Moss distribution along a rainfall gradient in semi-arid, non-Eucalypt woodland of Victoria, Australia by Christine Lorraine Tyshing BA, BSc (Hons) Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy Deakin University December, 2013 This thesis is dedicated to my parents Arthur and Nancy Hartshorne (dec.) and my aunt Alene (dec.), and also to my brother Darryl, who collectively, imbued me with a love for the natural world, and lovingly gave me the tools to see it. III Acknowledgements First, I would like to acknowledge and sincerely thank my supervisor, Dr Mary Gibson very much for her guidance, support and friendly encouragement throughout the different phases of the research and thesis preparation. I would also like to thank the staff at Deakin University (Burwood Campus), including, Michael Holmes, Clorinda Schofield, Linda Moon, the HDR section and the librarians. A big thank you goes to Khanh van Tranh who taught me how to use the Sputter-coater and Scanning Electron Microscope during my honours year, and in so doing, opened my eyes to another, amazing world full of intricate detail and beauty. My time at Deakin University has been enriched by those whom I have met there, including Cuong van Huynh, Chantal Carrigan, Aaron Floyed, Melanie Birtchnell. Bernadette Sinclair, Sharon Morely, Emma Carlos and Sarah Bradbury. Thanks also go to Tony Smith who accompanied me on two fieldtrips to Wyperfeld National Park. My family has provided me with their love, encouragement and great support during my rather long PhD candidature which has been punctuated with the events that life throws at you. My mother always believed in me and was very encouraging, and would often ask me how the moss identification etc was going. My two children have been great to discuss computer hassles with and my son Nicholas gave my studies a great start by providing, setting-up and maintaining my computer system for me. My daughter Anna has patiently shared many study related discussions with me and has been a great help with the formatting of the thesis. Christopher and Narelle have given me welcome encouragement, as have my brother Darryl, his wife Phyllis, and my cousins Wendy and Irene, while Jenny and Ross were instrumental in my ability to use the Wimmera/Mallee dust storm photograph. I have been very lucky to have you all on my team, and I thank you all very much. Special thanks go to my husband, Ray, who took care of things whilst I was away on extended field-trips. He has also had many opportunities during my candidature to IV refine his culinary skills to a high level of expertise, and he has also been a willing proof reader. I would also like to thank Dr Niels Klazenga (National Herbarium of Victoria) and Mr Graham Bell (State Herbarium of South Australia) for their technical support during my early candidature. Thanks also go to Parks Victoria for permission to carry out research in the amazing and beautiful Murray-Sunset National park, Wyperfeld National Park, and the Little Desert National Park, and to the Parks Victoria staff who were always helpful with my enquiries. I would also like to thank Mr. Nick Bade for permission to use the photograph he took and supplied of the Mallee/Wimmera dust storm which enveloped Melbourne on February 8th, 1983. My study-buddy and companion, the ancient cat, ‘Tatty’, lived to see the last chapter written, but only just. V Preface During the preparation of this thesis there have been some changes in the nomenclature for some mosses, e.g. the thesis contains two species, Microbryum brevicaule (Taylor) R.H. Zander and Microbryum starkeanum (Hedw.) R.H. Zander. Microbryum brevicaule now is known as Microbryum starkeanum; however, both names are maintained in the species list for this thesis. Other changes include: Barbula crinita Schultz = Pseudocrossidium crinitum (Schultz) R.H. Zander Barbula hornschuchiana Schultz = Pseudocrossidium hornschuchianum (Schulz) R.H.Zander Bryoerythrophyllum binnsii (R.Br.bis) Wijk & Margad. = Bryoerythrophyllum dubium (Schwägr.) P.Sollman Ceratadon purpureus (Hedw.) Brid. subsp. convolutus (Reichardt) Burley = Ceratadon purpureus (Hedw.) Brid. Tortula antarctica (Hampe) Wilson = Syntrichia antarctica (Hampe) R.H. Zander Tortula pagorum (Milde) De Not. = Syntrichia laevipila Brid. Tortula papillosa Wilson = Syntrichia papillosa (Wilson) Jur. These changes are listed in AUSMOSS, the Online Catalogue of Australian Mosses by the Royal Botanic Gardens, Melbourne, Victoria, which is available at: http://www.rbg.vic.gov.au/dbpages/cat/index.php/mosscatalogue/search The nomenclature followed in this thesis is that of Streimann and Klazenga (2002) for all species except those in Bryaceae, for which Spence and Ramsay (2000) was used. The list of species and authors is provided in Appendix 1 at the back. Four of the chapters in this thesis have been prepared as stand-alone papers for publication, so there is a degree of repetition, especially in parts of the Methods sections. To avoid further repetition, all references are placed together at the end of the thesis. VI All photographs, micrographs and figures are by Chris Tyshing except for the photograph of the Wimmera/Mallee dust storm over Melbourne, which was taken by Nick Bade in 1983. VII Abstract Moss distribution along a rainfall gradient in semi-arid, non- Eucalypt Woodland of Victoria, Australia. This thesis examines the distribution of moss assemblages of non-Eucalypt woodland along a rainfall gradient in semi-arid natural parkland of north-western Victoria, Australia to determine: 1. whether the moss flora showed a sequential change in species assemblage along the rainfall gradient and to identify the environmental factors that explained this; 2. the habitat exploitation of the moss flora along the rainfall gradient and to determine the environmental factors that explain this; 3. whether life-form, leaf architecture or life strategies contribute to the explanation of moss distribution along the rainfall gradient. Seventy-four species were identified with those of the Pottiaceae and Bryaceae dominating as is characteristic of arid and semi-arid regions across the globe. Seriation of moss assemblages occurred at the site level with rainfall, areal actual evapotranspiration, average maximum summer temperature and pH being pivotal factors determining moss distribution. This suggested the moss assemblages could be used as indicators of climate change within non-Eucalypt Woodland, and could be monitored to show which species were disappearing, which were surviving and whether migration of species occurred as conditions became hotter and drier. Examination of the moss flora at the habitat level also showed variation along the rainfall gradient within each habitat but soil texture also assumed a significant proportion of the explanation of moss distribution, suggesting habitat amelioration of conditions. Three habitats were examined: soil exposed to the sun, shaded soil and tree trunks. Soil exposed to the sun had greatest species richness while epiphytic species were very few. VIII The contribution of leaf architecture and life-form to distribution of moss assemblages was investigated. Both life-form and leaf architecture contributed to the explanation of the distribution of moss assemblages along the rainfall gradient with affiliation of specific adaptations occurring. Short turfs or small tufts were most common in the wetter regions while herd species were most prevalent in the driest regions, although both occurred throughout the length of the rainfall gradient. The most important adaptations were concave leaves, dry leaves that curve towards the stem, short turfs or small tufts, papillae/mammillae, dry leaves that twist and/or overlap, herds, specialised cells on borders of leaves and incrassate cells. Each moss was assigned a life strategy and colonists were found to dominate along the length of the rainfall gradient although annual shuttles also formed a strong component. Life strategies did not explain the distribution of moss assemblages along the rainfall gradient although other studies found life strategies were environmentally driven. In the process of examining life strategies, a novel method of spore dispersal was identified for Goniomitrium acuminatum Hook.f. & Wilson. Studies that examine the ecology of the moss flora of native vegetation in semi-arid Victoria, Australia are limited, other than what has been completed for the rangelands. The work presented in this thesis, therefore, fills an important knowledge gap and is all the more pertinent in this current time of climate change, particularly as much of Australia is arid and semi-arid. IX