Ecology and evolutionary adaptation in anthropogenically fragmented habitat

In this thesis, I carried out flora community composition surveys, pollinator observations, plant offspring fitness evaluation, and genetic analysis to investigate the impact of anthropo

General Introduction

Introduction

1.1.1 Anthropogenic habitat fragmentation and its impact on biodiversity and ecosystem function

Habitat fragmentation is defined as a process of breaking-up continuous habitats into several smaller fragments that are separated from one another by surrounding matrix (Feeley & Terborgh, 2008; Harrison & Bruna, 1999; Laurance et al., 2002; Wilcove et al., 1986; Wu,

2009) This process of habitat transformation can occur naturally over a long time through to natural factors such as waves or currents (Fonseca, 1998), fire (Franklin et al., 2002), waterfall, cascades (Fuller et al., 2015) Habitat fragmentation can also be formed by human activities such as road construction, logging, agricultural expansion, and urbanisation (Ribeiro et al., 2009; Skole & Tucker, 1993; Zuidema et al., 1996): this is termed anthropogenic habitat fragmentation The large scale of human activities are strongly correlated with habitat loss as habitat fragmentation creates several habitat patches that are smaller in size and more isolated at the endpoint of the fragmentation process (Fahrig, 2003; Ranta et al., 1998) Anthropogenic habitat destruction and modification significantly contribute to population decline and disappearance There is mounting evidence that the impact of population fragmentation varies depending on the species’ life history and autecology (Johnstone et al., 2010) The distribution and abundance of species can be significantly impacted by anthropogenic habitat fragmentation, which has implications for how well ecosystems function and provide related ecosystem services

Theoretical and empirical approaches have been taken in elucidating the effects of fragmentation on species immigration, survival and species diversity over the years (Bélisle et al., 2001; Best et al., 2001; Fischer & Lindenmayer, 2007; Hanski et al., 1996; Kurki et al.,

2000; MacArthur & Wilson, 1967) Many researchers focused on the degree of isolation of habitat patches, reduction of habitat amount and the decrease in patch sizes as the main effects of the fragmentation process on habitat patterns (Concostrina-Zubiri et al., 2018; ệckinger et al., 2009; Roque et al., 2017; Tabarelli et al., 1999; Young et al., 1993) Researchers have found that habitat fragmentation may adversely affect the original biodiversity (Fahrig, 2003; Zuidema et al., 1996) As the habitat becomes fragmented, many habitat fragments with different sizes and shapes are created (Baskent & Jordan, 1995) For example, the small fragments have proportionally more edge for a given amount of habitat (Fahrig, 2003) Therefore, small fragments may suffer more edge effects that could change the properties of these small landscapes, such as humidity, wind speed, temperature and soil nutrients (Matlack, 1993; Weathers et al., 2001) Furthermore, smaller fragments might be insufficient to sustain thriving populations, so some individuals may leave the habitat to enter the matrix or be forced into more isolated fragments; they may face increased predator risks in such habitats (Chalfoun et al., 2002; Zuidema et al., 1996)

The changes in habitat properties after the fragmentation process may alter species interaction (Taylor & Merriam, 1996), lead to species inbreeding (Kurki et al., 2000), affect dispersal success (Bélisle et al., 2001; King, 1999; With & Crist, 1995), reduce trophic chain length (Komonen et al., 2000) and increase infection rate and disease transmission (Allan et al., 2003; Ewers & Didham, 2006) Therefore, these processes could lead to the reduction of the overall reproductive rate and increased mortality rate (Fahrig, 2002) As a result, the persistence and stability of local species may be threatened (Andren, 1994; Wiens, 1976; Wolff et al., 1997), and the risk of local extinction could be increased (Bender et al., 1998; Bowers & Matter, 1997; Ewers & Didham, 2006)

Early research to understand the impacts of habitat loss and fragmentation was strongly influenced by the equilibrium theory of island biogeography (MacArthur & Wilson 1963) and

3 the meta-population dynamics (Levins & Extinction, 1970) However, fragmented habitats are not entirely comparable to oceanic islands, and fragments have frequently been shown to be genetically connected Over the past few decades, the discovery of strong context dependence in ecosystem responses, including spatial context dependence at multiple scales, time-lag population dynamics, and trait-dependent species responses, represents a significant advancement in our understanding of the effects of fragmentation (Coudrain et al., 2013) Additional insights have also emerged on species evolutionary dynamics (Eriksson, 2014) and its strong influence on species richness, abundance and (animal) behaviour and dispersal (Davies & Asner, 2014) following anthropogenic habitat fragmentation

1.1.2 Habitat fragmentation in biodiverse Southwest Western Australia Botanical

The Mediterranean climate region in Southwest Western Australia Botanical Province is the only region in Australia that has been listed as one of the 25 biodiversity hotspots in the world (Hopper & Gioia, 2004; Myers et al., 2000; Wilson et al., 2009) Over the past several decades, large native vegetation areas in the region have been cleared for agricultural purposes (Hobbs, 2001; Saunders, 1993) It was estimated that 70% of the vegetation in the area has been cleared and the ratio goes up to 90% in most of the Wheatbelt region (Hobbs, 1993; Wilson et al.,

2009), where 85% of remnant vegetation area is private property (Hobbs, 2003) The massive land clearing resulted in highly fragmented habitats with a small number of native flora and fauna remaining in some areas (Hobbs, 1993) These remnant fragments are usually severely disturbed by weed invasion, livestock grazing and trampling (Arnold & Weeldenburg, 1991) and by the changes in ecological and environmental conditions in the surrounding agricultural matrix (Hobbs, 2003) As a result, the number of species and populations in the fragmented habitats continues to decline, with 348 plant species being listed as rare and endangered in the Wheatbelt region (Hopper et al., 1990) It is believed that at least 70 plant species have become

4 extinct in Western Australia in general (Dixon, 1994) It is estimated that 24 plant species and

13 mammal species in the Wheatbelt region have disappeared after land clearing (Kitchener et al., 1980; Leigh et al., 1984) Despite several studies examined the effect of fragmentation on fauna and flora over the last 30 years in Western Australia (Abensperg-Traun et al., 1996; Arnold et al., 1995; Delnevo et al., 2019; Hobbs, 2001; Saunders, 1993; Yates et al., 2007) these studies mainly focus on the effect of fragmentation and land clearing on fauna such as birds (Fulton, 2013; Fulton & Lawson, 2021; Luck, 2002; Wilson & Recher, 2001), marsupials (Fulton, 2017; Pacioni et al., 2011), foxes (Marlow et al., 2014), and pythons (Pearson et al.,

2005) Significant knowledge gaps exist in how individual mechanisms of anthropogenic fragmentation effects, i.e patch size, time-lagged effects, species interaction across trophic levels, gene and energy flow dynamics, evolutionary dynamics, determining biodiversity and ecosystem function and service within an integrated theoretical frame.

Objective and overview of the thesis

This thesis aims to investigate the impact of anthropogenic habitat fragmentation on biodiversity and ecosystem function in Dryandra Woodland, and how species-specific responses to the changes in the fragment sizes and isolation distances Species response to the impact of fragmentation was determined by conducting community composition surveys, pollinator observation, greenhouse experiments and genetic analyses This thesis focuses on the response of plant communities to anthropogenic habitat fragmentation Four main objectives were presented in this research:

1 To determine how anthropogenic habitat fragmentation affects species diversity and composition in Dryandra Woodland

2 To determine how anthropogenic habitat fragmentation affects plant fitness and growth

3 To determine how anthropogenic habitat fragmentation affects plant reproduction and plant-pollinator interaction

4 To determine whether anthropogenic habitat fragmentation affects the genetic diversity and differentiation of particular species

I tested four hypothesises following the theory of island biogeography and population genetics

Hypothesis 1: Smaller and more isolated fragments will have lower species diversity and composition than larger fragments

Hypothesis 2: Seedlings of plants from larger and less isolated fragments should have better fitness (e.g growing faster) than those from smaller and isolated fragments

Hypothesis 3: Smaller and more isolated fragments will have lower species reproduction output and less efficient pollination

Hypothesis 4: Plant populations in smaller and more isolated fragments will have lower genetic diversity and higher genetic differentiation

To achieve these four mentioned objectives, I began implementing the first objective by conducting flora surveys in the selected fragments to investigate whether there are differences in species diversity and composition between big and small fragments as well as more isolated versus less isolated fragments (Chapter 2) Then, in Chapter 3, I assessed how the growth performance of the three common species: Banksia sessilis, Gastrolobium parviflorum, and

Bossiaea eriocarpa, are affected by fragment size and isolation distance (the second objective)

The growth performance of plant species in fragmented habitats has been considered to be affected by fragmentation in many studies from temperate, subtropical and tropical regions Still, very few studies have focused on the Mediterranean-climate regions Besides the potentially adverse effects on plant growth, habitat fragmentation could impact species reproduction and alter plant-pollinator interaction Therefore, in Chapter 4, I aimed to

6 determine the effect of fragmentation on plant-pollinator interactions of Banksia sessilis and reproduction of Banksia sessilis, Gastrolobium parviflorum and Bossiaea eriocarpa (the third objective) After assessing the impact of fragmentation on phenotypic traits, in Chapter 5, I used whole-genome sequencing techniques to investigate the impact of fragmentation on the genomic architecture of Banksia sessilis by comparing the genetic diversity and genetic differentiation between three large populations in three large fragments and three small populations in small fragments varying in degrees of isolation (the fourth objective) Finally, I summarise the results and findings and discuss the implications of the results in Chapter 6.

Study area

The Dryandra Woodland (32 o 48’S, 116 o 54’E) is located in the Narrogin district (Figure 1.1), about 160 km southeast of Perth in WA’s Wheatbelt (Burrows et al., 1987; Keighery & Keighery, 2012) The region experiences a typical Mediterranean climate characterised by cool wet weather in the winter and hot, dry weather in the summer The long term annual average rainfall is approximately 490 mm, which mainly occurs during the winter season, but some unpredictable rain accompanied by thunderstorms can also occur in the summer (Burrows et al., 1987; Garkaklis et al., 2004) Dryandra Woodland has various landforms, including lateritic plateaus, slopes, some occasional granite outcrops, flat sand plains and valleys with elevations ranging from 240 m to 440 m above sea level (Burrows et al., 1987) It consists of 24 distinct blocks that are separated from one another by tracts of agricultural land and are dispersed over an area of about 50 km from north to south The blocks range in size from 87 ha to 12,192 ha, making up the entire 27,947 ha area (Keighery & Keighery, 2012) Dryandra was connected to numerous vast areas in the east that contained native bush as well as to the main forest belt of the Darling Range as late as 1962 (Figure 1.2) Since then, many blocks in Dryandra Woodland

7 have been increasingly separated from these areas and each other by continued land clearing (Conservation Commission, 2011)

Dryandra Woodland has recently been listed as a National Park to recognise its biodiversity and cultural heritage significance The Dryandra Woodland contains over 800 native plants, 24 mammals, 98 bird, and 41 reptile species, including 15 plant species that have been designated priority species by the Department of Environment and Conservation (former Department of Wildlife and Parks Western Australia Government) Declared Rare and Priority Flora List (Department of Environment and Conservation, 2011) According to the Dryandra Woodland report of the Department of Environment and Conservation (2011), Western Australia's Wheatbelt once supported an incredibly diverse flora and fauna in which the most extraordinary floristic diversity existed in the Lateritic Plateau Woodlands, Dryandra, Petrophile Shrublands, Low Kwongan, Marri Woodlands and Lithic Complexes (Figure 1.3) However, the area occupied by these communities in the Dryandra forest is relatively small (Conservation Commission, 2011) Dryandra Woodland now is the last remnant of many species that have declined or disappeared from the surrounding cleared land, due to retaining large areas of old-growth forest Dryandra woodland comprises 12 different vegetation associations with 813 native plant species and supports ten species of threatened fauna in need of a particular protection (Keighery & Keighery, 2012) Consequently, Dryandra woodland provides an excellent system for research on the effects of anthropogenic fragmentation on an ecosystem with high conservation value

Figure 1.1: Insert shows the location of Dryandra Woodland, 160 km southwest of Perth,

Figure 1.2: Land clearing and fragmentation at Dryandra Woodland (1950 – 1993) ( Conservation Commission, 2011)

Figure 1.3: Some vegetation of Dryandra Woodland

References

Abensperg-Traun, M., Smith, G T., Arnold, G., & Steven, D (1996) The effects of habitat fragmentation and livestock-grazing on animal communities in remnants of gimlet Eucalyptus salubris woodland in the Western Australian wheatbelt I Arthropods

Allan, B F., Keesing, F., & Ostfeld, R S (2003) Effect of forest fragmentation on Lyme disease risk Conservation Biology, 17(1), 267-272

Andren, H (1994) Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review Oikos, 71(3) 355-366

Arnold, G., Weeldenburg, J., & Ng, V (1995) Factors affecting the distribution and abundance of Western grey kangaroos (Macropus fuliginosus) and euros (M robustus) in a fragmented landscape Landscape Ecology, 10(2), 65-74

Arnold, G W., & Weeldenburg, J (1991) The distributions and characteristics of remnant native vegetation in parts of the Kellerberrin-Tammin-Trayning and Wyalkatchem shires of Western Australia

Baskent, E Z., & Jordan, G A (1995) Characterizing spatial structure of forest landscapes

Canadian Journal of Forest Research, 25(11), 1830-1849

Bélisle, M., Desrochers, A., & Fortin, M.-J (2001) Influence of forest cover on the movements of forest birds: a homing experiment Ecology, 82(7), 1893-1904

Bender, D J., Contreras, T A., & Fahrig, L (1998) Habitat loss and population decline: a meta‐analysis of the patch size effect Ecology, 79(2), 517-533

Best, L B., Bergin, T M., & Freemark, K E (2001) Influence of landscape composition on bird use of rowcrop fields The Journal of Wildlife Management, 442-449

Bowers, M A., & Matter, S F (1997) Landscape ecology of mammals: relationships between density and patch size Journal of Mammalogy, 78(4), 999-1013

Burrows, N., McCaw, W., & Maisey, K (1987) Planning for fire management in Dryandra forest In Saunders, Denis A (Ed.) Nature conservation : the role of remnants of native vegetation.Chipping Norton, NSW : Surrey Beatty and Sons

Conservation Commission (2011) Dryandra Woodland Management Plan 2011 Conservation

Commission of Western Australia, Perth

Chalfoun, A D., Thompson III, F R., & Ratnaswamy, M J (2002) Nest predators and fragmentation: a review and meta‐analysis Conservation Biology, 16(2), 306-318 Concostrina-Zubiri, L., Martínez, I., & Escudero, A (2018) Lichen-biocrust diversity in a fragmented dryland: Fine scale factors are better predictors than landscape structure

Science of the Total Environment, 628, 882-892

Coudrain, V., Herzog, F., & Entling, M H (2013) Effects of habitat fragmentation on abundance, larval food and parasitism of a spider-hunting wasp PLoS One, 8(3), e59286

Davies, A B., & Asner, G P (2014) Advances in animal ecology from 3D-LiDAR ecosystem mapping Trends in Ecology and Evolution, 29(12), 681-691

Delnevo, N., van Etten, E J., Byrne, M., & Stock, W D (2019) Floral display and habitat fragmentation: effects on the reproductive success of the threatened mass‐flowering

Conospermum undulatum (Proteaceae) Ecology and Evolution, 9(19), 11494-11503

Dixon, K W (1994) Towards integrated conservation of Australian endangered plants—the

Western Australian model Biodiversity and Conservation, 3(2), 148-159

Eriksson, O (2014) Vegetation change and eco‐evolutionary dynamics Journal of Vegetation

Ewers, R M., & Didham, R K (2006) Confounding factors in the detection of species responses to habitat fragmentation Biological Reviews, 81(1), 117-142

Fahrig, L (2002) Effect of habitat fragmentation on the extinction threshold: a synthesis

Fahrig, L (2003) Effects of habitat fragmentation on biodiversity Annual Review of Ecology,

Feeley, K J., & Terborgh, J (2008) Direct versus indirect effects of habitat reduction on the loss of avian species from tropical forest fragments Animal Conservation, 11(5), 353-

Fischer, J., & Lindenmayer, D B (2007) Landscape modification and habitat fragmentation: a synthesis Global Ecology and Biogeography, 16(3), 265-280

Fonseca, M S (1998) Guidelines for the conservation and restoration of seagrasses in the

United States and adjacent waters (Vol 55): US Department of Commerce, National

Franklin, A B., Noon, B R., & George, T L (2002) What is habitat fragmentation? Studies in Avian Biology, 25, 20-29

Fuller, M R., Doyle, M W., & Strayer, D L (2015) Causes and consequences of habitat fragmentation in river networks Annals of the New York Academy of Sciences, 1355(1), 31-51

Fulton, G R (2013) Woodland birds persisting in least disturbed environment: Birds of

Dryandra Woodland Pacific Conservation Biology,

Fulton, G R (2017) Native marsupials as egg predators of artificial ground-nests in Australian woodland Australian Journal of Zoology, 65(3), 196-199

Fulton, G R., & Lawson, J (2021) Birds respond to woodland type, soil and mesic gradients in heterogeneous woodlands at Dryandra Australian Journal of Zoology,

Garkaklis, M J., Bradley, J., & Wooller, R (2004) Digging and soil turnover by a mycophagous marsupial Journal of Arid Environments, 56(3), 569-578

Hanski, I., Moilanen, A., & Gyllenberg, M (1996) Minimum viable metapopulation size The

Harrison, S., & Bruna, E (1999) Habitat fragmentation and large‐scale conservation: what do we know for sure? Ecography, 22(3), 225-232

Hobbs, R J (1993) Effects of landscape fragmentation on ecosystem processes in the Western

Hobbs, R J (2001) Synergisms among habitat fragmentation, livestock grazing, and biotic invasions in southwestern Australia Conservation Biology, 15(6), 1522-1528

Hobbs, R J (2003) The wheatbelt of Western Australia Pacific Conservation Biology, 9(1),

Hopper, S., van Leeuwin, S., Brown, A P., & Patrick, S (1990) Western Australia's endangered flora and other plants under consideration for declaration Department of

Hopper, S D., & Gioia, P (2004) The southwest Australian floristic region: evolution and conservation of a global hot spot of biodiversity Annual Review of Ecology, Evolution, and Systematics, 35, 623-650

Johnstone, C P., Reina, R D., & Lill, A (2010) Impact of anthropogenic habitat fragmentation on population health in a small, carnivorous marsupial Journal of Mammalogy, 91(6), 1332-1341

Keighery, G., & Keighery, B (2012) Vascular flora of Dryandra Woodland (LOL Gray and

Motangue State Forest) Western Australian Naturalists

King, A W (1999) Dispersal success on fractal landscapes: a consequence of lacunarity thresholds Landscape Ecology, 14(1), 73-82

Kitchener, D., Chapman, A., Muir, B., & Palmer, M (1980) The conservation value for mammals of reserves in the Western Australian wheatbelt Biological Conservation, 18(3), 179-207

Komonen, A., Penttilọ, R., Lindgren, M., & Hanski, I (2000) Forest fragmentation truncates a food chain based on an old‐growth forest bracket fungus Oikos, 90(1), 119-126 Kurki, S., Nikula, A., Helle, P., & Lindén, H (2000) Landscape fragmentation and forest composition effects on grouse breeding success in boreal forests Ecology, 81(7), 1985-

Laurance, W F., Lovejoy, T E., Vasconcelos, H L., Bruna, E M., Didham, R K., Stouffer,

P C., …Sampaio, E (2002) Ecosystem decay of Amazonian forest fragments: a 22‐ year investigation Conservation Biology, 16(3), 605-618

Leigh, J., Boden, R., & Briggs, J (1984) Extinct and endangered plants of Australia

Levins, R., & Extinction, M G (1970) Some mathematical problems in biology Extinction

Providence, Rhode Island: The American Mathematical Society, 77-107

Luck, G W (2002) The dynamics and conservation of a spatially subdivided avian population in a fragmented landscape Pacific Conservation Biology, 8(3), 159-169

MacArthur, R H., & Wilson, E O (1967) The Theory of Island Biogeography (Vol 1):

Marlow, N J., Thomas, N., Williams, A., Macmahon, B., & Lawson, J (2014) The diet of foxes (Vulpes vulpes) in fragmented wheatbelt reserves in Western Australia: implications for woylies (Bettongia penicillata) and other native fauna Conservation

Matlack, G R (1993) Microenvironment variation within and among forest edge sites in the eastern United States Biological Conservation, 66(3), 185-194

Myers, N., Mittermeier, R A., Mittermeier, C G., Da Fonseca, G A., & Kent, J (2000)

Biodiversity hotspots for conservation priorities Nature, 403(6772), 853-858 ệckinger, E., Franzộn, M., Rundlửf, M., & Smith, H G (2009) Mobility-dependent effects on species richness in fragmented landscapes Basic and Applied Ecology, 10(6), 573-578 Pacioni, C., Wayne, A., & Spencer, P (2011) Effects of habitat fragmentation on population structure and long‐distance gene flow in an endangered marsupial: the woylie Journal of Zoology, 283(2), 98-107

Pearson, D., Shine, R., & Williams, A (2005) Spatial ecology of a threatened python (Morelia spilota imbricata) and the effects of anthropogenic habitat change Austral Ecology, 30(3), 261-274

Ranta, P., Blom, T., Niemela, J., Joensuu, E., & Siitonen, M (1998) The fragmented Atlantic rain forest of Brazil: size, shape and distribution of forest fragments Biodiversity & Conservation, 7(3), 385-403

Ribeiro, M C., Metzger, J P., Martensen, A C., Ponzoni, F J., & Hirota, M M (2009) The

Brazilian Atlantic forest: How much is left, and how is the remaining forest distributed? Implications for conservation Biological Conservation, 142(6), 1141-1153

Roque, B B., Koptur, S., & Sah, J P (2017) The effects of habitat fragmentation on the reproduction and abundance of Angadenia berteroi Journal of Plant Ecology, 10(2) Saunders, D A (1993) A community-based observer scheme to assess avian responses to habitat reduction and fragmentation in south-western Australia Biological Conservation, 64(3), 203-218

Skole, D., & Tucker, C (1993) Tropical deforestation and habitat fragmentation in the

Amazon: satellite data from 1978 to 1988 Science, 260(5116), 1905-1910

Tabarelli, M., Mantovani, W., & Peres, C A (1999) Effects of habitat fragmentation on plant guild structure in the montane Atlantic forest of southeastern Brazil Biological Conservation, 91(2-3), 119-127

Taylor, P D., & Merriam, G (1996) Habitat fragmentation and parasitism of a forest damselfly Landscape Ecology, 11(3), 181-189

Weathers, K C., Cadenasso, M L., & Pickett, S T (2001) Forest edges as nutrient and pollutant concentrators: potential synergisms between fragmentation, forest canopies, and the atmosphere Conservation Biology, 15(6), 1506-1514

Wiens, J A (1976) Population responses to patchy environments Annual Review of Ecology and Systematics, 7(1), 81-120

Wilcove, D S., McLellan, C H., & Dobson, A P (1986) Habitat fragmentation in the temperate zone Conservation Biology, 6, 237-256

Wilson, B., Valentine, L., Kinloch, J., Sonneman, T., & Swinburn, M (2009) Habitat loss and fragmentation In Barbara A Wilson and Leonie E Valentine (Ed.) Biodiversity values and threatening processes of the Gnangara Department of Environment and

Wilson, K., & Recher, H F (2001) Foraging ecology and habitat selection of the yellow- plumed honeyeater, Lichenostomus ornatus, in a Western Australian woodland: implications for conservation EMU Austral Ornithology, 101(1), 89-94

With, K A., & Crist, T O (1995) Critical thresholds in species' responses to landscape structure Ecology, 76(8), 2446-2459

Wolff, J O., Schauber, E M., & Edge, W D (1997) Effects of habitat loss and fragmentation on the behavior and demography of gray‐tailed voles: Efectos de la pérdida y fragmentación de hábitat sobre el comportamiento y la demografía de ratones de campo de cola gris Conservation Biology, 11(4), 945-956

Wu, J (2009) Ecological dynamics in fragmented landscapes The Princeton guide to ecology

Yates, C J., Coates, D J., Elliott, C., & Byrne, M (2007) Composition of the pollinator community, pollination and the mating system for a shrub in fragments of species rich kwongan in south-west Western Australia Biodiversity and Conservation, 16(5), 1379-

