Assessing Local Patterns in Carnivore Occupancy and Richness by Camera Trapping: a Monitoring of Diversity at Verloren Vallei Na

Home , African clawless otter

Assessing local patterns in carnivore occupancy and richness by camera trapping: a monitoring of diversity at Verloren Vallei Nature Reserve, Dullstroom, Mpumalanga 2019

Daan Loock & Dr Lourens SwanepoeL

Daan Loock Pr. Sci.Nat (Ecology) Faculty of Natural and Agricultural Sciences, University of the Free State, 205 Nelson Mandela Drive, Park West, Bloemfontein, 930, South Africa

Associated member Wildlife and Resource Association (WRA)

*Email address: [email protected]. (DJE Loock)

Dr Lourens Swanepoel

Department of Zoology, School of Mathematical & Natural Sciences, University of Venda, P/ Bag X5050,

Thohoyandou 0950, South Africa

Affiliated with African Institute for Conservation Ecology and Genetics (AICEG)

(www.aiceg.org) & Wildlife Resource Association (WRA)

*Email address: [email protected] (L H Swanepoel)

Executive Summary

This is the second monitoring report after a thorough baseline assessment conducted during 2015 and the first monitoring assessment performed during 2018. This monitoring focused on occupancy of the carnivore diversity and not density like in the 2015 report. (Swanepoel, L.H., Matthews, W. 2015) From the 5th August to the 14th of September 2019, (40 days), the Verloren Vallei nature reserve outside Dullstroom were surveyed with camera traps. An effective camera trapping effort of 2000 trap nights were achieved. We detected 14 carnivore species, two species less than during the previous assessment conducted during 2018.

Another 13 non-carnivore species were detected excluding feral cats, domestic cattle, birds and humans.

There is a slight decline in the detection of both carnivore and non-carnivore species for the 2019 assessment.

One important detection is the small spotted cat ‘felid nigripes’ a near threatened species, which might be the highest elevated geographical recording for the species. New recordings for 2019 include the Cape

Clawless Otter ‘Aonyx capensis’ a near threatened species, and Meerkats ‘Suricata suricatta’. The assessment increased in geographical surface with the inclusion of the north-eastern area known as zone 6. This

implicated that we employed another 10 camera sites, increasing the 40 current sites tot 50 sites in total for

Verloren Vallei.

Contents

EXECUTIVE SUMMARY………………………………………………………………………………………………2

1. INTRODUCTION……………………………………………………………………………………………3

2. OBJECTIVES………………………………………………………………………………………………...7

3. METHODS…………………………………………………………………………………………………..7

I. Study Area………………………………………………………………………………………..7

II. Carnivore Species List…………………………………………………………………………….7

III. Camera Trapping…………………………………………………………………………………8

4. RESULTS…………………………………………………………………………………………………….9

5. DISCUSSION……………………………………………………………………………………………….20

6. REFERENCE…………………………………………………………………………………………………21

Suggested citation:

Loock, D.J.E. & Swanepoel, L.H. 2019. Assessing local patterns in carnivore occupancy and richness by camera trapping: a monitoring of diversity at Verloren Vallei Nature Reserve, Dullstroom, Mpumalanga

Introduction

Carnivores are a generic term for species that mostly predate on other species rather than plants or vegetation. Normally mammalian carnivores would be a vital part of ecosystems in maintaining and balancing the structure of the larger ecosystem. It is therefore very important for conservation managers to understand the dynamics of the carnivore diversity and to allow for reasonable natural fluctuations in numbers. (Ripple et al. 2014). Biodiversity loss and the decline in species can be ranked as the most important indicators in ecosystem change. One of the major limitations of this assessment is the methods used to detect species. A non-detection cannot really be understood as a species loss, but rather a non-detection or not present at the time of sampling. Non-detections over repetitive sampling may suggest that the species are not on site anymore. Local abundance or detection is the relative representation of species in this specific ecosystem and is measured by individuals found per sample. Therefor relative abundance for Verloren

Vallei would be calculated by dividing the species from one group by the total number of species from all groups. (Verberk, C.E.P et al. 2011) The central aim for Verloren Vallei will be to understand how communities are organized, by describing the general patterns of the structure of the communities and the changes over time.

