiri Gully, Whitiora 35B SNA 31 Claudelands Bush 37B, 38B SNA 32 Riverbank south of Miropiko 37B SNA 33 Seeley’s Gully 37B, 38B, 45B, 46B SNA 34 Peachgrove kahikatea 46B SNA 35 Mixed planted forest and totara forest near Golf Area 44B Hamilton Lake Domain SNA 36 Lake Rotoroa 44B, 45B, 54B, 55B SNA 37 Southwell Bush 38B SNA 38 Caldwell Native Bush 38B SNA 39 Waikato University kahikatea 47B SNA 40 Hillcrest kahikatea 47B SNA 41 Mangaonua Gully, Chelmsford Park 48B SNA 42 Mangaonua Gully, Silverdale 48B SNA 43 Temple View kahikatea 60B SNA 44 Graham Island (Te Motere o Kaipikau) 56B SNA 45 Riverbank east of Cobham Drive 56B SNA 46 River Island with turf vegetation 56B SNA 47 Mamaku-mahoe forest, Hamilton Gardens Riverbank 57B Mamaku-kamahi forest, Hamilton Gardens SNA 48 Riverbank kanuka opposite Hammond Park 57B SNA 49 Hammond Bush 58B, 57B SNA 50 Gully near Hammond Bush I 58B SNA 51 Gully near Hammond Bush II 58B SNA 52 Riverside kanuka, Hammond Park 58B SNA 53 Mangaonua streamside, Riverlea 58B SNA 54 Riverside kanuka, Peacocke 58B, 65B SNA 55 Mangakotukutuku gully, Te Anau Park 63B SNA 56 Mangakotukutuku gully, Peacocke 64B SNA 57 Mangaonua gully, Berkley 58B SNA 58 Mangaonua streamside, Riverlea 58B Site Number Name Map Number (refer to Planning Maps) SNA 59 Mangaonua gully arm, Riverlea 58B

Include: Whatukoruru Reserve Fitzroy (Peacocks Rd area) Long tailed bat roosting sites Schedule 9D Support Minor corrections required to Botanical names T37 Tilea Add listed but edit; Significant europeaa , should read: Tilia europea T66 Podocarpus Trees totara 'Avera' , should read Podocarpus totara 'Aurea' T37 Map 37B Tilia europea T69 Idesia polycarpo, should read: Idesia polycarpa T75.6 Tilin europea, should read: Tilia europea T66 Podocarpus totara 'Aurea' T69 Idesia polycarpa

T75.6 Tilia europea 13-1 Areas Support All Significant Natural Areas in Schedule 9C No tonly As per schedule specifically; with Historic would the asethetic visual appearance of Heritage Tecommunications and utility networks impair the All Significant Natural Areas in Schedule 9C Values or cahracter of these areas there maybe determental effects Visual Amenity to fauna , e.g. long tailed bats. Values 13-1 Areas Support All Significant Natural Areas in Schedule 9C No tonly As per schedule specifically; with Historic would the asethetic visual appearance of Heritage Tecommunications and utility networks impair the All Significant Natural Areas in Schedule 9C Values or cahracter of these areas there maybe determental effects Visual Amenity to fauna , e.g. long tailed bats. Values 24.2 Support 24.2.1a Financial contributions shall be applied in a fair As per schedule with editing change "compliments" not Complements Objectives and and equitable manner that: i. Is financially transparent. ii. Policies Complements (edit should be Compliments) Council’s 24.2.1a Financial other financial management policies. Financial contributions shall be applied in a fair and equitable manner that: Contributions i. Is financially transparent.

ii. Compliments Council’s other financial management policies. 24.3.1 Purpose Support Agree Agree with schedule of Financial Contribution 24.3.1 Purpose Support Agree Agree with schedule of Financial Contribution 24.3.3 Level of Support Reserves · The amount of money and/or land needed to As above Contribution ensure Council meets its agreed level of service for reserves as per Rule 24.4.1 · Whether the need for 24.3.3 Level of Contribution reserves has been identified in a relevant Structure Plan · Whether other funding sources, including Development a) The level of contribution shall be up to 100% of the actual or estimated costs, or Contributions, apply · Whether neighbourhood reserve land necessary to provide for the “Reasons for Financial Contribution” specified in capacity exists within 500m of the proposed activity · the table below. Whether there is a need to provide public access to the b) Actual or estimated costs may include: reserve · The need to protect significant natural features, biodiversity, sites of cultural, heritage or archaeological i. All reasonable costs incurred in providing the public utility service, transport value · The need for reserves to ensure appropriate infrastructure or reserve. amenity values · The effects of the development or subdivision on the capacity of reserves that serve a city- ii. Any reasonable costs in avoiding, remedying or mitigating any effects on the wide or greater function To meet the biodiversity environment of providing or upgrading the public utility service, transport obligations under Agenda 21 this is required to protect infrastructure or reserve. aBiodiversity. If hte cityis to achieve a minimum of 10% indigenous flora cover and increase fauna e.g. Tui bellbird c) The level of financial contribution shall include any associated costs including but and Kereru, long tailed bats then this is required with not limited to: financial support. Evaluating the Welfare Effects of Biodiversity on Private Lands: A Choice Modelling i. The reimbursement of legal costs incurred by Council in providing easements, Application encumbrances, covenants and the like. ii. The reimbursement of fees charged to Council by Government departments, local authorities and the suppliers of public utilities and infrastructure.

iii. Survey work.

iv. Any fees incurred by an adjoining local authority or network operator in processing the application.

v. Goods and Services Tax (GST).

d) The estimated or actual cost of financial contributions, whether provided or constructed by the Council or otherwise shall be calculated in accordance with, but not limited to, the following matters for consideration:

Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution Water Supply · Where an existing supply · Whether the new Systems is available, the cost of infrastructure/ connection with the upgrade/extension is already existing system identified as a growth-related · Where an existing supply project is available, but the · Whether other funding capacity of the system is sources, including inadequate to meet the development contributions, additional generated apply demand, the cost of · The quality and quantity of the connection and capacity supply upgrading of the existing · The effect any additional system connections may have on the · Where an existing supply existing system, its users is not available, the cost and/or on the quality and of providing for the quantity of the supply supply of water · The age and value of the existing infrastructure · The proximity and directness of the effect Stormwater · Where an existing outfall · Whether the new Disposal is available, the cost of infrastructure/ Services connection with the upgrade/extension is already existing stormwater identified as a growth-related system project · Where an existing outfall · Whether other funding is available, but the sources, including capacity of the system is development contributions, inadequate to meet the apply additional generated · The quality and quantity of the demand, the cost of supply Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution connecting and capacity · The effect any additional upgrading of the connections may have on the stormwater system existing system, its users · Where an existing outfall and/or on the quality and is not available, the cost quantity of the supply of providing a · The age and value of the stormwater system existing infrastructure · The proximity and directness of the effect Wastewater · Where an existing · Whether the new Disposal system is available, the infrastructure/ Services cost of connection with upgrade/extension is already the existing system identified as a growth-related · Where an existing project system is available, but · Whether other funding the capacity of the sources, including system is inadequate to development contributions, meet the additional apply generated demand, the · The quality and quantity of the cost of connection and supply capacity upgrading of · The effect any additional the existing system connections may have on the · Where an existing existing system, its users system is not available, and/or on the quality and the cost of providing a quantity of the supply sewerage system · The age and value of the existing infrastructure · The proximity and directness of the effect Transport · The costs of specific works · Whether the new needed to service the use, infrastructure subdivision or development infrastructure/ and Access and/or mitigate their effects upgrade/extension is already identified as a growth-related project Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution · Whether other funding sources, including development contributions, apply · The current standard and estimated carrying capacity of the transport network the subdivision, development or land use will connect to · The current number of users of the transport network and the estimated increase in number of users as a result of the subdivision, development or land use · The sensitivity and location of activities adjoining the transport corridor and adjacent to the subject site · Sight distances and the presence of blind spots along adjacent transport corridors and the standard and adequacy of intersections · The need to improve the transport network to accommodate additional traffic generated (taking into account both type and numbers of traffic generated) by the subdivision, development and/or land use · The benefit to the subdivision or land use arising from the improvement to the transport Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution network relative to the benefit to existing users and other members of the public · The estimated number of future users of the transport network, assuming degrees of development and growth anticipated by the standards in the District Plan · The likely route from the site to key locations in the City Reserves · The amount of money · Whether the need for reserves and/or land needed to has been identified in a ensure Council meets its relevant Structure Plan agreed level of service · Whether other funding for reserves as per Rule sources, including 24.4.1 Development Contributions, apply · Whether neighbourhood reserve capacity exists within 500m of the proposed activity · Whether there is a need to provide public access to the reserve · The need to protect significant natural features, biodiversity, sites of cultural, heritage or archaeological value · The need for reserves to ensure appropriate amenity values · The effects of the development or subdivision on the capacity of reserves that serve a city- Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution wide or greater function Offset Effects · The amount of money and/or land needed to offset any adverse environmental effects

24.3.3 Level of Support Reserves · The amount of money and/or land needed to As above Contribution ensure Council meets its agreed level of service for reserves as per Rule 24.4.1 · Whether the need for 24.3.3 Level of Contribution reserves has been identified in a relevant Structure Plan · Whether other funding sources, including Development a) The level of contribution shall be up to 100% of the actual or estimated costs, or Contributions, apply · Whether neighbourhood reserve land necessary to provide for the “Reasons for Financial Contribution” specified in capacity exists within 500m of the proposed activity · the table below. Whether there is a need to provide public access to the b) Actual or estimated costs may include: reserve · The need to protect significant natural features, biodiversity, sites of cultural, heritage or archaeological i. All reasonable costs incurred in providing the public utility service, transport value · The need for reserves to ensure appropriate infrastructure or reserve. amenity values · The effects of the development or subdivision on the capacity of reserves that serve a city- ii. Any reasonable costs in avoiding, remedying or mitigating any effects on the wide or greater function To meet the biodiversity environment of providing or upgrading the public utility service, transport obligations under Agenda 21 this is required to protect infrastructure or reserve. aBiodiversity. If hte cityis to achieve a minimum of 10% indigenous flora cover and increase fauna e.g. Tui bellbird c) The level of financial contribution shall include any associated costs including but and Kereru, long tailed bats then this is required with not limited to: financial support. Evaluating the Welfare Effects of Biodiversity on Private Lands: A Choice Modelling i. The reimbursement of legal costs incurred by Council in providing easements, Application encumbrances, covenants and the like. ii. The reimbursement of fees charged to Council by Government departments, local authorities and the suppliers of public utilities and infrastructure.

iii. Survey work.

iv. Any fees incurred by an adjoining local authority or network operator in processing the application. v. Goods and Services Tax (GST). d) The estimated or actual cost of financial contributions, whether provided or constructed by the Council or otherwise shall be calculated in accordance with, but not limited to, the following matters for consideration:

Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution Water Supply · Where an existing supply · Whether the new Systems is available, the cost of infrastructure/ connection with the upgrade/extension is already existing system identified as a growth-related · Where an existing supply project is available, but the · Whether other funding capacity of the system is sources, including inadequate to meet the development contributions, additional generated apply demand, the cost of · The quality and quantity of the connection and capacity supply upgrading of the existing · The effect any additional system connections may have on the · Where an existing supply existing system, its users is not available, the cost and/or on the quality and of providing for the quantity of the supply supply of water · The age and value of the existing infrastructure · The proximity and directness of the effect Stormwater · Where an existing outfall · Whether the new Disposal is available, the cost of infrastructure/ Services connection with the upgrade/extension is already existing stormwater identified as a growth-related system project · Where an existing outfall · Whether other funding is available, but the sources, including Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution capacity of the system is development contributions, inadequate to meet the apply additional generated · The quality and quantity of the demand, the cost of supply connecting and capacity · The effect any additional upgrading of the connections may have on the stormwater system existing system, its users · Where an existing outfall and/or on the quality and is not available, the cost quantity of the supply of providing a · The age and value of the stormwater system existing infrastructure · The proximity and directness of the effect Wastewater · Where an existing · Whether the new Disposal system is available, the infrastructure/ Services cost of connection with upgrade/extension is already the existing system identified as a growth-related · Where an existing project system is available, but · Whether other funding the capacity of the sources, including system is inadequate to development contributions, meet the additional apply generated demand, the · The quality and quantity of the cost of connection and supply capacity upgrading of · The effect any additional the existing system connections may have on the · Where an existing existing system, its users system is not available, and/or on the quality and the cost of providing a quantity of the supply sewerage system · The age and value of the existing infrastructure · The proximity and directness of the effect Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution Transport · The costs of specific works · Whether the new needed to service the use, infrastructure subdivision or development infrastructure/ and Access and/or mitigate their effects upgrade/extension is already identified as a growth-related project · Whether other funding sources, including development contributions, apply · The current standard and estimated carrying capacity of the transport network the subdivision, development or land use will connect to · The current number of users of the transport network and the estimated increase in number of users as a result of the subdivision, development or land use · The sensitivity and location of activities adjoining the transport corridor and adjacent to the subject site · Sight distances and the presence of blind spots along adjacent transport corridors and the standard and adequacy of intersections · The need to improve the transport network to accommodate additional traffic generated (taking into account both type and numbers of traffic generated) by the Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution subdivision, development and/or land use · The benefit to the subdivision or land use arising from the improvement to the transport network relative to the benefit to existing users and other members of the public · The estimated number of future users of the transport network, assuming degrees of development and growth anticipated by the standards in the District Plan · The likely route from the site to key locations in the City Reserves · The amount of money · Whether the need for reserves and/or land needed to has been identified in a ensure Council meets its relevant Structure Plan agreed level of service · Whether other funding for reserves as per Rule sources, including 24.4.1 Development Contributions, apply · Whether neighbourhood reserve capacity exists within 500m of the proposed activity · Whether there is a need to provide public access to the reserve · The need to protect significant natural features, biodiversity, sites of cultural, heritage or archaeological value · The need for reserves to Basis of Reasons for Financial Matters for Consideration in Contribution Contribution Determining Level and/or Nature of Financial Contribution ensure appropriate amenity values · The effects of the development or subdivision on the capacity of reserves that serve a city- wide or greater function Offset Effects · The amount of money and/or land needed to offset any adverse environmental effects

24.3.4 Credits Support Support Support above schedule 24.4.1 Support a) A minimum of 4ha per 1000 population shall be 24.4.1 Reserves Level of Service Reserves Level in part provided for passive activities. Should be a minimum of of Service 10ha per 1000 population if to achieve goal of minimum of a) A minimum of 10ha per 1000 population shall be provided for passive 10% indigenous vegetation cover. At 4ha this will require activities. the city to increase in population growth by 237, 500. This is unrealistic, even at 10ha it will still require an increase in b) A minimum of 3.5ha per 1000 population shall be provided for active outdoor population (developments to accomodate etc) 95,000. recreation and organised sports. c) Neighbourhood Reserves shall be a minimum of 0.5ha per 1000 population and shall be within 500m of residential dwellings.

d) Reserves shall be acquired in accordance with any relevant Structure Plan. 25.2.1 Purpose Support Trees are an important part of the city landsscape "City of 25.2.1 Purpose Trees" is one name Hamilton is known for large , rare and signifcant trees, nationally , and internationally. a) Earthworks refer to the disturbance of land by moving, removing, placing or replacing soil or earth by any means. Earthworks are a necessary part of land subdivision and development, but can result in adverse effects including accelerated erosion and sedimentation, contamination of fresh water, and increased risks from natural hazards. Earthworks can also impact on amenity values, including an unnatural look of the modified land.

b) Hamilton City is predominantly an urban environment. Trees make an important contribution to the health and wellbeing of the residents of the City and to the quality of the City’s landscape. Vegetation removal can impact on biodiversity and ecosystems within the City, and the urban amenity of the City.

c) The Waikato Regional Council and Waikato Regional Plan have primary responsibility under the Act for controlling land use for soil conservation and water quality. The District Plan has a supporting role, as the District Plan controls subdivision and development of land.

d) This chapter outlines earthworks and vegetation removal rules relating to the zones, and cross-references to chapters where more specific rules relating to earthworks and vegetation removal are outlined. 25.13.1 Support Reuired to achieve outcomes of a sustainable city - support schedule as outlined purpose. Purpose Agenda 21. 15.3 Rules – Support ac) Planting, trimming and maintenance of vegetation or Agree with schedule' Activity Status in part trees. Change in all text 'trimming' to 'pruning'. All pruning Table and remoal of vegetation within the 'Signifcant Natural Change all 'Trimming' in text to 'pruning' e.g. Areas' and including river and gully areas must have consideration to the effects on soil structure. Extnesive ac) Planting, pruning and maintenance of P P P P slips have occurred in areas along the river banks within vegetation or trees private and public land due to lack of consideration of soil structure (mostly Waikato sand - refer Bruce, J.G. 1979. Note Soils of Hamilton City, North Island, New Zealand. New 1. For activities and buildings in the Electricity Transmission corridors see Chapter Zealand Soil Survey Report 31. 65p + 1 sheet 1:20,000.), rill erosion and bank instability. In some instances the 2. All areas identified as river, gully and stream bank, specifically with soils of 'Waikato Sand' and management of pest plants will require a management plan 'Horitu Sandy Loam' dsignation under soil maps of to effectively eradicate pest plants while establishing Bruce, J.G. 1979. Soils of Hamilton City, North Island, New Zealand. New Zealand Soil Survey Report desired vegetation. In some instances this will require a 31. 65p + 1 sheet 1:20,000. long term approach. Land mangers will need resources to manage this sustainably e.g. it may take several years of manipulation to gain the desired outcome and therefore long term budgets and maintenance of these areas will need ot have committed funds to undertake. 15.6.3 Support Agree , protection and use of "Significant Natural Areas" 15.6.3 Organised Recreation in the Natural and Neighbourhood Open Space Zone Organised as outlined in Appendix 9, specifically require sensitive Recreation in treatment when utilisied for recreation use. a) Participants shall use existing walkways, cycleways, structures, buildings, facilities the Natural and and landform in the manner intended. Neighbourhood Open Space Zone 9.2.2 Objective Support Amenity values are important, planting of trees to offet 9.2.2 Objective and Policies and Policies carbon emissions and sequenting of carbon. We still ned to assess each Industrial ZOne as warming temperatures Objective Policies could limit the ability of tree to continue to effectively do 9.2.2 9.2.2a this. Reference: Melillo, J., Butler, S., Johnson, J., Mohan, The amenity levels of industrial Amenity levels within the Industrial Zone shall be J., Steudler, P., Lux, H., Burrows, E., Bowles, F., Smith, areas are to be enhanced. improved with the use of landscaping and R., Scott, L., Vario, C., Hill, T., Burton, A., Zhouj, Y, and screening, restrictions on site layout, enhanced Tang, J. Soil warming carbon-nitrogen interactions and design of buildings, ensuring orientation of buildings carbon-nitrogen budgets. Proceedings of the National towards the site frontage, and enhanced urban Academy of Sciences, May 23, 2011 DOI: design outcomes. 10.1073/pnas. 1018189108 Some land area that are Explanation zoned Industrial e.g. Pukete/Te Rapa , Maui St area Although lower standards of amenity are often characteristic of industrial conect directly to "natural Area' and park and open space locations, Plan provisions aim to enable a general improvement in the amenity use. Critical that these areas are not degraded but an of the City’s industrial locations. The purpose of this is to create functional athetic link is also made to connect landscapes between and attractive employment areas and to contribute to raising amenity levels these areas e.g. River to Pukete to Te rapa by vegetation within the City generally. as well as road linkages. This is to be achieved through resource consent being required for the construction of new buildings or alterations to existing buildings (other than minor alterations) to ensure improved urban design outcomes. There are also requirements for increased landscaping, articulation of building frontages and screening (particularly on the boundary of residential land and reserves), and site layout.

9.2.3 Objective Support Amenity values are important, planting of trees to offet 9.2.3 Objective and Policies and Policies carbon emissions and sequenting of carbon. We still ned to assess each Industrial ZOne as warming temperatures Objective Policies could limit the ability of tree to continue to effectively do 9.2.3 9.2.3a this. Reference: Melillo, J., Butler, S., Johnson, J., Mohan, The adverse amenity impacts The adverse effects of industrial activities shall be J., Steudler, P., Lux, H., Burrows, E., Bowles, F., Smith, of industrial activities on contained within the Industrial Zone boundary to R., Scott, L., Vario, C., Hill, T., Burton, A., Zhouj, Y, and residential and open space avoid adverse effects on amenity within other Tang, J. Soil warming carbon-nitrogen interactions and areas are to be avoided. zones, particularly the Residential, Special carbon-nitrogen budgets. Proceedings of the National Character and Open Space Zones. Academy of Sciences, May 23, 2011 DOI: 9.2.3b 10.1073/pnas. 1018189108 Some land area that are The establishment of noxious or offensive activities zoned Industrial e.g. Pukete/Te Rapa , Maui St area in locations near the boundary with Residential, conect directly to "natural Area' and park and open space Special Character and Open Space Zones, where use. Critical that these areas are not degraded but an there will be adverse amenity effects on these athetic link is also made to connect landscapes between locations, shall be avoided. these areas e.g. River to Pukete to Te rapa by vegetation Explanation as well as road linkages. Industrial activities can generate adverse amenity effects beyond the boundaries of the zone. These can have a particular impact on residential and open space areas where expectations for amenity are far higher.

The Amenity Protection Area is a key mechanism to protect residential sites where they are adjacent to land within the Industrial Zone. Industrial properties covered by the Amenity Protection Area are subject to additional standards. Enhanced management of noxious or offensive activities where they are near residential land uses is also a key aspect of the provisions.

9.2.4 Objective Support Reduces conflict between residential use and industrial, 9.2.4 Objective and Policies and Policies in part efficentcy and economy however the same rules should be applied in reverse, where there are areas of Amenity, Objective Policies Recreation or Natural Areas the impact of Industrial 9.2.4 9.2.4a facilites that would be in conflict need to be identifed and To optimise the benefits of the Logistics, freight-handling services and supportive mitigated. What will the limit of this type of development regionally significant freight activities and infrastructure shall be provided for be in th efuture? Continuaation of the use of CLass 1 soils village facility at Crawford Crawford Street Freight Village. within the CIty and surrounding area continues to impact Street. Limit future 9.2.4b on agriculture and horticultural businesses in the future. development of this site as Activities sensitive to the adverse affects of identifed on planning map. logistics and freight-handling facilities shall avoid locating in proximity to the Crawford Street Freight Village. Explanation The facility falls within the definition of regionally significant infrastructure within the Regional Policy Statement which provides for its protection, and requires that particular regard is given to the benefits that can be gained from its development and use. It provides connectivity between dairy manufacturing facilities in the region and further afield with the Auckland and Tauranga ports. The facility is a critical component in ensuring the efficiency of dairy manufacturing and export within the region. Limits must however be set to reduce continued land use in this manner where soils are identifed for horticullture or agriculture use.

9.3 Rules – Support I would like the current mixed nature of the industrial area Tighten the rules so that it will be very difficult for these types of activities to establish Activity Status in part to be preserved: a mix of service and light industrial and operate in this area. Specifically the Riverlea Road industrial area in close Table activity, and mixed residential and industrial. The close proximity to residential areas and some of the city’s most Significant Natural Areas. proximity of the Riverlea Road industrial area to residential areas and some of the city’s most Significant Natural Areas means that it is unsuitable for heavy industry, especially noxious and offensive activity, and activities generating large amounts of traffic, particularly heavy commercial vehicles. 11.1 Purpose Support Areas in close proximity require at least 50m offset form Include any development within areas bounding residential or Signifcant Natural Areas is in part the activity and planted densley to offset noise by a offset by landscape formation planted with indigenous ecologically sourced plants to nominal 10 decibels, a better approach if needed would be connect areas such as streams, gullies and 'Signifcant Natural Areas' to have mounded landscape formations with indigenous plantings to connect with 'Significant Natural Areas' and gullies and stream areas. 21.1 Purpose Support Agree with comments 21.1 Purpose

a) The Waikato-Tainui Raupatu Claims (Waikato River) Settlement Act 2010 establishes Te Ture Whaimana o Te Awa o Waikato – the Vision and Strategy for the Waikato River – as the primary directionsetting document for the management of the river (refer to Volume 2, Appendix 10). The Vision and Strategy forms part of the Regional Policy Statement. This chapter provides a mechanism for giving effect to the Vision and Strategy.

b) The Waikato River is an outstanding natural feature within the Waikato Region. Additionally, the Waikato River and gully systems form the most prominent landscape feature within Hamilton. These areas have economic, transport, recreational, ecological, amenity, landscape and cultural values. This chapter recognises that these values come together to contribute to the significance of the Waikato River corridor and gully systems. These values are collectively managed through a variety of approaches.

c) Management approaches that apply within the Waikato River corridor and gully systems are: significant natural areas (Chapter 20: Natural Environments); scheduled archaeological and cultural sites (Chapter 19: Historic Heritage); and natural hazard areas (Chapter 22: Natural Hazards). Where land is in public ownership the underlying zone is Open Space, while for areas that are in private ownership the underlying zone varies. Together the various management approaches address the values associated with the Waikato River corridor and gully systems.

d) The objectives and policies are achieved through various management approaches throughout the District Plan, so there are no rules contained within this chapter. The following objectives and policies must be read in conjunction with the relevant objectives and policies from other chapters. It should also be acknowledged that there are a number of non-District Plan methods that can be implemented with regard to the river corridor and gully areas (refer to Volume 2, Appendix 1, Section 1.8: Other Methods of Implementation). 21.1 Purpose Support Agree with comments 21.1 Purpose a) The Waikato-Tainui Raupatu Claims (Waikato River) Settlement Act 2010 establishes Te Ture Whaimana o Te Awa o Waikato – the Vision and Strategy for the Waikato River – as the primary directionsetting document for the management of the river (refer to Volume 2, Appendix 10). The Vision and Strategy forms part of the Regional Policy Statement. This chapter provides a mechanism for giving effect to the Vision and Strategy.

d) The objectives and policies are achieved through various management approaches throughout the District Plan, so there are no rules contained within this chapter. The following objectives and policies must be read in conjunction with the relevant objectives and policies from other chapters. It should also be acknowledged that there are a number of non-District Plan methods that can be implemented with regard to the river corridor and gully areas (refer to Volume 2, Appendix 1, Section 1.8: Other Methods of Implementation). 21.2.1 Support Agree with comments. 21.2.1 Objective and Policies Objective and Policies Objective Policies 21.2.1 21.2.1a The ecological, amenity, An integrated, holistic and co-ordinated approach landscape and cultural values of to management shall be used to protect, enhance the river corridor and gully and restore the natural, physical, cultural and system are restored and historical resources and character of the river protected. corridor and gully system. 21.2.1b Development and activities that have significant impacts on landform shall be controlled, particularly the:

i. Clearance of vegetation along the river and gullies.

ii. Filling of gullies, including the cumulative effects of such incremental filling. 21.2.1c The ecological functions of waterways shall be restored and protected by minimising the modification of natural watercourses and riparian margins. 21.2.1d The relationship of Waikato-Tainui with the Waikato River shall be restored and protected. 21.2.1e The relationships of the Waikato Region’s communities with the Waikato River shall be restored and protected. 21.2.1f The loss or disruption of corridors or connections provided by the Waikato River corridor and gully systems which link indigenous ecosystems and habitat fragments shall be avoided. 21.2.1g The connectivity and protective buffering of indigenous ecosystems provided by the Waikato River Corridor and gully system shall be maintained. Explanation The Waikato River corridor and associated gully system is the City’s key landscape and natural feature. It contains significant pockets of indigenous vegetation and provides an important ecological corridor and a wilderness experience within the City. The river has cultural significance for Waikato iwi and contains numerous sites of historical and cultural importance. The management approaches referred to in 21.2.1a include significant natural areas, scheduled cultural sites, the High, Medium and Low Flood Hazard Areas, Temple View Flood Hazard Area, Culvert Block Flood Hazard Areas, the Waikato Riverbank and Gully Hazard Area and Open Space Zone.

Hamilton City Bat Survey 2011-2012

WAIKATO TREE TRUST

HAMILTON CITY-WIDE BAT SURVEY 2011-2012

Table of Contents

Summary ...... 1 1 Introduction ...... 2 2 Methods & Materials ...... 2 2.1 Site descriptions ...... 2 3.2 Monitoring design and equipment ...... 3 3.3 Data collection and classification ...... 3 3.4 Statistical analyses...... 4 3 Results ...... 5 4.1 Presence/absence ...... 5 4.3 Habitat use ...... 6 4.4 Landscape and anthropogenic variables ...... 8 4 Discussion, Conclusions & Management Recommendations ...... 12 5 References ...... 16 6 Acknowledgments ...... 17 7 Appendix I ...... 18

Prepared by: Darren S. Le Roux & Noa N. Le Roux

Reviewed by: Associate Professor Stuart Parsons (University of Auckland) and Gerry Kessels

Except for the purposes of individual study or fair review for which the authors must be acknowledged, no part of this report may be copied, scanned, stored in any type of electronic file or retrieval system or published in any other form without the express written permission of Kessels & Associates Ltd.

Document Ref: \\server files\Hamilton City Council\Bat survey city wide\city wide survey report_240512

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 1

Summary

Reduced biodiversity in urban ecosystems is often attributed to the loss and fragmentation of suitable habitat for wildlife. To date, few studies have investigated if and how New Zealand bat species use urban environments. Only two bat species – long-tailed bats (Chalinolobus tuberculatus) and lesser short-tailed (Mystacina tuberculata) bats – are found in New Zealand, both of which are endemic and classified as nationally threatened by the Department of Conservation (DOC). We used bat detectors to conducted presence/absence surveys at 62 ‘green space’ habitats (0.7-92 ha) to better understand bat distribution and habitat use patterns in Hamilton City. Long-tailed bat activity was confirmed at 16 sites (25.8%), all of which were restricted to the most southern urban-rural fringe of the city. Although 14 of these habitats (87.5%) were classified as ‘riparian margins’ or ‘major gullies’ situated 0-100 m from the Waikato River (a major linear landscape feature), significantly higher pass rates were recorded at a rural indigenous forest remnant (Whewell’s Bush). Only six sites (<10%) showed any evidence of foraging activity and nightly activity patterns to suggest possible or likely roosting by bats. Habitat connectivity or distance to the Waikato River/major gullies emerged as the single most significant explanatory variable in our statistical model, highlighting the importance between habitat type and distance to the river/gullies for bats. Overall, bat activity significantly increased with: 1) decreasing distances from well-connected habitats and linear landscape features (gullies and river); and 2) increasing distances from the city centre and levels of human activity. Pass rates were consistently highest at habitats where houses, roads and street lights were lowest. Even slight increases in the number of roads and street lights resulted in decreases in pass rates of 86% and 70%, respectively. Riparian margins, with dense native and exotic trees and shrubs associated with riverine and gully landscapes, appear to be critical habitat, as bats depend on access to key resources associated with these environments. In particular these habitats provide: 1) mature exotic and native vegetation for roosting purposes; 2) emergent aquatic insect prey (e.g. mosquitoes) for foraging; 3) freshwater for drinking; and 4) linear landscape corridors for movement and navigation. Our results show the importance of maintaining, restoring and perpetuating these well connected, less developed habitats for long-tailed bats in Hamilton City. Habitat restoration and bat conservation efforts should thus strategically focus on preserving existing foraging and roosting habitats and improving habitat connectivity by reducing the effects of human activity (e.g. low light regimes). Future major growth cells for Hamilton city are currently focused on the southern urban-rural interface with several major roading and housing developments proposed. This study underscores the importance of making the urban landscape (both present and future) more permeable to long-tailed bats as well as protecting and enhancing existing well-connected bat habitats. If long-tailed bats are to be retained within Hamilton City in the face of ongoing urban expansion, major collective conservation decision-making by key stakeholders, as well as the implementation of multiple adaptive management strategies, will be required.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 2

1 Introduction

Better understanding how species, especially cryptic threatened populations, are distributed within the urban landscape and affected by human activities can guide decision-making by conservationists, wildlife managers, planners and developers. Bats are a challenging group of animals to monitor and manage due to their nocturnal and mobile nature (Fenton, 2003). In human-dominated environments bat management and conservation efforts remain absent or misguided due to a lack of knowledge about how bats use these ecosystems (O' Shea et al., 2003). Few studies have investigated if and how New Zealand bat species use human-dominated environments like major cities (Dekrout, 2009; Le Roux, 2010). It is often assumed that threatened species are not found in urban ecosystems, which is often the case, however, for cryptic species such as bats this may not always be true. Only two bat species are found in New Zealand, long-tailed bats (Chalinolobus tuberculatus) and lesser short-tailed bats (Mystacina tuberculata), which form the entirety of New Zealand’s native terrestrial mammal fauna (O'Donnell, 2005). Long-tailed bats are a nationally threatened species classified as ‘Nationally Vulnerable’ in the North Island (O’ Donnell et al., 2010) with ongoing population declines attributed to the loss and fragmentation of habitats and pest animal (e.g. cats and ship rats) predation and competition. Long-tailed bats are strict aerial insectivores that rely on 40 kHz frequency-modulated echolocation calls for navigation and foraging on the wing (Parsons et al., 1997). Individuals roost in hollows and under split bark typically associated with mature and dead native and exotic trees (O’ Donnell, 2001; Borkin & Parsons, 2011). Hamilton City (North Island, New Zealand) is one of the only known cities in New Zealand to still support bats within the City’s urban boundaries (Dekrout, 2009; Le Roux, 2010). This is despite the Hamilton Ecological District being one of the most degraded in New Zealand with c. 1.6% of the original native vegetation remaining (Clarkson & McQueen, 2004). To date, there has been no single city-wide survey to catalogue bat distribution and identify the factors which may influence this. The objectives of this study were to: 1) obtain a comprehensive understanding of bat distribution and habitat use patterns in and around Hamilton City; 2) develop a comprehensive, publicly accessible online bat distribution map and database; 3) determine which landscape features (i.e. habitat type and connectivity) and anthropogenic variables (i.e. housing, roading and street lighting) best explain bat distribution and habitat use patterns; and 4) provide strategic management recommendations for the ongoing conservation of bats in Hamilton City.

2 Methods & Materials

2.1 Site descriptions

Hamilton City (37°47’S, 175°17’E) is New Zealand’s fourth largest city with a total area of 9,800 ha that supports a population of c.150,000 people. We completed long-tailed bat presence/absence surveys at 62 sites in and around Hamilton City during austral spring and summer months (between September 2011 and January 2012), when bat activity is highest (Le Roux, 2010). Selected survey sites included both rural (N = 6; situated within 4 km of the city but outside of designated city boundaries) and urban (N = 56; within designated city boundaries) ‘green space’ habitats with features that had the highest potential of supporting roosting and/or foraging bats (i.e. mature native and exotic vegetation, proximity to water bodies and/or edge habitat). Sites were situated at varying distances (0-4,700 m) from the Waikato River – a major linear landscape feature bisecting the city. Four major gully systems are situated throughout the city. The Mangakotukutuku and Mangaonua gullies situated along the southern urban-rural interface of Hamilton City are the largest of the four gullies and together with Waikato River

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 3 form the single largest and most continuous ecotone in Hamilton. Conversely, the Kirikiriroa and Waitawhiriwhiri gullies (where no bat activity was detected in this survey) are situated within the urban matrix in highly developed areas in the northern part of the city. All sites were categorized into four major habitat types found throughout Hamilton. Habitats included: 1) Major gullies - well vegetated native and/or exotic corridor systems >50m from the banks of the Waikato River connecting habitats such as forest fragments with riparian margins; 2) Riparian margins - native and/or exotic vegetation immediately flanking (0-50m) the banks of the Waikato River; 3) Urban parklands - designated public recreational areas within the city’s boundaries dominated by large open grassy space, mature native and exotic vegetation, and/or artificial or natural waterbodies (e.g. lakes); and 4) Native forest remnants - urban and rural forest fragments <12ha in size dominated by mature native emergent vegetation (e.g. kahikatea (Dacrycarpus dacryiodes) and totara (Podocarpus totara)).

3.2 Monitoring design and equipment

Automated heterodyne bat detectors were used to remotely and passively record bat echolocation pulses (manufactured by DOC; Lloyd, 2009). Detectors are made with similar sensitivities and for the purposes of this study it was assumed that each detector had an equal chance of detecting echolocation pulses, although this may in fact vary with microhabitat and detector location (Le Roux, 2010). Detectors were calibrated to have the same time and date settings (NZST) and were pre-set to start monitoring 30 minutes before sunset until 30 minutes after sunrise. On average 15-20 detectors were available for use at one time, which limited the number of sites that could be concurrently surveyed. Each site was surveyed once for five consecutive nights, irrespective of weather conditions. The order in which sites were monitored was randomised. At each site, the distance between detector locations was at least 25 m apart to increase the chance of independent bat monitoring. All detectors were secured on trees and orientated upward at 30-45°C from the horizon. The number of detectors allocated to each site was dependent on the size of the habitat. Surveyed areas ranged in size from 0.7 to 92 ha. On average, for sites < 1 ha we allocated one detector (range 1- 2); for sites between 1 and 10 ha three detectors were allocated (range 1-9); for sites > 10 ha, seven detectors were deployed (range 2-10). Equipment failure, theft and limited availability of detectors meant that detector allocation per area was not always kept consistent. As a result bat activity measures were standardised and calculated per unit effort (i.e. passes/detector/night).

3.3 Data collection and classification

Recorded bat echolocation files were sorted by visual and auditory inspection of waveforms using Bat Search 1.02 Software ® (DOC, 2008, New Zealand; Lloyd, 2009). Individual sound files were sorted into: 1) echolocation passes, defined as a series of two or more high frequency echolocation pulses emitted in sequence by flying bats (Parsons et al., 1997); and 2) non-bat sounds (i.e. wind, rain or insect generated noise) that were discarded. All passes were classified into one of two echolocation categories: 1) search phase pulses with low pulse repetition rates (mean inter-pulse interval of c. 104 ms) likely used for commuting and/or locating prey; and 2) feeding buzzes consisting of a series of rapidly emitted pulses (mean inter-pulse durations of c. 4.5 ms) used to determine the range of prey prior to capture (Parsons et al., 1997). If a file contained one or more feeding buzzes, it was classified as a single feeding buzz only. All echolocation pulses were recorded with a date (day/month/year) and time (hour/minute/second) stamp.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 4

By assessing the amount, type and temporal peaks in nightly echolocation activity we were able to distinguish between three different ways in which bats were using habitats. Habitat use included: a) Commuting - sites with no feeding buzzes and ≤ 0.2 pass/detector/night. b) Foraging and possible periodic roosting - sites with feeding buzzes and ≥ 1 pass/detector/night with activity peaks recorded within the first hour after sunset and again before sunrise indicative of roost emergence and return. c) Foraging and likely regular roosting - sites with feeding buzzes and ≥ 20 passes/detector/night with clear bimodal peaks in activity after sunset and before sunrise indicative of roost emergence and return.

