The Pennsylvania State University

The Graduate School

ELEPHANTS WITHOUT BORDERS: HISTORICAL AND CONTEMPORARY

GENETIC CONNECTIVITY IN TANZANIA

A Dissertation in

Biology

by

George Martin Gwaltu Lohay

© 2019 George Martin Gwaltu Lohay

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

August 2019

The dissertation of George Martin Gwaltu Lohay was reviewed and approved* by the following:

Douglas R. Cavener Professor and Verne M. Willaman Dean, Eberly College of Science Dissertation Advisor

Stephen W. Schaeffer Professor of Biology, Associate Department Head of Graduate Education Biology Ombudsperson Chair of Committee

Katriona Shea Professor of Biology Alumni Professor in the Biological Sciences

John E. Carlson Professor of Molecular Genetics Director of The Schatz Center for Tree Molecular Genetics

George H. Perry Associate Professor of Anthropology and Biology

Anna B. Estes Assistant Research Professor, The Huck Institutes of the Life Sciences Adjunct Professor, The Nelson Mandela African Institution of Science and Technology Special member

*Signatures are on file in the Graduate School

iii ABSTRACT

African savanna elephants (Loxodonta africana) are ecologically important as ecosystem engineers and socio-politically as revenue earners for national economies and local communities. However, their population has declined due to poaching and loss of habitat as a result of an increase in the human population. Habitat loss and fragmentation makes most protected areas isolated because of blocking of wildlife corridors. This study covered four ecosystems (Serengeti, Tarangire-Manyara, Selous, and Ruaha) in Tanzania which have the largest elephant populations in the country to determine the extent of genetic diversity and population structure nuclear and mitochondrial DNA markers. We wanted to establish historical genetic connectivity using mitochondrial DNA and contemporary gene flow using microsatellite markers from DNA obtained non-invasively from fecal samples. We specifically wanted to determine if there is gene flow between the Serengeti and Tarangire-Manyara ecosystems and whether the genetic structure has substantially changed over the past 50 years. We assumed that the Greater Rift Valley between two ecosystems would also act a barrier to the gene flow. We collected 800 elephant fecal samples from the four ecosystems and performed genetic analyses at the Pennsylvania State University. Our results showed that the Serengeti elephants are genetically distinct from the Tarangire-Manyara. Elephants from Ngorongoro showed an admixture between the two ecosystems. We also identified that there was a higher genetic similarity of elephants between Ngorongoro and Lake Manyara compared to Lake Manyara and Tarangire. Also, Tarangire and Ruaha elephants shared the same population structure although they are more than 400 km apart. Within the Serengeti ecosystem, we identified two population clusters from south and north of the Serengeti. Our results suggest that even without any physical barriers, there is genetic differentiation. The analysis of nuclear and mitochondrial DNA showed significant population differentiation between the Ruaha and Selous ecosystems. We further found no evidence for female-mediated gene flow between Ruaha and Selous. Only 4% of elephants sampled in Ruaha shared a haplotype with the Selous Game Reserve.

iv We also developed a novel fecal-centric approach to assess the age and sex structure of elephants and validated it with a rapid demographic assessment. We compared the sex ratio of elephants between Serengeti National Park, Ngorongoro Conservation Area and Maswa Game Reserve which have different protection status. In Serengeti, the sex ratio for adult age classes was skewed in favor of females whereas, in Ngorongoro, the sex ratio was skewed in favor of males for elephants older than 25 years. Although poaching is the main explanation for the observed sex ratio in Serengeti, we speculate that differential survival rates between males and female could explain the differences in sex ratio, particularly for young elephants. Our findings provide baseline information about historical connectivity using the mitochondrial DNA and recent gene flow (using nuclear markers) between protected areas in Tanzania. This information may be used to inform laws to protect the existing wildlife corridors or to restore the blocked corridors. We have highlighted some wildlife corridors that may have been or are still very important for the elephants based on our data; these would be suitable targets for conservation and restorations

Keywords: African Savanna elephants, Population structure, genetic connectivity, corridors, fecal-centric, microsatellite markers, simple sequence repeats (SSRs), mitochondrial DNA (mtDNA), habitat loss, habitat fragmentation, wildlife corridors, age and sex structure, , Amelogenin gene (AMELX/Y)

v TABLE OF CONTENTS

LIST OF FIGURES ...... viii

LIST OF TABLES ...... xi

ACKNOWLEDGEMENTS ...... xii

: Introduction ...... 1

Background ...... 1 Elephant social structure ...... 1 History of the elephant population in the Serengeti ecosystem ...... 3 Threats facing elephant conservation ...... 4 Poaching ...... 4 Habitat loss and habitat fragmentation ...... 5 The role of a metapopulations in conservation ...... 6 Genetic impacts of small population size and fragmentation ...... 7 Microsatellite or Simple Sequence Repeats (SSR) ...... 7 Mitochondrial DNA ...... 8 Objectives...... 9 References ...... 12

: Genetic connectivity and population structure of African Savanna elephants in Tanzania ...... 21

Abstract ...... 21 Introduction ...... 22 Materials and methods ...... 24 Description of study areas ...... 24 The Serengeti ecosystem ...... 25 Tarangire-Manyara ecosystem ...... 26 Ruaha ecosystem ...... 28 Selous Ecosystem ...... 29 Field data sample collection ...... 29 DNA isolation ...... 30 Microsatellite Analysis ...... 31 PCR amplification and genotyping ...... 31 Genetic diversity and differentiation ...... 32 Population Structure ...... 33 Mitochondrial Sequencing and Analysis ...... 33 Results ...... 34 Microsatellite analyses ...... 34 Mitochondrial DNA analysis ...... 41 Discussion ...... 50 Connectivity between Tarangire and Manyara ...... 52

vi Ngorongoro-Manyara Corridor ...... 53 History of elephant re-colonization in northern Tanzania ...... 55 Implications for conservation ...... 56 References ...... 57

: Little evidence for female mediated-gene flow for the African savanna elephants between the Greater Ruaha and Selous ecosystems in Tanzania ...... 70

Abstract ...... 70 Introduction ...... 71 Materials and Methods ...... 73 Study Areas ...... 73 Field data sample collection ...... 74 Microsatellites ...... 75 Population Structure ...... 77 Mitochondrial Sequencing and Analysis ...... 77 Results ...... 78 Mitochondrial DNA ...... 82 Discussion ...... 88 Implications for conservation ...... 92 References ...... 93

: An accurate molecular method to sex elephants using PCR amplification of Amelogenin gene ...... 102

Abstract ...... 102 Introduction ...... 103 Materials and Methods ...... 104 Results and Discussion ...... 105 References ...... 108

: Assessment of age and sex structure of African savanna elephants in the Serengeti ecosystem using a novel fecal-centric method ...... 110

Abstract ...... 110 Introduction ...... 111 Materials and methods ...... 113 Study Areas ...... 113 Fecal-centric Approach ...... 116 DNA Isolation and Sex Determination ...... 117 Rapid Demographic Assessment Method ...... 118 Results ...... 119 Fecal-centric approach ...... 119 Rapid Demographic Assessment (RDA) ...... 121 Discussion ...... 123 Age and sex structure ...... 123 Poaching pressure ...... 126 Conclusion ...... 128 References ...... 129

vii : Early signs of genetic differentiation among African savanna elephants in the Serengeti and Tarangire-Manyara ecosystems in northern Tanzania ...... 137

Abstract ...... 137 Introduction ...... 138 Methods ...... 140 Results ...... 141 Discussion ...... 149 Implications for conservation ...... 151 References ...... 152 Supporting materials ...... 155

: Dissertation synthesis ...... 159

References ...... 165 APPENDICES ...... 168 Appendix A: Genotype data for the African Savanna elephants at 10 SSR loci from fecal samples collected between 2015 and 2017 in Tanzania and the GPS location where each sample was collected...... 168 Appendix B: Sequences of 33 mitochondrial DNA haplotypes identified in this study. We have included the frequency of each haplotype for each haplotype ...... 192

viii LIST OF FIGURES

Figure 2-1. Map of Tanzania showing four ecosystems where we collected elephant fecal samples...... 24

Figure 2-2. Map of protected areas in northern Tanzania showing the Serengeti and Tarangire-Manyara ecosystems...... 26

Figure 2-3. Map of Selous and Ruaha ecosystems in Tanzania showing sampling locations ...... 28

Figure 2-4. Hierarchical population STRUCTURE analysis for the African Savanna elephants in Tanzania using 10 microsatellite loci...... 36

Figure 2-5. Unrooted neighbor-joining tree for 688 African savanna elephants using 10 SSR loci. The numbers on the branches show bootstrap values (%). Serengeti ecosystem (blue), Selous (purple), Tarangire-Manyara (orange) and Ruaha (black). W=West, N=North, C-Central Serengeti...... 37

Figure 2-6. Principal coordinate analysis based on genetic distance obtained from 10 SSR loci for African Savanna elephants in Northern Tanzania. Using PCoA, we identified at least two clusters...... 38

Figure 2-7. Principal coordinate analysis for 10 SSRs using pairwise FST values in northern Tanzania...... 39

Figure 2-8. A correlogram showing spatial genetic autocorrelation (r) among elephants in northern Tanzania as a function of Euclidean distance...... 40

Figure 2-9. Relationship between genetic distance and geographic distance for African elephants in northern Tanzania. We have included the pairwise comparisons for each set of locations. For example, TNP& MAR are only 60 km apart but they have unexpectedly high FST value of 0.034...... 40

Figure 2-10. Map showing genetic relationship for African savanna elephants in Tanzania...... 46

Figure 2-11. Median-joining haplotype network based on mitochondrial DNA sequences from African savanna elephants in Tanzania. The size of circles is proportional to haplotype frequencies...... 47

Figure 2-12. A neighbor-joining tree constructed from 33 mitochondrial DNA sequences from this study and reference sequences from previous studies. Blue= East-central, Orange=Savanna wide, purple=southeast savanna...... 49

Figure 2-13. Wildlife corridors between Tarangire and Manyara, and between Lake Manyara and Ngorongoro ...... 52

ix Figure 2-14. Mitochondrial DNA haplotype distribution (subclades) between Ngorongoro and Lake Manyara. Within the NCA we did not observe elephants carrying southeast savanna (SS) subclade at the Crater...... 55

Figure 3-1. Map showing protected areas from which samples were obtained. Mikumi National Park is within the Selous ecosystem. Wildlife corridors between Selous and Udzungwa and Kilombero valley are closed due to agricultural expansion and human settlement (Jones et al., 2012)...... 74

Figure 3-2. Individual population assignment from Bayesian STRUCTURE and mtDNA subclades for African savanna elephants in central and southern Tanzania ...... 80

Figure 3-3. Principal Coordinate Analysis (PCoA) for 11 microsatellite loci ...... 81

Figure 3-4. Genetic spatial autocorrelation analysis for African savanna elephants in Ruaha and Selous ecosystems in Tanzania. Upper and lower error bars bound the 95% confidence interval about r as determined by bootstrap resampling (Peakall and Smouse, 2012)...... 82

Figure 3-5: A neighbor-joining phylogenetic tree showing the relationship among 12 haplotypes obtained from 55 individual elephants in the Ruaha and Selous ecosystems...... 85

Figure 3-6. Median-joining network reconstruction of L. Africana showing the genetic relationship among the cytochrome b haplotypes of mitochondrial DNA...... 86

Figure 4-1. Agarose gel electrophoresis (2% agarose) of PCR product using AMELX and AMELY primer sets for Loxodonta africana...... 106

Figure 5-1. Map of the Serengeti ecosystem in Tanzania, showing sampling location during the dry seasons of 2015-17 ...... 114

Figure 5-2. Age and sex structure of L.africana in the Serengeti ecosystem using fecal- centric and the rapid demographic assessment method...... 120

Figure 5-3 Operational sex ratio (OSR) and dependent ratio for SNP, NCA, and MGR using the rapid demographic assessment (RDA) and indirect method using dung circumference (Dung)...... 122

Figure 6-1. Pairwise comparison of FST value between sampling locations between young and old elephants ...... 141

Figure 6-2. Bayesian clustering using STRUCTURE program for different age groups of African Savanna elephants between the Serengeti and Tarangire ecosystems in northern Tanzania ...... 145

Figure 6-3. Bayesian clustering using STRUCTURE program for males and females of African Savanna elephants between the Serengeti and Tarangire ecosystems in north Tanzania ...... 146

x Figure 6-4. Principal coordinate analysis for (A) young <20 years and (B) old elephants >20 years in the Serengeti and Tarangire-Manyara ecosystems ...... 147

Figure 6-5. Correlogram showing spatial genetic autocorrelation (r) among (A) young (0- 20 years) and (B) old (25-60) African savanna elephants as a function of Euclidean distance...... 148

Figure 6-6. Correlogram showing spatial genetic autocorrelation (r) among (A) female (B) male African savanna elephants as a function of Euclidean distance...... 149

Figure 7-1. Recorded incident of crop raiding by elephants in Oldeani village adjacent to the Ngorongoro conservation Area (from left William Metamei, NCA Game Officer, a farmer and George Lohay): Photo credit: James Madeli ...... 161

xi LIST OF TABLES

Table 2-1 Genotyped African Savanna elephants obtained from 15 locations in Tanzania .... 35

Table 2-2. Distribution of 33 mtDNA for African elephants in northern and southern Tanzania...... 42

Table 2-3. Genetic diversity of African savanna elephants based on the sequence of 622 bp of mtDNA ...... 43

Table 2-4. Pairwise Genetic differentiation (FST) for the African savanna elephants in Tanzania ...... 44

Table 3-1. Genotyped African savanna elephants (n=98) obtained from Ruaha and Selous Game reserve in 2017...... 79

Table 3-2. Distribution of the 12 observed mitochondrial DNA (D-loop) haplotypes from a sample of 55 African savanna elephants from three localities in Tanzania...... 83

Table 3-3 Parameters of genetic diversity based on mitochondrial DNA sequence data for two populations of African elephants ...... 84

Table 3-4 List of haplotypes used in this study ...... 87

Table 4-1.The number of females and males identified using AMELX/Y primers. For each location, the number of males and females are shown for reference sex and AMELX/Y sexing method...... 106

Table 5-1. The number of elephants sampled at SNP, NCA and MGR between 2015 and 2017. The numbers in parentheses denotes the number of individuals as a proportion of the total population size (TAWIRI, 2015) ...... 121

Table 6-1. Genetic distance measures among elephant populations for two age groups; young (below diagonal) and old (above diagonal) ...... 142

Table 6-2. Analysis of molecular variance (AMOVA) of genetic diversity of African elephants using 10 SSRs loci ...... 143

Table 6-3. Population-specific FIS indices per polymorphic locus for young elephants ...... 143

Table 6-4. Population-specific FIS indices per polymorphic locus for old elephants ...... 144

xii ACKNOWLEDGEMENTS

First and foremost, I would like to express my sincere gratitude to my advisor, Dr. Douglas Cavener, for being a great mentor with his continuous tireless support throughout my program here at Penn State. I thank Dr. Cavener lab members particularly, Dr. Barbara McGrath, Jingjie Hu, Becky Bourne, Chelsea Hudson and Raza Ellahi for helping me with the laboratory analyses and providing a great support throughout my program, without them, this work would have not been possible. I also thank previous graduate students, Dr. Carrie Lewis, Dr. Siying and Brandon Wood for their support. I thank my undergraduate assistant, Thomas Schneyer, for helping in DNA isolation and determining sex for some samples. I thank Dr. Anna Estes for supporting this project especially for helping in grant proposal writing and providing a vehicle for fieldwork. A sincere gratitude to Dr. Casey Weathers and Dr. Di Wu from Dr. Carlson lab of the Penn State University for their assistance in the genetic analyses. A special thanks to my committee members: Dr. John Carlson, Dr. Stephen Schaeffer, Dr. Katriona Shea, Dr. Anna B. Estes, and Dr. George Perry, for their support and advice throughout my Ph.D. program. I thank the Amboseli Elephant Research Conservation Project, Tarangire Elephant Project, especially Dr. Charles Foley and his research assistant Mr. John Mkindi as well as Mr. Lameck Mkuburo and Trevor Jones from the Southern Tanzania Elephant Project for their training and support during data collection. I also extend my gratitude to all field assistants and park rangers from all protected areas where I conducted my fieldwork. I thank Mr. Nganana Papalay, Loliondo District officer for helping me getting the permits and other field logistics in Loliondo. I thank Dr. Dennis Ikanda, Deus Bwenge, and Dennis Teti from Kingupira for their support during data collection in Selous. I thank Edwin Kisega and my friend Shedrack Mungure for their help during the fieldwork in Grumeti and Ikorongo Game Reserve. Data collection because easy and fun, I couldn’t imagine doing it myself. I thank Mr. Dennis Minja for providing me with the accommodation in Serengeti during the field work. I finally thank my wife Grace Malley and our son Alvin, my parents, Mr. & Mrs. Martin Gwaltu, all my eight siblings: Mary, Josephine, Ester, Paul, Deogratius, Sylvia, Julius, and James, for their

xiii moral support throughout my program. To all my friends in Tanzania and at Penn State and all over the world for your moral support particularly, Dr. Sixtus Aguree, Dr. Anne Odele, Leslie Pobee, Anica Massas, Dr. Ruth Pobee and Sam Vohsen. I thank the Cleveland Metropark Zoo, Wildlife Conservation Society and the Huck Institutes of the Life Sciences at Pennsylvania State University for funding my research. I thank Tanzania Wildlife Research Institute, Tanzania National Park Authority, Tanzania Forest Agency Services (TFS) and Tanzania Wildlife Authority for providing permissions to conduct research in protected areas. I thank the Frankfurt Zoological society, particularly Mr. Gerald Bigurube for providing me with a vehicle for field work in 2015.

1

: Introduction

Background

Across Africa, savanna elephant (Loxodonta africana) populations have declined by 72%, from 1.3 million in 1979 to around 350,000 in 2014 (Bartáková et al., 2018; Cerling et al., 2016; Kikoti et al., 2015; Lee et al., 2013). Illegal hunting for ivory is the primary threat, despite a Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) international ban of the ivory trade. Tanzania is one of the most important countries for the conservation of elephants. However, the population declined from 100,000 in 2009 to 43,000 in 2014 (TAWIRI, 2015). Within Tanzania, the most significant number of elephants is found in the Selous and Ruaha ecosystems, although these protected areas have experienced a rapid decline in recent years. In northern Tanzania, particularly in the Serengeti and Tarangire ecosystems, the populations are growing (Foley & Faust, 2010; Jones et al., 2018; TAWIRI, 2014).

Elephant social structure

Elephants are highly intelligent, socially complex, matriarchal animals (Gobush et al. 2007), with a fusion-fission social structure (Hollister-Smith 2007). Females live in natal herds consisting of closely related females and their offspring. Family unit size typically ranges from 2 to 52 individuals (Moss and Lee, 2006). Female elephants are philopatric and remain with their natal herd (Archie et al., 2007); whereas males are ejected from the herd upon sexual maturity and subsequently facilitate gene flow between family groups (Archie et al., 2007, Hollister-Smith et al., 2007). Young males leave their natal groups between the ages of 8-12 (Foley et al. 2012, Gobush et al., 2007). After they move

2 out of the family groups, young males may form groups with other young males from the same family or a neighboring family and stay together. Males usually have a more extensive home range than females. Females start giving birth for the first time when they are about 12 years old (Poole et al., 1981), although this varies a lot by population density. Males become sexually mature at the age of 15.5 years, but their probability for paternity increases with age (Hollister-Smith et al., 2007). Puberty among African elephant males occurs between the ages of 9 and 15 (Short et al. 1967; Lee 1986). Males have one of the most extended delays in reproduction compared to other mammals. Although males reach sexual maturity at 15.5 years (Laws, 1969; Poole, 1994), males rarely mate until they are in their 30s (Poole, 1994), with larger and older bulls getting most of the mating opportunities (Poole 1987, 1994). The larger size of adult males and their periodic elevated levels of testosterone (called musth) gives them a competitive advantage over the younger males. Musth is a period of sexual activity in male elephants, signified by high testosterone levels, urine dribbling, green penis syndrome and swollen temporal glands (Poole & Moss, 1981). Females become sexually mature when they are ten years and give birth for the first time when they are 13.8 years (Lee et al.,1986). Females can give birth up to the age of 50 years, which means elephants have significant overlap between generations (Lee et al., 1986). Elephants are led by one adult female, usually the oldest in the group. Matriarchs are the repositories of social and ecological knowledge within elephant breeding herds (Evans & Harris, 2008). Matriarchs play a significant role in finding essential resources within their habitats and keep the memory of all-important areas such as water and food sources, dispersal areas, and migratory routes between different seasons of the year. When a matriarch dies, another older female in the groups takes the role. When a family group becomes too large, some females may split to form another family, not far from their natal families. Group sizes change in wet and dry seasons, with groups at the smallest family unit in dry season but coming together with bond groups when resources allow. Herd size increases spontaneously at the onset of each rains, and declines progressively through the dry season as food reserves diminish (Western & Lindsay, 1984). Elephants tend to aggregate when they experience threats. When threatened by poachers or human activity,

3 elephants sometimes form large aggregations, with smaller herds merging for safety (Eltringham & Malpas, 1980). Elephants are not territorial. Their home range overlaps with 2 other family groups. Elephants have an extensive home range between 10 and 10,738 km (Douglas-Hamilton et al., 2005). Blanc (2003) and Gadd (2005) established that more than 84% of elephants’ habitat is outside protected areas (Blanc 2003, Douglas-Hamilton 2005), though human habitat modification acts to increasingly restrict elephants to protected areas (Douglas-Hamilton et al., 2005)

History of the elephant population in the Serengeti ecosystem

African elephants have experienced population decline throughout Africa as a result of illegal poaching since the 19th century. Most recently, there were peaks in ivory poaching in the 1970s and ‘80s, and in the past decade. In 1979, there were about 1.3 million elephants in Africa ( Douglas-Hamilton 1979b) but they declined to around 352,271 by 2014 (the Great Elephant Census). In Tanzania alone, the population of elephants declined from 109,051 in 2009 to only 43,330 in 2014 (Mduma et al., 2014) - in five years about 85,181 elephants were killed in Tanzania. The most intensive poaching in the past decade occurred in Southern and Western Tanzania. The Serengeti ecosystem, like other protected areas in Africa, experienced elephant population declines due to ivory poaching. Sinclair et al., (2008) stated that in the 1860s there were more than 4000 elephants in Serengeti, but their number declined to zero in 1890. The number of elephants remained low until the establishment of Serengeti as a protected area in the 1950s. The elephant population increased to about 3000 in 1975 (Sinclair et al., 2008). However, heavy poaching hit Serengeti in the 1970s and 1980s which caused a decline to about 400 elephants (Sinclair et al., 2008). The population recovered following the CITES ban of the international ivory trade in 1989. Since then, the Serengeti elephants have been increasing. Reduced poaching followed by natural population growth and immigration of elephants from the Masai Mara in Kenya (Morrison et al., 2017) explain the increase in population. Although the Serengeti elephant population

4 is rebounding from the previous poaching episodes, efforts to protect them should be extended to address habitat loss and human-elephant conflicts which are other threats that affect elephant populations.

Threats facing elephant conservation

Poaching

Unlawful killing for ivory has driven the global decline in African elephants (Poole & Thomsen, 1989; Wittemyer et al., 2014) and remains the main challenge today. Poaching targets larger individuals for their more enormous tusk sizes (Mondol et al., 2014). Although poaching for ivory concentrates on single adult males (Gobush et al., 2008), poachers also target female matriarchs as they also have large tusks and are easier to find than solitary males (Poole, 1989). Poaching disrupts kin-based association patterns, decreases the quality of elephant social relationships, and increases male reproductive skew, with significant consequences for population health and the maintenance of genetic diversity (Archie & Chiyo, 2012). Heavily poached populations tend to have a skewed sex ratio in favor of females (Jones et al., 2018; Poole & Thomsen, 1989). Legal or illegal selective removal of old related matriarchs from elephant herds has a long-lasting negative consequence for the survivors (Gobush et al., 2008). For example, 75% of elephants in Mikumi National Park in Tanzania were poached before 1989. Fifteen years later, about 30% of groups had a single adult female, 33% were non-reproductive, group sizes were very small (2.2 animals), and only 14% of family groups had matriarchs > 30 years (Gobush et al. 2008). Thus, severe poaching alters behavioral patterns and social hierarchies (Gobush et al., 2008). Elephants in Tarangire National Park (Tanzania) were also heavily poached before 1989, but after improving protection, the population recovered (Foley and Faust, 2010). Elephant groups without matriarchs in Tarangire died during the drought in 1993 because they did not know where to go to access water but

5 groups with matriarchs older than 30 years survived (Foley et al., 2008). Since old females hold a unique social role in their families, their removal impairs groups social functioning and survival, elevates physiological stress and reduces reproductive output among females left behind (Gobush et al., 2008). Poaching has also been associated with an increase in the proportion of individuals without tusks. In less-disturbed populations, 2-4% of African elephants are tuskless. However, for heavily poached populations, their percentages significantly increase. For example, at Queen Elizabeth National Park in Uganda, tusklessness increased from 1-2% in the 1920s and 1930s to 10% in the 1990s, with 66% of females over the age of 40 years lacking tusks (Abe, 1996, Whitehouse, 2002).

Habitat loss and habitat fragmentation

An increase in human activities leads to habitat loss and fragmentation which harms biodiversity conservation. Habitat fragmentation is defined as a landscape-scale process involving both habitat loss and the breaking apart of habitat (Fahrig, 2003). Habitat loss has significant, consistently adverse effects on biodiversity which are thought to exceed habitat fragmentation effects (Fahrig, 2003; MEA 2005). Habitat loss is usually associated with habitat fragmentation because a continuous habitat is transformed into smaller patches which will have less total area compared to the original matrix (Wilcove, McLellan & Dobson 1986). Habitat fragmentation in itself, i.e., the splitting up and isolation of habitats independent of habitat loss, has weak effects on biodiversity (Fahrig 2003). However, both habitat loss and fragmentation lead to loss of wildlife corridors. Several definitions have been used to describe wildlife corridors. The Ninth U.S Circuit Court of Appeals defined corridors as “avenues along which wide-ranging animals can travel, plants can propagate, genetic interchange can occur, populations can move in response to environmental changes and natural disasters, and threatened species can be replenished from other areas” (Walker & Craighead, 1997). A corridor was also defined as a linear landscape element that connects two or more patches of natural habitat and function to facilitate movement (Soulé & Gilpin, 1991). The term “connectivity” has also been used

6 to describe the extent to which flora and fauna can move among patches, rather than the linear landscape element described as corridor (Hansson 1995). Sometimes corridors are referred to as habitat corridors, wildlife corridors or ecological structures (Hilty et al., 2006). Connectivity is inherently about the degree of movement of organisms or processes (Crooks and Sanjayan, 2006).

The role of a metapopulations in conservation

A metapopulation is a set of discrete populations of the same species, in the same general geographic area, that may exchange individuals through migration, dispersal or human-mediated movements (Akçakaya, Mills, & Doncaster, 2006; Hanski 1999). Metapopulation structure has significant consequences for population ecology, the behavior of individuals, and genetic structure and evolution (Hanski and Gilpin 1997). Metapopulation theory is of the most significant value when applied to species in a physically patchy environment with sufficiently large patches to support local breeding populations (Hanski 1999). Most protected areas are becoming isolated due to habitat loss and fragmentation (Newmark, 2008) The concept of connectivity can be viewed from several perspectives: metapopulation (Moilanen and Hanski 2006), landscape ecology (Taylor 2006), and genetic connectivity (Frankham, 2002). Taylor (1993) defined landscape connectivity as “the degree to which the landscape impedes or facilitates movement among resources patches.” Some physical features within landscapes such as mountains and rivers may impede connectivity of species. Loss of connectivity among the population is a significant concern to the conservation of species and biodiversity at large. Goals of corridors are to facilitate daily movements, seasonal movements (e.g., migration), and dispersal (genetic exchange, mate-finding) or long-term persistence of species (Hilty et al., 2006). Dispersal among relatively isolated populations in a metapopulation is vital for preserving genetic diversity and re-establishing populations in habitat fragments where species have become locally extinct (McCullough 1996). Most protected areas are too small or too isolated to maintain a viable population for many wide-ranging species, for example, terrestrial

7 carnivores (Paquet et al. 2006). In this thesis, we examined the genetic structure of African savanna elephants to inform the extent of connectivity of the metapopulation.

Genetic impacts of small population size and fragmentation

Inbreeding and loss of genetic diversity are unavoidable in a small and isolated population (Frankham, 1995). Inbreeding increases homozygosity, exposes deleterious alleles and thus reduces survival and reproduction (Frankham et al., 2002). Furthermore, the risk of mutation accumulation increases with population fragmentation (Higgins and Lynch, 2001).

Microsatellite or Simple Sequence Repeats (SSR)

Molecular markers such as microsatellites or simple sequence repeats (SSRs) are useful in genetic studies because they directly reveal changes at the DNA sequence level (Mahoney & Springer, 2009). SSRs are short tandem repeats of DNA between 2-5 base pairs (bp) nucleotide that is widely distributed throughout the genome. SSRs are among the most viable DNA sequences because they are mostly co-dominant, abundant in genomes and highly reproducible and some have high rates of transferability across species (Saha et al., 2004). For this reason, SSRs have been important marker system in cultivar fingerprinting, diversity studies, molecular mapping, and marker associated selection (Goldstein & Schlotterer, 1999) and they can provide abundant and precise genetic information on population structure analysis. The polymerase chain reaction (PCR) amplifies alleles at microsatellites, and the data can be used to quantify genetic variations within and between species (Abdul-Muneer, 2014). Furthermore, SSRs are suitable in population genetic studies because they are locus-specific in nature, highly polymorphic and a small amount of DNA is sufficient for analysis (Abdul-Muneer, 2014; Allendorf, Hohenlohe, & Luikart, 2010; Avise, 2010; Miah et al., 2013). They also have the highest level of discrimination between genotypes relative to other markers (Pejic et al., 1998).

8 Limitations of using SSRs includes the appearance of stutter bands, presences of null alleles and homoplasy but some genetic analysis software such as CERVUS and MICRO-CHECKER have been developed to detect genotype errors (Karaket & Poompuang, 2012; Van Oosterhout et al., 2004). It is well established that a decline in genetic variation reduces the ability of the population to adapt to environmental changes and therefore decreases its long term survival (Arif & Khan, 2009). The loss of genetic diversity also results in lower individual fitness and poor adaptability (Allendorf et al., 2010; Avise, 2010; Frankham, 1995). There is a link between the loss of genetic diversity and endangered species. For examples, remaining wild cheetah population has lost genetic diversity at an alarming rate over the past 30 years (Terrell et al., 2016), which is attributed to a severe population bottleneck that occurred ~12,000 years ago (Driscoll et al., 2002; O’Brien et al., 1985). Genetic variation can be assessed using allelic diversity and heterozygosity because the ability of a population to evolve is affected both by heterozygosity and the number of alleles present. Effects of genetic diversity have a significant impact on allelic variety because random genetic drift will eliminate the low- frequency alleles very rapidly (Arnold et al., 2001). Effects of reduction of genetic diversity may not be evident in heterozygosity but have significant impacts on allelic diversity because random genetic drift will have a tremendous effect on allelic diversity and genetic drift will eliminate the low-frequency alleles very rapidly (Allendorf, 1986).

Mitochondrial DNA

Mitochondrial DNA has been widely used in studies of population genetics because it is easily isolated the mutation rate is relatively high, does not undergo recombination and it is haploid and maternally inherited (Freeland, Kirk, and Peterson, 2011). MtDNA can be used to measure genetic diversity by assessing haplotype diversity. A haplotype is a multi- site (SNP) genotype. When a population passes through a bottleneck event, it loses haplotypes depending on severity of the bottleneck but this may not be evident with the microsatellites. For this reason, mtDNA is suitable to infer past demographic history (Nabholz et al., 2008). A group of organisms or taxa that trace back to a single common

9 ancestor is monophyletic group or clade. There are two very distinct clades of mtDNA forms in African elephants, designated as Forest and Savanna clades (Debruyne, 2005), origination from Forest and Savanna respectively (Ishida et al., 2011b). S clade mtDNA is only found among savanna elephants, but both savanna and forest elephants carry F clade mtDNA. Previous studies (Ahlering et al., 2012; Ishia et al., 2013) suggested that the Rift Valley separated the distribution of haplotypes in northern Tanzania (between Serengeti and Tarangire ecosystems). However, these studies were based on small sample sizes and did not cover all protected areas within these two ecosystems. Furthermore, the genetic relationship between elephants in northern and southern Tanzania is not known.

Objectives

Our overall goal was to use population genetics to identify potential historical genetic connectivity among elephant populations in Tanzania and compare this with an assessment of corridors. The immediate objectives of my research were to assess genetic diversity and population structure of elephants in Tanzania using microsatellite markers and mitochondrial data obtained noninvasively. We also wanted to determine the age and sex structure of elephants to assess if elephant populations have been impacted by poaching in the Serengeti ecosystem. The results of my thesis will help inform enact landscape-scale conservation aimed at preserving connectivity between the highly threatened elephant populations in northern Tanzania and those in southern Kenya, where complementary studies are taking place. To understand the genetic connectivity of elephants in Tanzania, we defined the following specific aims: Aim 1: Our first aim was to identify genetic connectivity and population structure of the African elephants in Tanzania using mitochondrial DNA and 11 microsatellites. We compared the genetic structure and the FST values to determine the extent of genetic differentiation among subpopulations. Here we wanted to compare the genetic structure of all subpopulations making up the Serengeti and Tarangire Manyara ecosystems about

10 known wildlife corridors. We used the mitochondrial DNA haplotypes to determine the historical genetic connectivity of elephants with other populations in Africa using previously known mitochondrial DNA haplotypes. MtDNA also provided insights into elephant social structure. We also wanted to test the hypothesis that two groups of elephants recently recolonized the Serengeti following poaching-induced extirpation: one from the south and another one from the north (Sinclair et al., 2008). We also asked if there was population genetic differentiation within the Serengeti ecosystem which does not have any known gene flow barrier. We hypothesized that there was no genetic differentiation in the Serengeti ecosystem.

Aim 2: We determined whether there was genetic differentiation between the Ruaha and Selous ecosystems and whether the isolation is contemporary or historical. We precisely wanted to test if the Eastern Arc Mountains, particularly the Udzungwa Mountains, impede gene flow of elephants. Genetic studies on lions (Smitz et al., 2018) and sable antelope (Pitra et al., 2002) have shown a distinct separation of subpopulations west and east of the Eastern Arc Mountains. Because elephants have much larger home ranges than lions and sable, it is likely that the mountains do not separate them.

Aim 3: We assessed the age and sex structure of elephants in the Serengeti ecosystem using rapid demographic assessment and a novel fecal-centric approach. The Serengeti ecosystem has several protected areas with different protection level and experienced heavy poaching in the last 30 years. We developed a molecular method to accurately identify the sex of individuals from fecal samples using the Amelogenin gene (AMELX/Y). We then used dung circumference to estimate the age of elephants and compare the results from both methods. We used the rapid demographic assessment (RDA) approach to validate the fecal centric approach in assessing the age and sex structure of elephant populations

11 Aim 4: We determined whether the elephant genetic structure has substantially changed over the past 50 years between the Serengeti and Tarangire-Manyara ecosystems. Here we hypothesized that young elephants (less than 25 years of age) are more genetically differentiated than the older individuals (above 25 years of age) due to restricted gene flow as a result of habitat loss and fragmentation. We estimated the age of elephants using dung circumference measure in the field and identified sex using AMELX/Y method. We assumed that the effects of human activities on the migratory corridor are recent. Thus, individuals older than 25 years experienced higher gene flow rate than the young elephants. Because female elephants stay in family groups as opposed to males who are mainly responsible for gene flow (Nyakaana and Arctander, 1999), we also compared genetic structure of males and females to determine if there was the difference in genetic structure. We expect to observe significant genetic differentiation in females than males.

12 References

Abe, E. (1996). Tusklessness among the Queen Elizabeth National Park elephants, Uganda. Pachyderm 22:46-47. Abdul-Muneer, P. M. (2014). Application of microsatellite markers in conservation genetics and fisheries management: recent advances in population structure analysis and conservation strategies. Genetics Research International, 2014, 691759. https://doi.org/10.1155/2014/691759 Ahlering, M. A., Eggert, L. S., Western, D., Estes, A., Munishi, L., Fleischer, R., Maldonado, J. E. (2012). Identifying source populations and genetic structure for savannah elephants in human-dominated landscapes and protected areas in the Kenya-Tanzania borderlands. PloS One, 7(12), e52288. https://doi.org/10.1371/journal.pone.0052288 Akçakaya, H. R., Mills, G., & Doncaster, C. P. (2006). The Role of Metapopulations in Conservation. Macdonald Key Topics in Conservation Biology, 5(September), 64–84. https://doi.org/10.1002/jcp.22422 Allendorf, F. W. (1986). Genetic drift and the loss of alleles vs heterozygosity. Zoo Biology, 5, 181–190. Retrieved from citeulike-article-id:540565 Allendorf, F. W., Hohenlohe, P. A., & Luikart, G. (2010). Genomics and the future of conservation genetics. Nature Publishing Group, 11(10), 697–709. https://doi.org/10.1038/nrg2844 Archie, E. A., & Chiyo, P. I. (2012). Elephant behaviour and conservation: social relationships, the effects of poaching, and genetic tools for management. Molecular Ecology, 21(3), 765–778. https://doi.org/10.1111/j.1365- 294X.2011.05237.x Archie, E. A., Hollister-Smith, J. A., Poole, J. H., Lee, P. C., Moss, C. J., Maldonado, J. E., … Alberts, S. C. (2007). Behavioral inbreeding avoidance in wild African elephants. Molecular Ecology, 16(19), 4138–4148. https://doi.org/10.1111/j.1365- 294X.2007.03483.x

13 Arif, I. A., & Khan, H. A. (2009). Molecular markers for biodiversity analysis of wildlife animals : a brief review. Animal Biodiversity and Conservation, 32(1), 9–17. https://doi.org/10.1007/s10722-009-9526-z Arnold, S. J., Pfrender, M. E., & Jones, A. G. (2001). The adaptive landscape as a conceptual bridge between micro- and macroevolution. Genetica, 112–113, 9–32. https://doi.org/10.1023/A:1013373907708 Avise, J. C. (2010). Perspective : conservation genetics enters the genomics era. Conservation Genetics, 11, 665–669. https://doi.org/10.1007/s10592-009-0006-y Bartáková, V., Reichard, M., Janko, K., Polačik, M., Blažek, R., Reichwald, K., McLean, I. (2018). Conserving elephants depend on a total ban of ivory trade globally. Biodiversity and Conservation, 13(March), 37–43. https://doi.org/10.1007/s10531-018-1534-x Blanc, J.J., Thouless, C.R., Hart, J.A., Dublin, H.T., Douglas-Hamilton, I., Craig, G.C. and Barnes, R.F.W. 2003. African elephant status report 2002. Occasional Paper of the IUCN Species Survival Commission 29. Gland: IUCN/SSC African Elephant Specialist Group. Cerling, T. E., Barnette, J. E., Chesson, L. A., Douglas-Hamilton, I., Gobush, K. S., Uno, K. T., Xu, X. (2016). Radiocarbon dating of seized ivory confirms rapid decline in African elephant populations and provides insight into illegal trade. Proceedings of the National Academy of Sciences, 113(47), 13330–13335. https://doi.org/10.1073/pnas.1614938113 Cox, C.Barry, Moore, Peter.D and Ladle, Richard (2016), Biogeography: An ecological and Evolutionary Approach, Wiley-Blackwell, pp. 106 Crooks, K. R., and Sanjayan, M., 2006. Connectivity conservation: maintaining connections for nature. Pages 1– 20 in K. R. Crooks and M. Sanjayan, editors. Connectivity Conservation. Cambridge University Press, Cambridge, United Kingdom . Debruyne, R. (2005). A case study of apparent conflict between molecular phylogenies: the interrelationships of African elephants. Cladistics, 21, 31–50.

