CAUGHT on CAMERA A community monitoring project in Wombat State Forest ViCTORiAN NATiONAl PARkS ASSOCiATiON The Victorian National Parks Association (VNPA) helps to shape the agenda for creating and managing national parks, conservation reserves and other important natural areas across land and sea.
The VNPA works with all levels of government, the scientific community and the general community to achieve long term, best practice environmental outcomes.
The VNPA is also Victoria’s largest bush walking club and provides a range of information, education and activity programs to encourage Victorians to get active for nature.
NATUREWATCH The VNPA’s NatureWatch program is a community-based biodiversity monitoring program which informs, educates and engages the community in conservation management and practices. The NatureWatch program actively builds links between community members, scientists and land managers, and develops scientifically based, practical projects that contribute to a better understanding of species and ecosystems, and the management of natural areas.
PROjECT PARTNERS Wombat Forestcare Wombat Forestcare is a community group dedicated to protecting and enhancing the natural ecosystems of the Wombat Forest and surrounding areas.
The Arthur Rylah institute for Environmental Research (ARi)
ARI is the biodiversity research base for the Department of Sustainability and Environment (DSE) in Victoria and is a leading centre for applied ecological research.
ACkNOWlEdGMENTS
Ryan Chick (ARI) for designing and building bait stations. Jenny Nelson, Ryan Chick and Michael Scroggie (ARI) for developing training material. VNPA team leaders and volunteers for collecting the data and running trips. Wombat Forestcare. Fiona Hamilton, Kerryn McTaggert, Nicholas Bauer, Amber Connor, Department of Sustainability and Environment for project support and assistance in site selection. Phoebe Macak (ARI) provided helpful comments on this report.
Project supported by: Project funded by the Australian Governments Caring for our Country Community Action Grants.
2 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest CAUGHT on CAMERA A monitoring project in the Wombat State Forest
Community monitoring of mammals in response to fire at Wombat State Forest, Victoria. Prepared by Caitlin Griffith and John Kotsiaris, Victorian National Parks Association and Matt Bruce, Meredith Kirkham and Richard Loyn (Arthur Rylah Institute for Environmental Research).
CONTENTS 3 CONTENTS
Summary ...... P5
1.0 Introduction ...... P6 Caught on Camera project ...... 6 Project background – fire ecology and monitoring ...... 6 Motion sensing cameras ...... 7
2.0 Methods ...... P8 Methods ...... 8 Discussion ...... 11
References ...... P14
4 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest CONTENTS 4 SUMMARY
he VNPA Caught on Camera Tproject trial in Wombat State Forest involved working with local community groups and scientists to establish community based monitoring of fauna using motion sensing cameras. The project addressed the question; ‘What mammal species are located at ‘Recently Burnt’, ‘Intermittently Burnt’ and ‘Long Unburnt’ sites in Foothills Forest and Forby Forest in Wombat State Forest? Mountain Brushtail Possums. This has been established to contribute to the larger question, ‘what is the impact of • Common Ringtail Possum volunteers and the images fire on mammals?’ displayed on the VNPA’s • Eastern Grey Kangaroo Facebook site have been a rich The trial ran in 2012, with a • Black Wallaby source of interest and delight to view to carrying out annual thousands of Victorians. monitoring over the long • Bush Rat term (>10 years). One year of 4 exotic mammals: Moreover, the images were monitoring with this method appropriated by artists, and in • Black Rat could not possibly answer this all the cultural aspects of the question. However, this report • Red Fox project have contributed to the outlines the results of the initial • Domestic Dog wider community’s awareness 1 year project trial. of, and enthusiasm for, the • Feral Goat animals of Wombat. Monitoring was carried out in two different forest types in After only one season of Wombat State Forest, Forby monitoring, no apparent Forest and Foothills Forest at trends have emerged, however sites categorised as ‘Recently monitoring has provided an Burnt’, ‘Intermittently Burnt’ and improved understanding of the ‘Long Unburnt’. A total of 10 mammals inhabiting Wombat native mammals and 4 exotic State Forest. mammals were recorded. As the project continues to run 10 native mammals: long term, we hope to continue to address the project question. • Short-beaked Echidna • Agile Antechinus • Koala Greater social outcomes • Common Wombat Caught on Camera has proven a very successful as a community • Mountain Brushtail Possum building enterprise. • Common Brushtail Possum The VNPA was never short of