3.4 Statistical analyses

For each site with confirmed bat activity (N =16) we calculated the mean number of search phase pulses and feeding buzzes/detector/night. By pooling data across these sites, we were able to calculate the mean number of passes for each hour after official sunset for: commuting habitats; foraging and possible periodic roosting habitats; and foraging and likely regular roosting habitats. To test the null hypotheses that bat activity (passes/detector/night) is not influenced by 1) habitat type; 2) habitat connectivity; and 3) anthropogenic variables, we used a bootstrap Generalised Linear Model (GLM) that incorporated both categorical (e.g. habitat type) and continuous variables (e.g. distance) into the analyses. This involved rank transforming pass rates to minimize the effect of extreme random spikes; calculating F-values in a GLM; and subsequently generating null F distributions using a bootstrap GLM to calculate P-values. To account for habitat connectivity and landscape features we measured both the distance of each site to the city centre and the distance of each site to the Waikato River or a major gully system. To account for the influence of anthropogenic variables, we estimated the density of three predominant variables according to the number/hectare and/or type of each variable for each surveyed site using a combination of aerial map inspections and physical counts (Table 1). The number of houses and streetlights/hectare was calculated from counts made within 50 m around the perimeter of each site. We also counted and classified roads within 50 m around the perimeter of each site. We ranked density values and allocated a score between 1 (lowest) and 5 (highest) for housing, roading, and street lighting for all sites. Because these three anthropogenic variables are interconnected and influence each other within the urban landscape (i.e. sites with high housing typical also have high roading and thus street lighting and vice versa), we needed to account for this issue as these data suffer from multicolinearity. To do this we calculated a single combined mean ‘human activity’ score which assimilated all three independent variables at each site. We thus tested the significance of all anthropogenic variables collectively in the GLM. Habitat type for all 62 sites was also incorporated as a variable in the GLM, however we also completed an additional analysis to determine if there were any differences in rank transformed pass rates across the four habitat types for sites with confirmed bat activity only. A bootstrap one-way ANOVA was used to do this. All statistical analyses were completed using bootstrap macros written in MINITAB® 13.30 (Minitab Inc. 2000). Bootstrap iterations run for each analysis were kept constant at 2000. The significance threshold was held at P < 0.05, although weak (P < 0.1) significant results/interactions were also considered. Means are presented with ± SEM. Untransformed data are displayed in all figures.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 5

Table 1 Scores between 1 (lowest) and 5 (highest) for the three anthropogenic variables characterising the urban landscape with corresponding ranges in the density and/or type of each variable (X = density for each site) Score # houses/ha # and type of # street roads lights/ha 1 0 ≤ X < 1 Few (1-3) rural 0 ≤ X < 1 roads 2 1 ≤ X < 5 Few (1-3) 1 ≤ X < 5 residential roads 3 5 ≤ X < 10 Main road OR 5 ≤ X < 10 surrounded by residential roads 4 10 ≤ X < 20 Main road AND 10 ≤ X < 20 surrounded by residential roads 5 20 ≤ X Major state 20 ≤ X highway

3 Results

4.1 Presence/absence

In total, bats were monitored at 254 fixed locations for 600 hours over 75 survey nights. This resulted in a combined surveyed area of 628.4 ha or c. 70% of the available ‘green spaces’ in and around Hamilton City. No lesser short-tailed bats were detected. Of the 62 sites surveyed 16 (25.8%) had confirmed long-tailed bat activity. See Figure 1, Appendix I, and the following URL address for the online public distribution map: http://maps.google.co.nz/maps/ms?vps=2&hl=en&ie=UTF8&oe=UTF8&msa=0&msid=2096835100367 76183064.0004b8677777cd753a43f

4.2 Habitat type All 16 bat habitats are restricted to the southern most urban-rural interface of the city and are situated within the same continuous ecotone (i.e. along the banks of the Waikato River and within or in close proximity to the two largest gully systems – Mangakotukutuku and Mangaonua; see Figures 1 and 2). Furthermore, 14 (87.5%) of the confirmed bat habitats were classified as ‘riparian margins’ (7 sites; 43.75%) or ‘major gullies’ (i.e. situated between 0-100 m from the Waikato River or a major gully which feeds into the Waikato River; Table 2). The only urban parkland (Resthills Park), and indigenous forest remnant (Whewell’s Bush1) with any bat activity were both 450 m from the nearest gully system. Bootstrap one-way ANOVA analysis was carried out using bat activity only and comparing and between habitat types. This revealed that pass rates were significantly higher (P = 0.02; F = 3.7) for Whelwell’s Bush compared with all other habitat types. This is despite native forest remnants being the most poorly

1 A 11.5 ha mature kahikatea dominated Scientific Reserve managed by the Department of Conservation

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 6 represented habitat type in the urban landscape and thus our data set. Major gullies (0.658 ±0.35) and urban parklands (0.086 ± 0.06) had the lowest pass rates (passes/detector/night) when compared to other habitat types. However, when ‘habitat type’ was incorporated into the more realistic bootstrap GLM it did not emerge as a significant explanatory variable (P = 0.11, F = 2.13).

4.3 Habitat use

A total of 1,898 echolocation passes were recorded, 94% of which were identified as search phase passes (6% feeding buzzes). Only 40 detectors recorded echolocation passes, 70% of which recorded a total of ≤10 passes over five nights while just 15% of detectors recorded a total of ≥50 passes over a five night period.

Of the 16 sites with bat activity, a majority (N =10; 62.5%) were identified as commuting habitats with low nightly activity (0.02 ± 0.009 mean passes/hour/night). This was characterised by search phase activity only, normally with one or two passes recorded randomly between the 2nd and 8th hours after sunset (Figure 4). The number of sites (N = 3) identified as foraging and possible periodic roosting habitats accounted for 18.7% of the bat habitats. At these sites activity was characterised by a relatively small series of nightly peaks in search phase passes (0.1 ± 0.08 mean passes/hour/night) and feeding buzzes (0.01 ± 0.008) occurring between the 1st and 11th hours after sunset. Similarly, the number of sites (N = 3) identified as foraging and likely regular roosting habitats accounted for 18.7% of the bat habitats or just 3% of the total sites surveyed. Nightly activity patterns at these three sites were the highest overall and were characterised by distinct bimodal peaks in search phase (2.35 ± 1.35 mean passes/hour/night) and feeding activity (0.53 ± 0.38) occurring between the 1st and 3rd hours after sunset and again between the 7th and 11th hours after sunset.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 7

Figure 1 Long-tailed bat distribution map for Hamilton City, highlighting major landscape and anthropogenic features as well as all 16 habitat locations with confirmed bat activity

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 8

Figure 2 Mean passes (± SEM) /detector/night for search passes and feeding buzzes recorded at all 16 sites with bat activity (site names from left to right correspond to site numbers 1-16 in Figure 1). Combined mean values (feeding and search) for each site are displayed above error bars and superscript numbers 1 and 2 correspond to habitats categorized as ‘foraging and possible periodic roosting’ and ‘foraging and likely regular roosting’, respectively, while values italised number correspond to ‘commuting’ habitats (see Figure 4).

4.4 Landscape and anthropogenic variables

Bat activity significantly (P < 0.001; F = 10.4) increased with a decreased distance to the Waikato River and/or major gully systems (Figure 3). This relationship was highly significant, but not necessarily linear in nature, as illustrated by the peaks in Figure 3. The only major exception to this trend was the high levels of pass rates recorded at the native forest remnant (Whewell’s Bush). Conversely, bat activity significantly (P = 0.02; F = 5.62) increased with an increase in distance from the city centre (Figure 3). Invariably, this meant that bat activity was highest at rural habitats (13.17 ± 7.57) compared with urban habitats (3.028 ± 2.39), despite an overall higher number of urban sites with bat activity (Figure 3). Pass rates more than doubled at sites > 9 km from the city centre compared with sites < 9 km from the city centre.

Overall, bat activity was highest at habitats that were allocated the lowest scores (1) for roading, street lighting and housing density (Figure 5). For roading in particular, pass rates dramatically decreased by 86% with a slight increase in variable score (i.e. from 1 to 2; Figure 5). A similar trend was observed for street lighting and housing with decreases of 70% and 42% between scores of 1 and 2, respectively. Tested in isolation in the bootstrap GLM, roading (P = 0.05, F = 4.02), street lighting (P = 0.98, F = 0.0) and housing (P = 0.33, F = 0.96) all emerged as weakly significant variables, however, as previously discussed these variables are inextricably interconnected and therefore these data suffer from multicollinearity. To overcome this we incorporated a more robust combined mean ‘human activity’ score accounting for these interactions. ‘Human activity’ was a significant (P = 0.01; F = 6.3)

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 9 explanatory variable of bat activity in this model, with increasing human activity resulting in reduced bat activity. There were no significant interactions between habitat type and distance from the city centre (P = 0.34) but there were weak significant interactions between ‘habitat type’ and mean ‘human activity’ (P = 0.09) and ‘habitat type’ and ‘habitat connectivity’ / distance to the Waikato River or major gully (P = 0.07).

Figure 3 Mean passes (± SEM) /detector/night for rural and urban habitats with confirmed bat activity considering the distance (m) of sites to the Waikato River or the nearest major gully system (A) and the distance (km) of sites to the city centre (B). Mean values for each site are displayed above error bars and the number of sites at each distance is presented in brackets (note: * on x-axis in Figure 3A indicates a change in scale, as bats were not detected at >500m).

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 10

Figure 4 Mean passes (± SEM) /detector /night for each hour after sunset for habitats categorised as commuting (A), foraging and possible periodic roosting (B); and foraging and likely regular roosting (C; note: y-axes scales vary)

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 11

Figure 5 Mean passes (± SEM) /detector /night for each score (1-5) allocated for all three anthropogenic variables (A) as well as the mean ‘human activity’ score (B). All 62 surveyed sites are considered.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 12

Table 2 Summary of the survey effort allocated to the four major habitat types along with corresponding mean passes/detector/night and rank means (bootstrap one-way ANOVA) for sites with bats. Habitat types with the same superscript letter above rank means do not have significantly (P > 0.05) different levels of bat activity. Habitat # sites Area (ha) # detectors # sites with Mean (±SEM) type surveyed (% surveyed (% allocated (% bats (% passes/detector total) total) total) totals) /night (rank means ± StDev)* Major 16 (25.81%) 114.1 (18.17%) 77 (30.31%) 7 (43.75%) 0.66 ± 0.35 gullies (6.72 ± 3.37) a Riparian 19 (30.64%) 184.1 (29.28%) 72 (28.35%) 7 (43.75%) 10.92 ± 5.09 margins (7.79 ± 4.33) a Urban 24 (38.71%) 311 (49.51%) 94 (37.01%) 1 (6.25%) 0.09 ± 0.06 parklands (4.83 ± 2.48) a Native 3 (4.84%) 19.1 (3.04%) 11 (4.33%) 1 (6.25%) 19.67 ± 14.99 forest remnants (12.97± 0.93) b Total 62 628.41 254 16

4 Discussion, Conclusions & Management Recommendations

As far as we are aware this is the most comprehensive acoustic bat survey undertaken to date in any New Zealand city. Presence/absence survey results revealed that the lesser short-tailed bat was not detected at any of the 62 sites surveyed in this study. This supports the assumption that this species is locally extinct in Hamilton City. This was anticipated as this species is critically endangered with a current known distribution limited only to a few managed mature indigenous forest reserves and predator free offshore islands (O’ Donnell et al., 2010). Long-tailed bats, the more widespread of the two native bat species found in New Zealand, were confirmed at 16 of the 62 sites surveyed. Long-tailed bat activity is thus confined to a relatively small number of sites with a distribution pattern restricted to the southern most urban-rural fringe of the city. This result is consistent with previous studies (Dekrout, 2009; Le Roux, 2010). The distribution pattern of bats in Hamilton City is likely to be partially explained by the presence of the most well connected ecotone in the southern urban-rural landscape, which is made up of the two largest gully systems (Mangaonua and Mangakotukutuku) that feed into the Waikato River and its associated riparian margins. Indeed, 14 (87.5%) of the confirmed bat habitats were classified as riparian margins or major gullies (7 sites or 43.75% for each) within 0-100 m of the Waikato River. The Waikato River is a major habitat connecting landscape feature which long-tailed bats are known to use as a corridor to move between habitats (Dekrout, 2009). This explains why habitat connectivity or distance to the Waikato River/major gullies emerged as the single most significant explanatory variable in our statistical model. This also explains the interaction between habitat type and distance to the river/gullies. Riparian margins, with dense native and exotic trees and shrubs associated with riverine and gully landscapes appear to be critical habitat, as individuals depend on access to key resources associated with these environments. In particular these habitats provide: 1) mature exotic and native vegetation for roosting purposes;

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 13

2) emergent aquatic insect prey (e.g. mosquitoes) for foraging; 3) freshwater for drinking; and 4) linear landscape corridors for movement and navigation, are most important to tree-dwelling aerial insectivorous species (Verboom et al., 1999; Fukui et al., 2006; Borkin & Parsons, 2011). Access to key resources (e.g. roosting sites) by bats in a highly modified landscape may be further constrained by human activities (Borkin & Parsons, 2011; Threlfall et al., 2012). Our results highlighted this, as bat activity significantly decreased with increasing distances to the city centre and habitats with the lowest scores for roading, housing and street lighting density had the highest bat activity. The effect of anthropogenic variables on bat activity is the most plausible explanation for why long-tailed bats do not utilise available foraging (Waikato River margins) and potential urban indigenous forest roosting habitats (e.g. Jubilee Park/Te papa Nui - also known as Claudelands Bush) extending north into the city. This is despite individuals being renowned for maintaining large home ranges and capable of sophisticated navigation (O’ Donnell, 2005). In fact, just 3 passes, or 0.1%, of bat activity were recorded downstream of Cobham Bridge – the first major well lit road crossing along the Waikato River corridor. These findings are consistent with a previous hand-held detector survey undertaken throughout Hamilton City in 2006-7 (Dekrout, 2009) and explains our statistical interactions between ‘habitat type’ and ‘human activity’. This conclusion is also supported by previous radio-tracking undertaken in Hamilton, which showed that although some individuals maintained large home ranges extending out into the rural landscape (mean 338 ha; range 0.8 to 7.3 km), most of their activity was concentrated within small core areas, with high roost fidelity (Dekrout, 2009). This is in contrast to bat use of unmodified native forests where long-tailed bat home ranges may be 1000’s of hectares in size with roost switching occurring on average every 1.8 days (O’ Donnell, 2001). It is not surprising then that individuals would concentrate their nightly activity, and thus optimise energetic expenditure, at a few core habitats where key resources are abundant and disturbance variables are lowest. This is common behaviour for many insectivorous bat species, particularly females that have higher thermoregulatory demands overall (e.g. Borkin & Parsons, 2011; Sedgeley, 2001; Safi et al., 2007). Recent research findings have highlighted the sensitivity of some echolocating bat species to anthropogenic variables. Stone et al., (2008) experimentally demonstrated that lesser horseshoe bats (Rhinolophus hipposideros) in the UK significantly reduced their commuting activity during high-light treatments compared with controls with no evidence of habituation. Kerth and Melber (2008) showed that Bechstein’s bats (Myotis bechsteinii) were reluctant to cross a major state highway in Germany and Schaub et al., (2008) demonstrated that greater mouse-eared bats (Myotis myotis) avoided foraging in high vehicle noise environments. Our results suggest that the restricted use of the urban landscape by long-tailed bats is partly explained by ‘barrier effects’ due to anthropogenic factors including roads, bridges and artificial lights. Across all habitats bat activity was highest at those sites with the lowest levels of street lighting, roading and housing density. Even slight increases in the number of roads and street lights in particular, resulted in considerable decreases in bat pass rates. Once again this is consistent with results from a previous study that found a significant negative correlation between long- tailed bat activity in Hamilton and housing (P = -0.863) and streetlight density (P = -0.961; Dekrout, 2009). Studies further afield have also found similar trends for urban sensitive insectivorous bat species (e.g. Threfall et al., 2012). A majority of the sites (62.5%) with bat activity in this study had very low pass rates with infrequent nightly use. Most of these sites included riparian margins and gully systems. Only six (37.5%) sites had any foraging activity at all. These same six sites (Whewell’s Bush, Hammond Bush2, Meridian 37 oaks, Hamilton Cemetery, Sandford Park, and The Narrow’s Golf Course) also had corresponding nightly activity patterns (i.e. bimodal peaks in activity) indicative of possible and likely roosting behaviour and likely support either singular and/or communal roosts. Sites with the most pronounced patterns of biomodality and the highest pass rates included just three habitats or only 3% of the total sites surveyed. All three of these habitats (Whewell’s Bush, Hammond Bush, Meridian 37 oaks) also support the highest concentrations of mature native and/or exotic vegetation which long-tailed bats are known to depend on for diurnal and nocturnal roosting, rearing young and breeding. Only two of these six sites are situated

2 The Hamilton City Council reserve formally known as Hammond Park

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 14 within the city’s boundaries, highlighting that bat roosting and foraging behaviour is more concentrated in rural environments rather than within the city’s boundaries. Bat activity was consistently highest at a low number of rural sites compared with much lower activity recorded at a higher number of urban sites. The fact that more habitats with bats were found within the city is likely an artefact of our sampling effort as we concentrated surveys at urban ‘green spaces’ within the city boundaries. Moreover, a vast majority of these sites had extremely low activity. Significantly higher levels of activity were recorded at Whewell’s Bush compared with all other habitat types. It is not surprising that the bat pass rate was highest at this site as this is the single largest (11.4 ha) mature native forest remnant within at least an 11 km radius of the city. This habitat also likely supports the highest concentration of hollow-bearing trees and thus roosting opportunities. Whewell’s Bush also differs from other sites in that it is further (450 m) from the Waikato River and major gullies. It is inevitable that rural habitats situated further from the city centre have fewer anthropogenic variables, which in turn likely makes the rural landscape more permeable to bat movement. This would allow individuals to occupy rural habitats like Whewell’s Bush, despite it being located further from core foraging and/or commuting linear landscape elements (e.g. Waikato River and gullies). It is becoming increasingly apparent that with more bat monitoring undertaken in the wider rural landscape, more rural indigenous forest fragments (exotic and native) are being identified as long-tailed bat habitats (e.g. Pukemokemoke Reserve, Maungatautari Reserve, Pirongia Forest Park). It is important to note that a failure to detect bats at a site does not necessarily mean that bats are entirely absent from that site or that they will remain absent going into the future. Ongoing bat monitoring and future city-wide surveys will invariably continue to improve our understanding of long- tailed bat distribution and habitat use patterns in Hamilton. Results from this survey should be used as a baseline data set to compare changes in bat distribution and habitat use patterns over longer time periods. Further experimental research is also needed to better elucidate the effect of light and roads on bat behaviour, which will enable more targeted mitigation and management measures to be formulated. Given that long-tailed bat populations are likely under additional pressure due to predation (Pryde et al., 2005) and competition by introduced species for roost sites (O'Donnell, 2000), further restriction of access to core habitats and disturbance/destruction of roosts through urban expansion is likely to exacerbate population declines. Results from this study have allowed us to develop three key strategic bat management and conservation recommendations for Hamilton. These recommendations should better inform and guide conservation planning and impact assessment and should not be viewed as being mutually exclusive (i.e. all recommendations and actions should be considered equally and collectively). They are: 1) Maintaining and perpetuating habitat connectivity: Our results emphasise the importance of well- connected habitat corridors and linear landscape features for long-tailed bats, with bat activity concentrated at sites 0-100 m from the Waikato River and major gullies along the southern urban-rural interface only. Enhancing and restoring habitat connections will likely facilitate bat movement into adjacent habitats and maintain existing habitat usage and commuting corridors. This might involve the following actions:  Ongoing gully restoration and enhancement;  Ongoing riparian restoration and enhancement;  Establishing appropriate buffer zones to retain gully and riparian values; and  Maintain linear landscape features and vegetation complexity (e.g. hedgerows and vegetation diversity) 2) Protecting existing bat habitats: The restricted site occupancy of bats in Hamilton underscores the importance of protecting habitats currently used by bats, particularly those supporting foraging and roosting resources. This might involve the following actions at existing bat habitats:  Long-term legal protection (e.g. Reserve Act or QEII open space covenants) of existing habitats;  Mature native and exotic vegetation retention (possible existing bat roosts);

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 15

 Young native and exotic vegetation retention (possible future bat roosts);  Active re-vegetation initiatives;  Pest management;  Artificial roost provision (bat boxes);  Bat monitoring incorporated as part of land use and subdivision consent conditions, in order to measure long-term and cumulative effects of developments and the effectiveness of associated mitigation measures; and  Offsets required if identified on-site threats to long-tailed bats cannot be avoided or sufficiently mitigated. 3) Reducing the effect of anthropogenic factors: Human variables, such as roads and artificial lighting, likely contribute to the restricted use of the urban landscape by long-tailed bats. Although this species is somewhat tolerant of human disturbances, given that it is found within the city’s boundaries, our results suggest that long-tailed bats remain relatively sensitive to human activities. We would argue that a complacent approach to urban development could result in further bat distribution and habitat use restrictions, particularly if subdivision and roading infrastructure development increases within the peri- urban and residential rural landscape south of Hamilton. Instead we urge that a ‘green infrastructure’ approach be incorporated in urban planning that focuses on reducing ‘barrier effects’ by making the urban landscape more permeable to bat movement. This might involve the following actions:

 Bridge and major arterial roading locations should be carefully considered and reviewed, following further research on bat disturbance, to prevent barriers to bat movement and habitat use;  Low impact residential lighting regimes (e.g. installing fewer more directional lights);  Low impact road and bridge lighting regimes;  Retain and enhance residential and roadside tree cover; and  Enhance vegetation connections between riparian/gully habitats and urban ‘green spaces’ (e.g. parklands). Major growth cells for Hamilton city are currently focused on the southern urban-rural interface with several major roading and housing developments proposed for the short to mid-term future (e.g. Floyd & Dekrout, 2009). This study underscores the importance of making the urban landscape (both present and future) more permeable to long-tailed bats as well as protecting and enhancing existing, well- connected bat habitats. Major collective conservation decision-making by key stakeholders, as well as the implementation of multiple adaptive management strategies, is likely needed if long-tailed bats are to be retained in Hamilton City in the face of urban expansion.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 16

5 References

Borkin, K. & Parsons, S. 2011. Home range and habitat selection by a threatened bat in exotic plantation forest. Forest Ecology and Management, 262, 845-852

Clarkson, B. D. & McQueen, J. C. 2004. Ecological restoration in Hamilton City, North Island, New Zealand. In: 16th International Conference Society for Ecological Restoration, pp. 1-6. Victoria, Canada.

Dekrout, A. 2009. Monitoring New Zealand long-tailed bats (Chalinolobus tuberculatus) in urban habitats: ecology, physiology and genetics. Unpublished PhD thesis, University of Auckland.

Fenton, M. B. 2003. Science and conservation of bats: Where to next? Wildlife Society Bulletin, 31, 6- 15.

Floyd, C. & Dekrout, A. 2009. Hamilton City Council – Proposed Peacocke Development: Review of submissions to proposed variation No. 14 – Peacocke Structure Plan. Kessels & Associates, Ltd.

Fukui, D., Murakami, M, Nakano, S. & Aoi, T. 2006. Effect of emergent aquatic insects on bat foraging in a riparian forest. Journal of Animal Ecology, 75 (6), 1252-1258

Kerth, G. & Melber, M. 2009. Species-specific barrier effects of a motorway on the habitat use of two threatened forest-living bat species. Biological Conservation, 142, 270-279.

Le Roux, D.S. 2010. Monitoring long-tailed bat (Chalinolobus tuberculatus) activity and investigating the effect of aircraft noise on bat behaviour in a modified ecosystem. Unpublished MSc thesis, Waikato University.

Lloyd, B. 2009. Bat call identification manual for use with digital bat recorders and BatSearch Software. (Ed. by The Department of Conservation). Wellington, New Zealand.

O'Donnell, C.F.J. 2000a. Conservation status and causes of decline in the threatened New Zealand Long-tailed bat Chalinolobus tuberculatus (Chiroptera: Vespertilionidae). Mammal Review 30, 89-106.

O'Donnell, C. F. J. 2001a. Home range and use of space by Chalinolobus tuberculatus, a temperate rainforest bat from New Zealand. Journal of Zoology, 253, 253-264.

O'Donnell, C. F. J. 2001b. Advances in New Zealand mammalogy 1990-2000: Long-tailed bat. Journal of the Royal Society of New Zealand, 31, 43-57.

O'Donnell, C. F. J. 2005. New Zealand long-tailed bat. In King CM ed: The handbook of New Zealand mammals. 2nd edition. Oxford University Press. Pp 98-109.

O’ Donnell, C.F.J., Christie, J.E., Hitchmough, R.A., Lloyd, B. 7 Parsons, S. 2010. The conservation status of New Zealand bats, 2009. New Zealand Journal of Zoology, 37 (4), 297-311

O' Shea, T. J., Bogan, M. A. & Ellison, L. E. 2003. Monitoring trends in bat populations of the United States and territories: status of the science and recommendations for the future. Wildlife Society Bulletin, 31, 16-29.

Parsons, S., Thorpe, C. W. & Dawson, S. M. 1997. Echolocation calls of the long-tailed bat: A quantitative analysis of types of calls. Journal of Mammology, 78, 964-976.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 17

Pryde, M.A., O’Donnell, C.F.J., Barker, R.J. 2005. Factors influencing survival and long-term population viability of New Zealand long-tailed bats (Chalinolobus tuberculatus): Implications for conservation. Biological Conservation, 126, 175-185.

Safi, K., Köning, B. & Kerth, G. 2007. Sex differences in population genetics, home range size and habitat use of the parti-colored bat (Vespertilio murinus, Linnaeus 1758) in Switzerland and their consequences for conservation. Biological Conservation, 137, 28-36.

Schaub, A., Ostwald, J. & Siemers, B. M. 2008. Foraging bats avoid noise. Journal of Experimental Biology, 211, 3174-3180.

Sedgeley, J. A. 2001. Quality of cavity microclimate as a factor influencing selection of maternity roosts by a tree-dwelling bat, Chalinolobus tuberculatus, in New Zealand. Journal of Applied Ecology, 38, 425-438.

Stone, E. L., Jones, G. & Harris, S. 2009. Street lighting disturbs commuting bats. Current Biology, 19, 1123-1127.

Threfall, C.G., Law, B. & Banks, P.B. 2012. Sensitivity of insectivorous bats to urbanization: Implications for suburban conservation planning. Biological Conservation, 146, 41-52

Verboom, B., Boonman, A.M. & Limpens, H.J.G.A. 1999. Acoustic perception of landscape elements by the pond bat (Myotis dasycneme). Journal of Zoology, 248 (1), 59-66

6 Acknowledgments

We especially wish to thank the Kessels & Associates team – notably, Derek Christie, Hannah Mueller, Ryan Clark, Britta Deichmann, Mahuru Robb and Gerry Kessels for field assistance, equipment provision, review and statistical analyses. We thank Gerard Kelly of the Waikato Tree Trust for assistance with funding applications and in-kind logistical services. We also thank The Department of Conservation, Waikato Regional Council and the Hamilton City Council for generously lending automated bat detectors and other essential resources. For review of the report and of the survey methods we thank Associate Professor Stuart Parsons of Auckland University.

This research was funded by the Hamilton City Council (Enviro Fund, Waikato Tree Trust), Waikato Regional Council, Trust Waikato, Waikato University (CBER) and Kessels & Associates, Ltd.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 18

7 Appendix I

Table 3 Summary table detailing key characteristics of all the 62 ‘green spaces’ surveyed.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 19

Table 3 (continued) Summary table detailing key characteristics of all the 62 ‘green spaces’ surveyed.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 20

Figure 6 Long-tailed bat distribution map. Surveyed areas in blue represent habitats where bats were not detected while habitats in red represent areas with confirmed bat activity.

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 21

Project Echo and Kessels & Associates Ltd May 2012 HAMILTON CITY-WIDE BAT SURVEY 2011/2012 22

Figure 7 Individual nightly activity patterns for all 16 sites with confirmed bat activity.

Project Echo and Kessels & Associates Ltd May 2012 Key Ecological Sites of Hamilton City Volume I

CBER Contract Report 121

Client report prepared for

Hamilton City Council

Toni S. Cornes, Rachel E. Thomson, Bruce D. Clarkson

Centre for Biodiversity and Ecology Research

Department of Biological Sciences

Faculty of Science and Engineering

The University of Waikato

Private Bag 3105

Hamilton, New Zealand

May 31st 2012

Email: [email protected]

Contents

Executive Summary ...... 1

Report Context and Overview...... 2 Overview ...... 2 Hamilton City Boundaries ...... 3 Ecology of Hamilton ...... 4 Climate ...... 4 Geology ...... 4 Landforms and Vegetation Types ...... 4 Fauna of Hamilton City ...... 5

Methodology ...... 8 Identification of Potential Sites ...... 8 Field Survey ...... 8 General Information ...... 8 Description of the Site ...... 9 Habitat and Vegetation Description ...... 9 Flora and Fauna ...... 10 Threats ...... 10 Human Associated Activities ...... 10 Climatic Conditions during Survey...... 10 Context/Nearby Site Information ...... 10 Management Recommendations ...... 10

Results ...... 12 City-wide Extent of Key Sites ...... 12 Extent of the Key Sites by Land Unit ...... 12 Analysis of Representativeness ...... 13 Analysis of Ecological Rankings ...... 14 Spatial Distribution of Key Sites ...... 15

Discussion ...... 17 Overview ...... 17 Spatial Distribution of Key sites ...... 17 Gully Systems ...... 17

iii

Peat lakes ...... 19 Riverbanks ...... 20 Reduced Key sites ...... 20 Nawton Wetland ...... 20 Whyte St kahikatea ...... 20 Pukete Wetland ...... 21 Results in a Regional Context ...... 21 Threats to the Key Sites ...... 22 Invasive Weeds ...... 22 Vertebrate Pests ...... 22 Surrounding Land Uses...... 22 Habitat Sustainability ...... 23 Threat Mitigation ...... 23 Weed Control ...... 23 Connectivity and Buffering ...... 23 Interactions with Landowners ...... 24 Long Term Strategy ...... 24 Further Recommendations ...... 26 Monitoring ...... 27

Conclusion ...... 27

Acknowledgements ...... 28

References...... 28 Appendix one: SNA Review – Site Visit Assessment Form ...... 32 Appendix two: Key Ecological Sites in Hamilton City …….……………………………………………………………..37 Appendix three: Ecological significance justification for key sites ...... 42 Appendix four: Species List ...... 47

Reviewed by: Approved for Release by:

Kembley Pudney Daniel Laughlin

Hamilton City Council University of Waikato

Executive Summary Ecological sites of significance previously identified in 2000 were reviewed in 2011. Natural vegetation in areas acquired by the city since 2000 was also surveyed to identify any new key sites. In total seventy key sites that met the Waikato Regional Council Regional Policy Statement criteria for ecological significance were identified across Hamilton City.

Of the original key sites, the total area covered by sites, average site size and overall quality of sites had increased between the 2000 and 2011 surveys. This was due to restoration efforts across the city by Hamilton City Council and the community. Vegetation restoration efforts have had other biodiversity and ecological benefits such as providing additional habitat for the city’s increasing tui population.

Key sites are not spread evenly across the city or across landform types. Most key sites are either in gullies or adjoining the Waikato River. Less than 1% of urban alluvial plains and peat bogs are key sites. Two sites on private land have degraded and no longer meet the ecological significance criteria in 2011.

The current survey utilised a standard methodology focused on vegetation types. There will be other significant sites not identified including sites with significant fauna values but a detailed and costly survey would be required to identify all such sites.

The 1.5% of the city area covered by key sites is well below the 10% minimum recommended to prevent biodiversity decline in urban areas. Areas where vegetation restoration has begun in the city have the potential to expand existing key sites or develop new sites if council and community efforts continue in the future. The Council and its restoration partners should continue to seek ways of increasing native vegetation cover in Hamilton City and restoration of the distinctive gully landform remains the best option.

Report Context and Overview

Overview In December 2010 the University of Waikato’s Centre for Biodiversity and Ecology Research was contracted by Hamilton City Council to review the significant natural areas (henceforth known as “key sites”) previously identified within the city boundaries. Any additional key sites in Hamilton City as a result of remediation work or extension of Hamilton City boundaries were also to be included. As with past reports, the significance of a site was based largely on the ecological significance of vegetation and there was no systematic attempt to identify significant fauna habitat or cultural values of a site.

The identification of Key Sites fulfils part of Hamilton City’s obligation under section 6 of the Resource Management Act 1991 (RMA) to “recognise and provide for the following matters of national importance: The protection of areas of significant indigenous vegetation and significant habitats of indigenous fauna”. The Waikato Regional Policy Statement (RPS) outlines a strategy for implementing the RMA in the Waikato Region, including the need to maintain and enhance indigenous biodiversity. The city could meet its obligations with district plans taking indigenous biodiversity into account when developing a local biodiversity strategy; creating reserves along the banks of lakes and water ways to provide linkages and enhance biodiversity; re-creating and restoring natural habitats in the district and managing activities e.g. subdivisions. Local biodiversity plans need to recognise indigenous biodiversity in their districts by identifying opportunities and priorities for re-creating, restoring and linking habitats. Key sites in the district need to be mapped, given a biodiversity value and have any protection or enhancement needs identified. To map, prioritize and develop strategies to enhance Hamilton City’s indigenous biodiversity, key sites inside the city boundary have been identified previously within three reports across four years. Those reports are described below.

In the Downs et al. (2000) report information was provided on the ecological character of Hamilton City, focusing on terrestrial ecosystems. The result of the corresponding survey identified 67 key sites within the Hamilton City boundary. A description of each of these sites was given with information on ecological characteristics, condition and spatial distribution in the contexts of both citywide and regional scale. Recommendations on management strategies and policy responses were also given.

Due to development in the north of Hamilton City, Stevens et al. (2002) produced a report on the vegetation of the Te Awa o Katapaki Gully and the adjoining river terrace. This report identified sites and features of ecological and environmental value in the study area, including the key sites identified by Downs et al. (2000).

Due to its proposed inclusion within Hamilton City the village of Temple View was surveyed by McQueen (2004) for areas of ecological significance, following the method of Downs et al. (2000). This resulted in one key site being identified.

The review of Hamilton’s District Plan triggers the need for a review of key sites to document any changes to area, quantity, vegetation type, biodiversity and quality of the sites of Hamilton City.

Hamilton City Boundaries Since 2000 the boundaries of Hamilton City have been extended to include new areas in the west, north and east of the city (Figure 1). Temple View village was transferred to Hamilton City from the Waipa district in July 2004. This area is bounded by Tuhikaramea Road and Wallace Road, extends around the campus of the now defunct Church College of New Zealand and up to Koromatua Road, enclosing the village of Temple View and some farmland. In 2011 two areas were transferred over from the Waikato District to Hamilton City (Hamilton City Council & Waikato District Council 2005). These areas were Ruakura in the south east and Te Rapa North in the north. Ruakura comprises of 730 hectares of land that includes Innovation Park and surrounding farmland. Ruakura is bounded by Greenhill Road in the north, the Mangaonua Gully on the west and the planned Waikato Expressway on the east. Te Rapa North is 240 hectares bounded by the Waikato River, Horotiu, and the planned Te Rapa bypass. Adding the 2011 boundary extensions to the 2004 city area increases Hamilton City from 9860 hectares to 11080 hectares (Hamilton City Council 2011).

Figure 1: Hamilton City boundary with post 2003 additions

Ecology of Hamilton

Climate Due to the sheltered inland location of the Waikato basin seasons consist of mild winters, warm, humid summers and frequent fog (Environment Waikato 2010). At the Hamilton Airport monitoring station in 2009 the average temperature was 13.2 ˚C, with 1088 mm of rainfall, and 2120 sunshine hours (National Institute of Water and Atmospheric Research 2010).

Although Hamilton City does not usually suffer extreme climatic events, in the summer/autumn of 2008 the whole Waikato Region suffered a severe drought, with rainfall the lowest it had been in approximately 100 years. This resulted in increased plant mortality across the city. Losses were seen in key ecological sites such as Claudelands Bush and Horseshoe Lake (Cornes et al. 2008; Cornes & Clarkson 2010).

Geology The bedrock of the Hamilton basin is comprised of greywacke basement rock that was eroded to a plain about 100 million years ago (mya). Peat swamps began to form as the surface warped and created depressions about 50 mya. By 30 mya the area that is now the Hamilton basin was submerged under the advancing sea, depositing sandstones and limestones on the seafloor. Differential uplift created the basins and uplands that now form the Waikato Region. Volcanic activity distributed ignimbrite and volcanic material across the Hamilton Basin, which was then shaped by the action of the Waikato River and its associated streams to create the hilly landscape that characterises the current Hamilton Basin (McCraw in Clarkson et al. 2002).

Landforms and Vegetation Types Hamilton City is comprised of four main landform units: hills, alluvial plains, gullies, and peatlands. In the past most of these areas were dominated by indigenous forest.

Low rolling hills and the foothills (9.5%) of ranges at the edge of Hamilton are generally comprised of late Quaternary parent material that used to be dominated by rimu-tawa forest and kauri-hard beech forest. On the footslopes of the low rolling hills the parent material is represented by poorly drained colluvium from Hamilton Ash and other deposits. The main vegetation type supported by this landform was pukatea-kahikatea forest (Clarkson et al. 2007).