14 Douglas-Hamilton I (1972). On the ecology and behaviour of the African elephant: the elements of Manyara. PhD thesis, Oxford University, UK. Douglas-Hamilton, I. 1979b. African Elephant Ivory Trade. Final Report to US Fish and Wildlife Service. Douglas-Hamilton, I., Krink, T., & Vollrath, F. (2005). Movements and corridors of African elephants in relation to protected areas. Naturwissenschaften, 92, 158– 163. https://doi.org/10.1007/s00114-004-0606-9 Driscoll, C. A., Menotti-Raymond, M., Nelson, G., Goldstein, D., & O’Brien, S. J. (2002). Genomic microsatellites as evolutionary chronometers: A test in wild cats. Genome Research, 12(3), 414–423. https://doi.org/10.1101/gr.185702 Eltringham, S., & Malpas, R. (1980). The decline in elephant numbers in Rwenzori and. African Journal of Ecology, 18, 73–86. Evans, K. E., & Harris, S. (2008). Adolescence in male African elephants, Loxodonta africana, and the importance of sociality. Animal Behaviour, 76(Walters 1987), 779–787. https://doi.org/10.1016/j.anbehav.2008.03.019 Fahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics, 34(1), 487–515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419 Foley, C., Pettorelli, N., and Foley, L., (2008). Severe drought and calf survival in elephants, Biological. Letters, 4, 541-544 Foley, C. A. H., & Faust, L. J. (2010). Rapid population growth in an elephant Loxodonta africana population recovering from poaching in Tarangire National Park, Tanzania. Oryx, 44(02), 205–212. https://doi.org/10.1017/S0030605309990706 Frankham, R. (1995). Conservation Genetics. Annual Review of Genetics, 29, 305–327. https://doi.org/10.1146/annurev.genet.29.1.305 Frankham, R., Ballou, J., & Briscoe, D. . (2002). Introduction to conservation genetics. Cambridge, UK: Cambridge University Press. Freeland JR, Kirk H, Petersen S. (2011) Molecular Ecology. Wiley & Sons; Chichester, UK

15 Gadd, M. E. (2005). Conservation outside of parks: attitudes of local people in Laikipia, Kenya. Environmental Conservation, 32(1), 50–63. https://doi.org/10.1017/s0376892905001918 Gobush, K. S., Mutayoba, B. M., & Wasser, S. K. (2008). Long-term impacts of poaching on relatedness, stress physiology, and reproductive output of adult female african elephants. Conservation Biology : The Journal of the Society for Conservation Biology, 22(6), 1590–1599. https://doi.org/10.1111/j.1523- 1739.2008.01035.x Goldstein, D. B., & Schlotterer, C. (1999). Microsatellites-Evolution and Applications. Heredity, 83, 633–635. https://doi.org/10.1037/h0089386 Hanski, Ikka, 1999. Metapopulation Ecology, Oxford University Press, UK Hanski, I., & Gilpin, M. . (1997). Metapopulation Biology: Ecology, Genetics, and Evolution. Academic Press, New York. Hanski, I., & Saccheri, I. (2006). Molecular-level variation affects population growth in a butterfly metapopulation. PLoS Biology, 4(5), 719–726. https://doi.org/10.1371/journal.pbio.0040129 Hansson, L. (1991). Dispersal and connectivity in metapopulations. Biological Journal of the Linnean Society, 42(1–2), 89–103. https://doi.org/10.1111/j.1095- 8312.1991.tb00553.x Higgins, K. and Lynch, M. 2001. Metapopulation extinction caused by mutation accumulation, Proc. Natl Acad. Sci, USA 98: 2928-2933 Hilty, J., W.Z. Lidicker Jr. and A.M. Merenlender. 2006. Corridor Ecology. Island Press, Washington D.C. Hollister-Smith, J.A., Poole, J.H., Archie E.A., Vance, E.A., Georgiadis, N.J., Moss, C. J and Alberts, S. C., Age, musth and peternity success in wild male African elephants, Loxodonta Africana, Animal Behaviour,74, 287-296 Hynes G, 1991. Mammoths, Mastodonts and elephants:Behavior, and the fossil records. Cambridge University Press, Cambridge, UK Ishida, Y., Demeke, Y., Van, P. J., De Groot, C., Georgiadis, N. J., Leggett, K. E. A., Roca, A. L. (2011). Distinguishing Forest and Savanna African Elephants Using

16 Short Nuclear DNA Sequences, Journal of Heredity, 102(5), 610–616. https://doi.org/10.1093/jhered/esr073 Jones, T., Cusack, J. J., Pozo, R. A., Smit, J., Mkuburo, L., Baran, P., Foley, C. (2018). Age structure as an indicator of poaching pressure: Insights from rapid assessments of elephant populations across space and time. Ecological Indicators, 88(November 2017), 115–125. https://doi.org/10.1016/j.ecolind.2018.01.030 Karaket, T., & Poompuang, S. (2012). CERVUS vs. COLONY for successful parentage and sibship determinations in freshwater prawn Macrobrachium rosenbergii de Man. Aquaculture, 324–325, 307–311. https://doi.org/10.1016/j.aquaculture.2011.10.045 Kikoti, A. P., Griffin, C. R., Wasser, S. K., Brown, L., Mailand, C., Mondol, S., Weir, B. S. (2015). Genetic assignment of large seizures of elephant ivory reveals Africa’s major poaching hotspots. Wildlife & Fisheries Conservation, 349 (Open Access Dissertations. Paper 108.), 84–88. https://doi.org/10.1126/science.aaa2457 Laws, R.M. (1969) Aspects of reproduction in the African elephant Loxodonta Africana, Journal of Reproductive Fertility, Supplement 6, 193-217 Lee, P.C. (1986) Early social development among African elephant calves. Nat. Geogr. Res. 2, 388–401 Lee, E. J., Lee, Y. H., Moon, S. H., Kim, N. Y., Kim, S. H., Yang, M. S., … Han, M. S. (2013). The identification of elephant ivory evidences of illegal trade with mitochondrial cytochrome b gene and hypervariable D-loop region. Journal of Forensic and Legal Medicine, 20(3), 174–178. https://doi.org/10.1016/j.jflm.2012.06.014 Lowe WH, Allendorf FW (2010) What can genetics tell us about population connectivity? Molecular Ecology, 19, 3038–3051. Mahoney, C., & Springer, D. A., (2009). Genetic Diversity, Nova Science Publishers, Incorporated MEA (2005), Millennium Ecosystem Assessment. Ecosystems and Human Well-Being: Biodiversity Synthesis, World Resources Institute, Washington, DC, USA

17 McCullough, D. R. 1996. Metapopulations and Wildlife Conservation. Washington, DC: Island Press. Mduma, H., Musyoki, C., Maliti, H., Kyale, D. M., Nindi, S., Hamza, K., … Kohi, E. M. (2014). Aerial total count of elephants and buffaloes in the Serengeti-Mara Ecosystem, 32. Retrieved from http://www.tanzaniaparks.com/useful_docs/Mara_Elephant_Count_Report.pdf Miah, G., Rafii, M. Y., Ismail, M. R., Puteh, A. B., Rahim, H. A., Islam, N. K., & Latif, M. A. (2013). A review of microsatellite markers and their applications in rice breeding programs to improve blast disease resistance. International Journal of Molecular Sciences, 14(11), 22499–22528. https://doi.org/10.3390/ijms141122499 Mondol, S., Mailand, C. R., & Wasser, S. K. (2014). Male biased sex ratio of poached elephants is negatively related to poaching intensity over time. Conservation Genetics, 15(5), 1259–1263. https://doi.org/10.1007/s10592-014-0603-2 Morrison, T. A., Estes, A. B., Mduma, S. A. R., Maliti, H. T., Frederick, H., Kija, H., … Kohi, E. M. (2017). Informing Aerial Total Counts with Demographic Models : Population Growth of Serengeti Elephants Not Explained Purely by Demography. A Journal of the Society for Conservation Biology, 00(October), 1–8. https://doi.org/10.1111/conl.12413 Moss, C. J., & Lee, P. C. (2006). Female Reproductive Strategies. In The Amboseli Elephants (pp. 187–204). University of Chicago Press. https://doi.org/10.7208/chicago/9780226542263.003.0012 Moss, CJ. & Poole, J.H. (1983) Relationships and social structure In African elephants. In: Primate Social Relationship.: An Integrated Approach. (Ed. R.A. Hinde). Nabholz, B., Mauffrey, J. F., Bazin, E., Galtier, N., & Glemin, S. (2008). Determination of mitochondrial genetic diversity in mammals. Genetics, 178(1), 351–361. https://doi.org/10.1534/genetics.107.073346 Newmark, W. D. (2008). Isolation of African protected areas. Frontiers in Ecology and the Environment, 6(6), 321–328. https://doi.org/10.1890/070003

18 O’Brien, S. J., Roelke, E., Marker, L., Newman, A., Winker, C. ., Meltzer, D., … Wildt, D. . (1985). Genetic Basis for Species Vulnerability in the Cheetah. Science, 227(4693), 1428–1434. Paquet, P.C., Alexander, S.M., Swan, P.L. & Darimont, C.T. (2006). The influence of natural landscape fragmentation and resource availability on connectivity and distribution of marine gray wolf (Canis lupus) population on the Central Coast, British Columbia, Canada. In Crooks and M.A. Sanjayan (eds), Connectivity Conservation, pp. 130-156. Cambridge University Press, Cambridge, UK. Pejic, I., Ajomne-Marsan, P., Morgante, M., Kozumplick, V., Taramino, G., & Motto, M. (1998). Comparative analysis of genetic similarity among maize inbred lines detected by RFLPs, RAPDs, SSRs and AFLPs, Theor Appl Genet, 97:1248-1255 Pitra, C., Hansen, A. J., Lieckfeldt, D., & Arctander, P. (2002). An exceptional case of historical outbreeding in African sable antelope populations. Molecular Ecology, 11(7), 1197–1208. https://doi.org/10.1046/j.1365-294X.2002.01516.x Poole. J.H. (1994) Sex differences in the behaviour of African elephants. In: The Differences Between the Sexes (Eds. R. Short &P. Lynch) Cambridge University Press Cambridge Poole, J. H., & Moss, C. J. (1981). Musth in the African elephant Loxodonta africana. Nature. Poole, J. H., & Thomsen, J. B. (1989). Elephants are not beetles. Oryx, 23(4), 188–198. Saha, M. C., Mian, R., Zwonitzer, J. C., Chekhovskiy, K., & Hopkins, A. A. (2004). An SSR- and AFLP-based genetic linkage map of tall fescue (Festuca arundinacea Schreb.). Theoretical and Applied Genetics, 110(2), 323–336. https://doi.org/10.1007/s00122-004-1843-1 Short, R. V., Mann, T. & Hay, M. F. 1967. Male reproductive organs of the African elephant, Loxodonta africana. Journal of Reproduction and Fertility, 13, 517e536. Sinclair, A. R. E., Hopcraft, J. G. C., Olff, H., & Mduma, S. A. R. (2008). Historical and Future Changes to the Serengeti Ecosystem, (1972), 7–46. Smitz, N., Jouvenet, O., Ligate, F. A., Crosmary, W. G., Ikanda, D., Chardonnet, P.,Michaux, J. R. (2018). A genome-wide data assessment of the African lion

19 (Panthera leo) population genetic structure and diversity in Tanzania. PLoS ONE, 13(11). https://doi.org/10.1371/journal.pone.0205395 Soulé, M. E., & Gilpin, M. . (1991). The theory of wildlife corridor capability. Nature Conservation: 2, 3–8. https://doi.org/10.1080/14635240.2007.10708095 TAWIRI. (2014). Aerial Total Count of Elephants and Buffaloes in the Serengeti-Mara Ecosystem. TAWIRI. (2015). Population Status of Elephant in Tanzania 2014.TAWIRI Aerial Survey Report. Taylor, P. D., Fahrig, K. H., & Merriam, G. (1993). Connectivity is a vital element of landscape structure. Oikos, 68(3), 371–373. Taylor, P. D., Fahrig, L., and With, K. A., 2006. Landscape connectivity: a return to the basics. In K. R. Crooks and M. Sanjayan, (eds). Connectivity conservation, pp 29– 43, Cambridge University Press, Cambridge, United Kingdom . The Great elephant census retrieved on http: www.greaterelephantcensus.com/final- report Terrell, K. A., Crosier, A. E., Wildt, D. E., O’Brien, S. J., Anthony, N. M., Marker, L., & Johnson, W. E. (2016). Continued decline in genetic diversity among wild cheetahs (Acinonyx jubatus) without further loss of semen quality. Biological Conservation, 200 (August), 192–199. https://doi.org/10.1016/j.biocon.2016.05.034 Van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M., & Shipley, P. (2004). MICRO- CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4(3), 535–538. https://doi.org/10.1111/j.1471-8286.2004.00684.x Walker, R., & Craighead, L. (1997). Analyzing Wildlife Movement Corridors in Montana Using GIS. Montana, The Magazine of Western History, (July 1997), 1– 11. Retrieved from http://gis.esri.com/library/userconf/proc97/proc97/to150/pap116/p116.htm Western, D., & Lindsay, W.K. (1994). Abdul-Muneer, P. M. (2014). Application of microsatellite markers in conservation genetics and fisheries management: recent

20 advances in population structure analysis and conservation strategies. Genetics Research International, 2014(691759), 1–11. https://doi.org/10.1155/2014/691759 Wilcove, D.S., McLellan, C.H. & Dobson, A.P. (1986) Habitat fragmentation in the temperate zone.Conservation Biology (ed. M.E. Soule), pp. 237–256. Sinauer, Sunderland, MA, USA Wilcox, Bruce, A., & Murphy, Dennis, D. (1985). Conservation Strategy: The Effects of Fragmentation on Extinction. American Society of Naturalists, 125(6), 879–887. Wittemyer, G., Northrup, J. M., Blanc, J., Douglas-Hamilton, I., Omondi, P., & Burnham, K. P. (2014). Illegal killing for ivory drives global decline in African elephants. Proceedings of the National Academy of Sciences, 111(36), 13117– 13121. https://doi.org/10.1073/pnas.1403984111

21

: Genetic connectivity and population structure of African Savanna elephants in Tanzania

George G. Lohay, T. Casey Weathers, Anna B. Estes and Douglas R. Cavener

Abstract

An increase in the human population is accelerating isolation of protected areas, threatening the sustainability of elephant populations and limiting gene flow between them. Although poaching of elephants remains a significant factor driving the decline of populations, habitat loss is another significant threat. Most wildlife corridors in Tanzania are endangered, and some are blocked completely. In this study, we assessed the genetic connectivity of elephants among subpopulations in the Tarangire-Manyara, Serengeti, Selous and Ruaha ecosystems using 11 microsatellite loci and mitochondrial DNA (mtDNA) to quantify gene flow among subpopulations. We collected 800 elephant fecal samples and successfully genotyped 711 at 11 loci and 558 mtDNA sequences from a hypervariable region of the control region. We used Bayesian model implemented in the STRUCTURE program to infer the number of subpopulations. Elephants from the Tarangire-Manyara ecosystem were genetically distinct from the Serengeti ecosystem. The mtDNA haplotype distribution suggests that The Great Rift Valley is a significant barrier to gene flow. Results from this study can be used to identify and prioritize wildlife corridors for protection.

Keywords: microsatellites, mitochondrial DNA, conservation genetics, population structure, gene flow, African Savanna elephants

22 Introduction

Habitat loss and fragmentation is a significant challenge in species conservation and one of the top threats to biodiversity (Henle et al. 2004). Human population growth near protected area (PA) boundaries is often higher than in comparable rural areas (Wittemyer et al. 2008). Population growth often brings changes in land use for agriculture and settlements, which leads to loss of buffer zones adjacent to, and corridors connecting PAs. This isolation of PAs decreases their effective size, limits gene flow between populations, and leads to increased human-wildlife conflicts. Elephant populations have been declining across Africa. Tanzania has lost over 60% (from 109,051 to 43,330) of its elephant population just in the past five years (Environmental Investigation Agency (EIA), 2014). While poaching and illegal ivory trade are the most severe immediate threat posed to the African elephants, range and habitat fragmentation remain a significant long-term threat to the species’ survival (CITES, 2014; Douglas-Hamilton, 1987; Wittemyer et al., 2014). Habitat fragmentation mainly affects far-ranging species, like the elephant. Elephants have extensive individual home ranges (10 to 10, 738 km2) and they show high fidelity to their home ranges and the corridors that they used traditionally (Desai and Baskaran, 1996). Habitat fragmentation resulting from human population growth and habitat conversion is a particular concern in Tanzania (Newmark, 2008) and threatens the connectivity of elephant populations that are becoming confined inside protected areas. Without connectivity, many PAs in Tanzania are too small to sustain their elephant populations and in many of these systems elephants disperse seasonally outside the protected areas (Estes et al., 2012). Genetic connectivity is the degree to which gene flow affects evolutionary processes within subpopulations (Lowe & Allendorf, 2010). Gene flow depends on movement between populations and the migrants mate within the new population. Genetic connectivity can be assessed using molecular markers. Molecular markers such as microsatellites also known as simple sequence repeats (SSRs) and D-loop haplotype are frequently adopted to determine the genetic diversity and population structure of species (Ramanadevi et al., 2013). SSRs loci have shown promise for assessment of genetic

23 diversity due to their high discriminatory power and comparatively low cost (Zong et al., 2015). Molecular genetics can assess the relationship between heterozygosity, level of inbreeding and the loss of alleles in a small population and the implications this might have for the adaptive potential of the remnant population (Frankel and Soule 1981; Allendorf, 1986). Understanding the level of genetic diversity can provide us with a theoretical basis to formulate future strategies for conservation (Zong et al., 2015). Low levels of genetic variation may limit a population’s future adaptability and evolutionary potential (Lacy, 1995) by predisposing it to disease (O’Brien et al., 1985) or compromising its ability to respond to significant environmental change that could seriously threaten its future survival. Serengeti and Tarangire-Manyara ecosystem are vital areas in Tanzania for the conservation of biodiversity. However, outside these protected areas, there is an increase in human population and the conversion of land to agriculture. For example, in the western Serengeti, human population growth between 1988 and 2002 was 3.5% per year, and the highest rates of agricultural conversion were close to the PA boundary (Estes et al., 2012). This growth is higher than the national average which is 2.7% per year. Human activities affect the movement of elephants within and between ecosystems by reducing the size of wildlife corridors or introducing connectivity barriers such as roads and human settlements. This limits the gene flow between the populations, threatening their long-term sustainability and in the Maasai Steppe, expanding cultivation towards Tarangire National Park (TNP) has severely restricted wildlife movements to dispersal areas outside the park by blocking their migratory corridors. Unfortunately, there is an overlap between land suitable for agriculture, migratory wildlife corridors, and wet season dispersal areas (Msoffe et al., 2011). Thus, the rapidity of rangeland conversion to farming presents significant threats to wildlife conservation and disrupts the ecosystem viability. Similarly, connectivity areas for the Tarangire-Manyara ecosystem (hereafter, TME) and the Serengeti ecosystem (henceforth, SE) may not be viable in the long term because of its increasing isolation by agricultural settlements (Mwalyosi, 1991). This work seeks to quantify genetic structure within and among elephant populations in northern Tanzania and

24 compare it with populations in Ruaha and Selous ecosystems to better understand past links and discontinuities and to inform future corridor conservation efforts.

Materials and methods

Description of study areas

To conduct a nationwide assessment of the genetic structure, we selected four major ecosystems that have the largest elephant populations in the country. These are Serengeti and Tarangire-Manyara in northern Tanzania and Ruaha and Selous ecosystems in central and southern Tanzania (Figure 2-1).

Figure 2-1. Map of Tanzania showing four ecosystems where we collected elephant fecal samples.

The Great Rift Valley separates the Serengeti ecosystem from the Tarangire-Manyara ecosystem in northern Tanzania, and the Eastern Arc Mountains separates Selous from Ruaha ecosystem. (NP=National Park, GR=Game Reserve).

25 The Serengeti ecosystem

The Serengeti ecosystem (SE) is in the north-east of Tanzania between 34°450– 35°500 E and 2°–3°200 S that covers several different conservation administrations (Ernest et al., 2012). The SE comprises of several protected areas including: Serengeti National Park (SNP), Ikorongo-Grumeti Game Reserves (IGGR), Maswa Game Reserve (MGR), Ngorongoro Conservation Area (NCA), Loliondo Game Controlled Area (LGCA) and Mwiba Wildlife Ranch (MWR) (Figure 2-2). The LGCA allows human activities such as pastoralism, farming, and settlement. The MWR is a privately-owned wildlife ranch which is bordered by NCA to the east, MGR to the north, and with SNP southern boundary being approximately 7 km north of the northern edge. The NCA is a different category of the protected area as it allows multiple land use with semi-nomadic Maasai pastoralists practicing traditional livestock grazing coexisting with wildlife (UNESCO, 2015). In the NCA, there is a Crater which supports a diverse number of wildlife species but is also a hotspot for safaris. No human settlement is allowed within the Crater and the Forest around the Crater rim. The Serengeti ecosystem has experienced several poaching episodes since the 1880s. The number of elephants in 1860 was recorded above 4,000 but in the year 1900, their number dropped to zero (Sinclair et al., 2008). The elephant numbers remained low until the 1950s when the Serengeti was established as a protected area with only 100 elephants (Sinclair, 2012). However, heavy poaching in the 1970s and 1980 and lost about 80% of its population (Sinclair et al., 2008). Yet, after the ban on international trade, the elephant population has recovered to about 6, 087 (TAWIRI, 2015).

26

Figure 2-2. Map of protected areas in northern Tanzania showing the Serengeti and Tarangire-Manyara ecosystems.

GCA=Game Controlled Area, GR=Game Reserve, NP=National Park, WR=Wildlife Ranch, CA=Conservation Area, MNP, Mountain National Park. The Great Rift Valley separates the Serengeti ecosystem from the Tarangire-Manyara ecosystem in northern Tanzania.

Tarangire-Manyara ecosystem

The Tarangire-Manyara ecosystem (TME) comprises three protected areas: Tarangire National Park (TNP), Lake Manyara National Park (LMNP) and Manyara Ranch (MANR) which acts a migratory corridor between the two parks (Figure 2-2). TNP and LMNP manage wildlife for tourism but, Manyara Ranch (MANR) is a private land

27 conservancy managed for livestock grazing and wildlife tourism. Most of the land in the TME (~85%) fall under community land managed as open areas, Game Controlled areas or wildlife management areas (Morrison et al., 2016). A rift valley wall (500-1000m) at the western edge of the TME may prevent the free movement of wildlife to and from SE. The TME elephants were also affected by poaching before a ban on international trade in 1989. Since then, the population recovered with the current population due to low poaching pressure and conducive climatic conditions (Foley and Faust, 2010). The current number of elephants in the TME is about 4,202 (TAWIRI, 2015) with an annual growth rate of 7% (Foley and Faust, 2010). Wildlife corridors between the TME and SE are under critical risk of being blocked completely (TAWIRI, 2010). The corridors between LMNP and NCA are essential for connecting the two ecosystems. Effects of habitat fragmentation on wildlife corridors between NCA and LMNP can be traced back in the 1940s when tsetse eradication allowed the expansion of human settlement in the Mbulu areas cutting the forest corridors (Homewood and Rodgers 1991). In the mid-1970s, reports showed that only a fragmented strip of forest remained from Kitete and Lositete down the Gregor rift wall to Mto wa mbu (Makacha and Fame, 1979). Also, the human settlement around Mangola area has affected the southern linkage to Lake Eyasi and Yaida Chini (Homewood and Rodgers, 1991). Increased human activities, the especially agricultural settlement around LMNP (Borner, 1985; Douglas-Hamilton, 1973; Mwalyosi, 1977) threatens the viability of wildlife corridors and the expansion of Mto wa Mbu agricultural settlement is of particular concern (Mwalyosi, 1991). The wildebeest in the TME is distinct from the Serengeti without mixing for thousands of years. Wildebeest have been migrating between TME and Lake Natron (Morrison et al., 2016; Morrison & Bolger, 2014).

28 Ruaha ecosystem

Ruaha National Park (RNP) is the largest national park in Tanzania (Figure 2-3), and it covers 20,226 km2 (TANAPA). The Ruaha NP experiences dry and rainy season with an annual rainfall of 516.7mm (Jones et al., 2018). The rainy season typically runs between December and May. The Greater Ruaha river and Ihefu wetland are the primary source of water in the ecosystem. Ruaha has diverse vegetation types from Miombo woodland that extends from Southern Africa to acacia woodlands (TANAPA website). The current number of elephants in the ecosystem is 8,272 (TAWIRI, 2015).

Figure 2-3. Map of Selous and Ruaha ecosystems in Tanzania showing sampling locations

GCA=Game Controlled Area, GR=Game Reserve, NP=National Park, MNP=Mountain National Park. The Eastern Arc Mountains separates the Selous from the Ruaha ecosystem. Between the two ecosystems are other protected areas that are important areas for facilitating genetic connectivity between the Ruaha and the Selous ecosystems.

29 Selous Ecosystem

Selous Game Reserve (SGR) is the largest game reserve in Africa located in the southeast of Tanzania at lat. 7°35′S, long. 38°15′E (Spong 2002) (Figure 2-3). Due to its size, Selous GR has eight administrative sectors. In this study, we focused on two sectors: Matambwe (MAT) and Kingupira (KPR) sectors. Matambwe is the only sector within the Selous GR that conducts photographic tourism. The rest of the SGR is designated for trophy hunting blocks. Rufiji River separates Matambwe from Kingupira sector. The current number of elephants in the Selous ecosystem is 15, 217 (TAWIRI, 2015). Between Ruaha and Selous ecosystems there is Udzungwa Mountain National Park (Udzungwas), part of the Eastern Arc Mountains with a high level of species endemism and richness (Burgess et al., 2007). Between Udzungwas and Selous GR is the Kilombero Valley (6,650 km2) which is an important bird area and important farming areas especially for rice and sugar cane (Hinde et al., 2001) (Figure 2-3).

Field data sample collection

We used DNA extracted from elephant fecal samples to genotype 11 SSR and mitochondrial DNA sequences of a hypervariable region in cytochrome b in the control region of D-loop. We collected about 5g of the outer portion of fresh dung and placed into a 50ml conical flask labeled with three initial letters of the protected areas followed by the number of sample in that location (for examples, samples from Ngorongoro Conservation Area (NCA) was labeled as NCA01). To ensure a higher concentration of DNA, we only collected samples that were deposited within the previous 12 hours. Scrapings were taken from the outer layer of the bolus, where most of the epithelial cells are found, following methods established by Ahlering et al., (2012). Immediately after collection of samples, we added Queens College buffer (20% DMSO, 0.25 M EDTA, 100m tris, PH 7.5 saturated with NaCl). We used new pairs of gloves for each sample to avoid cross-contamination between samples. Where possible, we collected 50 samples from each site, but we obtained

30 a minimum of 25 samples in small isolated populations and recorded the GPS of each. For each sampling location, we collected samples within a short period to minimize the chances of sampling closely related individuals and resampling the same individuals. For occasions where we observed elephants, we estimated their age and determined their sex before or after collecting their samples. We took photographs of each to crosscheck the accuracy of their use and their estimated age. We shipped the collected samples to the Pennsylvania State University for genetic analyses. Before shipping, we sealed well each tube containing sample using parafilm and placed it into a zip locked bag to avoid cross-contamination or leakage. We obtained four tissue samples from the Tanzania Wildlife Research Institute (TAWIRI) for elephants that died naturally to use them as a positive control for allele scores because their DNA quality is higher than that of fecal samples. We obtained all required permits from the United Republic of Tanzania, and the US Fish and Wildlife Service.

DNA isolation

We isolated DNA using company protocols with DNA extraction kits (QIAamp DNA Stool Mini Kit) with minor modifications (Eggert et al.,2005). We vortex fecal samples for 1min and poured directly into the 5ml tube. We increased the initial volume of samples from 200 µl to 1000 µl and added 1400 µl of buffer ASL. We added 25 µl of Proteinase K and vortexed the mixture thoroughly for 1min. We set an incubator shaker at 55°C and the samples were placed horizontally on a rack and secured by tying with a plastic wrap. We incubated samples with shaking for 12 hours. Other steps were the same as the QIAamp stool mini kit. To obtain a high concentration of DNA, we used a final elution of 100 µl. The DNA concentration was measured using a nanodrop and recorded the ratio of

A260/A280. DNA of a purity acceptable for downstream PCR amplification has an A260/A280 ratio of 1.7–1.9. Samples with the ratios out of this range were re-extracted.

31 Microsatellite Analysis

PCR amplification and genotyping

We performed PCR amplification of SSRs from fecal samples using QIAGEN reagents (QIAGEN multiplex PCR kit). We used 11 loci that had previously been found to be polymorphic in African elephants: LAT06, LAT08, LAT13, LAT24 (Archie et al, 2003), FH19, FH48, FH60, FH67 (Comstock et al, 2000), LA5, LA6 (Eggert, et al. 2000), and LafMS02 (Nyakaana et al, 2005) with some minor modifications (Kinuthia et al., 2015). The details of these markers are shown in (Table S1). We labeled all forward primers with one of the fluorescent dyes (FAM, NED or VIC). We used standard QIAGEN multiplex PCR protocol. We used 12.5 µl PCR reaction volume containing 6.25 µl of the master mix, 0.5 µl of primer mix (standardized to 2 µM), three µL of template DNA and 2.75 µl of millipure water. We performed PCR with the initial polymerase activation step at 95°C for 15 min, denaturation at 94°C for 30 seconds, annealing at 55°C or 58°C for 1:30 min, extension at 72°C for 1 min for 35 cycles. We did an initial evaluation of SSRs primers by agarose gel electrophoresis. We loaded onto 2% gel aliquots of 5 µl of each PCR product with electrophoresis run time of 45 minutes at 120V. We estimated fragment lengths for each sample (allele) from digital gel images. For genotyping, we grouped 11 SSRs into four panels for genotyping. All multiplex PCR samples were genotyped and separated by electrophoresis using ABI 3730xl DNA analyzer (Applied Biosystems) at Genomics Core Facility of the Pennsylvania State University. We analyzed the GeneScan results using GeneMapper® v. 5 (Applied Biosystems). We used elephant genotype data from tissue samples to set bins for each marker. Alleles were detected automatically by the Genemapper software but we manually verified before exporting the file. We genotyped twice inconsistent allele scores, and we considered all weaker peaks with less than 100 fluorescent units as failed, and we could not reliably score them (Okello, 2005)

32 Genetic diversity and differentiation

To account for allelic dropout, null alleles, and scoring error due to stuttering we used MICRO-CHECKER 2.2.3 (VAN OOSTERHOUT et al., 2004). We then corrected genotypes based on MICRO-CHECKER 2.2.3 results. We determined the number of alleles and their frequencies for all loci across all individuals in all the populations using GenAIEx 6.502 (Peakall and Smouse, 2012). We also used GenAIEX to export files into formats compatible with other genetic analyses software. We tested for deviations from expectations under Hardy-Weinberg equilibrium (HWE) and for linkage disequilibrium (LD) within protected areas in GENEPOP (Raymond & Rousset, 1995; Rousset, 2008), accessible online at http://gene-pop.curtin.edu.au/. For the two analyses, we used the Markov chain strategy with 1000 dememorizations, 1000 for combining independent test results across study location and we used the number of loci to determine the statistical significance test results. We applied a Bonferroni correction for multiple comparisons using a Holm-Bonferroni sequential correction for both HWE and LD tests (Hochberg, 1988; Holm, 1979). We used the IBD program (Bohonak, 2002) to test whether there is evidence for isolation-by-distance (IBD). We also plotted genetic distance (FST /1- FST) against geographic distance (km) to determine if the two variables are significantly correlated. We used FSTAT (Goudet, 1995) to calculate the inbreeding coefficient (Fis) and allelic richness (AR) which represents the number of alleles standardized to the smallest sample size in the study area. We used both ARLEQUIN (Excoffier et al., 2005) and

FSTAT (Goudet, 1995) to assess genetic diversity by estimating the expected (HE) and observed (HO) heterozygosities and the level of genetic differentiation(FST) between the sampling locations. We used POPTREE2 (Takezaki, Nei, & Tamura, 2010) to construct population trees to establish the genetic relationship between sites.

33 Population Structure

We used STRUCTURE 2.3 (Falush, Stephens, & Pritchard, 2003, 2007; Pritchard, Stephens, & Donnelly, 2000) to determine elephant population structure in our samples. This program uses a Bayesian clustering model to assign individuals to a population while simultaneously estimating population allele frequencies. This model aims to determine the number of subpopulatons K within the population, where K in most cases is unknown (Pritchard et al., 2000). We inferred K by running ten iterations for each K value from 1 to 10 using an admixture model with a LOCPRIOR option. We set a burn-in period at 1x106 and Markov Chain Monte Carlo (MCMC) repetition value of 1x106. We used STRUCTURE harvester to detect the best value of K and visualize the STRUCTURE results using Pophelper (Francis, 2017). From this point, we will use the term subpopulation to define groups of individuals identified through the STRUCTURE.

Mitochondrial Sequencing and Analysis

For mitochondrial (mt)DNA we used all 715 samples to capture all existing haplotypes in the sampling locations. We sequenced 630 base pair (bp) of cytochrome b gene using forward primers MDL5 5′- TTACATGAATTGGCAGCCAACCAG- 3′ and reverse primers MDL3 5′- CCCACAATTAATGGGCCCGGAGCG- 3' (Fernando & Lande, 2000). For samples that did not amplify successfully, we used a different reverse primer mtCR3 5′- GTC ATT AAT CCA TCG AGA TGT CTT ATT TAA GAG G- 3'. We performed PCR reactions with the initial polymerase activation step at 95°C for 3 min, denaturation at 95°C for 30 sec, annealing temperature at 60°C for 45 seconds, extension at 72°C for 30 sec for 35 cycles. Each PCR mixture contained 3μl of 5X Green GoTaq reaction buffer (Promega), a final concentration of 0.33μM for each of the primers, 0.13μM of dNTP (Quanta bio), Bovine serum albumin (BSA) 0.1μg/μl, and 3μl of DNA template of unknown concentration in a 15μl volume. For each PCR reaction with test samples, we also ran a negative control (no DNA) and a positive control using DNA from known tissue

34 samples. 6μl of the PCR product was used for electrophoresis in Tris-Acetate EDTA running buffer at 120V for 45 min on a 2% agarose gel stained with GelRed (Biotium). We sequenced PCR products using reverse primer only unless the sample did not have clean sequence results or had a unique haplotype. Sequence results in the trace file format were visually inspected using SnapGene® software 4.2.4 (from GSL Biotech; available at snapgene.com) by comparing to a reference sequence for the hypervariable control region of elephants (Hauf, 1999). We aligned all sequences using CLUSTALX2 (Larkin et al., 2007). We compared our sequences with other published studies to identify unique haplotypes (Ahlering et al., 2012; Debruyne, 2005; Eggert et al., 2008; Ishida et al., 2013). We constructed phylogenetic relationships among unique mtDNA haplotypes neighbor- joining implemented in MEGA (Kumar et al., 2008). Support for the nodes for each analysis was assessed using 1000 bootstrap iterations. We calculated haplotype diversity (h), nucleotide diversity (π), and Tajima D using Arlequin version 3.5 (Excoffier et al., 2005; Excoffier & Lischer, 2010) and constructed median-joining (MJ) network PopArt 4.8.4 (Leigh & Bryant, 2015).

Results

Microsatellite analyses

We successfully genotyped 711 (88.87%) of the 800 fecal samples collected between 2015 and 2017. We removed 23 samples with the same multi-locus genotypes identified using COLONY(Jones & Wang, 2010), and we retained with 688 samples. One locus, FH60 consistently showed evidence for the presence of a null allele for all sampling locations and deviated from HWE after applying a Bonferroni correction for multiple comparisons. We decided to remove FH60 locus from subsequent analyses. Genetic diversity indices are shown for seven zones and specific sampling locations within those zones (Table 2-1).

35 Table 2-1 Genotyped African Savanna elephants obtained from 15 locations in Tanzania

Ne Protected Area Zone N Na AR Ho uHe F FIS HWE LD Northern Serengeti 309 (223- NSE 212 11.00 9.78 0.72 0.73 -0.01 0.01 0.60 0.07 Ecosystem 464) North Serengeti NP (NSNP) NSE 103 9.40 6.35 0.71 0.73 0.01 0.03 0.60 0.00 Loliondo GCA (LGCA) NSE 35 6.90 5.89 0.70 0.71 -0.02 0.02 0.70 0.13 West Serengeti NP NSE 46 8.30 6.48 0.75 0.73 -0.06 -0.04 0.70 0.18 (WSNP) Grumeti-Ikorongo GR NSE 28 6.60 5.86 0.73 0.70 -0.08 -0.04 0.80 0.00 (GIGR) Southern Serengeti 310 (312- SSE 111 10.40 10.35 0.73 0.73 -0.03 -0.01 0.70 0.07 Ecosystem 524) Maswa GR(MGR) SSE 46 7.80 6.28 0.76 0.72 -0.09 -0.07 0.80 0.00 Mwiba Wildlife Ranch SSE 49 8.70 6.48 0.71 0.71 -0.02 0.01 0.80 0.02 (MWR) South Serengeti NP (SSNP) SSE 17 5.90 5.82 0.72 0.72 -0.04 -0.01 1.00 0.00 Ngorongoro CA (NCA) NCA 104 10.20 10.23 0.70 0.72 0.00 0.02 0.40 0.20 95 (75-122) Crater & Endulen (CRE) NCA 80 9.50 6.70 0.71 0.73 0.00 0.02 0.60 0.16 23.9 Ndutu (NDT) NCA 5.70 5.21 0.67 0.66 -0.06 -0.02 0.80 0.18 0 98.2 114 (88-150) Manyara (MAR) MAR 10.20 10.47 0.71 0.74 0.03 0.05 0.60 0.29 0 Lake Manyara NP 35.7 MAR 7.00 5.83 0.68 0.70 0.00 0.03 0.60 0.07 (LMNP) 0 62.5 Manyara Ranch (MANR) MAR 9.40 6.70 0.72 0.75 0.02 0.04 0.60 0.07 0 63.6 111 (79-170) Tarangire NP (TNP) TNP 8.50 9.22 0.71 0.73 0.01 0.03 0.70 0.02 0 47.8 114 (83-169) Ruaha NP (RNP) RNP 8.50 9.29 0.70 0.74 0.03 0.05 0.70 0.02 0 49.4 95 (64-161) Selous GR (SGR) SGR 9.20 10.41 0.68 0.74 0.08 0.09 0.40 0.11 0 21.5 Matambwe (MAT) SGR 7.40 6.92 0.66 0.73 0.07 0.10 0.80 0.04 0 27.9 Kingupira (KPR) SGR 7.30 6.49 0.69 0.74 0.05 0.07 0.40 0.07 0

Number of samples (N), number of different alleles (Na), allelic richness (AR), observed heterozygosity (Ho), unbiased heterozygosity (uHe), F=Fixation index, Fis=inbreeding coefficient, proportion of loci conforming to Hardy-Weinberg equilibrium (HWE) and proportion of locus pairs in significant (p<0.001) linkage disequilibrium (LD and mixed cohort effective population size (Ne) estimates are provided (CI=confidence interval).

36 In general, we observed high allelic richness and heterozygosity across sampling locations. Hierarchical STRUCTURE analysis indicated that the uppermost level of population structure was represented by two clusters (K=2; Figure 2-4). The NSE and SSE formed one cluster (blue), MAR, TNP, RNP and SGR (orange) formed another cluster. NCA elephants showed admixture between Tarangire and Serengeti ecosystems. Within the Serengeti ecosystem, two groups were identified (Figure 2-4 B, D & E).

Figure 2-4. Hierarchical population STRUCTURE analysis for the African Savanna elephants in Tanzania using 10 microsatellite loci.

Each individual is represented by a thin vertical bar partitioned into color segments representing individual’s ancestry into subpopulations. The optimum number of clusters (K) was obtained using the Evanno method (Evanno et al. 2005). Estimated membership coefficient plots (A-I) indicate the pattern of individual elephant cluster assignment for each of the hierarchical analysis. Samples are sorted from north to south as defined in Table 2-1.

37 Unrooted neighbor-joining tree for 688 African savanna elephants using 10 SSRs genotype data indicated three main clusters (Figure 2-5). The first cluster had individuals from the Serengeti ecosystem (blue), and the second cluster had individuals from the Tarangire-Manyara ecosystem and Ruaha NP, and the third cluster with individuals from Selous (Figure 2-5).

Figure 2-5. Unrooted neighbor-joining tree for 688 African savanna elephants using 10 SSR loci. The numbers on the branches show bootstrap values (%). Serengeti ecosystem (blue), Selous (purple), Tarangire-Manyara (orange) and Ruaha (black). W=West, N=North, C-Central Serengeti.

We further performed principal coordinate analysis (PCoA) to determine the number of clusters using FST values and genetic distance matrix between all pairs of individuals. We identified at least two clusters in northern Tanzania as shown in (Figure 2-

38 6 & 2-7). Samples ordinated close to one another are more closely related than those ordinated far away.

Figure 2-6. Principal coordinate analysis based on genetic distance obtained from 10 SSR loci for African Savanna elephants in Northern Tanzania. Using PCoA, we identified at least two clusters.

TME=Tarangire Manyara ecosystem, SE=Serengeti ecosystem and NCA=Ngorongoro conservation Areas.

39

Figure 2-7. Principal coordinate analysis for 10 SSRs using pairwise FST values in northern Tanzania.

All elephants from the Serengeti ecosystem were clustered together, consistent with the STRUCTURE results. Interestingly, the NCA elephants were grouped in the middle between the Tarangire-Manyara and Serengeti ecosystems. The long form of the locations are shown on Table 2-1.

To test for the presence of spatial structure for the genetic and geographic datasets, we conducted spatial autocorrelation (r) implemented into GenAIEX (Peakall & Smouse, 2012). We found evidence for the presence of spatial structure; the observed values of r were outside the upper and lower bound at a 95% confidence interval. Here, we tested a null hypothesis of no spatial structure for the genetic and geographic data sets. Elephants that are geographically close to each other show higher genetic similarity than distant elephants. Elephants within 120 km show genetic similarity with positive spatial autocorrelation (Figure 2-8). Beyond 120 km, elephants were less genetically similar (with negative r). We also found a positive correlation between genetic distance and geographic distance, r=0.3719, p=0.0008 (Figure 2-9). Elephants that are closer to each other show higher genetic similarity than those that are far away, as expected from random mating assumption.

40

Figure 2-8. A correlogram showing spatial genetic autocorrelation (r) among elephants in northern Tanzania as a function of Euclidean distance.

We defined distance classes every 30 km. Dotted lines indicate the 95% CI about the null hypothesis of no genetic structure. The error bars about r represent the 95% CI, as determined by bootstrapping (999 iterations)

Figure 2-9. Relationship between genetic distance and geographic distance for African elephants in northern Tanzania. We have included the pairwise comparisons for each set of locations. For example, TNP& MAR are only 60 km apart but they have unexpectedly high FST value of 0.034.

41 Mitochondrial DNA analysis

We analyzed a total of 558 mtDNA sequences of 622-bp. We identified a total of 33 haplotypes of which seven were published previously (Ahlering et al., 2012; Debruyne, 2005; Ishida et al., 2013) and 26 were unique to this study (Table 2-2)

42 Table 2-2. Distribution of 33 mtDNA for African elephants in northern and southern Tanzania.

Haplotype names have two parts: subclade names (EC=East Central, SW=Savanna wide and SS=Southeast Savanna) and location names SE=Serengeti, GR=Grumeti, LM=Lake Manyara, TA=Tarangire, MR=Manyara Ranch, NG=Ngorongoro, SG=Selous GR, RU=Ruaha, MG=Maswa GR. Haplotypes identified in by the previous studies are indicated with asterisks, the rest are new haplotypes unique to this study. Abbreviations for each of the sampling locations are indicated in Table 2-1.

Haplotype IGGR LMNP LGCA MANR MGR MWR NCA RNP SGR SNP TNP Total ECSE09* 25 4 24 6 11 13 29 135 3 250 ECSE11* 2 6 2 4 14 ECGR02 3 3 ECGR01 1 1 ECSE10 2 3 5 ECLM06 1 1 ECNG03* 9 1 1 2 2 15 ECSE08 1 1 ECLM03 4 4 ECSE07 1 1 2 ECMG01 1 1 SWSE03 1 2 3 SWTA01* 22 11 6 5 26 21 13 31 135 SWMR02 13 13 SWMR03 2 1 1 2 6 SWLM02 2 2 SWSE04* 1 4 3 7 5 20 SWSE02* 5 5 10 SWMR01 1 4 5 SSRU01 3 3 SSNG01* 6 3 2 6 4 4 25 SSSE01 1 1 1 3 6 SSNG02 1 4 1 6 SSSG01 5 5 SSSG02 5 5 SSSG03 4 5 SSSG04 1 1 SSSG05 4 4 SSSG06 1 1 SSSG07 1 1 SSSG08 1 1 SSSG09 1 1 SSSG10 2 2 Total 32 48 33 37 27 22 82 25 30 182 38 558

43 Haplotype diversity (h) ranged from 0.29 in RNP to 0.91 in SGR (Table 2-3). The nucleotide diversity ranged from 0.004 in RNP to 0.252 in NCA (Table 2-3). Overall nucleotide diversity was 2.63% which was similar to the wide continental level of 2.0% (Nyakaana et al., 2002).

Table 2-3. Genetic diversity of African savanna elephants based on the sequence of 622 bp of mtDNA

The number of samples (n), number of haplotypes (H), haplotype diversity (h), nucleotide diversity (π), and Tajima’s D with its corresponding P value for each region.

Location N H h π Tajima’s D P NSE 220 10 0.3549 0.145 -0.247 0.472 SSE 71 12 0.6833 0.242 4.596 1.000 NCA 82 9 0.762 0.252 5.134 1.000 MAR 84 10 0.790 0.208 3.6172 0.999 TNP 42 3 0.442 0.141 0.838 0.848 RNP 25 3 0.290 0.004 -0.498 0.355 SGR 30 12 0.910 0.1839 2.660 0.99

All Fst values, except between RNP and TNP for the mtDNA, showed significant genetic differentiation (Table 2-4). The FST values from the mtDNA were higher than the nuclear

DNA due to the differences in modes of inheritance. We also included normalized FST values by dividing the pairwise FST to the geographic distance between subpopulations.

We identified unexpected high normalized FST between MAR and TNP although they were less than 60 km apart (Table 2-4).

44

Table 2-4. Pairwise Genetic differentiation (FST) for the African savanna elephants in Tanzania

The FST based on SSRs are below the diagonal and mtDNA are above the diagonal. Significant levels of p ≤ 0.05 are indicated with an asterisk. We normalized the Fst values (in parentheses x104) to geographic distance (km) to show unexpected high population differentiation over the short distance (see also Figure 2-S2) Locations NSE SSE NCA MAR TNP RNP SGR NSE - *0.111 *0.229 *0.388 *0.588 *0.569 *0.353 (7.76) (14.77) (17.32) (21.54) (8.79) (4.84) SSE *0.007 - *0.019 *0.126 *0.288 *0.362 *0.207 (0.51) (2.88) (8.29) (15.74) (7.24) (3.35) NCA *0.009 *0.009 - *0.056 *0.165 *0.235 *0.161 (0.61) (1.31) (7.37) (12.22) (4.94) (3.18) MAR *0.021 *0.020 *0.013 - *0.11 *0.153 *0.143 (0.96) (1.34) (1.75) (18.33) (3.30) (2.79) TNP *0.040 *0.046 *0.028 *0.034 - 0.031 *0.339 (1.45) (2.53) (2.05) (5.70) (0.72) (7.55) RNP *0.021 *0.021 *0.016 *0.019 *0.028 - *0.384 (0.32) (0.41) (0.34) (0.40) (0.66) (11.53) SGR *0.027 *0.029 *0.026 *0.032 *0.041 *0.029 - (0.37) (0.46) (0.52) (0.62) (0.92) (0.87)

All haplotypes fall into two clades Forest (F), and Savanna (S) subdivided into three subclades Savanna-wide (SW), East-central (EC) and Southeast-savanna (SS) (Ishida et al., 2013). We identified eight haplotypes in Savanna wide subclade, 11 in East-central and 14 in the Southeast savanna. East-central haplotypes were most frequent in the SE, while the Savanna wide mainly was prevalent in the TME and RNP. Southeast savanna was dominant in the SGR (Figure 2-10).

45

46

Figure 2-10. Map showing genetic relationship for African savanna elephants in Tanzania.

A). Membership coefficient (%) obtained from STRUCTURE program showing the proportion of individuals in Cluster 1 (Blue) and Cluster 2 (Orange). B). The distribution of haplotypes grouped in three subclades (EC=East central (blue), SW=Savanna wide (orange) and SS=Southeast savanna (purple)) as described in Table 2-2.

47 A median-joining network for the mtDNA haplotypes shows 22 mutations between the F- clade and the S-clade and seven variations between Savanna-wide and the Southeast savanna which are both in S-clade (Figure 2-8). The rift valley wall between SE and TME may explain the haplotypes differences as observed in Figure 2-10. Savanna wide subclade was less common in the SE while it was the most common in the TME but we also observed remarkable differences in the haplotype distribution between MAR and TNP which are less than 60 km apart.

Figure 2-11. Median-joining haplotype network based on mitochondrial DNA sequences from African savanna elephants in Tanzania. The size of circles is proportional to haplotype frequencies.

The numbers 1-33 represent haplotypes identified in this study. The stepwise mutations between haplotypes are indicated by hatch marks. SSE=South Serengeti, MAR=Manyara Ranch and Lake Manyara, NCA=Ngorongoro Conservation Area, RNP=Ruaha National Park, SGR=Selous Game Reserve, NSE=North Serengeti and TNP=Tarangire National Park

Similarly, the East-central haplotypes were common on the SE which in the western side of the rift valley. 100% of Selous samples carried haplotypes with the southeast savanna (Figure 2-10). Our data suggest a clear pattern of haplotype distribution. The majority of

48 elephants in the SE carry haplotypes in the F-clades which is shared with the forest elephants (Ishida et al., 2013), and was not observed in Ruaha and Selous. Finally, we constructed a neighbor-joining tree from the mtDNA haplotypes identified from this study combined with known haplotypes previously published (Eggert et al., 2005, Debruyne, 2005; Ishida et al., 2013; Ahlering et al., 2012). Our phylogeny supports subdivision of the 33 haplotypes into three subclades (Figure 2-12).

49

Figure 2-12. A neighbor-joining tree constructed from 33 mitochondrial DNA sequences from this study and reference sequences from previous studies. Blue= East-central, Orange=Savanna wide, purple=southeast savanna.

Note: we included haplotypes from Garamba Reserve, for example, GR0013 (Ishida et al., 2013) from Democratic Republic of Congo and they were clustered with haplotypes from northern Tanzania. Numbers on the branches represent the percentage bootstrap values for 1000 replicates. We included forest elephant (2) and Asia (1) as outgroups Forest elephants were clustered together with the east central savanna which was common in Serengeti

50 Discussion

Herein, we provide evidence that the current population structure of elephants is the result of genetic differentiation between subpopulations in northern Tanzania. Both nuclear and mtDNA results revealed that the Tarangire-Manyara Ecosystem (TME) and the Serengeti Ecosystem (SE) are genetically distinct with low gene flow between them.

All pairwise FST values for SSRs and all but one mtDNA showed a significant genetic differentiation (p≤0.05). mtDNA haplotype distribution was significantly different between SE and TME. The Great Rift valley wall acting as a geographic barrier between TME and SE accounts for at least some of the distributions of mtDNA haplotypes and SSRs genotypes. However, habitat fragmentation and habitat loss between these two ecosystems are likely to have accelerated this genetic differentiation. For example, within the TME we identified two subpopulations suggesting limited gene flow between Tarangire and Manyara. Surprisingly, the Tarangire elephants showed higher genetic similarity with Ruaha elephants than the Serengeti elephants. This indicates that elephants frequently used the corridor between Tarangire and Ruaha before it was closed entirely in the past 30 years (Riggio et al., 2017; Caro et al., 2009). Our microsatellite analysis provides some insights into the presence of genetic differentiation within the SE. Elephants from northern Serengeti (NSE) formed one cluster whereas elephants from southern Serengeti (SSE) formed another cluster (Figure 2-4D).

