Category Archives: Community involvement

Addressing ghost nets in Australia and beyond – update of EMR feature

Britta Denise Hardesty, Riki Gunn and Chris Wilcox

[Update of EMR feature  – Riki Gunn, Britta Denise Hardesty and James Butler (2010) Tackling ghost nets: local solutions to a global issue in Northern Australia, Ecological Management & Restoration, 11:2, 88-98. https://onlinelibrary.wiley.com/doi/10.1111/j.1442-8903.2010.00525.x]

Key words.  derelict fishing nets, ghost gear, GGGI, Indigenous livelihoods

Figure 1. Dead turtle caught in a derelict ghost net. (Photo: Jane Dermer, Ghost Nets Australia)

Introduction. The focus of our 2009 feature was to highlight the work of Indigenous rangers in addressing the local but widespread problem of abandoned, lost or derelict fishing gear (ALDFG) in Northern Australia, particularly ‘ghost nets’ that are carried on the currents and continue to fish long after they are no longer actively used (Figs 1-4). We also aimed to raise awareness of the efforts required to address this complex issue, whilst highlighting the work of Indigenous rangers working in the region.  The feature reported ghost net removal efforts taking place in Australia’s Gulf of Carpentaria – which, by 2009, involved the removal of 5532 nets by over 90 Indigenous rangers from more than 18 Indigenous communities.  This highlighted the transboundary nature of the ghost gear issue, and identified that most nets likely originated from beyond Australia’s waters.

Figure 2. Napranum ranger Philip Mango releasing juvenile turtle trapped in ghost net. (Photo: Ghost Nets Australial)

Further work. Since 2010, the understanding of and approaches to addressing the derelict fishing gear issue have increased substantially. This has been reflected both in domestic efforts within Australia, and more broadly in the international community.

Domestically, in the last decade, the ranger program across northern Australia has evolved and grown, enabling more Indigenous people to remain culturally connected to their land and sea country through meaningful employment.  Ranger activities generally involve a range of restoration activities including feral and weed management, in addition to (for  coastal groups) ghost net removal. Across northern Australia, Indigenous ranger groups continue to remove nets on their country, demonstrating the success of the initial program supported by the Australian government. To date, nearly 15,000 ghost nets (three times the number reported in 2010) have been removed from the region. The net removal program has extended beyond Ranger groups working in the Gulf of Carpentaria to include the Torres Strait, the western part of the Northern Territory Coast, and parts of the Kimberly coastline in Western Australia.

Globally, the world is focused on the United Nations Sustainability Development Goals (SDGs) which aims to provide a ‘shared blueprint for peace and prosperity for people and the planet, now and into the future’ (https://sustainabledevelopment.un.org/sdgs).

A key focus for the SDGs is to help preserve the world’s oceans, a topic which touches on food security, poverty and economic growth, among other goals. Ensuring fishing practices are aligned with these goals includes reducing gear losses into the marine and coastal environment. In recognition of the issue and to end ALDFG, there is now a multi-stakeholder alliance of fishing industry, private sector, multinational corporations, non-government organizations, academics and governments, the Global Ghost Gear Initiative (GGGI), which is focused on solving the problem of abandoned, lost and derelict fishing gear worldwide. Both CSIRO and GhostNets Australia were founding members of this alliance and have been instrumental in engagement and scientific endeavours which inform the GGGI.

Fig 3. An enormous effort is invested by Indigenous rangers in removing ghost nets from beaches along the northern Australian coastline (Photo: World Animal Protection/Dean Sewell)

Based on collaborative research between GhostNets Australia and CSIRO, it was determined that the primary source of derelict nets washing ashore along Australia’s northern coastline was the Arafura Sea. Engagement with fishers in the region through a series of workshops identified that major causes of gear loss included snagging of nets and over-capacity in the region. We also identified opportunities to help resolve ghost net issues in the region, though stakeholder engagement, points of intervention and livelihood tradeoffs. Much of this overcapacity and overcrowding has been attributed to illegal, unreported and unregulated (IUU) fishing. Subsequently, Indonesia went through a substantial change in practices with regards to allowing foreign vessels in their waters, effectively closed their borders to foreign fisheries operators. Anecdotally, information from multiple ranger groups in Northern Australia suggests that this highly publicized and significant change in practice has resulted in a substantial decrease in the number of ghost nets washing ashore along at least part of the northern Australian coastline.

Another outcome from the collaborative research effort was a new understanding based on deep citizen science engagement and modelling to identify potential high risk areas where ghost nets were likely to cause the most harm to turtles. In this work, we were able to suggest interdiction points for ghostnets, before they entered the Gulf of Carpentaria where they were likely to kill wildlife. We also identified the nets that were most harmful to wildlife and we estimated that nearly 15,000 marine turtles had likely been killed by derelict nets in the region.

There have also been some technological improvements in this area. These fall into both reporting and in tracking nets. Electronic data collection has improved the quality of data collection and can ensure errors are minimised. Development of the tool has also been designed such that those with reduced literacy are also able to collect valuable information, a feature that can be important in many communities. Using icons and photos to help identify nets improved data reliability.

Also within Australia, alternative livelihoods programs such as Ghost Net Gear evolved into the Ghost Net Art Project where the art works have excited the International art community.  This has resulted in purchases by many internationally renowned purveyors of artwork including the British Museum, the Australian National Museum and the Australian Maritime Museum. Works from Indigenous artists can also be seen at Australia’s Parliament House, and exhibitions have taken place in Monaco, Alaska, Singapore and France as well as in numerous national and regional galleries around Australia. A commemorative stamp was even made from the Ghost Nets artwork that lives in the Australian National Museum.

Figure 4. Large nets can become entangled in coastal vegetation. (Photo: World Animal Protection/Dean Sewell)

Future directions. While GhostNets Australia has not formally continued as a non-governmental organization, many of the components initiated through the program have continued and grown through time, as exemplified above. This early work also helped springboard CSIRO’s engagement in capacity building with the Indonesian government to tackle Illegal, Unreported and Unregulated (IUU) fishing. This had led to a strong research collaboration relationship between the two countries, with a shared goal of reducing IUU fishing, building capacity on marine resource management, and improved monitoring, control and surveillance efforts in Indonesia.

CSIRO is also involved in an aerial (re)survey of the coastline across Northern Australia. In affiliation with World Animal Protection and Norm Duke and Jock Mackenzie from James Cook University, we are looking at changes in the number of ghost nets along the shoreline. Stereo images were recorded along the entire coastline and we are comparing ghost nets observed across the region with two other aerial surveys that have taken place in the last decade. The team have just completed flights (September 2019), so we are looking forward to analysing the images and comparing ghost net numbers across the region.

ContactDenise.hardesty@csiro.au; CSIRO Oceans and Atmosphere, Hobart, Tasmania, Australia. rikigunn1@outlook.com; chris.wilcox@csiro.au

Still repairing wetlands of the Lower Murray: continuing the learning – UPDATE of EMR feature

Anne Jensen

[Update to EMR feature – Jensen, Anne (2002) Repairing wetlands of the Lower Murray: Learning from restoration practice. Ecological Management & Restoration, 3:1, 5-14. https://onlinelibrary.wiley.com/doi/10.1046/j.1442-8903.2002.00092.x]

Key words:         Environmental water requirements, regeneration, wetlands, black box seedlings, Lower Murray Valley

Figure 1. Location of the Lower Murray Valley in South Australia (Map A. Jensen)

Introduction. As highlighted in the original EMR feature this summary is updating, in the Lower Murray Valley 1100 wetlands have been identified in 250 hydrologically-linked complexes (Fig. 1). They have undergone major changes to their water regime over the last 100 years, altering the timing, frequency and duration of floods. Wetlands at lower elevations have become permanently flooded by stable river levels and wetlands at higher elevations are ‘droughted’ by much reduced flooding. All would benefit from environmental watering, to fill gaps in breeding and regeneration cycles.

Our 2002 feature showed that, from 1998 to 2002, the not-for-profit conservation company Wetland Care Australia coordinated on-ground projects to repair priority wetlands in the Lower Murray. The Gurra Gurra project was the largest of these projects, with engineering works at 17 sites to restore multiple flowpaths through the 3000 ha floodplain complex.

Key funding from the National Heritage Trust terminated in 2002 and Wetland Care Australia relocated in 2003 to northern New South Wales, where project funding for wetland projects was still available. However, individuals involved with the Wetland Care Australia projects remained in the Lower Murray Valley in other jobs, so the intellectual property was retained and wetland conservation activities continued.

In 2002, the extent and severity of drought conditions in the Murray River Valley were just being recognised. By 2004, a survey estimated that >75% of the two main tree dominants in floodplain woodlands –  River Red Gum (Eucalyptus camaldulensis) and Black Box (E. largiflorens)  – were dead, dying or extremely stressed along 700 km of the Murray River Valley . The Millenium Drought (2000-2010) caused extreme stress to both ecological and human communities. Government agencies commenced emergency environmental watering from 2004 through the Living Murray program to limit catastrophic damage at eight iconic sites but millions of mature eucalypts were lost from floodplain woodlands along river valleys.

The Millenium Drought changed the governance context radically, with the Water Act 2007 establishing a new Murray-Darling Basin Authority and the Basin Plan. The Commonwealth Environmental Water Holder (CEWH) was able to purchase water for environmental use.

Nature delivered life-saving floods in 2010-12, which broke the drought and sent flows through the Gurra Gurra complex flowpaths, so the works completed back in 2000 finally fulfilled their function (Fig. 2). Water flowed through the pipes at Tortoise Crossing for 170 days in 2010-11 and again for 71 days in 2012.

Figure 2. The sign at the key Tortoise Crossing flow path explains that replacing three pipes with 160 pipes back in 2000 now allows 50 times more flow when the river floods, as seen at the flood peak in December 2016 (Photos A. Jensen)

The sequence of floods led to mass germination of Black Box at medium floodplain elevations, with mass River Red Gum seedlings at lower elevations. A range of studies show that the survival of these seedlings is critical to fill age gaps and replace the losses from the Millenium Drought, as survival rates from germination events in the 1970s and 1990s were very poor and the last successful mass recruitment of Black Box in the Lower Murray Valley was from the 1955-56 floods.

Following the floods in 2010-2012, conditions were dry in 2013-15 and the fields of mass seedlings began to dry out and die. A further short flood in 2016 watered the surviving fields of Black Box seedlings for at least two weeks, adding to prospects of survival and flowing through the Tortoise Crossing pipes for 75 days. However, conditions in 2018-19 and into summer 2019-20 are once again extremely dry, with stress appearing in mature trees and saplings dying off. The Lower Murray Valley is still recovering from the Millenium Drought, thus needing more frequent watering over a sequence of years to bring mature trees back to health and full seed production, so this is a significant setback.

Further works and activities since 2002. Since 2008, the environmental charity Nature Foundation SA (NFSA) has been undertaking environmental watering projects on smaller, privately-owned sites in the Lower Murray, many from the original Wetland Care Australia list. In the Lower Murray Valley, water needs to be lifted up to 3 m from the river channel to reach wetlands on the floodplain, requiring costly energy. This is done using irrigation techniques, including pumps, pipes and sprinklers. These smaller projects complement government agency projects using major infrastructure to deliver environmental water to much larger wetland complexes.

