Copyright and use of this thesis This thesis must be used in accordance with the provisions of the Copyright Act 1968. Reproduction of material protected by copyright may be an infringement of copyright and copyright owners may be entitled to take legal action against persons who infringe their copyright. Section 51 (2) of the Copyright Act permits an authorized officer of a university library or archives to provide a copy (by communication or otherwise) of an unpublished thesis kept in the library or archives, to a person who satisfies the authorized officer that he or she requires the reproduction for the purposes of research or study. The Copyright Act grants the creator of a work a number of moral rights, specifically the right of attribution, the right against false attribution and the right of integrity. You may infringe the author’s moral rights if you: - fail to acknowledge the author of this thesis if you quote sections from the work - attribute this thesis to another author - subject this thesis to derogatory treatment which may prejudice the author’s reputation For further information contact the University’s Director of Copyright Services sydney.edu.au/copyright
appendices
Appendices
Appendix 1
Primer of networks – a guide to the usage of, and terminology
describing networks in this thesis
A network (or interaction network/web) is a graphical representation of a collection of species and the interactions among them. Species can be referred to as nodes and interactions as links. Links refer the presence of an interaction and does not take into account the frequency of interaction.
There are several possible variations and different ways to portray a network of interactions:
• Interaction matrix: an adjacency matrix where plant species represent columns and animal species represent rows. The cell where a plant and animal species intersect can be marked by a ‘1’ to indicate the presence of and interaction and a ‘0’ to indicate no interaction. Alternatively, the cell can represent the number of interactions or a measure of the strength of the interaction. • Bipartite network (two-mode network): a network where only interactions occurring between plants and animals are considered. I.e. if animal b pollinates plant x, then b and x will be connected in a bipartite network. • Unipartite network (one-mode network): a network where only interactions occurring among plants or among animals are considered. A bipartite network can be collapsed into a unipartite network that portrays links among species that share a common interaction partner. I.e. if plant species x and y share pollinator species b, then x and y will be connected in a uni-partite network. • Quantitative/weighted network: a network where links are given a weight. In flower-visitor networks it refers the frequency of interaction. In pollen-transport networks it refers to the amount of pollen transported.
226 Appendices
• Qualitative/presence-absence/binary/unweighted network: a network where only the presence/absence of a link is shown.
Three different types of interactions are considered as networks in this thesis:
• Pollination network: links represent pollination interactions among species • Flower-visitor network: links represent flower-visitor interactions among species • Pollen-transport network: links represent the transport of a pollen species by an animal species.
On the ways to describe the patterns of species interactions in a network:
• Network function: as defined by the interaction type. The function of a pollination network is the movement of gametes effecting pollination and the acquisition of resources by visitors. • Network structure/topology: refers to the patterns (or geometric layout) in a network due to the way species interact with each other. • Network structural property: a specific structural pattern. E.g. nestedness • Network Indices: The variable to describe, characterize or measure some feature of network structural property. Synonymous with network metric and network statistic. See Dormann et al. (2009) for a more comprehensive list. • Forbidden link: two species that cannot interact due to trait incompatibility or spatio-temporal mismatch.
Network indices:
• P: plant species richness • A: animal visitor species richness • Network size (S): combined total of plant and animal species • I: total number of interactions • Iap: number of interactions between animal species a and plant species p. Also referred to as interaction strength. • Species degree: the number of links of a species
227 Appendices
Indices based on unweighted links (qualitative networks)
• Links per species: total number of links in a network divided by network size (L = I/S) • Connectance (C): proportion of realized links (C = AP) (Jordano 1987). • Nestedness (N): Tendency of specialised species to interact with a subset of the interaction partners of more generalised species (Bascompte et al. 2003). Nestedness implies both a high occurrence of degree asymmetry and the existence of a ‘generalised network core’ (relatively small group of highly connected species). A nestedness value of 1 equals perfect nestedness, 0 equals the opposite of nestedness of which there are several distinct patterns (Almeida-Neto et al. 2007), and randomness falls somewhere in between. • Degree asymmetry: the imbalance in the degree between two interacting species (Vazquez and Aizen 2004). • Degree distribution: frequency histogram of the degree of species in a network (Jordano et al. 2003). • Truncated power-law: a degree distribution of pollination networks are usually truncated power law which indicates that most species have few links and few species have many links, more than would be expected by chance. • Modularity/compartmentalisation: The existence groups of species with many intragroup links and few intergroup links (Guimera and Amaral 2005, Olesen et al. 2007). Modularity ranges from 0 (no modularity) and 1 (high modularity).