Young, A., Merriam, H., & Warwick, S (1993) The effects of forest fragmentation on genetic variation in Acer saccharum Marsh.(sugar maple) populations Heredity, 71(3), 277 Zuidema, P A., Sayer, J A., & Dijkman, W (1996) Forest fragmentation and biodiversity: the case for intermediate-sized conservation areas Environmental Conservation, 23(4), 290-297

Effect of anthropogenic habitat fragmentations on the floristic diversity and community

Abstract

Anthropogenic habitat fragmentation has been extensively studied worldwide It is considered a significant threat to biodiversity and ecosystem function, as it disturbs biological and physical processes in many terrestrial ecosystems When comparing the effects of fragment size vs other factors like habitat alteration, studies that look at how fragmentation affects species richness frequently produce ambiguous or conflicting results The reason for this is frequently identified as the fact that different species respond differently to the same environmental disturbances This study aims to determine the two main aspects of fragmentation, i.e fragment sizes and isolation distances, on the floristic diversity and community composition at Dryandra Woodland Sixty-three 10 m × 10 m vegetation plots were surveyed across the six selected habitat patches ranging from 54 ha to 1262 ha The number of species recorded varied between habitat patches, with an average of 12 species per patch Results indicated that species diversity was not correlated with fragment sizes or isolation distances However, species abundance was significantly affected by these two aspects This could be explained by a higher number of invasive species recorded in the small and isolated fragments This study suggests that small forest fragments in Dryandra Woodland could contain similar species diversity as big fragments if they are adequately protected

Introduction

Habitat fragmentation reduces a continuous habitat into several smaller, less continuous and isolated habitats (Franklin et al., 2002; Young et al., 1996) Fragmented habitats are formed mainly due to human activities like logging, land clearing for agricultural purposes, road construction, and urbanisation (Skole & Tucker, 1993; Zuidema et al., 1996) Habitats may also become fragmented by natural processes such as fire, wind or flooding, which can be barriers to exchanges of genetic material among habitats (Andren, 1994; Franklin et al., 2002) For instance, aquatic animals that depend on a river for dispersal between different habitats will not do so if part of the river dries up

Over the past 50 years, the island biogeography (MacArthur & Wilson, 1967) and metapopulation theory (Hanski & Parmesan, 1999) have been applied to explain fragmentation effects on immigration and dispersal, population survival and species richness (Fischer & Lindenmayer, 2007), population genetic consequences and the implication for species conservation These studies predominantly concentrate on population/area size and geographic isolation of populations that occupy habitat remnants (Van Treuren et al., 1991; Young et al.,

1993) Habitat fragmentation may negatively impact the original biodiversity of a forest at multiple levels, including genes, species, and species associations (Zuidema et al., 1996) There are likely to be two leading causes for the negative effect of fragmentation (Fahrig, 2003) Firstly, fragmented habitat contains many smaller patches that can be too small to maintain local populations of certain species; these species may not migrate through the non-habitat portion between patches Due to land-use changes, many common species are then forced into more isolated and smaller remnants As a result, populations of species in fragmented habitats could have fewer individuals compared to continuous habitats or before fragmentation (Zuidema et al., 1996); this threatens the persistence and stability of these species (Andren, 1994; Wiens, 1976; Wolff et al., 1997) Generally, habitat fragmentation can be implicated in

21 the local extinctions of species in a forest or the increased decline of species (de Souza & Brown, 1994) Secondly, fragmentation resulting in more patches will also produce more edges, resulting in edge effects The edge effects refer to changes in different indicators of landscape and physical processes such as humidity, wind speed, temperature, radiation, moisture, and soil nutrition through the patch (Matlack, 1993; Saunders et al., 1991; Weathers et al., 2001) Edge effects can affect individual persistence in general; however, some species may be favoured by, and therefore benefit from the disturbance from edge effects (Lamont et al., 2003)

Some empirical evidence also indicates that the effects of fragmentation per se are less severe than those of habitat loss, and in some cases, fragmentation may be favourable for specific species (Fahrig, 2003) Other studies have clearly indicated that habitat loss has adverse effects on biodiversity across taxa as it impacts species breeding (Kurki et al., 2000), dispersal success (Bélisle et al., 2001; Pither & Taylor, 1998; With & Crist, 1995), changes in species interaction (Taylor & Merriam, 1996), shortening of the trophic chain (Komonen et al.,

2000), and the number of large-bodied species (Gibbs & Stanton, 2001) The adverse effects of habitat loss on biodiversity can be either applied to directly measured factors such as species richness (Findlay & Houlahan, 1997; Gurd et al., 2001; Schmiegelow & Mửnkkửnen, 2002; Steffan-Dewenter et al., 2002; Wettstein & Schmid, 1999), abundance and distribution of the population (Best et al., 2001; Gibbs, 1998; Guthery et al., 2001; Hanski et al., 1996; Hargis et al., 1999; Hinsley et al., 1995; Lande, 1987; Venier & Fahrig, 1996) and genetic diversity (Gibbs, 2001), or to measured factors that are highly correlated with each other such as reproductive success and average patch size and percentage of forest cover, forest interior (Echeverría et al., 2007; Ross et al., 2002; Sáyago et al., 2018) Numerous studies have demonstrated that habitat loss rather than actual habitat fragmentation has a much greater

22 negative impact on biodiversity (Fahrig, 2003); however, habitat loss and fragmentation often co-occur, so they are both considered significant threats to biodiversity globally

2.2.1 The flora of South Western Australia (SWA) and habitat fragmentation

The South West Australian Floristic Region has a Mediterranean-type climate with typical sclerophyll forest, woodlands and shrublands (Kwongan) covering 300 000 km 2 area It is an internationally recognised biodiversity ‘hotspot’ (Wilson et al., 2009), with 7 380 native plant taxa have been documented so far (Hopper & Gioia, 2004) Still, extensive clearing, fragmentation and the alternation of native of ecosystems throughout the region pose a threat to this rich diversity (Hopper & Gioia, 2004)

Despite considerable resilience to fragmentation of insect-pollinator communities (Donaldson et al., 2002; Yates & Ladd, 2005) and the persistence of rich plant communities in many small fragments in The South West Australian Floristic Region (Gibson et al., 2000; Obbens et al., 2001), about 70% of the vegetation in this region has been cleared (Wilson et al., 2009) This ratio goes up to 90% habitat loss in some parts of the Wheatbelt (Hobbs & Saunders, 2012) Anthropogenic habitat fragmentation has led to the loss of both species and ecosystem services; isolation directly impacts the ecosystem by reducing the flow of resources and organisms between habitats and therefore has significant consequences for ecosystem functioning (Fahrig, 2003) However, many uncertain effects of fragmentation on species have not yet been identified and clarified (Hobbs & Yates, 2003; Lynch & Saunders, 1991; Yates et al., 2007)

Therefore, further empirical studies integrating investigations on pollination, pollinator visitation, reproduction and mating systems for various plant species with diverse life-history characteristics are needed (Hobbs & Yates, 2003; Whelan et al., 2000) The results from those further empirical studies could help develop a comprehension of how various flora and fauna are affected by habitat fragmentation and how genetic processes and demographics in plant and

23 animal communities could be influenced by species interaction Such information would be essential for the future conservation of biodiversity and managing fragmented habitat systems (Rathcke & Jules, 1993)

This chapter investigates how anthropogenic habitat fragmentation has affected species richness, abundance, and floristic composition in different habitat fragments at Dryandra Woodland It was hypothesised that large fragments would have higher species diversity and abundance than smaller fragments.

Methods

The study was conducted at Dryandra Woodland (32 o 48’S, 116 o 54’E) located in the Narrogin district, about 160 km southeast of Perth in WA’s Wheatbelt (Burrows et al., 1987; Keighery

& Keighery, 2012) Details of the study area such as location, climate, geography, and fragmentation history have been described in Chapter I

Six fragmented patches in Dryandra Woodland were selected based on some criteria such as location, area and accessibility Three large fragments were within the large continuous forest and open to visitors, while the other three small fragments were well protected as they are in natural reserve areas (Figure 2.1) All selected patches were in Dryandra Woodland, with the size ranging from 54 ha to 1262 ha The degree of isolation was defined as the distance to the nearest fragment

Figure 2.1: Locations of the 19 transects within Dryandra Woodland A, B, and C are the large fragments D, E, F are the small fragments

The survey was conducted in Spring 2015 at Dryandra Woodland Nineteen transects were established for vegetation surveys: each transect included three or four vegetation plots of 100 square meters each (10 m x 10m) and 100 m apart from each other and 100m apart from the edge to avoid edge effects The transect locations were selected to distribute evenly throughout the fragments to cover maximum of vegetation In each fragment, 10 — 12 plots were surveyed (Figure 2.1) The size of vegetation plots was determined using the Technical Guidance of the Environmental Protection Authority (Authority, 2015) and previous studies of Griffin et al

(1983) and George et al (1979) that suggested that a 10 m × 10 m quadrat vegetation plots

25 could cover about 80% of the floristic species present in a 100 m 2 plot A Garmin GPS device was used to record the locations of each plot In each plot, every visible plant individual was recorded and counted, and a voucher specimen was taken for each species Multiple vouchers were taken if the initial field identification was in doubt Aspects, including the dominant vegetation, functional type, and soil type, were also noted The total number of species and their relative abundances presented within the plot were therefore determined

Vegetation surveys were greatly restricted due to time limitations and lack of flowering and fruiting material After collecting from the plot, each species was pressed and attached to a cardboard sheet for later identification All specimens of plants were identified down to the species level by comparing with the models at the Dryandra Woodland herbarium, Western Australian Herbarium, and with the help of herbarium taxonomists at Western Australian Herbarium Voucher specimens were stored within the Department of Environment and Agriculture, Curtin University, Bentley

To determine the effects of fragment size and distance of isolation on species richness and abundance, the number of species present was used as the response variable, and patch size (large and small) and degree of isolation (far and near) were identified as predictor variables PRIMER v7 and R (R core team, 2018) statistical software programs were used to perform all statistical analysis

For measuring floristic diversity in different patches, four indices were calculated, including species abundance, species richness (d), Shannon – Weiner diversity index (H’), and species evenness (EH)

 Species abundance (N) was used to measure the total number of plant individuals found in the area

Species richness (d) was used as a diversity index to measure the number of species found in the areas.𝒅 = (𝑺−𝟏)

Where S is the total of species, N is a total individual

 Shannon–Weiner diversity index (H) (Shannon & Weiner, 1963) was calculated as a measure of species diversity

Where H’ is the Shannon–Weiner diversity index, pi is the relative frequency of the species i, and s is the total number of species

 Species evenness (EH) was used as a diversity index to assess how equal the community is numerically (Pielou, 1969)

𝒍𝒏𝑺 Where H = Shannon-Weaver diversity index, H max = ln (S), and S = Total number of observed species in the community

All variables were tested for normality with a Shapiro-Wilk test (α=0.05) and homogeneity of variance with a Levene test (α=0.05) before further analysis If, after testing for normality and homogeneity of variance the abundance of species was not normally distributed, it was then transformed to log(x+1)

Analysis of variance (ANOVA) tests of linear mixed-effects models was used to extract the effects of predictors (fragment size and isolation distance) and their interaction on responses (species abundance and richness) Non-metric multidimensional scaling (NMDS) was performed in PRIMER V7 using Bray – Curtis Similarity Matrix to calculate the community distance of plots among fragments and between fragments showing the relative abundance of species and overall differences in species composition Fragments close to each other are

27 expected to share more similar species compositions, and fragments far from each other would have less similar compositions Analysis of similarity (ANOSIM) was used to identify whether the species composition is similar within and between fragments and isolation distance and test for differences between unordered fragment groups Similarity and dissimilarity percentages (SIMPER) were used to identify the contribution of each functional group of plant species and the average similarity within fragments and average dissimilarity between fragments.

Results

2.4.1 Floristic analysis of the 2015 survey

The 66 plots that were examined within six fragments contained 157 different species in total (Table S2.2, Appendix) The 157 species represent 97 genera and 40 families Nine families (22.5% of the total) accounted for over 65% of species: Fabaceae, with 20 species present from nine genera; Myrtaceae, with 16 species present from eight genera; Proteaceae, with 15 species present from six genera; Asteraceae, with 14 species present from 12 genera; Poaceae, with eight species present from six genera; Ericaceae, with eight species present from four genera; Dilleniacae, with eight species present from one genus; Cyperaceae, with seven species present from five genera, and Asparagaceae, with seven species present from four genera Twenty-one families (52.5% of the total) were represented by a single taxon; Euphorbiaceae, Pittosporaceae, Boryaceae, Xanthorhoeacea, Pteridaceae, Polygalaceae, Restionaceae, Droseraceae, Haloragaceae, Primulaceae, Zamiaceae, Iridaceae, Phyllanthaceae, Amaranthaceae, Santalaceae, Stackhousiaceae, Celastraceae, Elaeocarpaceae, Araliaceae, Rhamnaceae, Lamiaceae, Hypericaceae Seven genera were represented by four or more species: Hibbertia with eight species; Eucalyptus with seven species; both Acacia and Banksia with six species; and Gastrolobium and Stylidium with four species A total of 21 genera

(21.6% of the total number of genera) were represented by one species Five specimens were not identified because the material was inadequate for proper identification

Among 157 species recorded during the survey, 146 species (93 % of the total species) were recorded as native species of the South Western Australia region and 11 species (7% of the total species) were invasive species The highest number of invasive species (5 species) was recorded in the smallest fragment and the smallest number of invasive species (1 species) was recorded in the biggest fragment Generally, the bigger fragments have lower percentage of invasive species (3%) compared to smaller fragments that have higher percentage of invasive species with over 7.4 % of the total Three species were listed as conservation Priority One Threatened Flora (critically endangered) species, one Priority Two species, and one Priority Four species (Table 2.1)

Table 2.1: List of species of conservation priority during the survey at Dryandra Woodland with Conservation Code indicated

Family Species Conservation code Note

Fabaceae Acacia insolita subsp recurva T Threatened Flora

The average number of species per fragment was 60, ranging from 40 species in fragment

E to 74 species in fragment D (Table 2.2) The average number of species per plot was 12 ± 0.8 SE, ranging from two species in fragment E (Transect 1, Plot 3 and Transect 2, Plot 6 and Plot 7) and fragment F (Transect 1, Plot 2 and Plot 3) to 38 species in fragment E (Transect 3, Plot 8) The number of species in these plots of both fragment E and fragment F was the lowest There were only two over-storey Eucalyptus and Santalum species predominated in these plots

29 and no understorey species and thus low species richness (2 species per plot) In contrast, Transect 3, Plot 8 in fragment E and Transect 2, Plot 5 was dominated by Melaleuca over- storey and species of Gastrolobium, Neurachne and Leucopogon under-storey having higher species diversity with 38 and 34 species per plot, respectively

Table 2.2: Summary of the total species, family in each fragment and their size and degree of isolation

Fragment Family Species Size (ha) Isolation (m)

2.4.2 Effect of fragmentation on species diversity and composition

The two-way ANOVA results showed that fragment size and isolation and their interaction had significant effects on the abundance of species The small and isolated fragments had a higher species abundance than did the large and less isolated fragments (Figure 2.2a; Table 2.3) Shannon and Evenness increased with fragment size (Figure 2.2 b, c; Table 2.3), while species richness was not affected by either fragment sizes or isolation distances (Figure 2.2 d; Table 2.3)

Figure 2.2: Effect of fragment size and isolation on the species diversity indices

Table 2.3: Summary of ANOVA test for the effects of fragment sizes and isolation and their interaction on diversity indices

Figure 2.3: Two-dimensional non-metric multidimensional scaling of species in large and small fragments A (blue), B (red) and D (pink) are the large fragments and C, E, F are the small fragments

The nMDS (stress = 0.15) indicated that the large habitat fragments shared more similarity in species composition than did the small fragments (Figure 2.3)

The two-way ANOSIM showed that species composition was significantly different between large and small fragments (p = 0.0001; R = 0.385; Table 2.4) Less isolated and more isolated fragments also harboured significantly different plant communities (p = 0.0002; R 0.262) SIMPER analysis showed that the average dissimilarity between large and small fragments was 90.17 %, and the three plant species Briza maxima, Neurachne alopecuroidea and Tetrarrhena laevis contributed most to the dissimilarity with 16.74 % of the total (Table 2.5) The average dissimilarity between isolated and less isolated fragments was 88.29 %, in which Briza maxima, Tetrerrhena laevis and Neurachne alopecuroidea were the three species that most contributed to the dissimilarity with 7.93 %, 3.92 % and 3.68 %, respectively (Table 2.6)

Table 2.4: Pairwise comparison of the species composition between fragments

A, B, D are the large fragments and C, E, F are the small fragments

Table 2.5: Average dissimilarity between big and small fragments

Table 2.6: Average dissimilarity between isolated and less isolated fragments

Discussion

Generally, the species diversity in the six fragments was not as diverse as recorded in the report of the Department of Environment and Conservation, which showed more than 800 native plant species in the Dryandra Woodland area Conservation Commission, 2011) I recorded only 157 plants recorded during the vegetation survey, and many of them were found in low abundance, with one or fewer individuals per 100 m 2 plots Many surveyed plots contained only woody species such as Eucalyptus wandoo with few or no understorey shrub species

2.5.1 Effects of fragment size and degree of isolation on species diversity

In their island biogeography theory, MacArthur and Wilson (1967) stated that larger fragments contain higher numbers of species than do smaller fragments Over the last 50 years, many studies have indeed reported that smaller fragments had fewer species than larger fragments (Benítez‐Malvido & Martínez‐Ramos, 2003; Hill & Curran, 2001; Iida & Nakashizuka, 1995; Lienert & Fischer, 2003; Metzger, 2000; Nilsson & Nilsson, 1982; Ross et al., 2002; Tabarelli et al., 1999; Zhu et al., 2004) Such a pattern has even been observed in mobile species such as butterflies and birds (ệckinger et al., 2009; ệckinger et al., 2010; Watson et al., 2005) Equally, many studies showed no effects of fragment size on the species richness (Echeverría et al., 2007; Ganivet et al., 2020; Meave & Kellman, 1994; Ochoa-Gaona et al., 2004; Weaver & Kellman, 1981) The results of my study have indicated that both fragment size and degree of isolation and their interaction did not affect the species richness among the six research fragments, though a low richness of plant species (between two and four species per 100 m 2 plot) was recorded in plots from small fragments

There are some possible explanations for the result of this study Firstly, small fragments may be more prone to edge effect The surrounding matrix could alter the microclimatic conditions such as moisture, wind speed, temperature, microbial activity (Matlack, 1993; Weathers et al., 2001), which could have an impact on both plant and animal species

41 composition in the fragments (Lahti, 2001; Valladares et al., 2006), and reduce the population size near the edges and increase extinction risk of local species (Tallmon et al., 2003) Studies that have considered the effect of small fragments ranging from under 1 ha (Metzger, 2000; Ross et al., 2002) to under 14 ha (Tabarelli et al., 1999; Watson et al., 2005) have shown the negative effect of small fragment size on species diversity However, the size of the small fragments in this study, ranging from 54.05 ha to 192 ha, are considerably different – bigger - compared to those in previous studies (e.g (Metzger, 2000; Ross et al., 2002; Tabarelli et al., 1999; Watson et al., 2005) Furthermore, all surveyed plots were located 100m from the edge Therefore, the edge effects on the species recorded might be minimised, and species diversity therefore maintained in the small fragments

Secondly, the forest disturbance caused by human activities causes direct threat to species diversity in this study as it has resulted in decreases tree density, basal area contribution and natural regeneration of species (Bhuyan et al., 2003; Gogoi & Sahoo, 2018; Rawal et al., 2012; Sagar et al., 2003) In particular, species richness will be more affected when anthropogenic disturbance interacts with the fragmentation (Ross et al., 2002) The small fragments in this study were usually located in the areas surrounded by the agricultural matrix and within the care of the landowners, so these fragments, despite being small in size, are less accessible and less affected by anthropogenic disturbance In contrast, the large fragments usually contain roads, bushwalk trails, camping sites and are easy to access and therefore they may have been adversely affected by human activities The anthropogenic disturbances might also lead to the rearrangement of native species and increase floristic homogenisation at local community levels (Dormann et al., 2007; Lôbo et al., 2011) For example, Dos Santos et al (2007) indicated that species similarity among large fragments was higher than the mean similarity in small fragments, which may explain why the large fragments in my study shared more species similarities (Figure 2.4)

Thirdly, the degree of isolation of the selected fragments and the nearest fragments in this study may be not far enough to allow identifiable effects on species diversity The isolation distance was divided into two categories: far (503 m – 1417 m) and near (70.8 m – 265 m), which may still be closer to each other compared to other studies where the fragments were located from 7.3 km to under 100 km far apart (Benítez‐Malvido & Martínez‐Ramos, 2003; Dos Santos et al., 2007; Tabarelli et al., 1999) Furthermore, the studied habitat fragments are located in an area with similar climate conditions and a similar topography and landforms (with elevations ranging from 301 m to 423 m) When fragments are close together, their physical environment may become more uniform (Dos Santos et al., 2007), and there may be less limitation on the exchange of organisms and dispersion process As a result, the species diversity between fragments with different sizes and isolation distances might be similar Finally, time since fragmentation could be a factor affecting species diversity between fragment sizes and isolation distance Dryandra Woodland has catered for different land uses such as recreation, education, nature conservation, and timber production The fragmentation process was recorded from 1950 to 1993, but then remained in the same state until the present (Conservation Commission, 2011) The time since fragmentation occurred in the area may not long enough to impose a significant effect on species diversity compared to other studies in which the fragmentation process occurred more than 100 years prior (Corlett, 1988; Ercelawn et al., 1998; Goldsmith et al., 2011) The fragmentation that occurred over the last 70 years at Dryandra may not have been long enough to have effects on many woody species with a long lifespan; for example, Eucalyptus species can live to over 300 years (McCarthy &

2.5.2 Effect of fragment sizes and isolation distance on species abundance and composition

Although the result of this study indicated that species diversity indices were not significantly related to fragment sizes and isolation distances and their interaction, species abundance was, however, significantly related to fragment sizes and isolation distances While the larger fragments shared more similarities in species composition (Figure 2.4), the abundance of the

Briza maxima, Tetrarrhena laevis, Neurachne alopecuroidea species in small fragments were the main contributors to the significant difference in abundance and composition of species between fragment size and isolation distance These three species contributed over 15% of the dissimilarity between fragment size and isolation distance (Table 2.6; Table 2.7) These species are considered grass-like or herb-like; Tetrarrhena laevis, Neurachne alopecuroidea are native species to Western Australia while Briza maxima is introduced African/European species These three species are distributed mostly in small fragments, especially the alien species Briza maxima, with hundreds of individuals being recorded in some 100 m 2 plots These locally abundant species that are rare in other areas may suggest that the effects of fragmentation could be species-specific (Wilsey et al., 2005) The higher number of alien species recorded in small fragments might be related to edge effects The small fragments have proportionally more edge areas which are vulnerable to plant colonization and occupation by exotic and weedy species from surrounding environments (Bierregaard et al., 1992; Echeverría et al., 2007; Gascon et al., 1999; Janzen, 1986) The changes in microclimatic near the edge could create suitable environments for the establishment and encouragement the spread of invasive species across the small fragment (Echeverría et al., 2007) Furthermore, the alien plants also outperformed native plants in terms of growth and competition in small fragments, thereby causing a decrease in the abundance of certain native species at both local and landscape-scale as the abundance of alien increased (Bernard‐Verdier & Hulme, 2019; Blood, 2001; Gigord et al., 1999; Oshima & Takahashi, 2020) This may explain why these species are widely

44 distributed in small fragments and contributed to the differences in species abundance and composition between fragments size and isolation distances

Some limitations of this study should be considered when the results are interpreted First, the vegetation survey was conducted in a single year that may not have been long enough to provide sufficient data to estimate the relationship between fragment sizes and isolation distances on species diversity, as the fragmentation process and their impact on biodiversity is a long-term process Furthermore, there was no previous data available on the floristic diversity in the six studied fragments, so it was impossible to compare these results with the result of previous surveys to estimate trends of changes in biodiversity However, this study set a benchmark and will provide a reference for future surveys Second, the studied fragments in this experiment were relatively small in number (six fragments) compared to the 24 available fragments in the area The results may not capture the large-scale effect of habitat fragmentation at Dryandra Woodland Third, all sampling plots were set up 100 m from the edge which may lead to the omission of different species in the edge areas, which may not be present within the fragment Finally, besides the two main factors - fragment sizes and isolation distances - considered in this study, other factors, such as fragment shape, soil property and humidity, and isolation history, should also be considered when studying the effect of habitat fragmentation.