Pairwise FST values between the SSE and NSE was significant (Table 2-4), suggesting genetic differentiation within the Serengeti ecosystem probably because there were at least two different sources of elephant populations following poaching induced extirpation (Sinclair, 2012). Analysis of the hypervariable control region in D-loop of the cytochrome b gene of mtDNA has frequently been used to assess nucleotide and haplotype diversity among mammals. The mtDNA is inherited in a haploid fashion because it is passed through the maternal lineage. The east-central subclade is in F clade which is also found among forest elephants. Thus, it is possible that some forest elephants moved to the SE before forest elephants diverged from the savanna elephants. The SE has two distinct populations that

51 arrived from outside the SE in the early 1960s as part of the population expansion that began in the 1950s after 100 years of near extermination due to the ivory trade 1840-1890 (Sinclair et al., 2008). Our results supports this the hypothesis because in the SE we observed two main subclades, Savanna-Wide and East-Central. The savanna wide subclade dominates the eastern rift valley, whereas the east-central subclade is predominantly found on the western side of the rift. because mtDNA is maternally inherited, the observed pattern reflects females’ complex social structure. Interestingly, Ruaha elephants which are 429 km away from Tarangire, share the same haplotypes with Tarangire elephants, and they are distinct from Selous GR (Figure 2-10), suggesting historical female-mediated gene flow between Ruaha and Tarangire. Unfortunately, wildlife corridors between Ruaha and Tarangire are now entirely blocked (Riggio et al., 2017). Nuclear loci also showed a high genetic similarity between Ruaha and Tarangire (Figure 2-4A & 2-5). In northern Tanzania, the rift valley separated elephants into two main subclades, i.e., east and west of rift valley. This pattern was observed by a previous study (Ahlering et al., 2012). Our study found similar patterns of haplotype and genotype frequencies between the SE and TME (Figure 2-11). Our results provide evidence for the gene flow between the two ecosystems, albeit at a reduced rate, which is likely due to recent habitat fragmentation. We also expected more recent divergence between Ruaha and Tarangire.

52

Figure 2-13. Wildlife corridors between Tarangire and Manyara, and between Lake Manyara and Ngorongoro

This information is based on the GPS collar data from elephants (Kikoti & Griffin, 2009), and wildebeest (Morrison et al., 2016; Morrison & Bolger, 2014). WMA=Wildlife Management Area. The Lake Manyara NP seem to be isolated from NCA and Tarangire

Connectivity between Tarangire and Manyara

Within the TME we detected significant genetic differentiation; MAR (Lake Manyara and Manyara Ranch) formed one cluster, and the Tarangire elephants formed a second cluster (Figure 2-3 F). There was significant estimates of genetic differentiation between MAR and TNP using both SSRs (FST, 0.034) and mtDNA (FST, 0.11) (Table 2-4). Haplotype diversity (H) was 0.79 in MAR and 0.44 in TNP (Table 2-4). Tarangire had only three haplotypes while MAR had ten haplotypes (Table 2-3). In Tarangire, one haplotype was carried by 31 individuals (81. 57%) of all samples and only two haplotypes were shared between TNP MANR and LMNP (Table 2-2).

53 Furthermore, when we compared haplotype distributions between the TME and SE, we found that most MAR elephants shared haplotypes with NCA elephants (Figure 2-14). We detected a significant difference in haplotype distribution. For example, in Tarangire, only four elephants were carrying a haplotype in the East Central subclade, but these haplotypes were common in MAR, subclade represented in blue color, (Figure 2-11). In TNP, there were no elephants carrying haplotypes in southeast savanna subclade but these haplotypes were observed in MAR (Figure 2-11). It appears there is no or limited female- mediated gene flow between MAR and TNP although there is no landscape barrier between them. Alternatively, there could be cultural or behaviorally mediated barriers to genetic connectivity among subpopulations. Females’ social behavior of remaining in their natal groups might drive discontinuities among subpopulations. The only viable migratory route between Tarangire and Manyara Ranch based on GPS data is the Kwa kuchinja corridor (Figure 2-13). Kwa kuchinja corridor has experienced a substantial increase in cultivation over the past decade and remains under heavy development pressure from agriculture and settlements (Morrison et al., 2016) thereby reducing connectivity between the Tarangire and Manyara Ranch populations.

Ngorongoro-Manyara Corridor

The NCA elephants show admixture because they are geographically very close to LMNP, which indicates that there was gene flow between them in the past, however, the corridor between these protected areas is probably now closed completely (Figure 2-13). The STRUCTURE results suggest a high genetic similarity between MAR and NCA

(Figure 2-4C). Also, the FST values between MAR and NCA was low for both nuclear loci

(FST=0.01) and mtDNA (FST=0.056) (Table 2-4). Haplotype distribution between the NCA and MAR is similar (Figure 2-11). Our study provides some insights into historical and contemporary gene flow between the two protected areas. Although the rift valley may impede the movement of elephants, there is enough evidence that there was gene flow between the two protected

54 areas. Historically it appears that there was little mixing between TNP and MAR but was with the NCA. The NCA elephants showed high genetic similarity with the MAR elephants. Now that the corridor between the NCA and Manyara are blocked, Lake Manyara is likely isolated. The Lake Manyara population is small, it is at a risk of extinction through inbreeding depression and population decline through demographic stochasticity (Gilpin & Soulé 1986). There is some speculation that Manyara elephants have smaller body size that the Tarangire and Ngorongoro elephants (field observations). At some point, the Manyara National Park had the highest known elephant density in Africa (Douglas‐Hamilton, 1973). Elephants declined from 500 in 1984 to about 150 in 1988 and then to 36 in 2007 and 34 in 2014 ( (Blanc et al., 2007; Kioko, Zink, Sawdy, & Kiffner, 2013; Prins, Jeugd, & Beekman, 1994; TAWIRI, 2015). Douglas-Hamilton (1973) reported that the Manyara elephant population was young and fertile and had expanding at the rate of 3-4% annually. However, the population size of elephants has remained around the 30 since 2007. Probably inbreeding depression as a result of genetic isolation has affected the population. Most African elephant populations were affected by poaching in 1980s but some protected areas such as Tarangire have recovered from poaching. Here, we argue that genetic factors may be affecting the Lake Manyara elephants.

55

Figure 2-14. Mitochondrial DNA haplotype distribution (subclades) between Ngorongoro and Lake Manyara. Within the NCA we did not observe elephants carrying southeast savanna (SS) subclade at the Crater.

History of elephant re-colonization in northern Tanzania

Mitochondrial DNA has been used to infer the ancestry of populations because it is inherited in haploid fashion. There is a hypothesis that elephants which re-colonized the SE came from two sources, one from the south and the other one from the north (Sinclair et al., 2008) based on anecdotal observations. Our mitochondrial DNA results demonstrate a clear pattern of different elephant groups recolonizing the Serengeti from north and south.

56 Our results affirm that there are at least two sources of populations that colonized the SE: elephants carrying F-clade and those with S-clade. Most elephants in the SE carry haplotypes in F-class clade whereas most elephants in the TME, Selous, and Ruaha carry haplotypes that fall in S-clade. Most elephants from the SE carry similar haplotypes with Bili Forest elephants in the Democratic Republic of Congo and Garamba elephants located in the Guinea-Congolian/ Sudanian area (Ishida et al., 2013). Other elephants in the SE share haplotypes with the elephants from the TME and other populations east of the rift valley wall.

Implications for conservation

The corridor between the Ngorongoro and Lake Manyara seems to be necessary for facilitating gene flow between the two ecosystems. The NCA has shown a high genetic similarity with MAR elephants Figure 2-4C. In recent years, the analysis of the structural connectivity of protected areas has been done at the national level (Riggio et al., 2017). Luckily, the Tanzanian government passed a wildlife policy aimed at protecting migratory wildlife corridors in 2008 (Tanzania, 2008) We strongly recommend an immediate action to restore wildlife corridors between Ngorongoro and Lake Manyara and between Tarangire and Manyara Ranch. Various studies have documented threats facing wildlife corridors in Tanzania (e.g., Newmark 2008, Jones et al., 2009, Lee et al., 2016, Morrison et al. 2014, and Riggio 2017). However, no mitigation efforts have yet taken place.

57 References

Ahlering, M. A., Eggert, L. S., Western, D., Estes, A., Munishi, L., Fleischer, R., Maldonado, J. E. (2012). Identifying source populations and genetic structure for savannah elephants in human-dominated landscapes and protected areas in the Kenya-Tanzania borderlands. PloS One, 7(12), e52288. https://doi.org/10.1371/journal.pone.0052288 Allendorf, F. W. (1986). Genetic drift and the loss of alleles vs heterozygosity. Zoo Biology, 5, 181–190. Blanc, J. J., Barnes, R. F. W., Craig, G. C., Dublin, H. T., Thouless, C. R., Douglas- Hamilton, I., & Hart, J. A. (2007). African elephant status report 2007: an update from the African elephant database. African elephant status report 2007: an update from the African elephant database. https://doi.org/10.2305/iucn.ch.2007.ssc-op.33.en Bohonak, A. J. (2002). IBD (Isolation by Distance): A Program for Analyses of Isolation by Distance, the Journal of Heredity, (93), 153–154. Borner, M. (1985). The Increasing Isolation of Tarangire National Park, Oryx, 19(2), 91– 96. https://doi.org/10.1017/S0030605300019797 Burgess, N. D., Butynski, T. M., Cordeiro, N. J., Doggart, N. H., Fjeldså, J., Howell, K. M., Stuart, S. N. (2007). The biological importance of the Eastern Arc Mountains of Tanzania and Kenya. Biological Conservation, 134(2), 209–231. https://doi.org/10.1016/j.biocon.2006.08.015 Caro, T., Jones, T., & Davenport, T. R. B. (2009). Realities of documenting wildlife corridors in tropical countries. Biological Conservation, 142(11), 2807–2811. https://doi.org/10.1016/j.biocon.2009.06.011 CITES. (2014). Elephant conservation, illegal killing and ivory trade, SC65 Doc., 1–38 Comstock, K. E., Wasser, S. K., & Ostrander, E. A. (2000). Isolation and characterization of polymorphic microsatellite markers in Abudefduf luridus (Pisces : Pomacentridae). Molecular Ecology, 9, 993–1011. https://doi.org/10.1046/j.1365- 294x.2000.00939-8.x

58 Debruyne, R. (2005). A case study of apparent conflict between molecular phylogenies: the interrelationships of African elephants. Cladistics, 21, 31–50. Desai, A., & Baskaran, N. (1996). Impact of human activities on ranging behavior of elephants in the Nilgriri Biosphere Reserve, South India. Journal of Bombay Natural History Society. Douglas-Hamilton, I. (1973). On the ecology and behaviour of the Lake Manyara elephants. African Journal of Ecology, 11(3–4), 401–403. https://doi.org/10.1111/j.1365-2028.1973.tb00101.x Douglas-Hamilton, I. (1987). African elephants: population trends and their causes. Oryx, 21, 11. https://doi.org/10.1017/S0030605300020433 Eggert, Lori S, Ramakrishnan, U., Mundy, N., & Woodruff, D. S. (2000). Polymorphic microsatellite DNA markers in the African elephants (Loxodonta africana) and their use in the Asian elephant (Elephas maximus). Molecular Ecology, 9, 2155– 2234. Eggert, L. S., Maldonado, J. E., & Fleischer, R. C. (2005). Nucleic acid isolation from ecological samples - Animal scat and other associated materials. Methods in Enzymology, 395, 73–87. https://doi.org/10.1016/S0076-6879(05)95006-4 Eggert, L. S., Rasner, C. A., & Woodruff, D. S. (2008). The evolution and phylogeography of the African elephant inferred from mitochondrial DNA sequence and nuclear microsatellite markers. Hungarian Quarterly, 49(April 2002), 1993–2006. https://doi.org/10.1098/rspb.2002.2070 Ernest, E. M., Haanes, H., Bitanyi, S., Fyumagwa, R. D., Msoffe, P. L., Bjørnstad, G., & Røed, K. H. (2012). Influence of habitat fragmentation on the genetic structure of large mammals: evidence for increased structuring of African buffalo (Syncerus caffer) within the Serengeti ecosystem. Conservation Genetics, 13(2), 381–391. https://doi.org/10.1007/s10592-011-0291-0 Estes, A. B., Kuemmerle, T., Kushnir, H., Radeloff, V. C., & Shugart, H. H. (2012). Land-cover change and human population trends in the greater Serengeti ecosystem from 1984–2003. Biological Conservation, 147(1), 255–263. https://doi.org/10.1016/j.biocon.2012.01.010

59 Excoffier, L., Laval, G., & Schneider, S. (2005). Arlequin (version 3. 0): An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 1, 47–50. https://doi.org/10.1177/117693430500100003 Excoffier, L., & Lischer, H. E. L. (2010). Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources, 10(3), 564–567. https://doi.org/10.1111/j.1755- 0998.2010.02847.x Falush, D., Stephens, M., & Pritchard, J. K. (2003). Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies. Genetics, 164(4), 1567–1587. https://doi.org/10.1111/j.1471-8286.2007.01758.x Falush, D., Stephens, M., & Pritchard, J. K. (2007). Inference of population structure using multilocus genotype data: Dominant markers and null alleles. Molecular Ecology Notes, 7(4), 574–578. https://doi.org/10.1111/j.1471-8286.2007.01758.x Fernando, P., & Lande, R. (2000). Molecular genetics and behavioral analysis of social organization in the Asian elephant (Elephas maximus). Behavioral Ecology and Sociobiology, 48, 84–91. Foley, C. A. H., & Faust, L. J. (2010). Rapid population growth in an elephant (Loxodonta africana) population recovering from poaching in Tarangire National Park, Tanzania. Oryx, 44(02), 205–212. https://doi.org/10.1017/S0030605309990706 Francis, R. M. (2017). pophelper: an R package and web app to analyse and visualize population structure. Molecular Ecology Resources, 17(1), 27–32. https://doi.org/10.1111/1755-0998.12509 Gilpin ME, & Soulé ME. (1986). Minimum viable populations: process of species extinction. Conservation Biology, the Science of Scarcity and Diversity, 19–34. Goudet, J. (1995). FSTAT (version 1.2): A computer Program to calculate F-Statistics. Journal of Heredity, 86(6), 485–486. Henle, K., Davies, K. F., Kleyer, M., Margules, C., & Settele, J. (2004). Predictors of species sensitivity to fragmentation. Biodiversity and Conservation, 13, 207–251.

60 Hinde, R. ., Corti, G. ., Fanning, E., & Jenkins, R. K. . (2001). Large mammals in miombo woodland, evergreen forest and a young teak (Tectona grandis) plantation in the KilomberoValley,Tanzania. East African Wildlife Society, 39, 318–321. Hochberg Y (1988) A sharper Bonferroni procedure for multiple tests of significance. Biometrika, 75, 800–803. Holm S (1979) A simple sequential rejective multiple test procedure. Scand. J. Stat., 6, 65–70. Homewood KM, Rodgers WA (1991). Masailand Ecology. Cambridge University Press, Cambridge, UK. Ishida, Y., Georgiadis, N. J., Hondo, T., & Roca, A. L. (2013). Triangulating the provenance of African elephants using mitochondrial DNA. Evolutionary Applications, 6(2), 253–265. https://doi.org/10.1111/j.1752-4571.2012.00286.x Jones, O., & Wang, J. (2010). COLONY: A program for parentage and sibship inference from multilocus genotype data. Molecular Ecology Resources, 10(3), 551–555. https://doi.org/10.1111/j.1755-0998.2009.02787.x Jones, T., Cusack, J. J., Pozo, R. A., Smit, J., Mkuburo, L., Baran, P., Foley, C. (2018). Age structure as an indicator of poaching pressure: Insights from rapid assessments of elephant populations across space and time. Ecological Indicators, 88, 115–125. https://doi.org/10.1016/j.ecolind.2018.01.030 Kinuthia, J., Harper, C., Muya, S., Kimwele, C., Alakonya, A., Muigai, A., Gatebe, E. (2015). The selection of a standard STR panel for DNA profiling of the African elephant (Loxodonta africana) in Kenya. Conservation Genetics Resources. https://doi.org/10.1007/s12686-014-0366-6 Kioko, J., Zink, E., Sawdy, M., & Kiffner, C. (2013). Elephant (Loxodonta africana ) Tarangire-Manyara Ecosystem , Tanzania. South African Journal of Wildlife Research, 43(1), 44–51. Kumar, S., Nei, M., Dudley, J., & Tamura, K. (2008). MEGA: A biologist-centric software for evolutionary analysis of DNA and protein sequences. Briefings in Bioinformatics, 9(4), 299–306. https://doi.org/10.1093/bib/bbn017

61 Lacy, R. C. (1995). Clarification of genetic terms and their use in the management of captive populations. Zoo Biology, 14(6), 565–577. https://doi.org/10.1002/zoo.143014060 Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., Mcgettigan, P. A., McWilliam, H., Higgins, D. G. (2007). Clustal W and Clustal X version 2.0. Bioinformatics, 23(21), 2947–2948. https://doi.org/10.1093/bioinformatics/btm404 Leigh, J. W., & Bryant, D. (2015). POPART: Full-feature software for haplotype network construction. Methods in Ecology and Evolution, 6(9), 1110–1116. https://doi.org/10.1111/2041-210X.12410 Lowe, W. H., & Allendorf, F. W. (2010). What can genetics tell us about population connectivity? Molecular Ecology, 19(15), 3038–3051. https://doi.org/10.1111/j.1365-294X.2010.04688.x Makacha S, Mollel CL, Rwezaura J (1979), The conservation status of the black rhinoceros (Diceros bicornis L.) in the Ngorongoro Crater, Tanzania. African Journal of Ecology, 17: 97–103. Morrison, T. A., Link, W. A., Newmark, W. D., Foley, C. A. H., & Bolger, D. T. (2016). Tarangire revisited: Consequences of declining connectivity in a tropical ungulate population. Biological Conservation, 197, 53–60. https://doi.org/10.1016/j.biocon.2016.02.034 Morrison, T., & Bolger, D. (2014). Connectivity and bottlenecks in a migratory. Fauna & Flora International, Oryx, 48(4), 613–621. https://doi.org/10.1017/S0030605313000537 Msoffe, F. U., Said, M. Y., Ogutu, J. O., Kifugo, S. C., De, J., Gardingen, P. Van, & Reid, R. S. (2011). Spatial correlates of land-use changes in the Maasai- Steppe of Tanzania: Implications for conservation and environmental planning. International Journal of Biodiversity and Conservation, 3(7), 280–290. Mwalyosi, R. B. B. (1991). Ecological evaluation for wildlife corridors and buffer zones for Lake Manyara National Park, Tanzania, and its immediate environment.

62 Biological Conservation, 57, 171–186. https://doi.org/10.1016/0006- 3207(91)90137-X Newmark, W. D. (2008). Isolation of African protected areas. Frontiers in Ecology and the Environment, 6(6), 321–328. https://doi.org/10.1890/070003 Nyakaana, S., Arctander, P., & Siegismund, H. R. (2002). Population structure of the African savannah elephant inferred from mitochondrial control region sequences and nuclear microsatellite loci. Heredity, 89(2), 90–98. https://doi.org/10.1038/sj.hdy.6800110 O’Brien, S. J., Roelke, E., Marker, L., Newman, A., Winker, C. ., Meltzer, D.,Wildt, D.(1985). Genetic Basis for Species Vulnerability in the Cheetah. Science, 227(4693), 1428–1434. Okello, J. B. A., Wittemyer, G., Rasmussen, H. B., Douglas-Hamilton, I., Nyakaana, S., Arctander, P., & Siegismund, H. R. (2005). Noninvasive genotyping and Mendelian analysis of microsatellites in African savannah elephants. Journal of Heredity, 96(6), 679–687. https://doi.org/10.1093/jhered/esi117 Peakall, R., & Smouse, P. E. (2012). GenALEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics, 28(19), 2537–2539. https://doi.org/10.1093/bioinformatics/bts460 Poole, J. H. (1987). Rutting Behavior in African Elephants : The Phenomenon of Musth Stable. Behaviour, 102(3), 283–316. Prins, H. H. T., Jeugd, H. P. V. Der, & Beekman, J. H. (1994). Elephant decline in Lake Manyara National Park, Tanzania. African Journal of Ecology, 32(3), 185–191. https://doi.org/10.1111/j.1365-2028.1994.tb00570.x Pritchard, J. K. J., Stephens, M. M., & Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics, 155(2), 945–959. https://doi.org/10.1111/j.1471-8286.2007.01758.x Ramanadevi, V., Thangaraj, M., Sureshkumar, A., & Subburaj, J. (2013). Detection of Genetic Variation and Genetic Diversity in Two Indian Mudskipper Species ( Boleophthalmus boddarti , B . dussumieri ) using RAPD Marker. Notulae Scientia Biologicae, 5(2), 139–143.

63 Raymond, M., & Rousset, F. (1995). GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal of Heredity, 86, 248–249. Riggio, J., & Caro, T. (2017). Structural connectivity at a national scale : Wildlife corridors in Tanzania. PLoS One, 12(11), 1–16. Rousset, F. (2008). GENEPOP’007: A complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources, 8(1), 103–106. https://doi.org/10.1111/j.1471-8286.2007.01931.x Sinclair, A.R.E, (2012), Serengeti Story: Life and science in the world’s greatest wildlife region, Oxford, University Press, UK. Sinclair, A. R. E., Hopcraft, J. G. C., Olff, H., & Mduma, S. A. R. (2008). Historical and Future Changes to the Serengeti Ecosystem. University of Chicago Press, (1972), 7–46. Takezaki, N., Nei, M., & Tamura, K. (2010). POPTREE2 : Software for Constructing Population Trees from Allele Frequency Data and Computing Other Population Statistics with Windows Interface. Molecular Biology and Evolution, 27(4), 747– 752. https://doi.org/10.1093/molbev/msp312 TANAPA website accessed February 12, 2019, http://www.tanzaniaparks.go.tz/ Tanzania, United Republic of (2008). The Wildlife Policy of Tanzania, The Ministry of Natural Resources and Tourism. TAWIRI. (2015). Population Status of Elephant in Tanzania 2014, TAWIRI Aerial Survey Report. Van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M., & Shipley, P. (2004). MICRO- CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4(3), 535–538. https://doi.org/10.1111/j.1471-8286.2004.00684.x Wittemyer, G., Elsen, P., Bean, W. T., Burton, A. C. O., & Brashares, J. S. (2008). Accelerated Human Population Growth at Protected Area Edges. Science, 321(5885), 123–126. https://doi.org/10.1126/science.1158900 Wittemyer, George, Northrup, J. M., Blanc, J., Douglas-Hamilton, I., Omondi, P., & Burnham, K. P. (2014). Illegal killing for ivory drives global decline in African

64 elephants. Proceedings of the National Academy of Sciences, 111(36), 13117– 13121. https://doi.org/10.1073/pnas.1403984111 Zong, J. W., Zhao, T. T., Ma, Q. H., Liang, L. S., & Wang, G. X. (2015). Assessment of Genetic Diversity and Population Genetic Structure of Corylus mandshurica in China Using SSR Markers. PLoS One, 10(9), e0137528. https://doi.org/10.1371/journal.pone.0137528

65 Table 2-S1.Characterization of 11 microsatellite loci for the African Savanna elephants

Original Dye Panel Fragment Ta Primer Source Primer Sequence 5’-3’ length ºC Eggert et al. NED 3 LA5-F (2000) GGGCAGCCTCCTTGTTTT 139-155 58 LA5-R CTGCTTCTTTCATGCCAATG 58 Comstock et VIC 3 FH60-F al. (2000) CAAGAAGCTTTGGGATTGGG 143-163 58 FH60-R CCTGCAGCTCAGAACACCTG 58 Archie et al. VIC 1 LaT24-F 2003 AAGTTGAGAGATCAGCAAAGCA 124-264 58 LaT24-R GATGTTCAGTCCTTCCTTAGCA 58 Eggert et al. VIC 1 LA6-F (2000) AAAATTGACCCAACGGCTC 145-177 58 LA6-R TCACGTAACCACTGCGCTA 58 Comstock et 6- 4 FH19-F al. (2000) GAAGCTCATGGTCAAGGTCAC FAM 185-207 58 FH19-R CTGCATACTCATCGAAGTCACC 58 Comstock et NED 1 FH67-F al. (2000) GCTTCTCTAGAAATGTGTATGC 88-110 55 FH67-R GGCGTATAGGATAGTTCCAC 55 LafMS02- Nyakaana et NED 2 F al. (2005) GAAACCACAACTTGAAGGG 136-168 55 LafMS02- R TCGCTTGTAAGAAGGCGTG 55 Comstock et NED 2 FH48-F al. (2000) GAGTCTCCATAATCAAGAGCG 166-182 55 FH48-R CCTCCCTGGAATCTGTACAG 55 Archie et al. 6- 1 LaT08-F 2003 ATGGACAGGCAGAAAGATTT FAM 177-230 55 LaT08-R TCCCAATAACAGGATAGCATT- 55 Archie et al. 6- 2 LaT06-F 2003 AGCCAGGCACATTAAGTGT FAM 270-398 55 LaT06-R CTCCTAGAAAAGGTTACCAC 55 Archie et al. NED 2 LaT13-F 2003 AGCTTCTGTAGGCTCTGA 216-272 55 LaT13-R ACTCGATAAACAGTGTTGA 55

66 Table 2-S2. Number of elephant fecal samples collected from each sampling locations in Tanzania

Location No. of samples No. samples genotyped collected Northern Serengeti (NSE) North Serengeti National Park (NSNP) 61 61 Western Serengeti National Park (WSNP) 50 48 Central Serengeti National Park (CSNP) 35 34 Loliondo Game Controlled Area (LGCA) 40 37

Ikorongo-Grumeti Game Reserve (IGGR) 61 30 Southern Serengeti (SSE) Southern Serengeti National Park (SSNP) 25 23 Maswa Game Reserve (MGR) 47 47 Mwiba Wildlife Ranch (MWR) 51 51 Ngorongoro Conservation Area (NCA) Ndutu (NDT) 25 25 Crater, Kakesio, and Oldeani (NCA) 101 89

Manyara (MAR) Lake Manyara National Park (LMNP) 64 39 Manyara Wildlife Ranch (MANR) 74 64 Tarangire National Park (TNP) 67 67

Ruaha National Park (RNP) 49 49 Selous Game Reserve (SGR) Matambwe (MAT) 22 22 Kingupira (KPR) 28 28 Total 800 714

67 A (K=2): All B, K=2: Serengeti ecosystem

C (K=2): NCA-TNP D (K=2): NSE & SSE

E (K=3): SSE&NCA F (K=2): MAR&TNP

68 G (K=1): NSE H (K=1): NCA

I (K=3): SGR&RNP J(K=3): MANR&LMNP

Figure 2-S 1. Optimum number of clusters (delta K) using the Evanno method (A-J) as shown in Figure 2-4

69

Figure 2-S 2 A graph shows Fst for microsatellite loci (blue) and mitochondrial DNA (red) normalized to geographic distance.

We found a few exceptions to the isolation by distance assumptions, for example, TNP & MAR were only about 60 km apart but the normalized Fst was high for both nuclear markers and mtDNA suggesting that the populations have not been mixing much. The highest normalized Fst was between NSE &TNP as expected, as shown in Figure 2-10. All the abbreviations for the sampling location is shown in Table 2-S2.

70

: Little evidence for female mediated-gene flow for the African savanna elephants between the Greater Ruaha and Selous ecosystems in Tanzania

George G. Lohay, Anna B. Estes and Douglas R. Cavener

Abstract

Loss of connectivity among animal and plant populations due to land-use change is one of the major threats to conservation. To initiate conservation programs to restore wildlife corridors, we need to understand the historical connectivity of population before fragmentation. However, there is a lack of knowledge about historical ecosystem connectivity with which to set priorities of restoration areas. We used both mitochondrial DNA (cytochrome b) in the control region and nuclear microsatellite markers to determine whether elephants from the Greater Ruaha ecosystem are genetically isolated from the Selous ecosystem. We sequenced 622 bp of the mitochondrial DNA control region and 11 microsatellite loci and performed genetic analyses to assess the genetic structure. Our results identified three subpopulations using nuclear markers, and all pairwise FST values were statistically significant. We identified 12 haplotypes which were all under Savanna- clade. Unexpectedly, there was only one haplotype that was shared between Ruaha (only one sample) and Selous. These results suggest that there was no evidence for female- mediated gene flow between the ecosystems. However, the microsatellite analysis indicated some level of gene flow between Ruaha and Selous, particularly, between Matambwe and Ruaha. Our study also revealed unidirectional colonization of elephants from Selous to Ruaha because the most frequent haplotype in Ruaha (84%) was not found in Selous, but one haplotype from Selous was found in Ruaha. The Eastern Arc Mountains seems to impede the movement of elephants. It is likely that even before the effects of human activities on wildlife corridors, elephant populations in this region were isolated.

71 Introduction

Fragmentation of natural habitats is a significant challenge in conservation biology and one of the top threats to biodiversity worldwide (Fahrig, 2003; Henle et al., 2004; Wilcox & Murphy, 1985; Haddad et al. 2015). Increasing connections between populations is becoming crucial in conservation genetics because wildlife corridors are being lost throughout the world at escalating speed because of increased human activities (Epps et al., 2013). Wildlife corridors are essential for maintaining the viability of isolated populations and conserving ecosystems functions (Beier and Nos 1998). Tanzania is among the few countries which have documented wildlife corridors at the national level (Caro, Jones, & Davenport, 2009; Riggio & Caro, 2017). However, current patterns of human settlement are causing unprecedented changes in connectivity among elephant populations (Epps et al., 2013). Therefore, there is a need to increase conservation efforts aiming at increasing connectivity between isolated population (Newmark, 1996, 2008; Soule, 1979). For the restoration of wildlife corridors, there is a need to understand historical and contemporary connectivity to determine populations that have been recently affected by changes in human land use. Female elephants are philopatric and remain with their natal herd (Archie et al., 2007); whereas males are ejected from the herd upon sexual maturity and subsequently facilitate gene flow between herds (Archie et al., 2007, Hollister-Smith et al., 2007). Several conservation genetic studies have used mitochondrial DNA, particularly the control region, to measure molecular diversity and identify conservation units for better management of species (Arif & Khan, 2009), population structure changes in sex-specific gene flow (Brandt, 2015). Furthermore, historical movement patterns can be determined using mitochondrial and nuclear DNA. The use of both mitochondrial DNA and microsatellite markers offers us a unique opportunity to understand past and contemporary connectivity of elephants in central and Southern Tanzania. The Greater Ruaha and Selous ecosystems are essential areas for elephants in Tanzania, accounting for 54% of the national elephant population. Unfortunately, these areas were heavily impacted by poaching in recent years. In Selous, the population of

72 elephants declined from 44,806 in 2009 to 15,217 in 2014 whereas in Ruaha the decline was from 34,664 in 2009 to 8,272 in 2014 (TAWIRI, 2015). To keep these elephant populations healthy, and enable long term survival, it has been suggested that it is vital to maintain connectivity between these populations (Jones et al., 2012). There was previously movement of elephants between the Greater Ruaha and Selous ecosystem in 2009, through corridors that connect to Udzungwa and Mikumi National Park, (Nahonyo, 2009). However, these elephant migratory routes, ranging and dispersal areas are threatened by habitat loss through the expansion of agriculture and human settlement, and poaching (Douglas-Hamilton et al., 2005; Galanti et al., 2006; Wittemyer et al., 2007). On the migratory routes poaching is more likely to occur than inside the protected areas (Nahonyo, 2009). In recent years, the wildlife corridors between Selous GR and Udzungwa Mountains National Parks have closed completely (Jones et al., 2012) and measures to restore three corridors namely, the Nyanganje, Ruipa, and Mwanihana-Magombera were proposed (Jones et al., 2012). However, restoration of corridors requires significant resources. For many long-lived species, such as elephants, it is critically important to understand connectivity because the rate at which habitat is being changed is sufficiently rapid and an evolutionary response to the changes is unlikely (Niko et al., 2016). There is a tendency to assume that elephants (and other wildlife) used to move freely across the landscape and mate in a panmictic manner before habitat loss. In this paper, we aim to determine if elephants from Ruaha are genetically isolated from Selous and whether the isolation is recent due to habitat loss or whether the populations were distinct even before the effects of human activities. We test the hypothesis that the Ruaha and Selous elephants have historically been genetically homogenous, which is now being threatened by corridors being closed.

73 Materials and Methods

Study Areas

Ruaha National Park (RNP) is the largest national park in Tanzania, and it covers 20,226 km2 (TANAPA). The Ruaha NP experiences dry and rainy season with an annual rainfall of is 516.7mm (T. Jones et al., 2018). The rainy season typically runs between December and May. The Greater Ruaha River and Ihefu wetland are the primary source of water in the ecosystem. Ruaha has diverse vegetation types from Miombo woodland that extends from Southern Africa to acacia woodlands (TANAPA website). We collected data in the eastern part of the Ruaha National park during the dry season from July 04-July 20, 2017. Between the Greater Ruaha and Selous ecosystem, there is a chain of mountains that runs from southern Kenya to southeast Tanzania (Figure 3-1). Selous Game Reserve (SGR) is the largest game reserve in African located in the southeast of Tanzania at lat. 7°35′S, long. 38°15′E (Spong 2002). Other protected areas adjacent to the SGR include Mikumi and Udzungwa Mountains National Parks, Kilombero Valley Ramsar site and wildlife management areas (Figure 3-1). Due to its size, Selous GR has eight administrative sectors. In this study, we focused on two sectors: Matambwe (MAT) and Kingupira (KPR) sectors. Matambwe is the only sector within the Selous GR that conducts photographic tourism. The rest of the SGR is designated for trophy hunting blocks. Rufiji River separates Matambwe from Kingupira sector.

74

Figure 3-1. Map showing protected areas from which samples were obtained. Mikumi National Park is within the Selous ecosystem. Wildlife corridors between Selous and Udzungwa and Kilombero valley are closed due to agricultural expansion and human settlement (Jones et al., 2012).

Field data sample collection

Fieldwork was conducted in Ruaha NP, Matambwe and Kingupira sectors in Selous GR from June to September 2017. Dung samples were collected at three locations. For each dung sample, about 5g of the external portion of fresh dung was collected and placed into a 50ml tube. We only collected fresh fecal samples from dung deposited within approximately 12 hours to make sure that the intestinal epithelial cells are still alive (should have wet potions). Scrapings were taken from the outer layer of the bolus, where most of the epithelial cells are found, the following methods established by Ahlering et al. (2012).

75 Samples were preserved using Queens College Buffer (20% DMSO, 0.25 M EDTA, 100m tris, PH 7.5 saturated with NaCl). To avoid cross-contamination between samples, new pairs of gloves were used for each sample collected. We recorded GPS coordinates for each sample collected. We measured dung circumference for at least three dung boli from each individual to estimate the age class of elephants (Hema et al., 2018) because dung circumference changes with age (Morgan & Lee, 2003). We aimed at collecting at least 25 samples from each of the three sampling location as it has been shown that 25-30 samples are enough for population genetic analyses (Hale et al. 2012). The collection of samples was done over a short period to avoid sampling closely related individuals or resample the same individuals.

We extracted DNA using the manufacturer’s recommended protocol with DNA extraction kits (QIAamp DNA Stool Mini Kit) with minor modifications, incubating overnight at 55°C,(Eggert et al., 2005). Fecal samples were vortexed for 1min and poured directly into the 5ml tube. We increased the initial volume of samples from 200 µl to 1000 µl, and 1400 µl of buffer ASL was added. 25 µl of Proteinase K was added, and the mixture was thoroughly vortexed for 1min. Incubator shaker was set at 55°C and the samples placed horizontally on a rack and tied with a plastic wrap. Samples were incubated for 12 hours. Other steps were the same as the QIAamp stool mini kit. To obtain a high concentration of DNA, we used a final elution of 100 µl. The DNA concentration was measured using a nanodrop and recorded the ratio of A260/A280. Pure DNA has an A260/A280 ratio of 1.7–1.9. We re-extracted samples that had low concentration.

Microsatellites

We selected 11 microsatellites which had previously been found to be polymorphic in African elephants LAT06, LAT08, LAT13, LAT24 (Archie et al, 2003), FH19, FH48, FH60, FH67 (Comstock et al, 2000), LA5, LA6 (Eggert, et al. 2000), and LafMS02 (Nyakaana et al, 2005) with some minor modifications (Kinuthia et al., 2015) as summarized in (Table 2-S1). Standard QIAGEN multiplex PCR protocol was used. We

76 used 12.5 µl PCR reaction volume containing 6.25 µl of the master mix, 0.5 µl of primer mix (standardized to 2 µM), three µL of template DNA and 2.75 µl of millipure water. We performed PCR with the initial polymerase activation step at 95°C for 15 min, denaturation at 94°C for 30 seconds, annealing at 55°C or 58°C for 1:30 min, extension at 72°C for 1 min for 35 cycles. Initial evaluation of SSRs primers was done by agarose gel electrophoresis. Aliquots of 5 µl of each PCR product was loaded onto a 2% agarose gel, with electrophoresis run time of 45 minutes at 120V. The fragment lengths for each DNA sample (allele) was estimated from digital gel images. For genotyping, we grouped 11 SSRs into four panels for genotyping. All multiplex PCR samples were genotyped and separated by electrophoresis using ABI 3730xl DNA analyzer (Applied Biosystems) at Genomics Core Facility of the Pennsylvania State University. The GeneScan results obtained were analyzed using GeneMapper® v. 5 (Applied Biosystems). Inconsistent allele scores were genotyped twice, and the weaker peaks with less than 100 fluorescent units were considered failed and could not be reliably scored (Okello et al., 2005).

To account for allelic dropout, null alleles and scoring error due to stuttering we used MICRO-CHECKER 2.2.3 (VAN OOSTERHOUT et al., 2004). We then corrected genotypes based on MICRO-CHECKER 2.2.3 results. The number of alleles and their frequencies were determined for all loci across all individuals in all the populations using GenAIEx 6.502 (Peakall and Smouse, 2012). We also used GenAIEX to export files into formats compatible with other genetic analyses software. We tested for deviations from expectations under Hardy-Weinberg equilibrium (HWE) and for linkage disequilibrium (LD) within protected areas in GENEPOP (Raymond & Rousset, 1995; Rousset, 2008), accessible online at http://gene- pop.curtin.edu.au/. For the two analyses, the Markov Chain strategy was used with 1000 dememorizations, 1000 for combining independent test results across study location and loci were used to determine the statistical significance test results. Loci that deviated from HWE were discarded for further analyses after applying Bonferroni correction of multiple comparisons. To determine if there is isolation by distance (IBD) we used IBD program (Bohonak, 2002) and plotted the graph of (FST /1- FST Vs geographic distance (km). FSTAT (Goudet, 1995) was used to calculate the inbreeding coefficient

77

(Fis) and allelic richness (AR) which represents the number of alleles standardized to the smallest sample size in the study area. We used both GENAIEX (Peakall & Smouse, 2012) and FSTAT (Goudet, 1995) to assess genetic diversity by calculating the expected (HE) and observed (HO) heterozygosities and the level of genetic differentiation(FST) between the sampling locations.

Population Structure

We used STRUCTURE 2.3 (Falush et al., 2003, 2007; Pritchard et al., 2000) to determine the elephant population structure in northern and southern Tanzania. This program uses a Bayesian clustering model to assign individuals to subpopulations while simultaneously estimating population allele frequencies. This model aims to determine the number of subpopulations K within the population, where K in most cases is unknown. We inferred the number of K by running ten iterations for each K value from 1 to 10 using an admixture model with a LOCPRIOR option. We set a burn-in period at 1x106 and Markov Chain Monte Carlo (MCMC) repetition value of 1x106. We used STRUCTURE harvester to detect the best value of K and visualize the STRUCTURE results using Pophelper (Francis, 2017). From this point, we will use the term “subpopulation” to define groups of individuals identified through STRUCTURE.

Mitochondrial Sequencing and Analysis

For mitochondrial (mt)DNA we used all 99 samples to capture all existing haplotypes in the sampling locations. We sequenced 630 base pair (bp) of cytochrome b gene using forward primers MDL5 5′- TTACATGAATTGGCAGCCAA CCAG- 3′ and reverse primers MDL3 5′- CCCACAATTAATGGGCCCGGAGCG- 3' (Fernando & Lande, 2000). For samples that did not amplify successfully, we used a different reverse primer mtCR3 5′- GTC ATT AAT CCA TCG AGA TGT CTT ATT TAA GAG G- 3'. We

78 performed PCR reaction with the initial polymerase activation step at 95°C for 3 min, denaturation at 95°C for 30 sec, annealing temperature at 60°C for 45 seconds, extension at 72°C for 30 sec for 35 cycles. Each PCR mixture contained 3μl of 5X Green GoTaq reaction buffer (Promega), a final concentration of 0.33μM for each of the primers, 0.13μM of dNTP (Quanta bio), Bovine serum albumin (BSA) 0.1μg/μl, and 3μl of DNA template of unknown concentration in a 15μl volume. For each PCR reaction with test samples, we also ran a negative control (no DNA) and a positive control using DNA from known tissue samples. 6μl of the PCR product was used for electrophoresis in Tris-Acetate EDTA running buffer at 120V for 45 min on a 2% agarose gel stained with GelRed (Biotium). We sequenced PCR products using reverse primer only unless the sample had unclean sequence results or had a unique haplotype. Sequence results in the trace file format were visually inspected using SnapGene® software 4.2.4 (from GSL Biotech; available at snapgene.com) by comparing to a reference sequence for the hypervariable control region of elephants (Hauf, et al., 1999). We compared our sequences with other published studies to identify unique haplotypes. We compared our sequences with other published studies to identify unique haplotypes (Ahlering et al., 2012; Debruyne, 2005; Eggert et al., 2008; Ishida et al., 2013). We trimmed sequences and collapsed haplotypes using FaBox (Villesen 2007). We aligned all sequences using Clustalw ( https://www.genome.jp/tools- bin/clustalw). We calculated haplotype diversity (h), nucleotide diversity (π), and Tajima D using dnaSP V6 (Rozas et al., 2017) and constructed median-joining network PopArt 4.8.4 (Leigh & Bryant, 2015). We inferred phylogenetic relationships among unique haplotypes using PAUP using the bootstrap method with 1000 bootstrap replications (Swofford, 2002).

Results

We successfully genotyped 99 samples using 11 SSRs (49 from Ruaha and 50 from Selous). Within the Selous GR, 22 samples were from Matambwe and 28 from Kingupira sectors. Two samples from Ruaha had the same multi-locus genotype; we removed one

79 individual. Eight loci in Ruaha and Kingupira and ten loci from Matambwe conformed to Hardy-Weinberg equilibrium assumptions. FH60 locus significantly deviated from HWE for all populations, and we removed it for subsequent analyses. In Matambwe no loci were in significant linkage disequilibrium while there were three and two loci, in Kingupira and Ruaha respectively, which were in significant linkage disequilibrium but we retained them for further analyses. We have summarized the allelic richness and other indices of genetic diversity in (Table 3-1)

Table 3-1. Genotyped African savanna elephants (n=98) obtained from Ruaha and Selous Game reserve in 2017.

Population N Na AR Ho uHe F FIS HWE LD KPR 28 7.00 6.61 0.623 0.715 0.132 0.130 8/11 3/55 MAT 22 7.45 7.35 0.635 0.740 0.127 0.145 10/11 0/55 RNP 48 8.64 6.87 0.654 0.731 0.104 0.106 8/11 2/55

Sample size (N), Mean number of alleles (Na), mean allelic richness (AR), observed heterozygosity (Ho), unbiased expected heterozygosity (uHE), proportion of loci conforming to Hardy–Weinberg equilibrium (HWE), proportion of locus pairs in significant (P < 0.000138) linkage disequilibrium (LD) and inbreeding coefficient (FIS)

All pairwise comparisons of genetic fixation (FST) and genetic differentiation (F’ST) between populations were statistically significant (p≤0.05) for the 11 neutral markers

(Table 3-2). The lowest FST of 0.018 was between KPR & MAT whereas the highest FST was between KPR & RNP (Table 3-2). The FST values between RNP & MAT was 0.026 which was lower compared to the FST value between KPR and RNP. Analysis of Bayesian clustering using the STRUCTURE program revealed three clusters (Figure 3-2). The optimum value of delta K (K=3) and mean of estimated likelihood proportion of data is shown in Figure 3-S1. Membership coefficient from STRUCTURE output for all three sampling locations for the three clusters is shown in Figure 3-2A.

80

Figure 3-2. Individual population assignment from Bayesian STRUCTURE and mtDNA subclades for African savanna elephants in central and southern Tanzania

(A) Individual population assignment from Bayesian STRUCTURE for 98 African elephants in central Tanzania, three clusters (K=3), (B) membership coefficients from the clustering analysis results based on 10 microsatellite loci, each pie chart represents a subpopulation and the colors of the pie chart represents each individual’s assignment probabilities to each of the three clusters. (C) Mitochondrial DNA haplotype distribution for the two main subclades: Savanna wide (SW) and Southeast savanna (SS). Ruaha National Park (RNP), Kingupira (KPR) and Matambwe (MAT) in Selous Game Reserve.

Results suggest the presence of gene flow or recent gene flow between RNP and MAT as they are geographically closer to one another than KPR & RNP. Principal coordinate analysis using Fst values supports clustering of subpopulations (Figure 3-3).

81

A

B

Figure 3-3. Principal Coordinate Analysis (PCoA) for 11 microsatellite loci

A) Genetic distance between individuals B) pairwise FST between subpopulations. MAT=Matambwe, KPR=Kingupira and RNP=Ruaha National Park

Spatial autocorrelation was positive for individuals within 126 km, although it was not significant. In all cases, spatial autocorrelation was not significantly different from zero (Figure 3-4). A positive correlation was found between interpopulation genetic differences

(linearized FST) and geographic distance at the microsatellite loci ( r=0.9006; p=0.1630) than at mtDNA (r= -0.1946, p=0.359) from 1000 randomization (Bohonak, 2002). Therefore, there was a higher effect of the isolation-by-distance in microsatellites than in

82 mtDNA. This implies that other factors could explain the observed haplotype subdivision such as the presence of the Eastern Arc Mountains.

