PhD_Conf.of.Cand.Presentation_ARU_Feb2011.v1_no animation

Slide 1:

Confirmation of approval as a candidate for PhD Predictors of Large Mammal Density in Biodiverse Forests of the Udzungwa Mountains, Tanzania Trevor Jones, Department of Life Sciences Trevor.Jones@anglia.ac.uk

Background:

Background The Sixth Extinction (Leakey & Lewin , 1995) Large mammal declines across Africa’s PAs ( Craigie et al, 2010) Why? - Multiple Causes of High Extinction Risk in Large Mammal Species ( Cardillo et al, 2005) “Species’ vulnerability to local extinction can be highly variable, likely to depend on both threat type and biology” (Isaac & Cowlishaw , 2004) Predictor variables show inconsistent patterns across forested sites, especially in fragmented landscapes (Bowers & Matter, 1997; Chiarello , 2000; Pyritz et al, 2010) “Eastern Afromontane Biodiversity Hotspot” is prime candidate for large mammal extinctions (Jones et al, 2005, 2008)

Central Question:

Central Question Which natural and/or anthropogenic variables best predict the distribution and abundance of large mammals across a fragmented landscape of highly biodiverse montane forests? Secondary Questions Species : Which variables predict abundance of three focal species (Red duiker, Abbott’s duiker, Elephant ) Site : Which forest sites are at risk of losing their mammal communities? Methodology : Are my adopted methods practical and repeatable for large mammal surveys and monitoring in other African forests?

Objectives:

Map distribution and estimate relative abundance of large mammals at 22 sites in 10 forest patches of the Udzungwa Mountains. Measure and evaluate potential ecological and anthropogenic factors affecting variation in large mammal presence and abundance in each forest. Evaluate relative importance of ecological and anthropogenic factors in predicting distribution and abundance of key large mammal species across the region, using principal component and other multivariate analyses. Assess validity and usefulness of the adopted ‘ recce transect’ method for zoological research and monitoring in African tropical forests. Assess local and regional threats to each large mammal species and community, and disseminate appropriate conservation recommendations. Objectives

Structure of thesis:

Structure of thesis 1: Introduction 2: Methods and Study Area 3: Generating variables and relative indices for each site 4: Comparison of densities of large mammals at different sites 5: Identifying predictors of mammal density I: PCA 6: Identifying predictors of mammal density II: GLMM 7: Towards a practical mammal survey method for remote African forests 8: Synthesis and conservation implications

Slide 6:

Study Area: Udzungwa Mountains

Slide 7:

Udzungwa Mountains 10,000 km 2 (study site within ~ 5,000 km 2 ) 300m to 2600m asl Long dry season (6 months) and variable wet season (6 months) Critical ecosystem services, especially watershed, hydroelectric power Extremely high biodiversity, strict endemism, restricted-range spp New vertebrate species discovered almost every year Only large E Arc site with viable closed large mammal populations (though not large enough for elephants) Important centre point of ecological connectivity among PAs of southern Tanzania

Slide 8:

Udzungwa Mountains

Habitats of the Udzungwa Mountains:

Habitat type Altitude ( m) Dominant tree species Description Grassland and wooded grassland (WG) 300–1500 Acacia spp., Brachystegia spp. Bracken and grassland with scattered trees Woodland (W) 300–2000 Low elevation: Commiphora spp., Adansonia digitata Mid to high elevation: A cacia spp., Uapaka kirkiana Deciduous woodland with low canopy (to 20 m) variable from very dense to open Lowland forest (LF) 300–800 Funtumia africana , Erythrophleum suaveolens , Treculia africana , Lettowianthus stellatus Forest with deciduous and semi-deciduous trees, canopy 15–25 m with emergents to 50 m Sub-montane forest (SF) 800–1400 Parinari excelsa , Felicium decipiens , Harungana madagascariense Moist forest with mainly evergreen spp , canopy 25–40 m with emergents to 50 m Montane forest ( MF) 1400–2600 Parinari excelsa , Ocotea usambarensis , Hagenia abyssinica , Syzygium sp . Evergreen moist forest, with canopy height progressively lower with altitude Habitats of the Udzungwa Mountains

Variables:

Variables Dependent Variables: Presence/Absence and Density of large mammals i ) Encounter Rates ii) Density estimates Independent (Predictor) Variables: Natural and Anthropogenic (e.g. Patch size, forest type, canopy cover, disturbance, distance to road, etc)

Density: Sampling Methods:

Density: Sampling Methods Recce Transects Line Transects Camera-trapping

Slide 12:

Body mass (kg) Plus...