Conclusions

The results of this study showed that fragment size and isolation distance did not have a significant influence on the total species diversity at the local scale but they significantly affected the species composition at Dryandra Woodland Furthermore, the study indicated that

45 the abundance of some species, particularly alien species, was significantly related to the change in fragment sizes, isolation distance and anthropogenic disturbance Therefore, if small fragments are properly protected and alien species are adequately managed, the species diversity in both small and large fragments could be maintained Generally, fragment size and isolation distance can still be used as indicators for the study of forest fragmentation over time and space

References

Andren, H (1994) Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review Oikos, 355-366

Authority, E P (2015) Technical guide–terrestrial flora and vegetation surveys for environmental impact assessment

Bélisle, M., Desrochers, A., & Fortin, M.-J (2001) Influence of forest cover on the movements of forest birds: a homing experiment Ecology, 82(7), 1893-1904

Benítez‐Malvido, J., & Martínez‐Ramos, M (2003) Impact of forest fragmentation on understory plant species richness in Amazonia Conservation Biology, 17(2), 389-400 Bernard‐Verdier, M., & Hulme, P E (2019) Alien plants can be associated with a decrease in local and regional native richness even when at low abundance Journal of Ecology,

Best, L B., Bergin, T M., & Freemark, K E (2001) Influence of landscape composition on bird use of rowcrop fields The Journal of Wildlife Management, 442-449

Bhuyan, P., Khan, M., & Tripathi, R (2003) Tree diversity and population structure in undisturbed and human-impacted stands of tropical wet evergreen forest in Arunachal Pradesh, Eastern Himalayas, India Biodiversity and Conservation, 12(8), 1753-1773 Bierregaard , R O., Lovejoy, T E., Kapos, V., dos Santos, A A., & Hutchings, R W (1992)

The biological dynamics of tropical rainforest fragments BioScience, 859-866

Blood, K (2001) Environmental weeds: a field guide for southeast Australia CH Jerram

Burrows, N., McCaw, W., & Maisey, K (1987) Planning for fire management in Dryandra forest In Saunders, Denis A (Ed.) Nature conservation : the role of remnants of native vegetation.Chipping Norton, NSW : Surrey Beatty and Sons

Corlett, R T (1988) Bukit Timah: the history and significance of a small rain-forest reserve

Environmental Conservation, 15(1), 37-44 de Souza, O F F., & Brown, V K (1994) Effects of habitat fragmentation on Amazonian termite communities Journal of Tropical Ecology, 10(2), 197-206

Donaldson, J., Nọnni, I., Zachariades, C., & Kemper, J (2002) Effects of habitat fragmentation on pollinator diversity and plant reproductive success in renosterveld shrublands of South Africa Conservation Biology, 16(5), 1267-1276

Dormann, C F., Schweiger, O., Augenstein, I., Bailey, D., Billeter, R., De Blust, G.,

Herzog, F (2007) Effects of landscape structure and land‐use intensity on similarity of plant and animal communities Global Ecology and Biogeography, 16(6), 774-787 Dos Santos, K., Kinoshita, L S., & dos Santos, F A (2007) Tree species composition and similarity in semideciduous forest fragments of southeastern Brazil Biological Conservation, 135(2), 268-277

Echeverría, C., Newton, A C., Lara, A., Benayas, J M R., & Coomes, D A (2007) Impacts of forest fragmentation on species composition and forest structure in the temperate landscape of southern Chile Global Ecology and Biogeography, 16(4), 426-439 Ercelawn, A C., Afrankie, J V L., Lum, S K., & Lee, S K (1998) Short-term recruitment of trees in a forest fragment in Singapore Tropics, 8(1/2), 105-115

Findlay, C S., & Houlahan, J (1997) Anthropogenic correlates of species richness in southeastern Ontario wetlands Conservation Biology, 11(4), 1000-1009

Franklin, A B., Noon, B R., & George, T L (2002) What is habitat fragmentation? Studies in Avian Biology, 25, 20-29

Ganivet, E., Unggang, J., Bodos, V., Demies, M., Ling, C Y., Sang, J., & Bloomberg, M

(2020) Assessing tree species diversity and structure of mixed dipterocarp forest remnants in a fragmented landscape of north-western Borneo, Sarawak, Malaysia

Gascon, C., Lovejoy, T E., Bierregaard Jr, R O., Malcolm, J R., Stouffer, P C., Vasconcelos,

H L., Borges, S (1999) Matrix habitat and species richness in tropical forest remnants

George, A., Hopkins, A., & Marchant, N (1979) The heathlands of Western Australia In:

Ecosystems of the World Vol.9A Heathlands and related shrublands Alsevier, 61-84

Gibbs, J P (1998) Distribution of woodland amphibians along a forest fragmentation gradient

Gibbs, J P (2001) Demography versus habitat fragmentation as determinants of genetic variation in wild populations Biological Conservation, 100(1), 15-20

Gibbs, J P., & Stanton, E J (2001) Habitat fragmentation and arthropod community change: carrion beetles, phoretic mites, and flies Ecological Applications, 11(1), 79-85 Gibson, N., Keighery, G., & Keigheryº, B (2000) Threatened plant communities of Western

Australia 1 The ironstone Journal of the Royal Society of Western Australia, 83, 1-

Gigord, L., Picot, F., & Shykoff, J A (1999) Effects of habitat fragmentation on Dombeya acutangula (Sterculiaceae), a native tree on La Réunion (Indian Ocean) Biological Conservation, 88(1), 43-51

Gogoi, A., & Sahoo, U K (2018) Impact of anthropogenic disturbance on species diversity and vegetation structure of a lowland tropical rainforest of eastern Himalaya, India

Goldsmith, G R., Comita, L S., & Chua, S C (2011) Evidence for arrested succession within a tropical forest fragment in Singapore Journal of Tropical Ecology, 27(3), 323-326 Griffin, E., Hopkins, A., & Hnatiuk, R (1983) Regional variation in mediterranean-type shrublands near Eneabba, south-western Australia Vegetation, 52(2), 103-127 Gurd, D B., Nudds, T D., & Rivard, D H (2001) Conservation of mammals in eastern North

American wildlife reserves: how small is too small? Conservation Biology, 15(5),

Guthery, F S., Green, M C., Masters, R E., DeMaso, S J., Wilson, H M., & Steubing, F B

(2001) Land cover and bobwhite abundance on Oklahoma farms and ranches The Journal of Wildlife Management, 838-849

Hanski, I., Moilanen, A., & Gyllenberg, M (1996) Minimum viable metapopulation size The

Hanski, I., & Parmesan, C (1999) Metapopulation Ecology Nature, 399(6738), 747

Hargis, C D., Bissonette, J A., & Turner, D L (1999) The influence of forest fragmentation and landscape pattern on American martens Journal of Applied Ecology, 36(1), 157-

Hill, J., & Curran, P (2001) Species composition in fragmented forests: conservation implications of changing forest area Applied Geography, 21(2), 157-174

Hinsley, S., Bellamy, P., Newton, I., & Sparks, T (1995) Habitat and landscape factors influencing the presence of individual breeding bird species in woodland fragments

Hobbs, R J., & Saunders, D A (2012) Reintegrating fragmented landscapes: towards sustainable production and nature conservation: Springer Science and Business Media Hobbs, R J., & Yates, C J (2003) Impacts of ecosystem fragmentation on plant populations: generalising the idiosyncratic Australian Journal of Botany, 51(5), 471-488

Hopper, S D., & Gioia, P (2004) The southwest Australian floristic region: evolution and conservation of a global hot spot of biodiversity Annual Review of Ecology, Evolution, and Systematics, 35, 623-650

Iida, S., & Nakashizuka, T (1995) Forest fragmentation and its effect on species diversity in sub-urban coppice forests in Japan Forest Ecology and Management, 73(1-3), 197-

Janzen, D (1986) The eternal external threat Conservation Biology, 257-285

Keighery, G., & Keighery, B (2012) Vascular flora of Dryandra Woodland (LOL Gray and

Motangue State Forest) Western Australian Naturalists

Komonen, A., Penttilọ, R., Lindgren, M., & Hanski, I (2000) Forest fragmentation truncates a food chain based on an old‐growth forest bracket fungus Oikos, 90(1), 119-126 Kurki, S., Nikula, A., Helle, P., & Lindén, H (2000) Landscape fragmentation and forest composition effects on grouse breeding success in boreal forests Ecology, 81(7), 1985-

Lahti, D (2001) The “edge effect on nest predation” hypothesis after twenty years Biological

Lamont, B., He, T., Enright, N., Krauss, S., & Miller, B (2003) Anthropogenic disturbance promotes hybridization between Banksia species by altering their biology Journal of Evolutionary Biology, 16(4), 551-557

Lande, R (1987) Extinction thresholds in demographic models of territorial populations The

Lienert, J., & Fischer, M (2003) Habitat fragmentation affects the common wetland specialist

Primula farinosa in north‐east Switzerland Journal of Ecology, 91(4), 587-599 Lôbo, D., Leão, T., Melo, F P., Santos, A M., & Tabarelli, M (2011) Forest fragmentation drives Atlantic forest of northeastern Brazil to biotic homogenization Diversity and Distributions, 17(2), 287-296

Lynch, J., & Saunders, D (1991) Responses of bird species to habitat fragmentation in the wheatbelt of Western Australia: interiors, edges and corridors

MacArthur, R H., & Wilson, E O (1967) The theory of island biogeography (Vol 1):

McCarthy, M A., & Lindenmayer, D B (1998) Multi-aged mountain ash forest, wildlife conservation and timber harvesting Forest Ecology and Management, 104(1-3), 43-56 Meave, J., & Kellman, M (1994) Maintenance of rain forest diversity in riparian forests of tropical savannas: implications for species conservation during Pleistocene drought

Metzger, J P (2000) Tree functional group richness and landscape structure in a Brazilian tropical fragmented landscape Ecological Applications, 10(4), 1147-1161

Nilsson, I N., & Nilsson, S G (1982) Turnover of vascular plant species on small islands in

Lake Mửckeln, South Sweden 1976–1980 Oecologia, 53(1), 128-133

Obbens, F., Davis, R., & Sage, L (2001) Vegetation, flora and recommendations for conservation management of Jingaring nature reserve: A" botanical gem" in the Western Australian wheatbelt Journal of the Royal Society of Western Australia, 84,

Ochoa-Gaona, S., González-Espinosa, M., Meave, J A., & Sorani, V (2004) Effect of forest fragmentation on the woody flora of the highlands of Chiapas, Mexico Biodiversity and Conservation, 13(5), 867-884 ệckinger, E., Franzộn, M., Rundlửf, M., & Smith, H G (2009) Mobility-dependent effects on species richness in fragmented landscapes Basic and Applied Ecology, 10(6), 573-578 ệckinger, E., Schweiger, O., Crist, T O., Debinski, D M., Krauss, J., Kuussaari, M.,

Summerville, K S (2010) Life‐history traits predict species responses to habitat area and isolation: a cross‐continental synthesis Ecology Letters, 13(8), 969-979

Oshima, K., & Takahashi, K (2020) Forest disturbances promote invasion of alien herbaceous plants: a comparison of abundance and plant traits between alien and native species in thinned and unthinned stands Biological Invasions, 22(9), 2749-2762

Pielou, E C (1969) An introduction to mathematical ecology New York, USA, Wiley-Inter- science Pither, J., & Taylor, P D (1998) An experimental assessment of landscape connectivity Oikos, 166-174

Rathcke, B J., & Jules, E S (1993) Habitat fragmentation and plant–pollinator interactions

Rawal, R S., Gairola, S., & Dhar, U (2012) Effects of disturbance intensities on vegetation patterns in oak forests of Kumaun, west Himalaya Journal of Mountain Science, 9(2), 157-165

Rose, P W (1993) Production of habitat hollows by wheatbelt Eucalypts: Final report save the bush research grant 1991/92-Project R053: Department of Conservation and Land Management

Ross, K A., Fox, B J., & Fox, M D (2002) Changes to plant species richness in forest fragments: fragment age, disturbance and fire history may be as important as area

Sagar, R., Raghubanshi, A., & Singh, J (2003) Tree species composition, dispersion and diversity along a disturbance gradient in a dry tropical forest region of India Forest ecology and Management, 186(1-3), 61-71

Saunders, D A., Hobbs, R J., & Margules, C R (1991) Biological consequences of ecosystem fragmentation: a review Conservation Biology, 5(1), 18-32

Sáyago, R., Quesada, M., Aguilar, R., Ashworth, L., Lopezaraiza-Mikel, M., & Martén-

Rodríguez, S (2018) Consequences of habitat fragmentation on the reproductive success of two Tillandsia species with contrasting life history strategies AoB Plants, 10(4), ply038

Schmiegelow, F K., & Mửnkkửnen, M (2002) Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest Ecological Applications, 12(2), 375-389

Shannon, C E., & Weiner, W (1963) The mathematical theory of communication Urban

Skole, D., & Tucker, C (1993) Tropical deforestation and habitat fragmentation in the

Amazon: satellite data from 1978 to 1988 Science, 260(5116), 1905-1910

Steffan-Dewenter, I., Münzenberg, U., Bürger, C., Thies, C., & Tscharntke, T (2002) Scale‐ dependent effects of landscape context on three pollinator guilds Ecology, 83(5), 1421-

Tallmon, D A., Jules, E S., Radke, N J., & Mills, L S (2003) Of mice and men and trillium: cascading effects of forest fragmentation Ecological Applications, 13(5), 1193-1203

Taylor, P D., & Merriam, G (1996) Habitat fragmentation and parasitism of a forest damselfly Landscape Ecology, 11(3), 181-189

Valladares, G., Salvo, A., & Cagnolo, L (2006) Habitat fragmentation effects on trophic processes of insect‐plant food webs Conservation Biology, 20(1), 212-217

Van Treuren, R., Bijlsma, R., Van Delden, W., & Ouborg, N (1991) The significance of genetic erosion in the process of extinction I Genetic differentiation in Salvia pratensis and Scabiosa columbaria in relation to population size Heredity, 66(2), 181

Venier, L., & Fahrig, L (1996) Habitat availability causes the species abundance-distribution relationship Oikos, 564-570

Watson, J E., Whittaker, R J., & Freudenberger, D (2005) Bird community responses to habitat fragmentation: how consistent are they across landscapes? Journal of Biogeography, 32(8), 1353-1370

Weaver, M., & Kellman, M (1981) The effects of forest fragmentation on woodlot tree biotas in Southern Ontario Journal of Biogeography, 199-210

Wettstein, W., & Schmid, B (1999) Conservation of arthropod diversity in montane wetlands: effect of altitude, habitat quality and habitat fragmentation on butterflies and grasshoppers Journal of Applied Ecology, 36(3), 363-373

Whelan, R J., Ayre, D J., England, P R., Llorens, T., & Beynon, F (2000) Ecology and genetics of Grevillea (Proteaceae): implications for conservation of fragmented populations Conservation Biology Series-Cambridge 253-270

Wiens, J A (1976) Population responses to patchy environments Annual Review of Ecology and Systematics, 7(1), 81-120

Wilsey, B J., Martin, L M., & Polley, H W (2005) Predicting plant extinction based on species‐area curves in prairie fragments with high beta richness Conservation Biology,

Wilson, B., Valentine, L., Kinloch, J., Sonneman, T., & Swinburn, M (2009) Habitat loss and fragmentation In: Barbara A Wilson and Leonie E Valentine(Ed.) Biodiversity values and threatening processes of the Gnangara Department of Environment and Conservation.With, K A., & Crist, T O (1995) Critical thresholds in species' responses to landscape structure Ecology, 76(8), 2446-2459

Wolff, J O., Schauber, E M., & Edge, W D (1997) Effects of habitat loss and fragmentation on the behavior and demography of gray‐tailed voles: Efectos de la pérdida y fragmentación de hábitat sobre el comportamiento y la demografía de ratones de campo de cola gris Conservation Biology, 11(4), 945-956

Yates, C J., & Ladd, P G (2005) Relative importance of reproductive biology and establishment ecology for persistence of a rare shrub in a fragmented landscape

Young, A., Boyle, T., & Brown, T (1996) The population genetic consequences of habitat fragmentation for plants Trends in Ecology and Evolution, 11(10), 413-418

Young, A., Merriam, H., & Warwick, S (1993) The effects of forest fragmentation on genetic variation in Acer saccharum Marsh.(sugar maple) populations Heredity, 71(3), 277

Zhu, H., Xu, Z., Wang, H., & Li, B (2004) Tropical rain forest fragmentation and its ecological and species diversity changes in southern Yunnan Biodiversity and Conservation, 13(7), 1355-1372

Zuidema, P A., Sayer, J A., & Dijkman, W (1996) Forest fragmentation and biodiversity: the case for intermediate-sized conservation areas Environmental Conservation, 23(4), 290-297

Effect of population size on plant growth in fragmented Dryandra Woodland

Abstract

Forest fragmentation increase isolation distance and decrease the size of the plant population among plant communities in fragments, acting as an important ecological filter for plant reproduction and growth In addition, small and isolated habitat fragments may experience losses in genetic variation within species In turn, this can cause lower rates of fruit and seed set, seed germination and lower biomass in recruits of affected species in fragments Here I examine seedling (recruits) growth performance in response to fragmentation in Banksia sessilis, Gastrolobium parviflorum and Bossiaea eriocarpa These are the foundation species of Dryandra Woodland, Western Australia Seeds of these species were collected from six fragments with sizes ranging from 54 to 1262 ha and an isolation distance from 70 m to 1417 m, over two consecutive years in 2016 and 2017 Seeds were then germinated in germination cabinets for a month Seedlings were moved into the greenhouse and raised there for a year under common conditions, and seedling performance was monitored Fragmentation, isolation, and their interaction effects on the growth of seedlings were detected but varied across sites by fragment size and isolation distance with higher germination rates and growth rates in larger and more isolated fragments in general

Introduction

Well-established consequences of anthropogenic habitat fragmentation are the loss of habitat, small population sizes and an increase in the isolation of remaining populations of the affected taxa (Ashworth & Martí, 2011; Fahrig, 2003; McGarigal & Cushman, 2002) Remaining habitat fragments have more edge habitat and larger edge-to-core habitat ratios Small habitat fragments are also more vulnerable to the physical effects of faster wind speed, lower humidity, higher temperature, and changes to soil nutrients These physical conditions synergistically alter important biological processes such as pollinator behaviour, species interaction and population structure of plant species in small fragments These changes increase physiological stresses on plant species near habitat edges (Chen et al., 1992; Fischer & Lindenmayer, 2007; Matlack, 1993; Weathers et al., 2001), which have the potential to reduce reproductive rates, lower flower set and fertility, as well as reduce fruit development and increase abortion rates (Aizen & Feinsinger, 1994)

Isolated remnant fragments may also experience lower pollinator pressure, reducing gene flow between populations (Aizen & Feinsinger, 1994; Roque et al., 2016; Young et al., 1996) Low levels of gene flow among plants in habitat fragments are known to cause inbreeding depression (Lande & Barrowclough, 1987; Mills & Smouse, 1994) Inbreeding depression further decreases genetic variability, which affects the ability of isolated populations to respond to environmental stressors (Frankham, 2005), ultimately reducing the fitness of subsequent generations (Lienert, 2004) Thus, habitat fragmentation reduces the reproductive success and regeneration of old-growth species in fragments (Aizen & Feinsinger, 1994; Rathcke & Jules,

1993) and increases the risk of pathogen infection (Lienert, 2004) The latter fragmentation effects increase the likelihood of species extinction in fragments (Frankham, 2005; Menges,

It has been widely documented that habitat fragmentation has negative effects on the fitness and diversity of plant species (Aguilar et al., 2006; Ashworth & Martí, 2011; Gibbs, 2001; Kurki et al., 2000) Individuals in small populations in isolated fragments are much less likely to mate with individuals in other fragments, thereby increasing the likelihood of self- pollination or mating within the isolated population (Aguilar et al., 2008; Jong et al., 1993; Ellstrand & Elam, 1993; Lienert, 2004; Rathcke & Jules, 1993) Thus, populations in fragments may become inbred Inbreeding depression reduces species fitness as expressed through measures of reproduction, such as fruit set, seed set and germination rate (Lienert, 2004; Menges, 1991) Inbred plants produce fewer flowers and fruits and set fewer seed sets with lower germination rates than outbred plants (Aguilar et al., 2006; Aizen & Feinsinger, 1994; Ashworth & Martí, 2011; Lienert, 2004; Mathiasen et al., 2007) Inbreeding depression in small fragmented populations further affects the fitness of future generations through lower growth rates, reflected as smaller leaf-size and root length, shorter plant height and lower plant biomass than is observed in populations in contiguous habitats (Hooftman et al., 2003; Kery et al., 2000; Lienert, 2004; Lienert et al., 2002)

Although many studies have evaluated the influence of habitat fragmentation on species richness and reproductive success of plants in temperate, subtropical and tropical regions (Cascante et al., 2002), only a few studies have focused on Mediterranean-climate regions Furthermore, most plant studies have looked at how population size affects offspring fitness (Ashworth & Martí, 2011; Le Cadre et al., 2008; Leimu et al., 2006; Menges, 1991) However, offspring fitness of plant species in fragments is affected by numerous factors, including fragment size, isolation, and edge effects, all of which need to be considered

In this chapter, I aim to determine whether Dryandra woodland fragment size and isolation distance in a Mediterranean-climate region affect plant species growth, including

60 germination rate, plant height, number of leaves, root length, fresh biomass, dry biomass, and root-to-shoot ratio I hypothesised that artificial habitat fragmentation negatively impacts population vigour, resulting in plant seedlings from small and isolated fragments growing slower than those in large and less isolated fragments.

Methods

Six study sites were selected and are described in Chapter 1 and Chapter 2 These sites were Dryandra Woodland fragments created since the 1950s by agricultural development in the region The fragments are all in the same biogeographic region with a Mediterranean climate - they range in size from 54 to 1262 ha and the isolation distance from 0.0708 km to 1.417 km (Figure 2.1)

Three species (30 individuals per species), Banksia sessilis, Gastrolobium parviflorum, and

Bossiaea eriocarpa, were selected, as they are plants native to Western Australia, and are common foundation species in all the selected fragments These three species are present in relatively high densities, which can provide an adequate sample size and are essential as a food source for birds and animals They differ in growth form, are from different taxonomic lineages, and have different life history characteristics (Table 3.1) Fruits were collected from December 2016 to March 2017 from 30 individual plants of each species that were similar in size and at least 5 m distant from the nearest conspecific individual

For the germination trials, I used 100 to 600 seeds per fragment, depending on the number of seeds collected per species For B sessilis, 100 undamaged and healthy seeds per fragment (10 seeds per maternal plant) were collected and sterilised for 15 minutes in a solution of 20 sodium hypochlorite and a drop of Tween 80 surfactant (polyoxyethylene sorbitan mono-oleate) After that, seeds were washed thoroughly three times using sterile deionised water (Downes et al

2010) After sterilising, the seeds were put in Petri dishes (150 mm diameter) that contain two layers of filter paper (Whatman No.1), five pieces of 2 cm 2 portions of Wettex sponge and 16 mL of sterile deionised water These Petri dishes were covered in Parafilm and put in a germination cabinet at 15 ° C with a daily cycle of 12 hours of light and 12 hours of darkness a month The germination rate was monitored and recorded every two days for four weeks

For the two remaining species, both of which have hard seed coats, after implementing the same sterilisation process as with B sessilis, the seed coat of G parviflorum and B eriocarpa was lightly scarified using sandpaper The seeds then were placed in 90 mm Petri dishes (depending on the seed size) that contain two layers of filter paper (Whatman No.1), four pieces of 2 cm 2 portions of Wettex sponge and 10 mL of sterile deionised water Germination rate was monitored in the same way as for B sessilis

Figure 3.1: Study species A: Banksia sessilis , B: Bossiaea eriocarpa, C: Gastrolobium parviflorum

Table 3.1: Summary of the selected species

Species Family Life form Flower Pollination Seed Leaves

Banksia sessilis Proteaceae Woody shrub or small tree 0.5- 4 m

Insects Dark round seed and hard seed coat

Bossiaea eriocarpa Fabaceae Small non- woody shrub 0.2 – 1 m

Insects Bare seeds Hairy narrow- oblong leaves, 10-27 mm long 1.5-4 mm width

After the germinating period, 30 seedlings of each species per fragment were transferred to a greenhouse and transplanted into PVC tube pots (5 cm diameter, 70 cm height, one seedling per pot) containing a mixture of 50% washed white sand and 50% organic fertiliser (without phosphorous P) All seedlings were planted in June 2017: starting from 7 June 2017 for B sessilis, 12 June for G parviflorum and 15 June for B eriocarpa The seedlings were watered

100 ml in one minute daily in the first ten days to establish them and then every two days for a year after that

After a year of growth in the greenhouse, all plants were harvested over four days (2 - 6 June 2018) Growth measures were: number of leaves of the plant, plant height, root length, total fresh biomass, fresh shoot biomass, fresh root biomass Plants were oven-dried at 70 ° C for 48 hours, after which total dry biomass and root: shoot ratio (dry root mass/dry shoot mass) were recorded Moisture content was derived from fresh versus dry tissue mass

All statistics were performed using the statistical software R (R core team, 2018) and PAST (version 3.21) Variables were tested for normality with a Shapiro-Wilk test homogeneity of variance with a Levene test To satisfy the assumption of normality, non-normally distributed growth variables for B sessilis and B eriocarpa were log(x+1) transformed, while variables for G parviflorum were square-root transformed The germination rate and mortality rate of these three species were arcsine transformed

Two-way ANOVA was performed on the following measurements: germination rate, number of leaves, root length, shoot length, fresh root biomass, shoot fresh biomass, total fresh biomass, root dry biomass, shoot dry biomass, total dry biomass, water content, root: shoot

65 ratio The analyses aimed to test for differences between fragments in (1) germination rate, with the prediction of lower rates from smaller fragments; (2) seedling mortality, with higher mortality rates predicted in smaller and more isolated fragments; and (3) whether there is isolation and fragment size effect on the seedling growth rates Post-hoc Tukey tests were used to evaluate differences among fragment size and isolation in the variables Significance was taken at p < 0.05.