In 2008-09, the primary purpose was to acquire water and use it to limit extreme environmental damage in the drought. In 2009 NFSA provided supplementary water for Little Duck Lagoon, one of the sites from the Wetland Care Australia Gurra Gurra project.

From 2012-19, NFSA has held a contract partnered with the Commonwealth Environmental Water Holder (CEWH) to deliver up to 10 GL/y of environmental water to selected sites. A priority for the NFSA Water for Nature program has been to sustain the mass germination triggered by the 2010-12 floods, watering fields of seedlings and saplings so they can fill the very large gap in age structure of Black Box populations. Stressed mature Black Box trees are being watered to improve their condition and volumes of seed produced. While delivering water to a defined wetland is relatively simple, with water pumped to an inlet point and allowed to pool in the wetland, watering scattered fields of seedlings and saplings on relatively flat floodplain land is a challenge, especially when they are in gaps between mature trees. The solution has been to use high-throw sprinklers (simulating rainfall) and operating them at night, to allow soakage into clay soils and to avoid evaporative loss during the day.

Since 2008, NFSA has delivered almost 13 GL of water to 97 watering sites in 20 wetland complexes, covering 27 different ecological targets across 12 habitat types. A total of 4.9 GL was delivered to 15 sites in 2017-18 and 1.55 GL was delivered in 2018-19 to 25 sites covering 126 ha. Rolling 5-year watering plans have been developed for each site, able to respond to annual water availability, Basin-wide priorities, environmental water requirements, climatic conditions, site watering history and feasibility of delivery.

One of the NFSA sites is Lyrup Lagoon in the Gurra Gurra complex, being watered to reduce accumulated salinity from groundwater inflows. Importantly, the infrastructure of the Central Irrigation Trust was used to deliver water to the lagoon. Thus, local irrigators are partners in delivery of water for regional environmental benefits and river health.

Figure 3. Watering guidelines developed by the Water For Nature program for stressed and healthy woodlands, for (a) River Red Gum and (b) Black Box (Water for Nature).

Further Results. The initial watering guidelines reported in the original EMR feature have been expanded through research and monitoring of responses to watering events, developing guidelines for timing and frequency of wetting and drying cycles to promote recovery in mature trees and support germination and survival of seedlings. These have been applied for each site in the rolling 5-year watering plans, which then determine the annual list of sites due for watering (see NFSA 5 year strategy and Fig. 3).

Watering by NFSA 2013-2019 has sustained Black Box seedlings and saplings through four dry summers, with watered plants 2-3 times taller than non-watered plants (Fig. 4). The Water For Nature monitoring report shows that, at NFSA sites, mature Black Box trees that have received periodic environmental water as determined by their 5-year watering plan during 2015-2019 were 21-46% (average 36%) better in health than adjacent non-watered sites, with denser, more vigorous canopies and the relative improvement was greatest during hotter and drier periods. The watering events thus provided water between natural floods to sustain growth in saplings and crop cycles in mature trees. Watering at other NFSA sites has provided vital habitat for vulnerable and endangered fauna including the Murray Hardyhead (Craterocephalus fluviatilis), Southern Bell Frog (Litoria raniformis), Regent Parrot (Polytelis anthopeplus) and Latham’s Snipe (Gallinago hardwickii).

Figure 4. Watered River Red Gum saplings at Thiele Flat, Loxton; November 2013 (top) and March 2018 (bottom). Note 2016 flood level mark on foreground trees (Photos A. Jensen)Lessons learned and future directions.

The significant benefits of environmental water have been demonstrated at NFSA’s Water For Nature sites, for floodplain vegetation communities and in temporary wetlands. Evolving research indicates that watering in late spring-early summer mimics peak flows in the natural water regime, coinciding with highest chances of breeding and germination events and thus ecologically ideal timing (See bibliography). Benefits are increased if seasonally filled wetlands are topped up in early summer, to ensure sufficient duration to sustain frog and waterbird breeding.

As well as ideal timing, studies have shown that watering at any time of the year can be beneficial, including enhancing soil moisture storage in the unsaturated zone and sustaining volume in bud and fruit crops. A key finding has been that watering in late autumn-early winter sustains soil moisture, priming sites to give an enhanced response to watering in the following spring-summer.

However, dry climatic conditions and political pressures to minimise water recovery volumes are combining to reduce availability of environmental water, with only very highest priority sites likely to receive water in the 2019-20 water year. Environmental water cannot create floods, it can only provide water to selected priority sites during dry times and enhance the benefits of any natural floods. Current volumes can only meet the requirements of a limited number of sites, leaving many sites without the water needed to sustain them through dry times or to recover from the severe impact of the Millenium drought.

Bureaucratic processes for approvals also hinder effective delivery of environmental water. With the water year coinciding with the financial year from July to June, water delivery stops in June to allow water accounts to be finalised. Approval to water in the following year can take 2-3 months, meaning no water can be delivered during the winter months for priming, missing the advantage of low evaporation rates and higher chances of piggy-backing on rainfall events.

Funding for environmental projects tends to be short term, leading to job insecurity for project managers, loss of continuity and project knowledge, and inability to complete watering sequences. Very significant volunteer resources are required to make these watering projects happen, including inputs from landholders who have donated electricity connections to the floodplain, transported diesel to re-fuel pumps, loaned pumps, tractors and irrigation equipment, plus use of irrigation and local government infrastructure to deliver water, and physical assistance and maintenance from local volunteer groups.

Practical on-ground watering knowledge is maturing well; what is needed now is sufficient water and ongoing consistent funding to support projects to deliver minimum environmental water requirements for the wetlands of the Lower Murray Valley. The pipes at Tortoise Crossing, installed in 2000 and only flooded twice, are more than ready for the next high flows to pour through!

Stakeholders and Funding bodies. The monitoring project was supported as part of the project Ecological Responses to Environmental Watering in the South Australian River Murray Valley, assessing the benefits of salinity interception schemes on floodplain vegetation, coordinated by Australian Water Environments for SA Water from March 2015 to June 2017. Continuing funding for monitoring in 2017-2019 was provided in a grant from the Ian Potter Foundation to Nature Foundation SA, as well as funding from the Commonwealth Environmental Water Holder (2018-19). Water for the environmental watering projects studied here was provided through annual allocations of water from the Commonwealth Environmental Water Office to Nature Foundation SA.  Water delivery was managed by the NFSA Water For Nature program through Program Manager Natalie Stalenberg. Practical support and site access was provided by Steve Clark, landholder and committee member for Water for Nature program, and landholders John and Bronwyn Burford.

Contact. Dr Anne Jensen, Environmental Consultant; Volunteer member, Water for Nature Committee, Nature Foundation SA; part-time consultant Wetland Ecologist for Water for Nature Program of Nature Foundation SA (7 Ford Street, Maylands SA 5069, Australia; Tel: +61 407 170 706; Email: ajensen@internode.on.net

Registration of domestic cats on Christmas Island, Indian Ocean: stage one to an eradication program for stray and feral cats to mitigate social and environmental impacts – UPDATE of EMR feature

 David Algar, Neil Hamilton and Caitlyn Pink

[Update to EMR article: Algar, David, Stefanie Hilmer, Don Nickels and Audrey Nickels (2011) Successful domestic cat neutering: first step towards eradicating cats on Christmas Island for wildlife protection. Ecological Management & Restoration, 12:2, 93-101. https://onlinelibrary.wiley.com/doi/10.1111/j.1442-8903.2011.00594.x]

Key words: domestic and feral cats, eradication program, cat de-sexing and registration, cat management, pet cat survey, local cat legislation

Figure 1. Stray cat on Christmas Island (Photo Neil Hamilton DBCA)

Introduction: In 2010 a ‘’Cat Management Plan’’ was commissioned by the various land management agencies on Christmas Island to mitigate the environmental and social impacts of cats (Felis catus) on the island (Fig 1). These impacts included contributing towards the decline of a number of native species through predation, as well as being a source of Toxoplasmosis gondii, a parasite that can lead to serious human health complications.

The plan proposed a strategy to eradicate cats entirely from the island as the domestic population died out and was adopted in late 2010. The essential first stage of the management plan was therefore the registration of all domestic cats. As part of this plan, amendments to the Local Cat Management Laws (Shire of Christmas Island Local Law for the Keeping and Control of Cats 2004 (WA)) under the Local Government Act 1995 (WA) were endorsed in August 2010. These revisions required that all domestic cats in the Shire of Christmas Island were legally bound to be de-sexed, tattooed, microchipped and registered with the Shire. The revisions were designed to limit domestic and stray/feral cat impact on the native fauna, promote responsible cat ownership, compliance and enforcement of cat management laws and prohibit the importation of new cats. Micro-chipping of domestic cats would enable the identification of those animals during trapping campaigns for stray and feral cats, so that they could be released rather than destroyed. De-sexing would prevent potential natal recruitment into the domestic, stray and feral populations. A survey of domestic cats was conducted prior to the veterinary program in October 2010 (see original feature), to guarantee that all domestic cats would be registered. One hundred and fifty-two cats were recorded during the initial survey in October 2010 of which 136 were registered as domestic pets.

Figure 2. Red-tailed Tropic-Bird with chick May 2012. (Photo Neil Hamilton DBCA)

Further works undertaken: Two further veterinary visits were conducted in May 2011 and 2012 following the domestic cat surveys to complete the veterinary program. Subsequent domestic cat surveys have been conducted each May in 2013, 2014, 2015 and 2016. In 2016 prior to the domestic cat survey, it came to our attention that a number of un-registered cats were being kept as pets. It was decided by the ‘’Christmas Island Cat Eradication Steering Committee’’ that a short-term amnesty on pet cat ownership be invoked so that these animals could also be de-sexed and registered. Following this amnesty, a final veterinary program was endorsed and fines were still issued to those residents who wanted their otherwise illegal cat to be de-sexed and registered, or unregistered cats could be handed in and euthanased without charge. Further domestic cat surveys were conducted in May 2017 and October 2018.

Further results to date: Since October 2010, 184 cats have been registered following the various veterinary programs. The survey conducted in 2018 recorded 66 registered cats remaining. The total number of domestic cats registered each year, the sex population structure, the number of new registrations and number deregistered are presented in Table 1, with the decline of two-thirds relatively steady over the years.

Table 1. Total number of domestic cats registered each year, the sex structure, the number of new registrations and number de-registered.

Date No. registered New/re-registers De-registers
  Total Female Male Total Female Male Total Female Male
October 2010 N/A N/A N/A 136 66 70 N/A N/A N/A
May 2011 138 69 69 18 10 8 16 7 9
May 2012 135 66 69 12 5 7 15 8 7
May 2013 111 53 58 0 0 0 24 13 11
May 2014 101 50 51 0 0 0 10 5 5
May 2015 87 45 42 0 0 0 14 5 9
May 2016 75 41 34 2 1 1 14 5 9
June 2016 93 49 44 18 8 10 0 0 0
May 2017 74 38 36 1 0 1 20 11 9
October 2018 66 36 30 0 0 0 8 2 6

Lessons learned and future directions: At the conclusion of the domestic cat survey in 2018, there were 66 registered cats present on the island. An additional seven domestic cats are known to have died before the planned 2019 domestic cat survey. Death of registered cats over the past nine years has been caused by a number of factors including: road fatalities; old age; disease; requests for cats to be euthanased for a variety of reasons and cats exported back to the mainland.