SUMMARY 5 1.0 INTRODUCTION
Caught on Camera to contribute to the larger impacts on fauna (Clarke 2008 question, ‘what is the impact of and MacHunter et al 2009). project fire on mammals?’ Without effective, repeatable, he VNPA Caught on Camera ongoing monitoring there will Tproject involves working continue to be limitations to with local community groups to Project stages how we understand the impacts establish long term monitoring of fire and how planned burning Over March – June 2012 the is used in the landscape. by the community, using motion VNPA’s NatureWatch program sensing cameras. In Wombat ran a one-season trial of Caught “Much monitoring of fauna is State Forest and Bunyip State on Camera, a citizen science of such a small scale and short Park, it involves looking at the project in Wombat State Forest duration that the statistical impact of fire on fauna, however in partnership with Wombat likelihood of detecting a when developing the project in Forestcare, a local community positive or negative effect of the management regime is minute. other locations, the project may group. This project has been Such shortcomings will only be look at variables other than fire established to run long term such as revegetation efforts or overcome through broad-scale and following this project trial, vegetation type. and/or long-term studies of fauna.” the project will be set up to run (Clarke 2008) The Caught on Camera project annual fauna monitoring using objectives in Wombat State motion sensing cameras. In order for land managers Forest and Bunyip State Park to improve planned burning are: techniques, particularly in This report response to our native fauna, • To provide crucial data on it is essential to understand the long term impacts of This report presents the the impact of fire on all life control burning on Australian results of the trial and forms, not just the impacts on fauna to land managers, recommendations for the the plants. This is particularly community groups, scientists development of ongoing relevant when working in a and Government. monitoring. fragmented landscape, such as Victoria. • To create working partnerships with “The ability of fire planners to Government, Researchers 1.1 Project meaningfully implement the and Community Groups to background – dual aspirations of protecting establish and run monitoring. life and property and achieving fire ecology and ecological goals is dependent on • To demonstrate and promote monitoring the availability of science and to Government the need evidence that informs operational for ongoing, strategic and “Altered fire regimes threaten processes, and monitoring that comprehensive monitoring in biodiversity and interact with influences future management.” response to fire. other threats in complex ways (MacHunter et al 2009) that are not yet fully understood.” This project specifically asks the (Department of Environment, Through establishing long question: Water, Heritage and the Arts term projects with good 2009) scientific basis, that are linked ‘What mammal species are to management of planned located at ‘Recently Burnt’, In Australia, we have a limited burning, it is possible for the ‘Intermittently Burnt’ and ‘Long understanding of the impacts community to be involved in Unburnt’ sites in Foothills Forest of fire on our biodiversity carrying out this monitoring and Forby Forest in Wombat (Clarke 2008 and MacHunter and to contribute to building State Forest? et al 2009). This is particularly our knowledge base on the the case when it comes to the impact of fire on fauna. This has been established
6 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest iNTROdUCTiON Common Wombat.