Results

Generally, only isolation distance affected the germination rate of B eriocarpa, and seeds from more isolated fragments had higher germination rates than those from less isolated fragments (Figure 3.2b) There was no interaction between fragment size and isolation on the germination rate of B eriocarpa (Table 3.2) By contrast, fragment size and isolation and their interaction significantly influenced the germination rate of G parviflorum The lowest germination rates were recorded from the small and more isolated fragments (Table 3.2; Figure 3.2c) while only isolation and the interaction fragment size and isolation significantly affected the germination rate of B sessilis (Table 3.2)

Table 3.2: The effect of fragment size and isolation on germination rate of B sessilis G parviflorum and B eriocarpa

B sessilis G parviflorum B eriocarpa df F p df F p df F p

Small and less isolated fragments particularly negatively impacted the germination rate of B sessilis and G parviflorum Less isolated fragments also negatively affect the germination rate of B eriocarpa (Table 3.3)

Table 3.3: The adverse effects of small and less isolated fragments on the germination rate of

B sessilis G parviflorum B eriocarpa df t p df t p df t p

The B sessilis seeds from large (>200 ha) isolated fragments had higher germination rates at 94.6 % (± 0.7 SE) than did seeds from isolated small fragments: 88.6 % (± 3.2 SE) Germination rates of B sessilis from less isolated fragments were higher in small than in large fragments: 94.3 % (± 1.4 SE) and 78.7 % (± 4.3 SE), respectively (Figure 3.2a)

The germination rate trends of G parviflorum were similar to B sessilis with complete

(100%) germination from larger isolated fragments than from small and more isolated fragments (95% ± 1.9 SE) In less isolated fragments, the germination rate was lower in small than large fragments, ranging from 73.78 % (± 3.08 SE) in small fragments to 98.95 % (± 1.04 SE) in large fragments (Figure 3.2c) In contrast, the germination rate of B eriocarpa seed declined with fragment size and isolation, ranging from 15.6% (± 1.78 SE) for small and less isolated fragments to 19.5 % (± 1.5 SE) for small and isolated fragments, compared to those from big and less isolated fragments and big and isolated fragments with 9.3 % (± 1.4 SE) and 15.3 % (± 2.9 SE), respectively (Figure 3.2 b)

Figure 3.2: Mean germination rates (%) of (a) B sessilis, (b) B eriocarpa, and (c) G parviflorum Data presented as mean ± SD

Banksia sessilis seedlings had the lowest mortality rate, with no seedlings lost from the six fragments In Gastrolobium parviflorum, 8.33% (5 out of 60) of seedlings from big and isolated fragments perished In small and isolated, and big and less isolated fragments, 26.67% (8 out of 30 seedlings; Figure 3.3) of seedlings died

Figure 3.3: Mortality of G parviflorum seedlings in relation to fragment size and isolation

Bossiaea eriocarpa species had a high mortality rate overall (73.9%) Seedlings from small and isolated fragments have the highest mortality rate (83.3%; 25 of 30 seedlings) Seedlings from both large and isolated fragments and large and less isolated fragments had the lowest mortality rate of this species (70%; 42 of 60 seedlings; Figure 3.4)

Figure 3.4: Proportion of perishing seedlings of B eriocarpa in six collection fragments

Fragment size significantly affected only four growth indices of 11 measured for B sessilis seedlings: the number of leaves, total dry biomass, shoot dry biomass, and root-to-shoot ratio (Figure 3.5a-d; p 0.05

Fragment size did not have a detectable effect on the growth development of Gastrolobium parviflorum In contrast, isolation had a significant impact on nearly all growth variables of Gastrolobium parviflorum (Figure 3.7 a-g; p0.05 in all cases) An interaction between fragment size and isolation had the same effect on growth variables of

Gastrolobium parviflorum as the isolation did Generally, almost all growth variables increased proportionally with isolation except the number of leaves, plant height, root length and root to shoot ratio

Figure 3.6 a-d: The effect of fragment size and isolation and their interaction on the number of leaves, plant height and root length of Gastrolobium parviflorum

Figure 3.7 a-f: The effect of fragment size and isolation on fresh biomass, shoot fresh biomass, root fresh biomass, total dry biomass, shoot dry biomass, root dry biomass and water content of Gastrolobium parviflorum Values sharing the same letters are not significantly different at p > 0.05

Figure 3.8 a-b: Effect of fragment size and isolation on the height and water content of

Bossiaea eriocarpa Data presented as means ± SE

There was no significant interaction between fragment size and isolation on the growth of B eriocarpa seedlings (Two-way ANOVAs, p>0.05 in all cases, Table 3.4) Only plant height was affected by isolation (Figure 3.8 a; p< 0.05) and water content was influenced by both fragment size and isolation (Figure 3.8 b; p< 0.01 in all cases), with seedlings in the less isolated fragments displaying higher height while seedlings in the small and less isolated fragment having more water content In general, fragment size and isolation and their interaction seem not to affect growth variables of Bossiaea eriocarpa except the water content

Table 3.4: The effect of fragment size and isolation and their interaction on the growth indices of Bossiaea eriocarpa Significant values are marked with an asterisk

Growth variables Size Isolation Size × Isolation p p p

Discussion

Overall, fragment size and isolation and their interaction had certain effects on the growth development of these species However, these two factors had different effects on the different species Isolation was shown to have more impact on all growth development including germination, mortality and growth rate of these three species compared to fragment size, which had only a few effects on the growth rates and germination of Banksia sessilis Species in larger and more isolated fragments displayed higher germination rates, higher growth rates and lower mortality rate

3.5.2 The effect of fragmentation on seed germination

The germination rates of B sessilis and B eriocarpa were significantly affected by the degree of isolation, displaying higher germination rates in more isolated fragments than in less isolated fragments Both fragment size and level of isolation, and their interactions, had significant impacted on the percentage of seed germination of G parviflorum, where big and isolated fragments had a higher germination rate than did the small and isolated fragments

All the selected model species were affected either by fragment size or isolation and their interaction This result concurs with the findings of other authors who have conducted comparisons of seed germination of Croton lachnostachyus, Rivina humilis, Aristotelia chilensis and Heliconia acuminata between large and continuous habitat fragments and small fragments, which have indicated that seeds germination rates from big and continuous fragments are at least two times higher than seeds from small fragments (Ashworth & Martí, 2011; Bruna, 2002; Valdivia & Simonetti, 2006)

3.5.3 The effect of fragmentation on seedling survival

Although B sessilis is killed by fire (Lamont et al., 1998) and by the dieback pathogen Phytophthora cinnamomi Rands, it is seen as an aggressive coloniser species that is widespread in the open forest throughout south-west Western Australia (Rockel et al., 1982) Meanwhile, the high survival rate of seedlings and their adaptability to the environment may contribute to the widespread of the species However, there was no statistically significant difference in survival rate between large and small fragments The mortality rates of G parviflorum ranged from 6.67 % to 26.67 % but the average mortality rate of this species was similar in both the small fragments and the large fragments with 17.50 % and 18.34 %, respectively There was a high mortality rate of B eriocarpa seedlings, ranging from 66.67 % to 83.33 %, in which seedlings from small fragments had a higher rate of mortality with an average of 79.16 % compared to those from large fragments with an average of 70%

3.5.4 The effect of fragmentation on plant growth

Fragment size and degree of isolation and the interaction between these two factors appear to differently influence the growth development of B sessilis and G parviflorum Fragment size did not have any effect on the growth development of G parviflorum and it affected only a few growth variables of B sessilis with higher total dry biomass and higher total shoot dry biomass of seedlings from small fragments In contrast, isolation had significant effects on both B sessilis and G parviflorum growth rate, which both displayed higher total fresh biomass, shoot fresh biomass, root fresh biomass, total dry biomass, shoot dry biomass and root dry biomass from more isolated fragments Of the three species, only growth development indices of B eriocarpa were not affected by either fragment size or degree of isolation

Species diversity may be negatively impacted by habitat fragmentation because it reduces gene flow (Aizen & Feinsinger, 1994) and hence genetic diversity (Honnay & Jacquemyn,

2007) However, the B sessilis growth rate recorded in this study was highest in the small and

79 isolated fragments It could be argued that although the fragment size is small, the population of B sessilis is still large enough to maintain the genetic diversity of this species This agrees with a previous study on several proliferating taxa which become hyper abundant in small fragments but have low abundance in mature forest regions of the Coimbra forest of the Brazilian north-eastern Atlantic forest (Santo-Silva et al., 2016) Furthermore, B sessilis is mainly pollinated by highly mobile birds, which capable of travelling over great distances between fragments and hence B sessilis likely has a bigger mating pool than species that are pollinated by short-distance travelling insects and small animals (Hopper & Moran, 1981; Murcia, 1996) As a result, the population of this species may have high outcrossing rates that can translate into enhanced progeny genetic diversity (González‐Varo et al., 2010)

In contrast, the seedlings of G parviflorum had a higher growth rate in large and isolated fragments Seedlings from large populations may have higher fitness levels as they may have fewer effects of inbreeding depression than those from small populations (Honnay & Jacquemyn, 2007) This result is congruent with other studies that reported that the population size of certain plant species strongly correlates with offspring fitness (Fischer & Matthies, 1998; Oostermeijer et al., 1994)

Not all species are impacted negatively by fragment size or isolation, and B eriocarpa in this study was one of those cases The seedling growth rate of this species was not affected by either fragment size or isolation This result concurs with the research on a gypsum endemic plant Centaurea hyssopifolia in which neither the measured functional traits nor their plasticity was affected by fragment size or connectivity: the size, growth rate, and leaf characteristics were comparable from all populations, (Matesanz et al., 2017)

3.5.5 Limitation of the study and directions for future study

Habitat fragmentation is a complicated process involving multiple factors simultaneously (Aguilar et al., 2008), analysing the fragment size and isolation may not, therefore, fully

80 represent all the contributing variables For example, edge effects which modify the physical variables like radiation, humidity, wind speed and soil nutrients (Chen & Franklin, 1990; Matlack, 1993; Weathers et al., 2001) and the biotic variables such as plant and animal species composition or pattern of competition and predation and parasitism (Benítez‐Malvido & Martínez‐Ramos, 2003; Lahti, 2001; Malcolm, 1994; Robinson et al., 1995; Valladares et al.,

2006) in fragments are seen to have negative impacts as they reduce the survival rate of trees such as Cecropia, Pourouma, Bellucia, and Vismia species (Nascimento & Laurance, 2004), and contribute to increased mortality in large trees (Laurance, 2000) These factors should also be considered in future studies on the fragmentation in this area This study focused on the effects of fragmentation on the growth development of the three focal shrub species, and the experiment was only carried out for one year To have a complete overview of fragmentation on plant species growth development, the investigation should ideally be taken continuously over several years with more species and various kinds of growth forms such as big trees, herbs, and grasses because species with various growth forms may respond to forest fragmentation differently Tropical trees such as Samanea saman are susceptible to forest fragmentation as seedlings from isolated fragmentation had smaller leaf areas and lower biomass than those from continuous populations (Cascante et al., 2002), forest fragmentation may also influence the forest’s regenerative capacity as it reduces the establishment of seedlings and increases seedlings mortality within fragments (Benitez‐Malvido, 1998) Fragmentation also reduces herbaceous seedling fitness of Phyteuma spicatum (Kolb, 2008) and Gentiana pneumonanthe in small populations (Oostermeijer et al., 1994) in a temperate climate Thus, it is necessary to conduct future research focusing on the effects of forest fragmentation on trees and herbaceous species in the Mediterranean climate to see whether species in these areas respond similarly or differently to those from tropical and temperate climates

The high proportion of seedlings that perished during the study also limited the final sample size, resulting in a sample size lower than planned to test treatment effects For future studies, larger sample sizes are suggested to increase confidence in determining whether a difference exists between the sizes of fragments and the degree of isolation on species growth development.

Conclusion

This study indicated that fragmentation and isolation of populations had certain effects on the seedlings of some species, but they responded to the fragmentation differently While B sessilis seedlings could develop well in small and isolated fragmentations, the seedlings of G parviflorum, in contrast, were affected by fragmentation as their growth indices were found to be higher in big fragments Bossiaea eriocarpa was the only species that was not affected by either fragment size or degree of isolation However, due to the small sample size of this species that survived (25%) for assessment, the result should be viewed with caution

References

Aguilar, R., Ashworth, L., Galetto, L., & Aizen, M A (2006) Plant reproductive susceptibility to habitat fragmentation: review and synthesis through a meta‐analysis Ecology Letters, 9(8), 968-980

Aguilar, R., Quesada, M., Ashworth, L., Herrerias‐Diego, Y., & Lobo, J (2008) Genetic consequences of habitat fragmentation in plant populations: susceptible signals in plant traits and methodological approaches Molecular Ecology, 17(24), 5177-5188

Aizen, M A., & Feinsinger, P (1994) Forest fragmentation, pollination, and plant reproduction in a Chaco dry forest, Argentina Ecology, 75(2), 330-351

Ashworth, L., & Martí, M L (2011) Forest fragmentation and seed germination of native species from the Chaco Serrano Forest Biotropica, 43(4), 496-503

Barrios Roque, B., Koptur, S., & Sah, J P (2016) The effects of habitat fragmentation on the reproduction and abundance of Angadenia berteroi Journal of Plant Ecology, 10(2), 340-348

Benitez‐Malvido, J (1998) Impact of forest fragmentation on seedling abundance in a tropical rain forest Conservation Biology, 12(2), 380-389

Benítez‐Malvido, J., & Martínez‐Ramos, M (2003) Impact of forest fragmentation on understory plant species richness in Amazonia Conservation Biology, 17(2), 389-400

Bruna, E M (2002) Effects of forest fragmentation on Heliconia acuminata seedling recruitment in central Amazonia Oecologia, 132(2), 235-243

Cascante, A., Quesada, M., Lobo, J J., & Fuchs, E A (2002) Effects of dry tropical forest fragmentation on the reproductive success and genetic structure of the tree Samanea saman Conservation Biology, 16(1), 137-147

Chen, J., & Franklin, J F (1990) Microclimatic pattern and basic biological responses at the clear-cut edges of old-growth Douglas-fir stands Northwest Environmental Journal,

Chen, J., Franklin, J F., & Spies, T A (1992) Vegetation responses to edge environments in old‐growth Douglas‐fir forests Ecological Applications, 2(4), 387-396 de Jong, T J., Waser, N M., & Klinkhamer, P G (1993) Geitonogamy: the neglected side of selfing Trends in Ecology and Evolution, 8(9), 321-325

Downes, K S., Lamont, B B., Light, M E., & van Staden, J (2010) The fire ephemeral

Tersonia cyathiflora (Gyrostemonaceae) germinates in response to smoke but not the butenolide 3-methyl-2 H-furo [2, 3-c] pyran-2-one Annals of Botany, 106(2), 381-384

Ellstrand, N C., & Elam, D R (1993) Population genetic consequences of small population size: implications for plant conservation Annual Review of Ecology and Systematics,

Fischer, M., & Matthies, D (1998) RAPD variation in relation to population size and plant fitness in the rare Gentianella germanica (Gentianaceae) American Journal of Botany,

Frankham, R (2005) Genetics and extinction Biological Conservation, 126(2), 131-140 Gibbs, J P (2001) Demography versus habitat fragmentation as determinants of genetic variation in wild populations Biological Conservation, 100(1), 15-20

González‐Varo, J P., Albaladejo, R G., Aparicio, A., & Arroyo, J (2010) Linking genetic diversity, mating patterns and progeny performance in fragmented populations of a Mediterranean shrub Journal of Applied Ecology, 47(6), 1242-1252

Honnay, O., & Jacquemyn, H (2007) Susceptibility of common and rare plant species to the genetic consequences of habitat fragmentation Conservation Biology, 21(3), 823-831

Hooftman, D A., Van Kleunen, M., & Diemer, M (2003) Effects of habitat fragmentation on the fitness of two common wetland species, Carex davalliana and Succisa pratensis Oecologia, 134(3), 350-359

Hopper, S., & Moran, G (1981) Bird pollination and the mating system of Eucalyptus stoatei

Kery, M., Matthies, D., & Spillmann, H H (2000) Reduced fecundity and offspring performance in small populations of the declining grassland plants Primula veris and

Gentiana lutea Journal of Ecology, 88(1), 17-30

Kolb, A (2008) Habitat fragmentation reduces plant fitness by disturbing pollination and modifying response to herbivory Biological Conservation, 141(10), 2540-2549

Kurki, S., Nikula, A., Helle, P., & Linden, H (2000) Landscape fragmentation and forest composition effects on grouse breeding success in boreal forests Ecology, 81(7), 1985-

Lahti, D (2001) The “edge effect on nest predation” hypothesis after twenty years Biological

Lamont, B B., Olesen, J M., & Briffa, P J (1998) Seed production, pollinator attractants and breeding system in relation to fire response—are there reproductive syndromes among co-occurring proteaceous shrubs? Australian Journal of Botany, 46(4), 377-385

Lande, R., & Barrowclough, G F (1987) Effective population size, genetic variation, and their use in population management Viable Populations for Conservation, 87, 124

Laurance, W F (2000) Do edge effects occur over large spatial scales? Trends in Ecology and

Le Cadre, S., Tully, T., Mazer, S J., Ferdy, J B., Moret, J., & Machon, N (2008) Allee effects within small populations of Aconitum napellus ssp lusitanicum, a protected subspecies in northern France New Phytologist, 179(4), 1171-1182

Leimu, R., Mutikainen, P., Koricheva, J., & Fischer, M (2006) How general are positive relationships between plant population size, fitness and genetic variation? Journal of

Lienert, J (2004) Habitat fragmentation effects on fitness of plant populations–a review

Lienert, J., Diemer, M., & Schmid, B (2002) Effects of habitat fragmentation on population structure and fitness components of the wetland specialist Swertia perennis

L.(Gentianaceae) Basic and Applied Ecology, 3(2), 101-114

Malcolm, J R (1994) Edge effects in central Amazonian forest fragments Ecology, 75(8),

Matesanz, S., Rubio Teso, M L., García-Fernández, A., & Escudero, A (2017) Habitat fragmentation differentially affects genetic variation, phenotypic plasticity and survival in populations of a gypsum endemic Frontiers in Plant Science, 8, 843

Mathiasen, P., Rovere, A E., & Premoli, A C (2007) Genetic structure and early effects of inbreeding in fragmented temperate forests of a self‐incompatible tree, Embothrium coccineum Conservation Biology, 21(1), 232-240

McGarigal, K., & Cushman, S A (2002) Comparative evaluation of experimental approaches to the study of habitat fragmentation effects Ecological Applications, 12(2), 335-345

Menges, E S (1991) Seed germination percentage increases with population size in a fragmented prairie species Conservation Biology, 5(2), 158-164

Mills, L S., & Smouse, P E (1994) Demographic consequences of inbreeding in remnant populations The American Naturalist, 144(3), 412-431

Murcia, C (1996) Forest fragmentation and the pollination of neotropical plants Forest

Nascimento, H E., & Laurance, W F (2004) Biomass dynamics in Amazonian forest fragments Ecological Applications, 14(sp4), 127-138

Oostermeijer, J., Van Eijck, M., & Den Nijs, J (1994) Offspring fitness in relation to population size and genetic variation in the rare perennial plant species Gentiana pneumonanthe (Gentianaceae) Oecologia, 97(3), 289-296

Robinson, S K., Thompson, F R., Donovan, T M., Whitehead, D R., & Faaborg, J (1995)

Regional forest fragmentation and the nesting success of migratory birds Science, 267(5206), 1987-1990

Rockel, B., McGann, L., & Murray, D I (1982) Phytophthora cinnamomi causing death of

Dryandra sessilis on old dieback sites in the jarrah forest Australasian Plant Pathology, 11(4), 49-50

Santo-Silva, E E., Almeida, W R., Tabarelli, M., & Peres, C A (2016) Habitat fragmentation and the future structure of tree assemblages in a fragmented Atlantic forest landscape

Valdivia, C E., & Simonetti, J A (2006) Decreased frugivory and seed germination rate do not reduce seedling recruitment rates of Aristotelia chilensis in a fragmented forest Plant Conservation and Biodiversity (pp 19-28): Springer

Valladares, G., Salvo, A., & Cagnolo, L (2006) Habitat fragmentation effects on trophic processes of insect‐plant food webs Conservation Biology, 20(1), 212-217

Effect of anthropogenic habitat fragmentation on plant reproduction

Abstract

Anthropogenic habitat fragmentation is believed to generally have adverse effects on plant species’ reproduction as it can reduce the diversity of pollinators in fragments and hence interrupt plant-pollinator interaction and lead to a reduction in species reproduction This study assessed the effect of anthropogenic habitat fragmentation on plant-pollinator interactions in

Banksia sessilis (Proteaceae) and the reproduction of B sessilis, Gastrolobium parviflorum

(Fabaceae) and Bossiaea eriocarpa (Fabaceae) in Dryandra Woodland, Western Australia Honeyeaters (Meliphagidae) were the main pollinators visiting B sessilis across all fragments The highest pollinator visit rate was recorded in the morning The visit rates and bird species’ composition were not significantly different between fragments with different sizes and isolation distances Similarly, the reproduction of B sessilis was not affected by fragmentation probably because bird pollinators effectively transfer pollen during foraging for nectar between different fragment sizes and isolation distances In contrast, the fragment size and isolation distance had adverse effects on species reproduction in both G parviflorum and B eriocarpa Plants of both species had fewer pollen grains in their flowers and lower seed set in smaller and more isolated fragments These insect-pollinated species could face limitations of pollen availability under fragmented conditions This study raises the possibility that various plant species may react to habitat fragmentation in a different way in terms of their reproductive success, which may be related to their pollination system and responsible pollinators

Introduction

Anthropogenic habitat fragmentation is considered one of the main factors influencing the persistence and viability of floristic populations (Aizen & Feinsinger, 1994; Hobbs & Yates,