Figure 3-4. Genetic spatial autocorrelation analysis for African savanna elephants in Ruaha and Selous ecosystems in Tanzania. Upper and lower error bars bound the 95% confidence interval about r as determined by bootstrap resampling (Peakall and Smouse, 2012).

Mitochondrial DNA

We successfully sequenced 55 samples from Ruaha and Selous Game Reserve. A total of 39 sites were found to be polymorphic for 591 bp region of mitochondrial DNA analyzed (Table 3-3). All haplotypes identified fell under the savanna clade (S-clade) (Debruyne, 2005). Within the S-clade, there are two subclades: Savanna-wide and Southeast-savanna (Ishida et al., 2013). Ruaha elephants had only three haplotypes (one unique) whereas in Selous GR there were ten haplotypes (eight unique) (Table 3-3). In Ruaha, one haplotype was carried by 84% of individuals sampled, and only one haplotype was unique to Ruaha. Only one haplotype was shared between Ruaha and Selous. In Selous, eight out of ten haplotypes were unique to Selous, and two were previously published.

83

Table 3-2. Distribution of the 12 observed mitochondrial DNA (D-loop) haplotypes from a sample of 55 African savanna elephants from three localities in Tanzania.

The vertical numbers indicate the positions of polymorphic sites relative to haplotype SSSG02. A dash (-) represents a deletion introduced to optimize alignment. Numbers on the right indicate haplotype frequency for each of the locations.

We observed approximately the same level of mitochondrial nucleotide diversity (π) of 1.036% to the wide continental level of 2.0% (Nyakaana et al., 2002). This low nucleotide diversity was attributed to low haplotype diversity in Ruaha. The highest haplotype diversity of 0.944 was observed in Matambwe whereas the lowest (0.29) was observed in Ruaha (Table 3-3). Ruaha was found to have the lowest nucleotide diversity compared to

Kingupira and Matambwe (Table 3-3). The FST value of 0.5925 was found between MAT and RNP, but the lowest was between MAT and KPR (Table 3-2). The value of Tajima’s D was 1.039 which was not significant at P≤0.05, meaning that our data fit with the assumptions of neutrality. The value of D is expected to be less than 0 under the bottleneck effect while a positive value indicates more intermediate frequency variants

84 Table 3-3 Parameters of genetic diversity based on mitochondrial DNA sequence data for two populations of African elephants

Location Sample size Number of Haplotype Nucleotide haplotypes (H) diversity (Hd) diversity (π)

Kingupira 21 6 0.862 0.1528 Matambwe 9 6 0.944 0.1874 Ruaha 25 3 0.290 0.0039 Total 55 12* 0.8276 0.1278 This table includes the number of the samples (N), number of haplotypes (H), haplotype diversity (Hd) and nucleotide diversity () *Some haplotypes were shared between Ruaha and Selous

Phylogeny of 591 bp supports the subdivision of mtDNA into two subclades. PAUP* program was used to obtain statistical support for the nodes in the tree using 1000 bootstrap replication and only nodes with ≥50% replicate bootstrap values are indicated). Reference for previously published haplotypes is shown in Table 3-4. The evolutionary relationship was inferred using a Neighbor-joining method.

85

Figure 3-5: A neighbor-joining phylogenetic tree showing the relationship among 12 haplotypes obtained from 55 individual elephants in the Ruaha and Selous ecosystems. The numbers between the branches represent the bootstrap percentage values for 1000 replicates. We indicated the haplotype frequencies with “x” for example, SSSG01x5 means that five individuals carried this haplotype. We included outgroup from the F- clade (Ishida et al, 2013), and Asian and forest elephants (Table 3-4)

86 A median-joining network also supports the subdivision of the subclades between Ruaha and Selous. The majority of Ruaha elephants carried one haplotype whereas in Selous we observed more haplotypes in southeast savanna (Figure 3-6). We included other known haplotypes from previously

Figure 3-6. Median-joining network reconstruction of L. Africana showing the genetic relationship among the cytochrome b haplotypes of mitochondrial DNA.

Size of the circles is proportional to haplotype frequencies, and the number of mutations separating the haplotypes is indicated with hatch marks. Numbers 1-26 represents the haplotypes used in this study including Asian elephant (18) and Forest elephant (21) (Table 3-4).

87 Table 3-4 List of haplotypes used in this study

S/N Haplotype Country of Accession Reference Name Origin No. 1 SE2051 Tanzania JQ438649 Ishida et al 2013 2 SWTA01 Tanzania JQ438685 Ishida et al 2013 3 SSRU01 Tanzania This study 4 SSSG03 Tanzania This study 5 SSNG02 Tanzania This study 6 SSSG01 Tanzania This study 7 SSSG02 Tanzania This study 8 SSNG01 Tanzania JQ438605 Ishida et al, 2013 9 SSNG02 Tanzania This study 10 SSSG06 Tanzania This study 11 SSSG07 Tanzania This study 12 SSSG10 Tanzania This study 13 SSSG08 Tanzania This study 14 SSSG09 Tanzania This study 15 SSSG05 Tanzania This study 16 KE4506 Kenya JQ438389 Ishida et al 2013 17 GR0042 DR-Congo JQ438333 Ishida et al. 2013 18 E.maximus India AB002412 Moro et al, 1998 19 TA2 Tanzania AY741070 Debruyne 2005 20 Addo5 South Africa AF527682 Eggert et al. 2002 21 L.cyclotis Ghana AF527678 Eggert et al 2002 22 Bo1 Botswana AY741074 Debruyne et al. 2005 23 SamburuA Kenya KC218474 Ahlering et al 2012 24 Mo1 Mozambique AY741076 Debruyne 2005 25 UG1 Uganda AY741323 Debruyne 2005 26 SA3 South Africa AY741320 Debruyne 2005

88 Discussion

This study presents the use of both mitochondrial DNA and nuclear markers to uncover the extent of genetic subdivision among African savanna elephants in central and southern Tanzania. Several studies such as (Ahlering et al., 2012; Eggert et al., 2008; Epps et al, 2011; Ishida et al., 2011; Nyakaana et al., 2002; Silvester Nyakaana & Arctander, 1999; Roca et al., 2015) have been conducted on the population genetics of African elephants. Fine-scale population genetic structure analyses combining mitochondrial and nuclear markers, have been done in Uganda (Nyakaana & Arctander, 1999) and Kenya (Okello et al., 2008). In this study, we use the same approach to ask whether mitochondrial and nuclear markers recapitulate the same genetic scenarios between the greater Ruaha and Selous ecosystems. Our genetic data strongly show that elephants from Ruaha and Selous are divided into two divergent mtDNA lineages: savanna-wide and southeast savanna subclades (Figure 3-5). Similarly, nuclear loci data show significant population differentiation between Ruaha and Selous. Even within the Selous GR, between Matambwe and Kingupira sectors, we detected two subpopulations, implying limited gene flow within the reserve. A small proportion of admixture was observed in Matambwe sector from the STRUCTURE analysis (Figure 3-2), and the pairwise FST values were lower between Matambwe and Ruaha than Kingupira and Ruaha using microsatellite loci. Male-mediated dispersal has been documented among elephant populations in both Uganda (Nyakaana & Arctander, 1999) and Kenya (Okello et al., 2008). In both studies, there was a lack of congruence between mitochondrial DNA and nuclear-based variation. Mitochondrial DNA data showed significant differentiation, whereas nuclear data showed low genetic subdivision between populations (Silvester Nyakaana & Arctander, 1999; Okello et al., 2008). This difference between markers was attributed to differences in the modes of mutations between mitochondrial and nuclear markers. Female elephants stay in family groups while males are ejected from groups after sexual maturity. The home range 2 for elephant family groups is between 15 and 52 km (Douglas‐Hamilton, 1973). Thus, mitochondrial markers would likely remain restricted to specific localities (Nyakaana and

89 Artander, 1999). Although we expected to see differences between haplotype distributions between Ruaha and Selous, there was only one haplotype shared between Ruaha (carried by one individual) and Selous (Figure 3-7 & Table 3-3). In Ruaha, 84% of elephants carried a haplotype in savanna-wide subclade while none of the Selous elephants had a haplotype in that clade. We found no significant correlation between geographic distance and genetic distance for mtDNA marker (r=-0.1946, p=0.359) and nuclear loci (r=0.9006, p=0.1630). Thus, the observed subdivision among the population could be explained by other factors, such as the presence of the Eastern Arc Mountains. A recent study on lions (Panthera leo) in Tanzania uncovered significant genetic differentiation between lions of Selous GR and Ruaha (Smitz et al. 2018). This differentiation may be a combined effect of both anthropogenic and environmental/climatic factors (Smitz et al., 2018). The presence of the Eastern Arc Mountain chains associated with the land use patterns may present a significant biogeographical barrier to lion dispersal (Smitz et al., 2018). Phylogenetic analyses based on mitochondrial DNA of sable antelope (hipotragus niger) identified unexpected clear, distinct lineages between Ruaha and Selous which was attributed to the presence of the Eastern Arc Mountains (Pitra et al., 2002). Furthermore, Pitra et al. (2002) found that the initial allopatric fragmentation is geographically consistent with the discontinuous distribution of miombo habitats inside and outside of this montane circle in East Africa. There is also a clear difference in the vegetation cover between the western (Ruaha) and eastern (Selous) side of the Eastern Arc mountains. The Ruaha ecosystem dominated by savanna vegetations while the Selous ecosystem is dominated by miombo woodlands. Our study also, showed a similar pattern of elephant divergence, particularly from mitochondrial DNA haplotypes. 84% of Ruaha elephants carried a haplotype which was not found in Selous, but some Ruaha elephants had haplotypes from Selous. This suggests unidirectional colonization from Selous to Ruaha. Although the microsatellite data show significant genetic differentiation between Ruaha and Selous, there is evidence of gene flow between them. It is likely that only males disperse between the Ruaha and Selous and not females’ groups based on the mtDNA data. Thus, our study found no evidence for female-mediated gene flow between the Greater Ruaha and Selous ecosystem. The

90 presence of the Eastern Arc Mountains between Selous and Ruaha probably acts a barrier for the movement of elephants and other species as shown by previous studies (Pitra et al. 2002, Smitz et al., 2018). A similar pattern was observed in the distribution of mitochondrial DNA haplotypes in northern Tanzania which were separated by the rift valley (Ahlering et al., 2012 and Lohay, et al. chapter 2). Elephants west of the rift carry different haplotypes from the east. However, in northern Tanzania, there were more shared haplotypes between the western and eastern side of the rift valley than between Ruaha and Selous that are separated by the Eastern Arc Mountains. Elephants can climb relatively steep slopes despite the rigidity of their other joints (Lindsay & Lee, 2006). However, elephants avoid areas with steep slopes (Wall et al., 2006). For that reason, mountain ranges may act as barriers for elephants (Epps et al., 2013; Wall et al., 2006). Elephants can take a risk to cross farmland such as moving between Udzungwa and Selous (Southern Tanzania Elephant -unpublished data). Elephants can negotiate relatively steep slope over the short distances, but long- distance movement over steep terrain may be restricted by energetic limitations (Wall et al., 2006). These barriers are semi-permeable to the movement of elephants (Sawyer et al., 2013). Evidence for the movement of elephants between Ruaha and Mikumi and Udzungwa Mountains National Park have been documented (Nahonyo, 2009). However, this movement does not necessarily lead to the exchange of genetic materials. It is possible that elephants disperse to these areas during a certain time of the years. The wildlife corridors connecting these ecosystems are completely blocked (Jones et al., 2012). While it is essential to restore the wildlife corridors that connect the two ecosystems, our data did not support the existence of significant female elephant gene flow between the two areas. Agricultural expansion and human immigration in the Kilombero valley (Figure 3-1) is one of the significant challenges facing wildlife corridors. A corridor between Ruaha and Selous does not seem to have been present historically based on the mitochondrial DNA analysis. Female elephants are philopatric and remain with their natal herd (Archie et al., 2007) whereas males are ejected from the herd upon sexual maturity and subsequently facilitate gene flow between herd (Archie et

91 al., 2007). Using nuclear markers, there was significant genetic differentiation but low FST between Matambwe and Ruaha suggesting some amount of gene flow (Table 3-2). Nuclear genes are not subject to the same inheritance limitations as mitochondrial DNA because males disperse nuclear markers (Brandt, 2014). Our data support male-biased gene flow and in fact so extremely male biased with little evidence for female-mediated gene flow (Nyakaana & Arctander, 1999; Okello et al., 2008). We also detected unidirectional colonization of elephants from Selous to Ruaha because we found 16% of Ruaha samples carried two haplotype (one haplotype was shared with the Selous but another haplotype was unique to Ruaha) in subclade which is found in Selous, but no elephants from Selous carried haplotypes found in Ruaha. We question whether the corridors between Ruaha and Selous were essential for Selous or Ruaha elephants. Our findings are supported by other studies on sable (Pitra et al., 2002) and lions (Smitz et al., 2018) that found significant genetic differentiation between Ruaha and Selous. For Sable antelope, the Selous population formed the same clade with the South African populations. Similarly, our phylogenetic analysis revealed that a haplotype from South Africa (Debruyne, 2005) clustered with the Selous subclade. The distribution of haplotypes from southeast savanna clades reduces from South to northern Tanzania (Lohay, chapter 2). Likewise, elephants carrying haplotypes in the Forest (F) clade are only found in Northern Tanzania and not in Ruaha and Selous. Thus, at least two groups of elephants colonized Tanzania, one from Southern African, another one from North-west Tanzania, likely they had their origin from the Democratic Republic of Congo (Lohay et al., chapter 2). Some savanna elephants carrying the F-clade mtDNA are a result of hybridization events between the forest females and savanna males after the two species diverged more than 500,000 years ago (Ishida et al. 2011). Our genetic analyses suggest limited gene flow between Ruaha and Selous using both mitochondrial and nuclear markers. The chain of the Eastern Arc Mountains seems to play a significant role in the subdivision of the mitochondrial DNA haplotypes and low gene flow between elephants.

92 Implications for conservation Maintaining genetic connectivity between protected areas is crucial for long term survival of species. Unfortunately, the corridors connecting the Ruaha and Selous ecosystems are entirely blocked. Strategies for restoration are needed (Jones et al., 2012). However, female elephants may have been isolated long before the current habitat fragmentation caused by humans. Lions and sable populations from Ruaha have been isolated for a long time too. Thus, restoration of corridors may facilitate the gene flow between these ecosystems, but more detailed genetic research needs to be conducted to cover the entire Selous ecosystem and Mozambique populations to reveal the most important corridors. Luckily, the wildlife corridor between Selous Game Reserve and Niassa Reserve in Mozambique was established in 1998 (Baldus and Hahn, 2009). This corridor is probably the most important corridor for elephants and other wildlife than the Ruaha-Selous corridors. Genetic evidence from Sable antelope and Lions supports that the Selous populations are genetically more closely related to Southern Africa than Ruaha. Our samples came from a small subarea of these large ecosystems, and there could be different subpopulations further south that we did not sample. Therefore, it is possible that we missed evidence of more significant gene exchange between Ruaha and Selous. Future studies should be conducted to cover the entire Selous ecosystem and Niassa Game Reserve in Mozambique.

93 References

Ahlering, M. A., Eggert, L. S., Western, D., Estes, A., Munishi, L., Fleischer, R.,Maldonado, J. E. (2012). Identifying Source Populations and Genetic Structure for Savannah Elephants in Human-Dominated Landscapes and Protected Areas in the Kenya-Tanzania Borderlands. PLoS ONE, 7(12), e52288. https://doi.org/10.1371/journal.pone.0052288 Archie, E. A., Hollister-Smith, J. A., Poole, J. H., Lee, P. C., Moss, C. J., Maldonado, J. E.,Alberts, S. C. (2007). Behavioural inbreeding avoidance in wild African elephants. Molecular Ecology, 16(19), 4138–4148. https://doi.org/10.1111/j.1365- 294X.2007.03483.x Archie, E. A., Moss, C. J. and Alberts, S. C. (2003). Characterization of tetranucleotide microsatellite loci in the African Savannah Elephant (Loxodonta africana africana), Molecular Ecology Notes, 3(2): 244–246. doi: 10.1046/j.1471- 8286.2003.00412.x. Arif, I. A. and Khan, H. A. (2009). Molecular markers for biodiversity analysis of wildlife animals : a brief review’, Animal Biodiversity and Conservation, 32(1): 9–17. doi: 10.1007/s10722-009-9526-z. Baldus, R. D. and Hahn, R. (2009). The Selous - Niassa Wildlife Corridor in Tanzania: Biodiversity conservation from the grassroots. Practical experiences and lessons from integrating local communities into trans-boundary natural resources management, CIC Technical Publication No.6, p. 48. Available at: http://cic- sustainable-hunting-worldwide.org/publications/transboundary.pdf. Beier, P., & Gregory, A. J. (2012). Desperately Seeking Stable 50-Year-Old Landscapes with Patches and Long, Wide Corridors. PLoS Biology, 10(1), e1001253. https://doi.org/10.1371/journal.pbio.1001253 Bohonak, A. J. (2002). IBD (Isolation by Distance): A Program for Analyses of Isolation by Distance, The Journal of Heredity, (93):153–154.

94 Brandt, A. L. et al. (2014). The elephants of gash-Barka, eritrea: Nuclear and mitochondrial genetic patterns, Journal of Heredity, 105: 82–90. doi: 10.1093/jhered/est078. Caro, T., Jones, T. and Davenport, T. R. B. (2009) ‘Realities of documenting wildlife corridors in tropical countries’, Biological Conservation, 142(11): 2807–2811. doi: 10.1016/j.biocon.2009.06.011. Comstock, K. E., Wasser, S. K. and Ostrander, E. A. (2000). Isolation and characterization of polymorphic microsatellite markers in Abudefduf luridus ( Pisces : Pomacentridae ), Molecular Ecology, 9: 993–1011. doi: 10.1046/j.1365- 294x.2000.00939-8.x. Debruyne, R. (2005). A case study of apparent conflict between molecular phylogenies: the interrelationships of African elephants, Cladistics, 21: 31–50. Douglas-Hamilton, I. (1973). On the ecology and behaviour of the Lake Manyara elephants, African Journal of Ecology, 11(3–4): 401–403. doi: 10.1111/j.1365- 2028.1973.tb00101.x. Douglas-Hamilton, I., Krink, T., & Vollrath, F. (2005). Movements and corridors of African elephants in relation to protected areas. Naturwissenschaften, 92(4), 158– 163. https://doi.org/10.1007/s00114-004-0606-9 Eggert, Lori S et al. (2000). Polymorphic microsatellite DNA markers in the African elephants ( Loxodonta africana) and their use in the Asian elephant (Elephas maximus), Molecular Ecology, 9: 2155–2234. Eggert, L. S., Maldonado, J. E., & Fleischer, R. C. (2005). Nucleic Acid Isolation from Ecological Samples—Animal Scat and Other Associated Materials. In Methods in Enzymology, 395: 73–82, https://doi.org/10.1016/S0076-6879(05)95006-4 Eggert, L. S., Rasner, C. A. and Woodruff, D. S. (2008). The evolution and phylogeography of the African elephant inferred from mitochondrial DNA sequence and nuclear microsatellite markers’, Hungarian Quarterly, 49: 1993– 2006. doi: 10.1098/rspb.2002.2070. Epps, C. W. et al. (2011). An empirical evaluation of the African planning in East Africa, Diversity and Distributions, 17: 603–612. doi: 10.1111/j.1472-4642.2011.00773.x

95 Epps, C. W. et al. (2013). Quantifying past and present connectivity illuminates a rapidly changing landscape for the African elephant, Molecular Ecology, 22(6):1574– 1588. doi: 10.1111/mec.12198. Excoffier, L., Laval, G. and Schneider, S. (2005). Arlequin (version 3. 0): An integrated software package for population genetics data analyses, Evolutionary Bioinformatics Online, 1: 47–50. doi: 10.1177/117693430500100003. Fahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity, Annual Review of Ecology, Evolution, and Systematics, 34(1): 487–515. doi: 10.1146/annurev.ecolsys.34.011802.132419. Falush, D., Stephens, M. and Pritchard, J. K. (2003). Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies, Genetics, 164(4): 1567–1587. doi: 10.1111/j.1471-8286.2007.01758.x. Falush, D., Stephens, M. and Pritchard, J. K. (2007). Inference of population structure using multilocus genotype data: Dominant markers and null alleles, Molecular Ecology Notes, 7(4), pp. 574–578. doi: 10.1111/j.1471-8286.2007.01758.x. Francis, R. M. (2017). Pophelper: an R package and web app to analyse and visualize population structure, Molecular Ecology Resources, 17(1): 27–32. doi: 10.1111/1755-0998.12509. Galanti, V., Preatoni, D., Martinoli, A., Wauters, L. A., & Tosi, G. (2006). Space and habitat use of the African elephant in the Tarangire–Manyara ecosystem, Tanzania: Implications for conservation. Mammalian Biology, 71(2), 99–114. https://doi.org/10.1016/j.mambio.2005.10.001 Goudet, J. (1995). FSTAT (version 1.2): A computer Program to calculate F-Statistics, Journal of Heredity, 86(6): 485–86. Hale, M. L., Burg, T. M. and Steeves, T. E. (2012). Sampling for Microsatellite-Based Population Genetic Studies: 25 to 30 Individuals per Population Is Enough to Accurately Estimate Allele Frequencies, PLoS ONE, 7(9). doi: 10.1371/journal.pone.0045170.

96 Hauf, J., Waddell, P.J., Chalwatzis, N., Joger, U., Zimmermann, F.K., 1999. The complete mitochondrial genome sequence of the African elephant (Loxodonta africana), phylogenetic relationships of Proboscidea to other mammals, and D- Loop heteroplasmy. Zoology, 102: 184–195. Hema, E. M. et al. (2017). First assessment of age and sex structures of elephants by using dung size analysis in a West African savannah, European Journal of Ecology, 3(1), pp. 1–8. doi: 10.1515/eje-2017-0001. Henle, K., Davies, K. F., Kleyer, M., Margules, C., & Settele, J. (2004). Predictors of species sensitivity to fragmentation. Biodiversity and Conservation, 13, 207–251 Hollister-Smith, J. A., Poole, J. H., Archie, E. A., Vance, E. A., Georgiadis, N. J., Moss, C. J., & Alberts, S. C. (2007). Age, musth and paternity success in wild male African elephants, Loxodonta africana. Animal Behaviour, 74(2): 287–296. https://doi.org/10.1016/j.anbehav.2006.12.008 Ishida, Y., Oleksyk, T. K., Georgiadis, N. J., David, V. A., Zhao, K., Stephens, R. M., Roca, A. L. (2011). Reconciling Apparent Conflicts between Mitochondrial and Nuclear Phylogenies in African Elephants. PLoS ONE, 6(6), e20642. https://doi.org/10.1371/journal.pone.0020642 Ishida, Y., Georgiadis, N. J., Hondo, T., & Roca, A. L. (2013). Triangulating the provenance of African elephants using mitochondrial DNA. Evolutionary Applications, 6(2), 253–265. https://doi.org/10.1111/j.1752-4571.2012.00286.x Jones, T., Bamford, A. J., Ferrol-Schulte, D., Hieronimo, P., McWilliam, N., & Rovero, F. (2012). Vanishing Wildlife Corridors and Options for Restoration: A Case Study from Tanzania. Tropical Conservation Science, 5(4): 463–474. https://doi.org/10.1177/194008291200500405 Jones, T., Cusack, J. J., Pozo, R. A., Smit, J., Mkuburo, L., Baran, P., Foley, C. (2018). Age structure as an indicator of poaching pressure: Insights from rapid assessments of elephant populations across space and time. Ecological Indicators, 88(November 2017): 115–125. https://doi.org/10.1016/j.ecolind.2018.01.030 Kinuthia, J., Harper, C., Muya, S., Kimwele, C., Alakonya, A., Muigai, A.,Gatebe, E. (2015). The selection of a standard STR panel for DNA profiling of the African

97 elephant (Loxodonta africana) in Kenya. Conservation Genetics Resources, 7(2): 305–307. https://doi.org/10.1007/s12686-014-0366-6 Leigh, J. W. and Bryant, D. (2015). POPART: Full-feature software for haplotype network construction, Methods in Ecology and Evolution, 6(9), pp. 1110–1116. doi: 10.1111/2041-210X.12410. Lindsay, W. K. and Lee, P. C. (2006). Habitat Use, Diet Choice, and Nutritional Status in Female and Male Amboseli Elephants, University of Chicago Press, USA Morgan, B. J. & Lee, P. C. (2003). Forest elephant (Loxodonta Africana cyclotis) stature in the Reserve de Faune du Petit Loango, Gabon. Journal of Zoology (Lond.) 259:337–344. Nahonyo, C. L. (2009). Feasibility Study on Elephant Movement between the Greater Ruaha Ecosystem and Selous Ecosystem in Central Eastern, Tanzania. Newmark, W. D. (1996). Insularization of Tanzanian Parks and the Local Extinction of Large Mammals’, Conservation Biology, 10(6): 1549–1556. doi: 10.1046/j.1523- 1739.1996.10061549.x. Newmark, W. D. (2008). Isolation of African protected areas, Frontiers in Ecology and the Environment, 6(6): 321–328. doi: 10.1890/070003. Newmark, W. D. and McNeally, P. B. (2018) ‘Impact of habitat fragmentation on the spatial structure of the Eastern Arc forests in East Africa: implications for biodiversity conservation’, Biodiversity and Conservation. Springer Netherlands, 27(6): 1387–1402. doi: 10.1007/s10531-018-1498-x Niko, Balkenhol, Samweli A. Cushman, Andrew Storfer and Lisette P. Waits, (2016), Introduction to Landscape Genetics- Concepts, Methods, Applications, John Wiley & Sons, Incorporated https://doi.org/10.1002/9781118525258.ch01 Nyakaana, S. et al. (2005). Six new polymorphic microsatellite loci isolated and characterized from the African savannah elephant genome, Molecular Ecology Notes, 5(2005): 223–225. doi: 10.1111/j.1471-8286.2005.00885.x. Nyakaana, S. and Arctander, P. (1999). Population genetic structure of the African elephant in Uganda based on variation at mitochondrial and nuclear loci:

98 Evidence for male-biased gene flow, Molecular Ecology, 8(7): 1105–1115. doi: 10.1046/j.1365-294X.1999.00661.x. Nyakaana, S., Arctander, P. and Siegismund, H. R. (2002). Population structure of the African savannah elephant inferred from mitochondrial control region sequences and nuclear microsatellite loci, Heredity, 89(2): 90–98. doi: 10.1038/sj.hdy.6800110. Okello, J. B. A., Wittemyer, G., Rasmussen, H. B., Douglas-Hamilton, I., Nyakaana, S., Arctander, P., & Siegismund, H. R. (2005). Noninvasive Genotyping and Mendelian Analysis of Microsatellites in African Savannah Elephants. Journal of Heredity, 96(6), 679–687. https://doi.org/10.1093/jhered/esi117 Okello, J. B. A., Masembe, C., Rasmussen, H. B., Wittemyer, G., Omondi, P., Kahindi, O., Siegismund, H. R. (2008). Population Genetic Structure of Savannah Elephants in Kenya: Conservation and Management Implications. Journal of Heredity, 99(5), 443–452. https://doi.org/10.1093/jhered/esn028 Van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M., & Shipley, P. (2004). MICRO- CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4(3), 535–538. https://doi.org/10.1111/j.1471-8286.2004.00684.x Peakall, R. and Smouse, P. E. (2012). GenALEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research-an update, Bioinformatics, 28(19), pp. 2537–2539. doi: 10.1093/bioinformatics/bts460. Pitra, C., Hansen, A. J., Lieckfeldt, D., & Arctander, P. (2002). An exceptional case of historical outbreeding in African sable antelope populations. Molecular Ecology, 11(7), 1197–1208. https://doi.org/10.1046/j.1365-294X.2002.01516.x Pritchard, J. K. J., Stephens, M. M. and Donnelly, P. (2000). Inference of population structure using multilocus genotype data, Genetics, 155(2): 945–959. doi: 10.1111/j.1471-8286.2007.01758.x. Raymond, M. and Rousset, F. (1995). GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal of Heredity, 86: 248–249.

99 Riggio, J. and Caro, T. (2017). Structural connectivity at a national scale : Wildlife corridors in Tanzania’, PLoS One, 12(11): 1–16. Roca, A. L., Ishida, Y., Brandt, A. L., Benjamin, N. R., Zhao, K., & Georgiadis, N. J. (2015). Elephant Natural History: A Genomic Perspective. Annual Review of Animal Biosciences, 3(1), 139–167. https://doi.org/10.1146/annurev-animal- 022114-110838 Rousset, F. (2008). GENEPOP’007: A complete re-implementation of the GENEPOP software for Windows and Linux, Molecular Ecology Resources, 8(1): 103–106. doi: 10.1111/j.1471-8286.2007.01931.x. Rozas, J., Ferrer-Mata, A., Sánchez-DelBarrio, J. C., Guirao-Rico, S., Librado, P., Ramos-Onsins, S. E., & Sánchez-Gracia, A. (2017). DnaSP 6: DNA Sequence Polymorphism Analysis of Large Data Sets. Molecular Biology and Evolution, 34(12), 3299–3302. https://doi.org/10.1093/molbev/msx248 Sawyer, H., Kauffman, M. J., Middleton, A. D., Morrison, T. A., Nielson, R. M., & Wyckoff, T. B. (2013). A framework for understanding semi-permeable barrier effects on migratory ungulates. Journal of Applied Ecology, 50(1), 68–78. https://doi.org/10.1111/1365-2664.12013 Smitz, N., Jouvenet, O., Ambwene Ligate, F., Crosmary, W.-G., Ikanda, D., Chardonnet, P., Michaux, J. R. (2018). A genome-wide data assessment of the African lion (Panthera leo) population genetic structure and diversity in Tanzania. PLOS ONE, 13(11), e0205395. https://doi.org/10.1371/journal.pone.0205395 Soule, M. E. (1979). Heterozygosity and developmental stability: another look. Evolution, 33(1), 396–401 Spong, G. (2002). Space use in lions, Panthera leo, in the Selous Game Reserve: Social and ecological factors, Behavioral Ecology and Sociobiology, 52(4): 303–307. doi: 10.1007/s00265-002-0515-x. Swofford, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4.0b10. Sinauer, Sunderland, MA. Thomas,

100 Takezaki, N., Nei, M. and Tamura, K. (2010) ‘POPTREE2 : Software for Constructing Population Trees from Allele Frequency Data and Computing Other Population Statistics with Windows Interface’, Molecular Biology and Evolution, 27(4), pp. 747–752. doi: 10.1093/molbev/msp312. TANAPA website accessed February 12, 2019, http://www.tanzaniaparks.go.tz/ TAWIRI (2015) .Population Status of Elephant in Tanzania 2014.TAWIRI Aerial Survey Report’. Villesen, P. 2007. FaBox: an online toolbox for fasta sequences. Molecular Ecology Notes 7:965–968 Wall, J., Douglas-Hamilton, I. and Vollrath, F. (2006) ‘Elephants avoid costly mountaineering’, Current Biology, 16(14): 527–R529. doi: 10.1016/j.cub.2006.06.049. Wilcox, Bruce, A. and Murphy, Dennis, D. (1985). Conservation Strategy: The Effects of Fragmentation on Extinction, American Society of Naturalists, 125(6): 879–887. Wittemyer, G. et al. (2007). Social dominance, seasonal movements, and spatial segregation in African elephants : a contribution to conservation behavior, Behavioral Ecology and Sociobiology, 61(12): 1919–1931. doi: 10.1007/s00265- 007-0432-0.

101 Supporting materials

Table 2-S3. Mean likelihood estimate log probability for K=1-10 and the highest value of delta K (k=3) using the Evanno method

102 : An accurate molecular method to sex elephants using PCR amplification of Amelogenin gene

George G. Lohay

Abstract

The use of molecular methods to identify the sex of elephants from non-invasive samples is essential for studies of population dynamics and population genetics. We designed a new technique for sex identification in elephants using Amelogenin (AMEL) genes. The X-Y homologs of AMEL genes are suitable for sex determination in pigs and some species in the family Bovidae. The use of AMEL genes was more successful than previous methods that relied on genes found exclusively on Y-chromosomes, such as SRY, to distinguish males from females. We designed a common forward primer and two reverse primers for X- and Y-specific AMEL genes to obtain 262bp and 196bp PCR amplicons from Y and X genes, respectively. We tested the primers for the identification of the sex of 132 elephants. Using our approach, the sex of 126 individuals (95.45%) matched the reference samples, while 6 (4.54%) did not match. This discrepancy observed was due to high grass content in which reduces the ability to accurately sex young individuals in the field. Through our stool samples results, we have shown that the use of only three primers for AMELX/Y provides a highly accurate PCR-based method for sex identification in elephants. The method is fast and shows more success than the SRY system by avoiding the inherent ambiguities of the previous PCR-based methods that made it difficult to distinguish between female samples and failed amplification reactions. Our sex identification method is non-invasive, and can be applied in population genetic studies and forensics tests with elephant species.

103 Introduction

Molecular genetics is a powerful tool with many potential applications for conservation biology and wildlife management (Ortega et al. 2004). One of these applications includes the development of simple and accurate methods to determine the sex of wildlife from non-invasive samples. African savannah elephants face population decline as a result of illegal poaching for their ivory. An accurate method to sex elephants non- invasively becomes crucial to determine sex ratio for ecological studies and for the identification of the forensic samples of elephant origin. Several methods have been developed that use PCR amplification of sex determination regions on sex chromosomes (Fernando and Melnick 2001; Gupta, Thangaraj, and Singh 2006; Vidya et al. 2007; Ahlering et al. 2011). A method of sex identification for elephants, that has been widely used on non- invasive sample types, is based on three pairs of PCR primers for amplification of the genes AMELY2, SRY1, and PLP1 (Ahlering et al. 2011). The method requires conducting multiplex PCR such that females exhibit a single band (PLP1) and males exhibit all three bands owing to the presence of Y-chromosome genes in males only. Although this method was reliable and more efficient than the previous methods (Ahlering et al. 2011), there were some incidences where amplification was not successful. Tests based on SRY amplification can be problematic as non-amplification of the Y-marker represents either female or a failed PCR (Peppin et al. 2010). Although this method (Ahlering et al. 2011) used two Y-markers (AMELY2 and SRY1) for males, there were occasions when one of the markers would fail to amplify, resulting in ambiguous results. A more accurate method was designed based on the existing differences in sequence length between X- and Y-specific Amelogenin (AMEL) genes in a variety of species (Weikard, Pitra, and Kuhn 2006). The X-Y homologs of AMEL genes (AMELX and AMELY) have been shown to be suitable for sex determination at the molecular level and have been used to identify sex in the family Bovidae (Weikard, Pitra, and Kuhn 2006) and in pigs (Sembon et al. 2008). A single primer pair can be designed to amplify both chromosome-specific forms of the genes. The aim of this paper was to design primers from

104 AMEL genes specific for elephants to accurately determine the sex of elephants from non- invasive samples.

Materials and Methods

Fresh elephant fecal samples were collected from external parts of dungs and stored in Queen's College buffer solution buffer (20% DMSO, 0.25 M EDTA, 100mM Tris, pH7.5, saturated with NaCl). We collected 132 fecal samples from the Serengeti and Tarangire-Manyara ecosystems in northern Tanzania (Table 1). Four tissue samples (2 males and 2 females) from elephants who died naturally were also included as positive controls for known sex. While in the field, an individual’s sex was determined through observations of genitalia, their overall body size, and the shape of the head and the thickness of the tusks, especially for adults. Photographs were taken for each individual to confirm their sexes. Through this close field observations, the sex of all 132 individuals were identified while collecting their fecal samples and these were used to test the new method. DNA was isolated using company protocols with DNA extraction kits (QIAamp DNA Stool Mini Kit) with minor modifications (Eggert, Maldonado, and Fleischer 2005). Three PCR primers were designed using SnapGene® software (GSL Biotech LLC), based upon the AMELX (Reference Sequence: NW_003573459) and AMELY (Genebank accession AY823322) in Loxodonta Africana by a common forward primer “AMELXY” 5'-TTCTGGAATCTGGTTTGAGGCT-3', X-specific “AMELX-R”5'-ATC TTT ACAACA AAA CAA TTG TTA ACC ATG CTC-3' and Y-specific “AMELY-R” 5'- TCAGATTCA GAGTTTCCT TCATGC AGTAG-3' reverse primers. Three primers would amplify both X and Y if present (i.e. males) yielding two DNA bands of different size, whereas females would generate only one band. The PCR reaction was performed the initial polymerase activation step at 95°C for 3 min, denaturation at 95°C for 30 sec, annealing temperature at 56°C for 45 seconds, extension at 72°C for 30 seconds for 35 cycles. Each PCR mixture contained 3μl of 5X Green GoTaq reaction buffer (promega) ,a final concentration of 0.67μM of forward primers and 0.33 μM for each of

105 the reverse primers, 0.13μM of dNTP (Quanta bio), Bovine serum albumin (BSA) 0.1μg/μl, and 3μl of DNA template of unknown concentration in a 15μl volume. For each PCR reaction with test samples, we also ran a negative control (no DNA) and a positive control using DNA from known tissue samples. 6μl of the PCR product was used for electrophoresis in Tris-Acetate EDTA running buffer at 120V for 45 min on a 2% agarose gel stained with GelRed (Biotium). Experiments were repeated twice to verify the results. Samples that gave ambiguous results were tested three times under the standard conditions described above. To confirm the origins of the PCR products, amplicons from at least three samples for each primer pair were sequenced and performed a BLASTn alignment search on NCBI database to confirm the sequences.

Results and Discussion

A total of 126 dung samples were successfully sexed using our AMELX/Y sexing method (Table 1). By this PCR-based method, 6 of the 132 individuals identified in the field as being of known sex, were shown to be misidentified (4.45% false-positive rate). Four of these were young elephants with age less than ten years. In the field, it is more difficult to accurately identify the sex of young elephants compared to older elephants due to limited visibility in long grass or savanna woodlands. A similar challenge was previously reported by (Jones et al. 2017). Therefore, all mismatches observed were likely due to observational errors in the field and not the failure of PCR-based sexing method.

106 Table 4-1.The number of females and males identified using AMELX/Y primers. For each location, the number of males and females are shown for reference sex and AMELX/Y sexing method.

Gender Reference sex AMELY/X sexing method Misidentification Females 67 64 3 Males 65 62 3 Total 132 126 6

As expected, known male samples produced two DNA amplification bands whereas only one DNA band was amplified from female samples (Figure.1). A similar method was used to identify the sex of species in family Bovidae (Weikard, Pitra, and Kuhn 2006) and pigs (Sembon et al. 2008) using tissue samples. Because we used DNA derived from scats, which were significantly degraded, we chose PCR priming sites that produced short fragment sizes of 262bp and 196bp for AMELY and AMELX respectively.

Figure 4-1. Agarose gel electrophoresis (2% agarose) of PCR product using AMELX and AMELY primer sets for Loxodonta africana.

Males (M) show two PCR bands (AMELX 196 bp and AMELY 262bp) whereas females (F) show only one PCR band (AMELX 196 bp). Lane M is 100bp DNA ladder (GBiosciences) and water was used as a negative control. An arrow indicates a 500bp size of the ladder. Lanes 1, 2, 3, 7 and 8 represent females and lanes 4, 5, 6, and 9 represents males

107 Our AMELX/Y sexing method is particularly useful for partially degraded DNA samples, such as those derived from fecal samples, because the amplicons produced are relatively short. The method also allows for the direct positive scoring of females. When the DNA quality is low, there is a chance of high PCR failure rates. Using the previous multiplexing method of Ahlering et al. (2011), we unexpectedly found a high number of mismatches from the reference samples, which made it difficult to distinguish males from females. A similar challenge was experienced in other studies. For example, a recent study could not determine the sex of some individuals (16%) due to repeated failure to discriminate DNA bands as either males or females (Lobora et al. 2017). Such a high failure rate results in a large amount of missing data needed to understand important aspects of population structure, such as effective population size. Our technique for identifying the sex of elephants can be used in a wide variety of studies aimed at the conservation of elephants, including those that address human-elephant conflicts, population dynamics, and anti-poaching efforts. Future improvements possible for this method, such as an automated assay, will further facilitate genetic analysis of free-ranging elephant populations and could make significant contributions to the management and conservation of other threatened species. In this paper, we described an improved technique to sex elephants using a non- invasive method. To our knowledge, this is the first work to report the use of this AMELX/Y method for sexing elephants based on DNA collected from fecal samples in the field. The use of this method will help researchers save time and resources and can be applied to other mammalian species. Using our method, we could easily distinguish males from females, thus the method is more reliable than the previous approached.

108 References

Ahlering, Marissa A, Frank Hailer, Melissa T Roberts, and Charles Foley. 2011. A Simple and Accurate Method to Sex Savannah, Forest and Asian Elephants Using Noninvasive Sampling Techniques, Molecular Ecology Resources 11 (5): 831–34. https://doi.org/10.1111/j.1755-0998.2011.03030.x. Eggert, Lori S., Jesús E. Maldonado, and Robert C. Fleischer. 2005. Nucleic Acid Isolation from Ecological Samples - Animal Scat and Other Associated Materials, Methods in Enzymology 395: 73–87. https://doi.org/10.1016/S0076-6879(05)95006- 4. Fernando, P, and D Melnick. 2001. Molecular Sexing Eutherina Mammals, Molecular Ecology Notes 1 (4): 350–53. https://doi.org/10.1046/j.1471-8278.2001.00112.x. Gupta, Sandeep K., Kumarasamy Thangaraj, and Lalji Singh. 2006. A Simple and Inexpensive Molecular Method for Sexing and Identification of the Forensic Samples of Elephant Origin, Journal of Forensic Sciences 51 (4): 805–7. https://doi.org/10.1111/j.1556-4029.2006.00154.x. Jones, Trevor, Jeremy J Cusack, Rocío A Pozo, Josephine Smit, Lameck Mkuburo, Paul Baran, Alex L Lobora, Simon Mduma, and Charles Foley. 2018. Age Structure as an Indicator of Poaching Pressure: Insights from Rapid Assessments of Elephant Populations across Space and Time, Ecological Indicators, 88(2018) 115–25. https://doi.org/10.1016/j.ecolind.2018.01.030. Lobora, Alex L, Cuthbert L Nahonyo, Linus K Munishi, Tim Caro, Colin M Beale, and Lori S Eggert, 2018. Incipient Signs of Genetic Differentiation among African Elephant Populations in Fragmenting Miombo Ecosystems in South-Western Tanzania Charles Foley, African Journal of Ecology, May: 1–40. https://doi.org/10.1111/aje.12534. Ortega, Jorge, María Del Rosario Franco, Brice A. Adams, Katherine Ralls, and Jesús E. Maldonado. 2004. A Reliable, Non-Invasive Method for Sex Determination in the Endangered San Joaquin Kit Fox (Vulpes Macrotis Mutica) and Other

109 Canids.Conservation Genetics 5 (5): 715–18. https://doi.org/10.1007/s10592-003- 1862-5. Peppin, Lindsay, Ross McEwing, Rob Ogden, Robert Hermes, Cindy Harper, Alan Guthrie, and Gary R. Carvalho. 2010. Molecular Sexing of African Rhinoceros. Conservation Genetics 11 (3): 1181–84. https://doi.org/10.1007/s10592-009-9912-2. Sembon, Shoichiro, Shun-ichi Suzuki, Dai-ichiro Fuchimoto, Masaki Iwamoto, Tatsuo Kawarasaki, and Akira Onishi. 2008. Sex Identification of Pigs Using Polymerase Chain Reaction Amplification of the Amelogenin Gene, Zygote (Cambridge, England) 16 (4): 327–32. https://doi.org/10.1017/S0967199408004826. Vidya, T. N.C., Surendra Varma, Nguyen X. Dang, T. Van Thanh, and R. Sukumar. 2007. Minimum Population Size, Genetic Diversity, and Social Structure of the Asian Elephant in Cat Tien National Park and Its Adjoining Areas, Vietnam, Based on Molecular Genetic Analyses.” Conservation Genetics 8 (6): 1471–78. https://doi.org/10.1007/s10592-007-9301-7. Weikard, Rosemarie, Christian Pitra, and CHRISTA Kuhn. 2006. Amelogenin Cross- Amplification in the Family Bovidae and Its Application for Sex Determination, Molecular Reproduction and Development 73: 1333–37. https://doi.org/10.1002/mrd.