1.2 Motion sensing Given this, it is important to see camera monitoring as cameras providing presence and absence Monitoring of fauna can be data for particular species, highly labour intensive. Motion rather than detailed data on sensing cameras provide the the size of populations. It is opportunity to gather data on also important to note that selected animal groups (e.g. this camera monitoring would small mammals, some arboreal provide useful information on mammals etc.) with much less selected fauna in response to labour than methods such as fire, but would not provide trapping. It is possible to select comprehensive data on sites for cameras and place population sizes etc. Camera cameras in these locations at monitoring is also a great regular intervals. The method community education tool, is also much less stressful where images of the different for the fauna than trapping species being ‘caught on and can provide data on the camera’ in the local area can be presence of certain species that regularly displayed. are unlikely to be caught in traps. However, motion sensing cameras do not have the ability to provide data on sizes of populations of particular fauna. iNTROdUCTiON 7 2.0 METHODS
Methods Study area he study was conducted Tin Wombat State Forest in central-west Victoria. The forest covers about 70,000 ha and has had a history of timber harvesting. Large scale harvesting ceased in 2002 (Macak et al. 2010). The vegetation is characterised as a mixed eucalypt foothill forest with the dominant overstory species being Messmate (Eucalyptus obliqua) in combination with Peppermints (E. dives and E. rubita) and Candlebark Gum (E. rubida) (Leversha 1996). The Wombat Land Use Study Sites (Time Since Fire) State forest is inhabited by State+Forest 0-10+years a number of native mammal National+Park+or+Nature+Conservation+Reserve 11-41+years species including key fire Other+Park+or+Conservation+Reserve 42++years response species such as Other+Public+Land Mountain Brushtail Possum, Plantation Agile Antechinus and Black Figure 1. Map of the study sites in the Wombat State Forest and their time since fire Wallaby (Macak et al. 2010; category. MacHunter et al. 2009).
• Inclusion of some sites with years since the last fire). Sites Sites planned control burns. were located in two different vegetation communities; The sites selected were a Figure 1 shows the location of Foothills forest and Forby forest. subset of sites established by the sites. Due to the constraints on site Wombat Forestcare and the selection parameters it was not Department of Sustainability The sites were selected to possible to achieve a balanced and Environment as Community represent three different time design with a similar number Research sites (Macak et since fire categories; recently of sites in each time since fire al. 2010). These sites were burnt (RB: 0 - 10 years since the and vegetation community established in a range of last fire), intermediately burnt categories. For example there vegetation types and have a (IB: 11 - 41 years since the last were fewer sites in Forby forest range of fire histories. fire and long unburnt (LU: 42+ and the IB category on Forby Due to time constraints 20 sites forest contained only one site. were selected for this study. Parameters for selection of sites Foothills Forest Forby Forest All sites included: RB IB LU RB IB LU • Ease of site access for Number of sites 3 4 6 3 1 3 20 community volunteers. Table 1. The number of sites in each fire category found in recently burnt (RB: 0-10 years since fire), intermediately burnt (IB: 11-41 years since fire) and long unburnt (LU: 42+ years • A range of fire histories. since fire) forest for the two vegetation communities in this study.
8 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest METHOdS Eastern Grey Kangaroo.
Table 1 shows the number of 3. Participants feel part of a VNPA and Wombat Forestcare sites in each category. team. volunteered every 3 weeks to pack down, move and set-up Survey methods – methods 4. Participants experience cameras and bait stations at the of involving and training Wombat State Forest. selected locations. community 5. Participants are more As this is a community project, Two (motion sensing cameras, informed about the VNPA heat-in-motion type), Reconyx volunteers undertook the on and aware we are a donation ground monitoring activities. Hyperfire were installed at supported organisation each site. At each site a 100m This required coordination and and more informed about training, led by the VNPA. transect was established and Wombat Forestcare. the cameras were installed at A community training activity The training activity was the 20 and 80m points along took place on 10th of March attended by over 50 eager the transect. They were attached 2012. participants who were provided to the nearest suitable tree to with project background, the designated point. A bait The objectives of the training cage containing six stainless activity were: organisational details of VNPA and Wombat Forestcare, visited steel tea strainers was attached 1. Participants learn how to a site in Wombat State Forest to a plastic garden stake and use motion sensing cameras and trained by ARI scientists in placed 2m from the camera with this projects methods. how to set-up bait stations and with the base of the cage 20cm cameras at a site. from the ground. Each tea 2. Participants sign up to carry strainer contained a mixture out monitoring. Following the training activity, of rolled oats, peanut butter community volunteers from and golden syrup. The camera