Indices based on weighted links (quantitative networks)
• Dependence: an estimate of the extent to which a species depends on another species (Jordano 1987, Bascompte et al. 2006). • Interaction strength asymmetry/dependence asymmetry: quantifies the imbalance in the interaction strength/dependence of two interacting species (Bascompte et al. 2006). Positive values indicate a higher specialisation of animals negative values indicate a higher specialisation of plants.
228 Appendices
• H2’ specialisation: a network-level measure of specialisation based on the deviation of a species’ realised number of interactions and that expected from each species’ total number of interactions
(Bluthgen et al. 2006). The H2’ ranges between 0 (no specialisation) and 1 (perfect specialisation). • d’ specialization: a species-level measure of specialisation which expresses how specialised a given species is in relation to what species are on offer (Bluthgen et al. 2006). Species-level equivalent
of H2’ specialisation. The d’ ranges between 0 (no specialisation) and 1 (perfect specialisation). • Generality: mean number of plant species visited per animal species (Brauman and Daily 2008, Dormann et al. 2009). See Dormann et al. (Dormann et al. 2009) for equation. Increasing value indicates a visitor species is visiting an increasingly diverse set of plants. • Vulnerability: mean number of animal species per plant species. Analogous to generality. • Evenness: a measure of the skewness in the distribution of interaction frequencies (Tylianakis et al. 2007). Also know as interaction evenness.
Processes driving network structure:
• Interaction neutrality: Occurrence of species interactions resulting from the random encounter of individuals, so that all individuals have the same probability of interacting regardless of identity (Vázquez et al. 2007, Krishna et al. 2008). • Trait-mediated: the likelihood of an interaction between two species depends on their phenotypic match. There two types of trait- mediated processes: trait complementarity, where the traits of two species correspond (e.g. a visitors nectar requirement matches that offered by a flower); and, exploitation barrier, where traits preclude interactions from happening (e.g. a long corolla length exclude short tongued visitors). Exploitation barrier can be a form of forbidden link.
Species role: the characterisation of a species depending on the pattern of their intra-module and inter-module links (Guimera and Amaral 2005). First defined by Guimera and Amaral (2005) using species ‘within module degree’ (z)
229 Appendices and ‘participation coefficient’ (P). Species with a similar pattern of links in the network are allocated the same role. Species role communicates the importance of a species in a network: how many other species rely on them, and what sort of effects to the network can be expected if they are lost. Four roles are defined, in this thesis:
• Hub: species with many links especially within their own module, but can have many outside of their module as well (z ≥ 2.5, 0 ≤ P ≤ 1). • Connector: network ‘glues’, linking many different modules together (z ≤2.5, P ≥ 0.62). • Peripheral: species with several links mainly within their own module, but can also link to another module (z ≤ 2.5, 0.05 < P < 0.62). • Ultra-peripheral: species often with only one or two links but only within their own module (z ≤0 2.5, P ≤0.05).
230 Appendices
231 Appendices
Appendix 2
A2-1. Abundance of invertebrates caught visiting flowers of each plant species using net sampling at each site and during each sampling period. Plant names follow the Australian Plant Name Index.