Low mounds or ridges of alluvial plains (58.6%) are characterised by moderately to well-drained alluvium from the Hinuera formation which predominantly supported mixed conifer-broadleaf forest. In shallow depressions or swales the alluvium has more silt and clay, and hence drains less readily. This created the boggy areas that were dominated by kahikatea semi-swamp forests. In lower terraces (2.4%) beside the Waikato River (3.1%) the alluvium has more sand and gravel and is better drained. This well drained area suited totara-matai-kowhai forest (Clarkson et al. 2007).

Gullies (7.0%) were formed about 15,000 years ago through a process called spring sapping. As the Waikato River cut down creating steep banks, aquifers were exposed. These eroded steep-sided troughs back from the river bank, which eventually became Hamilton’s gully system. Gullies have two main land units: the steep gully sides, and the gully floor. On the sides, soil material is well- drained, generally from the Hinuera formation and supported totara-matai-kowhai forest. The

4 gully floors are more poorly drained and were dominated by kahikatea-pukatea-swamp maire forest (Clarkson et al. 2007).

Peatlands include peat lakes (0.5%) and peat bogs (19.0%), all of which are generally very wet and poorly drained areas. These areas hosted a range of vegetation types including submerged vegetation, swamp forest, sedgelands, shrublands and restiad bogs (Clarkson et al. 2007).

Fauna of Hamilton City

Gully Systems Sprawling over approximately 770 hectares (7.0%) of Hamilton City is a series of gully systems. They include the four major gully systems of Kirikiriroa, Mangakotukutuku, Mangaonua and Waitawhiriwhiri, and numerous minor systems.

Collier et al. (2009) surveyed and assessed the value of these streams and gully systems in Hamilton’s Urban environment. Indigenous species found included the shortfin eel, longfin eel, banded kōkopu, giant kōkopu, inanga, common smelt, common bully, and torrentfish. Exotic species included koi carp, gambusia, catfish, and indeterminate trout. Many of the indigenous species found are rare or in decline and their presence in these urban environments lends great value to these gully systems.

Smith (2007) surveyed three gully streams within Hamilton City for mayflies, stoneflies and caddisflies. Twenty six species were found. Due to the sampling method this figure is likely to under-represent actual species diversity. Mangakotukutuku Gully had the highest diversity with 13 species found. Kirikiriroa Gully and Waitawhiriwhiri Gully had eleven and six respectively. In the Mangaiti section of Kirikiriroa Gully a new species of Oxyethira was discovered. It was suggested that the high species diversity for an urban area was due to the large number of vegetated gullies with riparian cover, which increases habitat complexity.

New Zealand’s Long-Tailed Bat is an endemic mammal that has persisted in Hamilton City, despite its disappearance from other cities in New Zealand. These bats are seriously threatened by habitat loss and rat predation. Echo-location detection has allowed surviving pockets of bats to be identified, particularly in Mangaonua Gully, Mangakotukutuku Gully, Hammond Bush and surrounding areas (Figure 2). Bats play an important role in ecosystems as aerial insectivores (Le Roux 2010). Echo-location studies have shown that populations of bats roosting outside the city use the gully systems of Hamilton to migrate to the Waikato River. This is an indicator of the importance of these gully systems for wildlife corridors in Hamilton City (Parsons & Dekrout in Collier et al. 2010; LeRoux & LeRoux, 2012).

Figure 2: Long-tailed bat distribution for Hamilton City: Confirmed bat activity with numbers while confirmed sightings are unnumbered (Waikato Regional Council (2010); LeRoux & LeRoux, (2012))

Waikato River New Zealand’s longest river is a key feature of Hamilton City. This wide single-path river cuts Hamilton City in two with its deep channel and provides an ecological corridor for the movement of both indigenous and exotic wildlife.

A survey by Hicks et al. (2005) of four locations on the Waikato River just south of Hamilton City showed the presence of brown trout, common bully, goldfish, inanga, koi carp, grey mullet, rudd, shortfin eel, and smelt.

Peat Lakes Hamilton is home to a number of peat lakes including Lake Rotoroa, Horseshoe Lake (within Waiwhakareke Natural Heritage Park), and Lake Rotokaeo. Lake Rotoroa is near the city centre and has a surface area of approximately 55 ha. It is relatively young, forming 16,000 years ago when the Waikato River changed its path. Horseshoe Lake (Lake Waiwhakareke) is a small peat lake located near the corner of Brymer and Baverstock Roads. It is the focus of a restoration project that is intended to span the next few centuries with the aim of improving the lake’s water quality and recreating the historic vegetation types that were once present in this area. In the future this lake could become suitable for the introduction of native mudfish, a species that is currently threatened in New Zealand waterways (Parks and Gardens Unit, Hamilton City Council 2010). Lake Rotokaeo (Forest Lake) is a shallow peat lake located in Minogue Park off Forest Lake Road. Eleven fish species have been found in these peat lakes, including natives and exotics (Table 1).

Table 1: Fish recorded in Lake Rotoroa, Horseshoe Lake, and Lake Rotokaeo Common Name Lake Rotoroa (Hicks Horseshoe Lake(Hicks, Lake Rotokaeo (Hicks, 2003) Osborne et al. 2005) Brijs et al. 2009) Brown bullhead catfish    Common bully   Gambusia    Goldfish   Longfin eel   Perch  Rudd   Shortfin eel    Tench  Brown Trout   Giant Kokopu   Frog tadpoles  

Birdlife The Ornithological Society of New Zealand regularly conducts a census of the bird species seen in different areas around the country. Monthly censuses are conducted at Hamilton Lake and Horseshoe Lake. Between January 2010 and January 2011 both land bird species and waterfowl were identified. Land bird species included Australasian harrier, blackbird, chaffinch, fantail, feral rock pigeon, goldfinch, grey warbler, magpie, mynah, silvereye, skylark, song thrush, sparrow, spur-winged plover, starling, welcome swallow and white doves. Water fowl included Australian coot, black shag, black swan, black teal, Canada goose, domestic duck, domestic white goose, farmyard duck, grey duck, grey teal, hybrid duck, kingfisher, little black shag, little shag, mallard duck, Muscovy duck, New Zealand dabchick, paradise shelduck, pied shag, pied stilt, pukeko, and white faced heron (Ornithological Society of New Zealand 2010a, 2010b, 2010c, 2011).

Other species known to be found in Hamilton include the Australasian shoveller, shining cuckoo, yellowhammer, white headed stilt, little pied cormorant, Caspian tern and notably, tui and bellbird (Fitzgerald & Innes 2009).

Local council, regional council and research institutions have combined efforts to increase native bird numbers within the city. Before 2007 tui had been regularly sighted in the city but none were known to permanently lived or bred there. Hamilton Halo was started in 2007 with the aim of re- establishing a population of tui within Hamilton City. It was intended that by increasing tui numbers outside the city using focused pest control in forest close to Hamilton visits of tui to Hamilton would increase leading to birds again nesting in Hamilton and becoming permanent residents. This operation has been successful in reducing predator numbers in selected forests and increasing tui with the city. Fifty bellbirds were reintroduced to Hamilton city in May 2010. In 2011 a bellbird fledgling was sighted, suggesting bellbirds may be successfully breeding in Hamilton (Waikato Regional Council, 2011).

Methodology Identification of Potential Sites To identify ecologically significant sites within Hamilton City boundaries satellite images were obtained and the updated boundaries overlaid using a Geographic Information System. Within these boundaries, potential sites were identified by vegetation cover. These sites along with those identified in previous reports were visited and assessed using a set of significance criteria as outlined in the Waikato Regional Policy Statement (2007) (Table 2).

Table 2: Ecological significance criteria (Sourced from http://www.ew.govt.nz/PageFiles/6777/rpsdecember07.pdf, pages 216- 217). Examples adapted for a Hamilton City context. Criteria Example Criteria 1 - Protected or Preserved Queen Elizabeth II National Trust Criteria 2 - Recommended for Protection Identified by DoC (1993) as being worthy of protection Criteria 3 - Threatened or Endemic Species Bat feeding site Habitat Criteria 4 - Under Represented A patch of wetland which is under-represented (or rare) in the Hamilton Ecological District Criteria 5 - Uncommon Before Settlement River islands Criteria 6 - Indigenous Wetland Habitat Contains (or is likely to contain) a natural wetland. Criteria 7 - Large Indigenous Habitat c. 3 ha podocarp forest remnant in Hamilton City Criteria 8 - Critical Aquatic Habitat Wetland with potential mudfish habitat Criteria 9 - Healthy Indigenous Vegetation Representative remnants of moderately dense podocarp forest Criteria 10 - Rare or Exceptional Representation Nationally rare Sporadanthus-Empodisma bog habitat Criteria 11 - Ecological Buffer Linkage or Corridor Indigenous forest connected with a gully system

Field Assessment

Once dominance of native vegetation or abundance of native plants of the site was established, a detailed assessment was conducted using the significant natural area (SNA) form given in Appendix one. The key points of the assessment are explained in the following paragraphs.

General Information

General information for identified sites included tenure, protection status, fencing, and matrix land-use. Each of these was given a number ranking between 1 and 4, 5, or 9 that corresponded to a particular criterion. For example, fencing can be ranked: (1) secure, intact fencing around the entire perimeter; (2) mostly fenced, areas where stock access is likely; (3) some fencing, one side, or large breaks; (4) no fencing. Owner details were also recorded including name, address, contact numbers, and email if available.

Description of the Site

These included but were not limited to: aspect, slope, condition (canopy, understorey, leaf-litter), unusual (or common) characteristics, surrounding land use, dominant vegetation communities, uncommon species, history of site/landowner’s comments.

Habitat and Vegetation Description

Habitat and vegetation descriptions were taken for each of the individual land units within each site. These descriptions included the unit number, hydrological regime, category, code, character, area, and vegetation description. Criteria for the classification of hydrological regimes are given in Table 2. Category and code were used to describe the vegetation of the unit. Each unit was classified into category A, B, C, or D, which correlate to wooded/treefern habitats (e.g. podocarp forest), grass/herb/moss habitats (e.g. sedgeland), bare habitats (e.g. rocky coast), and other habitats (e.g. roads/railways) respectively. Character refers to whether the unit is indigenous or exotic based on an estimate of the abundance of indigenous and exotic species. A proportion of greater than 50% qualifies the unit for classification into either one of these categories. The area of each unit was estimated and given a corresponding code (Table 3). Vegetation descriptions were based on Atkinson (1985) to give an indication of abundance, and presence or absence of tiers.

Table 3: Hydrological regime classification criteria Code Character Explanation 1 Terrestrial All dry areas of land not covered by a wetland hydroclass (see below) 2 Estuarine Coastal waters semi-enclosed by land and partially diluted with fresh water 3 Riverine Flowing waters contained within a channel: rivers, streams, and their margins. 4 Lacustrine Lakes or dammed rivers with open water 5 Palustrine All other non-tidal wetlands, small open water bodies, and vegetated wet ground.

Table 4: Area estimation codes Code Area Estimation Q < 5 x 5 m R 5 x 5 m S 10 x 10 m (100 m2) T 20 x 50 m (0.1 ha) U >0.1 < 1 ha V >1 < 5 ha W > 5 < 10 ha X > 10 < 25 ha Y > 25 < 100 ha Z > 100 ha

Flora and Fauna

Lists were made of the flora and fauna at the site, identified either by observations during the survey or from landowner’s previous observations. Special note was made of rare, threatened, or distinctive species at each site. Vegetation was described in more detail, with particular reference to the condition of each tier.

Threats

This referred particularly to pest plants (e.g. vine weeds) and animals (e.g. stock) which are, or have the potential to become, significant threats in the area. For each unit a rating of one to four was given for the abundance or cover of ground cover weeds, vine weeds, and shrub or tree weeds. A rating of one indicates a very common weed with over 50% ground cover. A rating of four indicates none present. Dominant species were noted and comments or suggestions for potential management were given. The main animal pests concerned in this survey were stock. Again, a one to four rating was given for each unit, describing the abundance and frequency of stock presence in the area. Management suggestions were also given.

Human Associated Activities

Evidence of human associated activities was recorded and given a rating for impact of the activity, and for the attitude of the involved parties toward remediation. Activity examples include: rubbish dumping, stock grazing, drainage, earth works, erosion, top dressing, fire, vegetation clearing, herbicide application, harvest/vegetation clearing, planting, animal pest control, domestic pets, and fencing. Recommendations for mitigation were given.

Climatic Conditions during Survey

An indication of humidity, cloud cover, wind and temperature was recorded for the time of the site survey in each location.

Context/Nearby Site Information

If applicable, scrub, forest, or wetland areas that were close to each surveyed site were recorded with information about dominant vegetation types, size and the condition.

Management Recommendations

Management recommendations follow Downs et al. (2000) but are limited because the survey is predominantly focussed on vegetation. Recommendations follow a canopy manipulation method with inter-planting of indigenous species where necessary. This method is recommended with consideration of the cost of weed removal, the benefits of canopy protection to native plants, and bank stability from plant retention on steep sites. Recommendations to increase native animal populations have not been given.

On the SNA sheets under Management Recommendations the management technique code (see below) is given followed by the species that needs removal and possible indigenous species to plant. For example a grey willow dominant forest would have M1: Grey willow. Planting: mahoe, pate, kahikatea.

Forests and Scrub Management action one (M1): Exotic dominant canopy Retain canopy dominant exotics until indigenous species in the understorey have developed enough to form a replacement canopy. When this occurs (and if safe to do so) canopy exotics can be poisoned and left to die where they stand. This means the structure of the canopy is still providing protection for the layer beneath. If the understorey is sparse or indigenous species are not dominant remove understorey exotics and plant natives.

Management action two (M2): Native dominant canopy Exotics in the canopy can be left to naturally senesce, while exotics under the canopy need removal. If exotics form dense patches in the canopy follow M1 method.

Management action three (M3): Exotics in the understorey Remove exotic weeds from the understorey, targeting species which could develop into canopy trees. If understorey has few indigenous species, undertake planting.

Management action four (M4): Exotic groundcover Remove invasive exotic groundcover and invasive exotic seedlings.

Shrubland

Management action five (M5): Exotic species As no understorey exists in a shrubland exotic species should be removed. If this forms large gaps in the canopy, indigenous species should be planted within gaps.

Open Wetlands

Management action six (M6): Exotic trees Invasive exotic trees need to be removed or poisoned and left to decay.

Management action seven (M7): Other exotics Remove invasive exotics from the wetland and plant indigenous species in any large gaps.

Exotic Vines over all vegetation types

Management action eight (M8): Exotic vines Where possible remove the exotic vines or stop their spread by cutting and pasting stems in mature canopy or if vines form self-supporting tangles or are growing with other exotics, spray and mechanical remove.

Ecological Rankings Using all information gathered from the survey an ecological ranking was determined for each site. Rankings followed the scale and criteria set out in Downs et al (2000), which comprised of a three point scale. Key sites were ranked 1, 2 and 3 for sites of very high, high and moderate ecological value.

Results City-wide Extent of Key Sites

Of the 67 Key Ecological Sites identified in Downs et al. (2000) 65 still fit the criteria. Two new key sites were located in the current survey within the 2000 city boundary. Three new key sites were found in the 2011 boundary extensions of Temple View and Te Rapa North. No sites were located in the 2011 Ruakura extension, although two existing key sites within Ruakura had their size increased due to the boundary change. SNA site sheets for each key site are given in Volume II. In Volume II some sites are grouped together due to site connectivity. A summary of all key sites is given in Appendix two and the justification for their inclusion is provided in Appendix three. A list of species noted in key sites is given in Appendix four.

The total area covered by key sites within the city has increased since 2000 due to new sites being added and area increases of previous sites resulting from restoration work (Table 5). The average size of key sites increased by 0.4ha (peat lakes excluded). In 2011 approximately 1.5% of the city is covered by key sites. When the 55.9ha of peat lakes within the sites is excluded this is adjusted to 1.0%. This gives a 0.2% increase in area of the city covered by key sites between 2000 and 2011.

Two key sites were removed from the list (totalling 1 ha) and five were added (totalling 7 ha). Only one site had a decrease in total area while 27 sites had an increase.

Table 5: City-wide extent of key sites 2000 2011 Total Area All Sites (ha) 126.9 163.8 Total Area All Sites (excl peat lakes) (ha) 71.0 107.9 Average Site Area (ha) 1.9 2.3 Average Site Area (excl peat lakes) (ha) 1.1 1.5 Total Area of Hamilton City (ha) 9427 11080 % City Area occupied by Key Sites (ha) 1.3 1.5 % City Area occupied by Key Sites (excl peat lakes) (ha) 0.8 1.0

Extent of the Key Sites by Land Unit Seventy key sites were located in Hamilton City. The majority of key sites were in gully systems (47%), which equates to 27% of the total area of key sites (Table 6). Peat lakes made up the largest area (34%) of key sites. River islands made up the lowest number of sites and smallest land area (Figure 2). As with the last survey, no river sites were included. Some key sites such as Waiwhakareke Natural Heritage Park include more than one landform.

Table 6: Frequency and Extent of the Key Sites by Land Unit Land Unit % No. of Sites Area of Sites (ha) % Total Site Area Alluvial Plain 14 19.9 12 Gully 47 44.7 27 Hillslope 7 10.9 7 Peat bog 3 7.3 5 Peat Lake 2 55.9 34 River Island 3 0.3 <1 Riverbank 24 24.8 15 TOTAL 100 163.8 100

River Island 0.2% Riverbank Alluvial Plain 15.1% 12.1%

Gully 27.3% Peat Lake 34.1%

Peatbog Hillslope 6.7% 4.5%

Figure 3: Extent of the Key Site Area by Land Unit Analysis of Representativeness

Alluvial plains are the dominant landform of Hamilton City (approximately 59%). Despite this only 0.30% of the landform was occupied by key sites (Table 7, Figure 3). This was the lowest coverage for a landform with key sites present. The two smallest landforms in Hamilton City, peat lakes and river islands, had 100% cover of key sites, while all others had below 10% cover.

Table 7: Representation of Key Sites by Land Unit Land Unit Area of Key Sites (ha) % City Area Est % City Area % Land Unit occupied by Key occupied by Land occupied by Key Sites Unit Sites Alluvial Plain 19.9 0.18 59.7 0.30 Gully 44.7 0.40 7.1 5.79 Hillslope 10.9 0.10 9.7 1.04 Peat bog 7.3 0.07 19.3 0.35 Peat Lake 55.9 0.50 0.5 100 River Island 0.3 <0.01 <0.1 100 Riverbank 24.8 0.22 2.4 9.35 River 0 0 3.1 0 TOTAL 163.8 1.48 100

60 A 0.6 B

50 0.5

40 0.4

30 0.3

20 0.2 Percentageofarea city Percentageofarea city 10 0.1

0 0.0

Figure 4: Representativeness of land units within A: the entire city and B: Key sites of the city

Analysis of Ecological Rankings

An increase in area across all ecological rankings has occurred from 2000 to 2011 (Table 8). The majority of sites are ranked 2 (Figure 4). Sites with an ecological ranking of 3 cover the least area. Fewer sites have an ecological ranking of 1 than in 2000. Although the total number of sites ranked 1 is the lowest of all three they make up 31% of the area covered by key sites. Sites with an ecological ranking of 2 also cover the largest area (Figure 5).

Table 8: Frequency and Area of Key Sites by Ecological Ranking

Ecological Number Area Area (excl. peat lakes) (ha) Ranking 2000 2011 2000 2011 2000 2011 1 10 8 22.6 36.1 20.2 31 2 29 36 85.7 106.4 32.2 55.6 3 28 26 18.6 21.33 18.6 21.33 TOTAL 67 70 126.9 163.83 71.0 107.93

11% 1 2 3 37%

52%

Figure 5: Frequency of Key Sites with each Ecological Ranking 2011

1 20% 2 29% 3

51%

Figure 6: Area of Key Sites (excl. peat lakes) with each Ecological Ranking Spatial Distribution of Key Sites

The majority of key sites are located on the banks of the Waikato River or in the Mangaonua Gully. Although the key sites are spread across the city from north to south and east to west, large areas of the city have no key sites within them (Figure 6). Areas with the fewest key sites are north-east of the city in north Rototuna; the eastern areas of Fairview Downs and Ruakura; north-west of the city in Burbush and Avalon; western areas such as Dinsdale and Frankton; and south of the city in Deanwell, Melville and southern Peacockes. The western side of the city has the fewest key sites, but all the peat lakes are found on this side of the city.

Figure 7: Location of key sites within Hamilton City

Discussion Overview

Seventy key sites are found in Hamilton City. Five sites have been added since 2000 as a result of restoration efforts within the city and boundary changes. Twenty seven of the original sites had an increase in the area covered, with only one having a reduction. The original key sites have increased in area by a total of 49.3 ha. Losses to key site cover were due to property development, subdivisions and degradation from weed invasion. Increases in number and area of key sites can attributed to council, community groups and private residents’ efforts to restore natural areas of Hamilton City. This has been helped by initiatives such as the Gully Restoration Programme and Waiwhakareke Natural Heritage Park. Sites under council or residential ownership were more likely to be improved than sites under industrial ownership.

Only 1.5% of the city is covered by ecologically significant land. In time it is possible other areas being restored around the city will meet the significance criteria outlined in Table 2. This takes place when planted vegetation matures, native trees regenerate and other native species colonise the area, as seen at Seeley’s Gully in eastern Hamilton.

Spatial Distribution of Key sites

Gully Systems

Most Hamilton City gullies are undeveloped and provide a unique natural landscape feature to the city. These areas provide the greatest opportunity to expand natural areas within Hamilton City and connect adjoining landowners to the natural environment on their doorsteps. Along with the ecological benefits of gullies, these areas form part of the stormwater drainage network of Hamilton City. Vegetated areas of scrub or forest also provide these gullies with structure for bank stabilization. Five major and seven minor gullies within the city boundary contain key sites. Of these gullies, the majority have been partially or completely disconnected from the Waikato River. This is due to preparations for development such as earthworks, road construction and vegetation removal. A management plan including six gully systems was published in 2007 (Craig 2007), while Donny Park has its own management plan (Parks and Gardens Unit, Hamilton City Council 2004). These plans only cover publically owned land. When deciding how to manage key sites within a gully, the whole gully system should be taken into account due to the vegetation connectivity and buffering. The main gullies and their key sites are discussed below.

Te Awa O Katapaki

This is the city’s most northern gully system. Although it contains only one key site it is the largest site within a gully. In the 2000 survey this gully was located within farmland but since 2001 residential development has been occurring and will continue around the majority of the gully. Farmland now only borders the north eastern section. Development has led to both positive and negative modification of the gully and its stream.

Hicks et al. (2001) surveyed the gully stream and found that while it was unpolluted, the stream had very high nutrient concentrations. Short finned eels were the dominant fish, with mosquito

17 fish and common smelt also present. They also found the stream had a relatively diverse macro- invertebrate fauna. Despite short term increases in sediment flow from development, it was hypothesized that the water quality would improve with the removal of cattle access to the stream.

Stevens et al. (2002) surveyed the vegetation of Te Awa O Katapaki gully. It was noted that this gully is a good candidate for restoration due to the large size of native dominant vegetation present, low weed species diversity, easy access, connectivity to the river, stream water quality and habitat diversity. Since then some of the recommendations have been put in place including putting part of the gully in public ownership, re-vegetating areas of wetland with indigenous species and the installation of some paths.

The current survey found cattle have now been excluded from the key site. Plantings have been installed partially around the edge of the kanuka forest and along the stream within the subdivision.

Kirikiriroa

Kirikiriroa gully and its tributaries run from Pukete Bridge to Gordonton Road and north to Somerset Heights. In the past the catchment was within a matrix of residential and farmland; it is now residential and parkland. This survey found six key sites located in the gully system, one more site than in 2000. Most of the sites are found either at the mouth or in the central part of the gully. For naturally establishing vegetation grey willow and cabbage tree are dominant canopy species on gully floors, while mahoe and wheki are dominant on the gully sides within many of this gully’s key sites. The area covered by these sites increased from 5.1 ha in 2000 to 11.7 ha due to vegetation restoration efforts in the area. These efforts have resulted in the mouth of the gully being connected to the Waikato River and the adjoining riverbank key site by indigenous vegetation. Planting has led to an increase in native dominant wetland and native scrub habitat in the gully.

Mangakotukutuku

Located in the south west of Hamilton, this is the city’s largest gully. Unfortunately it has the lowest key site cover of all the major gullies. All sites are under 1 ha in size and are the only sites found in the south west of the city away from the river. Both stream and vegetation restoration efforts are taking place within the gully system to improve indigenous biodiversity. As plantings are young they cannot yet be considered ecologically significant. It is clear that when these plantings do develop, they will significantly increase indigenous species habitat along the stream and improve ecological function of both terrestrial and aquatic systems.

Mangaonua

Located predominantly at the south-eastern boundary of the city this gully system contains the highest proportion of indigenous vegetation, the highest number of key sites and covers the largest total area (excluding peat lakes). Key site vegetation is directly connected to Hammond Bush vegetation at the mouth of the gully. There are four main sections of key sites on this gully; two found at the mouth, four in the Riverlea suburb, five in the Berkley suburb and two in the

Silverdale suburb. As in Kirikiriroa Gully, the naturally established canopy vegetation on the gully floor is dominated mainly by willow species, while on the gully sides mahoe and treeferns are dominant. Unlike the other sections of the gully the Riverlea section (industrial zone) seems to lack restoration efforts. Restoration efforts have increased the size of key sites within the gully system by approximately two hectares.

Waitawhiriwhiri

Located on the western side of the city, this is a long, highly fragmented gully close to Hamilton’s CBD. Two key sites are located in the gully within two kilometres of each other. Restoration is taking place within the cabbage tree-land dominant section of the Edgecumbe site. This gully provides an excellent opportunity to increase the indigenous vegetation cover in an area of Hamilton city where very few key ecological sites exist.

Other gullies

Apart from Gibbons Creek Gully, which has two key sites, the rest of the gully key sites are located across different gully systems within the city. These sites include Ranfurly Gully, Bankwood Gully and three Pukete gully sites. The gullies with the highest modification within Hamilton are in the Pukete/St Andrews area as many of the gullies and tributaries were filled by residential development decades ago.

Peat lakes

The three peat lakes are all represented in the five largest key sites within Hamilton City. The majority of larger indigenous vegetation species around peat lakes in Hamilton City has been planted but there are many apparent survivors among the marginal plants and ground cover, e.g. around Rotokaeo (Forest Lake). All of these lakes are shallow and eutrophic, which has caused problems with water quality and weed invasions. The main management goals for all of Hamilton’s lakes are to improve lake water quality and enhance indigenous vegetation within the lake’s catchment to levels that are representative of pre human settlement; this objective is constrained by a recognition that Rotoroa (Hamilton Lake) has an important recreational and amenity function, and all the lakes exist within an urban, modified environment.

Waiwhakareke Natural Heritage Park (Horseshoe Lake)

A large restoration effort has taken place at this site increasing the indigenous vegetation from 3ha to 16ha in size between 2004 and 2011. This accounts for almost all the increases in key site area on peatland, alluvial plain and hillslope landforms. When completed the majority of the lake’s catchment will be in indigenous forest. The objective of the Waiwharareke Natural Heritage Park is to create a self-sustaining, pest-free habitat sanctuary that represents the original ecosystem diversity of the Hamilton Basin within the 60ha. All plantings at this key site are under ten years old, with the majority under five (Parks and Gardens Unit, Hamilton City Council, 2010b).

Lake Rotokaeo (Forest Lake)

Plantings around this lake have continued to improve the habitats and condition of the ecosystem. Since 2002 the invasive weed Mexican water lily has been controlled and is no longer dominant in

19 the lake. The management plan states that pest plants need to be monitored and in some cases controlled. There are future plans to increase plantings of indigenous wetland species both in and around the lake (Parks and Gardens Unit, Hamilton City Council, 2009).

Lake Rotoroa (Hamilton Lake)

Three key sites are found at Hamilton Lake. At 54 ha the lake and its surrounds are Hamilton’s largest key site. Native riparian plantings have been increased around the lake since 2000. The largest increase in indigenous vegetation has been at the south side of the lake with the other sides having limited plantings but still having weed control. Invasive species Yellow flag iris and Egeria densa have been removed in the past. Monitoring and eradication of these species and other possible problem species such as water lilies will continue at this site (Parks and Gardens Unit, Hamilton City Council, 2010).

The 2010 Hamilton Lake Domain management plan includes priorities for management of the riparian vegetation. Key aims are; retain vegetation for wildlife habitat; increase riparian planting with eco-sourced plants; increase lake bank stabilisation by planting the margins; investigate the possibility of small scale artificial wetlands for stormwater control. There will still be some exotic amenity plants installed around the lake but the majority of new plantings will be native.

Riverbanks

No new sites were located on the banks of the Waikato River. The area covered has been extended by restoration planting and management. These riverbank areas help to provide ecological connectivity along the river and into adjoining gullies. Indigenous vegetation now connects the river bank to the mouths of the Kirikiriroa and Mangonua gullies. Vegetation also provides bank stabilization and amenity value along the river.

Reduced Key sites

Nawton Wetland (Farnborough Drive Reserve)

As stated in the Downs et al. (2000) report, the Nawton wetland site went through drastic clearance for a residential development in 2000. This site and areas of surrounding properties are prone to winter flooding. Vegetation clearance has also meant some peat shrinkage, which is affecting buildings and other structures (e.g. cracks in driveways). This is the only key site on terrestrial peatland without an associated surface water body. Even though Horseshoe Lake has a larger area of peatland vegetated with natural vegetation, Nawton wetland has a variety of species not found at the younger, planted Horseshoe Lake site. The ecological ranking of this site had reduced due to the high degree of modification.

Whyte St kahikatea

This site was already one of the smallest sites in Hamilton City in 2000. Since then subdivision of land has further reduced it and some large kahikatea trees were removed because of a safety hazard posed to the new buildings. Garden exotics have been planted in the area as well.

Pukete Wetland

The loss of this area as a key site seems to be through a lack of management leading to exotic weeds outcompeting indigenous species in the area. Raupo and indigenous open wetland plants are now absent while the exotic Glyceria maxima is now the dominant. There were still some native fern species under a canopy of grey willow.

Results in a Regional Context

Some 268 ecological districts have been recognised in New Zealand, on the basis of their differing climate, landform, soil, geology and biological features (Myers et al. 1987). They are commonly used as a spatial framework on coarse filter for significance assessment (Walker, et al. 2008) Hamilton City, at 11080 ha, makes up 7.0% of the 159375 ha of Hamilton Ecological District (ED). Hamilton ED is confined to the Hamilton Basin with some of the surrounding hills and foothills included (McEwen, 1987). Leathwick et al. (1995) found less than 2% of natural vegetation that once existed in the ecological district pre 1840s still remains. Wetlands and conifer forests were the dominant ecosystems of Hamilton ED before human settlement (Harding, 1997). These two ecosystems also suffered the highest percentage reduction through anthropogenic activities. All past vegetation types are less than 2% cover within the district.

Large stands of kahikatea forest are important in a regional context due to the reduction in conifer forest of any substantial size across the ecological district (Downs et al. 2000). The largest area of kahikatea forest within Hamilton ED is Whewells Bush (11.5 ha). Within Hamilton City, the largest kahikatea stand is Claudelands Bush at approximately 6.5 ha with the inclusion of recent plantings on the western boundary. As recognised in the 2000 report, all kahikatea stands with an ecological score of 1 should be considered regionally significant due to stand size, age and management.

The other sites with the highest ecological score are also regionally significant. As a consequence of widespread drainage of wetlands across New Zealand, Sporadanthus ferrugineus is only found naturally in three sites, all within the Waikato Region. For that reason the Sporadanthus- Empodisma restiad bog reconstructed at Waiwhakareke Natural Heritage Park can be considered regionally significant, even though it is planted. The two other regionally ecologically significant key sites are Key Sites 16.4 and 16.7 along the Waikato River. Key site 16.7 (Hammond Bush) is dominated by pukatea/swamp maire forest, which has been almost completely removed from the Hamilton ED landscape. Both Waiwhakareke Natural Heritage Park and Hammond Bush are managed by planting of indigenous species and undertaking weed and pest control. The steep riverbank of the kamahi-mamaku forest of Key Site 16.4 (Riverbank Mamaku-kamahi forest, Hamilton Gardens) makes management difficult but this inaccessibility also helps with its preservation. This forest type is rare along the banks for the Waikato River and is the only example of this vegetation type within the city.

The severe loss of indigenous vegetation from both Hamilton City and Hamilton ED makes protection and enhancement of all indigenous vegetation essential within the city, especially if connectivity between indigenous remnants can be enhanced. As stated in Downs et al. (2000) and Harding (1997), very few opportunities are left in the district for indigenous ecosystem protection; therefore managing what still exists is extremely important.

Threats to the Key Sites

Threats to key sites are the same as set out by Downs et al. (2000). Updated information is provided below.

Invasive Weeds

Thirty four percent of dominant key site canopy species were exotic. Four of these are in the banned plants register (National Plant Pest Accord): grey willow, crack willow, tree privet and Japanese honeysuckle. Of the 162 exotic species recorded in key sites, 38 are classed as weeds. Under the Waikato Regional Pest Management Strategy 26 recorded species are classed as weeds. According to Howell (2008) 94 of the exotic species found are ecological weeds. One key site has already been lost to weed invasion and others are at risk if weeds remain uncontrolled. Totara Park is one site where weed control would be particularly valuable to improve the chances of kahikatea surviving competition from grey willow, as grey willow will out-compete the kahikatea and prevent regeneration (Coleman, 2010).

Vertebrate Pests

Morgan et al. (2009) surveyed 18 Hamilton gully, parkland and residential sites. Signs of rats, mice, hedgehogs, cats, possums and rabbits were found in the study area. No mustelids were detected and are therefore absent or in low numbers. Rats, mice and possums were found in the highest abundance in gully sites. Gully sites were the only sites where all animals were detected. These animals cause browse and predation damage to indigenous flora and fauna. High vegetation cover, inclusion of waterways and connectivity between rural and urban zones were reasons given as to why pest numbers were higher in gullies. Innes et al. (2010) studied ship rat densities in both managed and unmanaged forest fragments. It was found when forest fragments were managed the vegetation and leaf litter increased; unfortunately, so did the number of rats. To successfully protect indigenous flora and fauna, vegetation stands need both plant and animal pest management.

Surrounding Land Uses

Knowing what effect the surrounding land matrix is having on an area is important for the overall management of a site. Not only does the land immediately adjoining the site have an effect but also the land use in the local catchment. Studies have shown different land uses affect hydrological regimes, water quality, biodiversity and micro-climate of forest fragments (Mensing et al. 1998; Hicks et al. 2001; Denyer et al. 2006; Gobel, et al. 2007, Miller, 2011). Negative land use impacts include farming increasing nutrient runoff into the area, subdivisions increasing water and pollutant runoff and open spaces such as parklands increasing the edge effects in vegetated areas thus decreasing moisture and habitat sustainability. Positive land use impacts include vegetated corridors facilitating species movement and providing shelter for fauna and pest control in forestry blocks benefiting indigenous flora and fauna adjoining forests. The majority of key sites are in landscape matrixes dominated by residential subdivisions, parkland or open space, and/or naturalised exotic or indigenous vegetation. As the landuse has changed from farmland to urban, nutrient runoff is likely to have decreased, but the stormwater runoff would have increased. This

22 can lead to increased erosion and flooding in an area. Increasing scrub and forest improves river and gully bank stability, decreases surface water flows, improves water quality and reduces edge effects.

Habitat Sustainability

Studies into habitat decline and biodiversity loss recognise that at least 10% of remnant habitat cover is needed across a landscape to stem large scale biodiversity loss (McIntyre and Hobbs, 1999; Rutledge, 2003; Drinnan 2005). Currently Hamilton City is approximately 970 ha short of this 10% target. Not only would Hamilton City require more high quality natural areas to achieve this threshold but habitat would need to cover a variety of landforms, ideally reflecting the proportion of landforms and vegetation types within Hamilton City. With this in mind, the alluvial plain and peat bog landforms are the most under-protected landform types of Hamilton City. These landforms have had the most residential and industrial development inside the city because of their accessible terrain. Priority should be given where possible to increasing the indigenous cover and biodiversity on these landforms as they have the highest risk of associated vegetation types disappearing from the city. Riverbank and gully landforms are the most likely to reach the 10% indigenous cover threshold as they only need 2 ha and 35 ha increases respectively. A high proportion of the riverbank and gully systems consist of wasteland vegetation. Therefore there is the potential for council and private land owners to convert this land to indigenous habitat. When Waiwhakareke Natural Heritage Park revegetation has been completed, 37 ha will be added across peat bog, alluvial plain and hillslope landform types.

Threat Mitigation

Threat mitigation to key sites is the same as set out by Downs et al. (2000). Updated information is provided below.

Weed Control

A broad range of management happens within Hamilton’s key sites from no management to intensive management. Downs et al. (2000) recommended focusing weed control at the highest ranked ecological sites. Of the eight highest ranked key sites six are in full or partial council ownership. Of these eight key sites Claudelands Bush, Southwell School, Horseshoe Lake, Forest Lake, Hammond Bush and Berkley kahikatea key sites all have weed control and planting of natives. Weeds such as wandering Jew, Mexican water lily and grey willow are either under continual control or have been eradicated from these sites. The two other highest ranked key sites, Burbush Road and the Riverbank Mamaku-kamahi forest at Hamilton Gardens are not actively managed at present. Burbush Road is fenced but had no signs of active management. Riverbank Mamaku- kamahi forest, Hamilton Gardens is on steep riverbank and it may be unrealistic to manage the whole site due to safety concerns and the risk of damaging the environment. Recommendations by Downs et al. (2000) are still relevant today.

Connectivity and Buffering

Due to the importance of the connectivity between key sites and buffering of these ecologically important zones many of the closer riverbank and gully sites are collated on to the same SNA

23 sheets. As stated above, surrounding land uses affect key sites with connected areas having the most influence. Therefore, buffering, closest seed sources and fauna movement need to be taken into consideration for management of key sites.

