This study explores the efficacy of drone-acquired RGB images and the YOLO model in detecting the invasive species Siam weed (Chromolaena odorata) in natural environments. Siam weed is a perennial scrambling shrub from tropical and sub-tropical America that is invasive outside its native range, causing substantial environmental and economic impacts across Asia, Africa, and Oceania. First detected in Australia in northern Queensland in 1994 and later in the Northern Territory in 2019, there is an urgent need to determine the extent of its incursion across vast, rugged areas of both jurisdictions and a need for distribution mapping at a catchment scale. This study tests drone-based RGB imaging to train a deep learning model that contributes to the goal of surveying non-native vegetation at a catchment scale. We specifically examined the effects of input training images, solar illumination, and model complexity on the model’s detection performance and investigated the sources of false positives. Drone-based RGB images were acquired from four sites in the Townsville region of Queensland to train and test a deep learning model (YOLOv5). Validation was performed through expert visual interpretation of the detection results in image tiles. The YOLOv5 model demonstrated over 0.85 in its F1-Score, which improved to over 0.95 with improved exposure to the images. A reliable detection model was found to be sufficiently trained with approximately 1000 image tiles, with additional images offering marginal improvement. Increased model complexity did not notably enhance model performance, indicating that a smaller model was adequate. False positives often originated from foliage and bark under high solar illumination, and low exposure images reduced these errors considerably. The study demonstrates the feasibility of using YOLO models to detect invasive species in natural landscapes, providing a safe alternative to the current method involving human spotters in helicopters. Future research will focus on developing tools to merge duplicates, gather georeference data, and report detections from large image datasets more efficiently, providing valuable insights for practical applications in environmental management at the catchment scale.

Hedge cactus (Cereus uruguayanus R. Kiesling; syn.: Cereus hildmannianus K. Schum.) is a columnar cactus that was introduced to Australia as an ornamental plant and has since become invasive in subhumid regions of Queensland and New South Wales. Compared with its congener, queen of the night (Cereus jamacaru DC.), which is currently invasive in both eastern and southern Africa, information on seed biology of C. uruguayanus is lacking. Experiments were conducted to study the effects of alternating day/night temperature, salt stress, water stress, and burial depth on germination and seedling emergence of four seed accessions of C. uruguayanus. Seeds were also subjected to a controlled aging test (CAT) to obtain an estimate of potential persistence under field conditions. The optimum temperature regime for germination of all accessions was 30/20 C. Germination decreased with an increase in sodium chloride (NaCl) concentration, but germination of all accessions (range 26% to 81%) occurred at 160 mM NaCl, indicating very high salt tolerance. Seed germination gradually decreased with an increase in water stress, but germination in all accessions (range 19% to 47%) occurred at −0.8 MPa. Seed viability and dormancy status were unaffected by exposure to salt level (320 mM NaCl) and water (−1.6 MPa) stress under which germination did not occur. Germination responses to all three factors were generally similar to those documented for C. jamacaru. The emergence of C. uruguayanus decreased with an increase in seed burial depth. The highest emergence (43%) was recorded for surface-sown seeds, and emergence was reduced to 0 at a burial depth of 2 cm. CAT results for two seed accessions indicated that seeds of C. uruguayanus are likely to demonstrate extended (>3 yr) persistence under field conditions, a prediction that is supported by evidence that germination of its small (2-mm) seeds is markedly reduced by burial.

Azadirachta indica (Neem tree) has become widely naturalised and invasive across many countries and regions including northern Australia. To aid management of A. indica where it has become a weed, a series of studies were undertaken to determine its potential soil seed bank persistence. In a field trial, packets of seeds were buried, retrieved periodically over two years and the seed viability assessed. Viability declined rapidly, with a single viable seed retrieved after 12 months burial and none thereafter. Burial depth, soil type, and pasture cover (present and excluded) significantly influenced viability (%) at 3- and 6-month retrievals. Similar data were obtained from repeated runs of a controlled ageing laboratory experiment, which categorized seeds as forming a ‘transient’ seed bank. In a third trial, fresh fruits were placed on the soil surface in replicated field enclosures over two consecutive years and seedling emergence monitored fortnightly. In both years there was no emergence from pasture excluded soil plots and emergence ceased after 2.3 and 8.4 months in plots with pasture present. A fourth (glasshouse) trial found most seeds will emerge from the soil when buried from 1 to 4 cm. However, more fatal germination than successful emergence was recorded for seeds buried at 8 cm. Seed desiccation and fatal germination are factors in A. indica developing a transient soil seed bank, and infestations require shorter-term control programs where seed input is prevented.

Parthenium hysterophorus L. is highly invasive across many countries, impacting agriculture, human and animal health and the environment. Despite a widespread invasive range, the persistence of P. hysterophorus seeds (achenes) in the soil profile has been identified as a gap in current knowledge. A long-term seed persistence field trial was under-taken, where packets of P. hysterophorus seeds were buried, retrieved periodically over 10 years and the viability assessed. Time, burial depth and soil type (clay or loam) significantly influenced seed viability. Overall viability was <5% after 5 years, and <0.2% of viable seed was recovered after 7 to 10 years. The decline was faster in seeds at 0 cm (on the soil surface) and buried at 20 cm and slower in seeds buried at 2.5 cm and 10 cm. At most retrieval times viability was higher in seeds retrieved from clay soil plots than from loam soil plots. Pasture cover (present or excluded) did not significantly influence the viability of seeds. Data from controlled ageing laboratory experiments indicated persistent seeds and complemented the field trial data. Based on the weight of 1000 seeds, less than 20 mm of rainfall could incorporate 50% of seeds into three soil types. Parthenium hysterophorus develops a persistent seed bank when incorporated into the soil and infestations will require longer-term control even if seed input is prevented.