Slide 13:

= 30 species +

Slide 14:

A.E. Bowkett & T. Jones, 2010 Verifying Dung Identification : Size of dung pellets of 3 sympatric forest antelope species: C. harveyi (n=83), N. moschatus (n=38), C. spadix (n=23) Pellet length (mm) Pellet width (mm) Pellet length:width ratio

Transects - Summary of data collected:

Transects - Summary of data collected Type Recording method Example species Discrete dung pile Perpendicular distance from centre of transect to centre of pile, with tape measure Antelopes, elephant, bushpig , buffalo, primates Large latrine or scattered dung piles Distance along transect recorded only Hyrax , fruit bats Animal seen Perpendicular distance from centre line of transect (with rangefinder or estimated) Primates (antelopes) Animal heard Perpendicular distance from centre line of transect, with rangefinder or estimated (Primates , antelopes) Animal hole Perpendicular distance from transect with tape measure Height and width of hole opening Recently active or old/inactive Aardvark , honey badger, giant pouched rat

Slide 16:

Sites sampled

Slide 17:

Forest Site Date of survey No. of recce transects Length (m) Number of line transects Length (m) Nyumbanitu Nyumbanitu E Sept 07 20 17,550 10 3,350 Nyumbanitu W Sept 07 6 5,200 4 850 Ukami Ukami Sept 07 12 9,150 7 1,250 Ndu-L'mero Vikongwa Nov 08 12 12,053 4 4,000 Ndundulu N Nov 08 12 11,020 2 1,850 Luhomero E Oct 08 12 12,071 5 4,496 Luhomero W Oct 08 10 10,124 2 1,850 Ng'ung'umbi Jul 10 7 9,740 - - Uzungwa Scarp US chini Jul 08 15 14,619 - - US juu Jul 08 10 10,458 - - New Dabaga New Dabaga N Oct 09 9 9,990 - - New Dabaga S Oct 09 12 11,228 - - Matundu Matundu W1 Jul 08 9 9,400 - - Matundu W2 Jul 08 12 11,710 - - Matundu Ruipa Jul 09 12 10,853 - - Iwonde Iwonde Sept 09 11 9,608 - - Nyanganje Nyanganje W Sept 09 12 11,549 - - Nyanganje E Aug 09 12 11,095 - - Mwanihana 3 Rivers Aug 08 15 15,200 4 3,950 Mizimu Sept 08 9 8,494 3 2,576 Kising'a-Rugaro K-Rugaro SE Oct 07 12 19,750 - - K-Rugaro SW Oct 07 6 10,122 - - Total 249 250,984 41 24,172 Transects completed

Sampling 1: Encounter Rates:

Site Total length (m) Diurnal primate sightings Angolan colobus seen Angolan colobus heard Red colobus seen Red colobus heard Sykes monkey seen Sykes monkey heard Mangabey seen Mangabey heard US chini 14,619 0.41 0.14 0.00 0.07 0.14 0.00 0.07 0.21 0.14 US juu 10,458 0.38 0.19 0.00 0.00 0.00 0.10 0.00 0.10 0.29 New Dabaga S 11,228 0.18 0.09 0.00 0.09 0.00 0.00 0.00 0.00 0.00 New Dabaga N 9,990 0.30 0.20 0.10 0.00 0.10 0.10 0.00 0.00 0.00 K-Rugaro SW 10,122 0.10 0.10 0.00 0.00 0.00 0.00 0.20 0.00 0.00 K-Rugaro SE 19,750 0.10 0.10 0.05 0.00 0.00 0.00 0.20 0.00 0.00 Matundu W1 9,400 0.96 0.11 0.43 0.21 0.43 0.64 0.21 0.00 0.00 Matundu W2 11,710 0.51 0.00 0.09 0.00 0.00 0.51 0.17 0.00 0.00 Matundu Ruipa 10,853 0.74 0.18 0.18 0.28 0.18 0.28 0.09 0.00 0.00 Nyumbanitu W 5,200 0.76 0.38 0.38 0.38 0.00 0.00 0.19 0.00 0.00 Nyumbanitu E 17,550 1.13 0.68 0.34 0.39 0.17 0.06 0.51 0.00 0.00 Ukami 9,150 0.44 0.22 0.00 0.22 0.33 0.00 0.66 0.00 0.00 Ndundulu N 11,020 0.82 0.45 0.27 0.36 0.36 0.00 0.45 0.00 0.00 Vikongwa 12,053 0.50 0.17 0.08 0.00 0.08 0.08 0.41 0.00 0.00 Luhomero W 7,200 0.97 0.69 0.28 0.14 0.00 0.14 0.00 0.00 0.00 Luhomero E 14,995 0.80 0.27 0.13 0.27 0.00 0.27 0.20 0.00 0.00 Ng'ung'umbi 9,740 0.72 0.21 0.00 0.21 0.00 0.31 0.41 0.00 0.00 Iwonde 9,608 0.52 0.21 0.10 0.21 0.21 0.10 0.10 0.00 0.00 Nyanganje W 11,549 1.04 0.35 0.26 0.52 0.00 0.17 0.35 0.000 0.00 Nyanganje E 11,095 0.27 0.18 0.36 0.09 0.09 0.00 0.18 0.00 0.00 3 Rivers 15,200 0.46 0.26 0.00 0.13 0.13 0.07 0.33 0.00 0.20 Mizimu 8,494 0.35 0.12 0.00 0.24 0.24 0.00 0.35 0.00 0.00 Sampling 1: Encounter Rates