Figure 1: This figure is a sattelite image of the study site with the locations of the detectors and detector numbers.

Figure 2: This figure indicates the location of camera trap and the number of images recorded for carnivores during the trapping survey.

Large carnivores are important in ecosystems where their impacts are mostly directly predation driven or indirectly fear driven. These impacts roles out in prey communities and later on in ecosystem structure and function (Roemer et al. 2009). However, most of the carnivores are not large especially at Verloren Vallei, they are medium to small carnivores known as meso-carnivores. These meso-carnivores are found in larger numbers and species diversity is greater than large carnivores making this group very important to study, especially in smaller reserves (Roemer et al. 2009; Prugh et al. 2009).

This study conducted on Verloren Vallei is the second monitoring study following a baseline study conducted during 2015, and one of the recommendations from the baseline assessment (Swanepoel, LH. 2016).

Verloren Vallei is a nature reserve located approximately 20 km outside Dullstroom, part of the

Steenkampsberg plateau. The landscape includes some of the most pristine grasslands and sensitive wetlands in Mpumalanga.

In order to report on the species diversity as a recommendation from the baseline assessment we made use of a camera trapping technique to collect our data set. Camera trap sites were located exactly on the same previous trap sites to repeat the previous trapping protocol (Figure 1.)

Objectives

I. To estimate carnivore species diversity as detected by camera trap surveys. II. To report on non-carnivore species diversity detected with camera trap surveys.

Methods

I. Study area

The monitoring assessment was conducted from the 5th of August to the 14th of September 2019 at Verloren

Vallei nature reserve approximately 20 kms north of Dullstroom in Mpumalanga (Figure 1). The reserve consists of 6061 hectares, with rocky outcrops, grass and wetlands, 18% or 1120 hectares is classified as wetland (Swanepoel, LH. 2016). The remainder of the surface area can be classified as Lydenburg Montane

Grassland (Grassland biome) (Mucina and Rutherford 2006).

The monitoring assessment had 50 camera trap stations deployed over roughly a surface area of 95% of the reserve (Figure 1). The reserve portion east of the larger reserve divided by the road was included during the 2019 survey. Camera traps were deployed during the entire span of the assessment. Inspections, changing of SD cards and batteries were done every 12 days.

II. Data collection

Camera traps were deployed evenly throughout the site and were situated throughout the site in order to represent all the described vegetation units on Verloren Vallei. Due to the sensitivity of the vegetation and the wetlands most of the cameras were deployed next to existing road infrastructure in order not to create new paths in the grasslands. After the collection of the pictures, the pictures were all named and described with a software package called ‘DIGIKAM’. All other statistical analysis was conducted with ‘R-Projects’ and

QGIS.

III. Carnivore species list

The monitoring effort is based on Mammal list prepared by Dr L Swanepoel for the basic assessment in

2015. Verloren Vallei nature reserve could potentially harbor 23 carnivore species (Table 2) 16 species are classified as ‘least concern’, five species as ‘near threatened’ and one specie as ‘data deficient’

(Swanepoel, LH. 2016).

