Department of Agriculture and Food Western Australia
Department of Agriculture and Food, Western Australia (DAFWA): Research LibraryNot a member yet
10920 research outputs found
Sort by
Training sensor-agnostic deep learning models for remote sensing: Achieving state-of-the-art cloud and cloud shadow identification with OmniCloudMask
No full textDeep learning models are widely used to extract features and insights from remotely sensed imagery. However, these models typically perform optimally when applied to the same sensor, resolution and imagery processing level as used during their training, and are rarely used or evaluated on out-of-domain data. This limitation results in duplication of efforts in collecting similar training datasets from different satellites to train sensor-specific models. Here, we introduce a range of techniques to train deep learning models that generalise across various sensors, resolutions, and processing levels. We applied this approach to train OmniCloudMask (OCM), a sensor-agnostic deep learning model that segments clouds and cloud shadow. OCM demonstrates robust state-of-the-art performance across various satellite platforms when classifying clear, cloud, and shadow classes, with balanced overall accuracy values across: Landsat (91.5 % clear, 91.5 % cloud, and 75.2 % shadow); Sentinel-2 (92.2 % clear, 91.2 % cloud, and 80.5 % shadow); and PlanetScope (96.9 % clear, 98.8 % cloud, and 97.4 % shadow). OCM achieves this accuracy while only being trained on a single Sentinel-2 dataset, employing spectral normalisation and mixed resolution training to address the spectral and spatial differences between satellite platforms. This approach allows the model to effectively handle imagery from different sensors within the 10 m to 50 m resolution range, as well as higher resolution imagery that has been resampled to 10 m. The OCM library is available as an open source Python package on PyPI
Managing lupin sclerotinia: know your risk and how to respond
No full textSclerotinia prevalence, particularly basal Sclerotinia (ground level infection) has increased significantly in Western Australia in the last five years. Since 2020, commercial crop surveys have found basal Sclerotinia more common in lupins than in canola, and it is difficult to manage. From 2021–2025 extensive research was conducted in the laboratory, glasshouse and field trials by research partners Department of Primary Industries and Regional Development (DPIRD), Centre for Crop Disease Management (CCDM), Mingenew Irwin Group (MIG) and lupin growers. Key new findings applicable to lupin growers are: Sclerotinia infects narrow-leaf lupin via two pathways: canopy and basal. Basal infection is more damaging, causing around 60% yield loss on infected plants, while canopy infection can cause up to 25% yield loss, mainly by infecting the main spike pods. On a paddock scale, canopy infection usually causes around 10% yield loss, increasing to 25% in a growing season with a wet spring and/or high disease pressure. A model from field data has been developed which allows for yield loss estimation based on disease incidence. Sclerotia contamination of harvested grain varies widely, but field research has found it is more common in trials with higher incidence of canopy infection. No consistently effective management strategies to reduce sclerotia grain contamination have been identified. An integrated disease management strategy for Sclerotinia canopy infection is being developed using data from 2021–2024 field trials. The approach includes agronomic, cultural, and chemical control options such as risk assessment (paddock, crop and season factors), delayed sowing, wider row spacing, lower seed rates, and applying a registered foliar fungicide from full flower to early pod emergence on the main spike. Employing a disease management strategy is only likely to be necessary and profitable in high-risk scenarios outlined in the Lupin Sclerotinia disease risk assessment guide. Research shows management is more economically viable in the medium-high rainfall zones of the Geraldton port zone, followed by the Kwinana North port zone. In contrast, the Albany port zone had limited disease incidence from 2021–2024, resulting in few significant responses to management. Effective options for reducing basal Sclerotinia infection in lupins are currently limited, as foliar fungicide applied during crop flowering is often ineffective. Ongoing research aims to identify the drivers for basal infection and potential management strategies. Gather annual data on the distribution and impact of Sclerotinia (canopy and basal) in commercial lupin crops in Western Australia. The unique WA environment requires that local research be undertaken, as it differs from other cropping regions in Australia. Expand our understanding of the epidemiology and the infection process of Sclerotinia in lupin. Identify the growing seasons and paddock scenarios where Sclerotinia is likely to be problematic and determine when preventative actions are necessary and profitable. Improve understanding of how cultural practices, (e.g. crop rotation, row spacing, plant density and sowing time) influence disease development and determine the effectiveness and optimal timing of fungicides for managing both canopy and basal Sclerotinia infection