Domestic cats will remain on Christmas Island for a number of years, with the youngest cat approximately three years of age. Initially, as reported in the 2011 feature, it was predicted that the island would be domestic cat-free by 2024 however, this is unlikely given the subsequent and final veterinary program in 2016.

Further amendments to the island’s cat local laws were adopted in 2018, following consultation with the community and the Christmas Island Cat Eradication Steering Committee. This included an increase in penalties for illegal unregistered cats and compulsory transfer of ownership procedures to prevent future movement of registered pet cats into the designated pet cat prohibited zone. This zone protects nesting habitat for the ground-nesting Red-tailed Tropic Bird (Phaethon rubricauda, Fig 2.), where cat predation led to 90% failure of fledgling rates pre-control. Subsequent cat management in this zone has been successful in improving fledgling survival (See 2012 report).

There are several benefits of repeating the domestic cat survey each year as pet numbers decline: continue program awareness to all residents; maintain community support and involvement; offer pet health advice; thoroughly check for illegal cats to report to the Shire and respond to stray cat reports within the township. This continued effort will help ensure there is little opportunity or temptation to obtain new kittens as illegal pets while later stages of the eradication are progressing, and responsible cat ownership is maintained until the domestic cat population has died out.

The goal of eradicating cats remains highly relevant and is supported by the island community, local land management agencies and the federal government. The feasibility of long-term success is high and the outcome is likely to provide valuable lessons for other jurisdictions with social and environmental issues surrounding the presence of feral and domestic cats.

Stakeholders and Funding bodies: This is a collaborative project between Western Australian Department of Biodiversity, Conservation and Attractions and Parks Australia. The authors would like to thank Parks Australia, Christmas Island Phosphates, Shire of Christmas Island, Department of Infrastructure, Transport, Cities and Regional Development and Australian Border Force for their financial, in-kind and logistical support. Special thanks to Robert Muller, Khaleisha Amin and Chris Su for their assistance in annual surveys. The warm welcome and assistance of the whole Christmas Island community during all domestic cat surveys has been appreciated.

Contact information: David Algar, Biodiversity and Conservation Science, Department of Biodiversity, Conservation and Attractions (Locked Bag 104, Bentley Delivery Centre, Western Australia, Australia 6983) Email: dave.algar@dbca.wa.gov.au

 

 

Ku-ring-gai Flying-fox Reserve Habitat Restoration Project at Gordon, 2000 – 2019 UPDATE of EMR feature

Nancy Pallin

[Update to EMR feature –  Pallin, Nancy (2001) Ku-ring-gai Flying-fox Reserve Habitat restoration project, 15 years on.  Ecological Management & Restoration 1:1, 10-20.  https://onlinelibrary.wiley.com/doi/10.1111/emr.12229]

Key words:         bush regeneration, community engagement, wallaby browsing, heat events, climate change

Figure 1. Habitat restoration areas at Ku-ring-gai Flying-fox Reserve within the urban area of Gordon, showing areas treated during the various phases of the project. Post-2000 works included follow up in all zones, the new acquisition area, the pile burn site, the ecological hot burn site and sites where vines have been targeted. (Map provided by Ku-ring-gai Council.)

Introduction. The aim of this habitat restoration project remains to provide self-perpetuating indigenous roosting habitat for Grey-headed Flying-fox (Pteropus poliocephalus) located at Ku-ring-gai Flying-fox Reserve in Gordon, NSW Australia (Fig 1).  The secondary aim was to retain the diversity of fauna and flora within the Flying-fox Reserve managed by Ku-ring-gai Council. Prior to works, weed vines and the activity of flying-foxes in the trees had damaged the canopy trees while dense weed beneath prevented germination and growth of replacement trees.  Without intervention the forest was unable to recover.  Natural regeneration was assisted by works carried out by Bushcare volunteers and Council’s contract bush regeneration team.  The work involved weed removal, pile burns and planting of additional canopy trees including Sydney Bluegum (Eucalyptus saligna), which was expected to cope better with the increased nutrients brought in by flying-foxes.

Figure 2. The changing extent of the Grey-headed Flying-fox camp from the start of the project, including updates since 2000. (Data provided by KBCS and Ku-ring-gai Council)

Significant changes have occurred for flying-foxes and in the Reserve in the last 20 years.

In 2001 Grey-headed Flying-fox was added to the threatened species lists, of both NSW and Commonwealth legislation, in the Vulnerable category.  Monthly monitoring of the number of flying-foxes occupying the Reserve  has continued monthly since 1994 and, along with mapping of the extent of the camp, is recorded on Ku-ring-gai Council’s Geographical Information System. Quarterly population estimates contribute to the National Monitoring Program to estimate the population of Grey-headed Flying-fox.  In terms of results of the monitoring, the trend in the fly-out counts at Gordon shows a slight decline.  Since the extreme weather event in 2010, more camps have formed in the Sydney basin in response to declining food resources.

In 2007, prompted by Ku-ring-gai Bat Conservation Society (KBCS), the size of the Reserve was increased by 4.3 ha by NSW Government acquisition and transfer to Council of privately owned bushland. The Voluntary Conservation Agreement that had previously established over the whole reserve in 1998 was then extended to cover the new area.   These conservation measures have avoided new development projecting into the valley.

From 2009 Grey-headed Flying-fox again shifted their camp northwards into a narrow gully between houses (Fig 2).  This led to human-wildlife conflict over noise and smell especially during the mating season. Council responded by updating the Reserve Management Plan to increase focus on the needs of adjoining residents.  Council removed and trimmed some trees which were very close to houses. In 2018 the NSW Government, through Local Governments, provided grants for home retrofitting such as double glazing, to help residents live more comfortably near flying-fox camps.

Heat stress has caused flying-fox deaths in the Reserve on five days since 2002. Deaths (358) recorded in 2013, almost all were juveniles of that year.  KBCS installed a weather station (Davis Instruments Vantage Pro Plus, connected through a Davis Vantage Connect 3G system) and data loggers to provide continuous recording of temperature and humidity within the camp and along Stoney Creek.  The station updates every 15 minutes and gives accurate information on conditions actually being experienced in the camp by the flying-foxes. The data is publicly available http://sydneybats.org.au/ku-ring-gai-flying-fox-reserve/weather-in-the-reserve/Following advice on the location and area of flying-fox roosting habitat and refuge areas on days of extremely high temperatures (Fig 3.) by specialist biologist Dr Peggy Eby, Council adopted the Ku-ring-gai Flying-fox Reserve 10 Year Management and Roosting Habitat Plan in 2018.  Restoration efforts are now focused on improving habitat along the lower valley slopes to encourage flying-foxes to move away from residential property and to increase their resilience to heat events which are predicted to increase with climate change.

Figure 3. Map showing the general distribution of flying-foxes during heat events, as well as the location of exclosures. (Map provided by Ku-ring-gai Council)

Further works undertaken.  By 2000 native ground covers and shrubs were replacing the weeds that had been removed by the regeneration teams and Bushcare volunteers.  However, from 2004, browsing by the Swamp Wallaby (Wallabia bicolor) was preventing growth of young trees and shrubs.  Bushcare volunteers, supported by KBCS and Council responded by building tree cages made from plastic-mesh and wooden stakes. Reinforcing-steel rods replaced wooden stakes in 2008.   From 2011, the Bushcare volunteers experimented with building wallaby exclosures, to allow patches of shrubs and groundcovers to recover between trees (Figs 3 and 4).  Nineteen wallaby exclosures have been built. These range in size from 7m2 to 225m2 with a total area of 846m2.   Wire fencing panels (Mallee Mesh Sapling Guard 1200 x 1500mm) replaced plastic mesh in 2018.  Silt fence is used on the lower 0.5m to prevent reptiles being trapped and horizontally to deter Brush Turkey (‎Alectura lathami) from digging under the fence.

The wallaby exclosures have also provided an opportunity to improve moisture retention at ground level to help protect the Grey-headed Flying-fox during heat events.  While weed is controlled in the exclosures south of Stoney Creek, those north of the creek retain Trad and privets, consistent with the 10 Year Management and Roosting Habitat Plan.

Madeira Vine (Anredera cordifolia) remained a threat to canopy trees along Stoney Creek for some years after 2000, despite early treatments.  The contract bush regen team employed sInce 2010 targeted 21 Madiera Vine incursions.

A very hot ecological burn was undertaken in 2017 by Council in order to stimulate germination of soil stored seed and regenerate the Plant Community Type (PCT) – Smooth-barked Apple-Turpentine-Blackbutt tall open forest on enriched sandstone slopes and gullies of the Sydney region (PCT 1841).  This area was subsequently fenced. The contract bush regeneration team was also employed for this work to maintain and monitor the regeneration in the eco-burn area (720 hours per year for both the fire and Madiera Vine combined).

Figure 4. Exclusion fence construction method. Pictured are Bushcare volunteers, Jill Green and Pierre Vignal. (Photo N Pallin).

Figure 5. Natural regeneration in 2018 in (unburnt) exclosure S-6 (including germination of Turpentines). (Photo N. Pallin)

Further results to date. The original canopy trees in Phase 1 and Phase 2 (1987 -1997) areas have recovered and canopy gaps are now mostly closed. Circumference at breast height measurements were taken for seven planted Sydney Blue gum trees.  These ranged from 710 to 1410mm with estimated canopy spread from 2 to 6m.  While original Turpentine (Syncarpia glomulifera) had circumferences from 1070 and 2350mm with canopy spread estimated between 5and 8m, those planted or naturally germinated now have circumference measurements between 420 and 980mm with canopy spread estimated from 1.5 to 3m.  A Red Ash (Alphitonia excelsa) which naturally germinated after initial clearing of weeds now has a circumference of 1250mm with a canopy spread of 5m.  Also three Pigeonberry Ash (Elaeocarpus kirtonii) have circumference from 265 to 405mm with small canopies of 1 to 2m as they are under the canopies of large, old Turpentines.  As predicted by Robin Buchanan in 1985 few Blackbutt (Eucalyptus pilularis) juveniles survived while the original large old trees have recovered and the Sydney Bluegum trees have thrived.

In the Phase 3 (1998 – 2000) area south of Stoney Creek the planted Sydney Blue Gum now have circumferences measuring between 368 and 743 (n7) with canopy spread between 2 and 6 m.  in this area the original large trees have girths between 1125 and 1770mm (n7) whereas trees which either germinated naturally or were planted now range from 130 to 678mm (n12).  These measurement samples show that it takes many decades for trees to reach their full size and be able to support a flying-fox camp.

Wallaby exclosures constructed since 2013 south of Stoney Creek contain both planted and regenerated species.  Eight tree species, 11 midstorey species, 27 understorey species and eight vines have naturally regenerated.  Turpentines grew slowly, reaching 1.5m in 4 years.  Blackbutts thrived initially but have since died. In exclosures north of the creek,  weeds including Large-leaved Privet,  Ligustrum lucidum,  Small-leaved privet,  L. sinense,  Lantana, Lantana camara,  and Trad, Tradescantia fluminensis) have been allow to persist and develop to maximise ground moisture levels for flying-foxes during heat events. Outside the exclosures, as wallabies have grazed and browsed natives, the forest has gradually lost its lower structural layers, a difference very evident in Fig 6.