Table 1: Potential species list and conservation status of carnivore species at Verloren Vallei nature reserve based on 2004 Red data book of mammals of South Africa (Red Data Book of the Mammals of South)

Species Common Name Sa Red Data Status 2015 2018 2019 Frequency

Canis mesomelas Black-backed Jackal Least concern ✅ ✅ ✅ high

Vulpes chama Cape Fox Least concern ✅ ✅ ✅ low

Caracal caracal Caracal Least concern ✅ ✅ ✅ low

Felis silvestris African Wild Cat Least concern ✅ ✅ ✅ medium

Atilax paludinosus Marsh Mongoose Least concern ✅ ✅ ✅ low

Cynictis penicillata Yellow Mongoose Least concern ✅ ✅ ✅ low

Civettictis civetta African civet Least concern ❌ ✅ ❌ low

Genetta genetta Small spotted genet Least concern ❌ ❌ ❌

Genetta maculata Rusty spotted genet Least concern ❌ ❌ ❌

Genetta tigrina Large spotted genet Least concern ❌ ✅ ❌ low

Ictonyx striatus Striped polecat Least concern ❌ ✅ ❌ low

Galerella sanguinea Slender Mongoose Least concern ❌ ✅ ✅ low

Ichneumia albicauda White-tailed Mongoose Least concern ❌ ❌ ❌

Mungos mungo Banded Mongoose Least concern ❌ ❌ ❌

Proteles cristatus Aardwolf Least concern ❌ ✅ ✅ low

Aonyx capensis Cape Clawless Otter Least concern ❌ ❌ ✅ Low

Panthera pardus Leopard Least concern ❌ ❌ ❌

Canis adustus Side-striped Jackal Near threatened ✅ ✅ ✅ medium

Leptailurus serval Serval Near threatened ✅ ✅ ✅ high

Hyaena brunnea Brown Hyaena Near threatened ✅ ✅ ❌ low

Mellivora capensis Honey Badger Near threatened ✅ ✅ ✅ medium

Felis nigripes Small spotted cat Near threatened ❌ ✅ ✅ low

Lutra maculicollis Spotted-necked Otter Near threatened ❌ ❌ ❌

Poecilogale albinucha African Weasel Data deficient ❌ ❌ ❌

Suricata suricatta Meerkat Least Concern ❌ ❌ ✅ Low

IV. Camera trapping

The monitoring assessment replicated the baseline study of 2015. The trapping survey followed guidelines for closed-population studies. Cameras were deployed within preset grids of 1.2km x1.2km. The average spacing between camera sites were 758 m (Swanepoel, LH. 2016). Fifty camera traps were deployed for

40 days resulting in 2000 trap nights. Cuddeback Ambush whiteflash cameras were used in the survey. Units were programmed with a one-minute delay between pictures. Cameras were positioned next to trail roads approximately 30 cm above ground level. All cameras were serviced once every two weeks. One camera was lost during the survey and was replaced immediately.

DigiKam software were used to catalogue images once downloaded from the SD cards. We used vegan

R-package to produce species accumulation curves and ordination plots.

Results Camera Trapping

During the monitoring assessment the trapping effort constitutes of 2000 trap nights from the 5th August to the 14th September 2019. The carnivore trapping success rate is 21.1%. The total carnivore species detected were 14 species and for the non- carnivore species 13 species, excluding domestic cattle, feral dogs and cats, humans and bird species (Table 2).

Table 2: Table indicates the trapping effort and the total amount of carnivore images recorded during the survey

Survey Description 2018 2019

Survey start data 2018/08/23 2019/08/05

Survey end date 2018/10/05 2019/09/14

Trap nights 1720 2000

Trap sites 40 50

Total Carnivores species 16 14

Total Carnivores 474 421

Total non-carnivore species 17 14

Total non-carnivore images 2006 949

Table 3: Indicates the number of species at various detectors

Carnivore Species detected Number of species detected at all traps Species detected at different traps

2018 2019 2018 2019

black_backed_jackal 320 261 35 41

serval 106 78 26 29

african_wild_cat 11 16 7 6

cape_fox 9 19 9 14

honey_badger 7 4 5 3

side_striped_jackal 6 16 4 14

yellow_mongoose 2 8 2 2

african_civit 2 0 2 0

aardwolf 1 7 1 5

large_spotted_genet 1 0 1 0

striped_poleca 1 0 1 0

small_spotted_cat 1 1 1 1

slender_mongoose 1 1 1 1

marsh_mongoose 1 6 1 3

caracal 1 8 1 6

brown_hyaena 1 0 1 0

Cape_clawless_otter 0 1 0 1

Meerkat 0 2 0 2

a) b)