East Fremantle, Town of - BEN sign map – 1 of 1
No full textBeach Emergency Number (BEN) Signage for the Town of East Fremantlehttps://library.dpird.wa.gov.au/gis_bens/1062/thumbnail.jp
Identifying capacity limitations and training needs using a stock assessment game
No full textThe technical capability of stock assessment analysts, along with characteristics of their operating environment, often limits the development of suitable population dynamics models and affects the accuracy of estimated quantities used for fisheries management. Following a series of training workshops focused on the Stock Synthesis and Stock Assessment Continuum Tool packages, Australian stock assessment scientists were invited to participate in a hypothetical stock assessment “Game” to explore the repercussions , for assessment, of different levels of experience and technical capability in an informal “consequence-free” manner. A fishery data set was generated using a simulation model that represented a stock distributed over 12 regions and harvested by three fishing fleets. The simulation model was made complex by including spatial structure, time-varying selectivity for some fleets, and changes over time in expected recruitment due to the effects of an environmental driver. The analysts self-organized into six (mostly within-agency) groups and reported estimates of current biomass, current depletion and advice regarding the possibility of local depletion. The results of the Game were used to evaluate the approaches used by the various groups and to identify areas where future training would be most beneficial. The results highlighted opportunities for additional training in spatially-explicit population dynamics modelling, the use of methods for pre-processing monitoring data to select appropriate fleet and population structures, as well as the use of methods to provide values related to growth and natural mortality. The groups treated the Game more seriously than was originally intended by the organizers, with several analysts concerned that any errors or assumptions that were mis-matched with the simulated reality may have brought embarrassment to themselves and their agency. Care should therefore be taken that simulation experiments intending to foster collaboration and learning do so in an explicitly understood risk-free environment. Overall, the Game proved valuable in contributing to the development of an Australian community of practice for stock assessment and identifying how to strengthen assessment capabilities
Response of wheat to phosphorus-enriched ironstone gravel
No full textContext Gravel fractions ( \u3e 2 mm) in soil are almost always excluded from laboratory analysis and glasshouse experiments as they are considered to be inert; however, the \u3e 2 mm fraction is always present in field experiments. Aims To determine whether the \u3e 2 mm fraction of ironstone gravel (IG) soil enriched with phosphorus (P) can supply P to wheat (Triticum aestivum L.). Methods An IG soil was separated into different size fractions ( \u3c 2, 2–4, 4–6, 6–8 and 8–10 mm), and adsorption and desorption experiments, volumetric moisture measurements and glasshouse experiments were conducted. Each of the \u3e 2 mm fractions were enriched with P to different levels and added to a sand culture, or to the enriched \u3c 2 mm fraction in different amounts (25%, 50% and 75% IG). Wheat was grown in pots and growth correlated to P added from enriched soil fractions, weighted Colwell P, soil solution P concentrations and volumetric water content. Key results The \u3c 2 mm fraction of the IG soil adsorbed more P than the \u3e 2 mm fraction of the IG soil likely due to its greater specific surface area. Volumetric water content decreased as gravel amount increased. Wheat was more responsive to P for larger compared to smaller gravel sizes. The P-enriched IG was able to support the growth of wheat in the absence of any other P source. For the same level of P enrichment, dry matter decreased as gravel amount increased. Conclusions The IG influences wheat growth through P retention and release and soil moisture. Volumetric water content can be reduced significantly by high gravel contents, leading to reduced wheat growth despite sufficient P fertility. Implications Depending on the nature of the soil matrix, soils with high amounts (~50%) of larger IG are likely to require lower P applications to optimise crop yield. Soil sampling strategies and laboratory testing need to consider how to practically include the \u3e 2 mm fraction during sample collection and analysis