Figure 6. Visible difference in density and height of ground cover north and south of Stoney creek. (Photo P. Vignal)

Coachwood (Ceratopetalum apetalum) were densely planted in a 3 x 15m exclosure under the canopies of mature Coachwood next to Stoney Creek in 2015. In 4 years they have reached 1.5m.  In this moist site native groundcovers are developing a dense, moist ground cover.

Madiera Vine, the highest-threat weed, is now largely confined to degraded edges of the reserve, where strategic consolidation is being implemented with a view to total eradication.

In the hot burn area, which was both fenced and weeded, recruitment has been outstanding. One 20 x 20m quadrat recorded 58 native species regenerating where previously 16 main weed species and only 6 native species were present above ground. A total of 20 saplings and 43 seedlings of canopy species including Eucalyptus spp., Turpentine and Coachwood were recorded in this quadrat where the treatment involved weed removal, burning and fencing  (S. Brown, Ku-ring-gai Council, July 2019, unpublished data).  Unfortunately, however, the timing and location of the burn did not take into account its impact on the flying-fox camp and there was some damage to existing canopy trees. It will be many years before the canopy trees, which are regenerating, will be strong enough to support flying-foxes.

Monitoring from the weather station and data loggers has shown that close to Stoney Creek on a hot day it is typically 2-3° C cooler, and 5-10% higher in humidity, than in the current camp area (pers. comm. Tim Pearson). During heat events the flying-foxes move to this cooler and moister zone, increasing their chances of survival.

Fauna observed other than flying-foxes includes a pair of Wedge-tail Eagle ( Aquila audax plus their juvenile, a nesting Grey Goshawk (Accipiter novaehollandiae) and a Pacific Baza (Aviceda subcristata).  Powerful Owl (Ninox strenua) individuals continue to use the valley. The presence of raptors and owls indicate that the ecosystem processes appear to be functional. Despite the decline of the shrub layer outside fenced areas, the same range of small bird species (as seen prior to 2000) are still seen including migrants such as Rufous Fantail ( Rhipidura rufifrons) which prefers dense, shady vegetation. The first sighting of a Noisy Pitta (Pitta versicolor) was in 2014.  Long-nosed Bandicoot (Perameles nasuta) individuals appear and disappear, while Swamp Wallaby remains plentiful.

Lessons learned and future directions. Climate change is an increasing threat to Pteropus species. On the advice of Dr Eby, Flying-fox Consultant, Council, KBCS and Bushcare Volunteers agreed to retain all vegetation including weeds such as Large-leaved Privet and Small-leaved Privet, patches of the shrub Ochna (Ochna serrulata) and Trad as a moist ground cover in the camp area and areas used by the flying-foxes during heat events.

Building cheap, lightweight fencing can be effective against wallaby impacts, provided it is regularly inspected and repaired after damage caused by falling branches. This style of fencing has the additional advantage of being removable and reusable.  It has been proposed that, to provide understory vegetation to fuel future burns in parts of the reserve away from the flying-fox camp, further such temporary fencing could be installed.

Ku-ring-gai Council has commenced a  program to install permanent monitoring points to annually record changes in the vegetation, consistent with the state-based  Biodiversity Assessment Method.

Stakeholders and Funding bodies. Members of KBCS make donations, volunteer for monthly flyout counts, Bushcare and present educational events with live flying-foxes. KBCS hosts the website www.sydneybats.org.au. Ku-ring-gai Council which is responsible for the Reserve has been active in improving management to benefit both residents and flying-foxes.  Ku-ring-gai Environmental Levy Grants to KBCS have contributed substantially to purchase of fencing materials and the weather station. http://www.kmc.nsw.gov.au/About_Ku-ring-gai/Land_and_surrounds/Local_wildlife/Native_species_profiles/Grey-headed_flying-fox

Thank you to Jacob Sife and Chelsea Hankin at Ku-ring-gai Council for preparing the maps and to volunteer Pierre Vignal for assistance with tree measurements, downloading data loggers and a photo.  Researcher,  Tim Pearson installed the weather station.

Contact information. Nancy Pallin, Management Committee member, Ku-ring-gai Bat Conservation Society Inc.  PO Box 607, Gordon 2072  Tel 61 418748109. Email:  pallinnancy@gmail.com

Is revegetation in the Sheep Pen Creek area, Victoria, improving Grey-crowned Babbler habitat? – UPDATE of EMR feature

Doug Robinson

[Update of EMR feature Robinson, Doug (2006) Is revegetation in the Sheep Pen Creek area, Victoria, improving Grey‐crowned Babbler habitat?  Ecological Management & Restoration, 7:2, 93-104.  https://doi.org/10.1111/j.1442-8903.2006.00263.x]

Key words: (<5 words): Monitoring, restoration, population ecology, woodland conservation

Figure 1. Location of babbler project works and other landcare works implemented since 1996 in the Sheep Pen Creek Land Management Group area and the two sub-districts used for the babbler study. (Source TFNVic)

Introduction: The Grey-crowned Babbler (Pomatostomus temporalis) (babbler) is a threatened woodland bird (classified as Endangered in the state of Victoria) that has declined substantially in overall distribution and abundance across much of its former range in southeastern Australia since European settlement.  Sheep Pen Creek Land Management Group area, in northern Victoria (Fig 1), was fortuitously the location of the largest known remaining babbler population in Victoria in the early 1990s (when this project began); and the focus of extensive land restoration programs from the 1980s onwards to help mitigate the impacts of erosion and dryland salinity, as well as biodiversity decline.  The original study, published in 2006, investigated the overall changes in tree cover across the district between 1971 and 1996 as a result of different land-management actions and responses of local babbler populations to those habitat changes.  The key finding was that in the Koonda sub-district which had a 5% overall increase in tree cover to 14% from 1971 to 2001, showed an increase in babbler numbers by about 30% (Table 1).   In the Tamleugh sub-district, tree cover increased by 1.3% to a total of 9%, with no change in babbler numbers.  The findings also showed that new babbler groups were preferentially colonizing new patches of vegetation established that suited their habitat needs.  Building on this research, the study concluded that future conservation programs needed to scale-up the extent of habitat restoration, target areas which were suitable for babbler colonization, and tailor incentive programs to assist with conservation of particular species.

Table 1. Changes in Grey-crowned Babbler numbers over time

Year Koonda Tamleugh
number of groups number of birds number of groups number of birds
1992 20 78 11 39
1993 20 89 10 34
1996 24 96 9 35
1997 24 102 8 30
1998 25 99 10 40
2000 26 97 10 43
2005 23 99 8 34

Further revegetation works undertaken. Since the initial study’s assessment of vegetation changes between 1971 and 1996, an additional 133 ha of vegetation has been restored or established as babbler habitat in Koonda district and 37 ha in the Tamleugh district (Figs 2 and 3, Table 2).  Extensive natural regeneration, supplemented by broadscale revegetation, has also occurred over more than 350 ha on five private conservation properties in the Koonda district,, contributing to substantial landscape change.  The wider landscape has also been identified as a statewide priority for nature conservation on private land, leading to increased conservation investment in permanent protection there by Victoria’s lead covenanting body – Trust for Nature.

Monitoring of outcomes: The monitoring that was carried out prior to the 2006 publication has not continued, leaving a knowledge gap as to how the population has fared in the context of the Millenium Drought and ongoing climate-change impacts. However, based on the original research’s initial findings, we conducted an experimental study with University of Melbourne to evaluate the effectiveness of habitat restoration in maintaining babbler survival. The study, published by Vesk and colleagues in 2015, compared the persistence and group size of babbler groups present in 1995 and subsequently in 2008 at a randomly selected set of stratified sites which had either had habitat works or none.  This study was conducted across a larger landscape of about 200,000 hectares which included Sheep Pen Creek Land Management Group area.  The study found that babbler group size decreased by about 15% over the 13 years at sites without restoration works.   At sites with restoration, average group size increased by about 22%, thereby effectively compensating for the overall reduction in numbers reported over that time.This increase also influenced subsequent demographic performance, with groups at restoration sites having higher breeding success and more fledglings than groups at control sites.

Another useful finding from this experimental study was the confirmation of the importance of particular habitat and landscape variables on babbler persistence.  In particular, abundance of large trees was a positive predictor of occupancy over time; and distance from the next nearest group was a negative predictor.

Figure 2. Changes in tree cover in the Koonda sub-district between 1971 (top),  and 2018 (bottom). (Source TFNVic).. (Source TFNVic)

Figure 3. Changes in tree cover in the Tamleugh sub-district between 1971 (top) and 2018 (bottom). (Source TFNVic)

Table 2.  Summary of additional habitat established or restored as part of the Sheep Pen Creek Grey-crowned Babbler project from 1996-2018, following the initial study period from 1971-1996.

District Number of sites Area (ha)
Koonda 62 133
Tamleugh 28   37
Other parts of landcare group and local babbler population area 29 103
Totals 119 273

Expansion of lessons to other districts: Building on the fundamental research conducted in Sheep Pen Creek Land Management Group area, similar habitat, landscape and babbler population assessments were subsequently undertaken in northwest Victoria near Kerang for the babbler populations found there.  Key results from these studies relevant to the initial Sheep Pen study were that the number of babbler groups in each sampled district was positively related to the proportion of woodland cover, especially the proportion of Black Box (Eucalyptus largiflorens) woodland habitat – the babblers’ preferred habitat in this region.  Conversely, the number of babbler groups was negatively associated with the proportion of land under intensive agriculture.  At the site scale, key positive predictors of babbler presence in Black Box habitat again included the abundance of large trees (> 60 cm dbh)

Lessons learned and future directions: The most valuable lesson learned since the initial paper was published was the power of the structured research project described above to evaluate the effectiveness of the babbler conservation program and inform future design and planning. The study further demonstrated the importance of taking a demographic approach to the species’ conservation needs, understanding what is happening across the whole population over time  and how habitat interventions can assist.  These lessons have since been applied usefully to other babbler projects  and more broadly to conservation of woodland birds.

The initial paper noted the importance of achieving landscape-scale change in vegetation extent, particularly in more fertile habitats. This has occurred to some extent within the Koonda district through a range of incentive programs, tender programs, covenanting programs and land purchase, but continues to achieve most gains on more infertile land. On fertile land, by contrast, there has been rapid land-use change to cropping over the past fifteen years, leading to reduced likelihood of those properties providing suitable habitat for babblers, as found in the study conducted in northwest Victoria.

The initial paper also suggested the benefit of developing tailored incentive programs for babblers and other threatened species with particular requirements to maximize potential conservation gains  and we suggest, based on Australian and overseas experiences,  that more specific incentive programs or more detailed criteria could assist.

Another important lesson learned was the difficulty in maintaining community-driven citizen-science monitoring, even with the best will in the world, without some over-arching organizational support and oversight.  We know that community monitoring for biodiversity conservation needs scientific input at the design and analysis stages; hence additional resources may also be required in terms of equipment or guidelines to help groups monitor effectively.  Modest government investments to conservation organisations with established biodiversity monitoring programs could usefully help address this issue.

Finally, the learnings from the Sheep Pen Creek Land Management babbler conservation project over nearly thirty years are that the landscape changes and that these changes are not always positive.  Land-use change is placing more pressure on  potential babbler habitat; and the eucalypt regrowth which was established and provided new nesting resources for a few years is now too tall to provide nesting habitat, but too dense and immature to provide suitable foraging habitat for another one hundred years.  Climate change is rapidly imposing constraints on the availability of food resources and breeding opportunities, exacerbated by increased competition for the same limited resources by exotic and native species.  For the Grey-crowned Babbler, the solution to all of these factors depends on ongoing commitment to the establishment or maintenance of their essential habitat needs and life-history requirements so that their life-cycle is provisioned for from generation to generation.