Primary consumers combined (for relative overall abundance; max. 16 species):

Primary consumers combined (for relative overall abundance; max. 16 species) Mean Encounters per km = species richness

Udzungwa red colobus :

Udzungwa red colobus Mean Encounters per km (n=62) (n=37)

Slide 21:

Sampling 2: “DISTANCE” sampling

Density Estimates from DISTANCE:

Density Estimates from DISTANCE Site n Density CV D LCL D UCL Model Truncation AIC US chini 98 2597 0.22 1653 4079 Half-normal 325 1051 US juu 133 5760 0.33 2835 11702 Half-normal 250 1423 New Dabaga S 387 14402 0.14 10714 19359 Half-normal 300 4158 New Dabaga N 119 2953 0.29 1544 5648 Half-normal none 1364 K-Rugaro SW 13 655 0.55 173 2476 Uniform none 119 K-Rugaro SE 122 5500 0.19 3665 8253 Half-normal 175 1157 Matundu W1 75 4904 0.30 2621 9178 Hazard 200 747 Matundu W2 154 7732 0.28 4304 13888 Half-normal none 1565 Matundu Ruipa 469 14435 0.18 9808 21246 Half-normal 350 5282 Nyumbanitu W 232 21297 0.18 14045 32291 Half-normal none 2448 Nyumbanitu E 309 8231 0.18 5719 11846 Half-normal 300 3304 Ukami 366 26029 0.15 19103 35467 Hazard 250 3603 Ndundulu N 392 17649 0.14 13112 23756 Half-normal 350 4179 Vikongwa 413 21466 0.12 16587 27779 Half-normal 250 4150 Luhomero W 96 4330 0.26 2504 7486 Half-normal none 1027 Luhomero E 111 4986 0.23 3051 8150 Half-normal none 1184 Ng'ung'umbi 116 6131 0.13 4615 8146 Half-normal 300 1175 Iwonde 443 20770 0.16 14786 29175 Half-normal 350 4862 Nyanganje W 338 14475 0.23 8901 23539 Half-normal 250 3528 Nyanganje E 150 4688 0.23 2873 7648 Half-normal 275 1625 3 Rivers 589 25534 0.12 19966 32654 Half-normal 200 5911 Mizimu 164 11681 0.24 6853 19909 Half-normal none 1663

Antelopes (dung piles, n=5289):

Antelopes (dung piles, n=5289) Dungs per km 2

Elephants (dung piles, n=1355):

Elephants (dung piles, n=1355) Dung piles per km 2

1. Generate Independent Variables:

1. Generate Independent Variables Predictor Data type Unit Habitat disturbance Scale Encounter rate Hunting pressure Ordinal Index of pressure Patch size Scale km 2 Canopy extent Scale km 2 Canopy cover Ordinal Categories 1-5 Habitat type Nominal Types 1-4 Altitude Ordinal Categories 1-5 Protective status Nominal Types 1-3 Distance to village Scale metres Distance to major road Scale metres Next Steps

Slide 27:

Next Steps, contd... Principal Components Analysis 3. Generalised Linear Mixed Model 4. Evaluate recce transects vs line transects

Slide 28:

Thankyou! To my supervisory team of Dawn, Guy & Nancy And to you all for not falling...

Slide 29:

Response to hunting Hunting Response Index = Encounter rate in heavily hunted Encounter rate in lightly hunted 1 = encounter rates the same (no impact) >1 = encounter rate higher in heavily hunted site <1 = encounter rate lower in heavily hunted site 0 = absent in heavily hunted site Additional Analyses [Johns & Skorupa, 1987; Isaac & Cowlishaw, 2004; Linder, 2008] Specialisation-Disturbance Hypothesis Density Compensation