Farm profit impacts of consecutive drought years under climate scenarios in southwestern Australia
No full textClimate projections for southern Australia indicate an increased likelihood of years of consecutive droughts. How resilient are farm businesses and their farming systems to recover from episodic drought? This study assesses the farm economic and financial impacts of consecutive droughts in southwest Australia. Bioeconomic simulation modelling of farm businesses with different farming systems at three locations is used to examine the long-term financial consequences of consecutive droughts. Various key factors affecting those consequences are examined. Farm location, farming system and starting equity are found to be the main determinants of how quickly a farm business can financially recover from the impact of consecutive droughts. Initial low equity greatly reduces the likelihood or speed of recovery from consecutive droughts, especially in low rainfall locations where crop dominant farming systems are commonplace. Projected future climate is shown to affect farms differently, based on the farm’s location and farming system. Farms in high rainfall locations are projected to benefit from future climate change and are more resilient and quicker to recover from consecutive droughts. Changes in sheep or grain prices, during and immediately after consecutive droughts, are shown to only marginally affect a farm’s long-term financial performance whereas rates of gain in crop improvement more strongly support the farm’s financial recovery
Evaluating Remotely Sensed Spectral Indices to Quantify Seagrass in Support of Ecosystem-Based Fisheries Management in a Marine Protected Area of Western Australia
No full textHighlights What are the main findings? Four spectral indices were identified as important for the quantification of seagrass within and adjacent to the MSC-certified Western Australia Enhanced Greenlip Abalone Fishery. The Normalised Difference Aquatic Vegetation Index (NDAVI) and Depth Invariant Index of the blue and green bands were the most important indices. Similar seagrass cover and distribution were observed inside and outside of the fishery area of operation. What are the implication of the main finding? The use of indices from free satellite products via Google Earth Engine workflows and automatic image annotation provides a rapidly repeatable method to support ecosystem-based fisheries management for this fishery. These findings may have broader applications for ecosystem monitoring across moderately deep (\u3c 20 m) fisheries and marine management areas. Abstract Understanding and monitoring benthic habitat distribution is essential for implementing ecosystem-based fisheries management (EBFM). Satellite remote sensing offers a rapid and cost-effective approach to marine habitat assessments; however, its application requires context-specific adjustment to account for environmental variability and differing study aims. As such, predictor variables must be tailored to the specific site and target habitat. This study uses Sentinel-2 Level 2A surface reflectance satellite imagery and stability selection via Random Forest Recursive Feature Elimination to assess the importance of remote sensing indices for mapping moderately deep (\u3c 20 m) seagrass habitats in relation to the Marine Stewardship Council-certified Western Australia Enhanced Greenlip Abalone Fishery (WAEGAF). Of the seven indices tested, the Normalised Difference Aquatic Vegetation Index (NDAVI) and Depth Invariant Index for the blue and green bands were selected in the optimal model on every run. The kernelised NDAVI and Water-Adjusted Vegetation Index also scored highly (both 0.92) and were included in the final classification and regression models. Both models performed well and predicted a similar cover and distribution of seagrass within the fishery compared to the surrounding area, providing a baseline and supporting EBFM of the WAEGAF within the surrounding marine protected area. Importantly, the use of indices from freely accessible ready-to-use satellite products via Google Earth Engine workflows and expedited ground truth image annotation using highly accurate (0.96) automatic image annotation provides a rapidly repeatable method for delivering ecosystem information for this fishery
Metabolic Responses to Salinity Identify a Role for Mitochondrial 2-Oxoglutarate Dehydrogenase in Wheat Tissue Tolerance