110 : Assessment of age and sex structure of African savanna elephants in the Serengeti ecosystem using a novel fecal-centric method

George G. Lohay, Anna B. Estes and Douglas R. Cavener

Abstract

Sex and age structure can tell us a lot about a population of animals. New methods for assessing sex and age structure in populations, especially if they are non-invasive can be very useful. We developed a fecal-centric method for assessing sex and age structure of elephants. We validated this method using a known rapid demographic assessment (RDA) method. We applied the method to areas with different histories of poaching and other human disturbances within the Serengeti ecosystem in northern Tanzania. We compared age and sex structure between Serengeti National Park (SNP), Maswa Game Reserve (MGR) and Ngorongoro Conservation Areas (NCA). We estimated the age of elephants by measuring the circumference of at least three dung boli per individual and determined sex using the AMELX/Y marker. We also used visual observations and photographs to estimate the age of elephants and grouped them in seven age classes using the RDA approach. We wanted to determine if the sex ratio in different age classes deviated from the expected 1:1 ratio and whether results from the fecal centric approach are comparable with the RDA method. We found a significantly skewed sex ratio for adults in favor of males for the Serengeti National Park using both methods. However, both the new method and the RDA in Serengeti, we found a significantly skewed sex ratio in favor of females in Serengeti for adult age classes. However, in NCA the sex ratio was skewed in favor of males for individuals older than 25 years. In MGR, the sex ratio did not deviate from the expected 1:1 ratio for almost all age classes using both methods. These results are impressive because they are consistent with this new approach. We can now thus can determine the age and sex structure of elephant populations directly without observing them. With the RDA method, there could be human errors in estimating the age of

111 elephants. We propose the use of fecal-centric approaches especially in areas where there is a small population size or areas where it difficult to observe the elephants. Poaching and differential survival rates between males and females could explain the skewness of sex ratio in Serengeti. In NCA, there were more adult males than females for individuals older than 25 years probably because of the high level of protection and low poaching pressure in the NCA. We did not observe any differences in sex ratios between age classes in MGR perhaps because this subpopulation has experienced significant poaching pressure, which affected both adult males and females. Keywords: Fecal-centric, age and sex structure, AMELX/Y, Rapid demographic assessment

Introduction

African elephant populations have declined by 72% from 1.3 million in 1979 to around 350,000 today (Bartáková et al., 2018; Cerling et al., 2016; Kikoti et al., 2015; E. J. Lee et al., 2013). Poaching targets large individuals for their higher tusk sizes (Mondol et al., 2014). Although poaching for ivory primarily targets single adult males (Gobush et al., 2008), poachers also target female matriarchs as they also have large tusks and are easier to find than solitary males (Poole, 1989). Selective removal of larger adult males and females from elephant populations affects age and sex structure. For a relatively unpoached population, adult sex ratios are only slightly biased in favor of females (Poole & Thomsen, 1989). Ivory poaching has an insidious effect upon elephant populations in terms of the fundamental aspects of population biology (Barnes & Kapela, 1991) such as age structure, the proportion of tuskless cows, operational sex ratio and dependent ratio. The dependent ratio is the number of dependents an adult female can recruit, and populations that have experienced high poaching pressure tend to have lower dependent to adult female rates than unpoached populations (Jones et al., 2018). The sex ratio of heavily poached populations can be significantly skewed towards females and young elephants (Jones et al., 2018; Poole & Thomsen, 1989). Changes in the proportion of tusklessness

112 among elephant populations may also be the result of poaching (Kioko et al., 2013). For example, the proportion of tuskless individuals increased in Ruaha between 1973 and 1988 and South Lungwa National Park as a result of heavy poaching (Barnes & Kapela, 1991; Jachmann & Billiouw, 1997; Jachmann et al., 1995; Jones et al., 2018). Operational Sex Ratio (OSR) is an essential ecological factor affecting the opportunity and direction of sexual selection in animals (Janicke & Morrow, 2018). Mobley (2014) defined OSR as the proportion of sexually active males in the population (cited in Janicke & Morrow, 2018). For the elephant population, OSR is a ratio of sexually active males > 25 years to the number of sexually mature females >10 years (Jones et al., 2018). Elephant populations that have experienced high poaching pressure tend to have lower OSR. Similarly, as a result of poaching the dependent ratio decreases with increasing poaching pressure (Jones et al., 2018). Currently, age of elephants is estimated using physical attributes based on shoulder height, back length, head and body shape, and size of tusks (Moss 1996). We used a protocol established by the Amboseli elephant project to visually estimate elephant age. When estimating the age of calves for both males and females, we use a range of 1 year because calves have a higher growth rate than adults. We also compare their size relative to their mothers or older siblings within a herd. We grouped elephants into seven age classes: 0-4, 5-9, 10-14, 15-19, 20-24, 25-34, and 35-50+ (Moss 1996). The use of rapid demographic assessment is entirely based on the judgment of the observer. Although there is a protocol to follow, there could be a bias in age estimation. We developed an alternative fecal-centric method to validate the rapid demographic assessment (RDA) method. We estimated the age of elephants based on dung circumference of at least three boli (Hema et al. 2017) and used a molecular technique (AMELX/AMELY) to determine sex using fecal DNA. The fecal centric approach may be appropriate in areas where it is hard to see elephants due to landscape or small population size. We can then assess the age and sex structure indirectly. This research aimed to determine if there was a different age and sex structure of elephants between Serengeti National Park (SNP), Ngorongoro Conservation Authority (NCA), and Maswa Game Reserve (MGR) which have different land use and poaching history and also to test the

113 success of our new fecal-centric method. We hypothesize that less poaching in better- protected areas should lead to more even sex ratios. We also assessed the age and sex structure of elephants between these protected areas to determine if the adult sex ratio is significantly different from the expected 1:1 ratio.

Materials and methods

Study Areas

The greater Serengeti ecosystem comprises Ikorongo-Grumeti and Maswa Game Reserve (MGR), Loliondo Game Controlled Area, Mwiba Wildlife Ranch, and Ngorongoro Conservation Area (NCA). The Serengeti National Park (SNP), the core of the ecosystem, is highly protected (Figure 5-1). Only photographic tourism is permitted. The elephant population at SNP experiences lower poaching pressure than other populations in southern Tanzania (T. Jones et al., 2018). NCA has multiple land uses, according to which humans, livestock, and wildlife coexist in an area of 8292 km2 (Runyoro et al., 1995). Within the NCA, there is a crater with an area of 250 km2 (Estes & Small, 1981; Oates & Rees, 2013). The NCA has four main vegetation types: forest; highland shrub and grassland; bushland and woodland; and plains grasslands (Homewood & Rodgers, 1991). Due to its complex ecological and sociological interactions, NCA is divided into six land use zones (Boshe, 1997). Ngorongoro Crater and the Northern Highlands Forest Reserve zones are highly protected, with limited human activity. Human settlement is allowed in other parts of NCA, including the short grass plains in the Ndutu, Oldupai, and Endulen/Kakesio woodland zones. The areas around the montane forests are evergreen and receive rainfall between 800-1200 mm, whereas the grassland and bushland areas receive between 400-600mm per annum (Estes et al., 2006). Most mammals are concentrated in the crater, only 4% of the entire NCA (Runyoro et al., 1995). The NCA has experienced a significant increase in woody vegetation coverage between 1975 and 2000,

114 including types of forest and bushland (Niboye, 2010). At MGR, trophy hunting takes place yearly between July and December. Mwiba Wildlife Ranch, established in 2007, is an open area in Makao village, at which photographic tourism and controlled trophy hunting take place. Mwiba is a potential dispersal area for wildlife from SNP, NCA, and MGR. For this study, we combined data from MGR and Mwiba Wildlife Ranch due to their proximity.

Figure 5-1. Map of the Serengeti ecosystem in Tanzania, showing sampling location during the dry seasons of 2015-17

SNP=Serengeti National Park, MGR=Maswa Game Reserve, NCA=Ngorongoro Conservation Area, MWR=Mwiba Wildlife Ranch, LGCA=Loliondo Game Controlled Area, IGGR=Ikorongo-Grumeti Game Reserve, and WMA=Wildlife Management Area. Dots indicate locations where elephants were observed or where fecal samples were collected (Source: USGS).

115 Ngorongoro Conservation Area (NCA) is in the multiple land use area within the Greater Serengeti ecosystem. Maasai people (pastoralists) had inhabited the NCA even before the NCA was established as a protected area in 1959. Elephants have been recorded to be present in NCA from 1958 (Oates & Rees, 2013). In the late 1950s and early 1960s, Masai killed 19 rhinoceros probably to generate income in response to severe drought (Fosbrooke, 1972). Eight elephants were killed in Endulen zone which resulted in the arrest of 210 poachers in NCA (Makacha et al., 1982). In the Crater, the number of elephants has been fluctuating between 0 and 87 between 1963 and 1995 (Moehlman et al., 1997, Estes and Small 1981). Homewood and Rodgers (1991) argued that poaching and poor relationship between NCA authority and the Maasai threatened the future of the elephant population although. Runyoro et al., (1995) concluded that poaching was not a significant threat for species other than rhinoceros. Human population in the NCA has increased from 10,633 in 1954 to 87,851 in 2012 (Kijazi, 1997, Masao et al., 2014). This increase in the human population may have impacted the abundance and distribution of elephants within the NCA. Only the Crater elephants have been recorded since 1963, and therefore there is insufficient data on the age and sex structure of elephants and their distribution in the NCA. Serengeti National Park (SNP) is highly protected although it experienced high poaching pressure before the international ivory trade ban through CITES in 1989. Since then, the population has increased to 5,160 while in the NCA the current estimated number of elephants is 199 (TAWIRI, 2015). More than 50% of elephants are outside the NCA but no records are available on the distribution or the age structure of elephants in the entire NCA. Maswa Game Reserve (MGR) and Mwiba Reserve are in the southern part of the Serengeti ecosystem. Trophy hunting takes place in MGR between July and December every year, while only photographic tourism is permitted in Mwiba Reserve (MWR). The MGR was established in 1962, but Mwiba Wildlife Reserve was established in 2008 when the Makao Village granted the derivative rights to Mwiba Holdings Limited for approximately 40 000 hectares. The Serengeti elephant population increased after recovering from heavy poaching before 1989. In contrast, most populations of African elephants outside the Serengeti experienced a rapid decline. Populations that have experienced heavy poaching tend to

116 have skewed adult sex ratio in favor of females, (Poole, 1989) lower OSR and a higher proportion of tuskless individuals (Jones et al., 2018). Given the fact that the SNP is experiencing the opposite of what is happening in most other areas of Africa, we do not expect the adult sex ratio to deviate significantly from 1:1. We also predict that in the NCA, which also has low poaching pressure, the adult sex ratio will not differ significantly from the expected 1:1 ratio.

Fecal-centric Approach

We used our newly developed fecal centric approach to obtain more information on elephant population structure from the areas where it was hard to observe elephants directly. At least three fresh and intact boli circumferences were measured for each dung pile which represents a single bowel movement from a single individual. We recorded the GPS location for each sample collected. Dung size has been proven to accurately classify elephants into age groups (Hema et al., 2017; Morrison et al., 2005). Fresh elephant fecal samples were collected from external parts of the dung as this portion is rich in epithelial cells and were stored in Queen's College buffers (20% DMSO, 0.25M EDTA, 100mM Tris, pH 7.5, saturated with NaCl) to preserve the sample for molecular analysis using a PCR- based AMELX/Y sexing method. To avoid cross-contamination between samples, we collected the fecal samples before measuring the dung circumference. To test our sexing method, we collected 132 samples from individuals with known gender and approximate age to use them as controls for developing molecular methods to sex individuals. Four tissue samples (two males and two females) from elephants that died naturally were also included as positive controls for individuals with known gender. To avoid including replicate samples into the analysis, we genotyped all samples at 11 microsatellite loci and removed samples with matching multilocus genotypes.

117 DNA Isolation and Sex Determination

DNA was isolated using a commercially available DNA extraction kit (QIAamp DNA Stool Mini Kit) with minor modifications to the manufacturer’s protocol (Eggert et al., 2005). Three PCR primers were designed using SnapGene® software (GSL Biotech LLC), based on AMELX (Reference Sequence: NW_003573459) and AMELY (Genebank accession AY823322) in Loxodonta Africana by a common forward primer “AMELXY” 5'-TTCTGGAATCTGGTTTGAGGCT-3', X-specific “AMELX-R”5'-ATC TTT ACAACA AAA CAA TTG TTA ACC ATG CTC-3', and Y-specific “AMELY-R” 5'- TCAGATTCA GAGTTTCCT TCATGC AGTAG-3' reverse primers. Three primers amplify both X and Y is present (i.e., males), yielding two DNA bands of different sizes, while females generate only one band. The PCR reaction was performed with the initial polymerase activation step set at 95°C for 3 mins, denaturation at 95°C for 30 sec, the annealing temperature at 56°C for 45 sec, and the extension at 72°C for 30 sec for 35 cycles. Each PCR mixture contained 3μl of 5X Green GoTaq reaction buffer (Promega), a final concentration of 0.67μM of forward primers (and 0.33 μM for each of the reverse primers), 0.13μM of dNTP (Quanta bio), Bovine serum albumin (BSA) 0.1μg/μl, and 3μl of a fecal DNA template of unknown concentration in a 15μl volume. PCR reactions with no DNA template served as negative controls, while reactions containing DNA from tissue samples of previously characterized individuals served as positive controls. PCR products were separated by electrophoresis in a Tris-Acetate EDTA running buffer at 120V for 45 min on a 2% agarose gel stained with GelRed (Biotium). Experiments were repeated twice to verify the results. Samples that gave ambiguous results were tested three times under the standard conditions described above. To confirm the origins of the PCR products, amplicons from at least three samples for each primer pair were sequenced and a BLASTn alignment search was performed using the NCBI database to confirm the sequences.

118 Rapid Demographic Assessment Method

The RDA has been successfully used to record the age, sex, and unique physical attributes of a large number of individuals and to record their group size (Jones et al., 2018; Poole, 1989). We primarily searched for elephants using the available road networks in these protected areas. On several occasions, it was necessary to drive off the road or search for elephants on foot at NCA, with support from field assistants who knew where the elephants were most likely to be encountered. After locating the elephants, we grouped them into age groups: 0-4, 5-9, 10-14, 15-19, 20-24, 25-34, and 35-50+ years. The age estimation was based on physical attributes as described by Poole (1989). We received training in 2015 from the Amboseli elephant research project and the Tarangire elephant project, which have monitored elephants since 1972 and 1993 respectively. We recorded the sex of each and took photos for each observation to verify the individual’s identity to avoid recounting individuals. For each sighting, we recorded the GPS location as well as any signs of poaching, such as the presence of wire snares, wounded elephants, or elephants with scars that might have resulted from illegal hunting. Because illegal hunters target elephants with tusks, we recorded the number of tusked and tuskless individuals. We used a one sample test of the proportions implemented in MINITAB statistical software to determine if the sex ratios for each age class were significantly different from the expected ratio. We calculated operational sex ratio and dependent ratio using the formula below (Jones et al., 2018).

Number of sexually competing males > 25 years Operational sex ratio (OSR) = Number of available females > 10 years

Number of elephants < 10 years Dependent ratio = Number of adult females > 10 years

119 Results

Fecal-centric approach

Using AMEX/Y approach, we successfully identified the sexes of 391, 108, and 86 individuals from SNP, NCA, and MGR respectively and we grouped the individuals into five age classes, as described in Hema et al. (2017). As expected, the known male samples produced two DNA amplification bands, while only one DNA band was amplified from the female samples (Figure 4-1). In SNP, the sex ratio of elephants older than 18 years was significantly skewed in favor of females whereas, in the NCA, the sex ratio was biased in favor of males. In MGR, there was no difference in the sex ratio for all age classes (Figure 5-3E&F). Selective poaching of males due to their larger tusks than females can explain the skewed sex ratio in favor of females in the SNP. In MGR, we did not observe any significant difference in the sex ratio between males and females probably because of heavy poaching. Initially poachers would select larger males but they would start targeting adult females as well. Thus, our results suggest that the MGR elephants have been affected by poaching more than the SNP and NCA elephant populations.

120

Figure 5-2. Age and sex structure of L.africana in the Serengeti ecosystem using fecal- centric and the rapid demographic assessment method.

Note. Age groups that significantly deviated from the expected 1:1 sex ratio at p<0.05 are indicated by asterisks. A & B= Serengeti National Park (SNP), C&D= Ngorongoro Conservation Area (NCA), E & F= Maswa Game Reserve (MGR)

121 Rapid Demographic Assessment (RDA)

We sampled individuals from 67 groups at SNP, 27 at NCA, and 11 at MGR to assess the age and sex observed during the dry seasons between 2015 and 2017. Due to the large population size at SNP, which made it hard to identify unique individuals, we only analyzed data collected in 2017. However, for NCA, we compiled data for all three years (Table 5-1).

Table 5-1. The number of elephants sampled at SNP, NCA and MGR between 2015 and 2017. The numbers in parentheses denotes the number of individuals as a proportion of the total population size (TAWIRI, 2015)

Dry Population # Sightings # Elephants # Males # Females #Gender season unknown 2017 SNP 67 608 (0.118) 229 328 51

2015-17 NCA 27 126 (0.602) 54 48 24

2017 MGR 11 68(0.2615) 29 26 13

We estimated the proportion of the sampled individuals using elephant population numbers from a recent survey, which showed 5,160 at SNP, 199 at NCA, and 257 at MGR (TAWIRI, 2015). At SNP, the sex ratio for young elephants in the age class of 0-4 and 5- 9 was not significantly different from 1:1. For all age classes of elephants, > 10 years sex was skewed in favor of females and was substantially different from the expected ratio of 1:1 (p <0.05) (Figure 5-3B). In general, the number of adults was fewer than younger elephants. At NCA, for the age classes 0-4, 5-9, 10-14, 15-19, 20-24 the sex ratio was not significantly different from 1:1. For the age classes of 25-34 and 35-60, there were more males than females, but only the sex ratio for the 35-60 age class was significantly different from 1:1 (Figure 5-3C & D). This unusually large number of adult males was mainly due to the number of elephants from the crater, where there were more males than females for

122 the 25-34 and 35-60 age classes (Table 5-S1). At MGR, the sex ratio for most age groups did not deviate from the expected rate (Figure 5-3 E & F). The highest value of OSR was observed at NCA, followed by MGR (Figure 5-4). The OSR was high at NCA because there were more adult males there than females. The OSR was higher at MGR than at SNP because there were more adult males than females observed particularly in Mwiba. Using dung circumference, the OSR is highest at NCA (0.8289), and it was low at MGR (0.0882). At SNP, the OSRs obtained using both methods were similar.

The dependent rates calculated using both RDA and dung were higher at MGR than at SNP and NCA, but they were lower at NCA than at SNP. The dependent ratios obtained using the indirect method (dung) were significantly lower than the ratios obtained using the RDA method (Figure 5-4).

Figure 5-3 Operational sex ratio (OSR) and dependent ratio for SNP, NCA, and MGR using the rapid demographic assessment (RDA) and indirect method using dung circumference (Dung).

123 MGR had higher dependent ratio using RDA probably because most adult females were poached which left orphans We recorded visible signs of poaching, such as the presence of wire snares and wounded animals (bullets/spears or snared) at all the sampling locations. At SNP, two females had lost two-thirds of their trunks due to wire snares in the Moru and Seronera areas. One male in the Kogatende area in the northern Serengeti had severe injuries from spears and bullets, which caused his death two days later. In the western Serengeti, we found 28 wire snares set by poachers along the Grumeti River. At NCA and MGR, we observed neither poaching signs nor wounded elephants. However, at MGR, the elephants immediately ran away as soon as they heard or saw a vehicle approaching, their flight distance was further, and they were more aggressive than the elephants at NCA and SNP. So, our new method will be very useful in these sorts of areas. Heavy poaching and hunting activities might have affected elephants’ behavior. For that reason, we observed very few elephants there compared to SNP and NCA (Figure 5-3), and we also found a very low proportion of tuskless elephants, 0.016 and 0.027 at SNP and NCA respectively.

Discussion

Age and sex structure

The elephant population at SNP is rapidly growing at a rate of 7% per annum (Morrison et al., 2017), despite the loss of about 80% before the international ban on ivory trade of 1989 by CITES (Sinclair et al., 2008). This increase is attributed to low poaching pressure, natural population growth, and immigration (Morrison et al. 2017). For a relatively unpoached population, the adult sex ratio is only expected to be slightly skewed in favor of females (Poole & Thomsen, 1989) because poachers targets males with larger tusks. Following the rapid population growth, we do not expect the sex ratio of elephants

124 to deviate from 1:1. SNP is highly protected and visits are restricted to photographic tourism. NCA has multiple land uses. Meanwhile, at MGR, trophy hunting takes place. For elephants older than ten years, we observed fewer males, females’ number was almost twice the number of males. Other studies on elephants have shown that males have a higher mortality rate than females (Moss, 2001; Turkalo et al., 2018). Another explanation for the observed pattern may be that males are poached more than females. However, SNP has experienced low poaching pressure for the past 30 years. At NCA, there were more adult males than females, especially in the crater (Figure 5-S1). Higher levels of protection in the Crater helps to enhance tourism experience but also protect endangered wildlife species such as rhinoceros. Our results were consistent using RDA and dung size methods (Figure 3 C & D). Ngorongoro Crater has different age and sex structure compared to the Serengeti National Park. In unpoached elephant populations, adult sex ratios are only slightly biased in favor of females (Poole & Thomsen, 1989). However, in heavily poached populations, due to the selective killing of males with enormous tusks, adult sex ratios are highly skewed in favor of females (Jones et al., 2018; Poole & Thomsen, 1989). In MGR, we did not observe significant deviation of sex ratios between males and females although the population experiences more poaching than SNP and NCA. Effects of poaching on sex ratio is visible for moderate poaching pressure. If a population is heavy poached, older females are also targeted by poachers. Wasser et al., (2015) found that there was an increase in the number of female elephants poached based on DNA analysis of seized elephant ivory from East Africa. Therefore, it is possible that heavily poaching in MGR went beyond a certain threshold which affected both males and females. For instance, during our study period, poachers in MGR even shot and killed a British pilot who was conducting anti- poaching patrols with a helicopter (Will Worley, 2016). An unusual observation in the crater indicates that there were more adult males there than females. This observation is consistent with previous studies, which observed more males than females, and they speculated that non-musth males avoid harassment from sexually active bulls by congregating here (Kabigumila, 1993). This study observed the presence of both musth and non-musth males (Table 5-S1) as we observed males from 25

125 years of age and above. This contradicts the findings of Kabigumila (1993) that there are only non-musth males on the crater floor. It has been established that males that were sexually inactive utilize areas different from those used by females to avoid harassment from sexually active males consorting with females and to replenish the energy spent due to sexual activity (Kabigumila, 1993; Poole & Moss, 1981). In this study, we observed both sexually active and inactive males in the Crater. The crater has about 60 lions (Kissui & Packer, 2004) and between 171 and 347 hyenas (Honer et al., 2005). We only observed a small number of elephant females in the crater (maybe residents) probably because they avoid such predators. All family groups were observed in the Lerai forest area (Acacia xanthophloea) within the crater. It is also possible that the Crater does not meet the nutritional requirements of females. An increase in non-invasive plant species on the Crater floor is one of the reasons for food scarcity at the Crater (Estes et al., 2006; Masao et al., 2015; Niboye, 2010; Oates & Rees, 2013). It could also be that the old males have weak teeth that cannot debulk trees easily, so they spend much of their time in the crater, where there is soft grass in the marshy area. The Crater is one of the most protected areas at NCA. It is possible that bulls avoid dangerous areas and take refuge in the crater. Although NCA has not experienced heavy poaching over the past decade, higher levels of protection within the Crater than outside the Crater may attract more elephants. The high level of protection in the crater is due to the presence of highly endangered black rhinoceroses, and it is a hotspot for tourism at NCA. Males are known to utilize a wider diversity of habitats than females (Stokke & Toit, 2002). However, the situation is different at NCA. The males there are more confined to the crater, and the family groups seem to utilize a wider range of habitats. NCA elephants have also adapted to utilize forest habitats, which is not typical for savanna elephants. The Ngorongoro forest is not easily accessible because its escarpment is 2,100-2,800m high. Therefore, it is hard to observe elephants in the forest, and no detailed studies have been done in this forest (Homewood and Rodgers, 1991). In these areas, a fecal-centric approach to determine age and sex structure is essential. A significant difference in elephant counts in the crater between for the dry and wet seasons 1963 and1995 (Estes & Small, 1981; Runyoro et al., 1995; Moehlman et al., 1997)

126 evidence the movement of elephants inside and outside the crater. The crater elephants show a strong genetic similarity to the Serengeti elephants (Lohay et al., chapter 2). However, during our fieldwork for all three years, no elephants were observed between the Oldupai Gorge, and the Serengeti. Most elephants occupy the southern part of the NCA. In this paper, we postulate that the areas in Endulen, Kakesio, Ndutu, and Oldeani are essential habitats for the conservation of elephants. Elephants have been observed more frequently in these areas than in previous years (W. Oleseki, personal communication, August 06, 2015). Assessment of vegetation cover between 1975 and 2000 indicated an expansion of the woodland south of NCA (Niboye, 2010). The number of elephants observed in those areas is probably due to the presence of woodland areas, which are suitable for elephants.

Poaching pressure

The elephant population in the Serengeti ecosystem increased rapidly after the international ban on ivory trade in 1989 (Sinclair et al., 2008). A recent study showed that there were no carcasses found in the NCA, but 75 were found in other parts of the Serengeti ecosystem (mostly in the western and northern SNP), of which 73% were found inside the park (TAWIRI, 2014). Most of these poaching hotspots are far from NCA. All carcasses but one were missing tusks, meaning that they had been poached (TAWIRI, 2014). During our study, we did not see signs of poaching at NCA. The crater is a very small fraction (4%) of the NCA landmass. We expect that most of the NCA elephants are to be found outside the Crater. Using RDA and the fecal-centric approach, we identified 102 and 108 unique elephants respectively, of which about 50% were found in the Crater. NCA is estimated to have had 199 elephants in 2014 (TAWIRI, 2015). In the mid-1980s, there were around 300 at NCA (Dublin & Hamilton, 1987), but in the crater, the mean number of elephants during the wet and dry seasons was 32 and 12 respectively between 1963 and 1995 (Runyoro et al., 1995; Moelhman et al., 1997). In the past, the Masai warriors occasionally speared elephants and black rhinos to prove their

127 bravery (Fosbrooke, 1972). However, the Masai at NCA started poaching for subsistence meat and trophies (skin, ivory, and rhino horns) for sale (Makacha et al., 1982). In recent years, incidences of elephant poaching have not been reported at NCA, which indicates that poaching is lower than it has been in the past. We found a skewed adult sex ratio in favor of males at NCA. This observation is quite unusual, even for populations that experience low poaching pressure. The high protection of elephants, particularly in the Crater, the immigration of older elephants in the crater, and low poaching may explain the higher number of older males at NCA. There were no elephants recorded in 1979 in Kakesio, but there were 175 elephants found between the Ndutu and Endulen zone (Makacha et al., 1982). However, elephants are observed more frequently recently in the Kakesio and Endulen areas than previous years (W. Oleseki, personal communication, August 06, 2015). NCA experienced a significant increase in woody vegetation cover between 1975 and 2000 (Niboye, 2010). These changes may have affected the movement and distribution of elephants between the Crater and other parts of the NCA and between the NCA and adjacent protected areas. The fraction of tuskless elephants is expected to increase due to selective hunting pressure (Jachmann et al., 1995). For example, the percentage of tuskless females in Selous, Ruaha, and Ugalla, which experience frequent poaching, were 6.3%, 7%, and 9.7% respectively (T. Jones et al., 2018). Our results showed that the proportion of tuskless females for SNP and NCA was less than 1% of all females older than 5 years old. At MGR, we did not observe any tuskless females, probably because we encountered a small number of elephants. These results were comparable with what was reported by Jones et al. (2018), which found less than 1% of tuskless elephants in the Serengeti and Tarangire. Although these populations were affected by heavy poaching in the mid-1980s, the proportion of tuskless individuals has remained low.

The number of calves is under-represented using the dung method because calves have a lower defecation rate, their dung degrades rapidly, and it is hard to see due to their small size (Hema et al., 2017). While the fecal-centric approach may not be suitable for determining the dependent ratio, the trend is consistent with what was observed using the RDA method.

128

Conclusion

Our new fecal-centric approach can be used to assess age and sex structure particularly in areas where it is hard to observe elephants. We validated our method using the RDA method and showed agreement with that method. Differential poaching and survivorship rates between males and females may explain this significantly skewed adult sex ratio observed at SNP. In contrast, at NCA, there were significantly more adult males than females. These results imply that SNP experiences more poaching pressure than NCA. However, the unexpected proportionally larger number of adult males in excess of females at NCA, particularly in the crater, suggests that older males immigrate into the crater seeking the protection and food. At MGR, there were very few adults older than 25 years old, probably due to poaching or because they moved to other areas of the ecosystem during the hunting season. Our observed higher OSR at NCA than at SNP and MGR reflects the larger number of males observed in the crater. This excess of males in the crater has been maintained over the past 30 years (Kabigumila, 1993). Unlike the previous study, which found only old non-musth males in the crater, this study found males that are both sexually active and inactive. We provide compelling evidence that a fecal-centric approach is well suited for determining the ages and sexes of elephants in field studies. Our modified AMELX/Y molecular sexing technique was simple and accurate for sex determination; however, it less useful to identify an individual’s age. Dung circumference measurements provided useful age estimates of adult elephants. However, the number of calves is under-represented due to their low defecation rate and reduced visibility due to their smaller size compared to adults. Thus, dung circumference can likely only provide reliable information on population age structure for individuals older than ten years old. We recommend the use of GPS collars on the NCA elephants to understand their movement within the ecosystem to secure their habitats.

129 References

Barnes, R. F. W., & Kapela, E. B. (1991). Changes in the Ruaha elephant population caused by poaching. African Journal of Ecology, 29(4), 289–294. https://doi.org/10.1111/j.1365-2028.1991.tb00466.x Bartáková, V., Reichard, M., Janko, K., Polačik, M., Blažek, R., Reichwald, K., McLean, I. (2018). Conserving elephants depend on a total ban of ivory trade globally. Biodiversity and Conservation, 13(March), 37–43. https://doi.org/10.1007/s10531-018-1534-x Boshe, J. I. (1997). Wildlife ecology. In Thompson, D. M. (ed.), Multiple Land-Use: The Experience of the Ngorongoro Conservation Area, Tanzania, IUCN, Gland, Switzerland and Cambridge, UK, pp. 71–80. Cerling, T. E., Barnette, J. E., Chesson, L. A., Douglas-Hamilton, I., Gobush, K. S., Uno, K. T., Xu, X. (2016). Radiocarbon dating of seized ivory confirms rapid decline in African elephant populations and provides insight into illegal trade. Proceedings of the National Academy of Sciences, 113(47), 13330–13335. https://doi.org/10.1073/pnas.1614938113 Chief Assignment Editor, 2017. Serengeti faces yet another silent but deadly poaching threat, eTurbo News, October 25, 2017 20:25, accessed May 19, 2019, https://www.eturbonews.com/168646/serengeti-faces-yet-another-silent-deadly- poaching-threat/ Estes, R. D., Atwood, J. L., & Estes, A. B. (2006). Downward trends in Ngorongoro Crater ungulate populations 1986-2005: Conservation concerns and the need for ecological research. Biological Conservation, 131(1), 106–120. https://doi.org/10.1016/j.biocon.2006.02.009 Estes, R. D., & Small, R. (1981). The large herbivore populations of Ngorongoro Crater. African Journal of Ecology, 19(1–2), 175–185. https://doi.org/10.1111/j.1365- 2028.1981.tb00661.x Fosbrooke, H. . (1972). Elephants in the Serengeti National Park: an early record. African Journal of Ecology, 150–152.

130 Gobush, K. S., Mutayoba, B. M., & Wasser, S. K. (2008). Long-term impacts of poaching on relatedness, stress physiology, and reproductive output of adult female african elephants. Conservation Biology: The Journal of the Society for Conservation Biology, 22(6), 1590–1599. https://doi.org/10.1111/j.1523- 1739.2008.01035. Hema, E. M., Di Vittorio, M., Petrozzi, F., Luiselli, L., & Guenda, W. (2017). First assessment of age and sex structures of elephants by using dung size analysis in a West African savannah. European Journal of Ecology, 3(1), 1–8. https://doi.org/10.1515/eje-2017-0001 Homewood KM, Rodgers WA (1991), Masailand Ecology. Cambridge University Press, Cambridge, UK. Jachmann, H., Berry, P. S., & Imae, H. (1995). Tusklessness in African elephants: a future trend. African Journal of Ecology, 33(3), 230–235. https://doi.org/10.1111/j.1365-2028.1995.tb00800.x Jachmann, H., & Billiouw, M. (1997). Elephant Poaching and Law Enforcement in the Central Luangwa Valley, Zambia. Journal of Applied Ecology, 34(1), 233–244. Janicke, T., & Morrow, E. H. (2018). Operational sex ratio predicts the opportunity and direction of sexual selection across animals. Ecology Letters, 21(3), 384–391. https://doi.org/10.1111/ele.12907 Jones, T., Cusack, J. J., Pozo, R. A., Smit, J., Mkuburo, L., Baran, P.,Foley, C. (2018). Age structure as an indicator of poaching pressure : Insights from rapid assessments of elephant populations across space and time. Ecological Indicators, 88(November 2017), 115–125. https://doi.org/10.1016/j.ecolind.2018.01.030 Kabigumila, J. (1993). Feeding habits of elephants in Ngorongoro Crater, Tanzania. African Journal of Ecology, 31(2), 156–164. https://doi.org/10.1111/j.1365- 2028.1993.tb00528.x Kijazi A, Mkumbo S, Thompson DM (1997) Human and livestock population trends. In: Thompson DM (ed) Multiple Land-Use: The Experience of the Ngorongoro Conservation Area, Tanzania, 169– 180. IUCN, Gland, Switzerland.

131 Kikoti, A. P., Griffin, C. R., Wasser, S. K., Brown, L., Mailand, C., Mondol, S.,Weir, B. S. (2015). Genetic assignment of large seizures of elephant ivory reveals Africa’s major poaching hotspots. Wildlife & Fisheries Conservation, 349(Open Access Dissertations. Paper 108.), 84–88. https://doi.org/10.1126/science.aaa2457 Kioko, J., Zink, E., Sawdy, M., & Kiffner, C. (2013). Elephant ( Loxodonta africana ) Tarangire-Manyara Ecosystem , Tanzania. South African Journal of Wildlife Research, 43(1), 44–51. Lee, E. J., Lee, Y. H., Moon, S. H., Kim, N. Y., Kim, S. H., Yang, M. S., Han, M. S. (2013). The identification of elephant ivory evidences of illegal trade with mitochondrial cytochrome b gene and hypervariable D-loop region. Journal of Forensic and Legal Medicine, 20(3), 174–178. https://doi.org/10.1016/j.jflm.2012.06.014 Makacha, S., Msingwa, M. J., & Frame, G. W. (1982). Threats to the Serengeti Herds, 437–444. Masao, C. A., Makoba, R., & Sosovele, H. (2015). Will Ngorongoro Conservation Area remain a world heritage site amidst increasing human footprint? International Journal of Biodiversity and Conservation, 7(9), 394–407. https://doi.org/10.5897/IJBC2015.0837 Mobley, K.B. (2014). Mating systems and the measurement of sexual selection. In: Animal Behaviour: How and Why Animals Do the Things They Do (ed. Yasukawa, K.). Praeger, Santa Barbara, pp. 99–144 Moehlman, P.D., Runyoro, V.A., Hofer, H., 1997.Wildlife population trends in the Ngorongoro Crater. In: Thompson, D.M. (Ed.), Multiple Land-Use: the Experience of the Ngorongoro Conservation Area, Tanzania. IUCN Protected Areas Programme, IUCN, Gland, Switzerland and Cambridge, UK, pp. 59–70. Mondol, S., Mailand, C. R., & Wasser, S. K. (2014). Male biased sex ratio of poached elephants is negatively related to poaching intensity over time. Conservation Genetics, 15(5), 1259–1263. https://doi.org/10.1007/s10592-014-0603-2 Morrison, T. A., Estes, A. B., Mduma, S. A. R., Maliti, H. T., Frederick, H., Kija, H., … Kohi, E. M. (2017). Informing Aerial Total Counts with Demographic Models :

132 Population Growth of Serengeti Elephants Not Explained Purely by Demography. A Journal of the Society for Conservation Biology, 00(October), 1–8. https://doi.org/10.1111/conl.12413 Morrison, T., Chiyo, P., Moss, C., & Alberts, S. (2005). Measures of dung bolus size for known-age African elephants (Loxodonta africana): implications for age estimation. Journal of Zoology, 266(2005), 89–94. https://doi.org/10.1017/S0952836905006631

Moss, C.J., 1996. Getting to Know a Population. Studying Elephants. African Wildlife Foundation, Nairobi, pp. 58–74 Moss, C. J. (2001). The demography of an African elephant (Loxodonta africana) population in Amboseli, Kenya. Journal of Zoology, 255(2), 145–156. https://doi.org/10.1017/S0952836901001212 Niboye, E. P. (2010). Vegetation Cover Changes in Ngorongoro Conservation Area from 1975 to 2000: The Importance of Remote Sensing Images. The Open Geography Journal, 3(14), 15–27. https://doi.org/10.2174/1874923201003010015 Oates, L., & Rees, P. A. (2013). The historical ecology of the large mammal populations of Ngorongoro Crater, Tanzania, east Africa. Mammal Review, 43(2), 124–141. https://doi.org/10.1111/j.1365-2907.2012.00211.x Poole, J. H. (1989). Mate guarding, reproductive success and female choice in African elephants. Animal Behaviour, 37, 842–849.Poole, J. H., & Thomsen, J. B. (1989). Elephants are not beetles. Oryx, 23(4), 188–198.

Runyoro, V. A., Hofer, H., Chausi, E. B. et al. 1995. Long-term costs of coexistence in female spotted hyenas: a life history trends in the herbivore populations of the Ngorongoro perspective, Crater, Tanzania. Evol. Ecol. 17: 315-331. Sinclair, A. R. E., Hopcraft, J. G. C., Olff, H., & Mduma, S. A. R. (2008). Historical and Future Changes to the Serengeti Ecosystem. University of Chicago Press, (1972), 7–46. TAWIRI. (2014). Aerial Total Count of Elephants and Buffaloes in the Serengeti-Mara Ecosystem.

133 TAWIRI. (2015). Population Status of Elephant in Tanzania 2014.TAWIRI Aerial Survey Report. Turkalo, A. K., Wrege, P. H., & Wittemyer, G. (2018). Demography of a forest elephant population. PLoS ONE, 13(2), 1–17. https://doi.org/10.1371/journal.pone.0192777 Will Worley, 2016. Roger Gower: Tanzania elephant poachers shoot dead British helicopter pilot in middle of mission, Independent, Sunday Jan 2016 00:09 https://www.independent.co.uk/news/world/africa/british-pilot-shot-dead-in-tanzania-on- anti-poaching-mission-a6844401.html accessed on May 19, 2019

134 Table 5-S 1 Age and sex structure of African Savanna elephants between the Serengeti and Ngorongoro in Tanzania between May and August 2015-17. Age classes were estimated from dung circumference, and their sexes were determined using the AMELX/Y method.

Location Age class Females Males Total p-value SNP 0-6 29 24 53 0.583 7-12 32 27 59 0.603 13-17 60 44 104 0.141 18-25 61 35 96 *0.01 25+ 51 28 79 *0.013 Total 233 158 391

NCA 0-6 5 4 9 1.00 7-12 3 3 6 1.00 13-17 13 9 22 0.523 18-25 14 20 34 0.392 25+ 8 29 37 *0.000 Total 43 65 108

MGR 0-6 3 6 9 0.508 7-12 8 7 15 1.00 13-17 15 16 31 1.00 18-25 12 9 21 0.664 25+ 7 3 10 0.344 Total 45 41 86

135 Table 5-S 2. Age and sex structure of elephants from SNP, NCA, and MGR

Population 0-4 5-9 10-14 15-19 20-24 25-34 35-50+

SNP All 252 78 58 48 36 53 32 Females 117 35 40 41 30 41 24 Males 135 43 18 7 6 12 8 P-value 0.284 0.428 *0.005 *0.000 *0.000 *0.000 *0.007

NCA All 29 10 11 10 13 14 15 Females 17 7 6 4 8 4 2 Males 12 3 5 6 5 10 13 P values 0.458 0.344 1.00 0.754 0.581 0.18 *0.007

MGR All 22 12 7 6 3 3 2

Females 10 4 5 4 2 1 0 Males 12 8 2 2 1 2 2 P values 0.832 0.388 0.453 0.687 1.00 1.00 0.50

136 Figure 5-S 1. Age and sex distribution of elephants between the (A) Crater and (B) Non- Crater (Kakesio, Endulen, and Ndutu) using the RDA method.

(A)

(B)

137 : Early signs of genetic differentiation among African savanna elephants in the Serengeti and Tarangire-Manyara ecosystems in northern Tanzania

George G. Lohay and Douglas R. Cavener

Abstract

Habitat loss and fragmentation pose a significant threat to the conservation and the sustainability of elephant populations. It is crucial to determine if existing wildlife corridors are still viable and facilitate gene flow. Loss of connectivity among populations results in a loss of genetic diversity, which may affect reproductive success and survival of populations. We asked whether there is genetic differentiation of elephants and if the genetic differentiation has changed over the past 50 years. We collected 615 fresh elephant dung samples from Serengeti and Tarangire ecosystems from which we extracted DNA for genetic analyses. We estimated the age of elephants by measuring the circumference of at least three dung boli per individual and identified sex using the AMELX/Y method. We first split samples into two major groups: young (0-20) and old 20+ and also subdivided into four age groups (0-12, 13-17, 18-25, and 25-60). We also performed a separate genetic analysis for males and females to determine if females reflect the effect of genetic isolation more than males because males facilitate gene flow between family groups and females stay permanently in family groups. We analyzed 11 microsatellite markers to amplify variable regions of DNA and genotyped all samples using ABI 3730 DNA analyzer. We used the program STRUCTURE to assign individuals to subpopulations based on their genotypes. Our results show early signs of genetic differentiation among young elephants. Although there were no significant differences in STRUCTURE results between age classes (K=2), we detected higher delta K for younger elephants than for the older elephants suggesting that the young elephants have stronger clustering than the older elephants. An

138 admixture of the two subpopulations was observed in Ngorongoro, suggesting either ongoing/recent gene flow between them. Interestingly no genetic clustering was detected using males older than 18 years, providing evidence for male-mediated gene flow between the two ecosystems. Also, older individuals show more admixture than the younger individuals suggesting the recent development of gene flow barriers. Although there is a clear separation of elephant populations between west and east of the rift valley, habitat loss due to human activities may reduce the rate of gene flow. Our results could be used to help enact landscape-scale conservation measures aimed at preserving connectivity between the slightly threatened elephant populations in northern Tanzania.

Introduction

Fragmentation of natural habitats is a significant challenge in conservation biology and one of the top threats to biodiversity (Hanski, 1999; Fahrig, 2003; Henle et al., 2004). Negative impacts of fragmentation result from the decrease in overall habitat availability and changes in spatial configuration and habitat quality of fragments (Ezard & Travis, 2006; Fahrig, 2003). Human population growth is one of the drivers of natural habitat loss and increased isolation (Jones et al., 2012; Newmark, 2008; Pereira et al., 2010; Rands et al., 2010). Indeed human-caused habitat fragmentation is a leading threat to biodiversity (Beier & Gregory, 2012) by decreasing habitats and increasing edge effects for species. Habitat loss and fragmentation have genetic effects on isolated species. Genetic consequences of habitat fragmentation on species include small population with lower genetic diversity which leads to genetic drift, higher risks of inbreeding, and lower evolutionary potential and consequently more elevated risk of extinction (Allendorf, 1986; Avise, 2010; Dixon et al., 2007; Frankham et al., 2002). African savanna elephant populations (Loxodonta Africana) have declined rapidly over the past few decades due to poaching. Habitat loss is another major threat affecting their population. For the long-term survival of elephant populations, there is a need to

139 maintain genetic connectivity between populations. Luckily, Tanzania has documented the status of wildlife corridors at the national level (Caro et al., 2009; Riggio & Caro, 2017). Wildlife corridors connecting protected areas between Tarangire-Manyara and Serengeti ecosystems are in a critical condition and are likely soon to be blocked completely (TAWIRI, 2010). To determine genetic consequences of habitat isolation on elephant population we need to know historical population genetic structure and compare that with the current genetic structure. Elephants, unlike most terrestrial mammals, have a long life span with an average between 60-65 years. The ability of elephants to live a long time provides us with a unique opportunity to use both their life history and their genetic information to address the conservation issues they face. Here we compared population genetic structure between young (0-20 years) and old (20-60) to determine whether there is an early sign of variation among populations using measures of genetic differentiation. We used the age of 20 years as a cut point because we could estimate their age from dung circumference with accuracy. We hypothesized that young elephants would be less genetically related to older elephants between the ecosystems because of habitat fragmentation. Because the blocking of wildlife corridors is a recent phenomenon, we assume that older elephants experienced more gene flow between them. The opposite hypothesis is that older adults in the Serengeti would show higher differentiation than younger elephants because the older elephants came from different regions (north and south) to recolonize the Serengeti after last major poaching. But this assumes that random mating in the Serengeti occurred after recolonizations among the two major colonizing groups. If random mating had occurred, then you would expect the younger elephants to be more closely related. We also wanted to determine if there is a difference in genetic population structure between males and females. Female elephants live in family groups while males leave their families after they reach sexual maturity. Male-biased gene flow has been documented among elephants (Silvester Nyakaana & Arctander, 1999; Okello et al., 2008). We hypothesize that females will be more genetically distinct than males (i.e. females will be more isolated than males)

140 Methods

We collected fecal samples from the Serengeti and Tarangire-Manyara ecosystem as described in chapter two. We estimated the age of elephants by measuring the diameter of at least three dung boli per individual and identified sex using AMELX/Y method. We first split samples into two major groups: young (0-20) and old. We used these two age groups because we could estimate elephants age from dung circumference with higher accuracy up to 20 years. In a different analysis we subdivided into four age groups (0-12, 13-17, 18-25, and 25-60). We used indirect method of measuring dung circumference to estimate the age of elephants. We also performed a separate genetic analysis for males and females to determine if females reflect the effect of genetic isolation more than males. We analyzed 11 microsatellite markers to amplify variable regions of DNA and genotyped all samples using ABI 3730 DNA analyzer. We used the program STRUCTURE to assign individuals to subpopulations based on their genotypes. We described detailed methods for DNA isolation, microsatellite fragment analysis, and genetic analyses are described in chapter 2. Female elephants have one of the most extended reproductive lives of any terrestrial mammals; they produce calves for 40 or even 50 years (Lee et al., 1982). They start giving birth at an average age of 13.8 years with a birth interval of 3 to 5 years (Moss & Lee, 2011). Males become sexually mature at the age of 15 years but their potential reproductive increases with their age. Analysis of molecular variance (AMOVA) was conducted to support the hypothesis of population structure due to isolation (Excoffier, Smouse, & Quattro, 1992) using Arlequin (Excoffier et al., 2005). We also compared fixation indices between populations calculated from Arlequin (Excoffier et al., 2005).