Jun-10 Nov-10 Jul-11 Total Total species abundanc Family Plant species FR KS MC FR KS MC FR KS MC richness e Aizoaceae Trianthema pilosa 26 59 3 1 63 152 35
Amaranthaceae Alternanthera sp. A 2 2 1
Amaranthaceae Ptilotus atriplicifolius 3 12 15 5
Amaranthaceae Ptilotus latifolius 1 15 2 34 15 2 69 27
Amaranthaceae Ptilotus polystachyus 7 11 18 12 48 11
Araliaceae Trachymene glaucifolia 103 91 120 13 19 47 393 70
Asteraceae Asteraceae sp. A 1 1 1
Asteraceae Calotis erinacea 9 1 129 1 37 6 183 35
Asteraceae Calotis plumothera 2 2 2
Asteraceae Helichrysum sp. A 2 1 3 3
Asteraceae Helipterum molle 6 25 31 6
Asteraceae Myriocephalus sp. A 1 1 1
Asteraceae Pterocaulon sphacelatum 31 52 83 21
Boraginaceae Halgania cyanea 7 7 6
Boraginaceae Trichodesma zeylanicum 19 1 11 7 27 55 21 18 159 28
Brunoniaceae Brunonia australis 6 13 19 12
Chenopodiaceae Chenopodiaceae sp A 6 6 1
Chenopodiaceae Salsola kali 1 1 1
Euphorbiaceae Adriana tormentosa 8 8 2
Euphorbiaceae Euphorbia drummondii 74 57 132 70 88 220 130 35 69 875 132
Fabaceae Acacia bivenosa 19 22 24 65 17
Fabaceae Acacia coriacea 3 3 3
Fabaceae Acacia dictyophleba 50 39 19 34 32 174 25
Fabaceae Acacia ligulata 60 2 138 27 93 320 43
Fabaceae Acacia stenophylla 1 1 1
Fabaceae Crotalaria cunninghamii 2 4 34 38 84 162 23
Fabaceae Crotalaria eremaea 1 10 8 10 84 22 28 30 193 61
232 Appendices Fabaceae Crotalaria smithiana 4 12 16 1
Fabaceae Fabaceae sp. A 3 3 2
Fabaceae Psoralea eriantha 2 2 2
Fabaceae Senna artemisioides 2 2 1
Fabaceae Senna pleurocarpa 9 2 11 8
Fabaceae Swainsona phacoides 1 2 3 2
Fabaceae Tephrosia rosea 5 4 65 44 42 1 9 170 30
Frankeniaceae Frankenia gracilis 7 7 4
Goodeniaceae Goodenia cycloptera 24 14 12 28 29 26 17 14 7 171 30
Goodeniaceae Lechenaultia divaricata 10 10 4
Goodeniceae Scaevola depauperata 3 1 21 82 8 47 1 163 29
Goodeniceae Scaevola parvibarbata 4 25 3 32 10
Goodeniceae Scaevola parviflora 19 5 1 25 11
Haloragaceae Haloragis gossei 4 4 1
Lamiaceae Dicrastylis costelloi 92 6 15 113 41
Lamiaceae Dicrastylis llewellynii 11 11 8
Lamiaceae Newcastelia cephalantha 16 122 138 26
Lamiaceae Newcastelia spodiotricha 146 172 318 36
Malvaceae Abutilon otocarpum 43 43 1
Malvaceae Rulingia loxophylla 1 1 1
Malvaceae Sida fibulifera 2 16 18 29 65 26
Malvaceae Sida trichopoda 5 2 7 6
Malvaceae Triumfetta winneckeana 3 68 71 10
Myrtaceae Eucalyptus pachyphylla 316 316 18
Portulacaceae Calandrinia balonensis 11 26 14 81 132 23
Proteaceae Grevillea juncifolia 8 1 6 15 6
Proteaceae Grevillea stenobotrya 221 61 3 285 37
Rubiaceae Oldenlandia pterospora 8 8 6
Sapindaceae Atalaya hemiglauca 1 1 1
Scrophulariaceae Eremophila macdonnellii 12 68 1 81 13
Scrophulariaceae Eremophila sp. A 2 2 2
Solanaceae Solanum sp A 10 3 13 4
Violaceae Hybanthus aurantiacus 2 2 2
Unknown 1 3 3 2
Total 407 248 411 636 1085 1067 451 479 436 5220 327 The five shades of cell colour indicate flowering intensity (darker shade equates to greater flowering intensity) of that species at that specific time and place.