Interactions with Landowners

Since the Downs et al. (2000) report there has been increased co-operation between council, landowners and community groups in the restoration and management of natural areas within Hamilton City. Forty two of the 70 key sites recognised have had some management input from the community as well as the council.

In 2001 the Gully Restoration Programme was initiated. The aim of this programme is to raise awareness and appreciation of Hamilton gully systems and actively promote and enable the physical restoration of the gullies. Through this programme the council has provided practical workshops and seminars, regular newsletters and approximately 2000 eco-sourced native plants annually to residents keen on restoring the sections of gullies on their properties (funded by both Hamilton City Council and Waikato Regional Council). Currently 800 residents, including schools, are on the gully programme mailing list and approximately 200 people have received plants from the council over the years (pers comm. Tim Newton, Hamilton City Council March 2011; Clarkson et al. 2012). This has led to enhancement plantings and weed control taking place in all the largest gully systems and some minor gullies in Hamilton City. This has contributed to the increase in size, connectivity and diversity within the gully systems.

Private and community group restoration has occurred at non-gully sites as well. Restoration efforts at Waiwhakareke Natural Heritage Park, Claudelands Bush and Hammond Bush have had a significant positive effect on those key sites. Both Waiwhakareke Natural Heritage Park and Hammond Bush have increased in size and quality from the 2000 survey. There are other projects around the city also enhancing both key sites and other areas with native planting and management.

Following the success of the Hamilton Halo project in increasing tui numbers within Hamilton, in 2010 Project Echo was established to gather information about the resident long-tailed bat population within Hamilton City. This project aims to discover the distribution, roosting sites and population density of Hamilton’s bat population. Although the programme is in its infancy, several new bat locations have already been discovered within the city (LeRoux 2010, LeRoux & LeRoux 2012). Having healthy old growth native forest within Hamilton City is essential to support tui and bat populations as well as many other fauna species.

Long Term Strategy

Some of the steps recommended by Downs et al. (2000) as part of the long term strategy of the key sites have been executed. Below are the recommendations from that report, and notes on implementation, and their relationship to the proposed Waikato Regional Policy Statement (RPS) Part B. All the recommendations are still relevant today and efforts to implement them should continue.

1. Increase the reserve network to include a range of representative examples of indigenous ecosystems for all the major land units. RPS section 11.1

Some gully key sites have been added to the council parks and reserves network. Indigenous vegetation types have been planted across a range of landform types at Waiwhakareke Natural Heritage Park.

2. Restoration of sites in council ownership. RPS section 11.1.1d

Since 2000 thirty-three of the key sites in full or partial council ownership have had some vegetation restoration management. This has varied in size from a small area of planting at the edge of a site to full site weed control and planting of large areas with new vegetation types.

3. Acquisition of reserve contributions from new subdivision developments. RPS section 11.1.1a

Parts of the Te Awa o Katapaki Reserve have come into council ownership in conjunction with the subdivision of the surrounding land, with more to come as development continues alongside the gully system. The Gordonton Road key site gullies within the Puketaha development are scheduled to come under council ownership in the future.

4. A broader strategy for the long term management of the key sites. RPS section 11.1.6

Management plans have been developed for Hamilton City parks and reserves which include key sites but there is still no overarching plan for all key sites in Hamilton City. This overarching plan is still necessary to properly protect and enhance the key sites. It would also help with providing information to landowners and the public about key sites.

5. Linkage of key sites through restoration of intervening areas. RPS section 11.1b and 11.1.1c

As recommended, areas between key gully sites and riverbank sites have received focused restoration efforts to increase linkages.

6. Graham Island also presents excellent possibilities for restoration as Hamilton’s only permanent river island. RPS section 11.1c and 11.1d

No restoration effort has apparently taken place on the island.

7. Suggested management for Lake Rotoroa includes keeping the wooded area weed free and monitoring of marginal weeds. RPS section 11.2.2b

Weed removal of region plants identified in the Regional Pest Management Strategy has been undertaken at this site, with ongoing monitoring of these species numbers.

8. Lake Rotokaeo; reducing the effects of trampling and cycling. RPS section 11.2.2b.

A raised boardwalk has been installed in the wettest areas around the lake.

9. Horseshoe Lake/Waiwhakareke Natural Heritage Park; the creation of the proposed ‘living museum’ and development of a comprehensive restoration plan. RPS section 11.2.2b.

The restoration of Waiwhakareke Natural Heritage Park has been underway for seven years now. An Operative Management Plan (2011) has been adopted incorporating a restoration plan with the overall objective of creating a self-sustaining, pest-free habitat sanctuary that represents the original ecosystem diversity of the Hamilton Basin.

10. Active management of Hamilton’s single peat bog (Nawton wetland), including maintenance of a high water table. RPS section 11.2.2b.

There has been no active management of the vegetation at the Nawton wetland site. Fortunately the site retains natural characteristics as the water table is still high. A boardwalk has been installed at the site.

11. Street-side plantings with a focus on indigenous species to create linkages through urban areas. RPS section 11.1.b and 11.1.1c.

All new major roads in Hamilton have had plantings, dominated by indigenous species, installed alongside them. This has greatly increased potential new natural corridors around Hamilton City. Unfortunately on some roads such as the Pukete section of Wairere Drive weed species such as ivy and agapanthus were used as well.

12. Further encouragement and support of restoration by school and community groups and individuals will also build connections between areas of ecological value. RPS section 11.1.5

As stated in the interactions with land owners section above, there has been an increased linkage and co-operation between the council and Hamilton City residents in the protection and promotion of indigenous areas of the city. This has led to community groups, schools and individual landowners helping in the management of many of the key sites within the city.

Further Recommendations

Many of the recommendations made in Downs et al. 2000 have been acted on by the council with positive results for the city and its vegetation. The following recommendations supplement these previous recommendations.

 Fauna survey required (RPS section 11.2): These key sites are limited to the ecological importance of vegetation within the city. No survey was undertaken in relation to habitat of any fauna species within the city. The RPS states significant habitat of indigenous fauna needs to be enhanced and maintained as well as indigenous vegetation. A survey needs to be completed and collated with existing data to understand the city’s fauna populations and habitats in order to produce a comprehensive key sites network. The requirements for protecting fauna can differ from protecting vegetation (e.g. maintaining of hydrological regimes for fish species), and these need to be taken into account in the management of Hamilton City’s natural areas.

 Enrichment plantings (RPS section 11.2.2a): There are many sites around Hamilton City that have been planted to develop areas of parkland into indigenous forest. These plantings consist of early succession species such as tarata, kanuka and karamu. In the future these sites

could become part of the network of ecologically significant sites within the city. To promote this progression MacKay et al. (2011) suggests enrichment planting of later succession species, be undertaken once the early succession species have developed; probably before 20 years has passed. This increases species richness and brings sites closer to the state of mature intact forests. The present survey also showed that sites with low human disturbance and quality maintenance (e.g. weed control) were in the best state.

 Increase native habitat (RPS section 11.1.1): To stem the loss of biodiversity within Hamilton City it will be necessary in the longer term to aim for the aspirational threshold of 10% minimum vegetation cover. To increase native vegetation cover the council, industry, iwi and community groups need to continue restoration efforts around the city and investigate how to most effectively reach the 10% threshold. While this needs to be done across all landform types realistically most progress can be made in the gully landform and 100% restoration of gullies would result in almost 10% native cover in Hamilton.

 Protection (RPS section 11.1.3 and 11.1.5): Loss of key sites has already occurred due to the lack of protection given to sites. Protection is needed for all sites, which could be by formal protection through the District Plan or other mechanisms, including incentive schemes.

Monitoring

Forty of the key sites identified by Downs et al. (2000) had permanent vegetation monitoring plots installed in 2010 using the iTree methodology (Nowak et al., 2003). These vegetation plots were located in sites that were accessible in terms of terrain or landowner permission. Monitoring plots also exist in Claudelands Bush and in the new plantings at Waiwhakareke Natural Heritage Park. Monitoring these sites in the future will give information about changes in vegetation structure, species richness and management effects. These should be re-measured at approximately five yearly intervals.

Eleven years have elapsed since the first key site survey. This is a longer time frame than the five years suggested by Downs et al. (2000). A rapid reconnaissance five yearly site check of the key sites to see if they are under threat may have prevented the loss of the Pukete Farm Park Gully site. It is therefore recommended that this style of monitoring survey of all key sites is undertaken in 2016.

Conclusion A little over 1% of Hamilton City’s area has been identified as key site habitat. This falls short of the 10% cover needed to reduce the most severe fragmentation threats. The lack of management or protection of some sites has led to the removal of two key sites from the network. A publicized overarching strategy for all key sites is needed to inform people of the existence and of these key sites to avoid further loss. Overall there has been an increase in ecologically significant land within Hamilton City and restoration projects already occurring will continue to increase this area. However, the currently

27 maturing vegetation will still need to be supplemented by aiming for achievement of the 10% threshold in the longer term. Increases need to take into account the preservation of different landforms and vegetation types representative of the city. Currently there is a distinct lack of key sites on alluvial terraces, Hamilton City’s most dominant landform. Although gullies and riverbanks around the city are a significant source of biodiversity and connectivity, most of the other key sites are very isolated within the city landscape. Increasing the connectivity of sites around Hamilton City and with the surrounding peri-urban districts will increase the flow of native plant and animal species through the city. The identification, protection and enhancement of key sites within Hamilton are important both to fulfil the City’s obligations under the RMA and RPS and because these ecologically significant areas are important in their own right. They contribute to ecosystem functions, natural local heritage and the liveability of the city. They enhance indigenous biodiversity and provide services such as enhancement of amenity features, form part of the stormwater system. Full indigenous vegetation improves bank stability and increases water quality. The work done to increase the quality and quantity of key sites across the city has already had positive results. Future work will reinforce those gains and contribute a healthy, sustainable environment for generations to come.

Acknowledgements Thank you to landowners for permission to view the native remnants on their properties. Thank you to Kemble Pudney, Daniel Laughlin, Emma Coleman and Catherine Bryan for comments on this document.

References Atkinson, I. A. E. (1985), "Derivation of vegetation mapping units for an ecological survey of Tongariro National Park, North Island, New Zealand", New Zealand Journal of Botany 23: 361-378. Clarkson, B. D., Clarkson, B. R. & Downs, T. M. (2007), Indigenous Vegetation types of Hamilton Ecological District, CBER Contract Report 68. Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato). Clarkson, B. D. & McQueen, J. C. (2004), "Ecological Restoration in Hamilton City, North Island, New Zealand", 16th International Conference, Society for Ecological Restoration, Victoria, Canada, Society for Ecological Restoration. Clarkson, B. D., Merrett, M. & Downs, T. M. (2002), Botany of the Waikato, Hamilton, Waikato Botanical Society Inc. Physical Environment by McCraw, J. 13-22. Collier, K. J., Aldridge, B. M. T. A., Hicks, B. J., Kelly, J., MacDonald, A., Smith, B. J. & Tonkin, J. (2009), "Ecological values of Hamilton urban streams (North Island, New Zealand): constraints and opportunities for restoration", New Zealand Journal of Ecology 33(2): 177-189. Coleman, E. J. (2010) "Mechanisms of interference between Kahikatea and Grey willow in the Waikato", unpublished MSc thesis, University of Waikato, New Zealand.

Cornes, T. S. & Clarkson, B. D. (2010), Assessment of vegetation condition and health at Claudelands Bush (Jubilee Bush, Te Papanui), CBER Contract Report No. 113, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Cornes, T. S., Wehi, P. M. & Clarkson, B. D. (2008), Waiwhakareke Restoration Plantings: Establishment of Experimental Monitoring Plots 2008. CBER Contract Report No. 86, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Craig A. (2007), Gully Reserves Management Plan, December 2007 amendment of Turner, J. (2001), Hamilton City Council.

Denyer, K., Burns, B. & Ogden, J. (2006) "Buffering of native forest edge microclimate by adjoining tree plantations", Austral Ecology, 31: 478-489.

Downs, T. M., Clarkson, B. D. & Beard, C. M. (2000), Key Ecological Sites of Hamilton City, CBER Contract Report No. 5., Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato). Drinnan, I. N. (2005), "The search for fragmentation thresholds in a southern Sydney suburb", Biological Conservation, 124: 339–349. Environment Waikato (2010), "Our Climate", Retrieved 6 January, 2010, from http://www.ew.govt.nz/environmental-information/About-the-Waikato-region/Our-natural- environment/Our-climate/. Fitzgerald, N. & Innes, J. (2009), Hamilton City Biennial Bird Counts, 2004-2008, Landcare Research Contract Report LC0910/099, Landcare Research, Hamilton.

Gobel, P., Dierkes, C. & Coldewey, W. G. (2007), "Storm water runoff concentration matrix for urban areas", Journal of Contaminant Hydrology, 91 (1-2): 26-42.

Hamilton City Council (2011) Hamilton City Council's 2010/11 annual report; a year in review, Hamilton City Council. Hamilton City Council & Waikato District Council (2005) Strategic agreement on future urban boundaries between Hamilton City Council and Waikato District Council, March 2005, Hamilton City Council and Waikato District Council. Harding M. (1997), Waikato protection strategy, a report to the forest heritage fund committee, Forest Heritage Fund, Wellington.

Hicks, B. J. (2003), "Biology and potential impacts of rudd (Scardinius erythrophthalmus L.) in New Zealand", Managing invasive freshwater fish in New Zealand.Proceedings of a workshop hosted by Department of Conservation, R. Munro. Hamilton, New Zealand, Department of Conservation: 49- 58. Hicks, B. J., Brijs, J. & Bell, D. (2009), Boat electrofishing survey of Lake Rotokaeo, Hamilton, CBER Contract Report No. 93, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato). Hicks, B. J., Ling, N., Osbourne, M. W., Bell, D. G. & Ring, C. A. (2005), Boat electrofishing survey of the lower Waikato River and its tributaries, CBER Contract Report No. 39, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Hicks, B. J., Osborne, M. W. & Ling, N. (2006), "Quantitative estimates of fish abundance from boat electro-fishing". Proceedings, Australian Society for Fish Biology annual conference, Darwin, NT. 11-15 July 2005. Hicks, B. J., Reynolds, G. B., Laboyrie, J. L. & Hill, C. D. H. (2001), Ecological and physical characteristics of the Te Awa O Katapaki Stream, Flagstaff, Waikato, CBER Contract Report No. 13, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Innes, J. , King, C. M., Bridgman, L., Fitzgerald N., Arnold G. & Cox N. (2010), "Effects of grazing on ship rat density in forest fragments of lowland Waikato, New Zealand", New Zealand Journal of Ecology, 34 (2): 227-232.

Le Roux, D. S. (2010), "Monitoring long-tailed bat (Chalinolobus tuberculatus) activity and investigating the effect of aircraft noise on bat behaviour in a modified ecosystem". unpublished MSc thesis, University of Waikato, New Zealand.

LeRoux, D. S. & LeRoux, N. N. (2012) Hamilton city-wide Bat survey 2011/2012, Project Echo and Kessels & Associates.

McIntyre, S. & Hobbs, R. (1999), "A framework for conceptualizing human effects in landscapes and its relevance to management and research models", Conservation Biology 13: 1282–1292.

McEwen, M., (1987) Ecological regions and districts of New Zealand, New Zealand Biological Resource Centre publication 5. Department of Conservation, Wellington.

MacKay, D. B., Wehi, P. M & Clarkson, B. D. (2011), "Evaluating Restoration Success in Urban Forest Plantings in Hamilton, New Zealand", Urban Habitats.

McQueen, J. C. (2004), Incorporation of Temple View into the Hamilton City Proposed District Plan - Field Investigation Results, CBER Contract Report No., Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Mensing, D. M., Galatowitsch, S. M. & Tester, J. R. (1998), "Anthropogenic effects on the biodiversity of riparian wetlands of a northern temperate landscape", Journal of Environmental Management, 53: 349-377.

Miller, K. T. (2011) "Understory restoration in Hamilton urban forests", Unpublished MSc thesis, University of Waikato, New Zealand.

National Institute of Water Atmospheric Research (2010), "New Zealand national climate summary", NIWA Media Release, 13 January, Auckland, NIWA: p. 1-38.

Nowak, D. J., Crane, D. E., Stevens J. C., Hoehn, R. E. (2003) "The urban forest effects (UFORE) model: Field data collection manual. USDA Forest Service, New York, U.S.A. Ornithological Society of New Zealand (2010a), Matuku: Waikato Region Newsletter February, Hamilton, Ornithological Society of New Zealand. Ornithological Society of New Zealand (2010b), Matuku: Waikato Region Newsletter June, Hamilton, Ornithological Society of New Zealand.

Ornithological Society of New Zealand (2010c), Matuku: Waikato Region Newsletter September, Hamilton, Ornithological Society of New Zealand. Ornithological Society of New Zealand (2011), Matuku: Waikato Region Newsletter February, Hamilton, Ornithological Society of New Zealand. Paris, B. (2010) Hamilton Halo Project Post Operational Report, EWDOCS n1796965, Environment Waikato.

Parks and Gardens Unit, Hamilton City Council (2004), Donny Park Management Plan, 2004 first review, Hamilton City Council.

Parks and Gardens Unit, Hamilton City Council (2009), Minogue Park Operative Management Plan October 2009, Hamilton City Council.

Parks and Gardens Unit, Hamilton City Council (2010), Operative Hamilton Lake Domain Management Plan, December 2010, Hamilton City Council.

Collier, K. J., Hamilton, D. P., Vant, W. N. & Howard-Williams, C. (2010) The Waters of the Waikato, Hamilton, Environment Waikato & Centre for Biodiversity and Ecology Research (The University of Waikato): Bats and the Waikato River by Parsons, S. & Dekrout, A. p. 230. Policy and Transport Group, Waikato Regional Council (2010), Proposed Waikato Region policy statement, November 2010, Waikato Regional Council.

Russell, P. (2010) Waikato Regional Pest Management Strategy Operational Plan 2010-2011. Environment Waikato.

Rutledge, D. (2003) "Landscape indices as measures of the effects of fragmentation: can pattern reflect process?", Department of Conservation Science Internal Series No. 98, Wellington, NZ Smith, B. J. (2007) Diversity of Adult Aquatic Insects in Hamilton Urban Streams and Seepages, Waikato Regional Council.

Stevens, M. I., Clarkson, B. D., McLean, D. A. (2002) Rototuna ecological survey, CBER Contract Report No. 22, Hamilton, Centre for Biodiversity and Ecology Research (University of Waikato).

Waikato Regional Council (2010), Project Echo Factsheet, brochure, Waikato Regional Council, Hamilton.

Waikato Regional Council (2011, July), ‘Bell’ of the town. Your Waikato: Your regional news update from Waikato Regional Council 70, 2

Walker, S.,Brower, A. L., Clarkson, B. D., Lee, W. G., Myers, S., Shaw, W. B., Stephens, R. T. T. (2008) “Halting indigenous biodiversity decline: ambiguity, equity and outcomes in RMA assessment of significance,” New Zealand Journal of Ecology 32(2)

Appendix one: SNA Review – Site Visit Assessment Form

Field sheets for inventory and quick assessment. Adapted from Ecobase and REA wetland field sheets, and Horizon Regional Council July 2007

Site number: Site Name: Recorder: Location Grid Ref/GPS: Date: Start Time: Finish Time:

General information for the site: Tenure: 5 Mixed / Multiple 4 Public property 3 Maori land 2 Private property 1 Leased Protection status 9 No legal - no managed protection (pest control) Estimated area (ha) legally 8 Reserve (TLA) - no managed protection protected (if not entire remnant): 7 Covenant (QEII) - no managed protection 6 Reserve (DoC) - no managed protection 5 No legal protection, but fenced and/ or pest controlled Protection measures (e.g. pest 4 Agreement, contract (RC) and managed protection* control, fencing): 3 Reserve (TLA) managed protection* 2 Covenant (QEII) and managed protection* 1 Reserve (DoC) and managed protection*

* managed protection is fenced and/or pest controlled Fencing 4 No fencing Year fenced (if known): 3 Some fencing (one side, large breaks) % 2 Mostly fenced (areas where stock access is likely) % 1 Secure, intact fencing around entire perimeter % Was the entire fence line seen? Yes / No Draw existing fences on map.

Matrix land-use 9 Dairying, cropping or horticulture Production method: 8 Sheep, beef or other agriculture Organic / Conventional 7 Lifestyle blocks 6 Urban subdivisions 5 Garden; parkland 4 Open space; Bare land; Recreation land 3 Plantation forestry 2 Coastal dunes 1 Permaculture tree lands; planted natives 0 Indigenous forest or scrub Owner Details: Name: Hamilton City Council Address: Phone Number(s):

feedback requested Yes/No

Significance Justification criteria to Identify SNA (Source from http://www.ew.govt.nz/PageFiles/6777/rpsdecember07.pdf, pages 216-217) The features that qualify the site for each criterion met (example): Significance criteria (RPS) For example…

Criteria_1 - Protected or Preserved Hamilton City Council park

Criteria_2 - Recommended for Protection Identified by DoC (1993) as being worthy of protection

Criteria_3 - Threatened or Endemic Species Habitat Bat feeding site.

Criteria_4 - Under Represented A patch of wetland which is under-represented (or rare) in the Hamilton Ecological District

Criteria_5 - Uncommon Before Settlement River islands

Criteria_6 - Indigenous Wetland Habitat Contains (or is likely to contain) a natural wetland.

Criteria_7 - Large Indigenous Habitat c. 3 ha podocarp forest remnant in Hamilton City

Criteria_8 - Critical Aquatic Habitat Wetland with potential mudfish habitat.

Criteria_9 - Healthy Indigenous Vegetation Representative remnants of moderately dense podocarp forest,

Criteria_10 - Rare or Exceptional Representation Nationally rare Sporadanthus-Empodisma bog habitat

Criteria_11 - Ecological Buffer Linkage or Corridor Indigenous forest connected with a gully system

Significance Criteria met:

A. Site Description (see definition site in EWDOCS#1709807):

Is there a stream running through the site (circle) Yes? No?

B. Habitat and Vegetation Description

Vegetation description

Unit Unit area

Code (Dominant species)

Habitat

Number

l Regime l

estimate Category

Character Hydrologica

C. Flora and Fauna C.1. Additional botanical information:

Include reference of rare, threatened or distinctive plant species seen or known to be, or have been present at the site (provide source and date of information e.g. SSBI, PNAP, botanical society. Also provide general comments on forest /vegetation composition e.g. dominant canopy species, understorey species etc.)

Rating (Tick appropriate level) for each unit. 3

information

Unit Unit 1 Unit 2 Unit

Forest/scrub

Canopy 1 Very sparse foliage, many large holes, dieback >20%. condition 2 Foliage sparse in some areas, canopy holes uncommon. Some dieback. 3 Foliage mostly dense, only occasional sparse areas, canopy holes rare, very

occasional dieback. 4 Abundant dense foliage over whole canopy, no canopy holes or dieback. Mid Tier 1 No browse palatable species 45cm-1.35m. Understorey bare. 2 Very few browse palatable species 45cm-1.35m. Scattered seedlings of

less palatable species. 3 Moderate browse palatable species 45cm-1.35m. Other species relatively

abundant. 4 Abundant browse palatable species and other species present. Ground 1 Bare soil, rock, >20% of forest floor. Ground vegetation (ferns, moss, Cover seedlings etc <45cm tall) absent of uncommon. Leaf litter on remainder of forest floor. 2 Scattered bare soil & rock. Ground vegetation <20%. Leaf litter on

remainder of forest floor. 3 Bare soil, rock absent or very uncommon. Ground vegetation 25%-50%.

Leaf litter on remainder of forest floor. 4 No bare soil or rock, or eroding soil. Ground vegetation, abundant, 50%-

100%. Leaf letter on remainder.

Wetland Unit : Description (please tick appropriate category) Lake Shallow water <2m Swamp Marsh Fen Bog Shrub-carr Grass/sedge Deciduous margin Saline Other (describe) meadow Water in the Yes No Evidence of water Yes No Unsure system? level changes? Degree of water Clear Water Quality Algal blooms Pollution turbidity Translucent opaque Plant Communities (enter % cover) Wetland Sedge/grass 100 Herbs Shrubs trees vegetation Vegetation Sedge/grass Herbs Shrubs Trees 100 bordering wetland (wetland margin) Pasture? Plant vigour in wetland Wetland Wetland margin

High√ Medium Low  High√ Medium Low  Invasive species in wetland: cover % 5 Distribution (circle) single patch >1 patch continuous Invasive species in margin: cover % 50 Distribution (circle) single patch >1 patch continuous

 Source of water: spring  surface flows  stream/river  precipitation only 

Erosion/Disturbance Wetland Wetland margin Comments No disturbance 34

C.2. Fauna Record all fauna species (exotic and native) seen (including sign) or heard during the survey.

Indicate whether species were seen (s), heard (h) or whether sign (such as faeces, footprint) was observed.

D. Threats

D.1. Pest plants

Estimate Notes for dominant Comments & suggested

(Tick appropriate level) Species etc

Unit Unit 2 Unit 3 Indicator Unit 1 Rating management

Ground 1 Very common, cover >50%

cover ground area. weeds 2 Common, 10%-50% ground

area. 3 Occasional, up to 10% ground

area. 4 None present. Vine 1 Very common, >50% canopy

weeds cover. 2 Common, 10%-50% canopy

cover. 3 Occasional, up to 10% canopy

cover. 4 None present. Shrub/Tree 1 Very common, <50%

Weeds understorey or canopy cover. 2 Common, 10%-50%

understorey or canopy cover. 3 Occasional, up to 10%

understorey or canopy cover. 4 None present.

D.2. Pest animals

If evidence of the same pest animal is present in different units, this needs to be indicated.

Indicator Estimate 2 Notes for Comments &

dominant suggested

(Tick appropriate level)

Unit Unit 1 Unit Unit 3 Rating species etc. management

e.g. Stock 1 Abundant fresh signs (droppings, major tracks and hoof prints) Stock heard or seen throughout area. 2 Common fresh sign but sometimes scattered. Occasional stock heard or seen. Confined to scattered areas on edge. 3 Sign uncommon. Sign is often old. Only near

edges. 4 No damage.

E. Human Associated Activities (Rubbish (organic or inorganic) dumping, Stock grazing, Drainage, Earth works, Erosion, Top dressing, Fire, Vegetation clearing, Herbicide application, Harvest / vegetation clearing, Planting, Animal pest control, Domestic pets, Fencing)

Unit Activity Impact Suggested Response

Notes

F. Climatic Conditions Humidity Clouds Winds Temp (oC) Dry Clear Calm Hot > 25 Moist -1/3 Cumulus Light breeze (leaves) Very warm 20-25 Mist -2/3 Cumulus Breeze (twigs) Warm 15-20 Fog Cirrus Windy (branches) Cool 10 - 15 Showers Alto stratus Storm (trees sway) Cold 1 - 10 Rain Stratus Frost < 0 Hail Nimbo stratus Sleet Cumulonimbus Snow

G. Context / Nearby Site Information (optional) Record details of other areas of scrub, forest or wetland in the vicinity. Include SNA site number (if applicable and known), dominant vegetation, approximate size and likely condition (e.g. grazed) if known, etc. Provide as much information as possible.

H. Photo Record (optional) Mark photo points on map with a cross. Indicate direction of photographs taken with arrow. Photo No. GPS Longitude / Easting GPS Latitude / Northing Description

I. Management Recommendations

What management activities would help to maintain or enhance this area? Specify and comment including activities such as pest control, fencing, weed control, time, planting buffers, threatened species protection and/or habitat restoration within a site. Comments:

Appendix two: Key Ecological Sites in Hamilton City; ordered by dominant landform and location Gullies and connected Riverbank sites Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new Te Awa O Katapaki Gully 2.1 River Road North Gully Gully Grey willow forest 8.3 8.3 1, 4, 7, 11 2 2 Kanuka/mahoe forest Pukete Gullies 2.3 Pukete Kanuka Gully I Gully Kanuka/kauri-rewarewa forest 1.5 1.6 9 2 2 Kauri-kahikatea scrubⁿ 2.6 Pukete Kanuka Gully II Gully Kanuka-willow/mahoe forest 0.2 0.3 4, 9 3 3 6.3 Totara Park Wetland Gully (Kahikatea)/grey willow-(cabbage tree) forest 1.7 1.7 1, 4 1 2 Mixed native and exotic treeland Kirikiriroa Gully and riverside 6.1 Riverbank North of Pukete Bridge Riverbank Mamaku-(alder)/mahoe forest 0.4 1.1 1, 4, 11 3 3 (alder)-(wattle)/mixed nativeⁿ scrub 6.2 Kirikiriroa Gully, Harrowfield Gully Mahoe-pate forest 0.6 2.2 1, 4, 7, 11 2 2 Mixed native shrublandⁿ Cabbage tree/flax-pasture grassland 3.1 Puketaha Astelia Gully Gully Grey willow-wheki forest 3.1 3.6 4, 7, 11 2 2 7.1 Kirikiriroa Gully Arm, adjacent to Gully Cabbage tree/ grey willow-hawthorn forest 0.6 0.6 4, 11 3 3 Gordonton Rd I Mixed native shrublandⁿ 7.2 Kirikiriroa Gully Arm, adjacent to Gully Cabbage tree/ grey willow-hawthorn forest 0.3 0.3 4, 11 2 3 Gordonton Rd II 7.3 Kirikiriroa Gully, Chartwell Gully Kahikatea/ mixed exotic and native forest 0.5 1.2 1, 4 2 2 Mixed native scrubⁿ 7.9 Kirikiriroa Gully: Mangaiti Gully Carex-flax sedgelandⁿ 3.8 4, 6, 7, 9, 11 2 Mixed native shrublandⁿ Fairfield Gullies 7.6 Donny Park Raupo Gully Raupo reedland 0.2 1.6 1, 4, 6, 7 3 2 Alder-willow treeland Mixed native shrublandⁿ Manuka shrublandⁿ 7.8 Ranfurly Park Kanuka Gully Kanuka forest 0.3 0.7 1, 6 3 2 (Cabbage tree-kahikatea)ⁿ/Carex-raupo sedge- reedland Mahoe-mamaku scrubⁿ mixed native shrublandⁿ 37

Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new Waitawhiriwhiri Gully 11.1 Waitawhiriwhiri Gully, Edgecumbe Gully Tree fern scrub 0.3 0.5 4, 11 3 3 Park Cabbage tree-land 11.2 Waitawhiriwhiri Gully, Whitiora Gully Mixed treefern/ mixed native treeⁿ forest 0.6 0.6 3, 11 3 3 Gibbon’s Creek Gully 11.6 Seeley’s Gully Gully Mixed native forestⁿ 2.2 2.2 1, 4, 6, 9 3 2 Honeysuckle-bindweed vineland/Raupo - Carex sedge- reedland 11.7 Peachgrove Kahikatea Gully Kahikatea/willow-mahoe forest 2.1 2.1 4, 6 2 2 (Kahikatea)/mahoe-cabbage tree forest Willow/Carexⁿ sedgeland Mangaonua Gully and Hammond Bush 16.7 Hammond Bush Riverbank Pukatea/swamp maire forest 1.8 3.3 3, 4, 7, 9, 11 1 1 Alder forest Tawa/titoki forest Mahoe/kanuka forest Machaerina/Phormium sedge/flaxland 16.8 Gully near Hammond Bush I Gully (Alder)/mahoe-lacebark forest 0.4 0.8 11 3 3 16.9 Gully near Hammond Bush II Gully Eucalypt-wattle/mixed nativeⁿ forest 0.2 1.5 3, 6, 11 3 3 Alder-pine-eucalypt/mahoe-mamaku forest Lemonwoodⁿ-eucalypt-kanukaⁿ forest Flax-cabbage tree/Carex flax/sedgelandⁿ 16.10 Riverside Kanuka, Hammond Park Riverbank Kanuka /mahoe forest 0.3 0.8 3, 7, 9, 11 3 3 Kanuka/toetoe scrubⁿ

16.11 Mangaonua Streamside in Gully Kanuka/ mahoe forest 0.9 0.9 3, 7, 9, 11 2 2 Riverlea I 16.12 Mangaonua Streamside in Gully Mahoe forest 0.1 0.1 3, 11 2 2 Riverlea II 17.6 Mangaonua Streamside in Gully Mahoe forest 0.2 0.2 3, 11 2 2 Riverlea III 17.7 Mangaonua Streamside in Gully Mahoe forest 0.2 0.2 3, 11 2 2 Riverlea IV 17.8 Mangaonua Streamside in Gully Mahoe-kanuka forest 0.4 0.4 3, 11 2 2 Riverlea V 17.5 Mangaonua Streamside in Berkley Gully Mahoe-pate-willow forest 0.6 0.7 3, 4, 7, 11 2 2

Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new 17.4 Mangaonua Gully Arm in Berkley III Gully Willow-mahoe-pate forest 0.4 0.7 3, 4, 7, 11 2 2 17.3 Berkley Kahikatea Gully Kahikatea forest 0.4 0.4 3, 4, 7, 9, 11 1 1 17.2 Mangaonua Gully Arm in Berkley II Gully Willow-mahoe-pate forest 0.5 0.5 3, 4, 7, 11 2 2 Lemonwood-totara-mahoe forestⁿ 17.1 Mangaonua Gully Arm in Berkley I Gully Willow-privet forest 0.2 0.3 3, 4, 7, 11 3 3 13.2 Mangaonua Gully, Silverdale Gully (Kahikatea)/treefern forest 2.2 4.3 7, 11 3 3 Grey willow/treefern forest 13.1 Mangaonua Gully, Chelmsford Gully Willow-treefern forest 0.7 1.6 3, 11 2 2 Mangakotukutuku Gully 16.14 Mangakotukutuku Gully, Te Anua Gully Eucalyptus-pine-(kahikatea)/ treefern- privet forest 0.3 0.6 10 3 3 Park 16.15 Kanuka Patch, Mangakotukutuku Gully Kanuka/privet forest <0.1 <0.1 10 3 3 Gully, Peacocke 16.16 Mangakotukutuku Gully Arm, Gully Grey willow forest 0.3 0.3 3, 10 3 3 Peacocke

Alluvial plain

Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new 1.2 Te Rapa North Kahikatea I Alluvial Plain Kahikatea forest 0.4 4 2 1.3 Te Rapa North Kahikatea II Alluvial Plain Kahikatea forest 0.6 3, 4 2 5.1 Burbush Rd Forest/ Perkins Bush Alluvial Plain Kahikatea forest 1.5 1.5 4, 9 1 1 6.4 Mooney St Kahikatea Alluvial Plain Kahikatea-mixed native forest 0.3 0.3 1, 4 2 2 10.2 Grove Park Kahikatea Alluvial Plain Kahikatea forest 0.1 0.1 1, 4 2 3 11.3 Claudelands Bush Alluvial Plain Kahikatea-(titoki)/mahoe forest 5.4 6.5 1, 4, 7, 9 1 1 (Kahikatea)/titoki-mahoe-pukatea forest Tawa forest Mixed native shrublandⁿ 12.1 Southwell Bush Alluvial Plain Kahikatea forest 0.9 1 4 1 1 12.2 Caldwell Native Bush Alluvial Plain Mixed native forestⁿ 0.3 0.3 9 3 3 12.4 Hillcrest Kahikatea Alluvial Plain Kahikatea forest 1.3 1.3 1, 4, 9 2 2

Riverbank and islands

Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new 2.2 Riverside Alder forest with Riverbank Alder-crack willow forest 1.6 1.9 3, 4, 11 3 2 treeferns, Hamilton North Carex germinata sedgeland Tree fuchsia-treefern -cabbage tree forest 2.4 Riverbank Mahoe scrub, Pukete Riverbank Mahoe-treefern forest 1.5 1.5 1, 9 3 3 2.5 Pukete Riverside Mamaku-mahoe Riverbank Mahoe-mamaku forest 1.2 1.2 4, 9 2 2 forest 2.7 Pukete Riverside Kanuka Riverbank Kanuka/mahoe forest 0.5 0.5 4 2 2 7.4 Riverbank opposite St Andrews Riverbank Alder/mahoe-black locust-treefern forest 1.5 1.5 11 2 2 Golf Course 7.5 St Andrews Kanuka Riverbank Kanuka/mahoe-silver fern forest 2.2 2.2 3, 4, 11 1 2 7.7 Riverbank opposite Ann St Riverbank Kanuka-treefern/mahoe-mapou forest 0.8 0.8 4, 9, 11 2 2 11.4 Riverbank south Miropiko Riverbank Kanuka/mahoe-karaka forest 0.1 0.1 3, 4 3 3 15.1 Graham Island (Te Motere o River Island Pampas grassland 0.3 0.3 4, 5 3 3 Kaipikau) Alder-wattle/ mahoe-privet forest Floodline vegetation 16.1 Riverbank east of Cobham Bridge Riverbank Mahoe-tree fern-kamahi forest 0.2 0.2 1, 3, 11 2 2 16.2 River Island, with turf vegetation River Island (Alder)-(grey willow)/ Paspalum-Glossostigma <0.1 <0.1 4, 5 3 3 herbfield 16.3 Mamaku-mahoe forest, Hamilton Riverbank Mahoe-mamaku forest 1.6 1.8 4, 9, 10, 11 2 2 Gardens 16.4 Riverbank Mamaku-kamahi forest, Riverbank Kamahi-mamaku forest 1.5 1.7 4, 9, 10, 11 1 1 Hamilton Gardens 16.5 Hammond Park – Northern End Riverbank Eucalyptus-blackwood/mahoe-mamaku forest 0.3 0.5 11 3 3 16.6 Riverbank Kanka opposite Riverbank Kanuka- privet - mamaku forest 2.4 2.4 4, 7, 11 1 2 Hammond Park 16.13 Riverside Kanuka, Peacocke Riverbank Kanuka/mahoe-privet forest 3.3 3.3 7, 11 3 3