METHOdS 9 Antechinus and Bush Rat. was positioned 50cm from the some doubt existed they Wren (80% of sites) and Eastern ground and aligned so that the were assigned to a more Yellow Robin (60% of sites) were bait station appeared in the generic category, for example the most commonly detected horizontal centre of the frame “Unidentified Brushtail Possum”. birds. and the bottom of the bait cage The data from both cameras was in the vertical centre of the at one site was combined and Table 2 shows the percentage frame. The vegetation between the percentage of sites in which of sites at which each species the camera and bait station, and each species was found was was encountered in the three one metre behind was cleared calculated for each fire category time since fire categories for within the field of view of the in each vegetation type. each vegetation community. camera to ensure that this did Some mammal species not obscure any photographs of showed interesting patterns in the number of detections animals. The cameras were left Results to operate for a minimum of 21 according to the time since days. Upon collection cameras A total of 13,927 images fire at a site. The Short-beaked were checked to see if they contained at least one animal. Echidna was only encountered were still operating and this was Fourteen mammal species were in long unburnt sites, whereas noted. identified from the photos (10 the Common Ringtail Possum native and 4 introduced) along showed this trend in Foothills Initial photo identification with 13 bird species (12 native forest but the opposite trend in was done by a trained VNPA and 1 introduced). The most Forby forest. Mountain Brushtail volunteer. The photos were commonly detected mammals Possums were encountered then sent to ARI for further were Black Wallaby (100% of more often in recently and identification. Where possible sites), Common Wombat (85% intermediately burnt sites in animals were identified to of sites) and Agile Antechinus Foothills forest but not in Forby species level. In cases where (75% of sites). Superb-fairy forest. The two rodent species
10 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest METHOdS Table 2. The percentage of sites in which PERCENTAGE OF SiTES iN WHiCH EACH SPECiES WAS FOUNd each species was found in recently burnt Foothills Forest Forby Forest All sites (RB), intermediately burnt (I) and long unburnt (LU) forest for the two vegetation Species RB IB LU RB IB* LU communities in this study. Mammals *There was only one site in this category. Short-beaked Echidna 0 0 17 0 0 33 10 Agile Antechinus 67 100 83 67 0 67 75 Koala 33 25 0 0 0 33 15 Common Wombat 100 100 83 67 0 100 85 patterns for some bird Mountain Brushtail 67 75 17 67 0 100 55 Possum species. For example, both Common Brushtail 0 25 0 0 0 33 10 the Australian Magpie and Possum the Grey Currawong were not Unidentified Brushtail 0 50 17 0 0 0 15 Possum encountered at long unburnt Common Ringtail 0 0 33 33 0 0 15 sites in Foothills forest which Possum differed from Forby forest where Eastern Grey Kangaroo 33 50 17 33 0 33 30 they were not encountered at Black Wallaby 100 100 100 100 100 100 100 recently burnt sites. The results Bush Rat 33 50 67 0 0 33 40 for White-winged Chough Black Rat 33 0 0 33 0 0 10 and Bassian Thrush were more Red Fox 67 50 50 0 100 67 50 consistent with both species not Domestic Dog 0 0 0 0 0 33 5 found in recently burnt sites in Feral Goat 0 0 0 33 0 0 5 both forest types. The opposite pattern was observed for the Birds introduced Common Blackbird, Laughing Kookaburra 0 50 0 33 0 0 15 which was only encountered at Superb Fairy-wren 33 100 83 100 100 67 80 long unburnt sites. White-browed 33 50 50 33 0 33 40 Scrubwren Spotted Quail-thrush 0 0 0 0 100 0 5 Grey Shrike-thrush 33 50 67 33 100 67 55 Discussion Australian Magpie 33 25 0 0 0 33 15 Grey Currawong 33 25 0 0 100 33 20 Mammals White-winged Chough 0 25 33 0 0 33 20 The results from this study Scarlet Robin 0 25 17 0 0 0 10 showed some patterns of Flame Robin 0 0 0 0 0 33 5 species presence with fire Eastern Yellow Robin 0 100 83 67 0 33 60 history and it will be interesting Bassian Thrush 0 25 50 0 0 33 25 to see if they persist after Common Blackbird 0 0 17 0 0 33 10 more sites are surveyed. These Unidentified bird 33 75 67 0 100 33 50 patterns were inconsistent across species and forest Unknown types. The two rodent species Unidentified animal 100 100 100 67 100 100 95 detected showed opposite patterns suggesting different responses to fire. Encounters showed opposite patterns Black Wallaby and Common of the native Bush Rat were regarding time since fire. Bush Wombat were encountered at biased towards intermediately Rats were only encountered at most (all in the case of Black and long unburnt forest. This long unburnt sites; whereas the Wallaby) regardless of time observation is consistent with introduced Black Rat was only since fire. the expert predictions made for encountered at recently burnt the two forest types (Foothills sites. Some species such as the There were also interesting