233 A2-2. Rarefaction curves for pan trap (broken line) and net sampling (solid line). Transects within each dune zone at each location at each site were pooled and used to generate average estimates (n = 12, ± SE).
Appendices
A2-3. List of bee species and abundance sampled by pan traps and nets.
Family Species Net sampling Pan traps Apidae Amegilla chlorocyanea 19 55 Apidae Exoneurella eremophila 58 53 Apidae Thyreus warooensis 3 Colletidae Brachyhesma barrowensis 44 Colletidae Chrysocolletes strangomeles 2 Colletidae Euhesma loorea 1 Colletidae Euhesma sp. D 31 14 Colletidae Euhesma sp. J 9 Colletidae Euhesma sp. T 2 3 Colletidae Euhesma sybilae 18 1 Colletidae Hylaeus albimandibulatus 2 Colletidae Hylaeus albozebratus 65 13 Colletidae Hylaeus calandriniae 12 33 Colletidae Hylaeus circinatus 3 Colletidae Hylaeus sp. AB 1 Colletidae Hylaeus sp. AG 1 Colletidae Hylaeus sp. AH 3 Colletidae Hylaeus sp. E 3 1 Colletidae Hylaeus sp. F 75 1 Colletidae Hylaeus sp. H 3 11 Colletidae Hylaeus sp. K 9 Colletidae Hylaeus sp. L 1 Colletidae Hylaeus sp. Z 12 3 Colletidae Hylaeus trimerops 13 Colletidae Hylaeus vittatifrons 4 Colletidae Hylaeus wyndhamensis 26 Colletidae Hylaeus zebrinus 4 48 Colletidae Leioproctus alloeopus 7 Colletidae Leioproctus argentifrons 2 1 Colletidae Leioproctus finkei 13 39 Colletidae Leioproctus impatellatus 8 2 Colletidae Leioproctus lucidicinctus 63 8 Colletidae Leioproctus sequax 98 174 Colletidae Leioproctus sp. A 25 61 Colletidae Leioproctus sp. AD 2 Colletidae Leioproctus sp. AE 29 3 Colletidae Leioproctus sp. B 4 Colletidae Leioproctus sp. S 17 Colletidae Leioproctus sp. V 1 Colletidae Leioproctus sp. W 4 4 Colletidae Leioproctus tarsalis 3 Colletidae Xanthesma flava 16 Colletidae Xanthesma lucida 42
235 Appendices
Colletidae Xanthesma nigrior 3 Halictidae Homalictus sphecodopsis 33 Halictidae Homalictus urbanus 340 346 Halictidae Lasioglossum alacarinatum 105 97 Halictidae Lasioglossum albopilosum 2 24 Halictidae Lasioglossum ebeneum 5 Halictidae Lasioglossum immaculatum 217 41 Halictidae Lasioglossum platychilum 120 82 Halictidae Lasioglossum cognatum 27 Halictidae Lasioglossum eremaean 18 40 Halictidae Lipotriches australica 41 3 Halictidae Lipotriches flavoviridis 73 15 Halictidae Lipotriches semipallida A 4 Halictidae Lipotriches sp 89 3 Halictidae Lipotriches sp AC 3 Halictidae Lipotriches sp G 1 Megachilidae Coelioxys reginae 5 1 Megachilidae Megachile apicata 2 Megachilidae Megachile aurifrons 1 Megachilidae Megachile barvonensis 3 Megachilidae Megachile bidentis 2 Megachilidae Megachile boharti 2 Megachilidae Megachile captionis 2 3 Megachilidae Megachile giddioorla 11 Megachilidae Megachile maculariformis 3 Megachilidae Megachile sequior 17 1 Megachilidae Megachile serricauda 6 1 Megachilidae Megachile tarltoni 1 Megachilidae Megachile warrambuccis 1
236 Appendices
A2-4. Partially nested ANOVA results comparing species richness and total abundance between methods (log transformed).