Stakeholders and Funding bodies:   Most of the targeted habitat works achieved for babblers in this landscape has occurred through funding support from the Australian government through its Natural Heritage Trust and Caring for our Country programs.  Broader habitat protection and restoration has occurred primarily with funding support to landholders from the Goulburn Broken Catchment Management Authority (GBCMA).  The Norman Wettenhall Foundation, along with GBCMA, was instrumental in enabling the research by University of Melbourne, which was also aided by the extensive voluntary support of Friends of the Grey-crowned Babbler.  Not least, local landholders continued to support the project and continue to protect or restore parts of their properties to assist with babbler conservation.

Contact information: [Doug Robinson, Trust for Nature, 5/379 Collins Street Melbourne, Victoria 3000, Australia.  dougr@tfn.org.au, (03) 86315800 or 0408512441; and  School of Life Sciences, La Trobe University, Bundoora, Victoria, Australia.

 

 

 

 

Restoration of Wollongong’s Tom Thumb Lagoon 25 Years On: UPDATE of EMR feature.

 Nicholas Gill

[Update of EMR feature: Gill, Nicholas (2005) Slag, steel and swamp: Perceptions of restoration of an urban coastal saltmarsh. Ecological Management & Restoration, 6:2, 85-93 https://onlinelibrary.wiley.com/doi/10.1111/j.1442-8903.2005.00224.x]

Keywords. coastal wetlands, urban green space, pollution, mangroves, volunteers.

Figure 1. Tom Thumb Lagoon and Greenhouse Park (a) 2008 and (b) 2017. (Source Google Earth)

Introduction. The 2005 feature was drawn from restoration work my students and I became involved in during the early 2000s at Tom Thumb Lagoon (TTL) – an estuarine wetland close to Wollongong’s CBD and adjacent to the Port Kembla industrial area and harbour. By that point Wollongong City Council (WCC), the Bushcare group Friends of Tom Thumb Lagoon (FTTL), industry, Conservation Volunteers Australia (CVA), and many volunteers had been variously working on the site since the early 1990s. After decades of impacts from industrial development, waste disposal, and neglect, this significant restoration effort encompassed removing landfill, reshaping the wetland with channels and shallow benches, revegetation, weeding, and the construction of access and viewing points. By the time we became involved and I wrote the 2005 paper, TTL and the adjacent Greenhouse Park (GHP; Fig 1), were substantially revegetated, aesthetically improved, and the saltmarsh wetlands were seen as ecologically valuable. Participants and stakeholders in the restoration project perceived that substantial progress and improvement had been made. They also perceived, however, that the project suffered from some issues common to such endeavours such as a lack strategic planning and monitoring of ecological outcomes.

Since this time, restoration and other work has continued at TTL and at GHP. The story of what has happened, however, is one of the dynamic and contextual nature of sites such as this. This is true in a biophysical sense of ongoing vegetation change, particularly the spread of Grey Mangrove (Avicenna marina), a native plant previously not occurring on the site but planted for perceived environmental benefits either in the 1990s, or around 2000. This spread (into what was previously saltmarsh and mudflats) arises from past decisions and, while providing benefits, is now potentially causing new problems as well as continuing debates about choices in restoration.  The social context has also been dynamic and influential, as priorities have shifted, as the funding environment has altered, and as the people and groups involved have changed. Finally, Tom Thumb Lagoon remains affected by the legacy of the industrial history of its location. Past waste disposal practices in the absence of regulation have led to pollution problems that have become of greater concern since the early 2000s.

Activities at Tom Thumb Lagoon and Greenhouse Park Today. The wetland area itself is adjacent to a capped waste disposal site that operated from the 1940s until the mid-1970s. This area is known now as Greenhouse Park and is being managed and developed as urban green space with more focus on fostering urban sustainability practices; any restoration work is nested within these foci. TTL and GHP were always associated through overlap between FTTL and GHP staff, and GHP facilities were a base for TTL activities. Today, however, personnel have changed, FTTL no longer exists and its key members are no longer associated with TTL, and TTL/GHP are managed as one site to a greater extent. The result of these factors, and of the achievements already made at TTL, have been a shift towards an emphasis on activities at GHP and a change in TTL activities from active restoration to maintenance. It is now GHP volunteers and associated WCC staff who undertake and oversee work at TTL. At GHP WCC has expended considerable resources in tree planting and expanding a permaculture garden. There is a shelter, outdoor kitchen, and pizza oven for volunteers, WCC and Wollongong firms compost green and food waste, and there are hopes for public, tourism, and event use. Around ten volunteers work at the site weekly. For the GHP staff and volunteers, activities at TTL itself today are largely limited to weeding, picking up litter, and feral animal control. Weeds and litter remain problems, partly due to TTL’s location at the bottom of an urban catchment. In addition, since 2005, frog ponds were installed at the eastern end of TTL for the endangered Green and Golden Bell frog (Litoria aurea), however, it is not clear if the ponds are effective. The non-native Giant Reed (Arundo donax) also remains well established at this end of TTL despite control attempts.

Shifts in support have meant that CVA bowed out of work at TTL/GHP in 2012. Previously their involvement had been via a wetlands program that relied on support from both industry (including Bluescope and NSW Ports, both operating adjacent to TTL) and government programs. Until 2012, in conjunction with WCC, CVA were revegetating the southern slopes of GHP (marked A in Fig. 1) and were removing weeds and litter from the saltmarsh. However, the funding that CVA relied on declined such that CVA was unable to continue at TTL/GHP.

Figure 2.  Eastern end of TTL looking south (a) 2002 and (b) 2019 (Photos Nick Gill)

The Mangroves are Coming. Apart from further revegetation at GHP, the most significant vegetation change at TTL has been the spread of Grey Mangrove. While approval to thin this species has been obtained in the past and some thinning did occur, it has not mitigated their current spread and density. Grey Mangrove spread is clearly seen for the period from 2002 to 2019 in Fig 2 which shows the eastern end of TTL and the southern end of the channel known as Gurungaty Waterway. Aerial photos further reveal changes from 2008-2017 where the largely east-west spread of mangroves along channels in TTL can be seen (marked B in Fig 1). Significant spread can also be seen north-south spread along Gurungaty Waterway over this period (marked C in Fig 1). As the 2005 paper records, not long after Grey Mangrove was planted in the late 20th or early 21st Century, its expansion was  soon causing concern for its consequences for the site’s mudflats, saltmarsh and tidal habitats although it appears to have largely remained confined to the channels and has no doubt generated some environmental benefits. In terms of its consequences on bird habitat, the long observations of local birdwatchers suggest that the expansion of Grey Mangrove has reduced the incidence of waders and shorebirds, particularly Black Winged Stilts (Himantopus himantopus) and also waterfowl and herons. Nonetheless, observers report that Grey  Mangrove colonisation is providing habitat for other birds, such as the Sacred Kingfisher (Todiramphus sanctus), the Nankeen Night-Heron (Nycticorax caledonicus), and the Striated Heron (Butorides striata). Elsewhere across more upland areas of TTL and GHP, the expansion of tree planting across GHP and TTL has seen a shift to birds favouring woodland habitats.

The expansion of Grey Mangrove is also implicated in flood risk, especially for the catchment of Gurungaty Waterway. A 2019 review of the Wollongong City Flood Study, suggests that low elevations and channel infrastructure, combined with sedimentation and flow limitations associated with the now dense mangroves (Fig. 3), have increased the likelihood of flooding in the urban catchment.

Figure 3.  Southern Gurangaty Waterway in (a) 2002 and (b) 2019. Note the steel footbridge on left of each photo. (Photos Nick Gill)

Industrial Legacies. The 2005 paper notes that saltmarsh restoration was an important part of the TTL work and that stakeholders saw the saltmarsh as a significant ecological element of TTL. Since 2004 coastal saltmarsh has been listed as an Endangered Ecological Community in NSW. From 2006, saltmarsh degradation prompted WCC to monitor the saltmarsh and analyse groundwater and soils.  This showed that the degradation was likely associated with ammonia leaching from the tip and causing nitrate pollution, and also with a hydrophilic layer of iron hydroxide in the soil causing waterlogging and contaminant absorption. The possible origins of this layer include past waste disposal practices from metal manufacturing.

These, however, are not the only legacies of past unregulated waste disposal and industrial activity. TTL is now a declared site of ‘significantly contaminated land’ by the NSW EPA. The 2018 declaration notes that site is contaminated by ‘polycyclic aromatic hydrocarbons (PAHs), petroleum hydrocarbons and other mixed contaminants from multiple sources including coal tar and lubricant oils’. At TTL elements of these can be visible as a film on the water surface and are among the substances leaching from GHP. Such substances are carcinogenic and exposure can cause a range of health problems. The presence of these materials in the groundwater has been known since the 1990s but from 2013 WCC began to monitor and map these materials. Monitoring points were installed along the wetlands at base of the old tip. Various remediation options for these contaminants, as well as for the nitrates and iron hydroxide layer, were proposed but action was not taken at this time for various reasons including disruption to the wetland, costs, and uncertainties regarding pollutant interception. As of 2019, the site is subject to a ‘Voluntary Management Proposal’ by WCC which includes the preparation of a remediation action plan by late 2019.

Future Directions. The last fifteen years have seen some aspects of restoration, such as tree planting, proceed and expand. By some measures this is continued progress of the original project. TTL/GHP is now a well-established urban green space with environmental and amenity value. However, concerns from the early 2000s about volunteer succession, the absence of a catchment approach to management, and the need to think more strategically about ecological trade-offs between management options have been realised to some extent. The spread of Grey Mangrove is the clearest example of this. In part, some of this is perhaps inevitable for a site with the history and setting of TTL/GHP; the management context has changed, participants and stakeholders have changed, and difficult legacy issues have assumed greater prominence and cost. Nonetheless, the challenge to manage the site with a clear strategy and goals remains.

Acknowledgements: For assistance with this update, I am indebted to several past and present WCC staff, particularly Mike McKeon. I was also helped by Adam Woods, formerly of CVA, and birdwatchers Penny Potter, Terrill Nordstrom, and David Winterbottom.

Contact. Nicholas Gill, School of Geography and Sustainable Communities Faculty of Social Sciences, University of Wollongong NSW 2522 Australia, Email: ngill@uow.edu.au

The biodiversity benefits of Greening Australia’s Saltshaker Project, Boorowa, NSW – UPDATE of EMR feature

David Freudenberger, Graeme Fifield, Nicki Taws, Angela Cailiss and Lori Gould

[Update of EMR feature – Freudenberger, David, Judith Harvey and Alex Drew (2004) Predicting the biodiversity benefits of the Saltshaker Project, Boorowa, NSW. Ecological Management & Restoration, 5:1, 5-14. https://doi.org/10.1111/j.1442-8903.2004.00176.x]

Key words: woodland restoration, monitoring, farmland rehabilitation, community engagement

Figure 1. Boorowa River Recovery project sites, south eastern NSW.