Figure 3: This figures indicates the species accumulation curve for a) carnivores and b) non-carnivore species detected during the survey. The x-axes indicate the number of trap sites, and the y- axes indicates the number of carnivore species detected.

a) Ordination plot for carnivores b) Ordination plot for non-carnivores

Figure 4: Figure three represents a normal ordination plots where a) Carnivores, and b) Non-carnivores are plotted. Trap sites are indicated in numbers and the species are indicated in text.

a) Species accumulation curve for carnivores Species accumulation curve for non-carnivores

Figure 5: This figures indicates the species accumulation curve for a) carnivores and b) non-carnivore species detected during the survey. The x-axes indicate the number of trap sites, and the y- axes indicates the number of carnivore species detected.

Relative abundance for carnivores 60

50 4849 50 4647 4445 4243 4041 3839 40 3637 3435 33 3233 3031 2829 30 2627 2425 2223 2021 1819 20 16 1617 15 1415 13 12 1213 13 1011 10 11 9 8 9 99 9 8 8 10 7 6 7 6 6 7 6 7 7 6 7 4 4 5 4 4 4 4 55 4 4 4 4 4 2 2 3 32 3 32 3 2 2 23 32332 3 3 23 22 32 22 2 323 3 2 32 1000000000000000000000000010000000000000000010000010100011000000000000000010101000010000000100000000000100100000010000001000100000000000000100100000100000000000010000000000001001000000000101001000110010000100000100100000010000000100011010010100010000010000001000010000100000000000000000000000100000000000000010000010000001000000100000000000000001000000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

camera aardwolf african_clawless_otter african_wild_cat black_backed_jackal cape_fox caracal honey_badger meerkat serval side_striped_jackal slender_mongoose water_mongoose yelow_mongoose

Figure 6: Relative abundance graph for carnivores

Relative abundance for non-carnivores 60

50 4849 50 4647 4445 4243 4041 3839 40 37 36 36 3637 34 3435 3233 3031 282829 28 30 2627 2425 2223 2021 21 21 19 1819 20 1617 1415 15 15 1213 13 1010 10 1011 10 10 10 10 10 1111 11 10 9 8 8899 8 9 89 8 9 8 10 6 6 7 6 6 7 7 7 6 67 77 6 7 45 4545 5 4 4 4 4 5 44 444 44 44 5 444 55 4 44 4 444 5 22 32 222 3 232 223 22 2 222 3 2232 222 33 22 2 22 3 232 2332 3 3 22 3 2 2 2 23 22 3 100101000001000001100010010010110000001001000000010001000000010000000100001001001010010000010100000010010000010010100000000010001000010000010100000101000000100000101101010100101001000000000000001010000010000000000001000000000010000000000000001001000100010000000000010000000100000100000001000000000000100000100000100000000000010000000000001000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

camera baboon black_wildebeest blesbok bushpig capehare capeporcupine greyduiker greyrhebok oribi springhare springbok steenbok zebra

Figure 7: Relative abundance graph for non-carnivores

Trapping site numbers Figure 8: This figure is a representation of the study site, indicating the vegetation groups and camera trap locations

Aardwolf African Clawless Otter Figure 9: This figures represents the number of images of Aardwolf and African Clawless Otters per trap.

African Wild Cat Black backed jackal Figure 10: This figure represents African Wild Cats and Black Backed Jackal per trap.

Serval Cape fox Figure 11: This figure represents Serval and Cape Foxes per trap.

Caracal Honey badger Figure 12: This figure represents Caracal and Honey Badgers per trap.

Meerkat Marsh Mongoose Figure 13: This figure represents Meerkat and Marsh Mongoose per trap.