Species richness Total abundance Source df MS F P MS F P Method 1 817.0 27.32 <0.05 1.634 38.00 <0.01 Trip 2 1662.9 53.64 <0.001 2.375 91.35 <0.001 Site 2 117.3 4.60 ns 0.314 9.81 <0.05 Dune 1 1094.5 243.22 <0.01 0.381 11.55 <0.05 Location(site) 3 25.5 1.42 ns 0.032 0.74 ns Method x Trip 2 366.8 33.35 <0.01 1.058 27.84 <0.01 Method x Site 2 44.6 1.49 ns 0.175 4.07 ns Method x Dune 1 5.1 0.11 ns 0.165 0.75 ns Method x Location(site) 3 29.9 1.67 ns 0.043 1.00 ns Trip x Site 4 179.8 5.80 <0.05 0.079 3.04 ns Trip x Dune 2 154. 8.24 <0.05 0.121 6.72 <0.05 Trip x Location(site) 6 31.0 1.73 ns 0.026 0.60 <0.05 Site x Dune 2 66.1 14.69 <0.05 0.124 3.76 ns Dune x Location(site) 3 4.5 0.25 ns 0.033 0.77 ns Method x Trip x Site 4 22.4 2.04 ns 0.215 5.66 <0.05 Method x Trip x Dune 2 42.3 1.51 ns 0.081 0.90 ns Method x Trip x Location(Site) 6 11.0 0.61 ns 0.038 0.88 ns Method x Site x Dune 2 42.3 0.95 ns 0.022 0.10 ns Method x Dune x Location(site) 3 44.5 2.49 ns 0.220 5.12 <0.01 Trip x Site x Dune 4 30.9 1.65 ns 0.033 1.83 ns Trip x Dune x Location(Site) 6 18.7 1.04 ns 0.018 0.42 ns Method x Trip x Site x Dune 4 3.2 0.11 ns 0.028 0.31 ns Method x Trip x Dune x Location(Site) 6 28.1 1.57 ns 0.090 2.09 ns Error 72 17.9 0.043
237 Appendices
A2-5. Species richness and abundance per transect (n = 24) of invertebrate orders (average ± SE). Hymenoptera split in bees, ants and wasps. Bees are grouped by Method x Trip x Site so mean values are from n = 8 transects.
Coleoptera
ptera Hemi
238 Appendices
A2-6. Pairwise comparisons for the significant ANOSIM method (a) x trip (b) x site (c) interaction comparing invertebrate assemblages collected using net sampling and pan traps.
Pairwise Test t P t P t P Method x Trip x Site a) Net, Pan FR KS MC Jun 10 3.344 <0.01 3.140 <0.01 2.483 <0.05 Nov 10 3.366 <0.01 2.913 <0.01 2.142 <0.05 Jul 11 2.979 <0.05 2.800 <0.01 2.160 <0.05 b) Net Jun 10, Nov 10 2.392 <0.05 2.777 <0.01 2.365 <0.05 Jun 10, Jul 11 2.594 <0.05 1.476 ns 1.997 ns Nov 10, Jul 11 2.758 <0.05 2.777 <0.05 2.055 <0.05 Pan Jun 10, Nov 10 2.192 <0.05 2.738 <0.01 3.961 <0.01 Jun 10, Jul 11 2.627 <0.05 2.597 <0.05 2.862 <0.05 Nov 10, Jul 11 2.649 <0.05 1.978 <0.05 2.382 <0.05 c) Net Jun 10 Nov 10 Jul 11 FR, KS 1.969 <0.05 2.403 <0.01 1.593 ns FR, MC 2.023 <0.05 1.800 <0.05 1.310 ns KS, MC 2.071 <0.05 1.779 <0.05 1.165 ns Pan FR, KS 3.525 <0.001 1.746 <0.05 1.784 <0.05 FR, MC 3.117 <0.01 2.628 <0.01 1.577 ns KS, MC 3.212 <0.01 1.961 <0.05 1.875 <0.05
239 Appendices
A2-7. Nonmetric multidimensional scaling (nMDS) ordination of assemblages of common invertebrate species. Solid symbols represent nets, hollow symbols represent pans, and shapes represent the sampling periods.
