Search

Mineral prospectivity mapping is a crucial technique for discovering new economic mineral deposits. However, detailed knowledge-based geological exploration and interpretations generally involve significant costs, time, and human resources. In this study, an ensemble machine learning approach was tested using geoscience datasets to map Cu-Au and Pb-Zn mineral prospectivity in the Cobar Basin, NSW, Australia. The input datasets (magnetic, gravity, faults, electromagnetic, and magnetotelluric data layers) were chosen by considering their association with Cu-Au and Pb-Zn mineralization patterns. Three machine learning algorithms, namely random forest (RF), support vector machine (SVM), and maximum-likelihood (MaxL) classification, were applied to the input data. The results of the three algorithms were ensembled to produce Cu-Au and Pb-Zn prospectivity maps over the Cobar Basin with improved classification accuracy. The findings demonstrate good agreement with known mineral occurrence points and existing mineral prospectivity maps developed using the weights-of-evidence (WofE) method. The ability to capture training points accurately and the simplicity of the proposed approach make it advantageous over complex mineral prospectivity mapping methods, to serve as a preliminary evaluation technique. The methodology can be modified with different datasets and algorithms, facilitating the investigations of mineral prospectivity in other regions and providing guidance for more detailed, high-resolution geological investigations. [ABSTRACT FROM AUTHOR]

Accurate pre-harvest yield forecasting of mango is essential to the industry as it supports better decision making around harvesting logistics and forward selling, thus optimizing productivity and reducing food waste. Current methods for yield forecasting such as manually counting 2–3% of the orchard can be accurate but are very time inefficient and labour intensive. More recent evaluations of technological solutions such as remote (satellite) and proximal (on ground) sensing have provided very encouraging results, but they still require infield in-season sampling for calibration, the technology comes at a significant cost, and commercial availability is limited, especially for vehicle-mounted sensors. This study presents the first evaluation of a "time series"—based remote sensing method for yield forecasting of mango, a method that does not require infield fruit counts and utilizes freely available satellite imagery. Historic yield data from 2015 to 2022 were sourced from 51 individual orchard blocks from two farms (AH and MK) in the Northern Territory of Australia. Time series measures of the canopy reflectance properties of the blocks were obtained from Landsat 7 and 8 satellite data for the 2015–2022 growing seasons. From the imagery, the following vegetation indices (VIs) were derived: EVI, GNDVI, NDVI, and LSWI, whilst corresponding weather variables (rainfall (Prec), temperature (Tmin/Tmax), evapotranspiration (ETo), solar radiation (Rad), and vapor pressure deficit (vpd)) were also sourced from SILO data. To determine the relationships among weather and remotely sensed measures of canopy throughout the growing season and the yield achieved (at the block level and the farm level), six machine learning (ML) algorithms, namely random forest (RF), support vector regression (SVR), eXtreme gradient boosting (XGBOOST), RIDGE, LASSO and partial least square regression (PLSR), were trialed. The EVI/GNDVI and Prec/Tmin were found to be the best RS and weather predictors, respectively. The block-level combined RS/weather-based RF model for 2021 produced the best result (MAE = 2.9 t/ha), marginally better than the RS only RF model (MAE = 3.4 t/ha). The farm-level model error (FLEM) was generally lower than the block-level model error, for both the combined RS/weather-based RF model (farm = 3.7%, block (NMAE) = 33.6% for 2021) and the RS-based model (farm = 4.3%, block = 38.4% for 2021). Further testing of the RS/weather-based RF models over six additional orchards (other than AH and MK) produced errors ranging between 24% and 39% from 2016 to 2020. Although accuracies of prediction did vary at both the block level and the farm level, this preliminary study demonstrates the potential of a "time series" RS method for predicting mango yields. The benefits to the mango industry are that it utilizes freely available imagery, requires no infield calibration, and provides predictions several months before the commercial harvest. Therefore, this outcome not only presents a more adoptable option for the industry, but also better supports automation and scalability in terms of block-, farm-, regional, and national level forecasting. [ABSTRACT FROM AUTHOR]

One of the most valuable and nutritionally essential agricultural commodities worldwide is milk. The European Union and New Zealand are the second- and third-largest exporting regions of milk products and rely heavily on pasture-based production systems. They are comparable to the Australian systems investigated in this study. With projections of herd decline, increased milk yield must be obtained from a combination of animal genetics and feed efficiencies. Accurate pasture biomass estimation across all seasons will improve feed efficiency and increase the productivity of dairy farms; however, the existing time-consuming and manual methods of pasture measurement limit improvements to utilisation. In this study, Sentinel-2 (S2) band and spectral index (SI) information were coupled with the broad season and management-derived datasets using a Random Forest (RF) machine learning (ML) framework to develop a perennial ryegrass (PRG) biomass prediction model accurate to +/−500 kg DM/ha, and that could predict pasture yield above 3000 kg DM/ha. Measurements of PRG biomass were taken from 11 working dairy farms across southeastern Australia over 2019–2021. Of the 68 possible variables investigated, multiple simulations identified 12 S2 bands and 9 SI, management and season as the most important variables, where Short-Wave Infrared (SWIR) bands were the most influential in predicting pasture biomass above 4000 kg DM/ha. Conditional Latin Hypercube Sampling (cLHS) was used to split the dataset into 80% and 20% for model calibration and internal validation in addition to an entirely independent validation dataset. The combined internal model validation showed R2 = 0.90, LCCC = 0.72, RMSE = 439.49 kg DM/ha, NRMSE = 15.08, and the combined independent validation had R2 = 0.88, LCCC = 0.68, RMSE = 457.05 kg DM/ha, NRMSE = 19.83. The key findings of this study indicated that the data obtained from the S2 bands and SI were appropriate for making accurate estimations of PRG biomass. Furthermore, including SWIR bands significantly improved the model. Finally, by utilising an RF ML model, a single 'global' model can automate PRG biomass prediction with high accuracy across extensive regions of all seasons and types of farm management. [ABSTRACT FROM AUTHOR]