Introduction

The Boorowa catchment in central NSW, like most of the wheat-sheep belt of eastern Australia, has been extensively cleared for agriculture.  Remnant woodland cover is less than 10% and highly fragmented into small patches, often less than 20 ha. As described in the 2004 article, there has been a documented decline in biodiversity across this region linked to declines in landscape function including dryland salinity and eucalypt dieback. In response to these declines, farmers in this catchment have been involved in land rehabilitation projects for over 25 years.  Many of these projects have been facilitated by Greening Australia, a national non-governmental organisation focused on protecting and restoring native vegetation.  Pioneering projects in the 1990s were often small in scale and lacked landscape scale targets.  To address this, Greening Australia collaborated with CSIRO to develop guidelines for catchment scale “enhancement activities” for the $1.8 Million “Saltshaker Project” that carried out ground works as described in Box 1 of the 2004 article (reproduced below). The project was based on a $845,000 grant from the Australian Government’s Natural Heritage Trust program and $1 Million in in-kind support from farmers, the Boorowa Shire, Boorowa Landcare and Greening Australia. This project ran for just two years (2000-2002), but it was hoped that the project would provide strategic guidance for decades to come.  This appears to be the case.

 Box 1. Priority ‘enhancement activities
1. Protect existing remnant vegetation by fencing out domestic livestock with a priority to protect 10 ha or larger remnants in the best condition (complex understorey).
2. Establish native understorey plants in those protected remnants requiring enhancement of habitat complexity.
3. Enlarge existing remnants to at least 10 ha.
4. Create linkages between fenced remnants and other protected remnants. Linkages should be at least 25 m wide, or 10 ha stepping-stones, particularly in those areas more than 1.5 km from other patches 10 ha in size.
5. Fencing and revegetation of at least 50 m wide along creeks and flow lines.
6. In recharge areas, revegetate in 2-ha blocks, or greater than eight row strips to intercept deep soil water moving down-slope.
7. Revegetate areas mapped as having a high risk of dryland salinity.
8. Block plantings in discharge areas with links to other saline reclamation works.

(Box reproduced with permission from the original feature]

During the Saltshaker project, bird surveys were conducted within 54 discrete patches of remnant woodland.  Bird species were identified that were particularly sensitive to loss of habitat area, simplification of habitat structure, and increase in habitat isolation. The Eastern Yellow Robin was the focal species for this catchment. It generally occurred in woodland patches larger than 10 ha that were no more than 1.5 km from other patches at least 10 ha in size, and had at least a moderate structural complexity made up of a healthy overstorey of eucalypts with an understorey of regenerating trees, shrubs, tussock grasses and fallen timber. The Saltshaker project predicted that many other woodland birds would co-occur if the habitat requirements of the Eastern Yellow Robin were met by patch and landscape scale enhancement activities.

Further works. The Saltshaker project was followed by many others since 2002. The largest project was “Boorowa River Recovery” that began in 2005 as a partnership managed by Greening Australia with the Lachlan Catchment Management Authority and the Boorowa Landcare Group.  Through a total investment of almost $2.2 million (in-kind included), this project rehabilitated or protected 640 ha of riparian area along 80 km of river including a continuous 29 km stretch of the Boorowa River above the town water supply dam (Figs 1 and 2). It involved more than 60 land managers who implemented on-ground works described in individual ten year management contracts. On-farm project size averaged 11.6 ha.

Other projects funded by a diversity of sources, particularly the Australian Government, have protected an additional 88 ha of woodland remnant, enhanced 353 ha of remnants, and revegetated 425 ha of native vegetation within the catchment.  Projects included Whole of Paddock Rehabilitation (WOPR).  All project activities linked to funding have been recorded in a detailed project management database held by Greening Australia. These additional projects were consistent with the enhancement activities recommended by the Saltshaker Project and described in the EMR feature.

Figure 2 (a) Before and (b) after willow removal in the Boorowa River Catchment. After willow removal, all sites were planted to a diversity of trees and shrubs.

Outcomes. There has been no comprehensive follow-up to the 2001 bird surveys across the Boorowa Catchment.  However since then, there is now a large and comprehensive scientific literature demonstrating dramatic increases in woodland birds in the revegetation areas in this region of southeastern NSW (reviewed in 2018). Most all the conservation and restoration activities in this catchment have likely led to an increase in woodland birds over the past 20 years.

Of all the Boorowa projects, the Boorowa River Recovery projects had sufficient funding for monitoring outcomes six years after project activities commenced. A sub-sample of 20 sites out of a pool of 47 were monitored for improvements in vegetation cover and density, macroinvertebrate abundance and stream bank stability. Planted shrub cover generally doubled at all sites as expected. Macroinvertebrate scores did not differ between treated and control sites, though activities did appear to improve stream bank stability (an indirect measure of reduced erosion).  Subsequent monitoring 12 years on showed further improvements in ecosystem function.

Since the Saltshaker Project finished, there has been no systematic monitoring of the hundreds of woodland remnants protected and enhanced by this project and subsequent ones.  However, landholders and staff anecdotally report indicative improvements in vegetation cover and wildlife habitat on the sites, and we can infer from a 2008 study that included woodland sites in the Boorowa Catchment, that significant ecological improvements are likely from fencing out livestock from woodland patches. This study found improvements included greater native floristic richness, native groundcover and overstorey regeneration within fenced sites compared to unfenced sites. Similarly, a 2009 study found that woodland sites in south eastern Australia, with livestock grazing removed, had a greater abundance of beetles and the opportunist ant functional group, a faster rate of litter decomposition, greater native plant richness, greater length of logs, and a better vegetation condition score.

Lessons learned. Long-term action with short-term funding. Natural resource management projects have been ongoing in the Boorowa catchment for over 25 years. But no single project has been funded for more than five years. This is the reality of natural resource management (NRM) in much of Australia.  Government NRM programs come and go with election cycles, but fortunately the commitment of landholders and local organisations persists.

Partnership model. All the projects before and after the Saltshaker Project have involved landholders working collaboratively with local agencies administering the diversity of funding. Most projects had a steering committee that proved a good way for stakeholders to have input through all stages of project, which was particularly important during project planning. Idealism needed to be balanced with practicality so bureaucracy was minimised while maintaining accountability. Good communication that recognised that no single view was more valuable than another was important, although full consensus was not always possible. Trust was enabled when processes were developed collectively. Skilled coordinators needed a clear understanding of their roles and care taken to not get involved in local politics.

Assessing outcomes. Developing a highly predictive understanding of ecological outcomes from NRM activities in catchments like Boorowa is a scientifically complex, expensive and long-term process. The confidence we can now claim for an increase in abundance and diversity of woodland birds in the Boorowa catchment stems from two types of monitoring. First is project monitoring of outputs like the 425 ha of revegetation known to have been established in the catchment. We know this from Greening Australia’s project management database (unfortunately there is no national database for this kind of outputs),  although satellite imagery should be able to pick up this output once plantings have a dense enough canopy. It is essential to know when and where project outputs like revegetation have occurred in order to then design scientifically rigorous studies to research ecological outcomes like increases in flora and fauna diversity and abundance. We have confidence that wildlife is colonising revegetation because research groups have conducted sophisticated statistical analyses of wildlife data from woodland revegetation in nearly 200 sites across south eastern Australia for over 15 years (summarised in a 2018 study).

Gaps in understanding. We know a lot about the ecological and social outcomes of NRM activities, but much less about improving the cost effectiveness of outputs such as revegetation and understory enhancements(see 2016 review). There are no recent published benchmarks on how much revegetation should cost in the face of variable climatic conditions, soil types and terrain.  More revegetation case studies need to be documented, but they need to include an accounting of costs.  The Australian restoration challenge is vast, funding always limited, so practical research and transparent accounting is sorely needed to reduce the cost of ecologically effective restoration.

Continuous re-learning. The many and diverse projects in the Boorowa Catchment are typical of NRM activities in Australian woodlands over the past 25 years. Each project has involved different agencies, many no longer exist or have changed their names (e.g. Catchment Management Authorities have morphed into Local Land Services in NSW). Each agency, including NGOs like Greening Australia, have a natural turn-over of staff. For example, only one staff member of Greening Australia involved in Saltshaker remains with the organisation.  Landholders tend to remain longer, but they too retire, sell out, and move on. Like education, every new staff member and every new landholder needs to learn the complex processes of successful catchment repair. This learning needs to be hands-on, hence funding for NRM activities and extension is needed in perpetuity (just like education). But experiential learning needs to be complemented with a diversity of learning resources such as the EMR journal, easily assessable reports (too many have disappeared from Government websites) and new media such as YouTube videos. Most importantly, communities of practice need to be perennially nurtured by a diversity of practitioners, experienced and less so.  There is much still to be learned and shared.

Stakeholders and Funding bodies:   The primary funding bodies for projects in the Boorowa catchment were the Australian Government, TransGrid, Alcoa Australia, the NSW Environmental Trust, and the former Lachlan Catchment Management Authority. These external funds were complemented by a diversity of in-kind support provided by farmers, Boorowa Shire Council, and other community members of the catchment.

Contact details. David Freudenberger, Fenner School of Environment and Society (Australian National University, Canberra, 0200, Australia, Email: david.freudenberger@anu.edu.au). GF, NT and AC can be contacted at Greening Australia, Kubura Pl, Aranda ACT 2614, Australia; and LG at GrassRoots Environmental, Canberra (http://www.grassrootsenviro.com/)

 

 

Ecological Restoration of Donaghys Corridor, Gadgarra, north Queensland – UPDATE of EMR feature

Nigel Tucker

[Update of EMR feature – Tucker, Nigel I. J. and Tania Simmons (209) Restoring a rainforest habitat linkage in north Queensland: Donaghy’s Corridor, Ecological Management & Restoration, 10:2, 98-112, https://onlinelibrary.wiley.com/doi/10.1111/j.1442-8903.2009.00471.x%5D

Keywords: Rainforest, corridor, regeneration, disturbance effects

Introduction. Complex notophyll vine forests of the Atherton Tablelands, particularly from basalt derived soils, have been significantly fragmented and degraded by human settlement over a 100yr period. Fragment isolation results in edge effects, exotic species colonisation, loss of genetic variability and species decline. During high rainfall events, eroding streambanks on farms mobilise sediments to the receiving environment of the Great Barrier Reef. Re-connecting isolated fragments to larger forest blocks through restored riparian corridors aims to reverse these effects through adaptive management. The restoration of Donaghys Corridor is an example of adaptive management, and its establishment was a key factor in the adoption of other local corridor projects.

As reported in the 2009 features, around 20,000 plants of selected local species were established in four yearly plantings (1995/96/97/98) along Toohey Creek, creating a continuous habitat corridor between the isolated Lake Barrine fragment (500ha) and the adjacent Gadgarra section of Wooroonooran N.P (80,000ha), both being part of the Wet Tropics World Heritage Area. The corridor is 1,200m in length and 100m wide, with three rows of Hoop Pine (Araucaria cunninghamiana) planted either side of the fenced corridor, which was established on lands largely owned by the Donaghy family. On completion, the corridor was secured through the Queensland Government’s declaration of Donaghys Corridor Nature Refuge, the State’s first Nature Refuge proclaimed over an ecologically restored site.

Ongoing recovery. In 2000, a vegetation survey of 3m x 5m plots in 12 permanent transects throughout the corridor showed regeneration had occurred upon canopy closure (Tucker and Simmons 2009).  Between 1995 and 1998, 119 native species had regenerated within the transects, mainly through vertebrate-mediated dispersal. The most recent (ongoing) survey, ca.20yrs after planting, indicates that regeneration has continued, and the majority of regenerating species are again vertebrate dispersed. There has also been a measurable increase in vegetation structural complexity, and a variety of life forms are present including ferns, orchids, vines, scramblers and canopy trees.