141 Results

We genotyped 575 individuals to determine if there is a substantial change in genetic differentiation between the Greater Serengeti (SE) and the Tarangire-Manyara ecosystem (TME). In this analysis, we only included samples with age estimated from dung circumference. For all 11 loci data, one locus, FH60 consistently deviated from the assumptions of Hardy-Weinberg Equilibrium and was removed for subsequent genetic analyses. We first compared fixation (FST) indices between young and old elephants for 5 locations (Figure 6-1) and then pooled data to compare FST for between three areas: TME,

NCA and SE (Table 6-2). There was no significant difference in FST values between the sampling locations between young and old individuals (Figure 6-1) although the FST values were slightly higher for the young individuals after pooling samples for the two ecosystems (Table 6-2). The lowest FST of 0.04 among the older elephants was observed between NCA and SE, while it was higher for young elephants, FST=0.01, (Table 6-2).

Figure 6-1. Pairwise comparison of FST value between sampling locations between young and old elephants

142 NSE=North Serengeti, SSE=South Serengeti, NCA=Ngorongoro Conservation Area Authority, MAR=Manyara Ranch and Lake Manyara National Park,TNP=Tarangire National Park Table 6-1. Genetic distance measures among elephant populations for two age groups; young (below diagonal) and old (above diagonal)

Tarangire-Manyara Ngorongoro Serengeti Tarangire-Manyara - 0.015* 0.021* Ngorongoro 0.010* - 0.004* Serengeti 0.023* 0.01* -

Note: Significance levels are indicated as *p < 0.05

Global FST (0.0173) for young elephants was slightly higher than the older individuals

(FST=0.014). We found more negative values of inbreeding coefficient FIS among young individuals (Table 6-4) than old (Table 6-5). FIS is a comparison of observed and expected heterozygosity within subpopulation. Negative values mean that there is an excess of heterozygotes. This could suggest more outcrossing in the younger generation. Analyses of molecular variance (AMOVA) show an increase in percentage variation among young elephants compared to the old (Table 6-3). Variation within individuals was higher 95.28% for young and 93.10% for older elephants (Table 6-3). Population-specific inbreeding coefficient (FIS) indicates that young elephants have more negative values than older individuals (Table 6-4). These results suggest that young elephants are less related than expected under random mating assumptions.

143 Table 6-2. Analysis of molecular variance (AMOVA) of genetic diversity of African elephants using 10 SSRs loci

Source of variation d.f Sum of squares Variance Percentage of components variation Among 2 (2) 26.36 (24.56) 0.063 (0.048) 1.73 (1.5) populations Among individuals 274 1003.88 (973.54) 0.108 (0.175) 2.99 (5.40) within populations (288) Within individuals 277 955 (881.5) 3.448 (0.17) 95.28 (93.10) (291) Total 553 1985.24 (1879.6) 3.618 (3.25) (581)

Note: Values in parenthesis are for older individuals

Table 6-3. Population-specific FIS indices per polymorphic locus for young elephants

Locus Average FIS SE NCA TME LAFMS02 0.04166 0.06341 0.17665 -0.10317 LA5 -0.03652 -0.06128 -0.08696 0.04471 LA6 0.06527 0.14988 -0.03653 -0.0805 LAT06 0.08203 0.08506 0.12611 0.04883 LAT08 -0.02131 -0.0492 -0.05685 0.07547 LAT13 -0.0156 -0.08235 0.00478 0.13089 FH19 0.10735 0.05994 0.09989 0.21985 LAT24 0.05649 -0.01163 0.17787 0.16014 FH48 0.14438 0.17727 0.30283 -0.06744 FH67 0.01221 -0.00704 -0.08867 0.11378

144

Table 6-4. Population-specific FIS indices per polymorphic locus for old elephants

Locus Average FIS SE NCA TME LAFMS02 0.04513 0.05412 0.04735 0.02995 LA5 0.15695 0.18956 -0.06061 0.22781 LA6 0.09485 0.21701 0.03726 -0.04395 LAT06 0.04328 0.05943 0.04959 0.01384 LAT08 0.03512 0.02738 -0.02722 0.09104 LAT13 0.08728 0.08818 -0.00313 0.14031 FH19 0.08784 0.02894 0.23568 0.08557 LAT24 0.10051 0.08552 0.12946 0.10492 FH48 0.11774 0.14041 0.10551 0.08833 FH67 -0.00871 0.0224 -0.05281 -0.02906

For all scenarios considered in STRUCTURE analyses, we identified at least two genetic clusters (Figure 6-2A-F). Although we did not detect a significant difference between STRUCTURE results among age groups, we used the height of delta K (∆K) as an indicator of the strength of the signal detected by structure (Evanno, Regnaut, & Goudet, 2005). The ∆K=90 for young and ∆K=15 for old group. When we run structure for other four age groups (0-12, 13-17, 18-25, and 25-60), ∆K was generally higher among young age groups than old (Figure 6-S1). For instance, ∆K=5 for individuals between 25 -60 years which suggests no genetic structure.

145

Figure 6-2. Bayesian clustering using STRUCTURE program for different age groups of African Savanna elephants between the Serengeti and Tarangire ecosystems in northern Tanzania

STRUCTURE analysis between females and males showed two clusters for females (∆K =8) or three clusters (∆K=6) but male showed only two clusters, ∆K= 20 (Figure 6-3). We then analyzed only males that are sexually mature 18-60 years and detected no clustering, K=1, (Figure 6-S1).

146

Figure 6-3. Bayesian clustering using STRUCTURE program for males and females of African Savanna elephants between the Serengeti and Tarangire ecosystems in north Tanzania

We also performed the principal coordinate analysis between the young and old groups to determine which groups are more or less similar. We used genetic distance data between individuals for each if the two groups. Our results suggest that older individuals showed show higher similarity (more clustered together) than the younger elephants (Figure 6-4)

147

Figure 6-4. Principal coordinate analysis for (A) young <20 years and (B) old elephants >20 years in the Serengeti and Tarangire-Manyara ecosystems

Similarly, spatial autocorrelation among females showed that after 180 km, females are less genetically related, but the spatial autocorrelation was not significantly different from zero for both age groups. We expected to observe the spatial autocorrelation (r) differences between males and females due to their difference in the social structure. Females are philopatric and stay in their natal groups while males leave their groups after reaching puberty.

148

Figure 6-5. Correlogram showing spatial genetic autocorrelation (r) among (A) young (0- 20 years) and (B) old (25-60) African savanna elephants as a function of Euclidean distance.

To test if there is a difference in spatial correlation among age groups, we performed spatial autocorrelation for an equal distance of 30 km to test the null hypothesis of no genetic structure. For both age groups, individuals within 30 km showed high genetic similarity (Figure 6-5). However, the young individuals showed a significant negative spatial autocorrelation suggesting that after 150 km elephants in these group are less genetically similar (Figure 6-5A). Among the older elephants, an individual’s spatial autocorrelation was not significantly different after the first two distance classes (Figure 6-5B, km 120).

149

Figure 6-6. Correlogram showing spatial genetic autocorrelation (r) among (A) female (B) male African savanna elephants as a function of Euclidean distance.

We defined distance classes every 30 km. Dotted lines indicate the 95% CI about the null hypothesis of no genetic structure. The error bars about r represent the 95% CI, as determined by bootstrapping (999 iterations)

Discussion

We detected early signs of genetic differentiation among African elephants in northern Tanzania. Elephants from the Tarangire-Manyara ecosystem were genetically distinct from the Serengeti elephants. Although we did not detect a significant difference

150 in population structure using the STRUCTURE program, we used the height of delta K (∆K) as an indicator of the strength of the signal detected by structure (Evanno et al., 2005). High ∆K indicates robust clustering whereas weak ∆K indicates weak or lack of clustering. For animals with long life span like elephants, it may take a few years to detect effects of genetic isolation. Our analyses consistently provide evidence for the presence of more genetic differentiation among young elephants than old. Similar studies in Western and central Tanzania showed higher genetic differentiation among younger elephants than older elephants due to habitat fragmentation of Miombo woodland (Lobora et al., 2017). In northern Tanzania, wildlife corridors are in critical condition and may be blocked entirely. For example, the only corridor remaining between Tarangire and Manyara is less than 5 km wide (Morrison et al., 2016). Similarly, corridors between Ngorongoro and Manyara are threatened by an increase in human activities (TAWIRI, 2010). For young elephants, we found no difference in FST values between NCA&TME and NCA& SE for young elephants (Table 6-2). For example, for older elephants, the FST value between SE and NCA was 0.004, but it was 0.01 for the young cohort. Even between SE and NCA, there is reduced gene flow. Population differentiation among the buffalos between the NCA and SE has been reported (Ernest et al., 2012). This genetic differentiation has been attributed to habitat fragmentation as a result of increased human activities in the NCA (Ernest et al., 2012). Male-biased gene flow has been documented among elephants (Nyaakaana and Arctander, 1999; Okello et al., 2008). Males older than 18 years did not show significant clustering suggesting a high genetic similarity between the SE and TME. We were not surprised to observe more genetic structure among females than males because females stay in their natal groups. Comstock (2002) predicted that genetic differentiation due to habitat fragmentation would be a conservation issue for African elephants. Indeed, we observed early signs of genetic differentiation due to reduced gene flow between subpopulations in northern Tanzania. Human activities have accelerated loss of connectivity between protected areas. Natural barriers such as mountains and the rift valley may affect connectivity among subpopulation Epps et al., (2013) stated that historical connectivity among elephant populations was strongly influenced by slope but not human

151 settlements, whereas contemporary connectivity was influenced most by human settlement. A clear separation between Tarangire and Serengeti ecosystems could be explained by the rift valley slope (Ahlering et al., 2012, Lohay, chapter 2).

Implications for conservation

Most wildlife corridors in Tanzania are vanishing. If we do not take action now most of these protected areas will be completely isolated which may have a long-term effect on the health of populations. Our study provides genetic evidence of reduced gene flow between these protected areas within the last 50 years. Luckily, the Tanzania government has a wildlife policy to protect migratory corridors wildlife (the United Republic of Tanzania, 2009). However, there is a lack of political will, to implement these policies because there are no acts. It should be noted that local people own most of the areas within the corridors. We need an approach which will also keep the interest and augment the wellbeing of local people. We suggest the use of the framework suggested by Jones et al., 2012 to restore these corridors. One approach would be to purchase areas along the corridors from private owners and fence them to reduce incidences of crop raiding by elephants. This exercise should involve all stakeholders to ensure all land owners receive compensation and helped to get another land elsewhere. Another way would be to build a bridge across Arusha- Babati highway between Manyara Ranch and Tarangire National Park to facilitate movement of wildlife and reduce accidents that would be caused by hitting wildlife that try to cross. Also, there are ways to increase set aside areas or enhance wildlife use particularly areas within the known migratory routes. Some areas could be upgraded to wildlife management areas.

152 References

Allendorf, F. W. (1986). Genetic drift and the loss of alleles vs heterozygosity. Zoo Biology, 5, 181–190. Retrieved from citeulike-article-id:540565 Avise, J. C. (2010). Perspective : conservation genetics enters the genomics era. Conservation Genetics, 11, 665–669. https://doi.org/10.1007/s10592-009-0006-y Beier, P., & Gregory, A. J. (2012). Desperately Seeking Stable 50-Year-Old Landscapes with Patches and Long, Wide Corridors. PLoS Biology, 10(1), e1001253. https://doi.org/10.1371/journal.pbio.1001253 Caro, T., Jones, T., & Davenport, T. R. B. (2009). Realities of documenting wildlife corridors in tropical countries. Biological Conservation, 142(11), 2807–2811. https://doi.org/10.1016/j.biocon.2009.06.011 Comstock, Kenine E, Georgiadis, Nicholas J, Pecon-slattery, Jill, Roca, Alfred L, Ostrander, Elaine A, O’Brien, Stephen J, & Wasser, Samuel K. (2002). Patterns of molecular genetic variation among African. Molecular Ecology, 2489–2498. Dixon, J. D., Oli, M. K., Wooten, M. C., Eason, T. H., McCown, J. W., & Cunningham, M. W. (2007). Genetic consequences of habitat fragmentation and loss: The case of the Florida black bear (Ursus americanus floridanus). Conservation Genetics, 8(2), 455–464. https://doi.org/10.1007/s10592-006-9184-z Ernest, E. M., Haanes, H., Bitanyi, S., Fyumagwa, R. D., Msoffe, P. L., Bjørnstad, G., & Røed, K. H. (2012). Influence of habitat fragmentation on the genetic structure of large mammals: evidence for increased structuring of African buffalo (Syncerus caffer) within the Serengeti ecosystem. Conservation Genetics, 13(2), 381–391. https://doi.org/10.1007/s10592-011-0291-0 Epps, C. W., Wasser, S. K., Keim, J. L., Mutayoba, B. M., & Brashares, J. S. (2013). Quantifying past and present connectivity illuminates a rapidly changing landscape for the African elephant. Molecular Ecology, 22(6), 1574–1588. https://doi.org/10.1111/mec.12198 Evanno, G., Regnaut, S., & Goudet, J. (2005). Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Molecular

153 Ecology, 14(8), 2611–2620. https://doi.org/10.1111/j.1365-294X.2005.02553.x Excoffier, L., Laval, G., & Schneider, S. (2005). Arlequin ( version 3 . 0 ): An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 1, 47–50. https://doi.org/10.1177/117693430500100003 Excoffier, L., Smouse, P. E., & Quattro, J. M. (1992). Analysis of Molecular Variance Inferred From Metric Distances Among DNA Haplotypes: Application to Human Mitochondrial DNA Restriction Data. Genetics, 131(2), 479–491. https://doi.org/10.3354/meps19828 Ezard, T. H. ., & Travis, J. M. . (2006). The Impact of Habitat Loss and Fragmentation on Genetic Drift and Fixation Time. Oikos, 114, 367–375. Fahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics, 34(1), 487–515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419 Frankham, R., Ballou, J. ., & Briscoe, D. . (2002). Introduction to conservation genetics. Cambridge, UK: Cambridge University Press. Hanski, Ikka, 1999. Metapopulation Ecology, Oxford University Press, UK Henle, K., Davies, K. F., Kleyer, M., Margules, C., & Settele, J. (2004). Predictors of species sensitivity to fragmentation. Biodiversity and Conservation, 13, 207–251. Jones, T., Bamford, A. J., Ferrol-, D., Hieronimo, P., Mcwilliam, N., & Rovero, F. (2012). Vanishing wildlife corridors and options for restoration : a case study from Tanzania, Tropical Conservation Science 5(4), 463–474. https://doi.org/10.1177/194008291200500405 Lee, P. C., Poole, J. H., Njiraini, N., Sayialel, C. N., & Moss, C. J. (1982). Male Social Dynamics: Independence and Beyond. University of Chicago Press. Leigh, J. W., & Bryant, D. (2015). POPART: Full-feature software for haplotype network construction. Methods in Ecology and Evolution, 6(9), 1110–1116. https://doi.org/10.1111/2041-210X.12410 Lobora, A. L., Nahonyo, C. L., Munishi, L. K., Caro, T., Foley, C., Prunier, J. G., Eggert, L. S. (2018). Incipient signs of genetic differentiation among African elephant populations in fragmenting miombo ecosystems in south-western Tanzania. African

154 Journal of Ecology, 56(4), 993–1002. https://doi.org/10.1111/aje.12534 Morrison, T. A., Link, W. A., Newmark, W. D., Foley, C. A. H., & Bolger, D. T. (2016). Tarangire revisited: Consequences of declining connectivity in a tropical ungulate population. Biological Conservation, 197, 53–60, doi.org/10.1016/j.biocon.2016.02.034 Moss, C. J., & Lee, P. C. (2011). Female reproductive strategies: individual life histories. The Amboseli Elephants: A Long-Term Perspective on a Long-Lived Mammal, University of Chicago Press, 187–204. Nei, M. (1972). Genetic distance between populations. The American Naturalist, 106(949), 283–292. Newmark, W. D. (2008). Isolation of African protected areas. Frontiers in Ecology and the Environment, 6(6), 321–328. https://doi.org/10.1890/070003 Nyakaana, & Arctander. (1999). Population genetic structure of the African elephant in Uganda based on variation at mitochondrial and nuclear loci: Evidence for male- biased gene flow. Molecular Ecology, 8(7), 1105–1115. https://doi.org/10.1046/j.1365-294X.1999.00661.x Okello, J. B. A., Masembe, C., Rasmussen, H. B., Wittemyer, G., Omondi, P., Kahindi, O.,Siegismund, H. R. (2008). Population genetic structure of savannah elephants in Kenya: Conservation and management implications. Journal of Heredity, 99(5), 443–452. https://doi.org/10.1093/jhered/esn028 Pereira, H. M., Leadley, P. W., Proença, V., Alkemade, R., Jörn, P. W., Fernandez- manjarrés, J. F., Sumaila, U. R. (2010). Scenarios for Global Biodiversity in the 21st Century. Science, 330, 1496–1501. Rands, M. R. W., Adams, W. M., Bennun, L., Butchart, S. H. M., Clements, A., Coomes, D.,Vira, B. (2010). Biodiversity Conservation: Challenges Beyond 2010. Science, 329(5997), 1298–1303. https://doi.org/10.1126/science.1189138 Riggio, J., & Caro, T. (2017). Structural connectivity at a national scale : Wildlife corridors in Tanzania. PLoS One, 12(11), 1–16. Tanzania, United Republic of (2008). The Wildlife Policy of Tanzania. TAWIRI. (2010). Tanzania elephant management plan 2010-2015.

155 Supporting materials

Figure 6-S 1. Evanno steps for detection of the true number of clusters, K

(A) Delta K [calculated as ∆K = m|L′′(K)|/ s[L(K)], providing the uppermost level of structure (B) Mean L(K) (± SD) over ten runs for each K value. The model considered here is a hierarchical island model using 10 SSR loci.

Young (0-20)

Old (20-60)

156

0-12 years

13-17 years

18-25 years

157 25-60 years

Females

Males

158 Males above 18 years ( K=1)

159 : Dissertation synthesis

This is the first broad genetic study of African elephants in Tanzania our research has covered four large ecosystems with a significant number of elephants. We have obtained multiple level genetic data which informs us about the extent of genetic diversity in Tanzania but we can also use this information to plan how best to manage elephant populations. Our data can be used to plan the restoration important wildlife corridors in Tanzania. Through this research, we have identified 26 unique mitochondrial haplotypes most of which were unique to some protected areas. These unique haplotypes together with nuclear markers can be used in forensic evidence for seized elephant ivory. We have revealed unexpected genetic relationship between subpopulations, for instance, between Ruaha and Tarangire although they are geographically far from each other. This chapter provides a comprehensive summary for our findings and provide recommendations for the management of elephant population in Tanzania. We were surprised to observe a significant genetic differentiation between Manyara and Tarangire. These protected areas are within (60 km) the same ecosystem. The movement of elephants between Manyara Ranch and Tarangire has been documented (Kikoti, et al., 2009). Other large mammals such as wildebeest (Morrison et al., 2016) and giraffes (Lee et al., 2016) have also been moving between Manyara and Tarangire. Despite this movement between Manyara and Tarangire, elephants do not seem to mix that much. Unfortunately, no data exist to understand the extent of gene flow before the wildlife corridors became less viable. To fully understand the impacts of habitat fragmentation on long-lived mammals like elephants we need to conduct further research from a population which is known to be isolated. Rubondo Island National Park would offer a unique opportunity to understand this phenomenon. Elephants in Rubondo island National Park in Tanzania, were introduced in the 1960s from Serengeti ecosystem. Currently there are about 102 elephants (Mwambola et al.,2016). This population has been isolated for more than 40 years. We recommend the use same nuclear and mitochondrial DNA markers to make a comparison with the rest of populations from this study.

160 We have shown that there is significant genetic differentiation between the Greater Serengeti and Tarangire-Manyara ecosystems. The rift valley separates the elephant populations into two distinct populations using both the mitochondrial DNA and nuclear data. Distribution of mtDNA haplotypes in northern Tanzania is consistent with previous work (Ishida et al., 2013; Ahlering et al., 2012). Since our study sampled a large area, we have identified 26 unique haplotypes which we would otherwise not have obtained if we had sampled only a few individuals. We have added more information on the current genetic diversity of elephants in Tanzania from which future references can be made. Ngorongoro elephants showed an admixture between the Serengeti and Tarangire ecosystem. Recent gene flow, especially between Lake Manyara National Park and the Ngorongoro elephants, explains this admixture. We are confident there was gene flow between Lake Manyara and Ngorongoro. Now that the corridors between the NCA and MAR are blocked, Lake Manyara is isolated. For the past ten years, the number of elephants at Lake Manyara has remained low, about 36 in 2007, and 34 in 2014 (Blanc et al., 2007; TAWIRI, 2014). In the long run, we think the Lake Manyara elephant populations will face local extinction if they remain isolated. Historically it appears that there was little mixing between Manyara and Tarangire. We recommend that the corridors between Ngorongoro and Lake Manyara be restored as suggested in (Riggio & Caro, 2017). The Kwa kuchinja corridor between Tarangire and Manyara Ranch is vanishing (Morrison et al., 2016). Another intriguing observation is the difference in the haplotype distribution within the NCA (Figure 2-14). It is not coincident that for all sampled elephants in the Crater there were no elephants carrying haplotypes in southeast savanna subclade but, in Oldeani, these haplotypes were common. It seems elephants from Oldeani have not been mixing with the Crater elephants that much. Perhaps the Oldeani Mountains and valleys between Oldeani and Crater impedes the movement of elephants. Although there was a significant genetic differentiation between Manyara Ranch and Tarangire this corridor is still vital for gene flow and dispersal for wildlife between different seasons. Areas around Essimingori forest around Makuyuni area seems to a critical dispersal area for wildlife. During our study, we also observed some incidences of crop raiding near Makuyuni village, likely by elephants from Manyara Ranch. Also, near

161 Oldeani village, adjacent to Ngorongoro, a male elephant destroyed crops during our field trip in 2017 (Figure 7-1). A continuous assessment of elephants using areas outside formally protected areas is required for effective conservation and address human-elephant conflict. We have provided evidence of two groups of elephants recolonizing the Serengeti after poaching wiped out in 1900 (Sinclair et al., 2008). Both mtDNA and microsatellite markers show at least two distinct populations one from the south and one from the north. Interestingly we found that most Serengeti elephants were carrying haplotypes which fall into the F-clade, which is shared with the forest elephants. We have provided empirical evidence for the hypothesis put forward by Sinclair (2012) about the recolonization of elephants. Ishida et al., (2011) suggested that many generations of hybridization between forest and savanna generated the current mitochondrial DNA pattern (i.e., the past success of female hybrids can be inferred by the F-clade mtDNA derived from forest elephants in northern Tanzania savanna herds). Since the majority of elephants in Serengeti carried haplotypes in F-clade, they must have been the result of hybridization between the two species

Figure 7-1. Recorded incident of crop raiding by elephants in Oldeani village adjacent to the Ngorongoro conservation Area (from left William Metamei, NCA Game Officer, a farmer and George Lohay): Photo credit: James Madeli

162 Traditional protected area systems have long been considered the most effective way of protecting wildlife in Tanzania. Indeed, most wildlife is found within these protected areas. However, the role of open areas, especially those adjacent to protected areas, is essential for dispersal areas for large mammals like wildebeest and elephants. Here want to emphasize that open area such as Mwiba and Manyara ranch are great examples of open spaces with abundant wildlife. For instance, we observed more elephants in Mwiba Ranch and Manyara Ranch than in adjacent protected areas (i.e., Maswa GR and Lake Manyara NP respectively). Mwiba Ranch, for instance, was established in 2008, Makao village provided land to a private investor for tourism activities. In recent years, the Mwiba investors constructed a Makao police station, one community center and implemented a water project (Batro Nakoli Ngilangwa, unpublished report). Similarly, in Manyara Ranch has become a significant wildlife dispersal and home range for many wildlife. Currently, the number of elephants in Manyara Ranch is around 70 elephants which are twice the number of elephants in Lake Manyara National Park ((TAWIRI, 2015). Kioko et al., (2013) suggested that the area between Manyara Ranch and Lake Manyara should be included in the Wildlife Management Areas to facilitate connectivity between Manyara ranch and Lake Manyara elephants, which seems to be isolated. Establishment of Wildlife Management Areas (WMA) in Tanzania has increased habitats and protection of wildlife species. For example, Lee (2018) documented significantly higher densities of several wild ungulate species and lower frequencies of domestic ungulates in the Burunge WMA compared to adjacent villages. Burunge WMA is within the Tarangire-Manyara ecosystem and provides potential connectivity between Tarangire and Lake Manyara National Park. In chapter three, we have shown that there is a significantly different population structure between Ruaha and Selous ecosystems. Based on the haplotype distributions there is no evidence for female-mediated gene flow between these two ecosystems. Nuclear markers showed low but significant FST value indicating limited gene flow between them. Based on our results we recommend for another study that will cover the entire Selous ecosystem to evaluate if there is evidence of gene flow between Selous and Niassa GR in Mozambique through the Selous-Niassa corridor. Probably, the Selous elephants are more genetically like Mozambique elephants than the Ruaha elephants. On the other hand, the

163 Ruaha elephants show high similarity (both mtDNA and nuclear markers) with Tarangire elephants. Thus, at least for elephants, the wildlife corridors between Ruaha and Tarangire through Swagaswaga Game Reserve seems to have been crucial although the migratory routes are entirely blocked (Riggio & Caro, 2017). In chapters four and five, we have validated the use of fecal centric approach with the rapid demographic assessment to determine the age and sex structure of elephants. We designed a new method using AMELX/Y primers which were more effective than the previous approach (Ahlering et al., 2011). The use of our new sexing method together with measurement of dung circumference to estimate the age of elephants can be used to assess age and sex structure particularly in areas where direct observation is difficult. This approach is effective for elephants older than 10 years. Although the rapid demographic method has been used in several studies, it is likely affected by observer bias. These errors can be minimized by using the fecal centric approach suggested in this work. Our results showed a skewed sex structure in favor of females for adult age classes in Serengeti but biased in favor of males in the Ngorongoro Conservation Area. Poaching of males could explain the trend observed in Serengeti. However, other factors such as differential survival rates between males and females in the Serengeti may have played a role in the observed deviations. In general, males have a lower survival rate than males (Turkalo et al., 2018; Wittemyer et al., 2013). However, in Serengeti, there is no long-term demography study to evaluate the survival rates of elephants. We recommend the establishment of a long-term research project to collect demography data to inform the management of changes that might happen in the park. Finally, in chapter six, we have identified early signs of genetic differentiation among elephants between the Greater Serengeti ecosystem and Tarangire Manyara ecosystems. Here we tested the hypothesis that the young elephants will be more genetically differentiated than the older elephants because of habitat fragmentation. We assumed that older elephants moved widely across the landscape and exchanged genes more than the young elephants which are more restricted to protected areas. For mammals with long life space and long generation time (of about 25 years), it takes a longer time to observe changes in population structure. However, we have seen some signs of genetic

164 differentiation which indicate that there is limited gene flow between these protected areas. Our results were comparable with other studies in western Tanzania that detected incipient signs of genetic differentiation among elephants because of habitat fragmentation in miombo woodlands (Lobora et al., 2018).

165 References

Ahlering, M. A., Eggert, L. S., Western, D., Estes, A., Munishi, L., Fleischer, R.,Maldonado, J. E. (2012). Identifying Source Populations and Genetic Structure for Savannah Elephants in Human-Dominated Landscapes and Protected Areas in the Kenya-Tanzania Borderlands. PLoS ONE, 7(12), e52288. https://doi.org/10.1371/journal.pone.0052288 Ahlering, M. A., Hailer, F., Roberts, M. T., & Foley, C. (2011). A simple and accurate method to sex savannah, forest and Asian elephants using noninvasive sampling techniques. Molecular Ecology Resources, 11(5), 831–834. https://doi.org/10.1111/j.1755-0998.2011.03030.x Blanc, J. J., Barnes, R. F. W., Craig, G. C., Dublin, H. T., Thouless, C. R., Douglas- Hamilton, I., & Hart, J. A. (2007). African elephant status report 2007: an update from the African elephant database. African elephant status report 2007: an update from the African elephant database. https://doi.org/10.2305/iucn.ch.2007.ssc- op.33.en Ishida, Y., Demeke, Y., Van, P. J., De Groot, C., Georgiadis, N. J., Leggett, K. E. A., Roca, A. L. (2011). Distinguishing Forest and Savanna African Elephants Using Short Nuclear DNA Sequences. Journal of Heredity, 102(5), 610–616. https://doi.org/10.1093/jhered/esr073 Ishida, Y., Georgiadis, N. J., Hondo, T., & Roca, A. L. (2013). Triangulating the provenance of African elephants using mitochondrial DNA. Evolutionary Applications, 6(2), 253–265. https://doi.org/10.1111/j.1752-4571.2012.00286.x Kioko, J., Zink, E., Sawdy, M., & Kiffner, C. (2013). Elephant ( Loxodonta africana ) Tarangire-Manyara Ecosystem , Tanzania. South African Journal of Wildlife Research, 43(1), 44–51. Lee, D. E., Kissui, B. M., Kiwango, Y. A., & Bond, M. L. (2016). Migratory herds of wildebeests and zebras indirectly affect calf survival of giraffes. Ecology and Evolution, 6(23), 8402–8411. https://doi.org/10.1002/ece3.2561

166 Lee, D. E. (2018). Evaluating conservation effectiveness in a Tanzanian community wildlife management area. Journal of Wildlife Management, 82(8), 1767–1774. https://doi.org/10.1002/jwmg.21549 Lobora, A. L., Nahonyo, C. L., Munishi, L. K., Caro, T., Foley, C., Prunier, J. G., … Eggert, L. S. (2018). Incipient signs of genetic differentiation among African elephant populations in fragmenting miombo ecosystems in south-western Tanzania. African Journal of Ecology, 56(4), 993–1002. https://doi.org/10.1111/aje.12534 Morrison, T. A., Link, W. A., Newmark, W. D., Foley, C. A. H., & Bolger, D. T. (2016). Tarangire revisited: Consequences of declining connectivity in a tropical ungulate population. Biological Conservation, 197, 53–60. https://doi.org/10.1016/j.biocon.2016.02.034 Mwambola, S., Ijumba, J., Kibasa, W., Masenga, E., Eblate, E., & Munishi, L. (2016). Population size estimates and distribution of the African elephant using the dung surveys method in Rubondo Island National Park, Tanzania. International Journal of Biodiversity and Conservation, 8(6), 113–119. https://doi.org/10.5897/IJBC2015.0873 Riggio, J., & Caro, T. (2017). Structural connectivity at a national scale : Wildlife corridors in Tanzania. PLoS One, 12(11), 1–16. Sinclair, A.R.E, (2012), Serengeti Story: Life and science in the world’s greatest wildlife region, Oxford University Press, UK. Sinclair, A. R. E., Hopcraft, J. G. C., Olff, H., & Mduma, S. A. R. (2008). Historical and Future Changes to the Serengeti Ecosystem. University of Chicago Press, (1972), 7– 46. TAWIRI. (2014). Aerial Total Count of Elephants and Buffaloes in the Serengeti-Mara Ecosystem. TAWIRI. (2015). Population Status of Elephant in Tanzania 2014.TAWIRI Aerial Survey Report. Turkalo, A. K., Wrege, P. H., & Wittemyer, G. (2018). Demography of a forest elephant population. PLOS ONE, 13(2), e0192777. https://doi.org/10.1371/journal.pone.0192777

167

Wittemyer, G., Daballen, D., & Douglas-Hamilton, I. (2013). Comparative Demography of an At-Risk African Elephant Population. PLoS ONE, 8(1), e53726. https://doi.org/10.1371/journal.pone.0053726

168

APPENDICES

Appendix A: Genotype data for the African Savanna elephants at 10 SSR loci from fecal samples collected between 2015 and 2017 in Tanzania and the GPS location where each sample was collected.

Sample Loc LafMS2 LafMS2 LA5 LA5 LA6 LA6 Lat06 Lat06 Lat08 Lat08 Lat13 Lat13 FH19 FH19 Lat24 Lat24 FH48 FH48 FH67 FH67 Lat Long