The characterisation of fuel distribution across heterogeneous landscapes is important for wildfire mitigation, validating fuel models, and evaluating fuel treatment outcomes. However, efficient fuel mapping at a landscape scale is challenging. Fuel hazard metrics were obtained using Terrestrial Laser Scanning (TLS) and the current operational approach (visual fuel assessment) for seven sites across south-eastern Australia. These point-based metrics were then up-scaled to a continuous fuel map, an area relevant to fire management using random forest modelling, with predictor variables derived from Airborne Laser Scanning (ALS), Sentinel 2A images, and climate and soil data. The model trained and validated with TLS observations (R 2 = 0.51 for near-surface fuel cover and 0.31 for elevated fuel cover) was found to have higher predictive power than the model trained with visual fuel assessments (R 2 = −0.1 for the cover of both fuel layers). Models for height derived from TLS observations exhibited low-to-moderate performance for the near-surface (R 2 = 0.23) and canopy layers (R 2 = 0.25). The results from this study provide practical guidance for the selection of training data sources and can be utilised by fire managers to accurately generate fuel maps across an area relevant to operational fire management decisions. [ABSTRACT FROM AUTHOR]

Land cover mapping from satellite images has progressed from visual and statistical approaches to Random Forests (RFs) and, more recently, advanced image recognition techniques such as convolutional neural networks (CNNs). CNNs have a conceptual benefit over RFs in recognising spatial feature context, but potentially at the cost of reduced spatial detail. We tested the use of CNNs for improved land cover mapping based on Landsat data, compared with RFs, for a study area of approximately 500 km × 500 km in southeastern Australia. Landsat 8 geomedian composite surface reflectances were available for 2018. Label data were a simple nine-member land cover classification derived from reference land use mapping (Catchment Scale Land Use of Australia—CLUM), and further enhanced by using custom forest extent mapping (Forests of Australia). Experiments were undertaken testing U-Net CNN for segmentation of Landsat 8 geomedian imagery to determine the optimal combination of input Landsat 8 bands. The results were compared with those from a simple autoencoder as well as an RF model. Segmentation test results for the best performing U-Net CNN models produced an overall accuracy of 79% and weighted-mean F1 score of 77% (9 band input) or 76% (6 band input) for a simple nine-member land cover classification, compared with 73% and 68% (6 band input), respectively, for the best RF model. We conclude that U-Net CNN models can generate annual land cover maps with good accuracy from proxy training data, and can also be used for quality control or improvement of existing land cover products. [ABSTRACT FROM AUTHOR]

Coastal wetland ecosystems, such as saltmarsh and mangroves, provide a wide range of important ecological and socio-economic services. A good understanding of the spatial and temporal distribution of these ecosystems is critical to maximising the benefits from restoration and conservation projects. We mapped mangrove and saltmarsh ecosystem transitions from 1991 to 2015 in south-eastern Australia, using remotely sensed Landsat data and a Random Forest classification. Our classification results were improved by the addition of two physical variables (Shuttle Radar Topographic Mission (SRTM), and Distance to Water). We also provide evidence that the addition of post-classification, spatial and temporal, filters improve overall accuracy of coastal wetlands detection by up to 16%. Mangrove and saltmarsh maps produced in this study had an overall User Accuracy of 0.82–0.95 and 0.81–0.87 and an overall Producer Accuracy of 0.71–0.88 and 0.24–0.87 for mangrove and saltmarsh, respectively. We found that mangrove ecosystems in south-eastern Australia have lost an area of 1148 ha (7.6%), whilst saltmarsh experienced an overall increase in coverage of 4157 ha (20.3%) over this 24-year period. The maps developed in this study allow local managers to quantify persistence, gains, and losses of coastal wetlands in south-eastern Australia. [ABSTRACT FROM AUTHOR]

Recent studies have suggested that due to climate change, the number of wildfires across the globe have been increasing and continue to grow even more. The recent massive wildfires, which hit Australia during the 2019–2020 summer season, raised questions to what extent the risk of wildfires can be linked to various climate, environmental, topographical, and social factors and how to predict fire occurrences to take preventive measures. Hence, the main objective of this study was to develop an automatized and cloud-based workflow for generating a training dataset of fire events at a continental level using freely available remote sensing data with a reasonable computational expense for injecting into machine learning models. As a result, a data-driven model was set up in Google Earth Engine platform, which is publicly accessible and open for further adjustments. The training dataset was applied to different machine learning algorithms, i.e., Random Forest, Naïve Bayes, and Classification and Regression Tree. The findings show that Random Forest outperformed other algorithms and hence it was used further to explore the driving factors using variable importance analysis. The study indicates the probability of fire occurrences across Australia as well as identifies the potential driving factors of Australian wildfires for the 2019–2020 summer season. The methodical approach and achieved results and drawn conclusions can be of great importance to policymakers, environmentalists, and climate change researchers, among others. [ABSTRACT FROM AUTHOR]