Aim Riparian ecosystems are regarded as vulnerable to the effects of climate change. Because of their reliance on passive dispersal to migrate from areas where conditions have become unfavourable, plants are particularly susceptible. On dryland river floodplains, the species diversity of herbaceous annuals is often high while that of structurally dominant woody perennials is low. We examined gene flow genetic structure and dispersal in Acacia stenophylla, a small perennial tree widely distributed throughout river systems of inland Australia. The role of the river corridor in shaping patterns of gene flow and genetic structure is also investigated. Location Murray‐Darling Basin, south eastern Australia Methods A total of 127 individuals, from 12 subpopulations located on seven rivers were genotyped at 13 microsatellite loci. Several population and landscape genetic tools were applied to the microsatellite data to evaluate spatial patterns of gene flow and genetic structure and make inferences regarding possible modes of dispersal. Results High gene flow and weak genetic structure was identified for the 12 subpopulations of A. stenophylla sampled, a surprising result given large distances between subpopulations. Pairwise genetic distance between subpopulations was low to moderate and could largely be explained (R2 = 0.68) by two variables: distance along the river and the proportion of no flow days. structure analysis revealed two genetic clusters. Subpopulations located on the Darling and Lower Balonne rivers were dominated by cluster one while subpopulations from the Warrego and Paroo rivers showed largely mixed ancestry with individuals descending from both clusters one and two. Main Conclusions These results indicate that the river corridor facilitates extensive gene flow between subpopulations of A. stenophylla in this system. Hydrochory appears to be the dominant process; however, upstream movements of propagules most probably via animal movement are sufficient to negate effects expected under unidirectional dispersal.

6 7 Duma florulenta and Acacia stenophylla are two ecologically important but 8 understudied species that naturally occur on the floodplains and riverbanks of Australia's 9 arid and semi-arid river systems. This paper describes the discovery and characterization 10 of 12 and 13 polymorphic microsatellite markers for D. florulenta and A. stenophylla 11 respectively. The number of alleles per locus for D. florulenta ranged from 2-12 with an 12 average of 6.1. Across all samples, observed and expected heterozygosities ranged from 13 0.026 to 0.784 and 0.026 to 0.824, respectively and mean polymorphic information 14 content was equal to 0.453. For A. stenophylla, the number of alleles per locus ranged 15 between 2 and 8 with an overall mean of 4.8. Across all samples, observed and expected 16 heterozygosities ranged from 0.029 to 0.650 and 0.029 to 0.761, respectively and mean 17 polymorphic information content was 0.388. The developed suites of 12 and 13 18 microsatellite markers for D. florulenta and A. stenophylla, respectively provide 19 opportunity for novel research into mechanisms of gene flow, dispersal and breeding 20 system and how they operate under the extreme variability these species are exposed to 21 in the environments in which they live. 22 23 PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.27027v1 | CC BY 4.0

Aim The conservation of plant species biodiversity has been identified as a crucial factor for the resilience of dryland ecosystems in the face of climate change and desertification. Duma florulenta (lignum) is a keystone species that facilitates biodiversity in the floodplains and wetlands of Australia's dryland river systems. This paper explores spatial genetic structure of lignum and investigates factors influencing dispersal and gene flow within and among river catchments of the northern Murray–Darling Basin. Location Northern Murray–Darling Basin, eastern Australia. Methods A total of 122 individual plants from subpopulations located on rivers in four adjacent catchments were genotyped using 10 microsatellite markers. Microsatellite data were then analyzed using population genetic techniques to evaluate levels of gene flow and genetic structure and identify factors influencing dispersal. Results Results suggest high levels of gene flow between lignum subpopulations of the northern Murray–Darling Basin. AMOVA revealed small but significant differences between subpopulations, and STRUCTURE analysis did not detect meaningful structure when sampling information was not provided. However, when sampling information was supplied using the LOCPRIOR model, three genetic clusters were identified. All Lower Balonne subpopulations were assigned to cluster 1 while a number of the other subpopulations showed mixed ancestry. Weak relationships were identified between pairwise genetic distance and geographic as well as river distance, although the R 2 value of the former was only half that of the latter. Main conclusions Patterns of genetic variation suggest frequent long‐distance overland gene flow largely as a result of the movement of seeds via floodwater. Therefore, maintenance of natural variability in flow regime is key both to maintain conditions favorable to recruitment and to promote dispersal and gene flow across the landscape. However, given future climate change projections persistence may be more reliant on the species ability to endure long periods of drought between flood events.

Duma florulenta and Acacia stenophylla are two ecologically important but understudied species that naturally occur on the floodplains and riverbanks of Australia’s arid and semi-arid river systems. This paper describes the discovery and characterization of 12 and 13 polymorphic microsatellite markers for D. florulenta and A. stenophylla respectively. The number of alleles per locus for D. florulenta ranged from 2-12 with an average of 6.1. Across all samples, observed and expected heterozygosities ranged from 0.026 to 0.784 and 0.026 to 0.824 respectively and mean polymorphic information content was equal to 0.453. For A. stenophylla, the number of alleles per locus ranged between 2 and 8 with an overall mean of 4.8. Across all samples, observed and expected heterozygosities ranged from 0.029 to 0.650 and 0.029 to 0.761 respectively and mean polymorphic information content was 0.388. The developed suites of 12 and 13 microsatellite markers for D. florulenta and A. stenophylla respectively provide opportunity for novel research into mechanisms of gene flow, dispersal and breeding system and how they operate under the extreme variability these species are exposed to in the environments in which they live.