Slender mongoose Side striped jackal Figure 14: This figure represents Slender Mongoose and Side Striped Jackal per trap.

Small spotted cat Yellow mongoose

Discussion This is the second monitoring effort after the camera trapping baseline assessment in 2015. The objectives of this study are not to determine any occupancy rates of any species, but rather to estimate species richness compared with previous surveys due to time and budget constraints.

It seems like most of the important species showed a decline in numbers detected. In terms of the carnivores, we reported a decline in the number of species for the 2019 survey of 2 species and for the non-carnivore species, also a decline of 3 species from 17 to 14 species. There are two new species reported for Verloren Vallei during the 2019 assessment, 1) the Cape Clawless Otter and 2) the Meerkat.

This survey proved a slight smaller diversity of smaller carnivores and emphasizes a failure of detection and not a true absence during the assessment.

This monitoring effort can easily be replicated every year. In order to achieve a more reliable occupancy result it is recommended to increase the trapping detectors and effort.

The species accumulation curves analyzed are used to estimate the number of species detected during a survey, and whether this curve indicates a sufficient or adequate yield representing the study area. The carnivore accumulation curve produced indicated an insufficient yield of species.

The species richness may improve or increase with a higher amount of camera points. It is estimated that the curve will level out with at least another 5 additional placements. The carnivore accumulation curve does not compare different sampling efforts like a rarefaction curve. For the non-carnivore’s accumulation curve, the top bend of the curve levels out meaning that even with an increased number of cameras, the detection of species will not increase. Therefor the trapping effort for non-carnivore’s is sufficient and implies an accurate number of species detected for

Verloren Vallei. (Figure 5)

The ordination plots for the carnivores demonstrates that most of the carnivore species are similar in terms of one variable ‘number of detections’. Most of the species plots low and clustered together due to the low number of detection, while only serval and black backed jackal plot further away from the cluster due to the higher number of detections. In case of the non-carnivores their ordination plots evenly distributed in the graph.

The relative abundance graphs indicate the relation between species detected relative to the number of trap sites detected. These graphs are prepared for both the carnivores and non- carnivores.

One additional feature to the 2019 report is the inclusion of the vegetation map, that describes the main vegetation groups, which was made available by the scientists from the Mpumalanga Tourism

and Parks Agency. With this new evidence we are still not able to get an understanding of preferred habitat for any of the carnivores. Most of the carnivores does not have a preference in terms of home ranges and habitat use.

References 1. Ripple, William J., James A. Estes, Robert L. Beschta, Christopher C. Wilmers, Euan G. Ritchie, Mark Hebblewhite, Joel Berger, et al. 2014. “Status and Ecological Effects of the World’s Largest Carnivores.” Science 343 (6167): 1241484. 2. Roemer, Gary W., Matthew E. Gompper, and Blaire Van Valkenburgh. 2009. “The Ecological Role of the Mammalian Mesocarnivore.” Bioscience 59 (2): 165–73. 3. Prugh, L. R., C. J. Stoner, C. W. Epps, W. T. Bean, and W. J. Ripple. 2009. “The Rise of the Mesopredator.” Bioscience. http://www.bioone.org/doi/abs/10.1525/bio.2009.59.9.9. 4. Swanepoel, 2016. Carnivore presence assessment in a protected area in the grassland biome based on camera trap surveys; Verlorenvallei Mpumalanga. 5. Mucina, L., and M. C. Rutherford. 2006. “The Vegetation of South Africa, Lesotho and Swaziland.” https://www.cabdirect.org/cabdirect/abstract/20073221934. 6. Friedman, Y., and B. Daly. 2004. “Red Data Book of the Mammals of South Africa: A Conservation Assessment: CBSG Southern Africa.” Conservation Breeding Specialist Group (SSC/IUCN), Endangered Wildlife Trust, Saxonwold, South Africa.