10 METHOdS 11 Feral Goat. and Forby Forest) investigated in Foothills Forest suggested intermediately burnt Foothills in this study (MacHunter et al. that, in contrast to the present forest, but this pattern was not 2009). It is also consistent with study, they are equally likely apparent in Forby forest. Expert previous studies that show to be encountered in recently judgement suggests that this Bush Rat numbers decline post burnt and long unburnt sites. species should slowly recover fire (Sutherland and Dickman after fire, with abundance 1999, Lindenmayer et al. 2008), The arboreal mammals detected maximised at long unburnt sites possibly due to a reduction in in this study showed differing (MacHunter et al. 2009). arthropod food and habitat patterns according to fire complexity (Sutherland and history. The Common Ringtail Some species such as the Black Dickman 1999). By contrast, Possum was only encountered Wallaby and the Common the Black Rat is advantaged in long unburnt sites in Foothills Wombat were found at most by disturbance and thus is forest but in recently burnt sites sites regardless of fire history. more likely to be found at in Forby forest. This species is This is similar to findings for modified sites, such as those predicted to be more common these species in the Victorian et more recently affected by fire at intermediate times since central highlands (Macak (Sutherland and Dickman 1999). fire (MacHunter et al. 2009) al. 2012). There is, however however this prediction has not evidence suggesting that Black In this study the Short-beaked been confirmed by empirical Wallaby will be negatively Echidna was more frequently data (Lindenmayer et al. 2008). impacted with increasing time encountered in long unburnt For the Mountain Brushtail since fire (Catlinget al. 2001). sites. There is sparse literature Possum the pattern also varied on the response of this species between forest types. This A number of introduced to fire. In a previous study species was more frequently mammal species were detected (Macak et al. 2012) conducted encountered in recently and in this study, of which the
12 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest METHOdS Black Rat (see above) and Red is becoming increasing new sites in the same forest Fox were the most commonly widespread in wildlife research types and with the same fire detected. The presence of Red (Meek et al. 2012). They are well histories, rather than expanding Fox at most sites is a concern suited to studies conducted by the study to include new forest as they are implicated in the volunteers due to the ease of types and fire histories, which decline of a number of native installation and the feedback would limit the potential for mammals especially in the 35 provided to volunteers via the robust data analyses. – 5500 g weight range (DEH photos produced (Macak et al. 2011). This includes a number 2012). Therefore, the value of It is hoped that this community of native species detected in such citizen science initiatives research will be undertaken this study such as the Common should not be judged on their annually to continue to Ringtail Possum, Mountain scientific merit alone, but also develop our understanding of Brushtail Possum and Bush on their ability to inform and anypatterns in relation to the Rat. It also includes species engage volunteers and the impact of fire on fauna. known from the Wombat State wider community. Forest but not detected in this study such as the Brush-tailed There are sophisticated statistical Phascogale which is listed as techniques that enable the robust vulnerable in Victoria. analyses of data from automated cameras (MacKenzie et al. 2002), but such analyses were beyond the scope of this project and thus Birds it is not possible to draw a causal Although this survey was link between the observations designed to detect ground- in this study and the fire history dwelling mammals a large variables. number of birds were also detected. This is consistent with previous studies using similar Recommendations lures and site preparation (e.g. Macak et al. 2012). Whilst This survey highlighted a there are some interesting number of interesting patterns trends for some species such with regard to fire history and as Australian Magpie, Grey native mammals in Wombat Currawong, Bassian Thrush and State Forest. However, further White-winged Chough they surveys are required to explore should be interpreted with these patterns and statistical caution as this study was not analyses are required to designed to survey birds. A establish a link between fire more robust approach would history and native mammal be to use indices of abundance occupancy in this system. Future such as that derived from the work should focus on adding to two hectare 20 minute count the current data set by sampling method (Loyn 1986).