Peat lakes, Peat bogs and hillslopes

Site Name Land Unit Main Vegetation Type(s) Area (ha) Criteria Ecol. Rank No. old new old new 5.2 Waiwhakareke Natural Heritage Peat Lake Open water with water lily 3.8 16.2 1, 3, 4, 6, 7, 2 2 Park (Horseshoe Lake) Peat Lake Baumea-raupo-kutakuta reedland 8, 9, 10, 11 Alluvial plain & Kahikatea-manuka-flax shrublandⁿ Peat bog Peat bog Cane rush-wire rush restiadlandⁿ Hillslope Manuka-kanuka-kohuhu-ribbonwood shrublandⁿ Kohuhu-kauri shrublandⁿ Hillslope 9.1 Nawton Wetland Peat bog Grey willow forest 2.2 0.9 1, 4 2 3 Manulka forest 9.2 Brymer Park Peat bog Paspalum-Carex-Baumea grass-sedgelandⁿ 1 1, 4, 6 3 Hillslope Mixed native scrublandⁿ 10.1 Lake Rotokaeo (Forest Lake) Peat Lake Manuka/Carex shrub-sedgelandⁿ 4.8 5.5 1, 4, 6, 7, 9 1 1 Peat Lake Baumea reedland Peat Lake & Manuka scrubⁿ Hillslope Peat Lake Kahikatea forest ⁿ Peat Lake (Kanuka-karamu)ⁿ/exotic herbfield

11.8 Mixed planted forest near Golf Hillslope Mixed native-exotic forestⁿ 0.6 0.6 1, 4, 9 3 3 Area, Hamilton Lake Domain 11.9 Planted Totara Forest, Hamilton Hillslope Totara forestⁿ 1.6 1.6 1, 4, 9 3 3 Lake Domain 11.10 Lake Rotoroa (Hamilton Lake) Peat Lake Raupo-Baumea reed-sedgeland 51.0 56.1 1, 4, 6, 7 2 2 Acacia forest 12.3 Waikato University Kahikatea Hillslope Kahikatea forest <0.1 <0.1 4 3 3 14.1 Templeview Kahikatea Hillslope Kahikatea forest 1.2 4 2

ⁿ indicates planted

Ecological Rank: 1=very high, 2=high, 3= moderate

Appendix three: Ecological significance justification for key sites; ordered by ecological score Ecological score 1

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 5.1 Burbush Rd Forest/ Perkins Best kahikatea forest in the west of the city Species: tawa, titoki Bush Feature: Understorey and shrub layers 5.2 Waiwhakareke Natural Heritage Second largest site, second largest peat lake Species: swamp maire, pukatea, + Larger size, increased species Park (Horseshoe Lake) and contains nationally rare planted Empodisma minus and habitat diversity Sporadanthus restiadland Feature: Wetland fringe 10.1 Lake Rotokaeo (Forest Lake) Diverse riparian vegetation and third largest Regeneration: kahikatea, swamp peat lake coprosma, manuka, kanuka 11.3 Claudelands Bush Best and largest kahikatea forest Species: kiekie, tawa, hangehange, titoki, Collospermum haastatum 12.1 Southwell Bush Fourth best kahikatea forest in the city and Species: tawa, titoki, hangehange few weed species present 16.4 Riverbank Mamaku-kamahi Best mamaku-kamahi forest. Second best Species: rewarewa, large kowhai, forest, Hamilton Gardens riverside forest Metrosideros fulgens 16.7 Hammond Bush Best riverside forest with rare vegetation type Species: Swamp maire, tree fuchsia, for the Waikato pukatea and numerous others 17.3 Berkley Kahikatea Best and oldest kahikatea forest in a gully Species: Tawa, titoki, largest kahikatea Corridor: Mangonua Gully

Ecological score 2

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 2.1 River Road North Gully Largest gully site. Third largest overall. Feature: Kanuka forest and a wetland forest Species: Large fuchsia, Astelia 2.2 Riverside Alder forest with Contains New Zealand passionfruit Feature: Carex wetland and fuchsia + Increased vegetation types treeferns, Hamilton North treefern forest 2.3 Pukete Kanuka Gully I Largest kanuka dominant forest in a gully Species: Large kanuka, fern diversity 2.5 Pukete Riverside Mamaku- Best example of mamaku-mahoe forest Species: fern and epiphyte diversity mahoe forest

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 2.7 Pukete Riverside Kanuka Second largest riverside kanuka forest in the Species: Wheki-ponga north 3.1 Puketaha Astelia Gully Largest Astelia population Species: Sedge and fern diversity 6.2 Kirikiriroa Gully, Harrowfield Best mahoe-pate forest in the north. Species: Wheki-ponga, Trichomanes Regenerating venosum, large mahoe and fuchsia 6.3 Totara Park Wetland Contains various aged kahikatea. Still have - Kahikatea known to not semi-swamp forest conditions. regenerate under willow canopy without management 6.4 Mooney St Kahikatea Fifth largest kahikatea forest in the west. Species: Large titoki Closest natural vegetation to Horseshoe Lake 7.3 Kirikiriroa Gully, Chartwell Large kahikatea present. Old plantings present. Established planted wetland 7.4 Riverbank opposite St Andrews Second best mahoe-mamaku forest in the Golf Course north 7.5 St Andrews Kanuka Best riverside kanuka forest in the north Species: New Zealand Passionfruit, - Large path installed reducing Metrosideros fulgens, fern diversity size 7.6 Donny Park Raupo Best raupo wetland in a gully + Willow control. Natives planted 7.7 Riverbank opposite Ann St Only kanuka/mahoe-mapou forest Species: kiekie, Metrosideros fulgens 7.8 Ranfurly Park Kanuka Kanuka forest in a gully in central city Species: large kanuka + Well established wetland 7.9 Kirikiriroa Gully: Mangaiti Developed planted Carex dominant wetland and native scrubland Largest wetland site. 11.6 Seeley’s Gully Diverse native planted gully Species: Clematis paniculata + Large well developed forest. (planted) Regenerating Feature: Carex and raupo wetland. Native tree, grass and fern regeneration Connectivity: Gibbon’s Creek 11.7 Peachgrove Kahikatea Regenerating kahikatea and native dominant Connectivity: Gibbon’s Creek wetland 11.10 Lake Rotoroa (Hamilton Lake) Largest site and largest peat lake Species: Dianella nigra 12.4 Hillcrest Kahikatea Third largest kahikatea forest. Developed understorey 13.1 Mangaonua Gully, Chelmsford Second largest Astelia population. Corridor Species: Wheki-ponga, swamp along Mangaonua Gully. coprosma (planted)

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 16.1 Riverbank east of Cobham Forest with kamahi in canopy, native Species: Koromiko, supplejack, Bridge dominant canopy Cordyline banksii 16.3 Mamaku-mahoe forest, Best mamaku-mahoe forest in south Species: Rhabdothamnus solandri, Hamilton Gardens large Fuchsia Diversity: Ferns, shrubs, sedges 16.6 Riverbank Kanuka opposite Largest kanuka forest in the south Species: Tawa - Privet now dominant in the Hammond Park sub-canopy 16.8 Gully near Hammond Bush I Native dominant gully connected to Diversity: Ferns + Now part of Hammond Bush. Hammond Bush Plantings. Regenerating 16.11 Mangaonua Streamside in Corridor. Kanuka/mahoe dominant forest in Size: Largest site on Riverlea section Riverlea I Mangaonua Gully of the gully 16.12 Mangaonua Streamside in Corridor along Mangaonua Gully Species: Fuchsia Riverlea II Diversity: Ferns 17.2 Mangaonua Gully Arm in Corridor along Mangaonua Gully Species: Fuchsia Berkley II Diversity: Ferns 17.4 Mangaonua Gully Arm in Corridor along Mangaonua Gully Species: Fuchsia Berkley III 17.5 Mangaonua Streamside in Corridor along Mangaonua Gully Species: Fuchsia Berkley Feature: Carex sedgeland 17.6 Mangaonua Streamside in Corridor along Mangaonua Gully Species: Fuchsia Riverlea III 17.7 Mangaonua Streamside in Corridor along Mangaonua Gully Diversity: ferns sedges Riverlea IV 17.8 Mangaonua Streamside in Corridor. Kanuka/mahoe dominant forest in Species: Large kanuka Riverlea V Mangaonua Gully 1.1 Te Rapa North Kahikatea II Fourth largest kahikatea forest in the west 1.2 Te Rapa North Kahikatea III Third largest kahikatea forest in the west 14.1 Templeview Kahikatea Western most kahikatea stand. Largest on a hillslope

Ecological score 3

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 2.4 Riverbank Mahoe scrub, Pukete Best example of riverside mahoe dominant scrub 2.6 Pukete Kanuka Gully II Second largest gully site in the west 6.1 Riverbank North of Pukete Native dominant steep riverbank. Connectivity Bridge with Kirikiriroa Gully 7.1 Kirikiriroa Gully Arm, adjacent Native dominant understorey in the north east Diversity: ferns to Gordonton Rd I 7.2 Kirikiriroa Gully Arm, adjacent Natives in understorey in the north east Species: Astelia grandis, wheki- - Not much in the understorey to Gordonton Rd II ponga Diversity: ferns 9.1 Nawton Wetland Terrestrial peatland swamp forest Species: Swamp coprosma, - Majority cleared. Exotic Lobelia anceps dominant canopy. 9.2 Brymer Park Wetland on terrestrial peat 10.2 Grove Park Kahikatea Smallest kahikatea forest Species: miro - -No regeneration or understorey 11.1 Waitawhiriwhiri Gully, Cabbage tree –land with restoration plantings, Edgecumbe Park Waitawhiriwhiri Gully 11.2 Waitawhiriwhiri Gully, Whitiora native-dominated vegetation, Waitawhiriwhiri Species: Adiantum fulvum, wheki- Gully ponga, kiekie 11.4 Riverbank south Miropiko Native dominant riverbank forest Species: Adiantum aethiopicum, Large kanuka + kamahi. 11.8 Mixed planted forest near Golf Old native plantings, regenerating understorey Area, Hamilton Lake Domain 11.9 Planted Totara Forest, Hamilton Planted totara forest Lake Domain 12.2 Caldwell Native Bush Planted native species with regeneration 12.3 Waikato University Kahikatea Small stand of secondary kahikatea forest 13.2 Mangaonua Gully, Silverdale Largest site on Mangaonua Gully, diverse Species: Tmesipteris elongata, regenerating understorey Chiloglottis cornuta, wheki- ponga, large kahikatea 15.1 Graham Island (Te Motere o Largest of two river islands Kaipikau) 16.2 River Island, with turf Smallest of two river islands. Native dominant herbfield vegetation

Site Name Justification Change in Ecological Score No. Main Other + or - Reason 16.5 Hammond Park – Northern End Buffer to highly ecologically significant site Species: Rhabdothamnus solandri, Hebe stricta, Cordyline banksii and hangehange 16.9 Gully near Hammond Bush II Connectivity between Hammond Park and Riverlea 16.10 Riverside Kanuka, Hammond Connectivity between Hammond Bush and Park Riverlea 16.13 Riverside Kanuka, Peacocke Second largest kanuka forest. Privet in canopy Species: kiekie, Adiantum cunninghamii 16.14 Mangakotukutuku Gully, Te Native vegetation Mangakotukutuku Gully Species: Large kahikatea Anua Park 16.15 Kanuka Patch, Kanuka forest Mangakotukutuku Gully Mangakotukutuku Gully, Peacocke 16.16 Mangakotukutuku Gully Arm, Native dominant understorey Mangakotukutuku Species: Wheki-ponga and swamp Peacocke Gully coprosma 17.1 Mangaonua Gully Arm in Connectivity along Manganoa Gully Species: kahikatea, Earina Berkley I mucronata, Rhabdothamnus solandri (planted)

Appendix four: Species Lists Plants Common Name Species Name Acer sp. Acer sp.* African clubmoss Selaginella kraussiana* agapanthus Agapanthus praecox subsp. orientalis* akeake Dodonaea viscosa alder Alnus glutinosa* allseed Polycarpon tetraphyllum* aluminium plant/weed Lamium galeobdolon* apple tree Malus x domestica* arum lily Zantedeschia aethiopica* asparagus fern Asparagus setaceus* Asplenium bulbiferum x flaccidum Asplenium bulbiferum x flaccidum Asplenium gracillimum Asplenium gracillimum Australian fireweed Senecio bipinnatisectus* bamboo Bambusa / Phyllostachys sp.* bamboo grass Oplismenus hirtellus var. imbecillis bamboo orchid Earina mucronata banana passionfruit vine Passiflora mixta* barberry Berberis darwinii* barnyard grass Echinochloa crus-galli* Baumea articulata Baumea articulata Baumea rubiginosa Baumea rubiginosa Baumea teretifolia Baumea teretifolia bear’s breeches Acanthus mollis* beggars' tick Bidens frondosa* Begonia sp. Begonia sp.* bindweed Convolvulus sp.* black locust Robinia pseudacacia* black maire Nestegis cunninghamii black nightshade Solanum nigrum* blackberry Rubus fruticosus* blackwood Acacia melanoxylon* blue lily pilly Syzygium oleosum* blue morning glory Ipomoea indica* bog nertera Nertera sp. bracken fern Pteridium esculentum broadleaf Griselinia littoralis broad-leaved dock Rumex obtusifolius* broad-leaved fleabane Conyza sumatrensis* broad-leaved plantain Plantago major* buddleia Buddleja davidii* bush flax Astelia fragrans bush rice grass Microlaena avenacea buttercup Ranunculus repens* button fern Pellaea rotundifolia cabbage tree Cordyline australis cabbage tree Cordyline banksii camellia Camellia japonica* cane rush Sporadanthus ferrugineus canna lily Canna indica* Carex dissita Carex dissita 47

Carex germinata Carex germinata Carex maorica Carex maorica Carex solandri Carex solandri Carex virgata Carex virgata cherry laurel Prunus laurocerasus* chestnut Castanea sativa* Chinese lantern Abutilon darwinii* Chinese privet Ligustrum sinense* Chinese windmill palm Trachycarpus fortune* cleavers Galium aparine* clematis Clematis paniculata climbing dock Rumex sagittatus* Collospermum hastatum Collospermum hastatum common lawn daisy Bellis perennis* common maidenhair Adiantum cunninghamii common tree daisy Olearia arborescens Coprosma propinqua Coprosma propinqua Coprosma propinqua × robusta Coprosma propinqua × robusta Coprosma rhamnoides Coprosma rhamnoides Coprosma sp. Coprosma sp. Coprosma spathulata Coprosma spathulata Coprosma tenuifolia Coprosma tenuifolia coral tree Erythrina crista-galli* Corokia sp. Corokia sp. crack willow Salix fragilis* cretan brake Pteris cretica cudweed Gamochaeta coarctata dandelion Taraxacum officinale* deadly nightshade Atropa bella-donna* Deparia petersenii Deparia petersenii Diplazium australe Diplazium australe dock Rumex sp.* egeria Egeria densa* elaeagnus Elaeagnus x reflexa* elephant ear Alocasia brisbanensis* Epacris sp. Epacris sp. Epilobium parviflorum Epilobium parviflorum* exotic broom Cytisus scoparius* exotic umbrella sedge Cyperus eragrostis* fatsia Fatsia japonica* fig Ficus carica* five finger Pseudopanax arboreus flax Phormium tenax fleabane Conyza canadensis* forget-me-not Myosotis arvensis* foxglove Digitalis purpurea* fragrant fern Microsorum scandens Gahnia sp. Gahnia sp. Gahnia xanthocarpa Gahnia xanthocarpa garden forget-me-not Myosotis sylvatica* garlic weed Allium triquetrum* gingko Gingko biloba* Glossostigma sp. Glossostigma elatinoides gooseberry Ribes uva-crispa* 48 gorse Ulex europaeus* grape Vitis vinifera* grey willow Salix cinerea* gully fern Pneumatopteris pennigera gum Eucalyptus sp.* gunnera Gunnera tinctoria* gypsywort Lycopus europaeus* hangehange Geniostoma ligustrifolium var. ligustrifolium hanging spleenwort Asplenium flaccidum hawkbit Leontodon taraxacoides* hawksbeard Crepis capillaries* hawthorn Crataegus monogyna* hebe cultivar Hebe sp. heketara Olearia rani hen and chicken fern Asplenium bulbiferum Himalayan cedar Cedrus deodara* Himalayan honeysuckle Leycesteria formosa* hinau Elaeocarpus dentatus holly Ilex aquifolium* honesty Lunaria annua subsp. annua* hook sedge Uncinia uncinata horse chestnut Aesculus hippocastanum* hounds tongue Microsorum pustulatum houpara Pseudopanax lessonii hydrangea Hydrangea macrophylla* Hypolepis distans Hypolepis distans ink weed Phytolacca octandra* Isolepis sepulcralis Isolepis sepulcralis* Isolepis sp. Isolepis sp. ivy Hedera helix* Japanese cedar Cryptomeria japonica* Japanese flowering cherry Prunus serrulata* Japanese honeysuckle Lonicera japonica* Japanese maple Acer palmatum* jasmine Jasminum polyanthum* Jerusalem cherry Solanum pseudocapsicum* jointed fern Arthropteris tenella jointed rush Juncus articulatus Juncus acuminatus Juncus acuminatus* Juncus effusus Juncus effusus* Juncus planifolius Juncus planifolius Juncus sp. Juncus sp.* kahikatea Dacrycarpus dacrydioides Kahili ginger Hedychium gardnerianum* kaikaiatua Rhabdothamnus solandri kaikomako Pennantia corymbosa kamahi Weinmannia racemosa kanono Coprosma grandifolia kanuka Kunzea ericoides karaka Corynocarpus laevigatus karamu Coprosma robusta karo Pittosporum crassifolium kauri Agathis australis kawaka Libocedrus plumosa 49 kawakawa Macropiper excelsum kiekie Freycinetia banksii king fern Ptisana salicina kiokio Blechnum novae zealandiae kiwakiwa Blechnum fluviatile kohuhu Pittosporum tenuifolium korokio Corokia cotoneaster koromiko Hebe stricta kowhai Sophora microphylla kowharawhara Astelia solandri kuta Eleocharis sphacelata lacebark Hoheria sexstylosa lacebark Hoheria populnea ladder fern Nephrolepis cordifolia* lance fern Blechnum chambersii lancewood Psuedopanax crassifolius large-leaved kowhai Sophora tetraptera Lastreopsis microsora Lastreopsis microsora Lastreopsis microsora subsp. pentangularis Lastreopsis microsora subsp. pentangularis Lastreopsis sp. Lastreopsis sp. leaf-less rush Juncus filicaulis* leather-leaf fern Pyrrosia eleagnifolia lemon Citrus limon* lemonwood Pittosporum eugenoides lesser joyweed Alternanthera denticulata* Lilaeopsis novae-zelandiae Lilaeopsis novae-zelandiae liquidambar Liquidambar styraciflua* loquat Eriobotrya japonica* lotus Lotus pedunculatus* macadamia Macadamia tetraphylla* tuhara Machaerina sinclairii macrocarpa Cupressus macrocarpa* magnolia Magnolia grandiflora* mahoe Melicytus ramiflorus maire-taiki Mida salicifolia mamaku Cyathea medullaris mangeao Litsea calicaris manuka Leptospermum scoparium mapou Myrsine australis marbleleaf Carpodetus serratus Metrosideros diffusa Metrosideros diffusa Metrosideros fulgens Metrosideros fulgens Metrosideros perforata Metrosideros perforata Mexican daisy Erigeron karvinskianus* milk tree Streblus heterophyllus milkweed Euphorbia peplus* mingimingi Leucopogon fasciculatus miro Prumnopitys ferruginea monkey musk Mimulus guttatus* montbretia Crocosmia x crocosmiiflora* moth plant Araujia sericifera* mountain flax Phormium cookianum naked ladies Amaryllis belladonna* Nandina domestica Nandina domestica* 50 nasturtium Tropaeolum majus* native umbrella sedge Cyperus ustulatus New South Wales warratah Telopea speciosissima* New Zealand passionfruit Passiflora tetrandra ngaio Myoporum laetum nikau Rhopalostylis sapida Norfolk Island pine Araucaria heterophylla* Nymphaea sp. Nymphaea sp.* NZ jasmine Parsonsia heterophylla NZ shore lobelia Lobelia anceps oak Quercus sp.* olearia Olearia sp. orange alstroemeria Alstroemeria aurea* oxalis Oxalis sp.* oxeye daisy Leucanthemum vulgare* Paesia scaberula Paesia scaberula pampas Cortaderia selloana* parataniwha Elatostema rugosum Parsonsia sp. Parsonsia sp. paspalum Paspalum sp.* pate Schefflera digitata patience Rumex patientia* pearlwort Sagina procumbens* pennyroyal Mentha pulegium* Petries starwort Callitriche petriei subsp. petriei Phoenix palm Phoenix canariensis* pigeonwood Hedycarya arborea pine Pinus sp.* pink bindweed Calystegia sepium subsp. roseata plantain Plantago lanceolata* poataniwha Melicope simplex pohuehue Muehlenbeckia australis pokaka Elaeocarpus hookerianus poplar Populus sp.* poroporo Solanum aviculare Prunus sp. Prunus sp.* psuedopanax Pseudopanax sp. Pteris sp. Pteris sp.* puka Griselinia lucida pukatea Laurelia novae zelandiae puriri Vitex lucens purple wind grass Lachnagrostis striata ramarama Lophomyrtus bullata rangiora Brachyglottis repanda Ranunculus sp. Ranunculus sp.* radiata pine Pinus radiata rasp fern Doodia australis raupeka Earina autumnalis raupo Typha orientalis red azolla Azolla filiculoides red dead nettle Lamium purpureum* redwood Sequoia sempervirens* reed sweet grass Glyceria maxima* rewarewa Knightia excelsa 51 rhododendron Rhododendron ponticum subsp. ponticum* ribbonwood Plagianthus regius rimu Dacrydium cupressinum round-leaved coprosma Coprosma rotundifolia royal fern Osmunda regalis* rye Secale cereal* sand coprosma Coprosma acerosa scarlet pimpernel Anagallis arvensis* scrambling fumitory Fumaria muralis* sheeps sorrel Rumex acetosella* shining spleenwort Asplenium oblongifolium sickle spleenwort Asplenium polyodon silk tree Albizia julibrissin* silver beech Nothofagus menziesii silver birch Betula pendula* silver fern Cyathea dealbata small flowered nightshade Solanum nodiflorum smooth shield fern Lastreopsis glabella Sonchus sp. Sonchus sp.* spindle tree Euonymus europaeus* stinking iris Iris foetidissima* stinking mayweed Anthemis cotula* strap fern Grammitis billardierei supplejack Ripogonum scandens swamp astelia Astelia grandis swamp blueberry Dianella haematica swamp coprosma Coprosma tenuicaulis swamp cypress Taxodium distichum* swamp kiokio Blechnum minus swamp mahoe Melicytus micranthus swamp maire Syzygium maire swamp millet Isachne globosa swamp sedge Carex secta sweet fern Pteris macilenta sweet vernal Anthoxanthum odoratum* tanekaha Phyllocladus trichomanoides tangle fern Gleichenia dicarpa taupata hybrid Coprosma repens hybrid tawa Beilschmiedia tawa tender brake Pteris tremula thin-leaved coprosma Coprosma areolata thistle Cirsium arvense* thornapple Datura stramonium* thread fern Blechnum filiforme titoki Alectryon excelsus Tmesipteris sp. Tmesipteris sp. toad rush Juncus bufonius var. bufonius* toetoe Austroderia fulvida toropapa Alseuosmia quercifolia totara Podocarpus totara tree fuchsia Fuchsia excorticata tree lucerne Chamaecytisus palmensis* tree privet Ligustrum lucida* true maidenhair Adiantum aethiopicum 52 turutu Dianella nigra umbrella palm Hedyscepe canterburyana* Uncinia banksii Uncinia banksii walnut Juglans ailantifolia* wandering Jew Tradescantia fluminensis* waratah Telopea sp.* water fern Histiopteris incisa water pepper Persicaria hydropiper* wattle Acacia spp.* wheki Dicksonia squarrosa wheki-ponga Dicksonia fibrosa white clover Trifolium repens* white maire Nestegis lanceolata wild radish Raphanus raphanistrum* wild strawberry Duchesnea indica* willow weed Persicaria maculosa* wineberry Aristotelia serrata wire rush Empodisma minus wiwi Juncus edgariae wonder tree Idesia polycarpa* woolly nightshade Solanum mauritianum* yew Taxus baccata* Yorkshire fog Holcus lanatus*

* Exotic species

Animals Common Name Class Species Name Australasian harrier Bird Circus approximans Australasian shoveller Bird Anas rhynchotis Australian coot Bird Fulica atra australis banded kokopu Fish Galaxias fasciatus bellbird Bird Anthornis melanura blackbird Bird Terdus merula* black shag Bird Phalacrocorax melanoleucos black swan Bird Cygnus atratus* black teal Bird Anthya novaeseelandiae brown trout Fish Salmo trutta* Canada goose Bird Branta canadensis* carp Fish Cyprinius carpio* Caspian tern Bird Sterna caspia catfish Fish Amieurus nebulosus* chaffinch Bird Fringilla coelebs* common bully Fish Gobiomorphus cotidianus common smelt Fish Retropinna retropinna domestic duck Bird Anas platyrhynchos domesticus* domestic goose Bird Branta sp.* fantail Bird Rhipidura fuliginosa feral rock pigeon Bird Columba livia* frog Amphibian Littoria aurea* gambusia Fish Gambusia affinis* giant kokopu Fish Galaxias argenteus goldfinch Bird Cardulelis carduelis*

53 goldfish Fish Carassius auratus* grey duck Bird Anas superciliosa grey mullet Fish Mugil cephalus grey teal Bird Anas gracilis grey warbler Bird Gerygone igata inanga Fish Galaxias maculatus kingfisher Bird Halcyon sancta little black shag Bird Phalacrocorax sulcirostris little pied cormorant Bird Phalarocorax sulcirostris little shag Bird Phalacrocorax melanoleucos long finned eel Fish Anguilla dieffenbachii long-tailed bat Mammal Chalinolobus tuberculatus magpie Bird Gymnorhina tibicen* mallard duck Bird Anas platyrhynchos* Muscovy duck Bird Cairina moschata* mynah Bird Acridotheres tristis* New Zealand dabchick Bird Poliocephalus rufopectus paradise shelduck Bird Tadorna variegate perch Fish Perca fluviatilis* pied shag Bird Phalacrocrax varius pied stilt Bird Himantopus himantopus pukeko Bird Porphyrio porphyrio rudd* Fish Scardinius erythrophthalmus shining cuckoo Bird Chrysococcyx lucidus short finned eel Fish Anguilla australis silvereye Bird Zosterops lateralis skylark Bird Alauda arvenis* song thrush Bird Turdus philomelos* sparrow Bird Passer domesticus spur-winged plover Bird Vanellus miles starling Bird Sturnis vulgaris* tench Fish Tinca tinca* torrentfish Fish Cheimarrichthys fosteri tui Bird Prosthemadera novaeseelandiae welcome swallow Bird Hirundo tahitica white doves Bird Streptopelia risoria* white faced heron Bird Ardea novaehollandiae white headed stilt Bird Himantopus himantopus leucocephalus yellow hammer Bird Emberiza citronella*

* Exotic species

UNIVERSITY OF WAIKATO

Hamilton New Zealand

Evaluating the Welfare Effects of Biodiversity on Private Lands: A Choice Modelling Application

Richard Yao and Pamela Kaval University of Waikato

Department of Economics

Working Paper in Economics 09/04

March 2009

Corresponding Author

Pamela Kaval Department of Economics University of Waikato, Private Bag 3105, Hamilton, New Zealand

Fax: +64 (7) 838 4331 Phone : +64 (7) 838 4045 Email: [email protected]; [email protected]

Abstract

Biodiversity loss is a global problem, especially in reference to private lands. In response, we investigated whether private land biodiversity may be attained by developing incentives which include funding landholders through the provision of native trees to enhance biodiversity on their own properties. Using choice modelling, we tested this hypothesis. A typical respondent was found to be better off, in terms of welfare, if there would be a biodiversity enhancing scheme in their locality. We also found that respondents in the upper northern regions of New Zealand were relatively more receptive in supporting biodiversity enhancement programmes on their properties, compared to those residing in the southern regions of the country.

Keywords Native biodiversity New Zealand Choice Modelling Community volunteers

JEL Classification Q57; Q2; Q25

Acknowledgements We would like to thank the Foundation for Research, Science and Technology (FRST) for funding this project. We would also like to thank Terry Parminter, Thomas Wilding, Bruce Burns, Kirsten Forsyth, Michelle Bird, Amber Bill, Tim Porteous, Caren Shrubshall, Frank Scrimgeour, Riccardo Scarpa, our 20 focus group participants, and the 709 anonymous respondents for their help with this project.

2 Introduction

Biodiversity loss is a global environmental problem, especially in reference to private lands (Pascual and Perrings, 2007; Theobald and Hobbs, 2002; Daily, 2000; Stoneham et al., 2000; Dale et al., 2000; Noss et al., 1997). In the case of New Zealand (NZ) and Australia, the most severely depleted ecosystems are on private agriculturally productive areas, as well as cleared land and land conversion areas (Dickson et al., 2005; Saunders and Norton, 2001; Norton, 2000; Buckley, 2002; Watkinson et la., 2000). This scenario represents the importance of biodiversity conservation and biodiversity management on private lands. Bennett (2003) suggests that private land biodiversity conservation may be attained by developing incentives such as funding landholders to protect or enhance biodiversity on their own properties. To examine the usefulness of funding private land biodiversity, it is important to ascertain if there is an incentive for private landholders to participate in biodiversity enhancement programmes on private lands.

To examine the value of biodiversity enhancement on private lands, economic valuation techniques can be used. The Convention on Biological Diversity (1998) recognised that the “economic valuation of biodiversity and biological resources is an important tool for well- targeted and calibrated economic incentive measures”. These valuation techniques are classified into two major categories: market and non-market valuation techniques. Since many benefits from biodiversity (e.g., better air quality, higher water quality, habitat provision) are not currently exchanged in existing markets, most biodiversity valuation studies use non-market valuation techniques (Bennett, 2003).

One non-market biodiversity valuation technique is the choice modelling (CM) method. CM is the most recently developed of the stated preference techniques (Rolfe and Windle, 2005; Dickson et al., 2005). CM aids in the estimation of biodiversity use and non-use values (Nunes and van den Bergh, 2001; Pearce, 2001). CM is typically conducted with a survey where a respondent is asked to state his/her preference between a set of environmental features or attributes at a given cost to the respondent, and another set of environmental features or attributes at a different cost (Whitten and Bennett, 2005).

Since CM examines the tradeoffs between alternatives with different sets of actions, CM is found to be an appropriate decision support tool (Rolfe and Windle, 2005). This is because policy decisions often involve the evaluation of several alternatives, with each alternative having a specific set of possible actions (Whitten and Bennett, 2005). Therefore, different alternatives would likely have different impacts on the environmental service in question. For instance, a set of actions for the operation of a more environmentally friendly winegrowing business can include an increase in the number of planted native plants surrounding the growing area, reduction in the amount of chemicals used for growing the crops, and cost of the environmental improvement to an individual. An example of a set of actions for an alternative might consist of a 25% increase in the number of natives, 20% reduction in the amount of chemicals used and a cost to the individual equal to $50/year. A second alternative might consist of a 50% increase in the number of natives, 10% reduction in

3 the amount of chemicals used and a cost to the individual equal to $75/year. The third alternative can be a no action alternative (or status quo alternative) which would have no change in the number of planted natives (existing number of native trees = 200 trees), no reduction in the amount of chemicals used and $0/year cost to an individual. A respondent can choose his most preferred alternative among the three alternatives presented (Hensher et al., 2005).

Many recent biodiversity valuation studies have elected to use the CM technique (Kerr and Sharp, 2007; Schou et al., 2006; Christie et al., 2006; Othman et al., 2004). This may be attributed to the fact that CM surveys elicit less biased responses than other available stated preference techniques (Hanley et al., 2001). Estimates from CM studies also offer advantages for use in benefit transfer1 studies, compared to other non-market valuation techniques (Bennett, 2006). One advantage of incorporating CM data into a benefit transfer study is that it can be used to value marginal changes in environmental attributes (Morrison and Bergland, 2006). Another advantage is that CM allows the transfer of a valuation function that permits adjustment for differences in site characteristics (Rolfe, 2006). These features offer more flexibility and a more useful basis for benefit transfer studies.

However, despite the advantages of the CM technique, the number of biodiversity valuation studies that have applied CM remains limited. In the case of NZ, only one CM biodiversity valuation study, to date, was reported (Kerr and Sharp, 2007). Kerr and Sharp (2007) used CM to estimate the values associated with the protection of indigenous species from the invasive wildling pines on NZ’s South Island. Other NZ biodiversity valuation studies have primarily used the contingent valuation method (CVM) (Yao and Kaval, 2008; Yao and Kaval, 2007). This dominance in the use of CVM in biodiversity valuation is evident, not only in NZ, but around the world (Yao and Kaval, 2007; Christie et al., 2006; Brander, et al., 2006; Woodward and Wui, 2001; Brouwer et al., 1999).

Current biodiversity valuation studies have focused on the biodiversity of parks, wetlands and public forests (e.g., Yao and Kaval, 2007; Woodward and Wui, 2001; Brouwer et al., 1999). Most of these biodiversity valuation study sites were located on government owned land or partly-private-partly-government owned areas. We did not find any biodiversity valuation studies that focused specifically on private land. In this regard, this study addresses this gap in the literature by using the CM valuation technique to answer the question of “Is biodiversity enhancement on private land important to NZ residents?” In this study, biodiversity enhancement refers to the planting of additional trees and shrubs that can provide a better developed habitat for native animals on private lands. In 2000, approximately 70% of NZ land was privately owned or occupied. Since so much land is privately held, private residents play an integral role in influencing the issue of biodiversity loss in NZ (Kneebone, 2000).

1 Benefit transfer is a non-market valuation technique that uses estimates from previous non-market valuation studies such as the contingent valuation method, travel cost method and hedonic pricing method. These valuation estimates are rescaled to suit the conditions of the transfer study site.

4 Analytic Models

In a CM survey, a respondent is presented with different sets of alternatives called choice sets. Each choice set may have between two to six different alternatives. Typically, each choice set contains a baseline alternative (also called the status quo alternative). The inclusion of the status quo alternative in each choice set is recommended to derive welfare-consistent estimates (Bateman et al., 2002, p. 251). Every alternative is composed of several characteristics or attributes (e.g., number of trees, cost). The alternative that has the best set of attributes, as perceived by the respondent, provides the largest positive change in utility to an individual. In a choice set, this “best” alternative is the most influential to the respondent’s well-being and is therefore selected as the ideal alternative by the respondent. From an analytical point of view, an individual’s utility difference has two components: the observed and the unobserved. The observed component of utility is the one that can be observed and modelled by the analyst and is treated as deterministic. The unobserved component is the one not included in the model and is therefore treated as stochastic. We express the utility (U) contribution of alternative j as:

U j = V j + ε j (1)

Where V j is the observed utility change, which can be translated into monetary amounts, whileε j is the change in utility unobservable to a researcher. These two components are generally assumed to be additive and independent of each other. Since Vj is observable, we use it as an approximation of the true change in utility of an individual. We express the influence of a good’s attributes to a person’s observed utility in a simple linear form as:

V j = β 0 j + β1 j f (X 1 j ) + β 2 j f (X 2 j ) + β 3 j f (X 3 j ) + ... + β Kj f (X Kj ) (2)

Where β 0 j represents the alternative specific constant (ASC), or the parameter estimate associated with the measured and observed attributes, β1 j is the parameter estimate associated with attribute X1 and alternative j. The explanatory variables, which in this case represent the attributes of the good in question, are represented as f (...) . This implies that the explanatory variables can be entered into the equation in different forms (e.g., linear, quadratic, log, square root, exponential). An attribute (e.g., price) can interact with another explanatory variable (e.g., attitude) and be entered into the right hand side of the equation

(e.g., β 4 j X 1 j X 3 j ) (Hensher et al., 2005; Louviere et al., 2000). In this choice modelling exercise, we use the Multinomial Logit (MNL) and the Random Parameter Logit (RPL) models.

Multinomial Logit Model

For our analysis, the observed change in utility V represents the estimated total observed benefit from the different combinations of attributes. Equation 2 reveals that the estimate for V depends upon the functional form of the explanatory variables, the level of attributes and

5 the magnitude of coefficient estimates. The most commonly used model for this estimation is the multinomial logit (MNL) model with a linear functional form. The error terms of the MNL model are assumed to be independent and identically Gumbel distributed. The MNL model is a closed-form model which can be estimated using the maximum likelihood approach. The MNL can be expressed as:

exp(X i β j ) Pi (j) = J (3) ∑exp()X i β k k =1

th The term Pi (j) represents the probability that individual i chooses the j alternative from J number of alternatives. Pi (j) is a function of the individual characteristics (Xi) and k unknown parameters (βk). In the MNL model, one of the several parameter vectors is normalized to zero since we only estimate utility differences (Hensher et al., 2005; Haab and McConnell, 2002). To make the notation simpler, the intercept term has been suppressed.

Coefficient estimates from the MNL model can be used to calculate the estimates of the change in welfare associated with a change in the level of an attribute (Hensher et al., 2005). This welfare measure can come in the form of an implicit price or the part worth of an attribute. The formula to calculate the part worth (PW) of an attribute is β PW = A (4) β M

Where β A represents the coefficient estimates for an attribute and β M is the negative of the coefficient of the monetary variable. A part worth value is usually expressed in monetary terms (e.g., $25 for every 5% increase in the number of endangered species protected).

Random Parameter Logit Model

In our MNL model, we assumed that the error term was independent and identically Gumbel distributed. However, the independent and identically distributed (IID) property can be limiting, since all information in the random components of the error term may not be identical in quantity. To relax the IID assumption, we instead used the random parameter logit (RPL) model. The RPL is also called the “mixed multinomial logit” or “mixed logit” model. The RPL model is a generalized version of the multinomial logit model that takes into account the correlations in the unobserved components of utility (Hensher et al., 2005). It accounts for heterogeneity in the preferences of respondents. In an RPL model, individual parameters for taste are allowed to have their own statistical distributions, since parameters are allowed to be specific for each respondent (Revelt and Train, 1998). More specifically, the RPL model uses a maximum simulated likelihood approach that allows explanatory variables to vary over respondents, allowing each random parameter to have a particular distribution (e.g., normal, triangular, uniform) (Hensher et al., 2005).