240 Appendices
A2-8. Comparison of assemblages of common invertebrate species sampled using PERMANOVA. Common invertebrates were arbitrarily defined as any species with a total abundance of 20 or greater of which there were 61 invertebrate species.
Source df MS Pseudo-F P(perm) Method 1 40968 24.90 <0.001 Trip 2 29644 12.16 <0.001 Site 2 14785 5.06 <0.0011 Dune 1 5127 2.29 ns Location(Site) 3 2919 1.77 <0.01 Method x Trip 2 23330 14.18 <0.001 Method x Site 2 11886 7.23 <0.001 Method x Dune 1 5444 2.46 <0.05 Trip x Dune 2 3314 1.60 ns Trip x Site 4 9534 3.91 <0.001 Site x Dune 2 3540 1.58 ns Trip x Location(Site) 6 2439 1.48 <0.01 Dune x Location(Site) 3 2238 1.36 ns Method x Site x Dune 2 4228 1.91 ns Method x Trip x Site 4 7351 4.47 <0.001 Method x Trip x Dune 2 2851 1.47 ns Method x Dune x Loc(site) 3 2213 1.35 ns Trip x Site x Dune 4 2508 1.21 ns Method x Dune x Loc(site) 6 2067 1.26 ns Method x Trip x Site x Dune 4 2948 1.52 ns Method x Trip x Dune x Lo(site) 6 1933 1.18 ns Residual 81 1645 1 = P-value determined using Monte Carlo sampling.
241 Appendices
Appendix 3
A3-1. Characteristics of the visitation, pollen-transport, and null pollen-transport networks when singletons were removed from the analysis.
Pollen-transport network Visitation Null model network Observed Mean (95% CI) Plant species 51 27 27.22 (22.00–32.00) Bee species 28 13 17.01 (13.00–21.00) Links 146 55 72.02 (64.00–80.00) Visits 654 95 95
Connectance 0.102 0.157 0.159 (0.123–0.210) Nestedness 0.885 0.807 0.857 (0.775–0.912)
H2' 0.501 0.783 0.469 (0.400–0.541) Vulnerability 3.323 1.577 4.111 (3.373–4.956) Evenness 0.864 0.736 0.975 (0.961–0.985)
Mean links per bee species 5.21 ± 2.28 4.23 ± 2.44 4.29 (3.45-5.43) Mean links per plant species 2.81 ± 0.55 2.43 ± 0.92 2.66 (2.24-3.17) Mean bee specialisation index d' 0.55 ± 0.06 0.67 ± 0.13 0.54 (0.46-0.62) Mean plant specialisation index d' 0.36 ± 0.07 0.44 ± 0.08 0.42 (0.35-0.49)
242 Appendices
A3-2. Examples of bees visiting various flowers. Clockwise from top left: Hylaeus zebrinus (Colletidae) visiting Calotis erinaceae (Asteraceae), bee = 5 mm; Hylaeus trimerops visiting Euphorbia dummondii (Euphorbiaceae), bee = 4 mm; Lasioglossum immaculatum (Halictidae) visiting Grevillea stenobotrya (Proteaceae), bee = 4 mm; Lipotriches flavoviridis (Halictidae) visiting Calandrinia balonensis (Portulacaceae) and Scaevola depauperata (Goodeniaceae), bee = 5 mm; Lasioglossum cognatum (Halictidae) visiting Wahlenburgia gracilis
(Campanulaceae), bee = 5 mm. Photos by Tony Popic.
243 Appendices
Appendix 4
A4-1: PERMANOVA results of main test assessing species composition differences among networks. Networks were sampled at the Location(Site) level during three periods (June 2010, November 2010, June 2011), and at three sites (FR, KS, MC – two Locations were nested within each site). Within each location there were also two dune zones (crest and swale).