limitations of survey methods The use of automated cameras, such as the motion sensing cameras used in this study,
METHOdS 13 REFERENCES
Australian Native Plants Society (Australia) Department of Sustainability and MacHunter, J., Menkhorst, P., and Loyn, website – http://anpsa.org.au/ Environment Website, Fire and Other R. (2009). Towards a Process for Emergencies, Fire Ecology – http:// Integrating Vertebrate Fauna into Biodiversity and Fire – Opportunities for www.dse.vic.gov.au/fire-and-other- fire Management Planning. Arthur VNPA volunteer monitoring programs emergencies/fires-and-biodiversity Rylah Institute for Environmental Research Technical Report Series No. Catering for the needs of fauna in fire Lindenmayer, D.B., MacGregor, C., Welsh, management: science or just wishful 192. Department of Sustainability and A., Donnelly, C., Crane, M., Michael, Environmnent, Heidelberg, Victoria. thinking? – Michael Clarke. D., Montague-Drake, R., Cunningham, Catling, P.C., Coops, N.C., and Burt, R.J. R.B., Brown, D., Fortescue, M., Dexter, MacKenzie, D.I., Nichols, J.D., Lachman, G.B., (2001). The distribution and abundance N., Hudson, M., and Gill, A.M. (2008). Droege, S., Royle, J.A., and Langtimm, of ground-dwelling mammals in Contrasting mammal responses to C.A. (2002). Estimating site occupancy relation to time since wildfire and vegetation type and fire. Wildlife rates when detection probabilities are vegetation structure in south-eastern Research 35: 395-408. less than one. Ecology 83: 2248-2255. Australia. Wildlife Research 28: 555- Loyn, R.H. (1986). The 20 minute search - A Meek PD, Ballard G and Fleming P (2012). 564. simple method for counting forest An Introduction to Camera Trapping for Cheal, D. (2010) Growth stages and birds. Corella 10: 58-60. Wildlife Surveys in Australia. PestSmart Toolkit publication, Invasive Animals tolerable fire intervals for Victoria’s Macak, P.V., Bruce, M.J., and Loyn, R.H. native vegetation data sets. Fire and Cooperative Research Centre, Canberra, (2012). Community finding fauna Australia. Adaptive Management Report No. - naturalist groups contributing to 84. Department of Sustainability and research on the response of fauna Sutherland, E.F., and Dickman, C.R. (1999). Environment, East Melbourne, Victoria, to fire. Black Saturday Victoria 2009 Mechanisms of recovery after fire by Australia. - Natural Values Recovery Program. rodents in the Australian environment: DEH. (2011). Predation by the European Department of Sustainability and a review. Wildlife Research 26: 405-419. Environment, Heidelberg, Victoria. red fox - key threatening process Wombat Forestcare Inc website – listing.
14 CAUGHT ON CAMERA – Monitoring Project in Wombat State Forest REFERENCES REFERENCES 15 Address: Level 3, 60 Leicester Street, Carlton, Victoria 3053. Phone: 03 9347 5188. Email: [email protected] Website: www.vnpa.org.au