Restored vegetation in 2000 was characterised by vegetation of even age and size classes and only a developing canopy was present (no sub-canopy). Recruitment was limited to the ground storey. Over 20yrs, total numbers of recruiting species have increased, along with canopy height, and the sub-canopy is now a distinguishable and measurable feature. To illustrate this change, species diversity and structure in two typical transects from the oldest (1995) and youngest (1998) plantings are shown in the table below. Figures are from the most recent survey (2019) and the bracketed numbers indicate comparative values in 2000.

Canopy

height

Sub-canopy

Height

Number of species Average number of species/plot Average number of species/plot – sub-canopy Average number of species/plot – ground storey
1995 19.9 (5) 7.5 (0) 84 (53) 22.6 (12.5) 8.3 13.8
1998 14.4 (2.5) 7.3 (0) 63 (15) 14.2 (1.6) 2.2 15.8

There has also been a significant difference in the distribution of regenerating vegetation. In 2000, regeneration was negatively correlated with edge, being concentrated in the central portion of each transect. Greater structural complexity and increased shading have significantly reduced the edge effect and regeneration is now distributed equally across the entire width of the corridor. This edge-effect reduction may partially result from the three Hoop Pine rows, now ca.15m tall, planted on each side of the corridor.

Figure 1.  Part of the 18m x 250m fence crossing Donaghys Corridor

Natural and man-made disturbance. Since establishment there has been both natural and anthropogenic disturbance. Occasional incursions by cattle have occurred, entering via fences sometimes damaged by branches falling from maturing corridor vegetation. In small areas incursions have visibly damaged regeneration but surveys show this has not significantly affected regeneration. Feral pig disturbance has also occurred but does not appear to have affected regeneration.

In 2006, corridor vegetation was damaged by severe tropical Cyclone Larry. Most stems lost crowns and some waters’ edge stems were permanently bent by floodwaters, but vegetation recovery was rapid and no weed invasion occurred. This infers a measure of resilience by restored vegetation to disturbance, and the distribution of regeneration described above supports this inference.

Anthropogenic disturbance has been more interventionist and not aligned to the original concept adopted by government, landholders, scientists and the community when the project commenced in 1995.  In 2017, the corridor’s upstream neighbour, with support from the DES but without consultation with the Donaghy family or other affected landholders, erected a chain mesh fence 250m long and 1.8m high across the western end of the corridor (see Figure 1). This is part of a larger fence which completely encloses mature forest at the western end of the corridor, including corrugated iron placed across the bed of Toohey Creek. Enquiries revealed the fence is part of an enclosure for a Cassowary (Casuarius casuarius johnsonii) rehabilitation facility, operated by Rainforest Reserves Australia (RRA) under a commercial arrangement with the Queensland Government.

Enhancing landscape permeability was the key reason for undertaking the Donaghys Corridor project, and the endangered Cassowary was a key target species; 53 Cassowary food plants were included in the original planting matrix of 100 species to encourage corridor utilisation. The Queensland Government notes that corridors are a key strategy in Cassowary conservation. In addition to blocking the movement of terrestrial vertebrates such as Cassowaries, Pademelon (Thylogale stigmatica) and Musky Rat Kangaroo (Hypsiprymnodon moschatus), construction of the enclosure has inadvertently fenced in a number of animals whose territories included part of the enclosure.

DES has advised that the fence is temporary and will be removed when restoration plantings on RRA lands are ‘sufficiently well-developed’ to support Cassowaries being rehabilitated.  It is unknown, however, when or through what processes this removal will occur. Resolution of the issue is anticipated.  However, such actions highlight the pitfalls associated with single-species conservation, and potential conflicts that might arise when responsibility for management of endangered species moves from the State to the non-scientific, commercially-focused private sector. Whilst iconic wildlife e.g., the Cassowary, can be effective in harnessing community and landholder participation in restoration, here it is clear that decision making and communication has been far from optimal, which may well lead to landholder and community disillusionment. In this case, the fence has also disrupted ongoing monitoring and evaluation. Planned re-survey of terrestrial vertebrate colonisation and movement has now been cancelled, given the unknown effect of the fence on wildlife passage and the behaviour of animals inadvertently trapped within the enclosure.

Lessons learned.  The project shows that sustained regeneration of native species can be achieved in restored tropical vegetation, along with increased structural complexity and functional resilience to natural disturbance.  However, the fencing incident shows that dysfunction in a restoration project can arise from totally unanticipated causes, potentially undoing well-established partnerships between government, community, scientists and landholders.

Contact.  Nigel Tucker, Director & Principal Environmental Scientist, Biotropica.  PO Box 866 Malanda QLD 4885 ; Email: nigeltucker@biotropica.com.au; Tel: +61 7 4095 1116.

 

 

 

Monitoring the Wunambal Gaambera Healthy Country Plan, Kimberley, Western Australia – UPDATE of EMR feature

[Update to EMR feature: Moorcroft, Heather, Emma Ignjic, Stuart Cowell, John Goonack, Sylvester Mangolomara, Janet Oobagooma, Regina Karadada, Dianna Williams and Neil Waina (2012) Conservation planning in a cross‐cultural context: the Wunambal Gaambera Healthy Country Project in the Kimberley, Western Australia,  Ecological Management & Restoration, 13:1, 16-25. See https://doi.org/10.1111/j.1442-8903.2011.00629.xk]

Key words: conservation planning, participatory conservation, Indigenous people, Kimberley

Figure 1. Location map of Wunambal Gaambera land and sea country. (Source: WGAC)

Introduction. The development of the Wunambal Gaambera Healthy Country Plan (HCP) was a key enabler for Wunambal Gaambera people to look after country (Fig 1) and occurred at an important time when native title rights to country were being secured. The plan came about through a partnership between Wunambal Gaambera Aboriginal Corporation (WGAC) and Bush Heritage Australia (BHA), who brought the planning approach and supported WGAC to develop a plan that met the needs of Traditional Owners. This plan then became the basis of a long term partnership between the two organisations to support implementation.

Healthy Country Planning, a term coined by this project, adapted the Open Standards for the Practice of Conservation (http://cmp-openstandards.org/ ) to an Indigenous native title community context. The approach has subsequently flourished in Australia, adopted by a further 30 indigenous groups ( see doi: 10.1111/emr.12267).

The unique partnership model established between BHA and WGAC, underpinned by Healthy Country Planning, has also expanded (https://www.bushheritage.org.au/what-we-do/aboriginal-partnerships). The program supports indigenous partners to achieve their community and conservation aspirations articulated through Healthy Country Planning.

Figure 2. Terrick Bin Sali handling a northern quoll (Dasyurus hallucatus). (Photo WGAC)

Conservation planning with a difference. Our experience has been that the Open Standards can be successfully applied to an Indigenous context but some important adaptation is required. For example:

  1. People, culture and country are inseparable in Australian indigenous worldviews. As such, Healthy Country Planning is much wider in scope than mainstream conservation or natural resource management planning. The vision in the HCP typically sets long-term goals that include aspirations for looking after country and achieving health conditions for landscapes, seascapes, plants and animals, as well as for people and culture (Figs 2-4). These outcomes are collectively referred to as “Healthy Country”. An indigenous plan will always recognize and value people in the landscape rather than as separate. Traditional Owners, and/or their culture, become a conservation target alongside species and habitats with ‘key cultural attributes’ established alongside ‘key ecological attributes’. HCPs also have a greater degree of management strategies that relate to people and culture than would be found in mainstream management plans, and culture and Indigenous Knowledge is incorporated into land management activities that could be considered mainstream, such as the inclusion of cultural rules within visitor management.
  2. The Open Standards include approaches and tools for stakeholder participation that lend themselves well to Indigenous peoples, however given that Indigenous lands are communally-owned and governed by unique law and culture systems, participation requirements are higher and more complex. Traditional systems of governance are often recognized under State and Federal Law (such as the Native Title Act) requiring additional process steps. In developing the HCP, WGAC undertook a process of involving all Traditional Owner families in various stages of planning and the plan was authorized at a Traditional Owner meeting.
  3. A Healthy Country Plan typically applies to the whole traditional estate of a Traditional Owner group regardless of tenure, consistent with the concept of “Country-based Planning”. WGAC initially dedicated stage I of the Uunguu Indigenous Protected Area over several aboriginal reserves in 2010, and later added exclusive possession lands following native title determination. The marine environment will also be added to the IPA once agreement is reached with State and Commonwealth Governments, as articulated in the Uunguu Wundaagu Plan.

Figure 3. Traditional Owners undertaking a ‘junba’ traditional dance workshop. (Photo: WGAC)

Monitoring & evaluation since 2012. Aspects of the monitoring and evaluation framework established by WGAC include the establishment of the Uunguu Monitoring & Evaluation Committee (UMEC) and the completion of mid-term evaluation of the HCP (see 2017 review doi:10.1111/emr.12257). The UMEC is composed of Traditional Owners and external experts who undertake key Healthy Country Planning steps and functions. A significant investment of time and money was required to support annual or biennial meetings “on country” (ie. on Wunambal Gaambera lands), however the review showed that the investment has proven worthwhile because WGAC has been able to implement the plan to a high degree.

In 2015 a major mid-term evaluation was undertaken to assess the progress of the HCP. It utilized some standard evaluation tools examining Open Standards process. It also incorporated some new approaches to include the views of the wider Traditional Owner community in the evaluation of the plan to ensure the plan was meeting their needs and following effective process with regard to governance and participation. The review made a number of observations as follows.

  1. Considerable progress had been made in the implementation of fire management strategies in the HCP to the point that the health condition of the “Right Way Fire” target was changed from ‘fair’ to ‘good’. Unmanaged wildfires have significantly reduced in size as a result of increased capabilities of the Uunguu Rangers to undertake burning. At the same time “right way” cultural principles have been incorporated into operations to ensure that Traditional Owner families are making decisions about burning and undertaking fire operations on their family areas.
  2. Significant progress has been made against a visitor management strategy. One of the key concerns of elders when developing the plan was unregulated visitation to important cultural sites by the expedition cruise industry and independent travelers. The Uunguu Visitor Pass, launched in 2017, requires visitors to obtain a permit to access authorized visitor locations as well as generating funds for looking after country. Over 90% of commercial tour operators have now registered for the UVP, with a similar independent traveler compliance, generating funds to support Traditional Owner participation in visitor management and delivering tour products.
  3. Progress was not made, however, in the health of cultural targets in the plan. As a result, the 2015 review recommended further investment and effort in culture strategies. Two important books have been completed to document elders’ knowledge of biodiversity and cultural places. There has been an increase in cultural activities including language and corroboree dance workshops and annual culture camps for school children. Law and culture dictates that visitors to country (indigenous or non-indigenous) need to be accompanied by the right local Traditional Owners when undertaking activities on country. In this way Healthy Country work supports large numbers of Traditional Owners to visit country and guide participation in the implementation of the plan. A number of actions have been undertaken to support Traditional Owners to live on country and remote ranger infrastructure has been developed.