No full textWheat is a staple crop crucial for global food security, but its production is significantly affected by salt stress. Exploring natural genetic diversity in wheat can identify ways to improve salt tolerance. We subjected five wheat genotypes: Mocho de Espiga Branca (enhanced tissue tolerance), Fretes (tissue tolerance), Wyalkatchem and Westonia (salt exclusion) and Westonia Nax1 (enhanced salt exclusion), to 150 mM NaCl for 8 days. We measured changes in biomass, photosynthesis, chlorophyll content, Na+/K+ ratios and protein abundance. Mocho maintained growth despite high tissue Na+, showing physiological tolerance supported by differential regulation of mitochondrial proteins, central carbon metabolism, the GABA shunt and compatible solutes. Mitochondrial complexome profiling revealed salt-induced instability of 2-oxoglutarate dehydrogenase complex (OGDC) and a hydroxyglutarate synthase orthologue (HglS). In vitro assays confirmed subtle but significant OGDC activity and stability differences in Mocho, which also retained higher TCA cycle enzyme levels in vivo. Whole-plant treatment with the OGDC inhibitor succinyl phosphonate reproduced salt-like reductions in chlorophyll and biomass, particularly in Mocho. These findings highlight distinct strategies of tissue tolerance and salt exclusion in wheat, emphasising OGDC\u27s role in Mocho\u27s salt tolerance and pointing to metabolic pathways that could improve tissue tolerance traits and support sustainable agriculture
Transcription factors – Insights into abiotic and biotic stress resilience and crop improvement
No full textNumerous crop traits are controlled by multiple gene-networks. These gene-networks play a crucial role in crop evolution, disease prevention, stress adaptation and other fundamental processes in different organisms. Transcription factors (TFs) are master regulators of gene-networks and therefore have been targets for genetic improvement in crops since the dawn of agriculture. Enhancement of quantitative traits through plant breeding often involves manipulation of several TF sites and altered RNA expression. Advancements in OMICS technology have significantly expanded our understanding of transcription factor (TF) binding sites in plants and their roles in various biological processes. This progress has facilitated the validation of TF-related mutations and alleles, offering breeders new opportunities to achieve rapid genetic gains in response to abiotic and biotic stresses. The crop improvements using TFs as master targets is irrespective of crop type, mode of inheritance, number of operative genes and their interactions. Here, we review some of the intensively studied families of TFs– bZIP, bHLH, NAC, ATAF, AP2/ERF, MYB, and WRKY for abiotic and biotic stress resilience in crops and their potential as targets for crop improvement. Breeders’ perspective on status and relevance of TFs in the current breeding programs, utilization of precision editing and prospects of using TFs as regular targets in future crop improvement is discussed
Large scale genome-wide association analysis identified QTLs associated with aluminum tolerance in chickpea
No full textChickpea has become an increasingly popular healthy food worldwide. Aluminum (Al) toxicity is a major hurdle for chickpea cultivation and yield improvement in acidic soils. However, the genetic mechanism of Al-tolerance in chickpea remains poorly understood. Here, we performed a large-scale hydroponics screening and SNP chip array genotyping of 1154 diverse chickpea accessions. Root lengths after 6 days cultivation under hydroponics in control (T0: pH 4.2) and Al treatment (T1: pH4.2, 15/20 μM Al3+) were measured. Root tolerance index (RTI = T1/T0) ranking revealed significant variations in chickpea Al-tolerance, with common Australian chickpea cultivars positioned in the low to medium range. Genome-wide association analyses revealed eight QTLs on chromosomes ca1 (CaAlt1-1), ca3 (CaAlt3-1), ca4 (CaAlt4-1, CaAlt4-2), ca5(CaAlt5-1), ca6 (CaAlt6-1), and ca7 (CaAlt7-1, CaAlt7-2) associated with T1, implying a multigenic genetic basis for Al-tolerance in chickpea. Specifically, CaAlt7-2 was associated with both T1 and RTI, whilst CaAlt4-2 was detected for T1 uniquely in the HatTrick x CudiB22C population. Al- tolerant and sensitive haplotypes for the identified QTLs were also identified. Organic acid transporter genes CaMATE2, CaMATE4, and CaALMT1 were found in proximal genomic regions to CaAlt7-2, CaAlt4-1, and CaAlt6-1, respectively. Further qRT-PCR in parental chickpea lines (HatTrick, Slasher, Gunas, CudiB) confirmed that CaMATE2 and CaMATE4 were strongly induced upon Al treatment. Interestingly, CaMATE2 was preferentially expressed in the upper part of the root, whilst CaMATE4 preferentially in the root tips, implying a potential complementary role in Al resistance. Their direct roles in Al tolerance and the potential alternative candidate genes near the QTLs require further investigation. This first report of QTLs on Al-tolerance in chickpea has substantially advanced our understanding of the genetic basis of Al tolerance in chickpea and will facilitate the rapid breeding of Al-tolerant chickpea cultivars for previously un-accessible acidic soils