6 The Choice Modelling Survey Design and Data

Survey Design

In this CM exercise, we created a hypothetical tree planting programme for the local government council. With the help of focus group meetings and suggestions from several experts, we identified four different attributes, or characteristics, of a tree planting programme on private land. These attributes and their corresponding levels are as follows:

1. Type of trees to plant o Levels: (1) non-native species only, (2) mixture of natives and non-natives, (3) native species only.

2. Type of council incentives to private landholders: o Levels: (1) get trees offered by the council for free; and (2) buy their own trees and have the specified amount reimbursed from councils upon the presentation of a purchase receipt.

3. Provision of free expert advice from councils about tree planting. This attribute remained the same for all alternatives, since local government councils in NZ continue to provide this service with or without the implementation of the hypothetical tree planting programme.

4. Values of trees and advice to the landholder. The values assigned included $45, $95, $120 and $145. The $45 refers to the value of an hour of tree planting advice provided by local government councils. The status quo alternative present in each choice set has the value of $45. This alternative did not offer any tree provisions, only advice. Alternatives that have values higher that $45 represent the changed alternatives, since they include both the value of the advice and the value of the trees provided by, or rebated from, councils. The changed alternatives were assigned different values, because different native and non-native trees can have different prices. The most expensive alternative is valued at $145. This value represents a biodiversity enhancing scheme where $100 of trees and $45 of tree planting advice can be obtained by the landowner. Therefore, the value of trees and advice attribute corresponds to the overall value of trees and advice that would be hypothetically provided by local government councils to residents.

In the hypothetical market created, we emphasized that the cost of the incentive package will be shouldered by the local government council. However, respondents were likely to be aware that residents pay local council taxes called annual rates (taxes) to fund local council’s environmental programmes (Environment Waikato, 2008a; Environment Waikato, 2008b)

7 Data Collection

The data collection process occurred between December 2006 and January 2008. A two- stage phone-mail survey was employed. The first survey stage involved the placement of phone calls to 3211 randomly selected NZ households listed in the White Pages telephone directory. A total of 1617 residents were contacted and 803 agreed to participate in the mail survey. These 803 residents were each sent a survey packet. Seven hundred nine (709) residents mailed completed surveys to us using the addressed freepost envelopes that we provided in the survey packet. This constitutes a mail survey response rate of 88.3%. In addition to the 709 returned surveys, we included in our database the responses of the 20 focus group participants, resulting in a total of 729 observations.

Of the 729 completed surveys, 618 had one CM choice set option, while 111 had two choice set options, for a total of 840 possible CM responses. However, not all people answered the CM questions. Seventy-three (73) choice sets were left unanswered by respondents. Eleven of these unanswered choice sets had a note saying that current local government council taxes were too high. We classified these responses as protest answers. Excluding these protest answers, we arrived at a total of 767 valid responses out of 840. We compiled these 767 valid responses into an MS Excel spreadsheet and called the sample the no-protest sample. The other sample, where we included all the responses (including protest responses), is called the with-protest sample. We used these two samples in our regression analysis.

Data Structure and Summary

In our analysis, we ran two different regression models: Model 1, which included the choice attributes only, and Model 2, which included both choice attributes and the socio- demographic characteristics (SDCs) of respondents. In Model 2, we interacted the alternative specific constant, as well as the price variable, with the SDCs. These SDCs were hypothesized to influence respondents’ choice behaviour. Table 1 presents a description of the attributes, the SDC variables and the indicator variables for survey regions. All of these variables were included in Model 2.

8 Table 1. Description of variables included in the regression Variables Description Attributes ASC Alternative specific constant (takes the value of 0 if status quo, 1 for the alternative) Price Price of the tree planting scheme to be shouldered by the local council (in 2007 NZ$) Native Type of trees/shrubs for planting (1 if purely native trees, 0 otherwise) Mixture Types of trees/shrubs for planting (1 if mixture of native and non-native trees, 0 otherwise) Rebate Indicator for preference of purchase rebates (1 if the respondent preferred to have a rebate, 0 otherwise) Socio-demographics Years at Property Number of years living at property Educational attainment Highest educational attainment (1-primary, 2-secondary, 3-tertiary, 4- postgraduate) Volunteer Willing to volunteer to plant native trees in their neighbourhood (1 if willing to volunteer, 0 otherwise) Urban Indicator for urban area (1 if property was in an urban area, 0 otherwise) Wellington Indicator for Wellington Region (1 if property was in Wellington, 0 otherwise) Bay of Plenty Indicator for Bay of Plenty Region (1 if property was in Bay of Plenty, 0 otherwise) Waikato Indicator for Waikato Region (1 if property was in Waikato, 0 otherwise) North Island Indicator for other North Island regions (other than Wellington, Bay of Plenty, and the Waikato) (1 if property was in other North Island region, 0 otherwise) South Island Indicator for the South Island regions (1 if property was in a South Island region, 0 otherwise)

Table 2 presents a summary of the SDC variables and regional indicator variables for the with-protest sample and the no-protest sample. The summary figures for the two samples are virtually the same. In this regard, we focus on discussing the summary statistics of the no- protest sample.

From the no-protest sample, we find that a typical respondent lived at their property for 10 years. More than half (57%) of the respondents had secondary schooling as their highest level of education, while 34% had tertiary and 7% attended up to the post-graduate level of schooling. The spirit of volunteerism in the sample seemed high, as reflected by the fact that 57% of the respondents would be willing to volunteer to plant native trees in their neighbourhood, such as public parks. Almost one-third (31%) of the respondents had properties located in rural areas. This statistic comes very close to the reported data from Statistics NZ (2006) which shows that 33% of NZ residents lived in rural areas in 2005. The Greater Wellington Region served as our priority survey area among the five survey regions. The highest proportion (33%) of respondents had their properties in this region. The other

9 four survey regions had smaller proportions of respondents: Bay of Plenty region (17%), Waikato region (18%), other North Island regions not including the Bay of Plenty, Wellington or Waikato regions (17%) and South Island regions (16%).

Table 2. Socio-demographic characteristics of respondents Characteristic All Responses Excluding Protest Responses Years living at property 10.46 (11.35) 10.02 (10.85) Volunteer to plant natives in their neighbourhood (e.g., 465 (56%) 433 (57% ) public parks) Property in rural areas 255 (31%) 235 (31%) Education Primary = 17 (2%) Primary = 14 (2%) Secondary = 476 (57%) Secondary = 430 (57%) Tertiary = 277 (33%) Tertiary = 258 (34%) Postgraduate = 59 (7%) Postgraduate = 58 (8%) Region Wellington = 275 (33%) Wellington = 251 (33%) Bay of Plenty = 142 (17%) Bay of Plenty = 132 (17%) Waikato = 150 (18%) Waikato = 136 (18%) North Island = 141 (17%) North Island = 127 (17%) South Island = 132 (16%) South Island = 121 (16%) No. of Responses 840 767 Note: Figures in parentheses are standard deviations or percentages of sample.

Results

In the CM survey, each respondent was presented with a choice set with four different alternatives. One of the four alternatives was the status quo which represented the current situation. In the status quo, a resident can avail of free expert advice from the local council. This tree planting advice was assigned a value of $45. No free trees were provided in this alternative. The status quo alternative was included in all three choice sets. Since the status quo was present in each choice set with four alternatives, if we assume that these alternatives have an equally likely probability of being chosen, each alternative would have a 25% chance of being selected. However, our results show that the status quo alternative only had a selection probability of 7%. This implies that a majority of the respondents preferred to have the non-status quo, or changed alternatives, where they will be able to get either tree purchase rebates or free trees from local councils.

Overall, the CM exercise had a total of six alternatives, which were shuffled across three choice sets. Using the average estimated probabilities from the MNL model, we identified the most preferred alternatives among the six. The alternatives with the highest average probabilities were mixture of native and non-native trees you purchase (27%) and natives you purchase (25%). The third and fourth most preferred alternatives were native trees from councils (19%) and mixture of native and non-native trees from councils (18%). This result indicates that, although many respondents preferred to receive free trees from local

10 councils, a greater proportion of these respondents prefer to purchase trees themselves and get those purchase amounts reimbursed from councils. The average probability of selecting the non-natives from councils was only 5%, making this alternative the least preferred among the six.

The MNL regression was run for both the with-protest sample and the no-protest sample. We initially ran Model 1, the model with CM attributes only. For both samples, the estimated MNL coefficients had the expected signs, consistent with economic theory (Table 3). The coefficients for the price variable (the monetary value of trees and advice) are negative and significant at the 99.9% confidence level. This implies that as the price of the government incentive scheme rises, the less it becomes preferred by the respondents. This result is interesting. Even though the respondents were told in the questionnaire that the council will fund the tree planting project on a respondent’s property, respondents may have also been aware that the funds to finance such projects would likely be derived from the annual rates they pay to the council. Having this notion, a typical respondent preferred to choose the cheapest (but most desired) option in a given choice set.

Table 3. Estimates for the attribute only Multinomial Logit Model Model 1 (Attributes Only) With-Protest Sample No-Protest Sample Estimates Part worth Estimates Part worth ASC 0.393 -0.408 (0.008) (0.149) Price -0.007 -0.018 (0.000) (0.000) Native 1.923 $ 291.34 2.136 $ 119.59 (0.000) (0.000) Mixture 1.990 $ 301.46 2.000 $ 112.03 (0.000) (0.000) Rebate 0.267 $ 40.50 0.323 $ 18.07 (0.006) (0.001) Pseudo R2 0.0243 0.0747 Adjusted R2 0.0228 0.0726 Log-likelihood -1177.08 -996.68 No. of observations 840 767 Note: Figures in boldface font are significant at the 90% confidence level or greater. Figures in parentheses are p-values.

Comparing the two samples, the regression summary statistics indicate that the no- protest sample had a higher log-likelihood value, as well as a higher pseudo-R2. This implies that the no-protest sample provided a better model fit than the with-protest sample. If a typical respondent was given the choice between purely natives, mixture of natives and non- natives and purely non-natives, results from the no-protest sample show that this respondent would likely not choose purely non-natives. With purely non-natives serving as the reference dummy variable, the coefficients for purely natives and mixture of natives and non-natives are

11 both positive and significant. In terms of magnitude, the coefficient for purely native is slightly higher than the coefficient for the mixture of natives and non-natives. Using the Wald test to check for the statistical difference between the two coefficients, we get a chi-square 0.10 value of 1.08. This value is lower than the chi square critical value of 2.71 ( χ1 = 2.71). This indicates that we fail to reject the fact that the two coefficients are the same. This implies that purely natives and mixture of natives and non-natives are similarly more preferred than purely non-natives. The part-worth values for these parameters are virtually the same, with $120 for purely native and $112 for the mixture.

The significantly positive coefficient for rebates denotes that respondents preferred to buy and choose their own trees, rather than simply choose from the local council’s available trees. Perhaps respondents feel that the rebate scheme gives them more flexibility to choose suitable trees from other tree nurseries for planting on their own properties, even though, the variety available could be the same. The part worth value for the rebate attribute is $18 in the no-protest sample.

Table 4 presents the regression estimates for Model 2. Model 2 represents the unrestricted MNL regression model where we included the attributes, SDCs, and the indicator variables for the location of the respondent’s property. The unrestricted model enabled SDC variables to explain the choice behaviour of respondents. The positive and significant coefficient estimate for the interaction variable ASC×Years in Property indicates that respondents who resided longer at their property would likely get a higher utility from a governmental tree planting scheme. The positive coefficient for Price×Rural indicates that, although respondents in general would prefer the cheaper alternative, the group of urban respondents would prefer the relatively cheaper planting scheme, compared to the rural respondents.

The interaction variable Price×Volunteer is also significantly positive, which implies that those who would be willing to volunteer, would likely choose the tree planting scheme with a relatively higher price. However, the interaction variable ASC× Education is not significant in this MNL model. To check for variations in choice behaviour between regions, we included indicator variables for the survey regions. Each indicator was interacted with ASC to capture whether residents prefer the changed alternatives in the regions. We dropped the interaction variable ASC×South Island as it would be represented in the constant. Results show that ASC×Wellington is not statistically different than ASC×South Island. The coefficients for ASC× Bay of Plenty, ASC× Waikato and ASC× North Island are all significantly positive, which implies that, geographically, residents in the upper North Island regions would likely be more welcoming of a local council biodiversity initiative compared with South Island respondents.

Using a series of Wald tests, we were also able to compare the coefficient for ASC×Wellington with the coefficients for ASC×Bay of Plenty, ASC×Waikato and ASC×North Island. The three Wald tests gave chi-square statistics above the critical values

12 0.1 0.1 for ASC×Bay of Plenty (χ1 = 7.96), ASC×Waikato (χ1 = 9.46) and ASC×North Island 0.1 ( χ1 = 6.04). These results indicate that the respondents in the three upper North Island regions would likely be more receptive of a private land biodiversity programme compared to the respondents in the Greater Wellington region.

Table 4. Estimates for the multinomial logit model with Socio-Demographic Characteristics Model 2 (With Socio-Demographics) With-Protest Sample No-Protest Sample Coefficient Part-worth Coefficient Part-worth ASC -0.612 -1.544 (0.170) (0.002) Price -0.019 -0.029 (0.000) (0.000) Native 1.855 $ 98.73 2.056 $ 70.59 (0.000) (0.000) Mixture 1.999 $ 106.44 1.973 $ 67.77 (0.000) (0.000) Rebate 0.244 $ 13.01 0.314 $ 10.79 (0.013) (0.002) ASC*Years in Property 0.021 $ 1.12 0.025 $ 0.86 (0.023) (0.010) Price*Education 0.003 $ 0.14 0.001 $ 0.04 (0.037) (0.512) Price*Volunteer 0.009 $ 0.46 0.010 $ 0.34 (0.000) (0.000) Price*Rural 0.004 $ 0.19 0.006 $ 0.19 (0.068) (0.056) ASC*Wellington 0.246 $ 13.10 0.233 $ 8.00 (0.617) (0.641) ASC*Bay of Plenty 1.265 $ 67.35 1.221 $ 41.94 (0.010) (0.014) ASC*Waikato 1.308 $ 69.65 1.267 $ 43.50 (0.007) (0.010) ASC*North Island 1.138 $ 60.60 1.156 $ 39.69 (0.021) (0.022)

Pseudo R2 0.0509 0.0971 Adjusted R2 0.0460 0.0906 Log-likelihood -1088.54 -871.39 No. of observations 807 767 Note: The indicator variable for the South Island region serves as the reference variable. Figures in boldface font are significant at the 90% confidence level or greater. Figures in parentheses are p-values.

13 We then check for differences between the coefficients of the three other North Island regional indicator variables using a Wald test. We obtained chi-square values of 0.02, 0.12, and 0.24, which indicate that the choice behaviour between the three other northern regions were not statistically different from one another. This scenario is consistent with their part worth values being virtually the same (Bay of Plenty = $42; Waikato = $44; and North Island = $40).

To examine if there were heterogeneity in respondents’ preferences, we ran RPL models. The RPL models followed the procedure described in Hensher et al. (2005, p. 632- 634). In this procedure, we initially identified which attributes (excluding price) and socio- demographic variables could be classified as random parameters. To accomplish this, a series of RPL regressions including all the attributes and SDC variables were conducted. We assigned different distributional forms to each explanatory variable to check which distribution (e.g., normal, triangular, uniform, log-linear) would provide the best regression model fit. From these regressions, variables found to have random parameters with significant standard deviation estimates (or demonstrated heterogeneity) were included in the final RPL models. We used the standard Halton sequence (SHS) with 200 random draws (Hensher et al., 2005).

For the RPL models, we used only the no-protest sample, since the MNL results point to this sample as having a better model fit. We estimated two RPL models, Model 1, which represents the restricted model (only has attributes as explanatory variables), and Model 2 was also called the unrestricted model, since it includes both attributes and SDC variables. Overall, estimates for both RPL Models 1 and 2 were found to be significant as indicated by the Chi-square values of 283.13 and 337.52, respectively. Chi-square values for RPL Models 0.10 0.10 1 and 2 far exceeded the critical Chi-square values of χ 6 = 10.64 and χ17 = 24.77 , respectively (Table 5). In addition, the pseudo-R2’s of RPL Models 1 and 2 of 0.1331 and 0.1634 are also higher than MNL Models 1 and 2 with 0.0747 and 0.0971, respectively. The above scenario implies that RPL models yield a better model fit than MNL models.

Despite the differences in the goodness of fit and model significance between the RPL and MNL models, almost all coefficient estimates and significance levels are very similar. The signs of the coefficients, as well as the magnitude of coefficient estimates, for the restricted MNL (no-protest sample) and the restricted RPL are virtually the same (Tables 3 and 5). In terms of the unrestricted models, the signs of the coefficients for both MNL and RPL are also the same. The only major difference between the two regression models is the estimate for the random parameter for the mean and standard deviation of the interaction variable price × education , which was positive and significant. Under the MNL model, the coefficient estimate for this interaction variable is positive, but not significant. The significantly positive coefficient of 0.016 for price × education implies that residents with a higher educational attainment tend to be more willing to choose the planting scheme with a relatively higher price. This was not captured in the MNL, since it did not account for

14 heterogeneity. The RPL model, which relaxed the IID assumption, allows the error components of different alternatives to be correlated, which accounts for heterogeneity.

Table 5. Random parameter logit results for sample with no protest responses Model 1 Model 2

(Attributes Only) (With Socio-Demographics) Parameter Standard Parameter Standard

Estimate Deviation Estimate Deviation ASC -0.437 -5.319 . (0.132) (0.004) Price -0.019 -0.086 (0.000) (0.002) Native 2.278 2.972 (0.000) (0.000) Mixture 2.126 3.128 (0.000) (0.000) Rebate 0.305 0.816 0.280 1.499 (0.006) (0.076) (0.052) (0.042) Price*Education 0.016 0.018 (0.036) (0.011) Price*Rural 0.016 0.032 (0.073) (0.104) ASC*Years in Property 0.039 (0.049) Price*Volunteer 0.022 (0.014) ASC*Wellington 0.952 (0.317) ASC*Bay of Plenty 2.585 (0.016) ASC*Waikato 2.721 (0.015) ASC*North Island 3.000 (0.015) Pseudo R2 0.1331 0.1634 Log-likelihood -921.72 -864.03 Chi-square value 283.13 337.52 No. of observations 767 745 Note: The indicator variable for the South Island region serves as the reference variable. Figures in boldface font are significant at the 90% confidence level or greater. Figures in parentheses are p-values.

Discussion and Conclusions

Although New Zealand was reported to have one of the world’s highest levels of threatened species (Hitchmough et al., 2007), New Zealanders increasingly participate in contributing to biodiversity enhancement for future generations with the intentions of preserving economic well-being and cultural wealth (MfE, 2007). We therefore tried to determine if there is an

15 incentive for a typical NZ resident to participate in government initiatives toward biodiversity enhancement on private land. In this study, we also attempted to ascertain if a typical resident finds biodiversity enhancement on private lands valuable. Our data suggests that government initiated biodiversity enhancement programmes using tree planting on private properties are valued by our sample of NZ residents. We found that our typical respondent would be better off in terms of welfare if there would be a biodiversity enhancing scheme on residential properties. This is exhibited by the part worth values of $120 for providing native trees to plant and $112 for providing a mixture of natives and non-natives to plant.

Our sample of residents would be willing to participate in the biodiversity scheme provided that this scheme would have the preferred attributes. From this choice modelling exercise, the two most preferred attributes were having purely native trees and having a mixture of natives and non-native trees. This represents a bimodal distribution of preference wherein there is a group who prefers to have purely native trees on their respective properties, while another group prefers a combination of natives and non-natives. The group who strictly prefers purely natives might have the notion that NZ native trees are very well adapted to the NZ environment and they do not have the tendency to become invasive or hard to control. Therefore, they consider native trees as the best strategy for enhancing indigenous biodiversity on private lands. The other group of respondents recognized that some non- natives can serve as good complements for natives. Perhaps they believe that some non- natives can have high aesthetic values (e.g., colourful leaves and flowers) and can also provide food and shelter for native animals (Salmon, 2003). On the other hand, purely non- natives were preferred by the least proportion of the respondents. This may be because some introduced plant species in NZ eventually became invasive (Weir, 2006; Pimentel, 2002; Froude, 2002; Environment Waikato, 2002; Fowler and Syrett, 2000).

Another feature of a biodiversity enhancing scheme valued by residents was having a rebate for purchasing trees they will plant to enhance biodiversity on their properties. This feature seems to provide residents with the freedom to choose, and/or flexibility to purchase, the type of trees they like. The part worth value of this flexibility was approximately $11. Respondents also preferred to have the least expensive alternative, where they would be able to plant trees on their property. Results show that the current situation (or status quo alternative), wherein local councils provide only free advice but no free trees to residents, appears to be insufficient. A typical resident seemed to have the need to be provided with both advice and subsidised trees to better encourage them to support biodiversity enhancement through the planting of trees. Ninety-seven (97) respondents who chose alternatives, which included additional trees on their property, shared their reasons why private properties should have trees. These reasons included: trees are good for the environment; trees attract birds and provide them shelter; trees attract native animals; trees provide aesthetic benefits; and their properties have space for additional trees.

This study examined how a set of identified characteristics or attributes of a hypothetical biodiversity enhancement programme influenced the preference behaviour of a sample of NZ residents. This study is hoped to serve as a starting point for future valuation

16 studies of biodiversity enhancement on private lands. Future CM studies may further examine the preference behaviour of household residents given other characteristics of trees to be planted, such as maximum height, canopy size and the amount of food that can be provided to native birds. Through focus group meetings and experts’ consultation, other relevant characteristics attributes can be identified. Other future studies might further investigate the preferences of rural and urban residents towards biodiversity on private land.

One limitation of this choice modelling exercise was that we developed our choice modelling scenario and bid design instrument based on experts’ opinion, what councils desired to know, and focus group meetings with respondents, to make the information more applicable to the councils we were working with. Construction may have been easier if we used the experimental design method (e.g., full factorial, fractional factorial, d-optimal) because there are several software programmes that could aid in developing the experimental designs for choice modelling (e.g., SPSS, GAUSS, SAS, MATLAB) (Johnson et al., 2007, Hensher et al., 2005; Burgess and Street, 2003; Kanninen, 2002). This was an important decision. We decided on the method we used specifically for our end users’ benefit.

References

Bateman, I.J., Carson, R.T., Day, B., Hanemann, M., Hanley, N., Hett, T., Jones-Lee, M., Loomes, G., Mourato, S., Özdemiroglu, E., Pearce, D.W., Sugden, R., Swanson, J., 2002. Economic valuation with stated preference techniques: A Manual, Edward Elgar, Ltd., Cheltenham, 458 pp.

Bennett, J., 2006. Choice modelling and the transfer of environmental values, in: Rolfe, J., Bennett, J., (Eds.), Choice Modelling and the Transfer of Environmental Values. Edward Elgar, Cheltenham, pp. 1-9.

Bennett, J., 2003. The economic value of biodiversity: a scoping paper. Paper presented to the national workshop The Economic Value of Biodiversity, 22-23 October 2003. Available online on 6 June 2008 at http://www.deh.gov.au/biodiversity/publications/index.html.

Brander, L., Florax, R.J.G.M., Vermaat, J., 2006. The empirics of wetland valuation: A comprehensive summary and a meta-analysis of the literature. Environmental and Resource Economics 33, 223-250.

Brouwer, R., Langford, I.H., Bateman, I.J., Crowards, T.C., Turner, R.K., 1999. A meta-analysis of wetland contingent valuation studies. Regional Environmental Change 1, 47–57.

Buckley, R., 2002. Private partnerships between tourism and protected areas: The Australian situation. The Journal of Tourism Studies 13, 26-38.

Burgess, L., Street, D., 2003. Optimal designs for 2 choice experiments. Communications in Statistics – Theory and Methods, 32.

Christie, M., Hanley, N., Warren, J., Murphy, K., Wright, R., Hyde, T., 2006. Valuing the diversity of biodiversity. Ecological Economics 58, 304-317.

17 Convention on Biological Diversity (1998) Conference of the Parties 4 Decision IV/10Bratislava, 4-15 May 1998. Available online on 9 June 2008 at http://www.cbd.int/decisions/cop- 04.shtml?m=COP-04&id=7133&lg=0

Daily, G.C., 2000. Management objectives for the protection of ecosystem services. Environmental Science and Policy 3, 333-339.

Dale, V.H., Brown, S., Haeuber, R.A., Hobbs, N.T., Huntly, N., Naiman, R.J. Riebsame, W.E., Turner, M.G., Valone, T.J., 2000., Ecological principles and guidelines for managing the use of land: a report from the Ecological Society of America. Ecological Applications 10, 639-670.

Dickson, R., Campbell, A., Buckley, R., Bennett, F., 2005. Making economic valuation work for biodiversity conservation. Land and Water Australia.

Environment Waikato, 2002. Waikato Regional Pest Management Strategy - Operative 2002 to 2007. Environment Waikato Policy Series 2002/06. Hamilton, New Zealand.

Environment Waikato, 2008a. Amendment to 2006-2016 Long-Term Council Community Plan Biosecurity Rate. Environment Waikato Regional Council, Hamilton, New Zealand.

Environment Waikato, 2008b. Envirocare: Your Environmental News Update from Environment Waikato. Environment Waikato Regional Council, Hamilton, New Zealand.

Fowler, S. V., Syrett, P., 2000. “Status of broom in New Zealand.” Plant Protection Quarterly, 15(4): 148.

Froude, V.A., 2002. Biological control options for invasive weeds of New Zealand protected areas. Science for Conservation 199, Department of Conservation, Wellington, New Zealand.

Haab, T.C., McConnell, K.E., 2002. Valuing environmental and natural resources: the econometrics of non-market valuation. New Horizons in Environmental Economics. Edward Elgar, Massachusetts, 326 pp.

Hanley, N., Mourato, S., Wright, R.E., 2001. Choice modelling approaches: A superior alternative for environmental valuation? Journal of Economic Surveys 15, 435-62.

Hensher, D.A., Rose, J.M., Greene, W.H., 2005. Applied Choice Analysis: A Primer. Cambridge University Press, New York.

Hitchmough, R, Bull L, Cromarty, P., (compilers), 2007. New Zealand Threat Classification System Lists. Accessed online on 26 July 2007 at: http://www.doc.govt.nz/upload/documents/science- andtechnical/sap236.pdf

Johnson, F.R., Kanninen, B., Bingham, M., Özdemir, S., 2007. Experimental Design for Stated Choice Studies, in: Kanninen, B.J. (Ed.) Valuing Environmental Amenities Using Stated Choice Studies. Springer, Dordrecht, pp. 159-202.

Kanninen, B., 2002. Optimal design for multinomial choice experiments. Journal of Marketing Research 39, 214-227.

Kerr, G.N., Sharp, B.M.H., 2007. The Impact of Wilding Trees on Indigenous Biodiversity: A Choice Modelling Study. Research Report No. 303, Agribusiness & Economics Research Unit, Lincoln University, New Zealand.

18 Kneebone, J., 2000. Biodiversity and Private Land. Final Report of the Ministerial Advisory Committee on Biodiversity and Private Land. Ministry for the Environment, Wellington.

Louviere, J.J., Hensher, D.A., Swait, J.D., 2000. Stated choice models: Analysis and application. Cambridge University Press, Cambridge.

MfE (Ministry for the Environment), 2007. Environment New Zealand 2007. Ministry for the Environment, Wellington, New Zealand. Morrison, M., Bergland, O., 2006. Prospects for the use of choice modelling for benefit transfer. Ecological Economics 60, 420–428.

Norton, D.A., 2000. Conservation biology and private land: shifting the focus. Conservation Biology 14, 1221-1223.

Noss, R.F., O'Connell, M.A., Murphy, D.D., 1997. The science of conservation planning: habitat conservation under the Endangered Species Act. Island Press, New York, New York, USA.

Nunes, P.A.L.D., van den Bergh, J.C.J.M., 2001. Economic valuation of biodiversity: sense or nonsense? Ecological Economics 39, 203– 222.

Othman, J., Bennett, J., Blamey, R. 2004. Environmental values and resource management options: a choice modelling experience in Malaysia. Environment and Development Economics 9, 803–824.

Pascual, U., Perrings, C.P., 2007. The economics of biodiversity loss in agricultural landscapes. Agriculture Ecosystems and Environment 121, 256–268.

Pearce, D., 2001. Valuing biological diversity: issues and overview. In: OECD: Valuation of Biodiversity Benefits; Selected Studies. Paris, OECD, pp. 27-44.

Pimentel, D., 2002. Biological Invasions: Economic and Environmental Costs of Alien Plant, Animal, and Microbe Species. CRC Press, New York. 384p.

Revelt, D., Train, K., 1998. Mixed logit with repeated choices: households’ choices of appliance efficiency level. Review of Economics and Statistics 80, 647-657.

Rolfe, J., Windle, J., 2005. Valuing options for reserve water in the Fitzroy Basin. The Australian Journal of Agricultural and Resource Economics, 49: 91-114.

Rolfe, J., 2006. A simple guide to choice modelling and benefit transfer, in: Rolfe, J., Bennett, J., (Eds.), Choice Modelling and the Transfer of Environmental Values. Edward Elgar, Cheltenham, pp. 10-27.

Salmon, P., 2003. The Reed Field Guide to Exotic Trees in New Zealand: Broadleaves and Conifers. Reed Publishing, U.K. 207 pp.

Saunders, A., Norton, D.A., 2001. Ecological restoration at Mainland Islands in New Zealand. Biological Conservation 99, 109-119.

Schou, J.S., Hasler, B., Nahrstedt, B., 2006. Valuation of biodiversity effects from reduced pesticide use. Integrated Environmental Assessment and Management 2, 174–181.

Statistics NZ. 2006. New Zealand Rural Residents. www.stats.govt.nz

19 Stoneham, G., Crowe, M., Platt, S., Chaudhri, V., Soligo, J., Strappazzon, L., 2000. Mechanisms for biodiversity conservation on private land. Victoria Natural Resources and Environment, Victoria, Australia.

Theobald, D.M., Hobbs, N.T., 2002. A framework for evaluating land use planning alternatives: protecting biodiversity on private land. Conservation Ecology 6(1): 5. Available online on 27 May 2008 at URL: http://www.consecol.org/vol6/iss1/art5

Watkinson, A., Freckleton, R. P., Dowling, P. M., 2000. Weed invasions of Australian farming systems: from ecology to economics, in: Perrings, C., Williamson, M., Dalmazzone, S., (Eds.), The Economics of Biological Invasions. Edward Elgar Publishing, Cheltenham, pp. 94-116.

Weir, B., 2006. Systematics, specificity, and ecology of New Zealand Rhizobia. PhD thesis. School of Biological Sciences. The University of Auckland.

Whitten, S., Bennett, J., 2005. Managing Wetlands for Public and Social Good. Edward Elgar, Cheltenham.

Woodward, R.T., Wui, Y.S., 2001. The economic value of wetland services: A meta-analysis. Ecological Economics 37, 257–270.

Yao, R., Kaval, P., 2008. Valuing Biodiversity Enhancement in New Zealand. Department of Economics Working Paper in Economics 08/06. University Waikato Management School, Hamilton, New Zealand.

Yao, R., Kaval, P., 2007. Non market valuation in New Zealand: 1974 through 2005. Working Papers in Economics 07/17, University of Waikato Department of Economics, Hamilton, New Zealand.

20 Ecological Restoration in Hamilton City, New Zealand

Ecological Restoration in Hamilton City, North Island, North New Zealand

Bruce D. Clarkson1 and Joanna C. McQueen1

Abstract

Hamilton City (New Zealand) has less than 20 hectares of high-quality, indigenous species dominated ecosystems, and only 1.6% of the original indigenous vegetation remains within the ecological district. A gradual recognition of the magnitude of landscape transformation has gathered momentum to the stage that there is now a concerted public and Figure 1. Aerial view of Hamilton City (showing the private effort to retrofit the City by restoring and Waikato River) and its location within the centre of reconstructing indigenous ecosystems. The initial northern New Zealand. (Photo courtesy of Hamilton focus was on rehabilitating existing key sites, but has City Council). shifted to restoring parts of the distinctive gully landform that occupies some 750 ha or 8% of the Bush (Fig. 2), floristically the richest of the Hamilton City. A new initiative at Waiwhakareke (Horseshoe indigenous remnants. Despite its small size (1 hectare), Lake) will involve reconstruction from scratch of a it supports an impressive 145 native plant species (de range of ecosystems characteristic of the ecological Lange 1996) and is regularly visited by Hemiphaga district over an area of 60 ha. This address will novaeseelandiae novaeseelandiae (kereru). Recently, examine a vision for ecological restoration in native Chalinolobus tuberculatus (long-tailed bats) have Hamilton City within the context of policy, also been recorded in the area (E. Ganley pers. comm. education, and community dimensions that have 2001). However, the widespread native nectar-feeding triggered a shift from traditional parks and gardens bird, Prosthemadera novaeseelandiae novaeseelandiae management to ecosystem management. (tui) (Fig. 3) is only a rare visitor to Hamilton City.

Introduction

The Hamilton Ecological District (159,376 ha) in the northern North Island of New Zealand is one of the most modified districts in New Zealand with only 1.6% of the indigenous vegetation remaining (Leathwick et al. 1995). At least 20% of its indigenous flora is threatened or extinct and more than one half of its indigenous bird species have gone. Hamilton City (9427 ha) (Fig. 1), in the centre of this transformed landscape, has only a few tiny remnants of the former indigenous forest cover, perhaps less than 20 hectares in total of high quality indigenous habitat. The largest remnant is Jubilee Park

(Claudelands Bush), a 5.2 hectare reserve comprising Figure 2. Hammond Bush, Waikato Riverbank, Dacrycarpus dacrydioides (kahikatea) forest (Whaley et Hamilton. Adjacent is the mouth of a small gully al. 1997). Another important remnant is Hammond system (part of the Mangaonua system), crossed by a 1.5m wide boardwalk. Indigenous evergreen vegetation 1 Centre for Biodiversity and Ecology Research, (appearing as dark green) is clearly distinguished from Department of Biological Sciences, The University of the deciduous exotic vegetation (mainly Salix cinerea). Waikato, Private Bag 3105, Hamilton, New Zealand

16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 1 Ecological Restoration in Hamilton City, New Zealand

Figure 3. The widespread native nectar-feeding bird, Prosthemadera novaeseelandiae novaeseelandiae (tui), sitting on Phormium tenax (flax). Photo courtesy of Max McRae.

Although the indigenous biodiversity resource is Figure 4. Gully systems of Hamilton City. The four very limited, Hamilton City does possess an extensive major systems (Kirikiriroa, Mangakotukutuku, network of gullies. These extend from the Waikato Mangaonua and Waitawhiriwhiri) are labeled. Peat River through many suburbs of the City, occupying Lakes and the Waikato River are also shown in blue. around 750 hectares or 8% of the City area (Downs et al. The city area is 9427ha. 2000) and are considered a unique feature of the Hamilton area (McCraw 2000). Four major gully systems (Kirikiriroa, Mangakotukutuku, Mangaonua and Waitawhiriwhiri) and numerous minor systems exist (Fig. 4). As McCraw (2000) has shown, the gullies are the result of the undermining of a geological formation of sand, silt, peat and gravel known as the Hinuera formation. Around 15,000 years ago, the Waikato River started to cut down through this material creating its present channel and as it deepened, springs were exposed along the riverbanks. As water drained from the surrounding land, these springs undermined the banks, in a process known as spring sapping, causing slips and creating a network of streams draining into the Waikato River. This process was repeated again and again giving rise to erosion and the formation of the steep-sided and intricate network of gullies that adjoin the river today. Despite being poorly treated in the past, Figure 5. Urban development adjacent to the the gullies have now been recognised (Clarkson & Mangakotukutuku gully system. Downs 2000) as the central focus of a city-wide restoration of indigenous ecosystems because they are Restoration and Policy Background the main wildlands in an otherwise entirely built landscape (Fig. 5). The significance of the gullies as a potential focus for landscape enhancement was highlighted as early as 1972 (McLeary 1972) but only in recent years has the move to make use of them gathered momentum. A Gully Protection Zone had been in force since 1989 but this related more to building requirements than protection of gullies per se and was ineffective at

16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 2 Ecological Restoration in Hamilton City, New Zealand protecting ecological features or preventing infilling of gullies associated with subdivision development. In recent years, to meet the requirements of the Resource Management Act (1991) and obligations of international agreements such as Agenda 21, the Hamilton City Council has undertaken a substantial information gathering exercise to improve knowledge of natural values within the City. This has included an evaluation of a range of ecological, landscape and hazard areas, and the preparation of a map based on the boundaries of these areas, termed an Environmental Protection Overlay (EPO). This map forms part of the District Plan, which also sets out policies and rules relating to components or layers of the EPO, developed in consultation with the community (Vare 2000). As well, Figure 6. Gully profile with locations of native plant a Gully Management Plan (2001) has been developed species (from Wall and Clarkson 2001). providing a comprehensive plan for restoration of the publicly owned gullies but recognising the need to also The benefits of gully restoration are numerous and involve the owners of adjoining privately owned gullies. wide-ranging. These include improved environmental, Investigation of ecological resources on public and aesthetic, scenic and cultural values. Native plants private lands included a comprehensive survey, in order perform important ecosystem functions. The nectar- to identify key ecological sites within the City (Downs producing Sophora microphylla (kowhai) flowers feed et al. 2000). Some 67 key sites were identified with a native birds, the fleshy fruits of the native conifers like total area of 76 ha and an average area of 1.1 ha. Extant Dacrydium cupressinum (rimu) feed the birds and in vegetation ranges from Kunzea ericoides (kanuka) forest turn are spread by them, and Hoheria sexstylosa of well-drained river and gully scarps to Typha (lacebark) grows into a good-sized nurse tree within five orientalis (raupo) reedlands fringing peat lakes, with years and flowers profusely every autumn. many intermediate types. Gully floor vegetation is Reintroducing a range of plant species once found in frequently dominated by the deciduous exotic tree Salix Hamilton gullies could also address the local shortage of cinerea (grey willow), though beneath this is often an native plant resources for rongoa (traditional Maori understorey dominated by indigenous plants including medicine) (McGowan 2000). As well as enhancing ferns, Melicytus ramiflorus (mahoe) and Cordyline terrestrial habitats, gully restoration benefits aquatic life australis (cabbage tree). Even where weeds are in streams. Riparian planting leads to cooler, more dominant, gullies still provide some important shaded streams with stable undercut banks - the ecosystem services, such as supporting desirable preferred habitat of some native fish species including wildlife. Native birds persist even in these highly Galaxias fasciatus (banded kokopu) and Gobiomorphus modified systems, with seven species, including Ninox huttoni (redfin bullies) (Hicks 2000). novaeseelandiae novaeseelandiae (morepork), Halcyon From the work of restoration pioneers, we know that sancta vagans (New Zealand kingfisher), Zosterops it is possible to have a good canopy cover of native trees lateralis lateralis (silvereye), Rhipidura fuliginosa established in a gully setting within 15-20 years (Fig. 7). placabilis (fantail) and Gerygone igata (grey warbler), But there is still much to accomplish. Prosthemadera widespread (Innes 2000). novaeseelandiae novaeseelandiae (Tui) are an icon for The original gully vegetation can be deduced from restoration success and a comparison with other North historical records, from the composition of extant Island cities showed it may be necessary to have almost remnants and from macrofossil deposits (Clarkson & 100 hectares of quality habitat within Hamilton City, or Clarkson 2000). This information provides a context for 1000 hectares within 10 kilometres of the City, to developing restoration goals and gives an indication of support resident tui. Increasing the current area of the diverse range of native plants appropriate for habitat towards these threshold figures would at least restoration plantings. Publications such as the Gully result in more regular tui visitors than at present. Restoration Guide (Wall & Clarkson 2001) and Botany of the Waikato (Clarkson et al. 2002) provide information on which plants to use, where to plant them (Fig. 6) and how to look after them.