Source df MS Pseudo-F P(perm) Time 2 25443 13.18 <0.01 Site 2 9567 3.96 <0.0011 Dune 1 29417 12.01 <0.05 Location(Site) 3 2416 1.33 <0.05 Time x Site 4 6002 3.11 <0.01 Time x Dune 2 8024 3.64 <0.01 Site x Dune 2 6131 2.50 ns Time x Location(Site) 6 1931 1.07 ns Location(Site) x Dune 3 2450 1.35 <0.05 Time x Site x Dune 4 3950 1.79 <0.05 Time x Location(Site) x Dune 6 2202 1.21 <0.05 Residual 36 1813 1 = P-value determined using Monte Carlo sampling due to insufficient permutations
A4-2: Pairwise tests of the significant Time x Site interaction.
Pairwise Test t P Time Jun 10, Nov 10 3.55 <0.05 Jun 10, Jun 11 3.38 <0.05 Nov 10, Jun 11 3.93 <0.05 Site FR, KS 2.23 <0.011 FR, MC 1.84 <0.011 KS, MC 1.90 <0.011 1 = P-value determined using Monte Carlo sampling due to insufficient permutations
244 Appendices
Appendix 5
A5-1: The realised roles of plant species and visitor species in all 18 flower-visitor networks grouped by roles. Adapted from Guimera and Amaral (2005).
hub Plants A 5 connector 4 peripheral ultraperipheral 3
2
1
0 z -within moduledegree -1
-2 0 B0.2 0.4 0.6 0.8 1 5 Visitors 4
3
2
1
0 z -within moduledegree -1
-2 0 0.2 0.4 0.6 0.8 1 P -par cipa on coefficient
245 Appendices
A5-2. The participation coefficient and within module degree were combined to produce one variable that described the importance of species in a network.
Toward the red colour (top right, values of up to 2) are hub and connector species, and toward the blue colour (bottom left, values of down to 0) are ultra-peripheral species.
C
246 A5-3. Rainfall, observed network dimensions, qualitative indices and quantitative indices for the 18 networks from spinifex grassland. Rainfall is the cumulative rainfall from the nine months before the network was sampled. See A1 for index details.
Observed network dimensions Qualitative Indices Quantitative Indices Interaction Visitor Plant strength Rainfall species species Network Links per Conn. Nested. Number Modularity asymmetry H2’ Time Site (mm) richness richness size Links Interact. species (0 – 1) (0 – 1) modules (0 – 1) (-1 – 1) (0 – 1) Generality Vulnerability Jun-10 FR1 447.2 33 12 45 64 176 1.42 0.16 16.0 6 0.48 0.18 0.48 2.71 6.09 Jun-10 FR2 447.2 51 15 66 98 231 1.48 0.13 12.2 6 0.51 0.07 0.44 3.26 6.88 Jun-10 KS1 360.4 38 9 47 55 130 1.17 0.16 15.2 6 0.53 0.40 0.67 1.74 10.59 Jun-10 KS2 360.4 35 12 47 54 118 1.15 0.13 11.4 8 0.55 0.10 0.65 1.90 6.52 Jun-10 MC1 311.4 34 13 47 58 217 1.23 0.13 9.2 8 0.50 0.16 0.43 2.66 6.67 Jun-10 MC2 311.4 38 13 51 60 194 1.18 0.12 9.3 6 0.54 0.13 0.45 2.28 6.92 Nov-10 FR1 654.2 67 20 87 110 251 1.26 0.08 9.5 9 0.61 0.07 0.65 2.00 6.63 Nov-10 FR2 654.2 70 19 89 119 385 1.34 0.09 9.6 10 0.56 0.21 0.56 2.55 7.34 Nov-10 KS1 589.2 63 20 83 116 599 1.40 0.09 15.5 10 0.56 0.23 0.60 2.79 6.00 Nov-10 KS2 589.2 67 26 93 119 486 1.28 0.07 8.6 12 0.61 0.14 0.65 2.37 6.23 Nov-10 MC1 526.2 94 24 118 176 634 1.49 0.08 8.8 9 0.55 0.20 0.54 3.13 8.86 Nov-10 MC2 526.2 73 15 88 121 433 1.38 0.11 12.3 7 0.60 0.27 