SNP166 NSE 143 143 144 144 164 172 280 280 186 202 236 244 187 189 216 216 170 170 96 96 -2.494898 34.819075

SNP4 NSE 139 147 139 152 164 164 365 380 190 206 244 248 187 189 220 224 170 174 96 98 -2.548760 34.836790

SNP167 NSE 143 147 144 144 164 172 276 284 194 210 227 244 176 187 216 224 173 175 96 96 -2.492959 34.818601

SNP168 NSE 145 149 144 152 164 164 280 280 194 194 240 240 187 195 216 236 173 173 98 98 -2.441237 34.799999

SNP169 NSE 143 145 144 152 164 172 284 361 190 202 244 248 187 195 224 232 174 174 96 106 -2.441029 34.800156

SNP170 NSE 147 151 144 152 164 172 276 384 190 206 240 244 184 193 220 228 170 174 96 104 -2.440736 34.800390

SNP171 NSE 138 145 139 144 164 172 280 361 198 202 240 248 187 189 216 220 170 173 96 96 -2.524698 34.887275

SNP172 NSE 143 147 144 144 164 172 361 361 194 198 236 236 187 187 212 220 170 172 98 98 -2.525353 34.886868

SNP173 NSE 138 149 139 144 164 172 280 361 190 202 218 240 187 193 212 220 170 170 98 106 -2.524698 34.887275

SNP174 NSE 145 145 144 152 164 172 276 280 194 214 236 240 176 187 220 224 173 179 98 98 -2.524906 34.886017

SNP175 NSE 147 147 145 152 164 174 280 337 198 214 236 244 187 187 216 232 170 170 92 106 -2.557637 35.125691

SNP176 NSE 138 145 144 152 164 174 276 350 202 217 240 244 187 187 216 220 168 174 96 98 -2.680658 34.800125

SNP177 NSE 145 149 144 144 164 174 280 350 194 194 244 252 187 191 216 224 172 179 98 98 -2.680704 34.800088

SNP178 NSE 147 149 144 145 164 164 280 280 194 206 236 236 187 189 220 224 166 174 96 96 -2.715930 34.775827

SNP179 NSE 145 145 144 145 164 164 280 361 198 202 240 244 187 195 224 228 170 175 96 98 -2.735615 34.837791

SNP180 NSE 147 149 144 144 164 172 280 361 190 194 240 244 187 195 224 228 170 175 96 98 -2.731072 34.835372

SNP181 NSE 145 151 144 144 164 172 280 280 194 202 227 236 187 195 212 216 170 174 98 98 -2.730523 34.833816

SNP182 NSE 139 147 139 145 164 172 280 280 206 206 236 244 187 195 220 220 170 173 96 96 -2.729346 34.833281

SNP183 NSE 147 147 152 152 164 172 280 361 194 194 220 236 187 187 220 220 173 173 96 96 -2.720494 34.825942

169

SNP2 NSE 147 147 152 152 164 194 280 280 198 206 236 236 189 189 220 224 170 170 102 106 -2.549200 34.837216

SNP201 NSE 147 151 144 144 164 172 280 361 198 214 244 244 189 189 216 220 170 175 96 98 -2.334949 34.838430

SNP202 NSE 143 149 144 144 164 172 284 361 190 214 236 244 187 187 216 216 170 173 98 104 -2.335012 34.838007

SNP203 NSE 147 151 144 144 164 172 361 361 198 202 240 244 189 190 208 220 170 175 96 96 -2.334959 34.838811

SNP204 NSE 145 147 144 144 164 156 276 284 198 214 0 0 189 195 232 236 170 174 96 96 -2.236930 34.921582

SNP205 NSE 145 147 144 144 164 172 280 357 194 202 240 240 189 189 228 232 173 173 96 96 -2.236941 34.921599

SNP206 NSE 149 149 152 152 164 172 280 357 202 206 236 244 184 187 216 220 174 175 98 98 -2.238329 34.928225

SNP207 NSE 147 151 144 152 164 172 280 369 194 206 236 240 189 205 220 220 170 174 92 98 -2.420649 34.934781

SNP208 NSE 147 151 144 152 164 172 280 280 194 214 236 240 189 205 216 220 170 172 92 96 -2.420655 34.934773

SNP209 NSE 138 145 144 152 164 172 276 276 194 194 244 244 187 189 220 236 170 172 96 96 -2.420661 34.934758

SNP3 NSE 139 147 144 144 164 194 280 361 190 206 236 240 189 189 140 216 172 174 98 104 -2.549170 34.837511

SNP119 NSE 147 147 144 152 164 172 280 369 198 202 236 240 189 189 224 228 168 169 96 96 -1.812986 35.220278

SNP130 NSE 138 138 144 152 164 172 361 365 194 202 240 248 187 187 216 232 169 169 96 96 -1.707509 35.200628

SNP149 NSE 139 145 139 144 164 172 276 276 186 217 236 248 187 187 224 228 170 174 92 98 -1.601485 35.083783

SNP160 NSE 139 145 152 152 164 164 280 280 198 202 244 248 187 189 216 224 170 174 96 104 -1.564732 34.995690

SNP162 NSE 147 147 144 144 164 172 276 361 198 198 240 248 189 205 0 0 170 174 96 98 -2.508436 34.863748

SNP163 NSE 147 149 144 144 164 172 361 361 194 202 236 240 187 187 212 224 170 170 96 98 -2.507558 34.866744

SNP164 NSE 143 147 152 152 164 172 346 353 194 214 236 240 187 205 220 224 174 179 98 98 -2.512365 34.868109

SNP165 NSE 145 149 144 152 164 164 280 284 202 210 244 248 187 187 212 216 174 174 96 98 -2.497148 34.819053

SNP99 NSE 147 149 144 152 164 172 280 361 194 198 236 248 190 195 224 224 170 174 96 98 -2.389424 34.819553

SNP115 NSE 138 138 144 144 164 163 280 284 186 194 240 248 184 189 220 220 168 169 96 106 -1.834475 35.228539

SNP210 NSE 147 147 144 152 164 164 280 280 198 202 236 240 189 189 216 236 173 175 92 96 -2.420643 34.934760

SNP211 NSE 145 149 144 144 164 164 280 361 190 202 236 240 187 187 220 220 173 173 96 98 -2.414218 34.938333

SNP212 NSE 147 147 144 144 164 164 276 346 190 198 232 240 187 195 220 236 173 173 94 98 -2.414214 34.938340

SNP213 NSE 145 147 144 152 164 164 280 365 194 198 236 248 187 189 220 236 172 175 96 96 -2.413390 34.938565

SNP214 NSE 149 149 152 152 164 172 280 280 194 198 240 244 187 193 224 232 169 174 96 104 -2.424687 34.950480

SNP215 NSE 145 149 144 152 164 164 361 365 190 190 240 240 187 191 131 212 170 174 94 96 -2.458197 34.978683

SNP216 NSE 145 149 144 152 164 172 280 365 194 206 236 244 189 195 212 220 170 174 98 106 -2.218310 34.962791

SNP217 NSE 145 147 144 144 164 172 280 280 194 214 236 244 187 189 212 228 170 174 98 106 -2.218175 34.962369

SNP218 NSE 147 151 144 152 164 172 280 357 194 214 232 240 187 187 208 220 170 174 98 98 -2.218175 34.962369

170

SNP219 NSE 149 151 152 152 164 172 280 361 194 214 236 248 187 189 216 224 170 174 96 98 -1.642520 34.904970

SNP220 NSE 139 147 139 152 164 172 361 361 206 214 240 244 184 193 216 232 173 175 98 98 -1.552042 34.881002

SNP221 NSE 139 145 139 144 164 172 280 361 190 206 236 240 187 187 216 244 173 175 98 98 -1.552033 34.880991

SNP222 NSE 147 149 144 144 164 172 361 361 190 202 240 248 187 187 216 216 170 173 96 106 -1.543310 34.866101

SNP223 NSE 145 145 144 152 164 164 361 365 202 202 248 248 191 199 216 228 170 174 96 98 -1.543441 34.866171

SNP224 NSE 145 147 144 152 164 164 280 361 190 202 248 248 187 191 224 228 173 175 98 98 -1.543431 34.866167

SNP225 NSE 145 147 144 152 164 164 280 361 206 206 236 244 189 189 220 228 173 173 96 106 -1.543330 34.866129

SNP226 NSE 145 145 144 144 164 164 284 361 198 198 236 240 187 189 212 224 170 173 96 96 -1.534014 34.724866

SNP227 NSE 145 151 144 144 164 164 288 384 198 206 236 240 187 189 224 232 170 173 96 96 -1.534293 34.724810

SNP228 NSE 147 149 144 152 164 172 279 279 198 206 236 244 187 189 220 232 172 174 96 98 -1.643131 35.178055

SNP229 NSE 139 143 139 152 164 172 353 353 202 206 240 244 187 189 220 232 170 174 96 96 -1.643260 35.178322

SNP230 NSE 143 143 144 144 164 172 280 280 194 214 236 244 189 205 220 224 168 169 96 96 -1.777309 35.177306

SNP116 NSE 138 151 144 152 164 163 284 361 206 214 244 248 187 189 220 220 170 174 96 106 -1.833248 35.228388

SNP117 NSE 149 149 144 152 164 163 280 280 194 202 240 244 187 189 228 236 170 174 96 98 -1.813232 35.220550

SNP118 NSE 138 147 144 144 164 172 280 369 198 214 240 244 189 189 212 228 169 169 96 98 -1.813218 35.220539

SNP120 NSE 147 149 152 152 164 163 280 357 198 202 240 244 187 189 228 228 170 174 96 98 -1.812973 35.220272

SNP121 NSE 147 147 144 152 164 163 280 369 198 210 236 244 187 195 220 224 169 174 96 98 -1.812722 35.220424

SNP122 NSE 138 138 144 144 164 172 357 361 198 206 240 240 187 195 220 236 168 170 96 96 -1.812400 35.220326

SNP123 NSE 145 145 144 144 164 172 280 361 194 206 236 236 187 195 212 220 170 174 98 104 -1.812707 35.220483

SNP124 NSE 138 138 144 144 164 163 280 369 194 194 244 248 187 205 208 220 170 170 96 106 -1.812444 35.220200

SNP125 NSE 138 138 139 139 164 172 280 280 198 214 236 248 187 187 220 224 169 169 96 96 -1.812451 35.220196

SNP126 NSE 138 145 144 152 164 163 280 280 206 206 236 236 189 195 224 228 170 174 96 101 -1.802404 35.217143

SNP127 NSE 139 145 139 139 164 172 280 280 194 202 236 236 187 193 212 220 173 173 96 98 -1.802404 35.217133

SNP128 NSE 139 145 139 139 164 163 280 361 194 194 236 240 187 189 220 232 170 170 96 96 -1.709620 35.201729

SNP129 NSE 139 147 144 145 164 172 280 361 194 194 240 244 187 189 228 228 173 173 96 96 -1.708470 35.200882

SNP131 NSE 139 145 144 152 164 172 280 361 206 214 236 244 187 187 232 232 170 175 96 98 -1.707513 35.200642

SNP132 NSE 139 145 139 139 164 163 280 361 194 214 236 240 187 189 232 232 168 169 96 96 -1.706464 35.200068

SNP133 NSE 147 151 144 152 164 163 280 365 190 206 240 248 189 189 224 228 170 174 96 106 -1.703904 35.274048

SNP134 NSE 149 151 152 152 164 169 280 357 190 202 240 240 193 193 224 224 170 174 96 106 -1.703904 35.274048

SNP135 NSE 147 149 144 152 164 172 280 280 202 214 240 244 189 193 216 220 170 174 96 96 -1.703992 35.273964

171

SNP136 NSE 139 145 144 144 164 163 280 361 190 202 240 244 189 189 208 224 173 173 96 96 -1.703992 35.273964

SNP137 NSE 147 147 144 144 164 172 280 280 198 202 236 244 193 195 212 224 170 170 96 96 -1.707994 35.273222

SNP138 NSE 138 147 144 144 164 172 276 280 194 202 240 244 187 195 232 236 170 174 96 98 -1.707994 35.273222

SNP139 NSE 151 151 144 152 164 172 280 361 190 202 244 244 187 193 208 220 170 170 96 98 -1.708002 35.273128

SNP140 NSE 138 145 144 152 164 172 280 361 198 202 240 244 189 193 212 224 166 170 96 98 -1.707983 35.273114

SNP141 NSE 139 149 144 144 164 172 0 0 194 198 236 236 189 193 208 228 170 174 96 98 -1.707862 35.273143

SNP142 NSE 147 147 144 144 164 163 280 280 194 202 240 244 189 195 208 220 166 170 96 98 -1.707858 35.273151

SNP143 NSE 147 151 144 144 164 172 280 280 194 202 236 236 184 189 208 220 166 174 98 98 -1.707814 35.273068

SNP144 NSE 139 147 144 144 164 172 280 280 198 202 236 244 187 189 208 228 170 170 96 98 -1.707814 35.272960

SNP145 NSE 138 145 144 144 164 164 280 280 190 202 236 244 189 189 208 220 170 174 96 104 -1.707718 35.273091

SNP146 NSE 139 145 139 144 164 172 276 280 186 202 240 244 176 189 208 216 170 174 98 98 -1.707702 35.273068

SNP147 NSE 138 151 139 144 164 172 276 276 190 217 240 248 189 195 220 220 175 175 92 96 -1.602191 35.083106

SNP148 NSE 141 149 139 144 164 172 276 276 206 217 240 248 187 205 224 224 170 170 96 106 -1.601515 35.083782

SNP150 NSE 138 149 139 152 164 172 276 276 210 217 248 248 187 195 220 224 170 174 96 96 -1.601131 35.083771

SNP151 NSE 147 149 139 152 164 172 276 276 190 217 248 248 187 187 224 224 175 175 92 96 -1.601736 35.084384

SNP152 NSE 139 145 144 144 164 172 276 361 206 214 240 240 189 193 216 224 170 174 98 98 -1.563531 34.995640

SNP153 NSE 141 145 152 152 164 172 276 276 206 206 244 248 189 193 220 220 169 174 92 92 -1.563115 34.995388

SNP154 NSE 139 145 145 152 164 172 271 271 190 210 244 248 187 189 216 220 170 174 92 96 -1.563043 34.995249

SNP155 NSE 139 141 152 152 164 172 276 280 206 220 244 248 189 193 220 220 170 174 92 92 -1.564722 34.995703

SNP156 NSE 145 145 152 152 164 172 357 361 182 214 240 244 187 191 216 220 169 174 98 101 -1.564733 34.995711

SNP157 NSE 145 151 139 152 164 172 280 365 198 202 218 248 187 199 216 224 170 174 98 104 -1.564732 34.995723

SNP158 NSE 139 139 144 145 164 172 361 361 182 206 236 244 187 189 212 216 169 174 96 101 -1.564730 34.995699

SNP159 NSE 139 145 144 145 164 172 280 280 202 206 244 244 189 189 216 220 170 174 96 98 -1.564741 34.995710

SNP161 NSE 147 147 152 152 164 164 280 280 198 221 240 240 189 191 212 216 174 175 98 98 -2.445636 34.861423

LGCA14 NSE 145 151 144 144 164 164 280 346 186 206 244 248 189 199 216 240 169 169 96 96 -2.112943 35.279210

LGCA15 NSE 147 151 144 144 164 172 280 280 194 206 236 244 193 195 216 220 174 174 98 104 -2.114357 35.281811

LGCA16 NSE 139 145 144 144 164 164 357 365 190 206 236 244 195 197 212 224 173 173 98 98 -2.115635 35.281678

LGCA17 NSE 145 145 144 144 164 172 361 365 206 214 236 240 189 193 220 224 170 174 96 106 -2.124443 35.282271

LGCA18 NSE 145 145 144 144 164 164 361 372 198 206 240 244 189 189 224 224 170 170 96 98 -2.051740 35.262214

LGCA22 NSE 139 145 144 144 164 164 357 361 194 202 244 248 187 189 216 232 170 170 96 98 -2.009274 35.251522

172

LGCA23 NSE 147 147 144 152 164 164 361 361 190 214 236 248 187 187 216 220 170 170 92 98 -2.009329 35.251579

LGCA24 NSE 145 149 152 152 164 172 280 365 190 202 236 240 187 187 212 236 170 174 98 102 -2.009391 35.251687

LGCA25 NSE 143 143 144 144 164 164 280 280 194 202 244 244 189 195 216 224 170 174 98 104 -2.009208 35.251615

LGCA26 NSE 139 147 144 144 164 172 280 361 194 194 240 248 187 187 216 224 170 174 96 101 -2.009018 35.251820

LGCA27 NSE 139 147 144 144 164 172 276 361 190 214 240 240 187 193 220 228 170 174 96 98 -2.009135 35.251961

LGCA28 NSE 141 145 144 144 164 164 280 384 198 214 240 248 187 189 224 224 170 174 98 101 -1.998921 35.252566

LGCA29 NSE 143 147 144 152 164 172 280 384 194 214 240 244 187 187 224 224 170 174 98 100 -1.998922 35.252547

LGCA30 NSE 139 143 144 152 164 172 280 384 194 214 240 244 187 187 224 224 170 174 98 101 -1.999002 35.252563

LGCA31 NSE 139 143 144 144 164 172 357 365 190 194 240 240 187 189 220 236 170 174 96 98 -1.998995 35.252573

LGCA32 NSE 143 145 144 144 164 172 357 365 190 194 240 240 187 189 220 236 170 174 96 98 -1.998963 35.252610

LGCA33 NSE 141 149 144 144 164 172 280 280 198 198 240 248 187 189 224 224 173 173 94 96 -1.998933 35.252660

LGCA34 NSE 139 145 144 144 164 164 357 361 194 202 244 248 187 189 216 232 169 169 96 98 -1.998761 35.252800

LGCA35 NSE 139 145 152 164 164 172 361 365 194 202 248 248 187 187 220 224 170 170 96 98 -1.996421 35.252029

LGCA36 NSE 139 147 144 144 164 172 280 361 194 194 240 248 187 187 220 228 170 174 96 101 -1.996859 35.251825

LGCA37 NSE 147 149 144 152 164 164 361 361 190 214 236 248 184 187 216 220 168 170 92 98 -1.996630 35.251917

LGCA38 NSE 149 151 144 152 164 164 276 353 186 206 240 244 193 195 224 224 170 174 96 98 -2.092617 35.209137

LGCA4 NSE 139 145 141 152 164 172 276 361 194 202 236 240 184 199 220 228 175 175 96 98 -1.946321 35.236162

LGCA40 NSE 147 151 144 152 164 172 280 280 190 198 236 240 189 193 216 228 173 173 94 96 -2.092433 35.209306

LGCA42 NSE 147 149 144 144 164 164 361 361 194 202 236 244 187 189 212 224 170 174 96 98 -2.091785 35.210252

LGCA43 NSE 139 145 144 152 164 172 284 288 194 194 216 248 187 187 216 216 173 175 96 98 -2.091785 35.210252

LGCA44 NSE 147 149 144 152 164 172 357 361 206 210 236 240 189 189 216 224 173 173 96 106 -2.091549 35.210482

LGCA45 NSE 149 151 144 152 164 172 276 280 186 198 240 244 187 187 216 220 173 173 96 98 -2.091729 35.210495

LGCA46 NSE 145 149 144 150 164 172 276 280 186 202 240 244 193 195 224 224 170 174 100 102 -2.091483 35.210946

LGCA47 NSE 145 147 152 164 164 172 280 365 190 190 244 244 187 195 224 224 170 174 96 98 -2.091428 35.211018

LGCA48 NSE 139 145 139 152 164 164 361 361 190 198 236 244 187 187 212 224 170 174 98 104 -2.090724 35.212366

LGCA49 NSE 149 151 144 152 164 172 276 357 206 206 244 244 193 195 224 224 174 174 94 98 -2.090724 35.212366

LGCA5 NSE 145 145 144 144 164 172 280 284 194 214 240 244 187 187 224 224 173 175 98 98 -1.944349 35.236717

LGCA50 NSE 151 151 144 152 164 172 276 357 206 206 244 244 193 195 224 224 174 174 94 98 -2.090724 35.212366

LGCA8 NSE 145 151 144 152 164 172 280 280 210 214 236 244 187 187 228 228 173 175 96 104 -1.944631 35.236657

SNP51 NSE 145 147 144 144 164 163 280 337 198 202 240 248 189 193 212 216 170 174 96 98 -2.421984 34.802793

173

SNP52 NSE 139 145 144 144 164 172 280 365 202 202 236 236 187 189 220 236 170 174 98 98 -2.422008 34.802971

SNP53 NSE 139 145 144 144 164 164 280 384 190 221 236 248 187 189 228 236 174 174 96 104 -2.219754 34.391325

SNP54 NSE 149 151 144 144 164 172 280 350 190 202 240 248 189 189 200 224 173 173 96 104 -2.219745 34.391281

SNP55 NSE 145 151 152 152 164 172 280 361 202 210 236 244 187 189 228 236 170 170 96 98 -2.219671 34.391565

SNP56 NSE 139 145 144 152 164 164 280 280 198 198 244 248 189 189 220 224 170 174 96 96 -2.220135 34.392500

SNP57 NSE 138 147 144 144 164 164 280 361 206 225 236 240 187 189 212 228 174 174 94 96 -2.220143 34.392486

SNP58 NSE 138 147 144 152 164 164 280 365 190 198 236 240 187 187 0 0 170 174 96 104 -2.219168 34.390679

SNP59 NSE 138 149 144 152 164 172 280 284 190 190 220 236 187 195 208 224 174 175 96 98 -2.219168 34.390679

SNP60 NSE 139 147 144 152 164 163 280 361 198 225 236 236 187 195 216 220 170 170 102 104 -2.219153 34.390674

SNP62 NSE 149 149 144 144 164 163 280 353 198 206 236 240 187 193 224 240 170 174 98 101 -2.218929 34.390604

SNP63 NSE 147 147 144 152 164 172 276 361 186 194 236 244 189 193 220 236 170 174 96 106 -2.218202 34.390075

SNP64 NSE 139 145 152 152 164 163 276 280 190 202 232 240 187 189 216 220 170 174 96 98 -2.218235 34.390078

SNP65 NSE 147 147 144 152 164 164 280 365 190 210 236 236 187 189 216 220 174 174 96 98 -2.218145 34.389197

SNP66 NSE 149 149 144 152 164 164 280 365 198 206 236 240 187 189 216 220 170 174 96 98 -2.218151 34.389186

SNP67 NSE 145 151 144 144 164 172 280 342 190 194 236 244 187 195 216 244 170 174 96 96 -2.217537 34.388976

SNP68 NSE 145 147 144 152 164 163 280 365 190 210 236 236 187 189 216 220 173 173 96 98 -2.215858 34.387273

SNP69 NSE 145 149 139 144 164 172 280 384 186 214 240 244 189 193 220 224 170 170 96 98 -2.213225 34.375515

SNP70 NSE 145 149 144 144 164 172 280 376 194 206 240 240 189 189 220 228 170 174 96 104 -2.104746 33.869884

SNP71 NSE 139 147 152 152 164 172 280 384 186 202 236 236 189 193 208 236 170 174 98 102 -2.104746 33.869884

SNP72 NSE 145 149 144 144 164 172 280 376 194 206 240 240 187 189 220 228 170 174 96 104 -2.104741 33.869895

SNP73 NSE 139 139 139 144 164 163 284 353 186 194 236 240 176 195 224 232 170 170 96 98 -2.104668 33.869971

SNP74 NSE 145 151 139 144 164 172 284 365 190 210 227 236 189 193 216 220 174 180 96 98 -2.104686 33.869962

SNP75 NSE 139 147 152 152 164 172 280 384 186 202 236 236 189 193 208 236 170 174 98 102 -2.104581 33.869964

SNP76 NSE 139 147 141 144 164 172 280 361 198 206 236 240 187 189 208 224 170 174 98 106 -2.104569 33.869973

SNP77 NSE 147 149 144 144 164 172 280 365 182 194 240 240 187 199 216 224 169 169 104 106 -2.104549 33.869953

SNP78 NSE 145 147 144 152 164 172 280 284 194 198 248 248 188 188 224 224 170 174 96 96 -2.104563 33.870027

SNP79 NSE 138 145 144 144 164 172 280 361 194 202 240 244 187 189 220 228 170 174 96 96 -2.103765 33.870133

SNP80 NSE 139 147 139 144 164 172 280 361 190 198 240 248 189 189 216 224 174 175 96 98 -2.103697 33.870287

SNP82 NSE 147 149 144 152 164 172 276 280 194 194 244 248 187 187 220 224 170 175 96 96 -2.103626 33.870331

SNP83 NSE 145 147 144 152 164 172 280 284 194 198 248 248 189 189 224 224 170 174 96 96 -2.104217 33.869828

174

SNP84 NSE 145 149 144 144 164 172 280 376 194 206 240 240 189 189 220 228 169 174 96 104 -2.104300 33.869836

SNP85 NSE 139 147 144 144 164 172 361 369 190 214 236 240 184 187 212 228 170 175 98 104 -2.104392 33.869869

SNP86 NSE 145 145 141 144 164 172 280 361 194 214 216 240 187 187 224 224 170 174 96 106 -2.104460 33.869860

SNP87 NSE 147 147 144 152 164 163 280 365 190 214 236 236 187 189 224 236 172 172 94 98 -2.104036 33.869026

SNP88 NSE 139 147 144 144 164 172 280 280 202 206 240 244 187 189 216 224 172 173 98 98 -2.104058 33.869036

SNP89 NSE 147 147 144 144 164 172 361 369 190 214 236 240 184 187 212 228 170 175 98 104 -2.104035 33.869045

SNP90 NSE 147 147 152 152 164 172 280 384 186 202 236 236 189 193 208 236 169 173 98 102 -2.104016 33.869035

SNP91 NSE 145 145 144 152 164 172 365 384 202 214 240 240 187 195 224 236 170 170 98 98 -2.095577 34.640853

SNP92 NSE 145 145 144 144 164 172 365 384 194 202 236 240 187 187 224 224 168 169 98 98 -2.096721 34.640275

SNP93 NSE 147 147 144 145 164 172 280 384 194 217 236 236 187 187 208 224 169 174 98 101 -2.096731 34.640289

SNP94 NSE 145 145 139 152 164 172 361 361 194 202 236 236 187 187 216 224 170 174 96 104 -2.096723 34.640296

SNP95 NSE 145 145 144 152 164 163 361 369 190 194 236 240 187 187 208 220 173 175 96 98 -2.096068 34.639677

SNP96 NSE 143 143 144 144 164 172 280 361 194 194 216 244 187 189 208 224 170 170 96 96 -2.093983 34.640936

SNP97 NSE 144 144 144 152 164 163 353 384 194 210 0 0 197 199 208 236 172 173 94 96 -2.093987 34.640935

SNP98 NSE 139 139 144 145 164 172 280 280 194 210 236 248 187 197 208 236 170 174 94 96 -2.093957 34.640941

GGR35 NSE 138 145 144 152 164 172 276 342 190 194 244 244 187 189 224 232 174 174 96 104 -2.116859 34.471190

GGR36 NSE 144 144 144 144 164 172 276 280 190 206 236 244 184 189 220 240 170 174 98 104 -2.116859 34.471190

GGR37 NSE 145 151 144 144 164 172 276 350 190 190 240 244 184 191 220 224 170 174 96 98 -2.116859 34.471190

GGR39 NSE 144 144 144 144 164 172 276 350 190 198 240 244 189 191 224 232 174 174 96 98 -2.116859 34.471190

GGR40 NSE 144 144 144 144 164 172 276 342 190 194 236 244 185 187 224 240 170 174 98 104 -2.117058 34.470780

GGR41 NSE 145 147 144 144 164 172 276 376 190 198 236 244 187 189 232 240 174 174 98 104 -2.120313 34.467409

GGR42 NSE 145 147 144 144 164 172 276 342 194 198 236 240 185 187 224 228 170 174 96 104 -2.120313 34.467409

GGR43 NSE 147 147 139 144 164 172 280 376 198 214 236 236 189 189 232 236 170 174 98 106 -2.120313 34.467409

GGR44 NSE 145 151 144 144 164 172 276 280 190 198 220 236 187 189 208 244 170 174 98 101 -2.120313 34.467409

GGR45 NSE 145 147 144 152 164 172 280 280 206 206 236 236 184 195 216 220 174 174 98 106 -2.120313 34.467409

GGR46 NSE 145 147 152 152 164 164 361 365 194 214 240 248 187 189 212 236 172 174 96 96 -2.080614 34.647085

GGR47 NSE 147 147 144 144 164 172 361 369 214 214 240 240 193 193 220 224 170 175 94 96 -2.080492 34.647671

GGR48 NSE 149 149 144 144 164 172 280 384 190 198 236 236 187 189 220 228 170 170 98 98 -2.067690 34.666865

GGR49 NSE 138 147 139 144 164 172 280 361 210 214 236 248 187 193 224 224 170 174 96 98 -1.990453 34.736237

GGR50 NSE 145 151 144 144 164 172 284 361 214 214 244 248 187 187 220 220 170 174 98 98 -2.050929 34.698741

175

GGR51 NSE 145 147 144 144 164 172 280 284 194 214 236 248 187 187 212 220 170 170 98 98 -2.048064 34.700564

GGR52 NSE 143 145 144 144 164 164 280 365 190 198 240 248 187 187 216 228 170 174 92 98 -2.048064 34.700564

GGR53 NSE 138 147 144 152 164 172 284 284 194 206 240 248 187 189 212 232 170 174 96 98 -2.047422 34.700579

GGR54 NSE 145 149 144 150 164 172 280 280 206 206 236 240 187 205 212 232 170 170 98 98 -2.140850 34.615848

GGR55 NSE 145 147 144 152 164 172 276 276 194 206 236 236 187 187 216 216 170 174 96 98 -2.141011 34.616150

GGR57 NSE 139 147 144 144 164 164 276 280 190 194 236 244 184 189 212 216 170 175 98 104 -2.140546 34.616166

GGR58 NSE 145 145 144 144 164 172 276 280 198 202 220 248 187 189 224 224 170 174 98 98 -2.140552 34.616199

GGR59 NSE 147 149 139 139 164 164 280 284 194 198 240 244 187 189 216 220 170 174 96 106 -2.140745 34.615428

GGR60 NSE 139 145 144 152 164 172 276 376 186 190 236 244 187 187 212 232 170 174 98 98 -2.140753 34.615439

GGR61 NSE 143 145 144 152 164 172 280 361 190 190 240 244 189 195 216 228 170 174 96 98 -2.140791 34.615234

GGR62 NSE 138 147 139 152 164 164 280 280 198 202 220 236 187 189 220 224 170 175 96 96 -2.142477 34.614909

GGR63 NSE 149 149 139 139 164 164 276 280 198 198 220 236 187 189 216 228 170 174 96 98 -2.141705 34.608051

GGR64 NSE 145 147 144 152 164 164 276 280 198 206 220 236 187 189 224 224 170 174 96 98 -2.140643 34.606852

MGR01 SSE 143 147 144 152 164 172 361 361 190 202 240 244 187 189 220 228 174 175 92 98 -2.972470 34.475780

MGR03 SSE 138 145 144 152 164 172 361 361 186 190 240 240 187 191 220 236 170 174 98 106 -2.972572 34.475401

MGR04 SSE 138 145 144 144 164 164 280 365 190 214 240 244 187 191 212 224 174 175 96 98 -2.972808 34.474556

MGR05 SSE 145 147 139 144 164 172 280 361 190 194 240 248 189 191 212 236 166 174 92 98 -2.972748 34.474579

MGR06 SSE 145 145 144 152 164 172 280 361 186 194 240 244 187 193 216 220 174 175 98 104 -2.972896 34.474352

MGR07 SSE 145 147 144 144 164 164 280 361 190 206 236 240 187 187 220 236 170 174 96 106 -2.973096 34.474159

MGR08 SSE 139 143 144 152 164 172 361 361 190 202 240 244 187 189 220 228 174 175 92 98 -2.973241 34.474121

MGR09 SSE 145 147 144 152 164 172 361 361 186 190 240 240 187 191 220 236 174 175 98 106 -2.973324 34.474113

MGR10 SSE 138 143 139 144 164 172 361 361 190 202 240 244 187 193 216 236 169 174 98 106 -2.973158 34.474288

MGR11 SSE 138 145 144 152 164 172 280 280 190 190 244 244 187 191 212 236 173 173 98 98 -2.973082 34.474357

MGR12 SSE 144 144 144 146 164 164 280 361 186 194 244 244 195 195 216 224 170 180 96 98 -3.061411 34.661242

MGR13 SSE 145 149 139 152 164 172 280 280 190 202 236 244 187 195 228 236 170 174 104 104 -3.063445 34.673821

MGR14 SSE 145 149 144 144 164 170 280 280 190 194 244 244 187 195 212 216 170 174 96 98 -3.063343 34.673732

MGR15 SSE 141 147 144 144 164 172 276 280 210 214 236 244 187 189 208 216 170 170 94 98 -3.063255 34.673456

MGR16 SSE 143 149 144 152 164 170 280 353 190 194 236 248 195 195 212 232 170 174 96 98 -3.062896 34.673673

MGR17 SSE 139 143 144 144 164 172 280 280 194 206 236 240 193 195 212 224 170 170 96 98 -3.062917 34.673689

MGR18 SSE 145 149 152 152 164 170 280 280 194 206 240 248 187 189 216 228 170 174 98 98 -3.063952 34.676500

176

MGR19 SSE 138 145 145 152 164 172 280 346 190 194 220 240 187 195 216 232 174 174 98 98 -3.064476 34.683031

MGR20 SSE 145 149 144 152 164 172 284 353 190 194 240 240 189 195 228 228 174 174 96 98 -3.064512 34.683175

MGR21 SSE 138 149 144 145 164 172 284 365 190 198 227 240 184 187 220 220 170 170 96 98 -3.069516 34.654942

MGR22 SSE 145 147 144 144 164 172 280 280 198 206 236 240 199 205 224 224 174 174 96 96 -3.299085 34.765288

MGR23 SSE 145 147 139 144 164 172 280 280 198 206 240 248 187 199 220 224 174 174 96 98 -3.298855 34.766447

MGR24 SSE 147 147 139 144 164 172 284 284 194 198 240 240 184 195 208 208 174 174 96 98 -3.186174 34.674953

MGR25 SSE 147 149 144 149 164 172 280 280 186 202 240 244 189 193 216 216 174 174 96 96 -3.186290 34.656107

MGR26 SSE 143 147 144 152 164 172 280 280 190 202 240 240 189 195 216 228 170 174 96 96 -3.186449 34.656129

MGR27 SSE 147 149 144 152 164 172 280 337 186 190 236 248 189 189 220 224 174 177 92 98 -3.186504 34.656113

MGR28 SSE 147 149 144 152 164 172 280 337 186 190 236 248 189 189 220 224 174 177 92 98 -3.186561 34.656088

MGR29 SSE 145 147 139 152 164 172 276 280 194 194 236 244 187 189 216 220 170 174 96 106 -3.186811 34.656110

MGR30 SSE 144 144 144 144 164 172 276 280 190 214 240 248 187 187 216 236 170 173 96 98 -3.187378 34.655958

MGR31 SSE 138 147 144 144 164 172 280 353 190 194 240 240 193 195 216 224 174 174 96 98 -3.187372 34.655971

MGR32 SSE 147 151 144 144 164 172 280 361 186 214 244 248 187 193 220 236 174 174 96 96 -3.187071 34.656009

MGR33 SSE 149 151 144 152 164 164 280 388 198 214 236 240 184 187 208 224 174 175 96 96 -3.123767 34.458203

MGR34 SSE 145 147 144 144 164 164 284 372 186 198 236 244 187 187 224 232 166 174 96 96 -3.123799 34.458258

MGR35 SSE 145 147 144 152 164 172 284 376 186 190 227 244 189 195 216 220 170 174 96 96 -3.123751 34.458241

MGR36 SSE 139 145 144 144 164 164 280 361 194 214 236 248 187 189 224 224 174 175 96 96 -3.123719 34.458207

MGR37 SSE 145 147 144 144 164 172 276 280 198 206 240 244 189 189 212 224 174 175 98 98 -2.957127 34.580140

MGR38 SSE 145 147 144 152 164 164 280 282 194 198 236 248 187 189 224 248 174 175 96 98 -2.957290 34.580222

MGR39 SSE 147 151 144 144 164 172 280 284 194 202 240 244 189 189 220 220 174 174 96 98 -2.957625 34.580186

MGR40 SSE 147 149 144 152 164 172 276 280 202 210 240 248 189 189 204 216 170 170 96 98 -2.958026 34.580338

MGR41 SSE 138 145 144 152 164 172 280 280 210 214 240 248 187 189 208 220 174 179 94 96 -2.958034 34.580372

MGR42 SSE 139 151 144 144 164 172 280 365 190 194 240 244 189 195 224 224 170 174 96 98 -2.958126 34.580357

MGR43 SSE 147 149 144 144 164 172 276 280 198 210 244 248 189 189 204 248 170 170 98 98 -2.959046 34.581150

MGR44 SSE 145 149 141 144 164 172 353 365 202 210 240 240 187 187 208 220 175 179 98 98 -3.001047 34.610813

MGR45 SSE 145 149 141 144 164 172 280 353 198 210 236 240 187 195 216 220 170 179 98 106 -3.001112 34.610861

MGR46 SSE 147 149 144 144 164 164 280 353 194 202 240 248 189 195 220 232 170 174 98 98 -3.001116 34.610771

MGR47 SSE 147 149 144 144 164 172 276 284 190 194 244 248 189 195 216 220 173 175 98 104 -3.001116 34.610771

MWR01 SSE 145 147 144 152 164 170 280 280 198 217 240 240 184 195 216 216 152 152 96 98 -3.463001 34.834051

177

MWR02 SSE 145 147 144 152 164 172 280 280 190 190 236 236 189 195 212 224 150 150 98 98 -3.445169 34.863347

MWR03 SSE 138 147 144 152 164 164 280 280 194 194 236 240 184 187 224 228 150 150 96 98 -3.444583 34.859653

MWR04 SSE 138 149 144 152 164 172 384 384 198 202 227 236 199 205 216 220 149 149 98 106 -3.444590 34.859493

MWR05 SSE 149 151 139 144 164 172 280 365 190 221 236 240 189 189 220 224 152 152 96 104 -3.436300 34.803492

MWR06 SSE 149 149 139 144 164 172 280 365 198 221 236 240 189 193 216 224 149 149 96 96 -3.436802 34.802413

MWR07 SSE 139 145 144 152 164 164 284 284 194 198 236 244 187 187 228 240 149 149 92 96 -3.436967 34.802396

MWR08 SSE 138 145 144 164 164 172 280 280 194 198 220 244 187 189 0 0 149 149 98 106 -3.436990 34.802501

MWR09 SSE 138 145 139 144 164 164 280 280 194 214 227 244 187 187 216 220 149 149 96 96 -3.437001 34.802158

MWR10 SSE 149 149 144 144 164 164 361 361 190 194 227 236 187 189 220 232 149 149 96 104 -3.437701 34.801979

MWR11 SSE 147 149 144 152 164 172 280 280 194 194 220 236 187 195 216 216 149 149 96 96 -3.437909 34.802053

MWR12 SSE 139 147 139 144 164 172 280 300 190 194 240 240 187 187 216 220 149 149 96 96 -3.438067 34.802069

MWR13 SSE 147 147 144 148 164 172 276 280 194 202 227 240 195 195 220 232 149 149 94 96 -3.438214 34.802098

MWR14 SSE 143 149 139 144 164 172 276 280 190 198 244 248 187 193 228 228 150 150 94 98 -3.437805 34.801892

MWR15 SSE 139 145 144 144 164 164 353 353 190 194 240 240 187 195 208 216 149 149 96 98 -3.437726 34.801844

MWR16 SSE 138 145 144 152 164 164 353 365 194 198 244 248 187 187 220 224 149 154 98 98 -3.438179 34.801422

MWR17 SSE 147 149 144 152 164 164 280 284 194 194 240 244 187 187 208 224 150 150 96 101 -3.438230 34.800257

MWR18 SSE 149 151 139 152 164 172 280 284 190 198 232 240 187 187 220 240 150 150 96 96 -3.438664 34.799780

MWR19 SSE 145 151 152 152 164 174 280 284 198 202 227 244 189 195 232 236 150 150 96 96 -3.438494 34.799665

MWR20 SSE 147 149 139 152 164 172 276 280 198 202 236 244 189 195 216 236 150 150 96 98 -3.438542 34.799361

MWR21 SSE 147 149 144 152 164 172 280 280 194 194 240 248 187 191 216 216 149 149 96 96 -3.438531 34.799284

MWR22 SSE 147 147 148 152 164 172 280 280 194 198 240 240 187 189 216 224 149 149 96 96 -3.535430 34.836443

MWR23 SSE 138 147 148 152 164 172 280 280 194 198 0 0 187 189 216 224 149 149 96 96 -3.535460 34.836289

MWR24 SSE 144 144 144 144 164 172 280 280 186 198 236 248 187 205 216 232 152 152 96 96 -3.535424 34.836206

MWR25 SSE 145 151 144 144 164 172 280 280 186 194 240 244 187 195 216 224 149 149 96 98 -3.535619 34.836364

MWR26 SSE 145 151 144 152 164 172 280 353 194 198 236 240 189 189 208 220 174 174 96 101 -3.502574 34.846389

MWR27 SSE 147 151 144 145 164 172 280 361 182 214 244 248 187 189 216 224 174 175 96 98 -3.535458 34.836378

MWR28 SSE 138 147 144 144 164 172 280 280 194 202 236 248 189 205 224 236 170 174 98 106 -3.535478 34.836366

MWR29 SSE 139 147 144 144 164 172 280 280 194 198 227 236 187 189 216 236 170 170 98 106 -3.535453 34.836279

MWR30 SSE 144 144 144 144 164 172 280 284 190 210 236 244 0 0 224 228 176 182 96 98 -3.535910 34.836639

MWR31 SSE 139 145 152 152 164 172 280 361 198 210 240 248 189 193 216 224 170 170 96 96 -3.495485 34.840011

178

MWR32 SSE 145 147 144 152 164 172 280 280 202 214 240 244 187 189 216 224 170 174 98 98 -3.458509 34.796581

MWR34 SSE 149 151 152 152 164 172 280 376 190 202 236 236 187 187 224 228 170 170 96 98 -3.446425 34.797576

MWR35 SSE 139 147 148 152 164 172 280 280 194 198 236 248 187 189 216 224 170 174 96 96 -3.446426 34.797596

MWR37 SSE 141 145 139 144 164 172 280 361 194 202 236 240 187 189 224 228 170 174 96 98 -3.446343 34.797519

MWR38 SSE 143 147 144 144 164 172 361 365 194 214 236 240 187 189 216 220 170 174 98 106 -3.458487 34.796340

MWR39 SSE 145 147 139 144 164 172 280 361 190 194 240 248 187 193 216 220 170 174 96 98 -3.458384 34.796567

MWR40 SSE 145 149 144 144 164 172 280 369 190 194 236 240 187 187 224 228 170 170 96 98 -3.458426 34.796613

MWR41 SSE 138 149 144 152 164 164 353 376 186 190 240 248 193 195 232 240 170 170 96 96 -3.458653 34.796628

MWR42 SSE 145 149 139 152 164 172 0 0 190 190 232 232 187 193 220 240 170 174 96 98 -3.458924 34.796781

MWR43 SSE 145 147 139 144 164 172 280 361 202 214 240 244 187 187 224 228 172 174 96 96 -3.458997 34.796795

MWR44 SSE 138 145 152 152 164 164 280 346 190 214 240 240 187 187 224 228 170 172 96 96 -3.458982 34.796846

MWR45 SSE 145 147 141 144 164 164 276 280 190 214 240 244 187 195 224 228 174 178 96 98 -3.459279 34.796920

MWR46 SSE 147 147 148 152 164 172 280 280 194 198 236 248 187 189 216 224 170 174 96 96 -3.458452 34.797094

MWR47 SSE 144 144 144 144 164 164 276 280 194 198 240 240 187 187 212 220 174 175 96 106 -3.458466 34.797098

MWR48 SSE 145 149 144 152 164 172 280 357 194 210 220 244 191 191 220 224 174 174 96 98 -3.494380 34.840519

MWR49 SSE 145 147 144 144 164 172 280 353 194 217 236 240 187 187 224 232 170 181 96 96 -3.494260 34.840478

MWR50 SSE 145 147 144 144 164 172 277 277 194 202 0 0 187 195 208 216 170 174 98 106 -3.494357 34.839969

MWR51 SSE 145 147 144 144 164 172 280 350 194 202 240 248 187 195 208 216 170 174 98 106 -3.494458 34.839931

SNP185 SSE 145 151 144 144 164 172 280 376 182 186 240 244 187 195 216 220 170 173 96 96 -2.662601 34.800512

SNP186 SSE 145 147 144 144 164 172 280 342 194 194 220 240 187 195 220 224 170 173 98 98 -2.661472 34.801035

SNP187 SSE 145 151 144 152 164 172 284 361 182 190 236 244 187 195 220 224 175 177 98 106 -2.654848 34.779896

SNP188 SSE 139 145 139 152 164 172 284 365 190 214 236 240 195 199 224 224 173 177 92 106 -2.675769 34.770304

SNP189 SSE 147 149 144 144 164 172 280 365 194 210 236 236 187 193 216 220 170 173 94 96 -2.675769 34.770304

SNP190 SSE 145 149 144 144 164 164 280 365 190 194 236 236 187 195 0 0 174 174 98 98 -2.677426 34.769134

SNP191 SSE 147 149 144 152 164 172 280 361 190 217 220 240 187 195 220 224 169 173 106 106 -2.677485 34.769080

SNP192 SSE 147 147 144 152 164 172 284 361 190 214 220 240 187 195 212 220 170 170 96 96 -2.677438 34.769223

SNP193 SSE 149 151 152 152 164 172 276 280 190 194 220 240 189 199 212 220 0 0 96 96 -2.716409 34.750017

SNP194 SSE 145 147 144 144 164 164 280 365 190 194 248 248 187 205 216 216 172 173 92 96 -2.764870 34.699408

SNP195 SSE 147 149 144 152 164 164 280 280 186 190 248 252 187 189 216 220 170 173 92 98 -2.813865 34.664493

SNP196 SSE 139 147 139 152 164 172 284 365 190 194 244 248 184 189 216 228 172 175 96 106 -2.813876 34.664698

179

SNP197 SSE 143 147 144 144 164 164 280 365 190 214 236 240 189 189 220 220 173 173 98 104 -2.816941 34.680556

SNP198 SSE 145 147 144 144 164 172 284 284 194 198 220 240 191 195 216 220 170 177 92 98 -2.816923 34.680542

SNP199 SSE 144 144 144 144 164 164 280 284 190 194 240 240 195 195 216 220 174 174 98 104 -2.816923 34.680542

SNP200 SSE 145 151 144 144 164 164 280 365 190 214 236 240 187 187 228 228 170 173 96 96 -2.816923 34.680542

TGTH SSE 147 147 144 144 164 172 280 361 186 194 244 244 187 189 220 224 169 173 96 96 -3.062917 34.673689

NCA100 NCA 145 145 139 139 164 174 280 284 190 210 236 244 187 187 224 228 175 182 96 98 -3.437433 34.943033

NCA101 NCA 145 147 139 144 164 172 280 280 198 206 240 240 187 199 220 224 174 174 96 98 -3.437411 34.942721

NCA102 NCA 147 147 144 152 164 172 365 384 202 214 240 240 193 193 224 224 170 170 98 98 -3.355238 35.099547

NCA103 NCA 139 151 144 144 164 172 284 350 198 198 232 244 187 189 232 232 170 174 98 98 -3.356153 35.098324

NCA105 NCA 147 149 139 152 164 172 280 361 194 194 240 244 187 187 212 228 170 170 98 98 -3.356165 35.098357

NCA106 NCA 138 147 144 152 164 164 280 280 190 194 236 244 187 193 220 224 170 174 94 98 -3.356224 35.098403

NCA12 NCA 145 151 144 152 164 172 280 280 190 190 236 240 189 195 212 216 168 169 96 96 -3.356186 35.211640

NCA13 NCA 145 145 144 144 164 169 280 284 190 198 240 244 189 195 212 236 169 169 96 96 -3.356273 35.211518

NCA14 NCA 139 139 139 144 164 164 280 361 194 198 240 240 187 189 224 236 173 173 96 98 -3.356299 35.211533

NCA17 NCA 139 145 139 144 164 170 280 353 198 206 244 248 184 195 224 228 170 170 98 104 -3.227670 35.489844

NCA18 NCA 145 149 144 144 164 170 280 353 186 198 240 244 184 195 220 224 170 170 98 104 -3.227656 35.489842

NCA23 NCA 138 144 144 144 164 171 275 280 194 206 236 248 187 195 132 132 169 175 96 96 -3.229694 35.516557

NCA24 NCA 138 144 144 152 164 171 276 280 194 206 236 240 187 195 132 132 169 175 96 98 -3.229694 35.516557

NCA25 NCA 138 147 144 150 164 171 280 280 194 206 236 236 195 195 132 132 173 173 96 96 -3.229693 35.516560

NCA26 NCA 138 145 144 144 164 171 276 280 198 206 236 252 187 195 132 140 173 173 96 96 -3.229673 35.516708

NCA30 NCA 138 147 144 144 164 171 276 280 190 206 236 240 195 195 132 140 166 172 96 96 -3.247179 35.506312

NCA31 NCA 138 144 144 144 164 171 276 280 194 206 236 236 187 187 132 140 175 175 96 96 -3.247179 35.506312

NCA32 NCA 138 147 144 144 164 171 276 280 194 206 236 240 187 189 132 140 169 169 96 98 -3.246207 35.507920

NCA34 NCA 139 145 139 144 164 164 280 361 194 198 240 240 189 189 224 236 173 173 96 98 -3.251191 35.224705

NCA35 NCA 139 145 144 152 164 172 284 365 194 202 236 244 187 195 216 216 170 175 96 98 -3.251090 35.224688

NCA36 NCA 139 145 139 144 164 164 280 361 194 198 240 240 189 189 224 236 172 173 96 98 -3.251108 35.224552

NCA37 NCA 145 145 144 144 164 164 284 284 194 198 244 244 189 195 216 236 170 170 96 98 -3.250962 35.224712

NCA38 NCA 139 145 139 144 164 164 280 361 194 198 240 240 189 189 224 236 173 173 96 98 -3.250549 35.224599

NCA39 NCA 139 145 144 144 164 164 284 361 190 194 240 244 189 197 212 220 170 174 96 96 -3.250254 35.224853

NCA40 NCA 139 139 144 152 164 169 280 280 198 202 244 252 195 205 228 236 169 169 96 98 -3.250226 35.224834

180

NCA41 NCA 147 147 144 152 164 172 280 357 198 210 244 244 189 195 220 224 170 170 94 98 -3.250530 35.225107

NCA42 NCA 145 145 144 152 164 172 284 365 194 202 236 244 187 195 216 216 170 175 96 98 -3.250862 35.225007

NCA43 NCA 139 147 144 152 164 172 280 376 190 202 240 240 187 187 212 224 170 174 98 104 -3.222693 35.512626

NCA44 NCA 147 149 152 152 164 172 280 280 190 198 236 240 187 187 220 220 170 175 96 96 -3.222622 35.512535

NCA45 NCA 139 145 139 144 164 172 280 346 194 198 244 248 184 195 212 216 174 179 96 98 -3.222621 35.512602

NCA46 NCA 146 150 144 152 164 164 288 361 194 198 240 252 187 195 212 244 172 173 98 104 -3.222605 35.512590

NCA47 NCA 147 151 144 144 164 172 288 337 198 206 244 248 187 187 208 216 175 175 96 98 -3.222605 35.512590

NCA48 NCA 138 145 144 152 164 172 280 357 194 206 236 248 189 189 212 220 170 170 96 106 -3.341600 35.522771

NCA49 NCA 145 145 139 144 164 172 280 280 194 202 232 256 187 187 220 224 169 169 96 98 -3.271507 35.588325

NCA50 NCA 145 145 139 144 164 172 279 280 194 206 0 0 187 195 220 224 170 174 96 106 -3.271591 35.588258

NCA51 NCA 145 151 139 144 164 169 280 280 194 206 232 256 187 195 220 224 170 174 96 106 -3.271653 35.587915

NCA52 NCA 139 147 144 150 164 172 276 337 0 0 220 244 193 205 208 220 174 175 96 98 -3.271802 35.587801

NCA53 NCA 139 149 144 144 164 167 276 280 198 206 244 248 187 189 220 224 174 175 98 98 -3.272083 35.587810

NCA54 NCA 145 145 139 144 164 174 280 357 194 206 240 244 187 205 224 224 170 178 96 101 -3.272339 35.587717

NCA55 NCA 145 147 139 144 164 169 280 280 190 198 240 248 187 187 216 216 174 178 96 98 -3.272427 35.587492

NCA56 NCA 145 147 139 144 164 172 280 284 190 194 240 244 187 195 216 228 170 174 96 98 -3.272345 35.587444

NCA57 NCA 145 147 139 144 164 169 280 280 190 198 240 248 184 187 216 216 174 178 96 98 -3.272167 35.586975

NCA58 NCA 145 145 139 144 164 169 280 353 186 210 240 244 187 187 208 220 168 170 96 106 -3.246728 35.480876

NCA59 NCA 145 145 139 152 164 169 280 353 186 210 240 244 187 187 208 220 168 170 96 106 -3.247235 35.479462

NCA60 NCA 145 145 144 152 164 172 276 361 190 202 236 240 189 189 224 224 170 174 96 98 -3.203775 35.510309

NCA61 NCA 139 139 144 152 164 172 280 280 194 206 236 240 187 188 224 228 170 170 96 96 -3.211200 35.507077

NCA62 NCA 147 149 144 152 164 172 337 353 186 194 232 240 189 189 216 228 174 174 98 101 -3.212284 35.601463

NCA63 NCA 138 147 144 152 164 172 276 353 202 214 236 248 184 205 220 232 174 174 94 101 -3.212286 35.601457

NCA64 NCA 145 145 139 139 164 164 280 280 0 0 240 240 184 187 212 216 170 170 96 98 -3.212286 35.601457

NCA65 NCA 138 145 144 144 164 172 276 279 190 198 232 240 187 195 220 220 170 172 96 98 -3.295719 35.556931

NCA66 NCA 138 145 144 144 164 198 280 361 194 194 236 248 187 195 220 220 170 170 96 106 -3.295719 35.556931

NCA67 NCA 145 145 144 144 164 198 280 280 186 190 240 240 184 198 220 220 170 170 98 106 -3.295719 35.556931

NCA68 NCA 138 138 139 144 164 198 279 279 195 201 240 240 187 187 220 220 168 170 96 106 -3.295719 35.556931

NCA69 NCA 139 145 139 152 164 172 280 353 194 206 240 248 187 195 220 220 175 179 96 98 -3.361391 35.499762

NCA70 NCA 139 145 144 144 164 164 280 280 194 194 227 240 187 187 220 220 170 170 96 96 -3.361391 35.499762

181

NCA71 NCA 138 147 139 152 164 198 276 365 194 206 244 248 187 198 220 220 170 174 96 98 -3.361391 35.499762

NCA72 NCA 138 147 144 144 164 172 280 353 194 206 236 236 184 189 220 220 170 174 96 96 -3.361267 35.498763

NCA73 NCA 138 147 139 144 164 172 280 280 194 206 236 252 187 187 220 220 174 175 98 98 -3.361316 35.498819

NCA74 NCA 138 147 139 144 164 198 280 357 186 194 220 236 184 198 220 220 174 174 98 98 -3.361174 35.498768