2003) It often alters ecological interactions and imposes strong selection pressures that may be used to promote the evolution of ecologically significant straits in isolated and small populations (Jacquemyn et al., 2012) Empirical research on how fragmentation affects species reproduction have found that fragmented habitat seems to have a negative impact on population persistence as it could disrupt plant-pollinator interactions, and decreases pollinator abundance and diversity in a fragment (Rathcke & Jules, 1993), thus leading to the reduction of seed set in plants (Jennersten, 1988) and decreased reproductive success in small and isolated populations (Brys et al., 2004) Moreover, fragmentation can also influence the fitness of individuals (Yates & Ladd, 2005) in the remaining patches by reducing reproductive output and offspring fitness (Jacquemyn et al., 2012), limiting dispersal (Bélisle, 2005), influencing foraging behaviour (Robinson & Holmes, 1982) As a result, reduced genetic diversity and genetic viability impacts the quality of pollen and the production of seeds (Aspinwall & Christian, 1992; Vargas et al., 2006)

The size of fragmentation and degree of isolation have been seen as key to determining species diversity as they are affected by plant pollinators interaction (Donaldson et al., 2002; Gilpin, 1986; Haila et al., 1993; Moody‐Weis & Heywood, 2001; Saunders et al., 1991; Wilcox

& Murphy, 1985) Large fragments or continuous habitat are likely to have more available resources and contain a greater species diversity such that they may attract more pollinators while small fragments are expected to have limited resources supporting fewer pollinators (Kearns et al., 1998; Kremen & Ricketts, 2000; Rathcke & Jules, 1993) Pollinators visit the flower for resources, therefore changes in remnant size, vegetation density, and habitat isolation may affect the frequency of visits to flowers (Agren, 1996; Fritz & Nilsson, 1994;

Groom, 1998; Sih & Baltus, 1987) Thus, a lower visit rate to plants in small populations may lead to pollen limitation causing a reduction of reproductive success and population viability The number, frequency and quality of pollinators attracted to the flowering plant also depend on the isolation distance of fragment The composition and frequency of flower visitors changes with the increase of isolation of fragmented habitats For example, the frequency of visits to flowers and the diversity of the pollinators will be reduced if the gap between populations or fragments exceeds the pollinators' foraging range (Harris & Johnson, 2004) The abundance and distribution of insect pollinators like euglossine and other solitary bees (Rathcke & Jules, 1993) are dependent on the degree of isolation and fragment size because they can travel only short distances from a few hundred meters (Byrne et al., 2007; Ottewell et al., 2009; Smith et al., 2017) to up to 2.4 km (Roubik & Aluja, 1983) Furthermore, these pollinators depend on one or a few plants so they are more vulnerable in isolated fragmentations where the diversity of plant species is lower than in continuous habitat and large fragments (Rathcke & Jules, 1993) Gibbs and Stanton (2001) have shown that, for beetles, the fragmented forests had one-third less species richness and three times less abundance when compared to continuous forests As a result, fragmentation appears to restrict pollinator movement between patches, thereby reducing gene flow and causing inbreeding (Lande & Barrowclough, 1987; Mills & Smouse, 1994) Inbreeding depression could further decrease the reproductive success of plant species as well as regeneration of species in fragmentations (Aizen & Feinsinger, 1994; Rathcke & Jules, 1993), by reducing reproductive output and offspring fitness (Jacquemyn et al., 2012), limiting dispersal (Bélisle, 2005) and influencing foraging behaviour (Robinson & Holmes, 1982), limiting opportunities for energy acquisition (McIntyre & Wiens, 1999) and influencing chance of predation for their pollinators (Dukas,

2005) Therefore, insect-pollinated plant species may face disadvantages through fragmentation of habitat as size limitation and distance between fragments will influence the

91 abundance and distribution of insect pollinators (Artz & Waddington, 2006) and alter ecological interactions (Whelan et al., 2000)

Plants that are pollinated by both insects and mobile birds, however, may be more resilient to fragmentation Bird species like the honeyeaters (Meliphagidae), the most important pollinating bird lineage in Australia, can distribute pollen over distances of 1-12.5 km (Saunders & De Rebeira, 1991; Yates et al., 2007) so the isolation of the patch may not serve a barrier of pollinator limitation Numerous investigations on how fragmentation affects vascular plants have revealed a little correlation between the size of fragment or distance of isolation on species diversity and plant reproduction (Haila et al., 1993; Kemper, 1997; Kemper et al., 1999; Kửchy & Rydin, 1997) Some plants may be less susceptible to fragmentation and adapted to varying environment than the other species (Kửchy & Rydin, 1997) especially perennial species, which can survive for a long time due to their enduring seed reserves banks or plant structures The population viability of perennial plants should not, therefore, be ignored when measuring the effect of habitat fragmentation on species diversity (Aizen & Feinsinger, 1994; Bond, 1994; Bronstein, 1995)

Some taxa of both plants and animals do surprisingly well in the anthropogenically altered landscape; for example, in urban habitats that tend to biologically homogeneous (McKinney,

2002, 2006) While many taxa may disappear, some taxa may adapt and alter their behaviour to fit the habitat Some (usually invasive/non-native species) even do better in such habitats than anywhere else: so called urban ‘avoiders’, urban ‘adapters’ and urban ‘exploiters’ (Bateman & Fleming, 2012; McKinney, 2006) Within taxa, there could be selection on variation in how individuals cope with their environment, particularly in unstable environments One would therefore expect greater variation in ‘personality’ traits and behavioural syndromes across individuals in animals (Wolf & Weissing, 2012), and in reproduction in plants in such habitats Habitat fragmentation should impose a strong selection

92 pressure, and species that adapt to such a pressure may have the capacity to adapt to such changes via individual heritable variation in behavioural characteristics reflected in personality syndromes, as well as through inherent species-specific reproductive and physiological tactics and traits

It is still challenging to estimate how fragmentation may impact plant species’ reproduction as pollinators respond differently to fragmentation This study aims to determine how the size of habitat fragments and the degree of isolation, and their interaction, might have influenced plant species reproduction output and pollinator assemblage The study focuses on the effect of different patch sizes and isolation on seed set viability, pollination efficiency and pollinator abundance and diversity My research questions are:

 Can the small and isolated fragments provide sufficient pollination service and a diverse pollinator assemblage?

 Do the small and more isolated fragments have fewer pollinator visits to open flowers than do the larger and less isolated fragments?

 How have fragment size and isolation distance influenced plant species reproduction output?

Method

4.3.1 Study species and experimental system design

Among many plants identified from the vegetation and plant community survey at Dryandra Woodland, three plant species, namely Banksia sessilis (Proteaceae), Gastrolobium parviflorum (Fabaceae) and Bossiaea eriocarpa (Fabaceae) that have sufficiently high abundance in each of the six remnant vegetation fragments, were selected for their reproductive response to habitat fragmentation

Pollinator assemblage was determined by observing the frequency of visits from mobile bird species on open flowers I paid special attention to honeyeater species (Meliphagidae), a major bird lineage that pollinate several different plant taxa in south-west Australia (Brown et al., 1997; Hopper, 1981; Keighery, 1982; Whelan & Burbidge, 1980) In each of six fragments, the frequency of visits from mobile birds to open flowers of Banksia sessilis and their foraging behaviour was recorded in the 2016 and 2017 flowering seasons In the beginning, the observer stood at the centre or the edge of the patch of Banksia sessilis observing birds visiting the flowers by using binoculars and cameras The species of each bird visiting open flowers was identified and their movement within a plant and between plants was recorded and followed until it was lost from the sight Each population was censused three times a day, morning (7.00 – 8.00), midday (10.00 – 11.00) and afternoon (1.00 – 2.00) by the same observer

To quantify levels of pollination, I harvested 10 – 30 flowers from observed plants of each species in the selected fragments and stored them in 70% ethanol before processing Styles from each of the selected flowers were isolated for pollen grain estimation Styles were fixed and stored overnight in individual microcentrifuge tubes containing acetic acid and ethanol (acetic acid: ethanol, 10:90) The submerged pistil tissue was softened in 1M NaOH solution overnight and washed three times with KPO4 buffer made with 50% glycerol, before being stained with 0.01% aniline blue in KPO4 buffer (dye) (Lu, 2011) The stained softened pistil was placed on a slide and then flattened under a cover slide with a gentle force The number of pollen grains on the stigma was counted under epifluorescence microscopy, and the average number of pollen grains per style per plant was determined

For each of the selected species in the six researched fragments, the number of fruits per plant was estimated for ten randomly selected plants Fruits per species per fragment were collected randomly from different branches of each plant, and seeds were extracted The number of seeds per individual plant was estimated, and seed weight and viability were measured

Due to the mean number of seed and mean seed mass-produced by a multi-seeded fruit of selected species being low, the mean seed mass per individual plant was chosen for measurement instead of mean seed mass per fruit For each species, the mean seed mass was obtained from up to ten different plants and only healthy seeds were chosen in each fragment Mean seed mass was measured and randomly collected from samples of 50 seeds per species per patch

All statistics were performed using the statistical software PRIMER V7 and R (R core team,

2018) All variables were tested for normality with a Shapiro-Wilk test and homogeneity of variance with a Levene test After testing for normality (Shapiro-Wilk test; α=0.05) and homogeneity of variance (Levene’s test; α=0.05), the following variables were not normally distributed; species abundance, the number of the pollen grain, frequency visit of birds These variables were log(x+1) transformed to reduce variation and use for further analysis

Two-way ANOVAs were performed to assess the seed weight, the number of pollen grain per pistil, frequency of visits from birds and pollinator diversity The number of honeyeaters and other birds that visited Banksia sessilis inflorescences across times of day and day of each patch were pooled to estimate the species abundance in 2016 and 2017 Two-dimensional non- metric multidimensional scaling (nMDS) was used to visualise the relative similarity and

95 relative groupings of species based on the distance between points using the Bray-Curtis similarity index Analysis of similarity (ANOSIM) was performed to determine whether the pollinator compositions were similar between fragments with different degrees of isolation Similarity percentages (SIMPER) were also used to identify the distribution of each bird species between the fragments Statistical significance was taken at P < 0.05.

Result

A total of 167 birds of ten species were recorded during two years study Among the ten observed bird species, eight belonged to the honeyeater family (Meliphagidae), and the other two belonged to the Acanthizidae and Zosteropidae family, respectively (Table 4.1) The four most abundant bird species were White-cheeked Honeyeater (Phylidonyris niger), Western Thornbill (Acanthiza inornata), Brown Honeyeater (Lichmera indistincta) and New Holland Honeyeater (Phylidonyris novaehollandiae) White-cheeked Honeyeater and Brown Honeyeater were recorded across all fragments In contrast, Singing Honeyeater (Gavicalis virescens), Red Wattlebird (Anthochaera carunculata) and Western Wattlebird (Anthochaera lunulata) were observed only at the largest fragment (A)

Table 4.1: The abundance of honeyeater and other bird species across six Banksia sessilis populations in two years censuses (2016 and 2017)

Figure 4.1: Some nectarivorous birds recorded in Banksia sessilis populations

A: Brown Honeyeater (Lichmera indistincta) B: Western Wattlebird (Anthochaera lunulata) C: White-cheeked Honeyeater (Phylidonyris niger) D: New Holland Honeyeater (Phylidonyris novaehollandiae) E: Western Thornbill (Acanthiza inornata) F: Western Spinebill

Figure 4.2: The number of birds recorded in each of the six Banksia sessilis population fragments during two consecutive years 2016 and 2017 a) The plot between the number of birds and fragment size, and b) The plot between the number of birds and the degree of isolation The value represents the sum of the birds recorded in each fragment

Despite a trend towards greater bird abundance in big fragments, the two-way analysis of variance (ANOVA) showed that fragment size and isolation did not significantly affect bird

99 species richness and abundance (p > 0.1) There was no significant and consistent correlation between the number of birds and fragment sizes in either 2016 or 2017 (r = - 0.5 and r = 0.18, respectively p > 0.05 Figure 4.2a) A similar result was also found in the correlation between the number of birds and the degree of isolation in the two consecutive years 2016 and 2017 (r

The nMDS (stress = 0) and cluster analysis (based on the Bray-Curtis index) indicated that there was a clear pattern of similarity in bird species composition between fragments The fragments were identified by two different groupings that shared 55% similarity showing that the large fragments (A and D) had a different pollinator composition to small fragments (F) and (E) with (B and C) (Figure 4.3, 4.4) However, the two-way ANOSIM result indicated that the composition of pollinator species did not differ between large and small fragments (Global

R = 0.25, p = 0.2), nor between isolated and less isolated fragments (Global R = 0.14, p = 0.4) Percentage of similarity and dissimilarity (SIMPER) showed that the average dissimilarity between large and small fragments was 42.33% in which three bird species, i.e Yellow-plumed Honeyeater, White-cheeked Honeyeater and New Holland Honeyeater, contributed more than 54.34% dissimilarity between large and small fragments Similarly, the average dissimilarity between isolated and less isolated fragments was 45.26%, and the Western Thornbill, White- cheeked Honeyeater and Brown Honeyeater were the three species sharing 55.31 % dissimilarity between isolated and less isolated fragments

Figure 4.3: Two-dimensional non-metric multidimensional scaling (nMDS) plots (Bray-Curtis similarity) showing similarities between fragments

Figure 4.4: Cluster analysis showing the similarity of the birds' species community composition at six different fragments (Blue branches indicated the large fragments and Red branches indicated small fragments)

A total of 167 nectarivorous birds visited Banksia sessilis flowers Of these, 77 were recorded within large fragments and 90 were recorded within small fragments Of the 167 birds, 102 were observed in isolated fragments, and 65 were observed in less isolated fragments The most recorded species in all fragments such as Brown Honeyeater, New Holland Honeyeater and White-cheeked Honeyeater were also the most active in foraging activities These species all probed multiple flowers on multiple individual Banksia sessilis plants (Figure 4.5)

Figure 4.5: The number of birds visiting different Banksia sessilis plants across six fragments

Although the number of birds visiting flowers in small and isolated fragments was higher than that in large and less isolated fragments, the total visits of birds was not statistically different between large and small fragments (F = 0.1, df = 1, P = 0.77) or between isolated and less isolated fragments (F = 1.23, df = 1, P = 0.38) Similarly, the mean number of visits per species did also not differ significantly between large and small fragments (F= 15.9, df = 1, P

> 0.05) and isolated and less isolated fragments either (F= 0.004, df = 1, P = 0.95)

The visits of nectarivorous birds within all fragments were most frequently observed in the morning (0700 – 0800), and they were more active in probing flowers during this time; the number of birds visiting flowers decreased during the day (Figure 4.6)

Figure 4.6: The total number of birds visiting flowers at a different time across the six fragments

The mean number of pollen grains per stigma varied between fragments ranging from 1 to 45 grains for B sessilis, from 1 to 75 grains for Bossiaea eriocarpa, and from 2 to 50 grains for

G parviflorum Fragment sizes and isolation distance had a significant effect on the pollen grain of B sessilis However, small and less isolated fragments had a higher mean number of pollen grains per stigma than those from large and isolated fragments (Table 4.2, Figure 4.7a) Fragment size and isolation distance and their interaction significantly affected the number of pollen grains of B eriocarpa The smaller and less isolated fragments had a lower number of pollen grains than those from larger and isolated fragments (Table 4.2, Figure 4.7c) Only isolation distance significantly affected the mean number of pollen grains of G parviflorum and plants in less isolated had more pollen grain per stigma (Table 4.2, Figure 4.7b)

Table 4.2: ANOVA table for two-factor analysis of fragment size and degree of isolation on the mean number of pollen grains per stigma

Bossiaea eriocarpa df F p df F p df F p

Figure 4.7: Effect of fragment size and isolation on the mean number of pollen grain of (a) B sessilis, (b) G parviflorum, and (c) B eriocarpa

Figure 4.8: Pollen grain on the stigma of (a) B sessilis, (b) B eriocarpa, and (c) G parviflorum under scanning electron microscopy (scale bar 200 àm)

4.4.5 Seed production and seed weight

The total seed production per B sessilis plant was unaffected by both fragment size and isolation distance (Table 4.3, Figure 4.9 a) In contrast, fragment size and isolation distance significantly affected seed production of G parviflorum with a higher number of seeds recorded in large and less isolated fragments (Table 4.3, Figure 4.9 b) Seed production of B eriocarpa was only affected by fragment size, and plants in larger fragments produced more seeds than those from smaller fragments (Table 4.3, Figure 4.9 c)

Table 4.3: ANOVA table for two-factor analysis of the effect of fragment size and degree of isolation on the seed production

Figure 4.9: Effect of fragment size and isolation on the seed production of (a) B sessilis, (b)

The seed weights of B sessilis and B eriocarpa species were unaffected by both fragment sizes and distance of isolation (Table 4.4, Figure 4.10 a, c) In contrast, G parviflorum was affected significantly by both fragment size and isolation distance, and plants in larger and less isolated fragments had a higher seed weight (Table 4.4, Figure 4.10 b)

Table 4.4: ANOVA table for two-factor analysis of the effect of fragment size and degree of isolation on the seed weight

Figure 4.10: Effect of fragment size and isolation on the seed weight of (a) B sessilis, (b) G parviflorum, and (c) B eriocarpa.

Discussion

The negative consequences of fragmented habitat on bird pollinator communities and on plant species’ reproduction have been reported in many studies (Aizen & Feinsinger, 1994; Brudvig et al., 2015; Cascante et al., 2002; Donaldson et al., 2002; Kolb, 2008; Lennartsson, 2002) The other studies, however, also indicated that some species may respond positively, and show resilience to fragmentation (Bruna & Kress, 2002; Donaldson et al., 2002; Roque et al., 2017; Sáyago et al., 2018; Steffan-Dewenter & Tscharntke, 1999; Yates et al., 2007; Yates & Ladd,

In this study, the effects of habitat fragmentation on the bird pollinators’ diversity and plant reproduction were not consistent across the three investigated plant species There was no significant difference in bird pollinators' abundance and richness across six fragments, though there was a trend for a decline in species diversity from large fragments to small fragments and from isolated to less isolated fragments Eight out of ten bird species were observed in the largest fragments (A) compared to only three bird species observed in the smallest fragments over two years of observation Similarly, the species composition was not

109 significantly different between fragments The results of this study is in line with the study of Yates et al (2007) that also noted that a small population of Calothamnus quadrifidus

(Myrtaceae) could support a similar honeyeater diversity as could large populations Honeyeaters can travel between fragments from 3.1 km to 12.5 km apart to visit a greater number of flowers to fulfil their energy needs (Saunders & De Rebeira, 1991; Yates et al.,

2007) Therefore, the bird species composition of small population fragments may have a similar amount of pollinator birds as large population fragments in this study

Although the pollinator bird diversity and composition across the six fragments were similar, the number of birds visiting the B sessilis populations in all fragments was not abundant in Dryandra Woodland with only hundreds recorded during the two years censuses

A similar result was also reported by Recher and Davis Jr (2011) who found only tens or hundreds of nectar-feeders in Dryandra Woodland, not thousands of birds as has been reported in other forests in south-western Australia (Recher & Davis, 2002; Recher & Davis Jr, 2010) The results may reflect the possibly smaller number of pollinator birds in the region of Dryandra Woodland

Birds may not be attracted to small habitat fragments as those habitats do not provide sufficient food sources Therefore, bird pollinators may make fewer visits to small population fragments (Jennersten, 1988; Lamont et al., 1993) Birds tend to visit multiple inflorescences within one plant in small population fragments and frequently move between plants in larger population fragments (Kolb, 2008; Yates et al., 2007) However, this study did not find any statistical differences in the total number of pollinator visits and the mean number of visits per pollinator to plants between different population fragments and by isolation distances Bird pollinators across all fragments rarely visited only one plant They spend most of their time visiting two or even three plants for nectar, and they probed several inflorescences during their forages, as reported previously (Collins & Grey, 1989; Yates et al., 2007) The greatest number

110 of visits of nectarivorous birds to B sessilis was recorded in the mornings, as was the most active foraging between plants, perhaps due to the fact that the highest nectar production of B sessilis was recorded at dawn and the majority of the flowers were opened throughout the day

(Lamont et al., 1998) This could also reflect bird behaviour: nectarivorous birds tend to be small and have high metabolic rates and a low capacity for energy storage; therefore, they need to actively feed in the morning after fasting all night (Ferguson et al 2019)

The effect of fragmentation on pollination varied widely among three species in this study, suggesting that species that are attractive to different pollinators may respond differently to fragmentation Indeed, a survey by Aizen and Feinsinger (1994) in the Chaco dry forest in Argentina found that pollination levels of hummingbird-pollinated hemiparasite Ligaria increased significantly as the forest transitioned from continuous to fragmented In contrast, the other insect-pollinated species showed a significant decline in pollination in smaller fragment sizes (Aizen Feinsinger, 1994) Steffan-Dewenter and Tscharntke (1999) also indicated that the abundance and richness of bees decreased with the increase of isolation, leading to the reduction in the seed set of mustard (Sinapis arvensis) and radish (Raphanus sativus) species In this study, B sessilis is mainly pollinated by nectarivorous birds (Collins

& Grey, 1989) These birds are highly effective in carrying pollen on their foreheads, throats or bill during foraging for nectar (Collins et al., 2008) These nectarivorous birds also have the ability to travel long distances between fragments to enhance outcrossing of their pollinated plant species (Lamont et al., 1998) Consequently, I found that B sessilis populations did not appear to experience pollinator limitation as the pollinators’ diversity in smaller fragments was similar to those in the larger fragments Therefore, this species might be adapted to fragmentation, as B sessilis plants even received a greater number of pollen grains on their stigma in smaller and more isolated fragments in this study

In contrast, G parviflorum and B eriocarpa are mainly pollinated by insects (Chandler et al., 2002; Ross, 2006) that are unable to forage across long distances between populations to transfer pollen (Kolb, 2005; Kwak et al., 1998) Furthermore, insect species may not be attracted to a small or fragmented population as it produces lower volumes of nectar and pollen (Cheptou & Avendaủo , 2006; Goverde et al., 2002; Mustajọrvi et al., 2001); therefore, insects may make fewer visits to small and fragmented populations, leading to a lower number of pollen grains on the stigma of individuals in small and isolated fragments This study did not survey insects that visit open flowers of G parviflorum and B eriocarpa species; however, I speculate that the lack of insect pollinators could be one of the reasons that lead to the lower pollen grain deposited on the stigma of these two species in small and isolated fragments

The similarity of pollen grain deposition on the stigma of B sessilis between all fragments and isolation distance would translate to a similar seed production per plant, and perhaps seed weight of this species The similarity in pollination efficiency, seed production, and seed weight could indicate that the mobile birds (i.e honeyeaters) were effective in pollination across all fragments and open space (Sáyago et al., 2018) However, the lower seed set of both

G parviflorum and B eriocarpa species recorded in either small or isolated fragments would suggest that a decrease in pollen quantity and/or quality could lead to a reduction in seed production in smaller fragments (Aizen & Feinsinger, 1994; Byers, 1995; Kolb, 2008; Sáyago et al., 2018) Furthermore, some other researchers reported that fragmentation did not affect species reproduction for one year but may manifest in the following years (Morgan, 1999; Tomimatsu & Ohara, 2002) Besides, other aspects caused by fragmentation could also affect pollination limitation and species reproduction; for example, vegetation cover, plant population size, plant size and plant density, and the edge effects on plants near the edges could influence flower production, and seed abortion rates etc (Aizen & Feinsinger, 1994; Brudvig et al., 2015; Donaldson et al., 2002; Kolb, 2005), and therefore their reproductive output

Conclusion

I found that the effects of anthropogenic habitat fragmentation on plant species reproduction varied across three plant species in Dryandra woodland Of the three species, the reproduction of B sessilis showed more resilience to the effect of fragmentation; this is likely to be due to this species’ reliance on pollination by highly mobile honeyeaters that can transfer pollen across long distances Such an effective pollination system would play a key role in maintaining the population viability of B sessilis in smaller and isolated fragments In contrast, species pollinated by insects, such as G parviflorum and B eriocarpa, are more vulnerable to fragmentation as they received lower quality and quantity of pollen These plants produced fewer seeds in small and isolated fragments, which may suffer a decline in population viability

However, as some aspects in this study was conducted for only one year, more precise conclusions about the effect of fragmentation on species reproduction would benefit from multiple years of observation and survey Furthermore, this study only conducted surveys and observations on bird pollinators’ visits to B sessilis flowers; therefore, future experiment should also explicitly examine visits by insects to open flowers of these three species to have a deeper comprehension of the effect of fragmentation on pollination efficiency and the reproduction of these species

References

Agren, J (1996) Population size, pollinator limitation, and seed set in the self‐incompatible herb Lythrum salicaria Ecology, 77(6), 1779-1790