16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 3 Ecological Restoration in Hamilton City, New Zealand

investments made in restoration projects can be very quickly compromised. Increased use of indigenous species in a range of settings, such as hedge plants, specimens in parks, and as integral components of home gardens, will positively influence and buffer the restoration projects being undertaken. For example, a survey of Cordyline australis (cabbage trees) growing in gardens and gullies in a suburb of Hamilton City showed that the planted garden trees were the most likely invasion source of the populations which had developed in the adjacent Kirikiriroa Gully (Clarkson 1999). The land-use change from rural (grazed) to City subdivision thus provided the opportunity for some native species to recolonise the gullies. Recent widespread establishment of indigenous plants along road ways and the ever expanding area of restoration plantings is therefore likely to increase the

Figure 7. The successfully restored gully of G. and A. opportunities for indigenous species to reinvade areas, Edgar, Hamilton. Restoration has been ongoing for 25 including those areas currently dominated by exotic years and has achieved an excellent canopy cover free of plants. weed species. Cordyline australis dominates the centre of the photo and there is a large specimen of Dacrycarpus dacrydioides behind. Wider Restoration Plans

Successful Gully Restoration It is apparent merely by inspecting a map of Hamilton City that the restoration of gullies cannot be The key ingredients for successful gully restoration conducted in isolation from other restoration projects. identified by Clarkson & Downs (2000) and Morris Not only do the individual gullies need to be restored (2000) are summarised here. It is sensible to use but they need to be linked to the Waikato River, the peat existing remnants or key sites with a significant lakes and the extant lowland forest remnants. Thinking indigenous component as nuclei for restoration projects. even more widely, the potential exists to establish Buffering issues and the development of corridors and further linkages and commence a regional-scale linkages can then be considered. A range of different restoration, by the establishment of riparian plantings or restoration strategies from complete weed clearance and corridors along the Waikato River to link with forest in replanting of the site to canopy manipulation of Salix the north and south. cinerea (willow) and other exotics is available. The The Gully Restoration Guide (Wall & Clarkson strategy adopted will vary depending on the budget 2001) and the Hamilton City Council Gully Reserves available, access to volunteer labour, and the need to Management Plan (2001) provide the information base protect extant specimens of native plants. The value of on which to commence a city-wide restoration, and there ecosourcing rests mainly in the superior performance of is an ever-increasing number of community group, locally sourced ecotypes or races, as well as the need to school and private gully restorations. A recent advent conduct authentic restoration in important reserves such has been the first advertisement for a housing as Hammond Bush. Native species propagated from subdivision proclaiming a gully enhancement approach. ecosourced plant material have been used successfully As at October 2003, some 187 ha of land in Hamilton in numerous restoration projects throughout the City. City is under restoration of some form, including 142 ha Matching species preferences to site conditions on public land and 45 ha on private land. The Hamilton through careful site selection is rewarded by greater City Council Community Planting Coordinator, has, in survival and growth of the species planted. There are 2003, supervised the planting of more than 25,000 also considerable benefits obtained by mimicking indigenous plants by the council and community groups. natural succession processes and it is necessary to A gully restoration database holds information on consider enrichment planting or spreading of seed of almost 500 gully properties where restoration is later successional species once a good canopy cover of underway or the owners have expressed an interest in trees has been established. Regular aftercare and weed beginning a restoration project. Gully restoration and pest control is all-important for the continuing seminars and practical workshops, native plant give- success of a restoration project. Controlling weeds and aways, a gully restoration newsletter and funding for pests can be expensive but without this management the school or community led projects are all incentives which have been used to promote the city-wide

16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 4 Ecological Restoration in Hamilton City, New Zealand restoration of gullies. To provide the ecological science Conclusion needed to underpin the restoration efforts, the University of Waikato has initiated several research projects. These By bringing together ecological understanding and include documenting the biodiversity values of key sites, the best-practice techniques of pest management, native assessing the success of 5-40 year old restoration plant propagation, planting and animal breeding and plantings, and seeking solutions to specific problems, recovery programmes, successful city-wide restoration for example pest fish and the weed Solanum is achievable. While it is important to consider the mauritianum (woolly nightshade). broader long-term restoration task, the higher-level Because of far-sighted decision making by the goals are best attained through a series of smaller, Hamilton City Council, there are currently unique manageable stages. Changing ecosystem dominance opportunities to considerably advance the wider from exotic deciduous trees to indigenous evergreen restoration plan. A Natural Heritage Park trees will assist but not completely solve some of the (Waiwhakareke, Horseshoe Lake) will be developed on current weed problems. Shifting the local seed rain some 60 ha of land owned by the Hamilton City balance from exotic to indigenous brings us closer to the Council, with five ecosystems representing a major threshold where it will be more likely that regeneration proportion of the original diversity of the Hamilton and recruitment will be of indigenous species rather than Ecological District being reconstructed. The five ubiquitous exotics. Consideration of all components of ecosystems have been delineated (McQueen & Clarkson the ecosystem, not just the trees, promotes a fully 2003) on the basis of the varied soils, topography and functioning restored ecosystem. Successful restoration water table of the area (Fig. 8). Over the next 20 years, projects will be more likely within the developing city- this project will significantly increase the area of wide strategy based on community support, indigenous vegetation within the City and provide a empowerment and partnerships. A convergence and valuable stepping-stone for native birds in the stretch of skills and technology is taking the City from a northwestern quadrant of the City, only 11 km from traditional parks, gardens and utilities management extensive indigenous forest habitat. Within this same approach into the realms of ecosystem management. time-span, there is also the possibility of erecting a pest Restoration or retrofitting within the City will proof fence around some areas, protecting native birdlife significantly contribute to an ecological transformation and plants from mammalian pests such as Rattus rattus, of the Waikato Region. Rattus norvegicus (rats), Mustela ermina (stoats), Mustela furo (ferrets), Trichosurus vulpecula (possums), Acknowledgements Erinaceus europaeus occidentalis (hedgehogs) and Oryctolagus cuniculus (rabbits), once these have been We gratefully thank the staff of the Hamilton City eradicated within. We intend to use the opportunities Council, in particular the Strategic Unit, Parks and provided by this project to conduct fundamental Gardens Unit and Sustainable Environment Team for information and support. Members of Tui 2000, ecological research on assembly rules for reconstructing ecosystems and to provide practical training for HEIRS, RAFT, Ecologic Foundation and many other undergraduate restoration ecology students. city environmental groups have also willingly supported this work. Staff and students of CBER have assisted in N the field and in the preparation of this paper. Beverley Clarkson and Carolyn King provided useful comments on the manuscript.

LITERATURE CITED

Clarkson, B.D. and Downs, T.M. 2000. A vision for the restoration of Hamilton Gullies. Pages 48-56 in Clarkson, B.D., McGowan, R., and Downs, T.M., editors. Hamilton Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City Council, 29-30 April 2000, Hamilton. The University of Waikato. Clarkson, B., Merrett, M. and Downs, T. 2002. Botany Figure 8. The proposed ecosystem reconstruction of of the Waikato. Waikato Botanical Society Inc., Waikwhakareke (60ha within the city). Hamilton. Kauri forest (ridge crest) Clarkson, B.R. and Clarkson, B.D. 2000. Indigenous Rimu/tawa forest (hillslope) vegetation types of Hamilton City. Landcare Kahikatea-pukatea forest (semi-swamp) Research File Note. Hamilton: Landcare Research. Harakeke (lake margin) Peat Lake/aquatic habitat 16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 5 Ecological Restoration in Hamilton City, New Zealand

Clarkson, F. 1999. Cabbage tree comeback. Rotorua Council, 29-30 April 2000, Hamilton. The Botanical Society Newsletter 32:17-24. University of Waikato. de Lange, P.J. 1996. Floristics and microclimate of Vare, M. 2000. Hamilton’s gullies – a policy response. Hammond Bush, a Hamilton Basin forest remnant. Pages 42-46 in Clarkson, B.D., McGowan, R., and Wellington Botanical Society Bulletin 47:65-80. Downs, T.M., editors. Hamilton Gullies – A Downs, T.M., Clarkson, B.D. and Beard, C.M. 2000. workshop hosted by The University of Waikato and Key Ecological Sites of Hamilton City. CBER sponsored by the Hamilton City Council, 29-30 Contract Report No. 5. Centre for Biodiversity and April 2000, Hamilton. The University of Waikato. Ecology Research, Department of Biological Wall, K., and Clarkson, B.D. 2001. Gully restoration Sciences, The University of Waikato, Hamilton. guide – A guide to assist in the ecological Hamilton City Council. 2001. Gully Reserves restoration of Hamilton’s gully systems. Hamilton Management Plan. Hamilton City Council, City Council, Hamilton. Hamilton. Whaley, P.T., Clarkson, B.D., and Smale, M.C. 1997. Hicks, B.J. 2000. Streams and rivers: aquatic life in Claudelands Bush: ecology of an urban kahikatea gully streams. Pages 20-26 in Clarkson, B.D., (Dacrycarpus dacrydioides) forest remnant in McGowan, R., and Downs, T.M., editors. Hamilton Hamilton, New Zealand. Tane 36:131-155. Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City Council, 29-30 April 2000, Hamilton. The University of Waikato. Innes, J. 2000. Birdlife, animal pests and their control. Pages 29-30 in Clarkson, B.D., McGowan, R., and Downs, T.M., editors. Hamilton Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City Council, 29-30 April 2000, Hamilton. The University of Waikato. McLeary, W.H. 1972. A study of the gully systems of the Waikato Basin with particular reference to those in and surrounding the City of Hamilton. Unpublished dissertation for the Diploma of Landscape Architecture in the University of Canterbury (Lincoln College). McCraw, J.D. 2000. Geology of Hamilton gullies. Pages 5-8 in Clarkson, B.D., McGowan, R., and Downs, T.M., editors. Hamilton Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City Council, 29-30 April 2000, Hamilton. The University of Waikato. McGowan, R. 2000. Plants for rongoa: traditional Maori medicine. Pages 27-29 in Clarkson, B.D., McGowan, R., and Downs, T.M., editors. Hamilton Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City Council, 29-30 April 2000, Hamilton. The University of Waikato. McQueen, J.C. and Clarkson, B.D. 2003. An Ecological Restoration Plan for Waiwhakareke (Horseshoe Lake), Scoping Report for Hamilton City Council. CBER Contract Report No. 29. Centre for Biodiversity and Ecology Research, Department of Biological Sciences, The University of Waikato, Hamilton. Morris, P. 2000. Restoring gullies: reflecting on experience. Pages 30-41 in Clarkson, B.D., McGowan, R., and Downs, T.M., editors. Hamilton Gullies – A workshop hosted by The University of Waikato and sponsored by the Hamilton City

16th Int’l Conference, Society for Ecological Restoration, August 24-26, 2004, Victoria, Canada 6

Hamilton City Council

Esplanade Feasibility Assessment of Mangaonua Stream & Margins - 155 Riverlea Rd

RIVERLEA ROAD ESPLANADE RESERVE FEASILBLITY ASSESSMENT 2

Table of Contents

1 Introduction ...... 3 1.1 Brief...... 3 1.2 Methods...... 3 2 Description of Ecological Features ...... 5 2.1 Landforms...... 5 2.2 Vegetation Description...... 5 2.2.1 Secondary broadleaved forest (MTF) ...... 5 2.2.2 Secondary small-leaved forest (K) ...... 5 2.2.3 Exotic forest (ET, ES, B, W, Pr)...... 8 2.3 Fauna...... 8 2.3.1 Bats...... 8 2.3.2 Birds...... 9 2.3.3 Fish ...... 9 2.4 Threats...... 9 2.4.1 Animals ...... 9 2.4.2 Plant Pests...... 9 3 Ecological Significance Assessment...... 9 3.1 Whaley Ecological Significance Assessment...... 9 3.2 Compliance with the EW Regional Policy Statement Criteria ...... 11 4 Esplanade and Access Assessment...... 12 4.1 Legislative Framework...... 12 4.2 Walkway Options ...... 12 4.3 Legal Protection Options ...... 16 5 Conclusions & Recommendations ...... 17 5.1 Existing Values and Risks ...... 17 6 References...... 18 7 Appendix I: Plant Species List ...... 19

Prepared by: Toni Cornes & Gerry Kessels Reviewed by: Britta Deichmann

This document and its contents is the property of © Kessels & Associates Limited. Any unauthorised employment or reproduction, in full or in part, in any format, is forbidden.

Document Ref: hcc/riverlea eco report 300407

www.kessels-ecology.co.nz

1 Introduction

1.1 Brief Hamilton City Council engaged Kessels & Associates Ltd to assess the need, justification and recommended boundary for esplanade reserve to meet the purposes of esplanade reserves as defined by the Resource Management Act 1991 for a property located at 155 Riverlea Rd, Hamilton. The broad considerations are: • existing and potential environmental and ecological values; • bank stability; and • continuity of environment and access to Hammond Park and up the Mangaonua Stream (the preferred standard is to accommodate walkway cycleway of 2.5 m with .25 flat area at either side – i.e. 3 m platform). 1.2 Methods Two site visits were carried out, on 22 February 2007 and 5 March 2007. The survey was based on the visual observation of two key biological indicators of terrestrial flora/fauna: • Indigenous flora distribution and diversity; • Identification of existing and potential indigenous fauna habitat. A plant species list was compiled from this (Appendix I), and a recent large-scale aerial photograph was used to produce a vegetation map (Figure 1). Areas of similar vegetation composition and structure were grouped into vegetation types following the descriptions of Leathwick et al. (1995). The site was surveyed for the presence/absence of key indigenous bird species. Positive identification of response calls or visual confirmations of these species were recorded. Long tailed bats are known to occupy suitable roosting sites in this locality. While no specific bat surveys were undertaken, suitable tree roost sites were looked for, and discussions were undertaken with Andrea Dekrout, who has conducted bats surveys throughout Hamilton as part of her PhD studies. No surveys were undertaken for terrestrial invertebrates, reptiles or frogs; however habitat opportunity for these taxa was assessed. An assessment of the ecological significance of the study area was undertaken using Environment Waikato’s Proposed Regional Policy Statement criteria for assessing sites of significant indigenous vegetation and habitats of indigenous fauna (Denyer & Shaw, 2002). The site has also been assessed based on criteria outlined by Whaley et al. (1995), which is summarised as follows: 1. Representativeness 2. Diversity and pattern 3. Rarity/special features 4. Naturalness/intactness 5. Size and shape 6. Inherent ecological viability/long-term sustainability 7. Buffering/surrounding landscape/ connectivity 8. Fragility and threat (threat process and agents, effects of proposed modification) 9. Management input (nature and scale/intervention necessary, restoration potential)

The site was visually assessed in terms of the erosion and bank stability potential. Kessels & Associates are not specialists in this field, however, and any comments made in this report should be verified by qualified geotechnical engineers. Figure 1 Location & Vegetation Map

2 Description of Ecological Features

2.1 Landforms The escarpment and terraces consist of alluvial rhyolitic sands deposited by the Waikato River1. These areas are known to be prone to erosion. Slumps could occur if vegetation is removed or during construction of structures in steep areas. The Mangaonua Stream runs directly adjacent to the south-west of the study area before connecting to the Waikato River. Residential and light industrial areas are found on the upper terraces to the north along Riverlea Road. 2.2 Vegetation Description The study area consists of seven more or less distinct vegetation communities which are mapped on Figure 1: 1. Kanuka forest (K) 2. Mahoe-tree fern forest (MTF) 3. Exotic trees (ET) 4. Bamboo (B) 5. Willow (W) 6. Privet (Pr) 7. Exotic scrub (ES) Appendix I provides a species list for the study area. The dominant vegetation types are described as follows:

2.2.1 Secondary broadleaved forest (MTF) The major canopy cover consists of indigenous mahoe and tree fern species (Photos 1+2). This vegetation type is common in secondary growth forests after clearance. The canopy species are mahoe, wheki, silver fern, mamaku fern, and Fuchsia. The understorey comprises mostly of wandering Jew and ivy. There are also some unvegetated areas of open sandy soils on the slope. Native canopy tree saplings and seedlings are present in the understorey showing some natural regeneration is occurring. Other understorey species include kiokio, blackberry, Doodia media, Asplenium oblongifolium, hangehange, pate, Carex solandri, Montbretia, Uncinia uncinata, nikau, karamu and Pteris macilenta.

2.2.2 Secondary small-leaved forest (K) This vegetation type covers the drier areas of the cliff face (Photos 3+4). The kanuka are large emergent canopy trees at this site. Under the kanuka is either the mahoe-tree fern vegetation or privet. On an open area of the upper terrace some kanuka saplings have been planted.

1 NZ Geological Survey Sheet 4 - Hamilton

Photo 1 – Mahoe-tree fern forest

Photo 2 – The understorey contains a high number of exotic species with some indigenous species

Photo 3 – Kanuka canopy

Photo 4 –Understorey of kanuka canopy

2.2.3 Exotic forest (ET, ES, B, W, Pr) There are four main exotic canopy species - pine, privet, bamboo and willow. The area also contains a few scattered ornamental trees. The pine and ornamental trees are old plantings on the edge of the forest beside the residential area. Some of these large trees are 60-70 years old and are possible bat roosting sites. Privet is the most abundant exotic species, being present in the canopy and the understorey. There is a dense area of bamboo at the southern end of the area. There are some small pockets of bamboo on the hillslope as well. Along the stream side are some grey willows. The middle terrace of the subject property looks as it was once grazed, forming an open groundcover of grass under a canopy of exotic trees and several smaller kanuka, with a large pine dominating the site. The exotic canopy species include mature Pinus radiata, Indian bean tree, poplar, tulip tree, sweet gum and macadamia. There were also a few planted kanuka in this area around 4m in height. Other groundcover species are wandering Jew, Solanum sp and blackberry. On the southern end of the terrace across an old fence line secondary broadleaved forest begins, while secondary small-leaved forest continues on the west and eastern sides.

Photo 5 – Bamboo canopy 2.3 Fauna

2.3.1 Bats In the nearby Hammond Bush long-tailed bats (Chalinolobus tuberculatus) have been found. The possibility of bats roosting in the large trees at this site should be explored. Foraging long-tailed bats frequent forest edges or low density regenerating kanuka and manuka forests. Furthermore the presence of several large, old pines trees in the locality would provide ideal roosting sites.

2.3.2 Birds Common open country birds are present such as magpie, quail and fantail. Birds noted during the site survey included mallard duck, fantail, grey warbler, welcome swallow, house sparrow, pheasant and pukeko. Tui are likely to utilise the forest areas on a seasonal basis.

2.3.3 Fish The fish fauna is typical of such streams in the Waikato (NZ Freshwater Fish Database). Short-fin eels are well represented in the Mangaonua Stream, but do tolerate highly modified environments. Giant kokopu are a Nationally Threatened Species (Hitchmough, 2002) and are likely to be found within this reach of the Stream (B Aldridge, pers comm)2. Banded kokopu are known to occur in Hamilton gully streams and are also likely to be present here. The occasional long-fin eel is present. This eel is less common in lowland streams than the shortfin as it prefers more intact mid & upland streams with good riparian margins. Long-fin eel is now listed as “Chronically threatened species in gradual decline” (Hitchmough, 2002). Other fish, such as inanga and smelt, will be present and adult brown trout may be found on occasion. 2.4 Threats

2.4.1 Animals Domestic livestock does not have access to the site. Possums are present in the area. Introduced predators, especially cats, mustelids and rats (both ship and Norwegian) are likely to be widespread and common.

2.4.2 Plant Pests Being in an urban area there is a large number of exotic species present. The main problem weeds are privet, wandering Jew, blackberry, Montbretia, woolly nightshade, hawthorn, bamboo, willow, bindweed, ivy, Jerusalem cherry, wattle, gorse and ladder fern.

3 Ecological Significance Assessment

3.1 Whaley Ecological Significance Assessment This assessment takes into account the natural values of the entire length of the escarpment, as well as linkages with upstream and downstream natural features. 1. Representativeness: The first criterion assesses what contribution the vegetation identified makes to the conservation of all indigenous ecosystems present in the natural landscape. The vegetation surveyed is modified but comprises predominantly of regeneration of secondary growth indigenous canopy with some exotic canopy. Nonetheless, Harding (1997) and Leathwick et al (1995) note that within the Hamilton Ecological District any indigenous vegetation of any kind is valuable as only 1.6 percent of the original vegetation cover and less than one percent of the original forest cover remains today. 2. Diversity and pattern: As Figure 1 indicates, exotic weeds are widespread in the canopy. Nonetheless, the presence of a diverse range of common indigenous and secondary growth species throughout suggests that the bank still retains natural indigenous features. 3. Rarity/special features: This site is near a known long-tailed bat nesting site (A Dekrout, pers comm.). All large, old or standing dead trees are potential bat roosting sites. Bats roost in the knots and crevasses of old-age trees. Ultrasound detection is needed to determine if bats are present in old trees before any are felled (Dekrout, 2005). No rare fauna or flora was found.

2 Masters student undertaking fish surveys within Hamilton gully steams

4. Naturalness/intactness: In terms of naturalness, the stream gully system can best be described as being modified, with elements of original ecosystem types still remaining. 5. Size and shape: The escarpment and terrace follows along the topography of the stream system. Its inherent viability comes from the fact that it is bio-physically linked to river and stream gullies upstream and downstream. It is of sufficient size to be self-sustaining. 6. Inherent ecological viability/long-term sustainability: The ecological viability of the gully and long term sustainability is linked to two major factors – linkages with habitat upstream and downstream and prevalence of plant pests. Plant pests are significantly limiting the long-term sustainability and regeneration of this habitat and will need control. 7. Buffering/surrounding landscape/ connectivity: The study area is an integral part of the river and stream ecosystem as a whole, providing extremely important links and buffering for fauna movement and native plant pollination. This bush adjoins Hammond Bush, which is currently being restored. City planning (Ware, 1998) envisages this site being connected with Hammond Bush pathway giving people access to the area. 8. Fragility and threat (threat process and agents, effects of proposed modification): At present the greatest threats to the stream bank are infestations of exotic weeds. Weeds will need to be controlled and removed. Since the escarpment consists of easily eroded sand, weed removal will have to be done carefully and over time to present slumps. Another problem is the site is at a higher risk of invasion by garden plants being situated in an urban area. There is a risk of residential or industrial development removing the vegetation cover. 9. Management input (nature and scale/ intervention necessary/restoration potential): Weed control, felling of large exotic trees which are not bat roosting sites and formal protection (in order to protect from future clearance and development) are necessary. A long-term plant and animal pest control and some replanting are required. The removal of exotics is a high priority but caution is required because of the unstable bank.

3.2 Compliance with the EW Regional Policy Statement Criteria Table 3 assesses the area with regard to the EW Regional Policy Statement Criteria for assessing ecological significance. Table 1 Assessment of the Mangaonua Stream Gully in the vicinity of 155 Riverlea Rd against Environment Waikato RPS Criteria for Significant Indigenous Ecosystems

Specific Criteria

1 specially set aside by statute or covenant for protection and preservation: Partially – Recreational Reserve vested in HCC along stream edge

2 recommended for protection: No

3 currently habitat for indigenous species or associations of indigenous species that are threatened or endemic to the Waikato Region : Yes - long fin eel, giant kokopu in stream; longtailed bats likely to be found in forest

4 under-represented (less than 10% of the known or likely original extent remaining) in an Ecological District, or an Ecological Region: Yes – All indigenous forest underrepresented within the Hamilton Ecological District

5 nationally uncommon before human settlement: No

6 wetland habitat for indigenous vegetation or fauna: No

7 It is an area of indigenous vegetation or naturally occurring habitat that is large relative to other examples in the Waikato Region of similar habitat types, and which contains all or almost all indigenous species typical of that habitat type: No

8 It is aquatic habitat that is a portion of a stream, river, lake, wetland, intertidal mudflat or estuary, and their margins, that is critical to the self sustainability of an indigenous species within a catchment of the Waikato Region and which contains healthy, representative populations of that species: No

9 It is an area of indigenous vegetation or habitat that is a healthy, representative example of its type because: its structure, composition, and ecological processes are largely intact, and if protected from the adverse effects of plant and animal pests and of adjacent landuse (e.g. stock, discharges, erosion), can maintain its ecological sustainability over time: No

10 It is an area of indigenous vegetation or habitat that forms part of an ecological sequence that is either rare or not common in the Waikato Region or an ecological district, or is an exceptional, representative example of its type; No

11 It is an area of indigenous vegetation or habitat for indigenous species (which habitat is either naturally occurring or has been established as a mitigation measure) that forms, either on its own or in combination with other similar areas, an ecological buffer, linkage or corridor, and which is necessary to protect any site identified as significant under Criteria 1-10 from external adverse effects: Yes – important corridor and linkage functions for the Waikato River Catchment

Significance Score (out of 11): 4.5 In summary, this escarpment and stream habitat has a high portion of native plant canopy species, provides habitat for native fish species, is possible long-tailed bat habitat and provides important ecological linkages for the wider Waikato River/ Mangaonua Stream ecosystem. The area also has great restoration potential with a good understorey of indigenous plants. It is considered to form part of an ecosystem mosaic which is of Regional Ecological Significance in terms of Environment Waikato’s Regional Policy Statement.

4 Esplanade and Access Assessment

4.1 Legislative Framework

The Resource Management Act 1991 (RMA) sets out the purposes of an esplanade reserve or an esplanade strip in Section 229 as follows: An esplanade reserve or an esplanade strip has one or more of the following purposes: (a)To contribute to the protection of conservation values by, in particular,— (i)Maintaining or enhancing the natural functioning of the adjacent sea, river, or lake; or (ii)Maintaining or enhancing water quality; or (iii)Maintaining or enhancing aquatic habitats; or (iv)Protecting the natural values associated with the esplanade reserve or esplanade strip; or (v)Mitigating natural hazards; or (b)To enable public access to or along any sea, river, or lake; or (c)To enable public recreational use of the esplanade reserve or esplanade strip and adjacent sea, river, or lake, where the use is compatible with conservation values.] In considering the proposed merits of setting aside land for esplanade reserve with respect to Section 229 it is therefore important to asses the ecological, site stability and amenity values of the locality. In 1998, Hamilton City Council undertook a feasibility study of a walkway along the Mangaonua Stream (HCC, 1998) which investigated a number of walkway options and broadly assessed these Section 229 values amongst other considerations. Three walkway options were assessed in “Segment One’ of this report, which incorporates the gully margin adjacent to 155 Riverlea Road (refer to Figure 2). The report notes the highly erosive nature of the slopes in this locality. 4.2 Walkway Options We consider that there are four options for a 3m wide walkway route in the vicinity of 155 Riverlea Road (Figure 3). These are: Option 1 – Closely following the edge of the Mangaonua Stream for the majority of the walkway route in this locality. Option 2 – Utilising the less steep upper slope partially traversing the lower terrace of 155 Riverlea Road and then easing down the terrace slope again to gain access to the stream flood plain. Option 3 – A mid slope option following more or less a path along the edge of the present Recreation Reserve boundary with 155 Riverlea Rd. Option 4– bridging across the Mangaonua Stream at some point before 155 Riverlea Rd, thereby accessing the marginal strip along the true left bank and linking up with a bridge as shown by HCC in their 1998 report (Figure 2). The relative disadvantages and advantages of each option are summarised in Table 2. We do not favour a mid slope option (Option 3). There is a steeply incised ridge and associated gully emanating in from the lower terrace of 155 Riverlea Road heading down into Mangaonua Stream. The vegetation clearance and bank excavation required in this location would pose considerable engineering challenges and would open up a large exposed cut batter and fill batter to erosion risk. This option would also require some 800m2 of clearance of the significant indigenous vegetation. This walkway alignment would also cut through the centre of this riparian margin vegetation, thereby resulting in a number of adverse ecological fragmentation effects.

Figure 2 Extract from Mangaonua Walkway Feasibility Report detailing walking options adjacent to Riverlea Rd

Table 2 Walkway Route Option Risk Assessment Matrix Option Advantages Disadvantages 1 • Avoids large gully. • Board-walking likely near stream • Requires less cut batters into ridge. edges and across gully. • Does not require private land. • Flood and scour risk high. • Does not fragment bush areas. • Will require cut batter into prominent ridge. • Good views of the stream and western banks. • Will require bamboo removal. 2 • Avoids large gully and ridge. • Requires acquisition or legal protection • Minimal significant indigenous vegetation of the lower portion of 155 Riverlea Rd clearance involved. and land on the adjacent properties. • High amenity vales as track passes by • Loss of privacy for the affected Indian bean tree and surrounding landowner. relatively flat ground. • Large pine and Liquidambar may be • Little, if any, board walking and/ or cut/fill OSH risk to passers by. batters required & hence low soil disturbance & slip risk. 3 • Avoids intrusion into 155 Riverlea Rd • Significant cut and fill batters and or board walking required – high erosion and slip risk created. • Clearance of large areas of significant indigenous vegetation • Fragmentation of significant indigenous vegetation. 4 • Avoids intrusion into 155 Riverlea Rd • Bridge construction expensive • Create ‘interesting’ and diverse walkway • Create privacy issue for owner on route western bank. • Likely to be significant erosion issues with a track/boardwalk construction. • Would require clearance of indigenous vegetation.

Figure 3 Walkway Route Options Adjacent to 155 Riverlea Road

Option 4, involving the bridging of Mangaonua Stream, does relocate any walkway away from the subject property to the south western bank. While not as steep as the true left bank of the stream, the banks on the western side are still quite steep and contain indigenous vegetation (predominantly mature mahoe), although they are subject to stock grazing. There would still be considerable erosion issues associated with construction of a track on this side of the stream and loss of privacy and security would be of concern to the landowner here as well. Land for this option is in Waikato District a joint project would be necessary to develop the walkway/cycleway. A foot bridge constructed here is likely to cost in the vicinity of $100,000. Option 1 would require cutting and board walk construction along the true right bank of the Mangaonua Stream. The stream in this location could accommodate a 3m wide walkway but in places significant cut batters would be required, and in others board walk would be the only practical option. In particular, the steep ridge would still require cutting into to obtain a 3m wide bench. The stream bank did not show any sign of slumping or slips but the banks are incised and construction would have to be undertaken carefully to prevent slumping (Photo 6). This option would create a pleasant stream side walk. While some vegetation removal would be required, Option 2 would not dissect the forest remnant as such.

Photo 6 – Mangaonua Stream as viewed from the true right bank adjacent to 155 Riverlea Rd Option 2 involves a walkway route up through 157 Riverlea Rd (north), or up the sharp ridge between 157 and 155, and then sidling around the steep gully within 155 along the middle terrace into 149 Riverlea Road or just below it, before entering the Recreation Reserve again into a large stream flood plain. This option appears to be the easiest to construct as it would avoid the gullies and ridges where cut and fill batters and/or board walking would be required. It also wouldn’t dissect the majority of the indigenous vegetation in this locality. The middle, flat terrace of 155 could be utilised as part of the walkway, which is quite a pleasant area underneath mature specimen trees, such as the mature Indian bean tree. The added benefit of securing a walkway via Option 2 would be that a portion of the existing indigenous vegetation would be included, thus strengthening the protection of this important ecological corridor from inappropriate management

© Kessels & Associates Ltd FINAL 300407 RIVERLEA ROAD ESPLANADE RESERVE FEASILBLITY ASSESSMENT 16 or use into the future. However, it is noted that not establishing a walkway route here would not preclude the landowner from covenanting the lower terrace to protect its ecological values. There are disadvantage with Option 2. Primarily the land is in private ownership and the affected owners are likely to be concerned about privacy and security issues, notwithstanding the fact that the land would have to be acquired or legally protected in some form prior to the walkway being constructed. There is a large old pine adjacent to this option. It would have to be limbed or removed, as it would pose risks to passers by. The pine may also provide a roosting site for long- tailed bats and would need to be surveyed and any bats found captured and trans-located prior to removal. 4.3 Legal Protection Options If one of the options chosen involves the walkway crossing private land, then some form of legal protection of the route would be required. The existing marginal strip along the true right bank of the Mangaonua Stream is vested in Hamilton City Council and protected under the Reserves Act 1977 as a Recreation reserve. If further land is required from private landowners for the provision of a walkway and/or the protection of significant natural values, formal legal protection is vital. Legal protection options can involve acquisition of land from private owners and vesting as a reserve in public ownership. In this case the best option would be to amalgamate the land into the existing Recreation Reserve. Legal protection can also involve a covenant (either under the QEII National trust Act, under the Reserves Act). Covenants, with suitable conditions and a willing landowner, can provide just as secure protection as a public reserve. The key, however, to successful covenant management is that the affected landowner must be a willing party and be completely happy with the proposed covenant conditions. Often, private landowners are happy to covenant an area of their land to protect natural values but are less comfortable in allowing unrestricted public access over that land. Another matter to consider is that covenants are a form of encumbrance on a title meaning that a more restrictive covenant may result in lower land value and greater difficulty in selling the property. The final point is that covenants are passed onto new owners after purchase. New owners may have different values and concerns than the original owners and thus management issues can arise with a change of ownership.

The RMA also provides for the creation of esplanade reserves (s231) esplanade strips (s232) or access strips (s237B). The legal ramifications of these access right options are outside the scope of this report. However, it does appear that the esplanade strip or access strip provisions of the RMA would provide a secure yet flexible solution to gaining access across the private land in question. As with covenants, the agreement of the affected landowner is required. Of concern is that esplanade strips and access strip agreements can be varied (s234 and s237c). To conclude, while a covenant or esplanade strip is a useful legal mechanism to protect important natural features or casual public access, where substantial infrastructure costs are involved, actual ownership of the land by Council would provide the most secure form of land tenure.

5 Conclusions & Recommendations

5.1 Existing Values and Potential Issues The indigenous vegetation remnants in the locality of and within 155 Riverlea Rd property are considered to be of ecological significance in terms of Environment Waikato Regional Policy requirements. Some form of protection is recommended for the areas of indigenous vegetation situated outside of the existing Recreation Reserve in order to restrict inappropriate future development and to enhance the ecological contiguity of this site with the nearby Hammond Bush and the upper reaches of the Mangaonua Stream. The landowner seems motivated in restoring this area. Weed control is necessary to keep native cover dominant and planting is required. Creating walkway linkages into the rural district is recognised as a key factor for the proposed Hamilton City walkway network (HCC, 1998). Securing a practical and environmentally benign walkway route along the Mangaonua Stream in this locality is considered to be an essential pre- requisite to linking the walkway system. The potential for erosion associated with walkway construction should not be under-estimated. As the slip at the end of Riverlea Rd and the slumping along Tauhara Park indicate, serious erosion can occur quickly within Hamilton gullies and is expensive to remedy. Legal protection of these gully faces gives Council surety that the risk of inappropriate management is reduced considerably. 5.2 Preferred Walkway Route Option Of the four walkway options considered, Option 3 is the least favoured because of the high erosion risk. Option 4, on the face of it, can be discounted because of cost and privacy issues for the landowner on this side of the stream. However, if Options 1 and 2 were found to be more expensive or landowner resistance was determined, then this option should be considered in more detail. Option 2 is slightly favoured over Option 1 as the flood and erosion risk are less, construction would be easier and the amenity values are potentially greater. If a walkway was to be constructed over private land, the land would need to be secured into Council ownership, as covenants or esplanade strips do not appear to provide as robust long-term legal protection. We do note that the private owners could legally covenant their river terrace vegetation, thereby adding to the ecological buffer and linkages along this side of the Mangaonua Stream.

6 References

Dekrout, A. 2005: The distribution, activity and habitat use of Long-tailed bats (Chalinolobus tuberculatus) in Hamilton City, New Zealand. Unpublished prepared for the Hamilton City Council. Leathwick, J. R, Whaley, K. J.; Clarkson, B. D.;. 1995: Assessment of criteria used to determine ‘significance’ of natural areas in relation to section 6(c) of the Resource Management Act(1991). Unpublished Landcare Research Contract Report LC9596/021 to Environment Waikato. 34 p. Ware, R. 1998. Mangaonua Walkway Feasibility Report. Hamilton City Council.

Whaley, K. J.; Clarkson, B. D.; Leathwick, J. R. 1995: Assessment of criteria used to determine ‘significance’ of natural areas in relation to section 6(c) of the Resource Management Act (1991). Unpublished Landcare Research Contract Report LC9596/021 to Environment Waikato. 34 p.