NCA75 NCA 139 147 144 144 164 164 280 284 186 194 236 252 187 187 220 220 174 174 98 106 -3.370750 35.546552

NCA76 NCA 139 145 144 144 164 172 280 284 186 206 236 252 187 195 220 220 170 174 98 106 -3.371706 35.545094

NCA77 NCA 139 147 144 150 164 172 280 361 186 210 236 244 195 197 220 220 170 174 96 98 -3.377261 35.483973

NCA78 NCA 147 149 144 144 164 172 280 337 186 206 236 244 187 193 208 232 174 175 98 98 -3.219298 35.585274

NCA79 NCA 147 147 139 144 164 198 280 353 198 198 240 240 189 195 216 220 174 174 96 98 -3.219507 35.585019

NCA80 NCA 145 147 150 150 164 172 280 284 190 194 227 236 187 187 216 220 170 175 96 106 -3.180178 35.499713

NCA81 NCA 145 151 144 144 164 172 280 280 202 202 240 256 187 205 212 220 170 174 96 98 -3.220210 35.513626

NCA82 NCA 147 151 144 152 164 164 284 284 198 202 236 240 191 195 220 232 174 174 98 98 -3.220210 35.513626

NCA84 NCA 147 149 139 152 164 172 361 376 198 202 236 240 187 195 224 232 170 174 96 98 -3.206707 35.595196

NCA85 NCA 147 151 144 152 164 172 280 365 210 214 236 240 189 189 232 236 170 172 96 98 -3.384267 35.014841

NCA86 NCA 145 145 144 152 164 172 280 376 202 206 240 248 187 189 220 236 170 174 98 104 -3.384256 35.014892

NCA87 NCA 145 147 139 152 164 172 280 280 190 194 232 244 187 189 212 232 170 174 96 98 -3.384189 35.014877

NCA88 NCA 145 147 144 144 164 164 284 361 186 194 236 240 195 199 216 220 170 174 96 106 -3.366043 35.013731

NCA91 NCA 147 147 144 144 164 164 284 353 194 194 236 240 189 195 216 228 170 174 96 106 -3.363936 35.012957

NCA92 NCA 147 149 141 144 164 172 284 361 194 194 244 244 189 189 208 224 174 175 98 98 -3.354448 35.021493

NCA95 NCA 145 149 139 152 164 172 280 361 194 194 236 240 195 195 224 228 170 170 94 96 -3.347256 35.039520

NCA96 NCA 145 151 144 144 164 172 280 280 194 206 244 248 187 195 220 224 174 178 98 98 -3.347256 35.039520

NCA97 NCA 145 147 144 152 164 172 280 280 206 214 236 244 187 195 220 228 170 174 98 98 -3.346723 35.039916

NCA98 NCA 145 145 144 152 164 172 280 353 198 206 240 240 189 195 220 220 170 170 96 96 -3.471711 34.995585

NCA99 NCA 147 147 144 152 164 172 361 365 194 194 236 236 187 195 212 220 170 170 96 98 -3.471699 34.995698

NCAK NCA 138 147 144 144 164 164 280 361 190 202 236 244 187 189 220 220 170 170 96 98 -3.206707 35.595196

NDT08 NCA 149 149 144 152 164 172 280 365 198 206 244 248 193 195 216 216 170 174 96 96 -2.965989 34.917964

NDT09 NCA 138 147 144 144 164 172 337 337 194 198 240 240 187 195 220 224 170 174 96 98 -2.965599 34.917691

NDT10 NCA 138 145 144 152 164 172 280 365 190 214 236 240 189 189 212 224 170 174 96 96 -2.965586 34.917676

NDT11 NCA 145 149 144 144 164 164 280 365 198 214 240 244 189 195 212 216 174 181 96 96 -2.965499 34.917708

NDT12 NCA 139 147 144 144 164 172 280 361 194 198 240 252 187 195 220 224 170 170 96 106 -2.965400 34.917665

182

NDT13 NCA 139 147 144 144 164 174 280 280 198 206 240 248 187 195 220 224 170 170 96 106 -2.965400 34.917665

NDT14 NCA 149 149 144 152 164 172 280 361 186 194 240 240 187 189 216 224 170 180 96 98 -2.965400 34.917665

NDT15 NCA 149 149 144 152 164 172 361 361 194 198 240 252 176 195 220 224 170 175 96 96 -2.965400 34.917665

NDT16 NCA 147 147 144 144 164 172 337 361 194 198 240 240 187 195 220 224 170 174 96 98 -2.965539 34.917875

NDT17 NCA 139 145 144 144 164 172 280 365 190 214 236 240 189 189 212 224 170 174 96 96 -2.965539 34.917875

NDT18 NCA 147 147 144 144 164 172 280 280 198 206 240 248 187 193 224 228 170 170 96 96 -2.997515 34.928223

NDT19 NCA 139 139 144 144 164 172 280 280 198 206 236 236 189 189 216 220 170 170 94 96 -3.017435 34.984980

NDT20 NCA 145 147 144 144 164 174 280 280 194 198 244 248 190 190 216 216 174 179 96 98 -3.017427 34.984963

NDT21 NCA 145 147 144 152 164 174 280 280 194 198 244 248 190 190 216 216 174 179 96 98 -3.017414 34.984969

NDT22 NCA 145 147 144 152 164 170 280 280 194 198 244 248 190 190 216 216 0 0 96 98 -3.020649 34.985391

NDT23 NCA 139 145 144 152 164 172 280 280 190 194 227 240 187 195 216 224 175 179 96 98 -3.021880 34.985541

NDT24 NCA 145 147 144 152 164 164 280 280 190 194 227 240 187 195 216 224 175 179 96 98 -3.027250 34.985209

NDT25 NCA 145 147 141 144 164 164 280 280 194 198 244 248 190 190 216 216 174 179 96 98 -3.028170 34.985080

NDT27 NCA 139 139 144 144 164 172 280 280 198 206 236 236 189 189 216 220 170 170 94 96 -3.022838 34.987159

NDT28 NCA 139 145 139 144 164 172 280 280 190 194 232 236 187 187 220 220 170 174 96 98 -3.017462 34.984994

NDT29 NCA 139 147 144 148 164 172 280 361 194 198 244 252 187 195 224 224 169 169 96 96 -2.965516 34.917022

NDT30 NCA 139 147 144 152 164 172 280 280 194 206 236 244 195 195 216 228 170 179 96 96 -2.965517 34.917021

NDT31 NCA 145 147 144 144 164 172 280 280 190 190 240 244 187 195 212 232 170 170 96 96 -2.965516 34.917022

NDT32 NCA 139 145 139 144 164 192 280 280 190 202 244 244 180 187 220 224 170 170 98 98 -2.965517 34.917021

LMNP44 MAR 139 145 139 144 164 164 388 388 194 198 240 248 187 201 220 232 170 174 96 98 -3.507938 35.783286

LMNP45 MAR 139 143 141 141 164 172 280 388 194 194 244 248 187 187 140 212 174 180 96 96 -3.507894 35.783210

LMNP46 MAR 139 145 150 152 164 172 280 388 190 198 244 248 187 189 220 224 170 176 96 96 -3.507894 35.783210

LMNP47 MAR 139 151 139 144 164 172 280 284 194 194 240 240 189 195 216 224 172 172 98 98 -3.507894 35.783210

LMNP48 MAR 139 145 144 144 164 164 280 284 194 198 240 240 187 195 224 224 174 174 98 98 -3.508060 35.783537

LMNP49 MAR 139 145 139 150 164 170 280 284 194 194 244 244 189 189 216 220 170 174 96 96 -3.508042 35.783541

LMNP50 MAR 139 145 148 152 164 172 280 353 194 194 244 248 187 201 220 224 170 170 96 106 -3.508029 35.783422

LMNP51 MAR 139 145 150 150 164 164 365 365 194 198 244 248 187 189 216 220 170 174 96 98 -3.508053 35.783341

LMNP52 MAR 139 145 139 152 164 164 280 280 190 198 232 248 187 189 216 224 170 174 96 104 -3.501753 35.785076

LMNP53 MAR 139 145 139 144 164 170 280 280 194 194 244 244 187 189 216 220 175 175 96 96 -3.501324 35.784926

LMNP54 MAR 139 145 139 152 164 164 276 280 194 198 244 244 189 189 0 0 174 174 96 96 -3.501002 35.784688

183

LMNP55 MAR 139 145 139 144 164 192 280 388 190 198 244 244 187 191 140 216 175 175 98 104 -3.500979 35.784673

LMNP56 MAR 138 147 139 144 164 172 361 365 194 198 236 252 195 195 212 224 174 174 96 98 -3.418295 35.835321

LMNP57 MAR 145 147 139 144 164 164 280 284 190 198 236 240 187 189 212 224 170 174 96 98 -3.418295 35.835321

LMNP59 MAR 147 151 139 144 164 164 280 284 190 194 240 240 187 187 216 216 172 172 98 98 -3.504601 35.785696

LMNP60 MAR 145 147 139 144 164 164 280 284 190 194 240 244 187 187 208 220 175 175 96 98 -3.504229 35.785807

LMNP61 MAR 147 151 144 144 164 172 280 284 194 202 240 240 187 187 208 208 174 174 98 98 -3.504264 35.785848

LMNP62 MAR 147 151 139 144 164 164 280 353 194 202 220 244 187 189 220 224 174 174 96 106 -3.502112 35.785273

LMNP64 MAR 145 147 139 144 164 172 280 337 190 194 240 244 187 187 216 220 170 170 96 98 -3.501758 35.785133

LMNP65 MAR 145 145 150 152 164 174 280 280 186 194 220 236 185 195 216 220 170 170 96 96 -3.501372 35.784954

LMNP66 MAR 138 149 139 144 164 174 284 380 190 198 236 252 189 195 204 208 170 174 96 103 -3.501372 35.784954

LMNP67 MAR 151 151 139 139 164 172 280 280 190 198 244 244 185 195 204 220 170 174 96 96 -3.416141 35.819308

LMNP68 MAR 139 147 139 144 164 172 0 0 198 198 240 240 189 195 208 212 170 170 96 96 -3.430493 35.806214

LMNP69 MAR 145 145 139 144 164 172 280 353 186 202 236 244 185 187 212 220 170 175 96 96 -3.430493 35.806214

LMNP70 MAR 138 151 144 144 164 172 284 380 194 198 240 240 189 195 216 224 172 175 98 98 -3.487520 35.783180

LMNP71 MAR 147 151 139 139 164 172 280 369 194 194 236 244 187 187 216 220 174 179 96 98 -3.487968 35.783268

LMNP72 MAR 145 147 144 144 164 174 280 346 186 194 220 236 187 195 216 224 174 175 96 96 -3.487841 35.783240

LMNP73 MAR 145 147 139 139 164 164 280 280 190 198 232 248 187 189 216 224 170 174 96 104 -3.487841 35.783240

LMNP74 MAR 145 147 144 144 164 172 353 361 190 206 240 240 187 187 216 220 170 174 96 104 -3.493452 35.782735

LMNP76 MAR 147 151 139 144 164 172 280 280 194 194 240 244 187 189 220 224 170 174 96 106 -3.501096 35.783933

LMNP77 MAR 138 145 139 139 164 172 284 284 186 194 236 244 189 189 220 220 170 170 96 96 -3.442947 35.803584

LMNP78 MAR 147 149 144 152 164 164 337 353 202 206 240 240 189 205 212 224 170 174 96 104 -3.484081 35.784675

LMNP79 MAR 139 147 144 144 164 172 284 353 190 206 240 240 187 187 220 220 174 175 98 104 -3.484238 35.784589

LMNP80 MAR 147 147 139 144 164 164 284 284 202 206 232 236 187 195 220 224 174 174 92 104 -3.484371 35.784586

LMNP81 MAR 147 151 139 144 164 164 280 353 0 0 220 244 187 189 220 224 174 174 96 106 -3.487188 35.783149

LMNP83 MAR 147 151 139 144 164 164 276 280 194 202 220 244 187 189 220 224 174 174 96 106 -3.503041 35.782880

LFR01 MAR 145 147 139 144 164 164 280 280 204 204 240 244 187 187 208 220 170 174 96 104 -3.506651 36.093355

LFR02 MAR 139 145 144 144 164 164 280 280 190 198 236 244 195 195 212 216 170 174 96 96 -3.505756 36.095419

LFR05 MAR 147 151 144 144 164 164 276 280 186 194 236 240 195 195 131 216 170 174 98 104 -3.506916 36.090964

LFR06 MAR 151 151 139 144 164 164 276 280 194 198 0 0 189 197 131 212 170 174 96 98 -3.503832 36.071692

LFR07 MAR 147 147 144 144 164 164 276 280 186 194 232 236 187 187 0 0 170 174 98 104 -3.502455 36.071347

184

LFR09 MAR 139 145 148 148 164 164 280 280 206 225 226 226 187 193 220 224 170 174 96 104 -3.463317 36.134883

MANR01 MAR 139 145 144 144 164 164 337 361 190 198 236 244 195 195 212 216 170 174 96 96 -3.563043 36.007288

MANR02 MAR 145 145 139 139 164 172 280 353 198 198 244 244 187 189 220 224 170 170 98 101 -3.562924 36.008027

MANR03 MAR 139 147 144 152 164 172 280 280 194 198 227 240 195 197 212 220 174 176 96 104 -3.563013 36.007726

MANR04 MAR 139 147 144 152 164 170 288 361 198 210 236 240 187 195 208 232 170 174 92 101 -3.563032 36.007695

MANR05 MAR 139 147 144 150 164 192 353 353 190 194 240 240 187 187 212 220 174 178 98 98 -3.562744 36.007495

MANR06 MAR 139 151 144 150 164 174 353 353 190 194 244 248 189 193 220 220 174 178 98 104 -3.561606 36.008929

MANR07 MAR 147 151 144 144 164 172 276 280 186 194 236 236 195 195 131 216 170 174 98 104 -3.561606 36.008929

MANR08 MAR 139 145 139 139 164 192 284 337 194 194 236 244 189 190 208 232 170 178 92 104 -3.561885 36.009741

MANR09 MAR 139 145 139 139 164 192 284 337 194 194 236 244 189 191 208 232 170 178 92 104 -3.561742 36.011793

MANR10 MAR 139 147 150 152 164 192 353 353 194 194 240 240 184 187 212 220 174 180 98 98 -3.561709 36.012430

MANR11 MAR 139 145 145 152 164 170 280 280 194 194 236 240 195 195 212 216 180 180 96 106 -3.554048 36.013726

MANR12 MAR 139 139 145 152 164 172 284 284 190 206 240 256 184 189 212 224 170 170 96 106 -3.554052 36.013730

MANR13 MAR 138 144 139 139 164 172 0 0 194 198 238 238 187 197 212 224 170 170 96 106 -3.554527 36.013975

MANR14 MAR 139 158 139 152 164 172 276 280 194 198 232 244 189 205 212 212 174 174 106 106 -3.567043 36.009277

MANR15 MAR 139 145 139 152 164 164 276 280 194 198 236 248 187 191 140 216 170 180 96 96 -3.566570 36.010844

MANR16 MAR 139 151 150 152 164 192 276 280 194 206 236 244 187 195 212 220 170 181 104 106 -3.570732 36.023459

MANR17 MAR 145 151 139 144 164 164 284 361 194 206 244 256 190 195 212 220 170 174 98 101 -3.570539 36.023484

MANR18 MAR 139 147 144 144 164 172 276 280 194 206 236 244 189 195 212 220 180 180 98 104 -3.569785 36.022578

MANR19 MAR 139 145 141 141 164 192 280 284 194 206 244 248 176 191 131 216 180 180 96 106 -3.569785 36.022578

MANR20 MAR 139 147 139 139 164 172 280 284 194 206 244 248 182 187 220 232 180 180 96 96 -3.569785 36.022578

MANR21 MAR 139 145 144 144 164 172 284 361 194 214 236 240 187 189 220 232 174 180 96 96 -3.569871 36.021414

MANR22 MAR 145 151 144 144 164 172 276 280 198 206 236 244 187 201 220 232 180 180 96 96 -3.567748 36.015519

MANR23 MAR 145 151 139 144 164 172 276 280 198 202 244 244 187 193 212 216 172 175 96 106 -3.568173 36.014890

MANR24 MAR 145 151 144 144 164 172 276 280 194 194 240 240 195 195 208 232 180 180 98 104 -3.568176 36.014878

MANR25 MAR 139 139 139 144 164 164 276 369 198 206 241 241 191 195 216 220 173 177 101 104 -3.567320 36.008427

MANR27 MAR 139 145 152 152 164 172 280 361 198 206 241 241 187 189 216 216 174 180 96 96 -3.599874 35.983995

MANR28 MAR 139 139 144 150 164 172 353 361 198 206 241 241 187 195 208 208 170 174 104 104 -3.566353 36.005715

MANR29 MAR 139 151 144 152 164 192 284 361 198 198 244 244 187 189 212 216 170 180 98 106 -3.564782 36.006000

MANR30 MAR 147 151 139 144 164 192 280 353 194 202 220 244 187 189 220 224 174 174 96 106 -3.558215 36.010081

185

MANR31 MAR 139 139 144 144 164 170 276 280 198 202 240 248 195 195 216 220 170 174 96 96 -3.520035 36.044148

MANR33 MAR 147 151 144 144 164 172 280 337 198 202 236 240 187 195 212 220 170 170 96 96 -3.571498 36.005862

MANR34 MAR 145 151 144 144 164 172 346 353 194 194 240 240 195 195 208 232 172 175 98 104 -3.567857 36.004111

MANR35 MAR 145 151 144 152 164 164 353 353 194 210 244 244 195 197 228 228 174 174 96 106 -3.567237 36.005663

MANR36 MAR 147 151 144 144 164 172 280 337 198 217 236 240 187 195 212 220 170 170 96 96 -3.560617 36.003799

MANR37 MAR 145 151 144 144 164 172 337 361 202 202 244 248 187 195 216 216 170 175 96 96 -3.557825 36.023220

MANR38 MAR 144 151 144 144 164 172 353 353 194 202 236 240 184 195 212 212 174 174 98 98 -3.558538 36.023029

MANR39 MAR 139 147 139 152 164 172 280 357 194 194 220 236 195 197 216 228 170 174 96 106 -3.561505 36.019344

MANR40 MAR 138 144 144 144 164 164 280 361 194 198 236 240 189 191 208 236 170 174 96 98 -3.561505 36.019344

MANR41 MAR 139 145 144 144 164 172 280 361 194 198 236 240 189 191 208 208 170 174 96 98 -3.561465 36.019553

MANR42 MAR 147 151 144 144 164 172 280 280 194 194 236 240 189 191 212 232 174 178 96 106 -3.561465 36.019553

MANR43 MAR 145 147 141 144 164 164 276 280 194 194 236 236 189 195 208 236 175 175 96 101 -3.564950 36.014145

MANR44 MAR 138 149 144 152 164 172 280 361 178 198 236 236 189 195 220 228 174 174 101 106 -3.564829 36.014584

MANR45 MAR 149 151 144 152 164 164 276 361 194 206 236 240 195 195 208 220 172 179 92 96 -3.562549 36.004192

MANR46 MAR 149 151 144 152 164 172 280 337 194 194 236 244 189 195 228 232 170 174 104 106 -3.562197 36.002884

MANR47 MAR 147 151 144 145 164 172 337 361 198 206 229 240 187 195 208 228 170 174 92 96 -3.562197 36.002884

MANR49 MAR 138 147 144 152 164 164 284 337 198 198 227 236 187 195 212 216 170 174 96 98 -3.562197 36.002884

MANR50 MAR 139 147 139 144 164 170 284 353 198 202 240 240 189 195 224 232 174 174 96 96 -3.562197 36.002884

MANR51 MAR 147 147 144 144 164 170 284 353 198 202 240 240 189 195 224 232 174 174 96 96 -3.562197 36.002884

MANR52 MAR 147 147 144 144 164 164 276 276 198 202 236 236 183 187 0 0 170 170 96 106 -3.562197 36.002884

MANR53 MAR 144 147 144 152 164 164 280 337 198 202 240 240 187 195 232 232 174 177 96 104 -3.562197 36.002884

MANR54 MAR 145 147 144 144 164 172 280 280 194 194 236 236 187 191 212 224 174 174 98 106 -3.562197 36.002884

MANR55 MAR 147 147 152 152 164 164 284 361 194 202 240 240 191 195 208 216 170 174 92 104 -3.562197 36.002884

MANR56 MAR 145 149 144 152 164 172 280 361 194 214 232 244 195 195 224 228 170 174 96 98 -3.562197 36.002884

MANR57 MAR 147 149 144 152 164 172 284 337 190 206 236 244 187 189 212 224 170 175 98 104 -3.562197 36.002884

MANR58 MAR 149 151 144 152 164 172 280 361 178 198 236 236 189 195 220 228 174 174 101 106 -3.562197 36.002884

MANR59 MAR 138 145 139 139 164 172 280 361 0 0 236 244 189 191 216 236 174 174 96 101 -3.562197 36.002884

MANR60 MAR 145 147 139 144 164 172 280 337 202 206 236 244 193 195 216 220 174 175 96 96 -3.562197 36.002884

TNP11 TNP 145 151 144 144 164 164 280 284 194 202 240 244 187 187 212 212 174 175 96 104 -3.843506 36.053029

TNP24 TNP 145 151 144 144 164 172 276 280 198 198 240 240 187 201 208 236 170 172 92 96 -3.797874 36.068093

186

TNP25 TNP 139 149 139 152 172 198 276 353 198 198 240 256 191 198 212 228 170 174 0 0 -3.798055 36.068749

TNP26 TNP 145 145 139 144 164 184 280 346 202 214 236 248 193 195 212 220 170 174 96 106 -3.901247 36.081821

TNP27 TNP 145 145 144 144 164 198 280 280 190 194 236 248 175 198 208 236 170 174 96 104 -3.900666 36.078607

TNP28 TNP 145 145 144 144 172 198 280 280 190 194 240 244 189 197 212 228 170 174 0 0 -3.799000 36.064383

TNP29 TNP 145 151 139 144 164 172 280 346 206 214 220 240 189 195 212 228 170 170 96 106 -3.759978 36.026812

TNP30 TNP 145 151 139 144 164 172 280 346 206 214 220 240 189 195 212 228 170 170 96 106 -3.756797 36.028863

TNP31 TNP 145 147 144 152 164 164 353 361 194 194 220 236 184 189 212 212 170 172 96 98 -3.756579 36.028916

TNP32 TNP 138 145 144 152 164 172 280 353 198 206 236 240 187 189 220 220 170 178 96 98 -3.793028 36.054175

TNP33 TNP 145 145 152 152 172 172 280 353 190 202 252 256 187 197 216 220 170 175 96 98 -3.731188 35.995822

TNP34 TNP 145 145 152 152 164 184 280 353 194 198 236 248 176 189 140 216 170 174 98 106 -3.885208 35.937460

TNP35 TNP 145 145 144 152 164 198 280 361 190 210 240 244 189 198 224 236 170 174 98 106 -4.087502 35.979440

TNP36 TNP 145 147 144 144 172 198 280 280 194 194 236 248 189 198 220 220 170 174 98 106 -4.089211 35.979950

TNP37 TNP 145 151 139 139 164 198 346 353 194 194 240 244 187 198 212 224 170 175 96 98 -4.089322 35.979987

TNP38 TNP 145 145 139 144 172 198 280 280 186 214 236 252 198 198 131 140 170 174 83 96 -4.089389 35.980046

TNP39 TNP 145 147 139 152 164 198 280 280 194 198 236 240 198 198 212 224 170 172 96 98 -4.089419 35.980075

TNP40 TNP 139 145 144 144 198 198 280 280 0 0 236 248 198 198 212 224 170 172 83 96 -4.089212 35.980011

TNP41 TNP 139 145 150 152 164 170 280 365 186 194 236 244 187 195 224 228 170 170 96 98 -4.036481 36.003043

TNP42 TNP 147 149 144 144 164 170 280 280 186 194 240 244 195 195 224 224 170 172 98 106 -4.086434 36.114880

TNP43 TNP 145 147 144 150 164 170 280 346 194 194 240 248 195 195 220 224 170 174 92 98 -4.086561 36.115126

TNP44 TNP 145 149 139 144 164 164 280 361 186 206 236 240 195 195 212 224 170 170 96 98 -4.086511 36.115089

TNP45 TNP 145 149 144 152 164 164 280 280 194 210 236 240 191 191 220 220 170 174 98 104 -4.086076 36.114954

TNP46 TNP 145 149 144 152 172 172 280 353 198 202 236 248 176 189 224 224 170 172 98 98 -4.111867 35.944489

TNP47 TNP 147 151 139 144 164 164 280 280 202 206 240 244 187 201 131 220 170 174 96 98 -4.104573 35.967682

TNP48 TNP 145 145 139 139 164 198 280 280 198 198 244 248 198 198 212 212 170 174 96 104 -4.104611 35.967420

TNP49 TNP 139 145 139 152 164 198 280 280 214 214 236 248 189 198 131 236 170 174 96 96 -4.103161 35.969526

TNP50 TNP 139 145 152 152 198 198 280 353 178 178 227 248 198 198 140 220 170 174 96 98 -4.102677 35.970452

TNP51 TNP 139 145 168 168 164 198 280 280 186 210 236 240 198 198 208 224 170 174 98 98 -4.102613 35.970789

TNP52 TNP 145 145 139 144 164 198 280 280 206 206 227 227 198 198 131 131 170 174 96 96 -4.102692 35.970894

TNP53 TNP 139 147 139 152 163 163 280 280 186 186 236 252 189 193 224 228 170 170 96 98 -4.102663 35.970909

TNP54 TNP 139 147 144 144 164 198 280 353 198 202 236 248 187 198 220 224 170 174 96 101 -4.102936 35.970922

187

TNP55 TNP 145 151 144 152 164 198 276 280 194 202 240 244 198 198 224 224 170 174 101 106 -4.102972 35.970963

TNP56 TNP 139 147 139 144 164 198 280 280 198 214 227 236 198 198 212 220 170 174 96 96 -4.102965 35.970871

TNP57 TNP 147 147 152 152 164 164 271 276 186 186 236 248 198 198 0 0 170 172 96 98 -4.103023 35.970913

TNP58 TNP 139 147 139 152 164 198 280 280 186 186 236 252 189 198 224 224 170 170 96 98 -4.103084 35.971014

TNP59 TNP 139 145 152 152 164 172 280 353 190 210 244 256 189 198 228 236 170 174 96 106 -4.103352 35.971064

TNP60 TNP 145 145 144 152 164 198 280 361 206 206 236 244 198 198 220 220 170 174 98 104 -4.103261 35.971027

TNP61 TNP 139 139 144 152 164 164 276 280 194 194 244 248 176 191 216 224 170 174 98 104 -4.010891 36.092307

TNP62 TNP 139 149 139 152 164 164 280 280 186 206 240 248 176 191 216 224 170 174 98 98 -4.010891 36.092307

TNP63 TNP 147 151 144 152 164 164 284 284 198 202 236 240 187 195 212 220 170 175 98 106 -4.011005 36.092035

TNP64 TNP 147 147 144 152 172 172 284 357 194 202 236 240 195 197 220 220 170 175 104 106 -4.011005 36.092035

TNP66 TNP 147 151 144 152 164 170 284 284 198 202 236 240 187 195 212 220 170 175 98 106 -4.011146 36.092017

TNP67 TNP 138 149 139 152 164 172 284 365 194 206 240 240 187 191 220 220 174 179 98 106 -4.011146 36.092017

TNP68 TNP 139 145 144 152 164 164 280 350 206 210 244 248 189 198 220 224 168 170 98 98 -4.011141 36.092014

TNP69 TNP 139 147 144 144 164 198 353 361 198 202 236 248 187 198 220 224 170 174 96 101 -4.081344 35.964621

TNP70 TNP 139 145 144 152 164 198 276 280 194 206 232 236 198 198 212 212 170 174 98 104 -4.080658 35.964908

TNP71 TNP 145 145 144 152 164 198 280 353 194 194 227 236 198 198 131 212 170 170 98 104 -4.080702 35.964871

TNP72 TNP 147 151 144 152 164 198 280 337 190 194 240 248 187 198 224 224 170 170 98 104 -4.078755 35.967088

TNP73 TNP 145 145 144 144 164 192 280 280 198 206 236 236 197 197 220 220 170 170 96 104 -4.075887 35.969028

TNP74 TNP 139 147 144 144 163 164 280 280 194 198 232 236 195 195 212 220 170 174 96 98 -4.174953 36.108595

TNP75 TNP 139 147 144 152 164 198 280 361 194 198 244 256 187 198 220 224 170 174 96 96 -4.174951 36.108583

TNP76 TNP 139 147 144 144 164 172 280 353 198 214 236 236 189 195 212 220 170 174 83 98 -4.174701 36.108238

TNP77 TNP 145 151 144 144 164 164 280 353 190 194 244 244 175 195 220 224 170 174 98 106 -4.174666 36.108243

TNP78 TNP 139 147 144 144 164 164 280 353 198 214 236 256 191 195 212 212 170 170 98 98 -4.174286 36.108092

TNP79 TNP 138 151 144 152 164 198 284 353 194 194 236 240 189 198 212 212 170 170 98 106 -4.083982 36.118558

TNP80 TNP 139 147 144 144 150 164 280 284 194 198 236 236 195 195 212 228 170 170 98 106 -4.083540 36.118697

TNP81 TNP 139 139 148 148 164 198 280 280 206 206 236 244 187 198 212 228 170 170 96 101 -4.083528 36.118786

TNP82 TNP 145 145 139 144 164 170 280 353 194 206 236 236 191 195 212 224 170 170 96 96 -3.797766 36.068748

TNP83 TNP 139 145 139 144 164 192 346 353 194 214 236 240 189 195 212 212 170 174 96 98 -3.797861 36.068002

TNP84 TNP 139 145 144 152 164 198 276 353 198 202 240 240 187 198 220 220 170 176 96 96 -3.799160 36.066345

TNP85 TNP 145 151 144 144 164 192 280 357 194 194 240 240 195 195 232 232 170 174 96 98 -3.799243 36.066242

188

TNP86 TNP 139 145 152 152 164 192 280 284 194 206 236 236 195 195 212 212 170 174 96 104 -3.798985 36.064107

TNP9 TNP 147 149 150 152 164 164 280 353 190 198 248 248 187 195 212 232 170 174 98 98 -3.872970 36.066567

RUNP01 RNP 138 147 139 144 164 172 268 353 198 210 236 240 189 205 216 220 170 175 96 98 -7.686914 34.924863

RUNP02 RNP 147 147 144 144 164 172 353 357 194 198 240 244 189 205 220 236 170 175 96 104 -7.686885 34.924888

RUNP03 RNP 144 151 139 144 164 198 280 280 198 198 236 236 189 198 216 228 175 178 98 101 -7.686892 34.924902

RUNP04 RNP 138 144 144 152 164 192 280 284 186 221 236 248 189 195 212 236 170 174 96 98 -7.686883 34.924913

RUNP05 RNP 144 149 139 144 164 164 280 284 198 210 236 240 189 205 216 228 170 174 96 98 -7.686892 34.924934

RUNP06 RNP 145 151 139 145 164 198 280 280 198 198 236 236 189 198 0 0 174 174 96 96 -7.687005 34.924895

RUNP07 RNP 147 151 144 145 164 164 280 280 198 198 236 236 187 189 216 216 174 174 96 98 -7.686815 34.924855

RUNP08 RNP 138 145 139 144 164 198 353 357 194 210 244 256 189 189 216 224 174 174 98 101 -7.686815 34.924855

RUNP09 RNP 138 147 139 144 164 172 353 353 198 210 236 240 189 205 216 220 170 175 96 98 -7.683660 34.927705

RUNP10 RNP 145 145 139 139 164 198 353 353 198 210 236 240 189 205 216 228 174 174 96 98 -7.684540 34.927950

RUNP11 RNP 145 151 139 145 164 198 280 280 190 198 236 236 189 198 216 228 174 174 96 96 -7.684687 34.928018

RUNP12 RNP 138 147 144 152 164 172 280 353 194 210 236 240 187 187 204 224 173 175 92 98 -7.682115 34.939370

RUNP13 RNP 147 147 144 152 164 172 280 280 190 202 236 236 187 195 212 224 170 174 98 101 -7.668717 34.940339

RUNP14 RNP 143 147 139 144 164 172 280 357 194 198 240 252 187 187 224 236 175 175 98 98 -7.667039 34.936944

RUNP15 RNP 145 151 144 152 164 198 280 280 194 198 240 240 187 189 208 224 170 174 96 98 -7.642187 34.920516

RUNP16 RNP 145 149 144 152 164 172 361 384 194 198 236 240 189 195 224 224 170 175 96 104 -7.642195 34.920540

RUNP17 RNP 145 149 141 152 164 198 280 280 186 186 236 244 187 198 220 220 170 174 94 98 -7.603039 34.887808

RUNP18 RNP 138 151 139 141 164 172 361 384 198 206 240 244 189 197 212 220 170 175 92 104 -7.600367 34.903565

RUNP19 RNP 147 147 139 145 164 172 280 284 186 194 240 244 187 187 224 236 170 174 96 101 -7.600367 34.903565

RUNP20 RNP 138 147 139 150 164 172 280 280 206 217 240 240 187 189 220 220 170 170 96 96 -7.600784 34.907208

RUNP21 RNP 144 147 139 144 164 172 280 280 194 198 240 240 187 189 212 216 170 170 98 98 -7.600733 34.906536

RUNP22 RNP 138 147 139 152 164 172 280 280 198 206 232 240 187 189 212 220 170 175 98 106 -7.598866 34.903167

RUNP23 RNP 145 147 144 144 164 172 280 280 198 198 240 240 184 187 216 236 170 174 92 98 -7.598863 34.903171

RUNP24 RNP 145 147 152 154 164 198 276 280 190 190 240 240 189 195 220 228 170 175 98 106 -7.598769 34.903158

RUNP25 RNP 147 149 144 144 164 172 280 280 202 217 240 240 189 195 224 224 170 170 96 98 -7.598866 34.903167

RUNP26 RNP 147 151 144 152 164 172 280 280 198 202 240 260 187 195 220 224 170 170 96 96 -7.598863 34.903171

RUNP27 RNP 138 145 139 144 164 164 353 353 198 198 240 244 189 195 212 216 166 175 104 104 -7.598769 34.903158

RUNP28 RNP 138 151 139 139 164 172 284 353 198 198 240 244 189 197 212 220 170 175 92 104 -7.598439 34.902928

189

RUNP29 RNP 147 151 144 152 164 164 280 280 202 202 240 260 187 189 224 224 170 170 96 98 -7.598402 34.902942

RUNP30 RNP 145 149 144 144 164 164 353 353 198 198 240 256 187 195 216 224 170 170 96 96 -7.598235 34.903063

RUNP31 RNP 139 145 144 144 164 172 337 353 190 202 220 240 189 195 224 224 170 174 96 104 -7.600984 34.901363

RUNP32 RNP 138 151 139 144 164 172 280 353 190 194 236 240 189 189 212 228 170 175 104 106 -7.600939 34.901381

RUNP33 RNP 145 149 139 144 164 172 280 280 198 206 236 248 189 189 220 228 175 179 96 106 -7.601390 34.910099

RUNP34 RNP 147 147 144 144 164 172 280 369 194 202 240 240 187 187 224 232 170 174 98 98 -7.668392 34.940275

RUNP35 RNP 145 145 152 152 164 169 286 357 190 198 236 244 187 187 212 228 174 174 96 106 -7.681106 34.930030

RUNP36 RNP 147 149 144 144 164 172 280 280 186 202 236 236 189 205 212 216 174 174 96 96 -7.680317 34.929788

RUNP37 RNP 147 149 144 144 164 172 280 365 194 194 236 240 189 193 216 220 174 179 96 106 -7.680135 34.929894

RUNP38 RNP 138 151 144 152 164 198 280 280 194 198 240 240 189 205 212 228 170 174 96 96 -7.680098 34.929889

RUNP39 RNP 145 147 144 152 164 198 284 357 229 229 240 240 189 195 228 232 170 170 98 98 -7.658746 34.987449

RUNP40 RNP 151 151 141 152 164 198 280 280 198 198 244 244 187 189 216 228 175 175 94 98 -7.635022 34.996774

RUNP41 RNP 145 147 144 152 164 170 280 284 194 210 240 244 187 198 216 220 174 175 96 96 -7.625605 34.996762

RUNP42 RNP 144 151 139 145 164 198 280 280 0 0 244 244 187 189 208 216 174 175 96 98 -7.627088 34.994111

RUNP43 RNP 149 149 141 152 164 198 284 369 186 186 236 236 187 198 224 224 170 170 98 101 -7.615160 35.006326

RUNP44 RNP 138 147 147 154 164 169 282 282 186 186 244 244 187 189 208 216 170 170 96 104 -7.589517 35.010639

RUNP45 RNP 139 151 139 152 164 164 282 284 198 206 236 236 187 187 224 224 170 174 98 106 -7.571047 35.026366

RUNP46 RNP 145 145 144 144 164 169 282 282 194 194 240 248 187 198 220 220 174 174 92 98 -7.568284 35.025169

RUNP47 RNP 145 147 139 152 164 172 276 280 194 206 240 244 189 195 216 220 170 174 96 98 -7.560732 35.011777

RUNP48 RNP 139 147 139 144 164 198 280 280 194 194 240 240 187 198 212 212 170 174 96 98 -7.561250 35.012056

MSGR01 SGR 145 145 144 144 164 198 280 292 194 206 240 244 195 195 208 224 170 176 98 98 -7.684613 38.133654

MSGR02 SGR 138 145 139 152 164 172 280 280 214 214 248 256 187 189 212 220 170 174 96 96 -7.682956 38.129067

MSGR03 SGR 145 145 144 144 164 192 337 361 276 280 236 252 187 187 228 228 174 174 0 0 -7.683035 38.128628

MSGR04 SGR 138 145 139 150 164 164 276 357 0 0 240 252 187 187 212 228 174 174 98 98 -7.683400 38.127878

MSGR05 SGR 145 147 144 150 164 192 276 280 206 210 236 248 187 187 208 212 174 174 96 96 -7.683508 38.127918

MSGR06 SGR 138 145 152 152 164 172 280 353 202 210 240 248 187 195 216 228 166 174 96 96 -7.683869 38.128153

MSGR07 SGR 138 145 144 144 164 164 357 361 202 206 240 252 187 189 212 224 174 174 98 98 -7.684092 38.127946

MSGR08 SGR 145 151 141 150 164 172 0 0 194 198 236 252 184 189 212 224 174 174 98 98 -7.684359 38.127793

MSGR09 SGR 144 147 144 152 164 172 357 357 194 214 236 248 184 189 220 224 170 170 96 98 -7.683025 38.126818

MSGR10 SGR 144 147 144 152 164 164 280 280 210 214 236 248 187 195 212 224 170 174 96 98 -7.682989 38.126818

190

MSGR11 SGR 138 149 144 144 164 169 280 361 194 202 244 244 187 187 216 228 174 174 98 98 -7.680814 38.126525

MSGR12 SGR 138 149 144 144 164 169 280 280 194 202 244 244 187 187 216 228 174 174 98 98 -7.680832 38.126510

MSGR13 SGR 145 145 139 139 164 198 280 365 186 198 240 248 184 187 216 224 170 174 96 98 -7.680762 38.126517

MSGR14 SGR 145 149 144 144 164 172 280 280 194 198 260 260 187 189 123 123 0 0 94 104 -7.678586 38.166654

MSGR15 SGR 145 151 142 142 164 172 280 280 194 198 252 260 184 189 232 232 168 168 96 96 -7.662342 38.133172

MSGR16 SGR 144 151 142 142 164 172 280 365 192 192 244 260 184 187 140 140 168 174 96 96 -7.662367 38.133219

MSGR17 SGR 149 151 142 144 164 172 280 280 192 192 260 260 184 191 216 220 0 0 96 104 -7.623109 38.104246

MSGR18 SGR 143 145 144 144 164 172 280 353 194 221 240 240 187 195 216 220 174 174 98 101 -7.623098 38.104252

MSGR19 SGR 138 147 144 144 164 172 280 280 202 206 236 244 195 195 220 224 170 170 96 98 -7.572124 37.898436

MSGR20 SGR 138 145 139 144 164 172 337 361 194 198 240 248 187 187 216 224 170 174 96 98 -7.572059 37.898360

MSGR21 SGR 145 145 144 144 164 172 280 376 186 206 236 244 193 195 208 216 174 178 98 101 -7.584070 37.837331

MSGR22 SGR 145 147 139 145 164 164 280 280 194 198 240 256 187 189 216 216 172 174 87 96 -7.533154 37.766789

KPR01 SGR 145 145 139 139 164 172 276 280 186 206 252 256 187 198 216 224 170 170 98 101 -8.449560 38.597930

KPR02 SGR 139 145 144 144 164 163 276 280 186 206 240 244 188 188 216 224 174 174 98 104 -8.449560 38.597930

KPR03 SGR 139 145 144 144 164 172 280 280 194 194 244 244 195 195 224 236 172 174 98 104 -8.447919 38.600565

KPR04 SGR 138 147 144 144 164 198 337 365 194 214 240 244 187 189 212 232 174 174 98 98 -8.476557 38.570803

KPR05 SGR 138 145 139 156 164 198 353 357 194 210 244 256 189 189 216 224 174 174 98 101 -8.476335 38.570982

KPR06 SGR 151 151 144 144 164 172 276 280 198 214 244 256 184 184 216 224 170 174 96 98 -8.448682 38.600084

KPR07 SGR 145 149 152 156 164 198 280 280 198 202 244 244 195 195 228 236 174 174 98 98 -8.464123 38.552585

KPR08 SGR 145 145 139 156 164 198 280 280 186 206 252 256 187 187 216 224 170 170 98 101 -8.464123 38.552585

KPR09 SGR 145 145 152 152 164 172 280 280 198 225 240 244 187 187 208 208 174 174 96 98 -8.464744 38.551758

KPR10 SGR 138 145 152 152 164 198 276 346 186 202 240 244 189 198 131 220 170 170 96 104 -8.464770 38.551890

KPR11 SGR 145 149 156 156 164 198 276 276 198 202 244 244 184 195 228 236 175 175 98 98 -8.464893 38.551856

KPR12 SGR 145 145 141 144 164 198 276 280 221 225 240 248 187 189 220 236 174 174 96 104 -8.465114 38.551742

KPR13 SGR 145 145 144 145 164 198 276 280 190 194 244 252 198 198 216 220 170 172 101 104 -8.465125 38.551639

KPR14 SGR 138 144 144 152 164 172 284 365 214 221 240 248 184 189 224 236 174 174 96 98 -8.465374 38.551540

KPR15 SGR 138 151 139 152 164 198 276 276 214 214 240 244 187 198 240 240 173 173 98 104 -8.465539 38.551537

KPR16 SGR 145 147 144 144 164 198 392 392 198 198 240 240 187 198 140 220 172 174 96 96 -8.465762 38.551315

KPR17 SGR 145 145 144 144 164 172 392 392 198 206 244 252 187 189 216 232 170 172 101 101 -8.466027 38.550809

KPR18 SGR 138 145 144 144 164 175 307 365 198 198 240 244 187 187 0 0 174 174 96 98 -8.466262 38.550871

191

KPR19 SGR 138 145 141 144 164 198 280 372 186 202 244 248 187 189 220 224 170 170 96 104 -8.466424 38.550892

KPR20 SGR 138 145 144 144 164 172 280 280 186 202 244 252 187 189 212 224 170 174 98 101 -8.465790 38.551231

KPR21 SGR 145 145 144 145 164 198 280 372 194 214 248 252 184 187 216 220 170 172 101 104 -8.463146 38.553408

KPR22 SGR 138 145 144 144 164 198 357 357 186 202 240 248 195 198 220 224 170 170 96 104 -8.463235 38.554035

KPR23 SGR 138 145 144 144 164 198 280 372 186 202 240 248 187 189 220 220 170 174 96 104 -8.462903 38.554180

KPR24 SGR 145 149 139 144 164 172 337 361 194 202 240 240 187 187 228 232 174 174 94 98 -8.462836 38.554159

KPR25 SGR 145 145 144 152 164 172 357 365 202 221 240 252 187 189 216 220 174 174 96 104 -8.462919 38.554213

KPR26 SGR 145 151 144 144 164 198 280 280 192 214 240 248 187 189 216 236 174 174 96 98 -8.462951 38.554204

KPR27 SGR 138 145 141 152 164 198 357 365 186 202 240 248 187 189 220 224 170 174 96 104 -8.462852 38.555577

KPR28 SGR 149 149 152 152 164 164 280 280 190 194 240 252 187 187 208 220 174 174 96 98 -8.462518 38.555585

192

Appendix B: Sequences of 33 mitochondrial DNA haplotypes identified in this study. We have included the frequency of each haplotype for each haplotype

>ECSE09x250

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECSE11x14

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCACGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECGR02x3

TAGAATATCCTTACATCATTATCGGCCAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCATT AAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTATC AATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCTC

>SWTA01x135

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGCTCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTACA

193

TAGCACATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SWMR01x5

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGCTCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTACA TAGCACATTACTGAGAAATCTCTAGTCACCAAGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTG TCCCTC

>SSRU01x3

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG03x5

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCGCATCACTGAGAAATCTCTAGTCTCCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSE01x6

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGCCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

194

>SWMR02x13

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAACCCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTACA TAGCACATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SWMR03x6

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGCTCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTACA TAGCACATTACTGAGAAATCTCTAGTCGCCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>ECLM06

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTGCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECNG03x15

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

195

>SWLM02x2

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGCTCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTACA TAGCACATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCGCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECLM03x4

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTAGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCAAGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECSE07x2

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>SSNG02x6

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTGCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

196

>SWSE04x20

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTAGCCCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SWSE02x10

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTTCGGTCCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCCCTTGATCGTGCA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTATCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>ECMG01

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTCCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SWSE03x3

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCTTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAAGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGCCCATGGATATTGTTCACCCACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATTACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

197

>ECGR01

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTA TCAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATA TATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTC AAGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTA CATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCC

>ECSE10x5

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTATCA AGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGTAC ATAGCGCATTACTGAGAAATCTCTAGTCGTCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

>ECSE08

TAGAATATCCTTACATCATTATCGGCCAAATAGCCTCCATTCTATACTTCTCCATTATTCTAGCTTTCTTGCCAATTGCAGGAATAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGACATTACAATGGTCTTGTAAGCCATAAATGAAAACTATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAAACTGAAGTTCTTCTTAAACTATTCCCTGCAAGCAAACCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATAT ATATTACTAACTATACTTAATCTTACATAGACCATACCATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACCAATGTAT CAAGTCATACTCATGTAGATTCACAGATCATGTTCTAGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGATCGT ACATAGCGCATTACTGAGAAATCTCTAGTCATCATGCATATCACCTCCAACAGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTC GTGTCCC

>SSSG01x5

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCGCATCACTGAGAAATCTCTAGTCACCATGCATATCACCTGCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGT GTCCCT

198

>SSSG02x5

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSNG01x6

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG04

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCGCATCACTGAGAAATCTCTAGTCGCCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG06

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTCACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TGGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

199

>SSSG07

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTCACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG10x2

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACGGTTGAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG08

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCGCCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

>SSSG09

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGTCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATAT ATTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCA AGTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATCATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAGCTCTTGGTCGTACA TAGCACATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

200

>SSSGR05x5

TAGAATATCCTTACATCATTATTGGTCAAATAGCCTCAATTCTATATTTCTCCATTATCCTAGCCTTCCTGCCAATTGCAGGAGTAATCGAAAACTACCTCAT TAAGTAACCCCTATAGTATAAGATATTACAATGGTCTTGTAAGCCATAAATGAAAGCCATTTTCTAAGGGTATTCAGGGAAGAGGTCCACTTACCTCGCTAT CAATACCCAAAACTGAAATTCTTCTTAAACTATTCCCTGCAGGCAATCAACCCGCTATGTATATCGTGCATTAAATGCTTGTCCCCATACATAATGATATATA TTACTAACTATACTTAATCTTACATAGACCATACTATGTATAATCGTGCATCACATTATTTACCCCATGCTTATAAGCAAGTACTGTTTAACTAATGTGTCAA GTCATATTCATGTAGATTCACAGGTCATGTTCTGGTTCATGGATATTATTTACCTACGATAAACCATAGTCTTACATAGCACATTAAAGCTCTTGGTCGTACA TAGCGCATCACTGAGAAATCTCTAGTCACCATGCATATCACCTCCAATGGTTGTACCTTAACTACCTACCTCCGAGAAACCATCAACCCGCCCATCTTCGTGT CCCTC

201

VITA

George Martin Gwaltu Lohay Education The Pennsylvania State University University Park, PA Ph.D. Candidate in Biology, GPA: 3.6/4 Expected August 2019

Uganda Martyrs University Jinja, Uganda B.A. in Philosophy, GPA: 4.2/5 May 2014

University of Dar es salaam Dar es salaam, Tanzania B. Sc. Wildlife Science and Conservation 3.4/5 December 2009

Research experiences Ph.D. Researcher| The Pennsylvania State University August 2014-July 2019 • Collected, extracted DNA, sequenced and genotyped 800 elephant samples • Designed a method to determine the sex of elephants using AMELX/Y gene • Mapped the mitochondrial DNA haplotypes for elephants in Tanzania Research assistant| Serengeti Lion Project October 2009-July 2011

Teaching experience Graduate teaching assistant| The Pennsylvania State University, August 2014-July 2019 • Spring 2016, Teaching Assistant BIOL 220W Biology of communities • Instructor Swahili Spring and Fall 2016SWA001, SW003, SWA003 Publications and Presentations • Lohay, G. M. G., An accurate molecular method to sex elephants using PCR amplification of Amelogenin gene (To be submitted) • Lohay, G. M. G, Weathers, T. C, Estes, B. A., and Cavener, D.R., Genetic connectivity and population structure of African Savanna elephants in Tanzania (To be submitted) • Lohay, G. M. G, Weathers, T. C, Estes, B. A., and Cavener, D.R., Little evidence for female mediated-gene flow for the African savanna elephants between the Greater Ruaha and Selous ecosystems in Tanzania (To be submitted) • Lohay, G. M. G and Cavener, D.R., Assessment of age and sex structure of African savanna elephants in the Serengeti ecosystem using a novel fecal-centric method (To be submitted) Awards and Grants Huck Life Sciences Institute ($9,450), Wildlife Conservation Society, Fellowship (US $50,000), Cleveland Metroparks Zoo, African Seed Grants (US $3400) and 2017 Graduate exhibition second position, Penn State Graduate School (US $250)