Artz, D R., & Waddington, K D (2006) The effects of neighbouring tree islands on pollinator density and diversity, and on pollination of a wet prairie species, Asclepias lanceolata (Apocynaceae) Journal of Ecology, 94(3), 597-608

Aspinwall, N., & Christian, T (1992) Pollination biology, seed production, and population structure in queen‐of‐the‐prairie, Filipendula rubra (Rosaceae) at Botkin Fen, Missouri American Journal of Botany, 79(5), 488-494

Bateman, P W., & Fleming, P (2012) Big city life: carnivores in urban environments Journal of Zoology, 287(1), 1-23

Bélisle, M (2005) Measuring landscape connectivity: the challenge of behavioral landscape ecology Ecology, 86(8), 1988-1995

Bond, W (1994) Do mutualisms matter? Assessing the impact of pollinator and disperser disruption on plant extinction Royal Society, 344(1307), 83-90

Bronstein, J L (1995) The plant—pollinator landscape Mosaic Landscapes and Ecological

Brown, E., Burbidge, A., Dell, J., Edinger, D., Hopper, S., & Wills, R (1997) Pollination in

Western Australia: a database of animals visiting flowers Western Australian Naturalists Club: Perth

Brudvig, L A., Damschen, E I., Haddad, N M., Levey, D J., & Tewksbury, J J (2015) The influence of habitat fragmentation on multiple plant–animal interactions and plant reproduction Ecology, 96(10), 2669-2678

Bruna, E M., & Kress, W J (2002) Habitat fragmentation and the demographic structure of an Amazonian understory herb (Heliconia acuminata) Conservation Biology, 16(5),

Brys, R., Jacquemyn, H., Endels, P., Van Rossum, F., Hermy, M., Triest, L., Blust, G D

(2004) Reduced reproductive success in small populations of the self‐incompatible

Primula vulgaris Journal of Ecology, 92(1), 5-14

Byers, D L (1995) Pollen quantity and quality as explanations for low seed set in small populations exemplified by Eupatorium (Asteraceae) American Journal of Botany, 82(8), 1000-1006

Byrne, M., Elliott, C., Yates, C., & Coates, D (2007) Extensive pollen dispersal in a bird‐ pollinated shrub, Calothamnus quadrifidus, in a fragmented landscape Molecular Ecology, 16(6), 1303-1314

Cascante, A., Quesada, M., Lobo, J J., & Fuchs, E A (2002) Effects of dry tropical forest fragmentation on the reproductive success and genetic structure of the tree Samanea saman Conservation Biology, 16(1), 137-147

Chandler, G T., Crisp, M D., Cayzer, L W., & Bayer, R J (2002) Monograph of

Gastrolobium (Fabaceae: Mirbelieae) Australian Systematic Botany, 15(5), 619-739

Cheptou, P O., & Avendaủo V, L G (2006) Pollination processes and the Allee effect in highly fragmented populations: consequences for the mating system in urban environments New Phytologist, 172(4), 774-783

Collins, B G., & Grey, J (1989) Preferential foraging by honeyeaters in the Jarrah forest of

Collins, B G., Walsh, M., & Grey, J (2008) Floral development and breeding systems of

Dryandra sessilis and Grevillea wilsonii (Proteaceae) Australian Journal of Botany, 56(2), 119-130

Donaldson, J., Nọnni, I., Zachariades, C., & Kemper, J (2002) Effects of habitat fragmentation on pollinator diversity and plant reproductive success in renosterveld shrublands of South Africa Conservation Biology, 16(5), 1267-1276

Dukas, R (2005) Bumble bee predators reduce pollinator density and plant fitness Ecology,

Ferguson, S M., Gilson, L N., & Bateman, P W (2019) Look at the time: diel variation in the flight initiation distance of a nectarivorous bird Behavioral Ecology and Sociobiology, 73(11), 1-7

Fritz, A.-L., & Nilsson, L A (1994) How pollinator-mediated mating varies with population size in plants Oecologia, 100(4), 451-462

Gilpin, M (1986) Minimum viable populations: Processes of extinction in ME Soulé editor

Conservation Biology: The Science of Scarcity and Diversity Sinauer Associates, Sunderland, MA,

Goverde, M., Schweizer, K., Baur, B., & Erhardt, A (2002) Small-scale habitat fragmentation effects on pollinator behaviour: experimental evidence from the bumblebee Bombus veteranus on calcareous grasslands Biological Conservation, 104(3), 293-299

Groom, M J (1998) Allee effects limit population viability of an annual plant The American

Haila, Y., Saunders, D., & Hobbs, R J (1993) What do we presently understand about ecosystem fragmentation Nature Conservation, 3, 45-55

Harris, L F., & Johnson, S D (2004) The consequences of habitat fragmentation for plant– pollinator mutualisms International Journal of Tropical Insect Science, 24(1), 29-43 Hobbs, R J., & Yates, C J (2003) Impacts of ecosystem fragmentation on plant populations: generalising the idiosyncratic Australian Journal of Botany, 51(5), 471-488

Hopper, S D (1981) Honeyeaters and their winter food plants on granite rocks in the central wheatbelt of Western Australia Wildlife Research, 8(1), 187-197

Jacquemyn, H., De Meester, L., Jongejans, E., & Honnay, O (2012) Evolutionary changes in plant reproductive traits following habitat fragmentation and their consequences for population fitness Journal of Ecology, 100(1), 76-87

Jennersten, O (1988) Pollination in Dianthus deltoides (Caryophyllaceae): effects of habitat fragmentation on visitation and seed set Conservation Biology, 2(4), 359-366

Kearns, C A., Inouye, D W., & Waser, N M (1998) Endangered mutualisms: the conservation of plant-pollinator interactions Annual Review of Ecology and Systematics, 29(1), 83-112

Keighery, G (1982) Bird-pollinated plants in Western Australia Pollination and Evolution,

Kemper, J (1997) The effects of fragmentation of South Coast Renosterveld on vegetation patterns and processes University of Cape Town

Kemper, J., Cowling, R M., & Richardson, D M (1999) Fragmentation of South African renosterveld shrublands: effects on plant community structure and conservation implications Biological Conservation, 90(2), 103-111

Kửchy, M., & Rydin, H (1997) Biogeography of vascular plants on habitat islands, peninsulas and mainlands in an eascentral Swedish agricultural landscape Nordic Journal of Botany, 17(2), 215-223

Kolb, A (2005) Reduced reproductive success and offspring survival in fragmented populations of the forest herb Phyteuma spicatum Journal of Ecology, 93(6), 1226-

Kolb, A (2008) Habitat fragmentation reduces plant fitness by disturbing pollination and modifying response to herbivory Biological Conservation, 141(10), 2540-2549

Kremen, C., & Ricketts, T (2000) Global perspectives on pollination disruptions

Kwak, M M., Velterop, O., & van Andel, J (1998) Pollen and gene flow in fragmented habitats Applied Vegetation Science, 1(1), 37-54

Lamont, B B., Klinkhamer, P G., & Witkowski, E (1993) Population fragmentation may reduce fertility to zero in Banksia goodii a demonstration of the Allee effect Oecologia,

Lamont, B B., Olesen, J M., & Briffa, P J (1998) Seed production, pollinator attractants and breeding system in relation to fire response — are there reproductive syndromes among co-occurring proteaceous shrubs? Australian Journal of Botany, 46(4), 377-385 Lande, R., & Barrowclough, G F (1987) Effective population size, genetic variation, and their use in population management Viable Populations for Conservation, 87, 124

Lennartsson, T (2002) Extinction thresholds and disrupted plant–pollinator interactions in fragmented plant populations Ecology, 83(11), 3060-3072

McIntyre, N E., & Wiens, J A (1999) Interactions between landscape structure and animal behavior: the roles of heterogeneously distributed resources and food deprivation on movement patterns Landscape Ecology, 14(5), 437-447

McKinney, M L (2002) Urbanization, biodiversity, and conservation The impacts of urbanization on native species are poorly studied, but educating a highly urbanized human population about these impacts can greatly improve species conservation in all ecosystems Bioscience, 52(10), 883-890

McKinney, M L (2006) Urbanization as a major cause of biotic homogenization Biological

Mills, L S., & Smouse, P E (1994) Demographic consequences of inbreeding in remnant populations The American Naturalist, 144(3), 412-431

Moody‐Weis, J M., & Heywood, J S (2001) Pollination limitation to reproductive success in the Missouri evening primrose, Oenothera macrocarpa (Onagraceae) American Journal of Botany, 88(9), 1615-1622

Morgan, J W (1999) Effects of population size on seed production and germinability in an endangered, fragmented grassland plant Conservation Biology, 13(2), 266-273 Mustajọrvi, K., Siikamọki, P., Rytkửnen, S., & Lammi, A (2001) Consequences of plant population size and density for plant-pollinator interactions and plant performance

Ottewell, K M., Donnellan, S C., Lowe, A J., & Paton, D C (2009) Predicting reproductive success of insect-versus bird-pollinated scattered trees in agricultural landscapes

Recher, H., & Davis, W (2002) Foraging profile of a Salmon gum woodland avifauna in

Western Australia Journal of the Royal Society of Western Australia, 85(2), 103-111 Recher, H F., & Davis Jr, W E (2010) The foraging behaviour of woodland birds along the mulga-eucalypt line on Mt Gibson station, Western Australia during late winter and spring Western Australian Journal of Ornithology, 2, 29-41

Recher, H F., & Davis Jr, W E (2011) Observations on the foraging ecology of honeyeaters

(Meliphagidae) at Dryandra Woodland, Western Australia Amytornis: Western Australian Journal of Ornithology, 3, 19-29

Robinson, S K., & Holmes, R T (1982) Foraging behavior of forest birds: the relationships among search tactics, diet, and habitat structure Ecology, 63(6), 1918-1931

Roque, B B., Koptur, S., & Sah, J P (2017) The effects of habitat fragmentation on the reproduction and abundance of Angadenia berteroi Journal of Plant Ecology, 10(2)

Ross, J (2006) A conspectus of the Western Australian Bossiaea species (Bossiaeeae:

Roubik, D W., & Aluja, M (1983) Flight ranges of Melipona and Trigona in tropical forest

Journal of the Kansas Entomological Society, 217-222

Saunders, D., & De Rebeira, C (1991) Values of corridors to avian populations in a fragmented landscape In: Saunders, D.A.; Hobbs, Richard J (eds.) Nature conservation 2: the role of corridors Chipping Norton, N.S.W: Surrey Beatty and Sons Pty Ltd

Sih, A., & Baltus, M.-S (1987) Patch size, pollinator behavior, and pollinator limitation in catnip Ecology, 68(6), 1679-1690

Smith, J P., Heard, T A., Beekman, M., & Gloag, R (2017) Flight range of the Australian stingless bee Tetragonula carbonaria (Hymenoptera: Apidae) Austral Entomology,

Steffan-Dewenter, I., & Tscharntke, T (1999) Effects of habitat isolation on pollinator communities and seed set Oecologia, 121(3), 432-440

Tomimatsu, H., & Ohara, M (2002) Effects of forest fragmentation on seed production of the understory herb Trillium camschatcense Conservation Biology, 16(5), 1277-1285 Vargas, C F., Parra‐Tabla, V., Feinsinger, P., & Leirana‐Alcocer, J (2006) Genetic diversity and structure in fragmented populations of the tropical orchid Myrmecophila christinae var christinae 1 Biotropica, 38(6), 754-763

Whelan, R J., Ayre, D J., England, P R., Llorens, T., & Beynon, F (2000) Ecology and genetics of Grevillea (Proteaceae): implications for conservation of fragmented populations Conservation Biology Series-Cambridge, 253-270

Whelan, R J., & Burbidge, A H (1980) Flowering phenology, seed set and bird pollination of five Western Australian Banksia species Australian Journal of Ecology, 5(1), 1-7 Wilcox, B A., & Murphy, D D (1985) Conservation strategy: the effects of fragmentation on extinction The American Naturalist, 125(6), 879-887

Wolf, M., & Weissing, F J (2012) Animal personalities: consequences for ecology and evolution Trends in Ecology and Evolution, 27(8), 452-461

Yates, C J., & Ladd, P G (2005) Relative importance of reproductive biology and establishment ecology for persistence of a rare shrub in a fragmented landscape

Effect of anthropogenic habitat fragmentation on population genomic architecture of

Abstract

It has been proposed that plant populations in small and isolated habitats created by anthropogenic activities, may have lower genetic variation levels and higher genetic differentiation This study examines the level of genetic diversity and genetic differentiation of

Banksia sessilis (Proteaceae), a common species in Dryandra Woodland, to determine whether this species has been negatively affected by anthropogenic fragmentation or demonstrates genetic resilience to fragmentation effects I investigated the effect of anthropogenic habitat fragmentation, mainly fragmentation sizes and degree of isolation, on the level of genetic diversity and genetic differentiation of B sessilis within and between fragments Using high throughput whole-genome shot-gun sequencing, I generated genomic profiles with an average of 887,577 single nucleotide polymorphism (SNPs) for 48 individuals of B sessilis sampled from six habitat fragments varying in their size and degree of isolation The results showed that genomic diversity was not correlated with habitat size or degree of isolation and the six populations were not genetically differentiated I speculate that this species may be favoured by artificial fragmentation as Banksia sessilis is a species that aggressively colonises disturbed and open spaces (Rockel et al., 1982) I argue that this species is genetically resilient to current levels of anthropogenic fragmentation There appears to be sufficient gene exchange between patches, through between-population seed dispersal, to maintain genetic diversity However, the temporal context must be considered, and it is also possible that insufficient time has passed since the fragmentation of Dryandra Woodland for genetic effects to manifest

Key words: whole-genome shot-gun sequencing, single nucleotide polymorphism (SNPs),

Banksia sessilis, anthropogenic habitat fragmentation, Dryandra Woodland

Introduction

Habitat fragmentation and destruction of continuous habitat are recognised as one of the main threats to the biodiversity of terrestrial habitats globally (Fischer & Lindenmayer, 2007; Henle et al., 2004; Novacek & Cleland, 2001; Sala et al., 2000) Fragmentation breaks up continuous habitat into smaller and more isolated patches which may affect interactions between plants and animals (Aizen & Feinsinger, 1994), resulting in smaller population sizes and greater population isolation by limiting their connectivity through dispersal (O'Connell et al., 2006; Young et al., 1993) Smaller and more isolated populations may attract fewer pollinators resulting in a restriction of the movement of pollen (Morgan, 1999), and hence a reduction of gene flow between populations in the fragments (Aizen & Feinsinger, 1994; Young et al.,

1996), and ultimately lower or loss of genetic variation by genetic drift (Frankham, 1996; O'Connell et al., 2006)

Lower genetic variation in small and isolated populations usually leads to a reduction in the outcrossing rate of adult plants and increases the inbreeding coefficient of offspring (Aguilar et al., 2008) Inbreeding may increase homozygosity of recessive deleterious mutations, which could potentially lead to inbreeding depression, reducing the fitness, viability and fecundity of individuals and hence a population in a habitat fragment (Charlesworth & Willis, 2009; Couvet, 2002; Keller & Waller, 2002; Lienert, 2004; Reed et al., 2003; Young et al., 1993).Genetic variation is, therefore, considered an important source of adaptations for populations in the face of fluctuations in the environment in both the short term and in evolutionary ability to react to future changes in the environment in the long term (Booy et al., 2000; Lienert, 2004; Nason et al., 1997; Young et al., 1996)

Plant species may differ in their response to habitat fragmentation and the susceptibility of particular plant species to genetic effects of fragmentation may be influenced by many factors, such as time since fragmentation, mating system, and mechanism of pollination (Aguilar et al.,

2008; Honnay et al., 2007; Vranckx et al., 2012) However, it is also likely that not all species are negatively influenced by fragmentation Several empirical studies on woody plants have shown that there was no evidence of fragmentation effect on genetic parameters (Collevatti et al., 2001; Gapare & Aitken, 2005; Lemes et al., 2003; Muir et al., 2004) In some circumstances, gene flow can even increase among populations when fragmentation occurs (Young et al., 1996) Therefore, understanding the effect of anthropogenic habitat fragmentation on population genomics of species is essential for developing conservation strategies and management of any remaining small and isolated populations (Mijangos et al., 2015; Picó & Van Groenendael, 2007)

Since the late 1980s, microsatellites markers, also known as simple sequence repeats (SSRs) or short tandem repeats (STRs), have been extensively employed to estimate genetic diversity and genetic differentiation in population genetic studies (Fischer et al., 2017; Guichoux et al., 2011) The application of microsatellites has been extensively used in studies on the genetic consequences of habitat fragmentation on many different species including trees (Bradbury & Krauss, 2013; Zhang et al., 2012), shrubs (Llorens et al., 2018), herbaceous plants (Fischer et al., 2017; Gentili et al., 2018), grassland plants (Sullivan et al., 2019), and bird species (Hermes et al., 2016) However, due to the limited number of microsatellite markers used, the marker determines the bias as well as high variance in estimates of diversity derived from microsatellites

In recent years, with the rapid reduction in cost of next-generation sequencing (Bragg et al., 2015) and the development of new methods incorporating with next-generation sequencing that enables the running of hundreds of samples with a small amount of genomic material within a short period of time and reasonable expense (Peterson et al., 2012), and allows the use of even a small number of samples per population to detect genetic differentiation (Willing et al., 2012) Compared to microsatellite markers, genome-wide single-nucleotide

124 polymorphisms (SNPs) measures the genome-wide genetic diversity (Fischer et al., 2017; Haasl & Payseur, 2011; Queirús et al., 2015; Vọli et al., 2008) Furthermore, SNPs require only a small number of sample sizes for accurate estimation of allelic frequency (Guichoux et al.,

2011) Therefore, genome-wide SNPs analysis is highly recommended for inferring and comparing genetic diversity within and between populations as its reliability and cost- effectiveness (Fischer et al., 2017)

In this chapter, I focused on the genomic architecture of Banksia sessilis in Dryandra Woodlands, Western Australia I used high throughput whole-genome shot-gun sequencing to generate SNPs profile and estimated the genomic diversity level and genomic differentiation of Banksia sessilis between three larger fragment populations and three smaller fragment populations varying in degrees of isolation I tested two hypotheses:

(i) The levels of genetic diversity of big B sessilis populations from large fragments of habitat are higher than those of populations from smaller fragments

(ii) B sessilis populations between large fragments of habitat have lower genetic differentiation than do those populations between smaller fragments of habitat.

Methods

Banksia sessilis (Proteaceae), until 2007 this species was known as Dryandra sessilis, is a shrub or small tree up to 5m tall This species is widely distributed across the southwest corner of Western Australia (Collins et al., 2008) and is often found in eucalyptus woodland (Marchant

& Menadue, 1987) Banksia sessilis is killed by fire (nonsprouter) and regenerates from seeds in the following wet season It can produce high numbers of flowers and up to 600 seeds per plant (Lamont et al., 1998) Therefore, B sessilis is seen as a key food source providing nectar

125 and insects for bees, honeyeaters, and other bird species during the winter months (Arnold,

Young leaves were collected from eight randomly chosen individual mature B sessilis plant from each of six selected patches (see Figure 2.1) Each plant was at least ten m apart from each other in order to prevent sampling of individuals that have genetically related The collected leaf samples then were refrigerated immediately for storage Genomic DNA was extracted using the Cetyltrimethylammonium bromide (CTAB) protocol described by He et al

(2004) About 100 àg of genomic DNA was used for library construction and high throughput sequencing Library construction and sequencing were carried out at Novogene Sequencing (Beijing) following their standard protocol The raw sequence was filtered by the following criteria: (a) reads with >2% unidentified nucleotides (N) or with poly-A structure; (b) reads with ≥40% bases having low quality

5.3.2 Reads mapping and variants calling

The 48 whole-genome sequences were individually mapped to a Banksia attenuata reference genome (He et al., 2019) using CLC Genomics Workbench 9.0 (https://www.qiagenbioinformatics.com/) following its protocols, including further local realignment The B attenuata reference genome includes 33,452 contigs with a minimum of

500 base pairs as described by He et al (2019) For each sample, a binary alignment map (bam) file was exported from CLC Genomics Workbench for SNP variants calling

SNP variants calling was implemented with GATK tool “HaplotypeCaller” algorithm (Van der Auwera et al., 2013) using the BAM read mapping file as the input file B attenuata reference genome (606 Mb, only with contigs > 1000 bp) was used as a reference genome (BAref.fasta) A genomic VCF file (GVCF) was obtained for each sample GVCF record the

126 variants for each genomic position of 606 MB reference were recorded, so that combining variants from all samples could be performed in a later stage

SNP calling code was written as (sample S1_01 as an example, BAref.fasta as Banksia attenuata reference genome): gatk java-options "-Xmx30G" HaplotypeCaller -R BAref.fasta

-I Sample_S101.bam -O Sample_S101.g.vcf.gz -ERC GVCF" Individual GVCF files were combined in two steps using CombineGVCFs function in GATK tool The first step was the combination of all samples from a site (a population), and the second was combining six populations for downstream analysis

Code used for the first-step combination was as follows (population S1 as an example): gatk java-options "-Xmx30G" CombineGVCFs -R BAref.fasta variant Sample_1_1.g.vcf.gz variant Sample_1_2.g.vcf.gz variant Sample_1_3.g.vcf.gz variant Sample_1_4.g.vcf.gz variant Sample_1_5.g.vcf.gz variant Sample_1_6.g.vcf.gz variant Sample_1_7.g.vcf.gz variant Sample_1_8.g.vcf.gz variant Sample_1_10.g.vcf.gz -O sample_1.g.vcf"

Code used for the second-step combination was as follows: gatk java-options "-Xmx30G" CombineGVCFs -R BAref.fasta variant Sample_1.g.vcf variant Sample_2.g.vcf variant Sample_3.g.vcf variant Sample_4.g.vcf variant Sample_5.g.vcf variant Sample_6.g.vcf -O sample_dry.g.vcf"

The combined variants were filtered for further analysis Non-variable variants and Indels were filtered out Only sites with a minimum of 40 samples being recorded and BIALLELIC loci were kept Meanwhile, only sites with calling quality greater than 30 (QUAL > 30.0) and sequence depth greater than 10 (DP > 10) sites were kept The filtering code was as follows:

127 gatk java-options "-Xmx30G" CombineGVCFs -R BAref.fasta variant dry.g.vcf excludeNonVariants minFilteredGenotypes 40 restrictAllelesTo BIALLELIC selectTypeToExclude INDEL -O dry_filtered.g.vcf select "QUAL > 30.0 & DP > 10"

5.3.3 Population genomic diversity and genomic differentiation

Genome-wide nucleotide diversity (π), Watterson’s theta (θ) and Tajima’s D were used to estimate population genomic diversity and demographic history Nucleotide diversity was used to measure the degree of polymorphism within a population Nucleotide diversity is denoted by π (pi) which is based on the average number of nucleotide differences per site between two DNA sequences Watterson’s theta (θ) is based on the total number of segregating sites in a sample of sequences drawn at random from a population (Brown et al., 2004) Nucleotide diversity measures the genomic variation and Watterson’s theta (θ) is similar to expected heterozygosity Nucleotide diversity may be used to monitor diversity within or between ecological populations Nucleotide diversity can be estimated directly from examination of DNA sequences or calculated using molecular marker data (e.g SNP) Tajima’s D was used to infer the history of population demography Tajima’s D is the comparison between the average number of pairwise differences and the number of segregating sites in a sample (Tajima, 1989)

A positive result of Tajima’s D at genome-wide level is consistent with a scenario in which population size is expected to decrease, while a negative D indicates population expansion following a bottleneck (Biswas & Akey, 2006; Tajima, 1989) Nucleotide diversity (π) and Tajima’s D for each population were estimated using TASSEL 5 (Bradbury et al., 2007) Comparison of population genomic diversity and ANOVA were carried out in TASSEL 5 Significance was taken with P < 0.05

Genetic differentiation (FST) was calculated between populations, and exact tests to determine whether FST significantly differ from zero (FST being significantly greater than zero indicate significant genetic differentiation) Estimated FST and exact tests were implemented using Genepop (Rousset, 2008) Significance level was set at P < 0.05

In order to determine whether habitat fragment size and degree of isolation have any effect on the genomic diversity of the six B sessilis population, correlation analysis between fragment size for genomic diversity and degree of isolation were performed using PAST (Hammer et al.,

2001) Significance was set at P < 0.05 The selected fragments varied in their size and degree of isolation so logarithmic transformations were performed in order to reduce variation

To further evaluate the population genetic structure of the six B sessilis population, a