7 Appendix I: Plant Species List

Gymnosperms Pinus radiata Pine tree Podocarpus totara totara Dicotyledons Calystegia sepium pink bindweed Catalpa bignonioides Indian bean tree Cirsium vulgare Scotch thistle Coprosma robusta karamu Corynocarpus laevigatus karaka Crataegus monogyna hawthorn Digitalis purpurea foxglove Euphorbia lathyris caper spurge Fuchsia excorticata tree fuchsia Galeobdolon luteum aluminium plant Geniostoma rupestre var. ligustrifolium hangehange Hedera helix Ivy Hoheria populnea lacebark Jasminum polyanthum jasmine Kunzea ericoides kanuka Leycesteria formosa Himalayan honeysuckle Ligustrum lucidum tree privet Ligustrum sinense Chinese privet Liquidambar styraciflua sweet gum Liriodendron tulipifera tulip tree Lonicera japonica Japanese honeysuckle Macadamia integrifolia macadamia nut Macropiper excelsum kawakawa Melicytus ramiflorus mahoe Myrsine australis mapou Paraserianthes lophantha brush wattle Physalis peruviana cape gooseberry Phytolacca octandra inkweed Populus sp poplar Rubus fruticosus blackberry Salix cinerea grey willow Schefflera digitata pate Solanum mauritianum woolly nightshade Solanum pseudocapsicum Jerusalem cherry Solanum nigrum black nightshade Sophora microphylla kowhai Tradescantia fluminensis wandering Jew Tropaeolum majus garden nasturtium Ulex europaeus gorse Weinmannia racemosa kamahi

Monocotyledons Carex solandri Cordyline australis cabbage tree Cortaderia sellosana pampas Crocosmia x crocosmiiflora Montbretia Phyllostachys sp bamboo Rhopalostylis sapida nikau Uncinia uncinata Ferns Asplenium flaccidum hanging spleenwort Asplenium oblongifolium shining spleenwort Asplenium polyodon sickle spleenwort Blechnum novae-zelandiae kiokio Blechnum chambersii Blechnum filiforme thread fern Cyathea dealbata silver fern Cyathea medullaris mamaku tree fern Dicksonia squarrosa wheki Doodia media rasp fern Histiopteris incisa water fern Microsorum pustulatum hound’s tongue Nephrolepis cordifolia ladder fern Pneumatopteris pennigera gully fern Pteridium esculentum bracken Pteris macilenta sweet fern Pyrrosia eleagnifolia leather leaf fern

Jerry M. Melilloa,1, Sarah Butlera, Jennifer Johnsona,b, Jacqueline Mohana,c, Paul Steudlera, Heidi Luxa,d, Elizabeth Burrowsa,e, Francis Bowlesf, Rose Smitha, Lindsay Scotta, Chelsea Varioa,g, Troy Hilla,h, Andrew Burtoni, Yu-Mei Zhouj, and Jim Tanga aThe Ecosystem Center, Marine Biological Laboratory, Woods Hole, MA 02543; bBiology Department, Stanford University, Palo Alto, CA 94305; cSchool of Ecology, University of Georgia, Athens, GA 30602; dHarvard Forest, Petersham, MA 02543; eRutgers University, Piscataway, NJ 08901; fResearch Designs, Lyme, NH 03768; gDepartment of Biological Sciences, Dartmouth College, Hanover, NH 03755; hSchool of Forestry and Environmental Studies, Yale University, New Haven, CT 06511; iSchool of Forest Resources and Environmental Science, Michigan Technological University, Houghton, MI 49931; and jInstitute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China

Edited by William H. Schlesinger, Cary Institute of Ecosystem Studies, Millbrook, NY, and approved April 12, 2011 (received for review December 5, 2010)

Soil warming has the potential to alter both soil and plant pro- square meters—has to be heated in situ over a long enough time to cesses that affect carbon storage in forest ecosystems. We have capture the effects of warming on plant and soil carbon stocks at the quantified these effects in a large, long-term (7-y) soil-warming forest-stand level. Second, evaluating belowground carbon-cycle study in a deciduous forest in New England. Soil warming has responses to warming requires quantification of the relative con- resulted in carbon losses from the soil and stimulated carbon gains tributions of microbial respiration and root respiration to total soil in the woody tissue of trees. The warming-enhanced decay of soil respiration, the flux commonly measured in soil-warming studies. organic matter also released enough additional inorganic nitrogen Here we report the changes in net carbon storage in trees and into the soil solution to support the observed increases in plant soil in a mixed hardwood forest ecosystem in central Massachusetts carbon storage. Although soil warming has resulted in a cumula- in response to a 5 °C increase in soil temperature. The study in- tive net loss of carbon from a New England forest relative to a cludes one large area (30 × 30 m) in which the soil was heated and control area over the 7-y study, the annual net losses generally an adjacent area (30 × 30 m) as a control. We present results after decreased over time as plant carbon storage increased. In the 8 y, including one pretreatment year followed by 7 y of soil seventh year, warming-induced soil carbon losses were almost warming. We quantified changes in plant carbon storage by using totally compensated for by plant carbon gains in response to direct measurements of tree growth. To assess changes in soil warming. We attribute the plant gains primarily to warming- carbon storage, we measured soil respiration, fine-root respiration, induced increases in nitrogen availability. This study underscores and fine-root biomass. We also measured changes in nitrogen the importance of incorporating carbon–nitrogen interactions in availability in response to soil warming, as this ecosystem is nitrogen atmosphere–ocean–land earth system models to accurately simu- limited (24). This information has provided us with insights into the late land feedbacks to the climate system. importance of carbon–nitrogen interactions in determining net carbon storage in forests in response to soil warming. climate system feedbacks | ecological stoichiometry | forest carbon budget | forest nitrogen budget | global climate change Results and Discussion Soil warming has resulted in carbon losses from the soil and has stimulated carbon gains in the woody tissue of trees. Over the 7 y he world’s forests account for more than half of the organic of treatment, the cumulative warming-induced net flux of carbon carbon stored on land (1). Currently, forests of the temperate T has been from the forest to the atmosphere, but the magnitude zone are actively accumulating carbon in large enough quantities of the flux has diminished over time as a result of the increase in to affect the global carbon budget (1, 2). A number of phe- tree growth rate in the heated area. nomena may be contributing to this enhanced carbon accumu- – Pretreatment measurements of carbon budget indexes showed lation (3 8), including recovery from historical land use (e.g., the control and heated areas to be very similar. In 2002, the − abandoned agricultural land reverting to forested land), carbon pretreatment year, tree carbon was 106 Mg·ha 1 in the control − dioxide (CO2) fertilization of photosynthesis, increased nitrogen area and 109 Mg·ha 1 in the heated area. Woody increment was − − (N) deposition, and climate change. 1.73 Mg·C·ha 1 in the control area and 1.68 Mg·C·ha 1 in the In the future, climate change is likely to play a major role in heated area. Total soil respiration rates—the combination of root − the carbon balance of temperate forests and other land ecosys- and microbial respiration—were 6.4·Mg·Cha1 for control area − tems, although the sign and magnitude of the resulting feedbacks and 5.8 Mg·C·ha 1 for the heated area (Fig. S1). During the 7-y to the climate system are uncertain (9, 10). The projected warming treatment period, however, total soil respiration from the heated of between 1.1 °C and 6.4 °C over the next 100 y (11) could affect area was consistently higher than from the control area (Fig. 1). the carbon balance of terrestrial ecosystems by altering bio- We estimate that fine-root (<1 mm diameter) respiration aver- geochemical processes such as plant photosynthesis and micro- – aged 26% of total soil respiration in the control area and 18% in the bial respiration (2, 12 14). heated area over the 7 y of treatment (Fig. 1). These root respiration Soil warming experiments conducted in a variety of ecosystems, estimates reflect the effects of warming on both respiration rates per including forests, have shown short-term losses of soil carbon as CO2 and acceleration of nitrogen cycling rates, leading to an increase in the availability of nitrogen to the vegetation (15–20). The – Author contributions: J.M.M., P.S., and F.B. designed research; J.M.M., S.B., J.J., J.M., principles of ecosystem stoichiometry (21 23) suggest that, in P.S., H.L., E.B., R.S., L.S., C.V., T.H., A.B., and Y.-M.Z. performed research; F.B. and A.B. forest ecosystems, the redistribution of a relatively small amount contributed new reagents/analytic tools; J.M.M., S.B., J.J., J.M., H.L., E.B., R.S., L.S., C.V., of this newly available nitrogen from the soil to the trees could T.H., A.B., Y.-M.Z., and J.T. analyzed data; and J.M.M. and S.B. wrote the paper. result in a substantial increase in carbon storage in woody tissues. The authors declare no conflict of interest. Until now, direct empirical evidence to evaluate the effects of This article is a PNAS Direct Submission. soil warming on carbon budgets of forest ecosystems has been 1To whom correspondence should be mailed. E-mail: [email protected]. lacking. Determining ecosystem-level responses to warming is dif- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. ficult for at least two reasons. First, a large area—hundreds of 1073/pnas.1018189108/-/DCSupplemental.

9508–9512 | PNAS | June 7, 2011 | vol. 108 | no. 23 www.pnas.org/cgi/doi/10.1073/pnas.1018189108 Although soil warming caused a loss of carbon from the soil, it concurrently stimulated a gain of carbon in the vegetation (Fig. 2). The forest in the study area has been regrowing since a major, stand-replacing hurricane in 1938, which leveled most of the trees in both the control and the heated areas. Over the course of the study, carbon storage in the vegetation of the control area ranged − between 1.7 and 3.6 Mg·C·ha 1, with a mean for the 7 y of 2.2 − Mg·C·ha 1. For forest stands of similar age and composition in other parts of the Harvard Forest, Barford et al. (31) reported a biometrically determined carbon storage rate in the live vege- − tation of 1.7 Mt·C·ha 1 for the period 1993–2000. The annual rate of carbon storage in the vegetation of the heated area was greater than in the control and ranged between − − 2.4 and 5.1 Mg·C·ha 1, with a mean for the 7 y of 3.2 Mg·C·ha 1. The rate of carbon storage in the vegetation increased over time. Fig. 1. Total annual soil CO2 efflux partitioned into soil organic matter loss, root respiration, and fine-root decomposition for the heated and control Compared with the control, the net annual ecosystem carbon · · −1 balance resulting from warming shifted from a substantial carbon areas (in Mg C ha ). − loss early in the experiment (2.0 Mg·C·ha 1) to near zero (0.0 − Mg·C·ha 1) in year 7 (Fig. 3). unit of root mass and total root mass. Over the course of the study, Integrated over the 7-y period, the warming-induced soil carbon soil warming has resulted in increased respiration rates per unit of losses have been greater than the warming-induced vegetation root mass, whereas it has led to decreased fine-root mass. carbon gains. The cumulative carbon loss from the soil that was − Our estimate of the relative contribution of root respiration to induced by warming over the 7 y of treatment was 13.0 Mg·C·ha 1, − total respiration in the control area falls between two earlier and the warming-induced vegetation gain was 7.0 Mg·C·ha 1 (Fig. estimates made in unheated areas in other similarly structured 4). Since the start of the experiment, the equivalent of 54% of the deciduous stands at the Harvard Forest (18, 25). One of the carbon released from soils in response to warming has been taken up studies, which used trenched plots to estimate the relative con- and stored in trees in the heated area. Thus, although warming has tributions of root and microbial respiration to total soil respiration, resulted in a net positive feedback to the climate system, the mag- reported that root respiration accounted for about 20% of total nitude of the feedback has been substantially dampened by the in- soil respiration (18). Bowden et al. (25), also using a trenched-plot crease in storage of carbon in vegetation. technique in a Harvard Forest deciduous stand that was not sub- Increases in vegetation carbon storage in the heated area are likely due, for the most part, to the warming-induced increase in ject to warming, reported that root respiration accounted for 33% − − of total soil respiration. These Harvard Forest results are in the net nitrogen mineralization of about 27 kg·N·ha 1·yr 1. This lower half of the range of the relative contribution of root respi- represents a 45% increase relative to the nitrogen mineralization ration to total soil respiration reported in the literature. A review rate in the control area (Fig. 5). by Hansen et al. (26) shows that, globally, estimates of the con- We used the principles of ecosystem stoichiometry to explore tribution of root respiration to total soil respiration range widely— whether or not the increase in net nitrogen mineralization in the from 5 to 100%—depending on forest type, experimental setting, heated area was large enough to support the measured increase season, and time step of the analysis. in carbon storage in the trees growing there. When carbon is Fine-root respiration increased in the heated area relative to the stored in plant tissues, a small amount of nitrogen is also stored, control in the first 2 y of warming by an average of 19%. By the with the mass ratio of C:N specific to plant tissue type. In the third year of heating, however, fine-root respiration in the control area surpassed that in the heated area by 9%, and this difference increased in magnitude in subsequent years due to a progressive decrease in fine-root biomass in the heated area (Table S1). We estimate that fine-root biomass in the top 10 cm of soil decreased by 62% in the heated area during the 7-y study. Other soil-warming studies in northern hardwood forests have shown similar dramatic decreases in fine-root biomass in response to warming (27, 28). The reduction of fine-root biomass with soil warming is consistent with current ideas that link carbon allocation in plants to the availability of nitrogen and other soil resources (29, 30). The logic is that, as nitrogen becomes more available with warming, trees do not have to allocate as much carbon resource belowground to acquire nitrogen, and so fine-root biomass decreases. Microbial respiration associated with soil organic matter decay was the largest component of soil respiration in both the heated and the control areas during the pretreatment and treatment

phases. By difference (total soil respiration minus root respira- ECOLOGY tion), we estimate that, over the 7-y treatment period, microbial respiration accounted for 74% of the total respiration in the control area and 82% in the heated area (Fig. 1). We also estimate that, over this period, the warming-induced increase in microbial respiration resulted in a reduction of the total soil carbon pool to a depth of 60 cm by 14.7% relative to the control. Carbon dioxide emissions from ﬁne-root decay increased in the heated area each year with the growth of the ﬁne-root de- Fig. 2. (A) Annual vegetation carbon storage in the heated and control − tritus pool. The additional root litter decay in the heated area, areas (in Mg·Cha 1). (B) Annual vegetation in carbon storage delta (heated − however, was a small percentage (∼1%) of total soil respiration. minus control) (in Mg·C·ha 1).

Melillo et al. PNAS | June 7, 2011 | vol. 108 | no. 23 | 9509 Fig. 3. The average annual effect of soil warming on the net carbon bal- Fig. 5. Net nitrogen mineralization in the control and heated areas. Bars ± ance of the forest stand (ecosystem carbon flux) expressed as the difference represent mean net nitrogen mineralization rates of subplots (n = 10) 1SE · · −1· −1 between the warming-induced carbon loss from the soil (soil organic matter in kg N ha yr . decay) and the gain in the above- and belowground perennial tissues of the −1 canopy trees (vegetation carbon storage) (in Mg·C·ha ). These values are between 4 and 7 d sooner in trees in the warmed area. This relative to the control area. Note that the ecosystem carbon flux value for – · · −1 translates to a 3 4% increase in the mean growing season length year 7 is near 0 Mg C ha . of 161 d reported for deciduous stands at the Harvard Forest (34). A slightly longer growing season may be interacting with wood of deciduous trees at the Harvard Forest, the mass ratio of greater nitrogen availability to enhance plant productivity and carbon stored per unit of nitrogen stored is ∼300:1 (e.g., 32, 33). carbon storage. However, as nitrogen is considered to be a major On the basis of this ratio, we estimate that the amount of ni- factor limiting plant growth in this system (24, 35), we attribute − − trogen required to store 1,000 kg·C·ha 1·yr 1 in new woody most of plant carbon gain to changes in the nitrogen cycle. − − growth resulting from warming is about 3.3 kg·N·ha 1·yr 1. This Although we believe that the warming of soils as the climate − amount of nitrogen is about 12% of the additional 27 kg·N·ha 1· changes will affect carbon cycling in forest ecosystems by increasing − yr 1 made available to the trees growing in the heated area. nitrogen availability and lengthening the growing season, we rec- − Our data show that most of the remaining 23.7 kg·N·ha 1 of ognize that the carbon balance of forest ecosystems in a changing climate will also depend on other factors that will change over the the newly available nitrogen in the heated area has entered century (e.g., 36–38). For example, carbon storage in woody tissue a rapidly cycling nitrogen pool that moves between the soil and will also be affected by changes in water availability, changes in the vegetation. We have measured an annual average increase in availability of other nutrients such as phosphorus, the effects of nitrogen mass in the green canopy in the heated area relative · · −1 increased temperature on both plant photosynthesis and above- to the control of 22.5 kg N ha , which accounts for almost all ground plant respiration, and the atmospheric concentration of of this newly available nitrogen in the rapidly cycling pool. This CO2. Reductions in soil moisture and the increased plant respira- accounting of the newly available nitrogen resulting from warm- tion associated with warming will tend to reduce carbon storage in ing is consistent with our observations that there has been no mid-latitude forests, whereas moderate increases in soil moisture evidence of nitrogen losses to groundwater or the atmosphere and increased concentrations of CO2 will likely increase carbon (as nitrous oxide) from either the heated or the control areas. storage in these systems, especially if nitrogen limitation is relieved. Although we think that most of the increased carbon storage It is important to recognize that the relief of nitrogen limita- in the trees is related to the warming-induced acceleration of the tion by the redistribution of nitrogen from the soil to the vege- nitrogen cycle, we also observed a lengthening of the growing tation has limits set by the size of the soil nitrogen pool and its season with warming. Using a threshold number of 50% of the accessibility to the microbial community. Results from individual buds on the trees opening, we estimate that bud-break occurs field studies (39, 40) and meta-analyses (41, 42) support the argument that, for land ecosystems to sustain large, long-term carbon accumulations in response to rising atmospheric CO2, nitrogen inputs will have to increase and/or nitrogen losses will have to decrease. With model simulations, we have demonstrated the importance of including nitrogen in coupled atmosphere–ocean–land earth system models by comparing terrestrial carbon uptake in response to increased surface temperatures using two versions of our global biogeochemistry model, Terrestrial Ecosystem Model (the carbon only and coupled carbon–nitrogen versions), in the Massachusetts Institute of Technology’s Integrated Global Systems Model framework (43). A change in terrestrial carbon uptake with increased surface temperatures was observed when nitrogen was included, leading to a net sequestration of carbon in the plant–soil system and a reduced CO2 feedback to the climate system. Other research groups have obtained similar results when they incorporated their carbon–nitrogen models into atmosphere–ocean– land earth system models (44–46). Conclusions To date, the idea that warming-induced redistribution of nitrogen Fig. 4. The cumulative effect of soil warming on the carbon balance of the from soil to vegetation can alter the carbon budget of normally − ecosystem after 7 y of warming in Mg·ha 1. Increases in growing season nitrogen-limited forest ecosystems has been an untested hypoth- length may also contribute to vegetation carbon storage (not shown in this esis (12, 23, 44–46). The results presented here provide empirical figure). These values are rounded to the nearest tenth of a Megagram (Mg). support for this concept and underscore the importance of in-

9510 | www.pnas.org/cgi/doi/10.1073/pnas.1018189108 Melillo et al. cluding both plant and soil carbon–nitrogen interactions in mak- then on. Hourly modeled values were then summed to determine annual ﬂ ing projections of land carbon balance in a warmer world. CO2 ef ux values.

Materials and Methods Root Biomass and Respiration. We used fine-root biomass and respiration data fi Site Description and Approach. The research site is an even-aged, mixed de- to assess the contribution of ne roots to total soil respiration in response to ciduous forest in central Massachusetts (42° 28′ N, 72° 10′ W). It is dominated warming at the ecosystem level. Root biomass was measured in the organic – by Quercus rubra and Quercus velutina (42% of basal area) and Acer rubrum and mineral soil horizons (0 10 cm depth) at Barre Woods from April (29%) with lesser components of Fraxinus americana (11%) and occurs on through November 2008. Roots were extracted from soil cores taken from soils of the Canton series. As in an earlier, smaller (6 × 6 m) soil-warming a subset of plots within each of the experimental areas and separated by size < – study nearby (18), we used buried resistance cables to heat the soil. In large class ( 1mmand1 3 mm diameter). Roots were then dried at 60 °C for 2 d (30 × 30 m) heated and control (unheated) areas, we carried out a set of and weighed. To assess the degree to which root respiration adjusts to fi · biogeochemical and plant phenology and growth measurements. The bio- warmer soil temperature regimes, speci c root respiration rates (nmol · −1· −1 geochemical measurements include the emissions of CO from the soil sur- CO2 g s ) were measured from May through October in 2008 and in 2009. 2 fi fi < face to the atmosphere, carbon accumulation in the vegetation, and in situ Speci c respiration rates for ne roots ( 1 mm) from control and heated net nitrogen mineralization. For all of the stoichiometric analyses, we ap- areas were measured both at a common reference temperature of 18 °C and plied a pretreatment correction factor to differentiate between the effects at the ambient soil temperature of the control or heated (5 °C) areas in each of heating and preexisting microsite differences. The pretreatment correc- experiment. Roots were cut from the top 10 cm of soil, brushed free of soil, tion factor adjusts the initial heated data to equal the control. and immediately placed in a respiration cuvette where respiration rates were measured using an infrared gas analyzer. A Q10 value was determined Historical records, stone walls, and soil horizon characteristics indicate that for the heated and control areas for each month sampled on the basis of the the area was used for either pastureland or low-intensity agriculture before reference and ambient temperatures (48). 1908. White pines dominated the site by the early 1900s, but were destroyed Monthly fine-root respiration rates were calculated using the average in a hurricane in 1938. Blowdowns were salvaged and the area was left to monthly temperature on each area and the monthly Q10 values for the heated recover and regrow naturally to its current state: a relatively young stand of − and control areas. We then calculated the daily fine-root respiration (g·C·m 2) mixed hardwoods. Soils are mainly of the Canton series (coarse-loamy over by multiplying the respiration rate by the fine-root biomass for each month sandy or sandy-skeletal, mixed, semiactive, mesic Typic Dystrudepts) with sampled. We used the modeled daily soil respiration values to determine the a surface O horizon pH of 5.2 and subsurface mineral horizon pH of 5.5. The − − percentage of total soil respiration accounted for by the roots. Analyses in- average bulk density is 0.37·g·cm 3 in the organic layer and 0.78·g·cm 3 in cluded fine roots less than 1 mm. Assuming that the percentage of total the mineral layer. The climate is cool, temperate, and humid. The mean respiration accounted for by roots has remained stable in the control areas weekly air temperature varies from a high of about 20 °C in July to a low of through time, we used a linear relationship to determine the change in root about −6 °C in January. Precipitation is distributed evenly throughout the contribution to total respiration in the heated area through time. The aver- year and annually averages about 108 cm. age percentage of total respiration accounted for by roots in the heated and During the summer and fall of 2001 about 5 km of heating cable was control areas was then applied to annual respiration values. buried by hand to minimize disturbance in a 30- × 30-m area. Cables were buried at a 10-cm depth, spaced 20 cm apart. An adjacent 900-m2 area was Root Carbon Inputs. We used fine-root biomass data from 2007 and 2008 in the delineated to serve as the control area, with a 5-m buffer in between the heated and control areas to examine the root loss in the heated area through two areas. Pretreatment biogeochemical data were collected during the time, assuming that root biomass in the control area has remained stable 2002 growing season to compare the baseline biogeochemistry for the two through time and that the root biomass in the heated area in the pretreatment large areas, control and treatment (SI Text; Figs. S1 and S2). This approach year equaled that in the control area in 2008. Root biomass estimates for the has been commonly used in large-scale ecosystem manipulations such as the control area in 2007 were not significantly different from those measured in Hubbard Brook Ecosystem Study (47). Results from an earlier soil-warming 2008. Over the course of the study, we estimated a 62% decrease in fine roots in experiment confirmed that the soil disturbance associated with the in- the heated area relative to the control (Table S1). We used root carbon values stallation of heating cables has had no effect on soil temperatures or soil from fine roots sampled in 2008 to estimate carbon input from roots to soil in respiration and only minor and variable impacts on soil moisture, net ni- response to warming. Carbon accounts for 46% of fine-root biomass. trogen mineralization (Fig. S3), and shrub growth (18). When supplied with 240 volts of alternating current, the heating cables − Root Litter Decomposition. Using the estimates of root biomass loss in response have a power output of 3.6 W·m 1 and produce a power density of about − to the warming calculated above, we determined the additional carbon pulse 77 W·m 2. There are 160 resistance heating cables, each of which is ∼30 m as a result of root litter decay in the heated area each year. The root detritus long. A typical resistance per cable is 96 ohm. The heated area is monitored pool was calculated as the root litter from the previous year plus the pulse of with 80 thermistors and 6 moisture probes, and the control plot has 12 new root biomass for the current year minus the root decomposition from the thermistors and 6 moisture probes. Every minute heating cables are turned previous year. The root decomposition was calculated as: on and off automatically to maintain a 5 °C temperature differential between heated and control areas. This heating method works well under a Root Litter Decomposition ¼k Rp Heffect variety of moisture and temperature conditions. Heating commenced on the treated area in May 2003. For all analyses in where –k is a decay rate constant (49), Rp is the root detritus pool for that fl the pretreatment and treatment periods, rates were calculated on the basis of year, and Heffect is the proportional heating effect on total CO2 ef ux for “warming years,” defined as consecutive 12-mo periods beginning in May that year (heated area respiration rate for given year/average control area 2002 (e.g., May 2002–April 2003). respiration rate from 2002 to 2009).

Total Soil Respiration. The net flux of CO2 was measured monthly from April Soil Organic Matter Decay. We combined the analyses above to estimate total through November in nine subplots in each of the heated and control areas. soil organic matter decay as a result of warming. Soil organic matter de- On each sampling date, fluxes were measured at early morning and after- composition was calculated as: noon intervals to capture low and high respiration rates. Although we did SOM ¼ Total Respiration Root Respiration Root Litter Decomposition not measure CO2 fluxes in the winter, we used a temperature model to

fl ECOLOGY extrapolate out data to daily uxes. From 2002 to 2009, daily rates of soil with the components calculated as described in Total Soil Respiration, Root respiration were modeled as: Biomass and Respiration, and Root Litter Decomposition. Changes in the soil carbon pool were calculated as: ¼ β β1T FCO2 0e − − ¼ þ 2 1 Soil C Poolloss SOMloss Root litter Decompositionloss Root C Inputsgain where FCO2 is the rate of soil CO2 efflux in mg·Cm ·hr and T is the soil temperature in degrees C measured at 4 cm. Soil temperature was measured with the components calculated as described above. adjacent to the gas sampling chambers at the time of sampling. β0 and β1 represent year- and treatment-specific parameters, which were calculated by We calculated the percentage of the soil carbon pool that was lost over the linear regression analysis in SAS (version 9.1.3). The model was driven by study in the warmed area period using Gaudinski et al.’s (50) total carbon measurements of soil temperature collected at 6-h intervals in each of the stock estimates at the Harvard Forest and our soil organic matter loss esti- experimental areas from 2003 to June 2008 and hourly measurements from mates (see above).

Melillo et al. PNAS | June 7, 2011 | vol. 108 | no. 23 | 9511 Vegetation Carbon Storage. Allometric equations were applied to monthly of the soil was placed in 100 mL of 2 M KCl, extracted for 36 h, and filtered. > measurements of dendrometer bands on all trees 5 cm diameter at breast The extracts were analyzed for NO3-N and NH4-N using a Lachat QuikChem height to calculate changes in above- and belowground woody biomass FIA+ 8000 Series Flow Injection Analyzer. carbon and vegetation carbon storage over time (51). Carbon inputs from the additional root litter in the heated area were subtracted from the Statistical Methods. Hourly modeled total soil respiration values for each 5- × woody biomass carbon delta for each year to account for the loss of fine- 5-m subplot (nine heated, nine control) were summed over the year and root mass due to warming. averaged to obtain estimates of total annual soil respiration for each treatment. Similarly, net nitrogen mineralization rates were averaged across the Nitrogen Mineralization. Using the in situ buried bag incubation method, we 10, 5- × 5-m subplots in the heated and control areas to obtain average an- measured the rates of net nitrogen mineralization and nitrification (18). nual rates. For both soil respiration and net nitrogen mineralization, annual Incubations were 5 wk in length from April to November and for 5 mo average rates were compared between the heated and control areas using through the winter. Soils were separated into organic and mineral layers Friedman’s test, a nonparametric repeated-measures analysis in SAS (v. 9.1.3). and sieved through a 5.6-mm screen to remove rocks and roots. The organic horizon had an average depth of 1.4 cm, and we analyzed the remaining 8.6 ACKNOWLEDGMENTS. This research was supported by National Science Foun- cm of mineral soil in the core. A subsample of the soil was weighed and dation Grant DEB-0620443 (Harvard Forest Long Term Ecological Research) dried at 105 °C for 24 h for soil moisture analyses. Approximately 10 grams and Department of Energy Grants-DE-FC02-06-ER641577 and DE-SC0005421.

1. Goodale CL, et al. (2002) Forest carbon sinks in the Northern Hemisphere. Ecol Appl 26. Hansen PJ, Edwards NT, Garten CT, Andrews JA (2000) Separating root and soil 12:891–899. microbial contributions to soil respiration: A review of methods and observations. – 2. Canadell JG, et al. (2007) Contributions to accelerating atmospheric CO2 growth from Biogeochemistry 48:115 146. economic activity, carbon intensity, and efficiency of natural sinks. Proc Natl Acad Sci 27. Redmond DR (1955) Studies in forest pathology. XV. Rootlets, mycorrhizae, and soil USA 104:18866–18870. temperatures in relation to birch dieback. Can J Bot 33:595–627. 3. Thomas RQ, et al. (2010) Increased tree carbon storage in response to nitrogen 28. Pregitzer KS, King JS, Burton AJ, Brown SE (2000) Responses of tree fine roots to – deposition in the US. Nat Geosci 3:13–17. temperature. New Phytol 147:105 115. 4. Caspersen JP, et al. (2000) Contributions of land-use history to carbon accumulation in 29. Rastetter EB (2011) Modeling coupled biogeochemical cycles. Front Ecol Environ 9: – U.S. forests. Science 290:1148–1151. 68 73. 5. Hurtt GC, et al. (2002) Projecting the future of the U.S. carbon sink. Proc Natl Acad Sci 30. Rastetter EB, Agren GI, Shaver GR (1997) Responses of N-limited ecosystems to increased – USA 99:1389–1394. CO2: A balanced-nutrition, coupled-element-cycles model. Ecol Appl 7:444 460. 6. Pacala SW, et al. (2001) Consistent land- and atmosphere-based U.S. carbon sink 31. Barford CC, et al. (2001) Factors controlling long- and short-term sequestration of atmospheric CO in a mid-latitude forest. Science 294:1688–1691. estimates. Science 292:2316–2320. 2 32. Nadelhoffer KJ, et al. (1999) Nitrogen deposition makes a minor contribution to 7. Schimel DS, et al. (2001) Recent patterns and mechanisms of carbon exchange by carbon sequestration in temperate forests. Nature 398:145–148. terrestrial ecosystems. Nature 414:169–172. 33. Hooker TD, Compton JE (2003) Forest ecosystem carbon and nitrogen accumulation 8. McGuire AD, et al. (2001) Carbon balance of the terrestrial biosphere in the twentieth during the first century after agricultural abandonment. Ecol Appl 13:299–313. century: Analyses of CO , climate and land-use effects with four process-based 2 34. Waring RH, et al. (1995) Scaling gross ecosystem production at Harvard Forest with ecosystem models. Global Biogeochem Cycles 15:183–206. remote sensing: A comparison of estimates from a constrained quantum-use 9. Cox PM, Betts RA, Jones CD, Spall SA, Totterdell IJ (2000) Acceleration of global warming efficiency model and eddy correlation. Plant Cell Environ 8:1201–1213. – due to carbon-cycle feedbacks in a coupled climate model. Nature 408:184 187. 35. Magill AH, et al. (2000) Long-term nitrogen additions and nitrogen saturation in two 10. Friedlingstein P, et al. (2001) Positive feedback between future climate change and temperate forests. Ecosystems (N Y) 3:238–253. – the carbon cycle. Geophys Res Lett 28:1543 1546. 36. Van Cleve K, Oechel WC, Hom JL (1990) Response of black spruce (Picea mariana) fi 11. Alley R, et al. (2007) Climate Change 2007: The Physical Scienti c Basis, Summary for ecosystems to soil temperature modificationin interior Alaska. Can J ForRes 20:1530–1535. Policymakers (Cambridge University Press, New York). 37. Schimel D, et al. (2000) Contribution of increasing CO2 and climate to carbon storage 12. Melillo JM, Callaghan TV, Woodward FI, Salati E, Sinha ESK (1990) Effect on by ecosystems in the United States. Science 287:2004–2006. Ecosystems in Climate Change: The IPCC Scientific Assessment, eds Houghton JT, 38. Zhao M, Running SW (2010) Drought-induced reduction in global terrestrial net Jenkins GJ, Ephraums JJ (Cambridge University Press, New York), pp 283–310. primary production from 2000 through 2009. Science 329:940–943. 13. Melillo JM, Kicklighter DW, McGuire AD, Peterjohn WT, Newkirk K (1995) Global 39. Oren R, et al. (2001) Soil fertility limits carbon sequestration by forest ecosystems in change and its effects on soil organic carbon stocks. Report of the Dahlem Workshop a CO2-enriched atmosphere. Nature 411:469–472.

on the Role of Nonliving Organic Matter in the Earth’s Carbon Cycle, eds Zepp R, 40. Norby RJ, Warren JM, Iversen CM, Medlyn BE, McMurtrie RE (2010) CO2 enhancement Sonntag C (John Wiley & Sons, Oxford), pp 175–189. of forest productivity constrained by limited nitrogen availability. Proc Natl Acad Sci 14. Shaver GR, et al. (2000) Global warming and terrestrial ecosystems: A conceptual USA 107:19368–19373. framework for analysis. Bioscience 50:871–882. 41. Hungate BA, Dukes JS, Shaw MR, Luo Y, Field CB (2003) Atmospheric science. 15. Peterjohn WT, et al. (1994) The response of trace gas fluxes and N availability to Nitrogen and climate change. Science 302:1512–1513. elevated soil temperatures. Ecol Appl 4:617–625. 42. van Groenigen K-J, et al. (2006) Element interactions limit soil carbon storage. Proc – 16. Rustad LE, Fernandez IJ (1998) Soil warming: Consequences for foliar litter decay in Natl Acad Sci USA 103:6571 6574. a spruce-fir forest in Maine, USA. Soil Sci Soc Am J 62:1072–1081. 43. Sokolov AP, et al. (2008) Consequences of considering carbon/nitrogen interactions – 17. Luo Y, Wan S, Hui D, Wallace LL (2001) Acclimatization of soil respiration to warming on the feedbacks between climate and the terrestrial carbon cycle. J Clim 21:1 21. in a tall grass prairie. Nature 413:622–625. 44. Thorton PR, et al. (2009) Carbon-nitrogen interactions regulate climate-carbon cycle 18. Melillo JM, et al. (2002) Soil warming and carbon-cycle feedbacks to the climate feedbacks: Results from an atmostphere-ocean general circulation model. Biogeosci Discuss 6:3303–3354. system. Science 298:2173–2176. 45. Zaehle S, et al. (2010) Carbon and nitrogen cycle dynamics in the O-CN land surface 19. Eliasson PE, et al. (2005) The response of heterotrophic CO2 flux to soil warming. Glob model: 2. Role of the nitrogen cycle in the historical terrestrial carbon balance. Global Change Biol 11:167–181. Biogeochem Cycles, 24:GB1006 10.1029/2009GB003522. 20. Shaw RM, Harte J (2001) Response of nitrogen cycling to simulated climate change: 46. Gerber S, Hedin LO, Oppenheimer M, Pacala SW, Shevliakova E (2010) Nitrogen cycling Differential responses along a subalpine ecotone. Glob Change Biol 7:193–210. and feedbacks in a global dynamic land model. Global Biogeochem Cycles, 24:GB1001 21. Melillo JM, Gosz JR (1983) Interactions of biogeochemical cycles in forest ecosystems. 10.1029/2008GB003336. The Major Biogeochemical Cycles and Their Interactions, eds Bolin B, Cook RB (John 47. Likens GE, Bormann FH, Johnson NM, Fisher DW, Pierce DW (1970) Effects of forest – Wiley & Sons, New York), pp 177 222. cutting and herbicide treatment on nutrient budgets in the Hubbard Brook 22. Sterner RW, Elser JJ (2002) Ecological Stoichiometry: The Biology of Elements from Watershed-Ecosystem. Ecol Monogr 40:23–47. Molecules to the Biosphere (Princeton University Press). 48. Burton AJ, Melillo JM, Frey SD (2008) Adjustment of forest ecosystem root respiration 23. Rastetter EB, McKane RB, Shaver GR, Melillo JM (1992) Changes in C storage by as temperature warms. J Integr Plant Biol 50:1467–1483. terrestrial ecosystems: How C-N interactions restrict responses to CO2 and temperature. 49. McClaugherty AC, Aber JA, Melillo JM (1982) The role of fine roots in the organic Water Air Soil Pollut 64:327–344. matter and nitrogen budgets of two forested ecosystems. Ecology 63:1481–1490. 24. Aber JD, Nadelhoffer KJ, Steudler P, Melillo JM (1989) Nitrogen saturation in 50. Gaudinski JB, Trumbore SE, Davidson EA, Zheng S (2000) Soil carbon cycling in Northern forest ecosystems. Bioscience 39:378–386. a temperate forest: Radiocarbon-based estimated of residence times, sequestration 25. Bowden RD, Nadelhoffer KJ, Boone RD, Melillo JM, Garrison JB (1993) Contributions rates and partitioning of fluxes. Biogeochemistry 51:33–69. of aboveground litter, belowground litter, and root respiration to total soil 51. Jenkins JC, Chojnacky DC, Heath LS, Birdsey RA (2003) Nation-scale biomass estimators respiration in a temperate mixed hardwood forest. Can J For Res 23:1402–1407. for United States tree species. For Sci 49:12–35.

9512 | www.pnas.org/cgi/doi/10.1073/pnas.1018189108 Melillo et al.