0.67 2.31 8.30 Jun-11 FR1 519.2 39 11 50 60 261 1.20 0.14 12.2 6 0.55 0.28 0.40 2.91 6.05 Jun-11 FR2 519.2 47 10 57 69 190 1.21 0.15 16.6 6 0.56 0.35 0.62 2.21 7.71 Jun-11 KS1 577.8 46 13 59 70 199 1.19 0.12 19.3 8 0.60 0.09 0.56 2.96 4.77 Jun-11 KS2 577.8 62 14 76 88 280 1.16 0.10 11.5 10 0.63 0.30 0.57 2.49 7.90 Jun-11 MC1 393.0 34 12 46 52 256 1.13 0.13 11.4 6 0.58 0.11 0.61 2.58 4.47 Jun-11 MC2 393.0 38 14 52 58 180 1.12 0.11 13.0 8 0.61 0.13 0.58 3.14 4.26 A5-4: The observed number of plant and visitor species in each role across all 18 networks. The expected values are based on the null that the number of plant and visitor species in each role would be the same proportion as the total number of observed plants and visitors. Observed and expected numbers were compared using a chi-square test.
Observed plant Observed visitor chi- Network Role species (expected) species (expected) square d.f. Significance Hub 62 (14) 0 (48) 211.76 1 P < 0.001 Connector 14 (22) 81 (73) 3.39 1 P > 0.05 Peripheral 67 (51) 156 (172) 6.97 1 P < 0.01 Ultra-peripheral 129 (185) 692 (635) 22.54 1 P < 0.001 Total 272 929
Appendices
A5-5: The average number of bee species that visited zygomorphic plant species (black
2 symbols) increased with rain (F(1,17) = 32.53, P < 0.001, r = 0.65) but not with
2 actinomorphic species (F(1,17) = 0.73, P > 0.05, r = 0.00; red symbols; a). The average role of zygomorphic plant species increased with network bee species richness (F(1,17) =
2 38.77, P < 0.001, r = 0.69), but not for actinomorphic plant species (F(1,17) = 0.67, P > 0.05, r 2 = 0.00; a). Each symbol represents a different network.
A 6 a 5
4 Zygomorphic 2 3 r = 0.65 species 2
1 Average # bee species visi ng plant 0 200 300 400 500 600 700 9 month cumula ve rainfall
1.4 B b 1.2
1
0.8 Zygomorphic r2 = 0.57 0.6 Average role 0.4
0.2
0 0 5 10 15 20 25 30 Bee species richness
249 Appendices
A5-6: Relationship between flower size and the variation in visitor size. Lower flower sizes may restrict the size of visitors that can visit, but no difference in the variation in visitor size was observed as flower size changed.
6 Jun-10 Nov-10 5 Jun-11
4
3
2
1 Standard devia on of visitor size 0 0 0.5 1 1.5 2 2.5 Log(flower size)
A5-7: The distribution of visitor sizes varied among sampling periods. There was a peak in visitors between 3-5 mm for all three sampling periods. June and November 2010 also had a peak in visitors around 8-9 mm but June 2011 did not.
0.50
0.45 Jun-10 0.40 Nov-10 0.35 Jun-11 0.30 0.25 0.20
Propor on of total 0.15 0.10 0.05 0.00 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Visitor body size (mm)
250 Appendices
A5-8: Average role of plant species according to their nectar available pooled across networks. There was no significant difference in average role of plants with different nectar volumes (ANOVA, F(2,233) = 0.254, P = 0.78).
0.9 0.8 0.7 0.6 0.5 0.4 n=41 n=133 n=62
Average Role 0.3 0.2 0.1 0 Absent Glisten More than glisten
Available nectar
251 Appendix