Principal Component Analysis was conducted using TASSEL 5 using SNP profiles of the 48 samples (Bradbury et al., 2007) The first two components were extracted, and an X-Y plot was created with the first two principal components The phylogenetic relationship among the

48 samples was investigated using RAxML following a Maximum likelihood procedure (Stamatakis, 2014)

Results

Whole-genome shotgun sequencing generated an average of 16.7 million clean reads, varying from 12.3 to 20.8 million for the 48 samples (Table 5.1) Thus, an average of 2.51 Giga base (Gb), ranging from 1.84 to 3.12 Gb of clean DNA sequences was obtained for the 48 samples, representing a 3.1 to 5.2 × coverage, assuming a 650 Mega base (Mb) genome size for Banksia species (see, He et al., 2019) An average of 84.9% of the read pairs in all samples was properly mapped to the B attenuata reference genome

Table 5.1: Summary of genomic data for the 48 Banksia sessilis samples

Number of bases (MB) Coverage (×)

After filtering, a total of 887,577 polymorphic SNP markers were discovered within the 48 samples from the six sites On average, 15.4 % of sites were heterozygous in each sample, and the average minor allele frequency was 0.0883 Table 5.2 details the number of heterozygous sites in each sample

Table 5.2: Number of heterozygous sites in each sample

Sample Name Number of Sites Number Heterozygous Proportion Heterozygous

Across the six populations, genome nucleotide diversity (π) varied from 0.1282 in population S6 to 0.1394 in S1, with an average of 0.1337 (Table 5.3) Tukey's pairwise comparison of genome nucleotide diversity π revealed significant differences in genome nucleotide diversity (π) between populations, except for S1-S3, S2-S4, S2-S5, S4-S5 pairs where genome nucleotide diversity (π) remained statistically similar (Table 5.4) ANOVA analysis partitioned 3% of total genetic diversity to between-population, while the majority of genome nucleotide diversity existed within populations (Table 5.5) Another parameter of genetic diversity, theta (θ), also showed a difference between the six populations, with S1 being most diverse, and S6 being the least diverse population (Table 5.3) Tukey's pairwise comparison showed the significant difference in genetic diversity theta (θ) between the populations, except for S1-S3, S2-S5 population pairs (Table 5.6) ANOVA analysis also partitioned the majority of total genetic diversity existing within populations (Table 5.7) Tajima’s D was negative across the six populations, with -1.5751 in S6 and -2.0049 in S3 (Table 5.3)

Table 5.3: Nucleotide diversity in the six populations S.D.: standard deviation

Table 5.4: Genomic diversity comparison - Tukey's pairwise comparison of genome nucleotide diversity π P was shown in the upper triangle

Table 5.5: ANOVA of population genome nucleotide diversity Pi (π)

Sum of sqrt df Mean square F

Table 5.6: Genomic diversity comparison- Tukey's pairwise comparison of genome diversity

Theta (θ) P was shown in the upper triangle

Table 5.7: ANOVA of population genomic diversity Theta (θ)

Sum of sqrs df Mean square F p (same)

For the pairwise FST analysis, I calculated the genetic distances between each pair of populations to evaluate the levels of genetic differentiation between populations Population genetic differentiation was small, with FST varying between 0.0051 and 0.0099 (Table 5.8) The exact test, however, suggested the genetic differentiation between any pair of populations was not significant (P > 0.05)

Table 5.8: Coefficient of genetic differentiation (FST) between pairwise populations

Correlation analysis between population size and genetic diversity has shown no correlation (Figure 5.1 a, b, c) A similar result was also found in the correlation analysis between the degree of isolation and genetic diversity (Figure 5.2 a, b, c) with P > 0.05 in all cases (Table 5.9)

Table 5.9: Correlation analysis between population sizes, degree of isolation on genomic diversity

Pop size vs genetic diversity 0.325 0.528 0.506 0.312 0.29 0.568

Figure 5.1: Correlation between population size and genomic diversity parameters in the six populations of Banksia sessilis: (a) correlation between population size and genetic diversity theta (θ); (b) correlation between population size and genome nucleotide diversity Pi (π); (c) correlation between the population size and Tajima’s D

Figure 5.2: Relationship between the degree of isolation and genomic diversity parameters in the six populations of Banksia sessilis : (a) correlation between the degree of isolation and genetic diversity theta (θ); (b) correlation between the degree of isolation and genome nucleotide diversity Pi (π); (c) correlation between the degree of isolation and Tajima’s D

Principal components analysis (PCA) was used to visualise the population structure among the six populations (Figure 5.3) PCA plot showed complete overlapping of S4, S5 and S6 if the first two principal components were plotted A phylogenetic analysis of the samples further illustrated the genetic structure among the populations (Figure 5.4) Samples from populations S2, S3, S4, S5 and S6 were clustered into independent groups, while it seems these populations were derived from population S1

Figure 5.3: Plot of Principal Component Analysis of the 48 Banksia sessilis samples showing the population structure of the six populations

Figure 5.4: Phylogenetic relationship among the 48 samples of Banksia sessilis The branch length does not scale to the evolutionary rate

Discussion

Many researchers have reported negative effects of fragmentation on genetic diversity of various plant species Vranckx et al (2012) found that small population size was associated with the reduction of genetic diversity parameters such as expected heterozygosity, number of alleles and percentage of polymorphic loci for woody plants Other research conducted on calcareous grassland and cow-wheat also indicated that decreases in habitat area can reduce genetic diversity and have long term effects on plant populations’ genetic diversity (Crichton et al., 2016; Jacquemyn et al., 2010) A meta-analysis concluded that, overall, genetic diversity of species is negatively impacted by human-caused habitat fragmentation (Schlaepfer et al

(2018) The assessment of the impact of fragmentation on genetic diversity across six populations of Banksia sessilis in this study has shown that genetic diversity generally differed between the populations and that the majority of genetic diversity existed within populations This study also indicated that the population from the largest fragment was most genetically diverse, while the lowest genetic diversity was recorded in the population from the population in smallest fragments Despite differences in the genetic diversity between the six populations, the analysis indicated that both population size and degree of isolation were not correlated with the overall population genetic diversity of Banksia sessilis This finding is consistent with the studies conducted on other plant species where the authors found no significant effect or correlation of either fragmentation size or population size and connectivity on genetic diversity (He et al., 2004; Honnay et al., 2007; Llorens et al., 2018; Matesanz et al., 2018; Matesanz et al., 2017; Vargas et al., 2006)

The lack of the effects of fragmentation on genetic diversity across the six populations of

Banksia sessilis could be explained by the absence of restriction on gene flow among

148 populations in fragmented habitats The gene movement between populations takes place through pollen and seeds dispersal (Ouborg et al., 1999) Extensive pollen dispersal in the shrub species Calothamus quadrifidus has been reported by Byrne et al (2007) where pollen was found to be transferred to a maximum distance of 5 km by mobile birds Banksia reproductive success depends highly on mobile birds to deliver outcrossing pollen (Collins & Rebelo, 1987; Ritchie & Krauss, 2012) Nectarivorous birds such as honeyeaters (Meliphagidae) are the main pollinators for Banksia sessilis (Collins et al., 1990; Collins & Grey, 1989) These species can travel long distances between fragments from at least 3.1 km to 12.5 km (Ford et al., 2000; Saunders & Rebeira, 1991) and typically carry a higher pollen load compared to other pollinators such as small animals, invertebrates and insects that only travel for shorter distances with lower pollen amount (Collins et al., 1984; Lamont et al., 1998) As a result, these honeyeater species may enhance outcrossing and plant species that are pollinated by these highly mobile birds may have a greater mating pool than those pollinated by insects (Hopper, 1981; Lamont et al., 1998; Murcia, 1996) These mobile bird species were frequently recorded visiting Banksia sessilis, as I reported in Chapter 4 The flower of this species is considered as the most important source of nectar for honeyeaters (Byrne et al., 2007; Recher & Davis, 2011) Pollen of Banksia sessilis flowers are transferred effectively when honeyeaters visit flowers for nectar as pollen is captured on their forehead, throats, and beaks Furthermore, these birds tended to travel more regularly between neighbouring plants’ inflorescences than between the same plants’ inflorescences as I observed in Chapter 4 and as has been reported previously (Collins et al., 1984; Collins et al., 2008; Llorens et al., 2004) Therefore, the abundance of gene exchange and high genetic connectivity within and between the populations, which is supported by the abundance of pollinators like honeyeater species and their ability to transfer pollen, could have led to genetic rescue and maintain the population size of Banksia sessilis at a level that could reduce the impact of fragmentation

Gene flow may also rely on the seed dispersal events within and between populations Long-distance and between-population seeds dispersal has been reported; for example (He et al., 2004) have shown that seed dispersal events of Banksia hookeriana can spread farther than 2.5 km by wind and birds Seed dispersal distance in Prunus mahaleb population was nearly as twice the distance of pollen dispersal, i.e 990 m of seed dispersal compared to 548 m of pollen dispersal (Garcia et al., 2007) Seed dispersal over long distances could be an important factor in sustaining the resilience of populations in the landscape and promoting migration of species in the face of changes in environmental conditions (Alsos et al., 2007) Although the dispersal ability of B sessilis seeds has not been recorded to date, the seed shape and weight is very like that of B hookeriana and the dispersal ability by winds of B sessilis is likely to be comparable to B hookeriana

Another possibility to explain the limited effects of fragmentation on the genetic diversity of B sessilis could be the insufficient time and insufficient generations since fragmentation occurred Although the land clearing that reduced the native bushland areas to the isolated remnants in the area occurred over the last fifty years (Conservation Commission, 2011), it may be not long enough to detect the differences in genetic diversity or the loss of genetic diversity within populations of this species Theoretically, the negative effects of fragmentation on genetic diversity will increase over time in decades or even centuries and after multiple generations (Barrett et al., 1991; Hamrick, 2004; Lowe et al., 2005; Young & Clarke, 2000) The negative effect of fragmentation on genetic diversity may simply not be significant for fragments that have been more recently isolated (Collevatti et al., 2001; Schlaepfer et al., 2018; Vargas et al., 2006) But for those fragments older than 50 years, the species’ genetic diversity may respond negatively (Rivera‐Ortíz et al., 2015) Therefore, the effects of fragmentation on the genetic diversity of Banksia sessilis should be monitored over a longer term

Low levels of genetic differentiation (FST varying from 0.0051 to 0.0099) among the six B sessilis populations in this study show that these populations are genetically similar These populations may have historically originated from the same large population (Honnay et al.,

2007), which is consistent with the phylogenetic relationship analysis (Figure 5.4) showing the populations S2, S3, S4, S5 and S6 were derived from population S1 The very low genetic differentiation also indicated that there was no restriction of gene flow and low levels of genetic drift among populations inhabiting fragmented habitats (Hamrick, 2004; Koizumi et al., 2006; Matesanz et al., 2018; Mathiasen et al., 2007) The abundance of gene flow could be due to the high level of pollinators visiting flowers as well as seed dispersal between populations The abundance of gene flow could also indicate that the degree of isolation may not be a barrier to pollinators foraging for nectar The genetic differentiation was also in line with the Tajima’s D results where the negative of Tajima’s D was found in the regions across the six populations and varied significantly from zero, ranging from – 1.5751 to – 2.0049 The negative results indicated that there was a bottleneck followed by recovery and expansion of the population size after fragmentation events (Acosta et al., 2019; Tajima, 1989).

Conclusion

Even though many plant species in small and isolated fragmentations have low levels of genetic diversity, and high levels of genetic differentiation among populations, B sessilis in this study showed the reverse tendencies The genetic diversity of B sessilis across the six study populations appears to have the capacity to maintain population viability, even though genetic diversity differs between the six populations The six populations are not significantly genetically differentiated, even though some populations are marginally more diverse than the others Sufficient gene exchange through pollen and possibly seed dispersal between patches,

151 the relatively short time since fragmentation, and expansion ability after fragmentation could play a crucial role in maintaining genetic diversity and reducing negative genetic consequences such as inbreeding and genetic drift in small fragmented populations of Banksia sessilis The results of this study implies that maintenance of genetic connectivity can be achieved through pollinator conservation The absence of effects of population size and degree of isolation on genetic variations shown in this study suggests that future studies on population size and degree of isolation of this species should be carried out at larger spatial scales using various theoretical frameworks

References

Acosta, J J., Fahrenkrog, A M., Neves, L G., Resende, M F., Dervinis, C., Davis, J M.,

Kirst, M (2019) Exome resequencing reveals evolutionary history, genomic diversity, and targets of selection in the conifers Pinus taeda and Pinus elliottii Genome Biology and Evolution, 11(2), 508-520

Alsos, I G., Eidesen, P B., Ehrich, D., Skrede, I., Westergaard, K., Jacobsen, G H.,

Brochmann, C (2007) Frequent long-distance plant colonization in the changing Arctic Science, 316(5831), 1606-1609

Arnold, G (2003) Bird species richness and abundance in wandoo woodland and in tree plantations on farmland at Baker's Hill, Western Australia EMU Austral Ornithology,

Barrett, S., Kohn, J., Fal, D., & Holsinger, K (1991) Genetics and conservation of rare plants

Oxford University Press on Demand

Biswas, S., & Akey, J M (2006) Genomic insights into positive selection Trends in Genetics,

Booy, G., Hendriks, R., Smulders, M., Van Groenendael, J., & Vosman, B (2000) Genetic diversity and the survival of populations Plant Biology, 2(04), 379-395

Bradbury, D., & Krauss, S L (2013) Limited impact of fragmentation and disturbance on the mating system of tuart (Eucalyptus gomphocephala, Myrtaceae): implications for seed- source quality in ecological restoration Australian Journal of Botany, 61(2), 148-160

Bradbury, P J., Zhang, Z., Kroon, D E., Casstevens, T M., Ramdoss, Y., & Buckler, E S

(2007) TASSEL: software for association mapping of complex traits in diverse samples Bioinformatics, 23(19), 2633-2635

Bragg, J G., Supple, M A., Andrew, R L., & Borevitz, J O (2015) Genomic variation across landscapes: insights and applications New Phytologist, 207(4), 953-967

Brown, G R., Gill, G P., Kuntz, R J., Langley, C H., & Neale, D B (2004) Nucleotide diversity and linkage disequilibrium in loblolly pine Proceedings of the National Academy of Sciences, 101(42), 15255-15260

Charlesworth, D., & Willis, J H (2009) The genetics of inbreeding depression Nature

Collevatti, R G., Grattapaglia, D., & Hay, J D (2001) Population genetic structure of the endangered tropical tree species Caryocar brasiliense, based on variability at microsatellite loci Molecular Ecology, 10(2), 349-356

Collins, B., Grey, J., & McNee, S (1990) Foraging and nectar use in nectarivorous bird communities Studies in Avian Biology, 13, 110-121

Collins, B G., & Grey, J (1989) Preferential foraging by honeyeaters in the Jarrah forest of

Collins, B G., Newland, C., & Briffa, P (1984) Nectar utilization and pollination by

Australian honeyeaters and insects visiting Calothamnus quadrifidus (Myrtaceae) Australian Journal of Ecology, 9(4), 353-365

Collins, B G., & Rebelo, T (1987) Pollination biology of the Proteaceae in Australia and southern Africa Australian Journal of Ecology, 12(4), 387-421

Collins, B G., Walsh, M., & Grey, J (2008) Floral development and breeding systems of

Dryandra sessilis and Grevillea wilsonii (Proteaceae) Australian Journal of Botany, 56(2), 119-130

Couvet, D (2002) Deleterious effects of restricted gene flow in fragmented populations

Crichton, R J., Dalrymple, S E., Woodin, S J., & Hollingsworth, P M (2016) Conservation genetics of the annual hemiparasitic plant Melampyrum sylvaticum (Orobanchaceae) in the UK and Scandinavia Conservation Genetics, 17(3), 547-556

Fischer, M C., Rellstab, C., Leuzinger, M., Roumet, M., Gugerli, F., Shimizu, K K.,

Widmer, A (2017) Estimating genomic diversity and population differentiation–an empirical comparison of microsatellite and SNP variation in Arabidopsis halleri BMC

Ford, H A., Geering, D., & Ley, A (2000) Radio-tracking trials with regent honeyeaters

Xanthomyza phrygia and other honeyeaters Corella, 24(3), 25-29

Frankham, R (1996) Relationship of genetic variation to population size in wildlife

Gapare, W J., & Aitken, S N (2005) Strong spatial genetic structure in peripheral but not core populations of Sitka spruce [Picea sitchensis (Bong.) Carr.] Molecular Ecology,

Garcia, C., Jordano, P., & Godoy, J A (2007) Contemporary pollen and seed dispersal in a

Prunus mahaleb population: patterns in distance and direction Molecular Ecology, 16(9), 1947-1955

Gentili, R., Solari, A., Diekmann, M., Duprè, C., Monti, G S., Armiraglio, S., Citterio, S

(2018) Genetic differentiation, local adaptation and phenotypic plasticity in fragmented populations of a rare forest herb PeerJ, 6, e4929

Guichoux, E., Lagache, L., Wagner, S., Chaumeil, P., Léger, P., Lepais, O., & Petit, R J

(2011) Current trends in microsatellite genotyping Molecular Ecology Resources, 11(4), 591-611

Haasl, R J., & Payseur, B A (2011) Multi-locus inference of population structure: a comparison between single nucleotide polymorphisms and microsatellites Heredity, 106(1), 158-171

Hammer, ỉ., Harper, D A., & Ryan, P D (2001) PAST: paleontological statistics software package for education and data analysis Palaeontologia Electronica, 4(1), 9

Hamrick, J L (2004) Response of forest trees to global environmental changes Forest

He, T., Krauss, S L., Lamont, B B., Miller, B P., & Enright, N J (2004) Long‐distance seed dispersal in a metapopulation of Banksia hookeriana inferred from a population allocation analysis of amplified fragment length polymorphism data Molecular Ecology, 13(5), 1099-1109

He, T., Lamont, B B., Enright, N J., D'Agui, H M., & Stock, W (2019) Environmental drivers and genomic architecture of trait differentiation in fire‐adapted Banksia attenuata ecotypes Journal of Integrative Plant Biology, 61(4), 417-432

Henle, K., Lindenmayer, D B., Margules, C R., Saunders, D A., & Wissel, C (2004) Species survival in fragmented landscapes: where are we now? Biodiversity and Conservation,

Hermes, C., Dửpper, A., Schaefer, H M., & Segelbacher, G (2016) Effects of forest fragmentation on the morphological and genetic structure of a dispersal-limited, endangered bird species Nature Conservation, 16, 39

Honnay, O., Adriaens, D., Coart, E., Jacquemyn, H., & Roldan-Ruiz, I (2007) Genetic diversity within and between remnant populations of the endangered calcareous grassland plant Globularia bisnagarica L Conservation Genetics, 8(2), 293-303 Hopper, S D (1981) Honeyeaters and their winter food plants on granite rocks in the central wheatbelt of Western Australia Wildlife Research, 8(1), 187-197

Jacquemyn, H., Roldán-Ruiz, I., & Honnay, O (2010) Evidence for demographic bottlenecks and limited gene flow leading to low genetic diversity in a rare thistle Conservation Genetics, 11(5), 1979-1987

Keller, L F., & Waller, D M (2002) Inbreeding effects in wild populations Trends in Ecology and Evolution, 17(5), 230-241

Koizumi, I., Yamamoto, S., & Maekawa, K (2006) Decomposed pairwise regression analysis of genetic and geographic distances reveals a metapopulation structure of stream‐ dwelling Dolly Varden charr Molecular Ecology, 15(11), 3175-3189

Lamont, B B., Olesen, J M., & Briffa, P J (1998) Seed production, pollinator attractants and breeding system in relation to fire response — are there reproductive syndromes among co-occurring proteaceous shrubs? Australian Journal of Botany, 46(4), 377-385 Lemes, M R., Gribel, R., Proctor, J., & Grattapaglia, D (2003) Population genetic structure of mahogany (Swietenia macrophylla King, Meliaceae) across the Brazilian Amazon,

157 based on variation at microsatellite loci: implications for conservation Molecular Ecology, 12(11), 2875-2883

Llorens, T., Ayre, D., & Whelan, R (2004) Evidence for ancient genetic subdivision among recently fragmented populations of the endangered shrub Grevillea caleyi (Proteaceae)

Llorens, T M., Ayre, D J., & Whelan, R J (2018) Anthropogenic fragmentation may not alter pre‐existing patterns of genetic diversity and differentiation in perennial shrubs

Lowe, A., Boshier, D., Ward, M., Bacles, C., & Navarro, C (2005) Genetic resource impacts of habitat loss and degradation; reconciling empirical evidence and predicted theory for neotropical trees Heredity, 95(4), 255-273

Marchant, N G., & Menadue, M (1987) Flora of the Perth region Western Australian

Herbarium, Department of Agriculture, WA

Matesanz, S., García-Fernández, A., Limón-Yelmo, A., Gómez-Fernández, A., & Escudero, A

(2018) Comparative landscape genetics of gypsum specialists with naturally-patchy distributions reveal their resilience to anthropogenic fragmentation Perspectives in Plant Ecology, Evolution and Systematics, 34, 1-9

Matesanz, S., Rubio Teso, M L., García-Fernández, A., & Escudero, A (2017) Habitat fragmentation differentially affects genetic variation, phenotypic plasticity and survival in populations of a gypsum endemic Frontiers in Plant Science, 8, 843

Mathiasen, P., Rovere, A E., & Premoli, A C (2007) Genetic structure and early effects of inbreeding in fragmented temperate forests of a self‐incompatible tree, Embothrium coccineum Conservation Biology, 21(1), 232-240

Mijangos, J L., Pacioni, C., Spencer, P B., & Craig, M D (2015) Contribution of genetics to ecological restoration Molecular Ecology, 24(1), 22-37

Morgan, J W (1999) Effects of population size on seed production and germinability in an endangered, fragmented grassland plant Conservation Biology, 13(2), 266-273 Muir, G., Lowe, A J., Fleming, C C., & Vogl, C (2004) High nuclear genetic diversity, high levels of outcrossing and low differentiation among remnant populations of Quercus petraea at the margin of its range in Ireland Annals of Botany, 93(6), 691-697

Murcia, C (1996) Forest fragmentation and the pollination of neotropical plants Forest

Nason, J., Aldrich, P., & Hamrick, J (1997) Dispersal and the dynamics of genetic structure in fragmented tropical tree populations Tropical forest remnants: Ecology,

Management, and Conservation of Fragmented Communities, 304-320

Novacek, M J., & Cleland, E E (2001) The current biodiversity extinction event: scenarios for mitigation and recovery Proceedings of the National Academy of Sciences, 98(10), 5466-5470

O'Connell, L., Mosseler, A., & Rajora, O (2006) Impacts of forest fragmentation on the mating system and genetic diversity of white spruce (Picea glauca) at the landscape level Heredity, 97(6), 418

Ouborg, N., Piquot, Y., & Van Groenendael, J (1999) Population genetics, molecular markers and the study of dispersal in plants Journal of Ecology, 87(4), 551-568

Peterson, B K., Weber, J N., Kay, E H., Fisher, H S., & Hoekstra, H E (2012) Double digest RADseq: an inexpensive method for de novo SNP discovery and genotyping in model and non-model species PloS one, 7(5), e37135

Picó, F X., & Van Groenendael, J (2007) Large‐scale plant conservation in European semi‐ natural grasslands: a population genetic perspective Diversity and Distributions, 13(6), 920-926

Queirós, J., Godinho, R., Lopes, S., Gortazar, C., De la Fuente, J., & Alves, P (2015) Effect of microsatellite selection on individual and population genetic inferences: an empirical study using cross‐specific and species‐specific amplifications Molecular Ecology Resources, 15(4), 747-760

Recher, H F., & Davis Jr, W E (2011) Observations on the foraging ecology of honeyeaters

(Meliphagidae) at Dryandra Woodland, Western Australia Amytornis: Western Australian Journal of Ornithology, 3, 19-29

Reed, D H., Lowe, E H., Briscoe, D A., & Frankham, R (2003) Inbreeding and extinction: effects of rate of inbreeding Conservation Genetics, 4(3), 405-410

Ritchie, A L., & Krauss, S L (2012) A genetic assessment of ecological restoration success in Banksia attenuata Restoration Ecology, 20(4), 441-449

Rivera‐Ortíz, F., Aguilar, R., Arizmendi, M., Quesada, M., & Oyama, K (2015) Habitat fragmentation and genetic variability of tetrapod populations Animal Conservation, 18(3), 249-258

Rousset, F (2008) genepop’007: a complete re‐implementation of the genepop software for

Windows and Linux Molecular Ecology Resources, 8(1), 103-106

Rockel, B.A., McGann, L.R & Murray, D.I.L (1982) Phytophthora cinnamomi causing death of Dryandra sessilis on old dieback sites in the jarrah forest Australasian Plant Pathology, 11, 49–50.