Figure 4. Uunguu Rangers during a cultural fire walk. (Photo WGAC)

Future Directions. The first 8 years of the HCP implementation has focused on building WGAC’s land and sea management capacity and resources, with funding from WGAC,  the Australian Governments Indigenous Protected Area and Indigenous Ranger Programs and from Bush Heritage Australia. Traditional Owners currently hold certificate level roles of rangers, tourism workers and construction workers but further work is now needed to support Traditional Owners to take on roles that require higher levels of training. There is also a need to support the development of a wider range of livelihoods to support Traditional Owners to live and/or work on country.

Contact information. Wunambal Gaambera Aboriginal Corporation, PMB 16 Kalumburu via Wyndham WA 6740, Australia.

Email: info@wunambalgaambera.org.au

Web: www.wunambalgaambera.org.au

Facebook: https://www.facebook.com/uunguulife/

 

 

 

 

 

 

Developments in Big Scrub Rainforest Restoration: UPDATE of EMR feature

Tony Parkes, Mark Dunphy, Georgina Jones and Shannon Greenfields

[Update of EMR feature article: Parkes, Tony, Mike Delaney, Mark Dunphy, Ralph Woodford, Hank Bower, Sue Bower, Darren Bailey, Rosemary Joseph, John Nagle, Tim Roberts, Stephanie Lymburner, Jen Ford and Tein McDonald (2012) Big Scrub: A cleared landscape in transition back to forest? Ecological Management & Restoration 12:3, 212-223. https://doi.org/10.1111/emr.12008]

Key words: Lowland Subtropical Rainforest, ecological restoration, seed production, landholder action, corridors

Figure 1a. Rainforest regenerators undertake camphor injection, leaving bare trees standing creating light and an opportunity for seed in the soil to naturally regenerate. (Photo © Envite Environment)

Figure 1b Aerial photo showing camphor conversion by injection
(Photo © Big Scrub Regeneration Pty. Ltd.)

Introduction. The Big Scrub, on the NSW north coast, was once the largest tract of Lowland Subtropical Rainforest (LSR) in Australia. It was reduced to less than 1% of its original extent by he end of hte 19th century after clearing for agriculture. Big Scrub Landcare (BSL) is a non-profit organisation dedicated to improving the long-term ecological functionality of what remains of this critically endangered ecosystem –  lowland subtropical rainforest.  Our 2012 EMR feature reported on remnant restoration and revegetation works overseen by BSL to 2012. At that time, 68 remnants were identified as significantly affected from the impacts of environmental degradation including weed invasion and cattle access. These remnants had been undergoing treatments, with 20 substantially recovered and on a ‘maintenance’ regime.  Approximately 900,000 trees had been planted to establish 250 ha of young diverse well-structured rainforest.  A comparatively small area of forest dominated by the highly invasive exotic, Camphor Laurel (Cinnamomum camphora) (Camphor), which  has colonised much of the Big Scrub landscape had been converted to early phase LSR by skilled removal of a range of weeds and facilitating natural regeneration. 

Progress since 2012. Substantial progress in restoring critically endangered lowland subtropical rainforest in the Big Scrub has been achieved over the past seven years in the following areas.

  • Assisted regeneration of remnants has continued and become more focused
  • Re-establishment of LSR through plantings has expanded
  • Camphor conversion has developed in scale and techniques
  • Greater security of funding has been achieved
  • Community engagement has greatly improved and expanded
  • Genome science is being applied to produce seed with optimal genetic diversity for rainforest restoration.

Assisted regeneration of remnants. This work continues to be the major focus of on-ground restoration work. About 2000 regenerator days (9 years Full Time Equivalent) of work has been undertaken in 45 remnants. BSL’s remnant restoration program has become more strategic, with more focus on Very High Conservation Value (VHCV) remnants, particularly those in the NSW National Parks Estate, including the VHCV sites in Nightcap National Park (NP) including Big Scrub Flora Reserve, Minyon Falls and Boomerang Falls; Andrew Johnston’s Scrub NR; Snow’s Gully Nature Reserve (NR); Boatharbour NR; Victoria Park NR and Davis Scrub NR, plus the Booyong Flora Reserve. Rehabilitation work at these sites is prioritised in the major new four-year Conservation Co-funding project funded jointly by BSL and the NSW government’s Saving our Species program. Big Scrub Foundation (BSF) funding has enabled BSL to continue maintenance work in remnants that have reached or are approaching the maintenance stage.

Monitoring outcomes has become more rigorous and has demonstrated ongoing improvements in vegetation structure, with decreasing levels of weed invasion and improvements in native species cover.

BSL’s partner Envite Environment, with some assistance from BSL, is creating an important linkage between Nightcap NP and Goonengerry NP by the restoration of rainforest through the progressive removal of weeds that had dominated the 80 ha Wompoo/Wanganui corridor between these two NPs.

 Re-establishment of rainforest by planting. The area of LSR is being re-established by planting on cleared land has also continued to expand.   In the last 7 years  more than 0.5 million rainforest trees have been planted in the Big Scrub region, contributing to the restoration of another 175 ha of LSR, expanding total area of re-established rainforest by another 13%. While landscape-scale landholder driven work is inevitably opportunistic rather than strategic, the establishment of new patches of LSR enhance valuable stepping-stone corridors across the Big Scrub. Since 2012 the number of regenerators working fulltime in the Big Scrub region has increased by approximately 50%.  Another trend that has strengthened in the last 7 years is that larger plantings are now being carried out by well-resourced landowners. This is accounting for about 40% of the annual plantings. Offsets for residential development account for another 40% of trees planted. The remaining 20% is made up by small landowners, cabinet timber plantations, large-scale landscaping, and other planting of Big Scrub species. This is a significant change from the more dominant grant-based small landowner/Landcare group plantings prior to 2012.

 Camphor conversion. Larger areas of Camphor forest are being converted to rainforest, with project areas increasing substantially from less than a hectare to ten and twenty hectares. BSL estimates that more than 150 ha of Camphor forest are currently under conversion. Some landowners underake camphor injection which leaves bare trees standing, creating light and an opportunity for existing native seedlings and seed in the soil (or seed dropped by perching birds) to naturally regenerate (Fig 1). Others are choosing the more expensive option of physically removing the Camphor trees and carefully leaving the rainforest regrowth (Fig 2).  Improved techniques and landholder capacity building continue to progress and camphor conversion is now a significant component of rainforest restoration.

BSL alone is facilitating the conversion of almost 40 ha of Camphor forest to LSR funded by two 3-year grants from the NSW Environmental Trust, together with contributions from the 19 landholders involved in these projects. The ecological outcomes being achieved are significant and less costly than revegetation via plantings.

Figure 2a. Camphor forest under conversion using heavy machinery leaving rainforest regrowth intact (Photo © Big Scrub Landcare)

Figure 2b. Aerial photo showing camphor conversion by removal
(Photo © Big Scrub Landcare)

Greater security of funding. Australian Government funding for biodiversity conservation is at a very low level. Competition for existing NSW state government funding is increasing. BSL therefore has continued to  develop new strategies for fund raising to ensure continuity of its long-term program for the ecological restoration of critically endangered LSR in the Big Scrub and elsewhere. Ongoing funding of at least $150,000 annually is needed to ensure the great progress made  over the past 20 years in rehabilitating remnants is  maintained and expanded to new areas of large remnants. These funds finance weed control and monitoring; weeds will always be a part of the landscape and an ongoing threat to our rainforest remnants.

Establishment of the Big Scrub Foundation in 2016 was a major development in BSL’s fund raising strategy. The Foundation received a donation of AUD $1M to establish a permanent endowment fund that is professionally invested to generate annual income that helps finance BSL’s remnant care program and its other activities. Generous donors are also enabling the Foundation to help finance the Science Saving Rainforest Program.

Figure 3a. Australian gardening celebrity Costa Gregoriou at a Big Scrub community tree planting (part of the 17th annual Big Scrub Rainforest Day) in 2015 (Photo © Big Scrub Landcare)

Figure 3b. Founder of the Australian Greens political party Bob Brown and Dr. Tony Parkes at the 18th annual Big Scrub Rainforest Day in 2016. (Photo © Big Scrub Landcare)

Community engagement. The  Big Scrub Rainforest Day continues to be BSL’s  major annual community engagement event, with the total number of attendees estimated to have exceeded 12,000 over the past 7 years; the 2016 day alone attracted more than 4000 people (Fig 3). Every second year the event is held at Rocky Creek Dam.  A new multi-event format involving many other organisations has been introduced on alternate years.

BSL’s Rainforest Restoration Manual has been updated in the recently published third edition and continues to inform and educate landowners, planners and practitioners.

BSL in partnership with Rous County Council produced a highly-commended book on the social and ecological values of the Big Scrub that has sold over 1000 copies. BSL’s website has had a major upgrade: its Facebook page is updated weekly; its e-newsletter is published every two months. BSL’s greatly improved use of social media is helping to raise its profile and contribute to generating donations from the community, local businesses and philanthropic organisations to fund its growing community education and engagement work and other activities.

Science saving rainforests program. BSL, the Royal Botanic Gardens Sydney, the BSF and their partners have commenced an internationally innovative program to apply the latest DNA sequencing and genome science to establish plantations to produce seed of key species with optimal genetic diversity for the ecological restoration of critically endangered lowland subtropical rainforest. This program will for the first time address the threat posed by fragmentation and isolation resulting from the extreme clearing of Australia’s LSR, which is estimated to have resulted in the destruction of 94% of this richly biodiverse Gondwana-descended rainforest.

Many  key  LSR species are trapped in small populations in  isolated remnants  that  lack the genetic diversity needed to adapt and survive in the long term, particularly faced with climate change Necessary  genetic diversity is also lacking in many key species in the 500 ha of planted and regrowth rainforest. The first stage of the program, already underway, involves collecting leaf samples from approximately 200 individual old growth trees in 35 remnant populations across the ranges of 19 key structural species of the ‘original’ forest. DNA will be extracted from the leaf samples of each species and sequenced. The  latest genome science will be applied to select the 20 individual trees of each species that will be cloned to provide planting stock with optimal genetic diversity for the establishment of a living seed bank in the form of a plantation that will produce seed  for use in restoration plantings. As the individual trees in the restoration plantings reproduce, seed with appropriate genetic diversity and fitness will be distributed across the landscape. The project focuses on key structural species and thus helping the survival of Australia’s critically endangered Lowland Subtropical Rainforest in the long term.

Lessons learned and current and future directions. A key lesson learned some five years ago was that BSL had grown to the point where volunteers could no longer manage the organisation effectively. BSL took a major step forward in 2015 by engaging a part-time Manager, contributing to BSL’s continuing success by expanding the scope, scale and effectiveness of its community engagement activities and improving its day to day management.

The principal lesson learned from BSL’s on-ground restoration program is to focus on rehabilitation of remnants and not to take on large planting projects, but rather support numerous partnered community tree planting events. Large grant-funded multi-site tree planting projects are too difficult to manage and to ensure landholders carry out the necessary maintenance in the medium to long term.

Acknowledgements.  BSL acknowledges our institutional Partners and receipt of funding from the NSW government’s Saving our Species program, NSW Environmental Trust and Big Scrub Foundation.

Contact:  Shannon Greenfields, Manager, Big Scrub Landcare (PO Box 106,  Bangalow NSW 2479 Australia; . Tel: +61 422 204 294; Email: info@bigscrubrainforest.org.au Web: www.bigscrubrainforest.org.au)