Sala, O E., Chapin, F S., Armesto, J J., Berlow, E., Bloomfield, J., Dirzo, R., Kinzig, A

(2000) Global biodiversity scenarios for the year 2100 Science, 287(5459), 1770-

Saunders, D., & De Rebeira, C (1991) Values of corridors to avian populations in a fragmented landscape

Schlaepfer, D R., Braschler, B., Rusterholz, H P., & Baur, B (2018) Genetic effects of anthropogenic habitat fragmentation on remnant animal and plant populations: a meta‐ analysis Ecosphere, 9(10), e02488

Stamatakis A Version, R (2014) RAxML version 8: a tool for phylogenetic analysis and post- analysis of large phylogenies Bioinformatics, 30(9), 1312-1313

Sullivan, E R., Barker, C., Powell, I., & Ashton, P A (2019) Genetic diversity and connectivity in fragmented populations of Rhinanthus minor in two regions with contrasting land-use Biodiversity and Conservation, 28(12), 3159-3181

Tajima, F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism Genetics, 123(3), 585-595

Vọli, ĩ., Einarsson, A., Waits, L., & Ellegren, H (2008) To what extent do microsatellite markers reflect genome‐wide genetic diversity in natural populations? Molecular Ecology, 17(17), 3808-3817

Van der Auwera, G A., Carneiro, M O., Hartl, C., Poplin, R., Del Angel, G., Levy‐Moonshine,

A., Thibault, J (2013) From FastQ data to high‐confidence variant calls: the genome analysis toolkit best practices pipeline Current Protocols in Bioinformatics, 43(1), 11.10 11-11.10 33

Vargas, C F., Parra‐Tabla, V., Feinsinger, P., & Leirana‐Alcocer, J (2006) Genetic diversity and structure in fragmented populations of the tropical orchid Myrmecophila christinae var christinae 1 Biotropica, 38(6), 754-763

Vranckx, G., Jacquemyn, H., Muys, B., & Honnay, O (2012) Meta‐analysis of susceptibility of woody plants to loss of genetic diversity through habitat fragmentation

Willing, E.-M., Dreyer, C., & Van Oosterhout, C (2012) Estimates of genetic differentiation measured by FST do not necessarily require large sample sizes when using many SNP markers PloS One, 7(8), e42649

Young, A., & Clarke, G (2000) Genetics demography and viability of fragmented populations

Young, A., Merriam, H., & Warwick, S (1993) The effects of forest fragmentation on genetic variation in Acer saccharum Marsh.(sugar maple) populations Heredity, 71(3), 277 Zhang, X., Shi, M.-M., Shen, D.-W., & Chen, X.-Y (2012) Habitat loss other than fragmentation per se decreased nuclear and chloroplast genetic diversity in a monoecious tree PLoS One, 7(6), e39146

Summary, conclusions, significance and recommendations

Summary of major results and conclusions from this thesis

Anthropogenic habitat fragmentation has occurred, and is still occurring, across the globe It has been extensively studied worldwide as it is, rightly, considered a significant threat to biodiversity This thesis provides a comprehensive analysis of the ecological and evolutionary response of floristic species to anthropogenic habitat fragmentation at Dryandra Woodland, a recently listed National Park in Western Australia’s Wheatbelt I investigated diversity, composition, growth and reproduction, pollinator assemblage, population genetic diversity and connectivity for several plant species, in the context of anthropogenically fragmented habitat at Dryandra Woodland I draw the overall conclusion that these species responded differently under fragmented conditions Certain floristic species of Dryandra Woodland such as Banksia appears to be resilient to anthropogenic habitat fragmentation so far, but other species like

Gastrolobium and Bossiaea eriocarpa may be adversely affected by anthropogenic fragmentation

Reduction in species richness, abundance, and composition at the patch scale due to habitat loss and fragmentation for plants has been reported in previous studies (Echeverría et al., 2007; ệckinger et al., 2009; ệckinger & Smith, 2006; Ross et al., 2002; Tabarelli et al., 1999) Understanding how fragmentation affects different plant species in terms of species richness, abundance and compositions is essential for conserving and managing fragmented habitat systems in a meaningful way Therefore, in Chapter 2, I examined the species richness, abundance, and composition in habitats with contrasting patch sizes and degrees of isolation I found significant effects of fragment size and degrees of isolation on species abundance Smaller and less isolated fragments had higher species abundance It is likely that smaller patches may be more penetrable by alien weedy species contributing to high species

163 abundance In contrast, species richness was unaffected by the either fragment size or degree of isolation

In Chapters 3 and 4 and I investigated how fragmentation have affected the growth performance of seedlings and species reproduction of Banksia sessilis (Proteaceae),

Gastrolobium parviflorum (Fabaceae) and Bossiaea eriocarpa (Fabaceae) Previous studies

(Hooftman et al., 2003; Kery et al., 2000; Lienert, 2004; Lienert et al., 2002) have documented the negative impact of fragmentation on the growth performance of future generations in plants

In Chapter 3, I detected negative effects of population size and degree of isolation on the growth of seedlings of the three chosen plant species, but the effect varied between the sites and by species with higher germination rates and growth rates in plants from larger and from more isolated populations in general

In Chapter 4, I also identified the negative effects of fragmentation on the reproduction of two plant species, Gastrolobium parviflorum and Bossiaea eriocarpa (Fabaceae) The negative effects of fragmentation on these insect-pollinated species could disrupt plant-pollinator interactions (Rathcke & Jules, 1993) and, therefore, reduce productive output and productive success of the species (Byrne et al., 2007; Jacquemyn et al., 2012; Jennersten, 1988) In contrast, the reproduction of bird-pollinated species such as Banksia sessilis was more resilient to fragmentation Bird visits to open flowers, and bird species composition, was similar between fragment sizes and isolation distances in this study, which is likely to be due to the effective activities of mobile birds across all fragments and open space (Sáyago et al., 2018) I conclude that the response of plant species appears to be related to their capacity of exchanging genes between populations Those species that could disperse their genes effectively and across long-distance are likely to be more resilient to fragmentation

After investigating the effect of fragmentation on species diversity (Chapter 2) and species fitness (Chapter 3 and Chapter 4), in Chapter 5 I further examined the genetic diversity and genetic differentiation of the six populations of Banksia sessilis to investigate whether fragmentation has significantly influenced the genetic structure and local adaptation of this species The fragmentation process has been shown to be correlated with genetic diversity and genetic structure in several empirical studies (Aguilar et al., 2008; Frankel & Soulé, 1981; Lienert, 2004; Nason et al., 1997) I measured the population genetic diversity and genetic differentiation using single nucleotide polymorphisms (SNPs) derived from whole-genome shotgun sequencing I found some populations were more genetically diverse than others, but none of the Banksia sessilis populations were genetically differentiated Furthermore, neither population sizes nor isolation distance has yet appeared to affect the genetic diversity of this species, which could be explained by sufficient gene exchange between fragments through highly effective bird pollinators and also long-distance seed dispersal I conclude that Banksia sessilis can be considered genetically resilient or have low sensitivity to fragmentation.

Significance and implications of this thesis

The results of this thesis indicate that the effect of fragmentation on the plant species in Dryandra Woodland over the past 60 years appears to depend on species-specific responses Although small and isolated fragments may still have the capability to maintain biodiversity of species comparable to the larger ones, to some extent the insect-pollinated species in small and isolated fragments have been affected by anthropogenic fragmentation in term of species fitness, whereas bird-pollinated species appear to be more resistant to the effects of fragmentation

Understanding how plant species respond to anthropogenic fragmentation is one of the keys to managing fragmented habitats with the aim of restoring or maintaining the biodiversity of the landscape The research presented here has indicated that small bushy shrubs such as

Gastrolobium parviflorum and Bossiaea eriocarpa are more sensitive to fragmentation compared to tree species which have less been affected by fragmentation These sensitive species need more protection and management in order to maintain or increase their population sizes in small and isolated habitats and to minimize the further negative impact of fragmentation Future research should also focus more on the effect of fragmentation on insect- pollinated shrub species to have sufficient data to allow optimal strategies for managing and conserving biodiversity in fragmented areas.

Recommendations for future study

Anthropogenic habitat fragmentation can have particular effects on those species with limited mobility, low range of seed dispersal and that heavily depend on a certain kind of pollinator, and these species may face a possibility of extinction in small fragments However, anthropogenic fragmentation of contiguous landscapes is often a relatively recent phenomenon and its long-term effects on species require more, and continuous, monitoring Therefore, research on the effects of anthropogenic fragmentation is a long-term process and requires the evaluation of many factors for a better understanding of anthropogenic fragmentation as well as effective management and conservation Based on the result of my study, I would propose future research in the areas below:

 Further surveys of floristic composition to monitor the fixed plots that have been surveyed in this study to detect any other changes in species diversity and composition over time under fragmented conditions Furthermore, species in the other fragments

166 should be surveyed in greater detail to have an overview of species diversity and composition in the area

 Further study on the growth and development of species under fragmented conditions should be undertaken with a variety of species including herbs, shrubs and trees from diverse families and with various life-forms

 Besides my survey of bird pollinators, further research into the availability and compositions of insect pollinators in the fragmented habitat should be undertaken The survey data of both insect and birds pollinators will help to detect any changes in pollinator availability and composition between fragment size and isolation distance that may be occurring

 Lack of effect of anthropogenic fragmentation on genetic diversity and population differentiation observed in this study would suggest that further research should take into broader spatial scales with different genetic approaches to detect genetic diversity and population differentiation of Banksia sessilis species, and, indeed, other species: genetic research on plants from diverse families and with various life-forms are recommended

 Further research should monitor the appearance of invasive species in the fragments and update the list of invasive species Research is also required to investigate the risk and impact of these species to allow a more precise guide for early warning, prevention and management of invasive species in fragmented habitats

References

Echeverría, C., Newton, A C., Lara, A., Benayas, J M R., & Coomes, D A (2007) Impacts of forest fragmentation on species composition and forest structure in the temperate landscape of southern Chile Global Ecology and Biogeography, 16(4), 426-439 Frankel, O., & Soulé, M E (1981) Conservation and Evolution: CUP Archive

Hooftman, D A., Van Kleunen, M., & Diemer, M (2003) Effects of habitat fragmentation on the fitness of two common wetland species, Carex davalliana and Succisa pratensis Oecologia, 134(3), 350-359

Jacquemyn, H., De Meester, L., Jongejans, E., & Honnay, O (2012) Evolutionary changes in plant reproductive traits following habitat fragmentation and their consequences for population fitness Journal of Ecology, 100(1), 76-87

Jennersten, O (1988) Pollination in Dianthus deltoides (Caryophyllaceae): effects of habitat fragmentation on visitation and seed set Conservation Biology, 2(4), 359-366

Kery, M., Matthies, D., & Spillmann, H H (2000) Reduced fecundity and offspring performance in small populations of the declining grassland plants Primula veris and

Gentiana lutea Journal of Ecology, 88(1), 17-30

Lienert, J., Diemer, M., & Schmid, B (2002) Effects of habitat fragmentation on population structure and fitness components of the wetland specialist Swertia perennis

L.(Gentianaceae) Basic and Applied Ecology, 3(2), 101-114

Nason, J., Aldrich, P., & Hamrick, J (1997) Dispersal and the dynamics of genetic structure in fragmented tropical tree populations Tropical forest remnants: Ecology, Management, and Conservation of Fragmented Communities, 304-320 ệckinger, E., Franzộn, M., Rundlửf, M., & Smith, H G (2009) Mobility-dependent effects on species richness in fragmented landscapes Basic and Applied Ecology, 10(6), 573-578 ệckinger, E., & Smith, H G (2006) Landscape composition and habitat area affects butterfly species richness in semi-natural grasslands Oecologia, 149(3), 526-534

Ross, K A., Fox, B J., & Fox, M D (2002) Changes to plant species richness in forest fragments: fragment age, disturbance and fire history may be as important as area

Table S2.1: GPS coordinates of the 63 plots surveyed within Dryandra Woodland in 2015

Table S2.2: Species identified from 66 plots during the 2015 survey at Dryandra Woodland

Family Genus Species Life form

Asteraceae Ursinia Ursinia anthemoides Herb

Asteraceae Lagenoflora Lagenoflora huegelii Herb

Asteraceae Pterochaeta Pterochaeta paniculata Herb

Asteraceae Brachyscome Brachyscome iberidifolia Herb

Asteraceae Olearia Olearia rudis Shrub

Asteraceae Hypochaeris Hypochaeris glabra Herb

Asteraceae Rhodanthe Rhodanthe citrina Herb

Asteraceae Podolepis Podolepis lessonii Herb

Asteraceae Helichrysum Helichrysum leucopcidium Herb

Asteraceae Rhodanthe Rhodanthe manglesii Herb

Asteraceae Waitzia Waitzia nitida Herb

Asteraceae Helichrysum Helichrysum leucopsideum Herb

Asteraceae Arcfotheca Arcfotheca celendula Shrub

Asteraceae Sonchus Sonchus asper Herb

Amaranthaceae Ptilotus Ptilotus manglesii Herb

Araliaceae Trachymene Trachymene pilosa Herb

Lomandra micrantha subsp micrantha Herb

Asparagaceae Lomandra Lomandra effusa Herb

Asparagaceae Laxmannia Laxmannia squarrosa Herb

Asparagaceae Thysanotus Thysanotus manglesianus Herb

Asparagaceae Chamaexeros Chamaexeros serra Herb

Asparagaceae Lomandra Lomandra collina Herb

Asparagaceae Thysanotus Thysanotus patersonii Herb

Asparagaceae Lomandra Lomandra micrantha Herb

Boryaceae Borya Borya sphaerocephala Herb

Boryaceae Borya Borya lacinata Herb

Campanulaceae Wahlenbergia Wahlenbergia multicaulis Herb

Campanulaceae Lobelia Lobelia heterophylla Herb

Caryophyllaceae Petrorhagia Petrorhagia dubia Herb

Caryophyllaceae Cerastium Cerastium glomeratum Shrub

Casuarinaceae Allocasuarina Allocasuarina campestris Shrub

Casuarinaceae Allocasuarina Allocasuarina humilis Shrub

Celastraceae Stackhousia Stackhousia pubescens Herb

Centrolepidaceae Centrolepis Centrolepis aristata Herb

Cyperaceae Lepidosperma Lepidosperma drummondi Herb

Cyperaceae Lepidosperma Lepidosperma leptostachyum Herb

Cyperaceae Mesomelaena Mesomelaena preissii Herb

Cyperaceae Mesomelaena Mesomelaena tetragona Herb

Cyperaceae Caustis Caustis dioica Herb

Cyperaceae Schoenus Schoenus curvifolius Herb

Dilleniaceae Hibbertia Hibbertia commutata Shrub

Dilleniaceae Hibbertia Hibbertia montana Shrub

Dilleniaceae Hibbertia Hibbertia exasperata Shrub

Dilleniaceae Hibbertia Hibbertia acerosa Shrub

Dilleniaceae Hibbertia Hibbertia hibbertioides Shrub

Dilleniaceae Hibbertia Hibbertia enervia Shrub

Hibbertia hibbertioides var pedunculata Shrub

Droseraceae Drosera Drosera macrantha Herb

Elaeocarpaceae Tetratheca Tetratheca virgata Shrub

Ericaceae Leucopogon Leucopogon fimbriatus Shrub

Ericaceae Styphelia Styphelia tenuiflora Shrub

Ericaceae Leucopogon Leucopogon polymorphus Shrub

Ericaceae Andersonia Andersonia bifida Shrub

Ericaceae Andersonia Andersonia caerulea Shrub

Ericaceae Astroloma Astroloma epacridis Shrub

Ericaceae Astroloma Astroloma pallidum Shrub

Euphorbiaceae Beyeria Beyeria leschenaultii Shrub

Fabaceae Bossiaea Bossiaea eriocarpa Shrub

Fabaceae Gastrolobium Gastrolobium trilobum Shrub

Fabaceae Acacia Acacia lasiocarpa var sedifilia Shrub

Fabaceae Gastrolobium Gastrolobium parviflorum Shrub

Fabaceae Gastrolobium Gastrolobium reticulatum Shrub

Fabaceae Dillwynia Dillwynia laxiflora Shrub

Fabaceae Acacia Acacia browniana Shrub

Fabaceae Gastrolobium Gastrolobium microcarpum Shrub

Fabaceae Acacia Acacia chrysocephala Shrub

Fabaceae Mirbelia Mirbelia dilatata Shrub

Fabaceae Acacia Acacia browniana var intermedia Shrub

Fabaceae Daviesia Daviesia decipiens Shrub

Fabaceae Acacia Acacia insolita subsp insolita Shrub

Fabaceae Hovea Hovea chorizemifolia Shrub

Fabaceae Daviesia Daviesia decurrens Shrub

Fabaceae Acacia Acacia pulchella var pulchella Shrub

Fabaceae Gompholobium Gompholobium cyaninum Shrub

Fabaceae Acacia Acacia acuminata Shrub

Fabaceae Gompholobium Gompholobium confertum Shrub

Fabaceae Gompholobium Gompholobium marginatum Shrub

Chorizema aciculare subsp aciculare Shrub

Goodeniaceae Dampiera Dampiera obliqua Shrub

Goodeniaceae Lechenaultia Lechenaultia formosa Shrub

Goodeniaceae Lechenaultia Lechenaultia biloba Shrub

Goodeniaceae Dampiera Dampiera lavandulacea Shrub

Goodeniaceae Goodenia Goodenia scapigera Herb

Goodeniaceae Velleia Velleia trinervis Herb

Haemodoraceae Conostylis Conostylis aculeata subsp aculeata Herb

Haemodoraceae Conostylis Conostylis setigera Herb

Haemodoraceae Haemodorum Haemodorum laxum Herb

Haloragaceae Glischrocaryon Glischrocaryon aureum Herb

Hemerocallidaceae Caesia Caesia parviflora Grass

Hemerocallidaceae Dianella Dianella revoluta Herb

Hemerocallidaceae Tricoryne Tricoryne elatior Herb

Hypericaceae Hypericum Hypericum gramineum Herb

Iridaceae Patersonia Patersonia occidentalis Herb

Lamiaceae Hemigenia Hemigenia incana Shrub

Myrtaceae Beaufortia Beaufortia incana Shrub

Myrtaceae Leptospermum Leptospermum erubescens Shrub

Myrtaceae Melaleuca Melaleuca seriata Shrub

Myrtaceae Verticordia Verticordia serrata Shrub

Myrtaceae Calytrix Calytrix leschenaultii Shrub

Myrtaceae Eucalyptus Eucalyptus albida Tree

Myrtaceae Melaleuca Melaleuca radula Tree

Myrtaceae Eucalyptus Eucalyptus wandoo Tree

Myrtaceae Eucalyptus Eucalyptus astringens Tree

Myrtaceae Eucalyptus Eucalyptus incrassata Tree

Myrtaceae Eucalyptus Eucalyptus rudis Tree

Myrtaceae Eucalyptus Eucalyptus accedens Tree

Myrtaceae Eucalyptus Eucalyptus marginata Tree

Myrtaceae Corymbia Corymbia calophylla Tree

Myrtaceae Melaleuca Melaleuca incana subsp incana Shrub

Orchidaceae Pterostylis Pterostylis hamiltonii Herb

Orchidaceae Pterostylis Pterostylis sanguinea Herb

Orchidaceae Caladenia Caladenia flava Herb

Orchidaceae Caladenia Caladenia falcata Herb

Orchidaceae Pterostylis Pterostylis picta Herb

Phyllanthaceae Phyllanthus Phyllanthus calycinus Shrub

Pittosporaceae Billardiera Billardiera variifolia Shrub

Poaceae Neurachne Neurachne alopecuroidea Herb

Poaceae Eragrostis Eragrostis curvula Grass

Poaceae Austrodanthonia Austrostipa elegantissima Herb

Poaceae Austrodanthonia Austrodanthonia setacea Herb

Poaceae Austrostipa Austrostipa hemipogon Herb

Poaceae Tetrarrhena Tetrarrhena laevis Herb

Poaceae Briza Briza maxima Grass

Polygalaceae Comesperma Comesperma volubile Herb

Primulaceae Lysimachia Lysimachia arvensis Weed

Proteaceae Banksia Banksia sphaerocarpa Shrub

Proteaceae Petrophile Petrophile seminuda Shrub

Proteaceae Hakea Hakea incrassata Shrub

Proteaceae Petrophile Petrophile ericifolia Shrub

Proteaceae Adenanthos Adenanthos cygnorum Shrub

Proteaceae Hakea Hakea prostrata Shrub

Proteaceae Grevillea Grevillea tenuiflora Shrub

Proteaceae Synaphea Synaphea interioris Shrub

Proteaceae Petrophile Petrophile striata Shrub

Proteaceae Petrophile Petrophile serruriae Shrub

Proteaceae Banksia Banksia proteoides Shrub

Banksia squarrosa sub Squarosa Shrub

Proteaceae Grevillea Grevillea leptobotrys Shrub

Proteaceae Banksia Banksia senssilis Shrub

Proteaceae Banksia Banksia nobilis Shrub

Proteaceae Banksia Banksia armata Shrub

Proteaceae Hakea Hakea undulata Shrub

Pteridaceae Cheilanthes Cheilanthes austrotenuifolia Herb

Restionaceae Desmocladus Desmocladus flexuosus Herb

Trymalium ledifolium var lineare Shrub

Santalaceae Santalum Santalum murrayanum Shrub

Stylidiaceae Stylidium Stylidium caricifolium Herb

Stylidiaceae Stylidium Stylidium leptophyllum Herb

Stylidiaceae Stylidium Stylidium piliferum Herb

Stylidiaceae Stylidium Stylidium amoenum Herb

Thymelaeaceae Pimelea Pimelea suaveolens Shrub

Thymelaeaceae Pimelea Pimelea imbricata var piligera Shrub

Thymelaeaceae Gnidia Gnidia pinifolia Shrub

Xanthorrhoeaceae Chamaescilla Chamaescilla corymbosa Herb

Xanthorrhoeaceae Chamaescilla Chamaescilla spiralis Herb

Xanthorrhoeaceae Xanthorrhoea Xanthorrhoea drummondii Grass

Xanthorrhoeaceae Xanthorrhoea Xanthorrhoea preissii Grass

Zamiaceae Macrozamia Macrozamia riedlei Tree

Flora collection licence and permission

Table S3.1: GPS coordinates of the area where Banksia sessilis, Gastrolobium parviflorum and Bossiaea eriocarpasamples were collected

180 seedlings (30 seedlings per fragment) of each species were sown into 180 PVC tubes with

5 cm diameters and 70 cm height All seedlings of three species were grown in June 2017 in a greenhouse at Curtin University, Perth, Western Australia, seedlings were watered 100 ml every day in the first ten days and then two days for a year thereafter to allow them to establish The temperature of the greenhouse was monitored by using evaporative cooling systems to cool down the temperature in the greenhouse during the hot summer days in 2018

After a year of growing in the greenhouse, plants were harvested and washed to remove soils and then drained naturally Then the growth indices were measured including the total number of leaves per plant, plant height (measured from root collar to terminal shoot), root length (to the tip of the largest root), total wet biomass (including leaves, roots and shoots), wet shoot biomass (without root), wet root biomass (without shoot and leaves) Plants were oven-dried at 70 0 C for 2 days (48 hours), after which total dry biomass was recorded, root to shoot ratio was calculated by dividing dry root mass to dry shoot mass and moisture content was obtained as total wet biomass – total dry biomass

Tiêu đề	Ecology and Evolutionary Adaptation in Anthropogenically Fragmented Habitat
Tác giả	Ha Huy Hoang
Người hướng dẫn	Assoc. Prof. Bill Bateman, Dr. Tianhua He, Dr. Paul Nevill
Trường học	Curtin University
Chuyên ngành	Ecology
Thể loại	thesis
Năm xuất bản	2022
Thành phố	Perth

Định dạng
Số trang	203
Dung lượng	3,95 MB