Smoke plumes are the first things seen from space when wildfires occur. Thus, fire smoke detection is important for early fire detection. Deep Learning (DL) models have been used to detect fire smoke in satellite imagery for fire detection. However, previous DL-based research only considered lower spatial resolution sensors (e.g., Moderate-Resolution Imaging Spectroradiometer (MODIS)) and only used the visible (i.e., red, green, blue (RGB)) bands. To contribute towards solutions for early fire smoke detection, we constructed a six-band imagery dataset from Landsat 5 Thematic Mapper (TM) and Landsat 8 Operational Land Imager (OLI) with a 30-metre spatial resolution. The dataset consists of 1836 images in three classes, namely “Smoke”, “Clear”, and “Other_aerosol”. To prepare for potential on-board-of-small-satellite detection, we designed a lightweight Convolutional Neural Network (CNN) model named “Variant Input Bands for Smoke Detection (VIB_SD)”, which achieved competitive accuracy with the state-of-the-art model SAFA, with less than 2% of its number of parameters. We further investigated the impact of using additional Infra-Red (IR) bands on the accuracy of fire smoke detection with VIB_SD by training it with five different band combinations. The results demonstrated that adding the Near-Infra-Red (NIR) band improved prediction accuracy compared with only using the visible bands. Adding both Short-Wave Infra-Red (SWIR) bands can further improve the model performance compared with adding only one SWIR band. The case study showed that the model trained with multispectral bands could effectively detect fire smoke mixed with cloud over small geographic extents.

This study establishes the use of the Earth Observation Data for Ecosystem Monitoring (EODESM) to generate land cover and change classifications based on the United Nations Food and Agriculture Organisation (FAO) Land Cover Classification System (LCCS) and environmental variables (EVs) available within, or accessible from, Geoscience Australia’s (GA) Digital Earth Australia (DEA). Classifications representing the LCCS Level 3 taxonomy (8 categories representing semi-(natural) and/or cultivated/managed vegetation or natural or artificial bare or water bodies) were generated for two time periods and across four test sites located in the Australian states of Queensland and New South Wales. This was achieved by progressively and hierarchically combining existing time-static layers relating to (a) the extent of artificial surfaces (urban, water) and agriculture and (b) annual summaries of EVs relating to the extent of vegetation (fractional cover) and water (hydroperiod, intertidal area, mangroves) generated through DEA. More detailed classifications that integrated information on, for example, forest structure (based on vegetation cover (%) and height (m); time-static for 2009) and hydroperiod (months), were subsequently produced for each time-step. The overall accuracies of the land cover classifications were dependent upon those reported for the individual input layers, with these ranging from 80% (for cultivated, urban and artificial water) to over 95% (for hydroperiod and fractional cover). The changes identified include mangrove dieback in the southeastern Gulf of Carpentaria and reduced dam water levels and an associated expansion of vegetation in Lake Ross, Burdekin. The extent of detected changes corresponded with those observed using time-series of RapidEye data (2014 to 2016; for the Gulf of Carpentaria) and Google Earth imagery (2009–2016 for Lake Ross). This use case demonstrates the capacity and a conceptual framework to implement EODESM within DEA and provides countries using the Open Data Cube (ODC) environment with the opportunity to routinely generate land cover maps from Landsat or Sentinel-1/2 data, at least annually, using a consistent and internationally recognised taxonomy.

This article presents our research in the prelaunch phase of the Kanyini mission, which aims to implement an energy-efficient, AI-based system onboard for early fire smoke detection using hyperspectral imagery. Our approach includes three key components: developing a diverse hyperspectral training dataset from VIIRS imagery, groundwork in band selection and AI model preparation, and developing an emulation system. We adapted and evaluated our previously developed lightweight convolutional neural network model, VIB_SD, to meet the computational constraints of satellite deployment. The emulation system tests various onboard AI tasks and processes. Our comprehensive experiments demonstrate the feasibility and benefits of employing onboard AI for fire smoke detection, significantly improving downlink efficiency, energy consumption, and detection speed.

A globally relevant and standardized taxonomy and framework for consistently describing land cover change based on evidence is presented, which makes use of structured land cover taxonomies and is underpinned by the Driver‐Pressure‐State‐Impact‐Response (DPSIR) framework. The Global Change Taxonomy currently lists 246 classes based on the notation ‘impact (pressure)’, with this encompassing the consequence of observed change and associated reason(s), and uses scale‐independent terms that factor in time. Evidence for different impacts is gathered through temporal comparison (e.g., days, decades apart) of land cover classes constructed and described from Environmental Descriptors (EDs; state indicators) with pre‐defined measurement units (e.g., m, %) or categories (e.g., species type). Evidence for pressures, whether abiotic, biotic or human‐influenced, is similarly accumulated, but EDs often differ from those used to determine impacts. Each impact and pressure term is defined separately, allowing flexible combination into ‘impact (pressure)’ categories, and all are listed in an openly accessible glossary to ensure consistent use and common understanding. The taxonomy and framework are globally relevant and can reference EDs quantified on the ground, retrieved/classified remotely (from ground‐based, airborne or spaceborne sensors) or predicted through modelling. By providing capacity to more consistently describe change processes—including land degradation, desertification and ecosystem restoration—the overall framework addresses a wide and diverse range of local to international needs including those relevant to policy, socioeconomics and land management. Actions in response to impacts and pressures and monitoring towards targets are also supported to assist future planning, including impact mitigation actions.

To comprehensively support national and international initiatives for sustainable development, land cover products need to be reliably and routinely generated within operational frameworks. Coupled with consistent semantics and taxonomies, ensuring confidence in mapping land cover for multiple time periods, facilitates informed decision-making at scales appropriate to multiple policy domains. The United Nations Food and Agriculture Organisation (FAO) Land Cover Classification System (LCCS) provides a taxonomy that comparable at different scales, level of detail and geographic location. The Open Data Cube (ODC) initiative offers a framework for operational continental-scale land cover mapping using analysis-ready Earth Observation data. This study utilised the FAO LCCS framework and the Landsat sensor data through Digital Earth Australia (DEA; Australia’s ODC instance) to generate consistent and continent-wide land cover mapping (DEA Land Cover) of the Australian continent. DEA Land Cover provides annual maps from 1988 to 2020 at 25 m resolution. Output maps were validated with ∼12,000 independent validation points, giving an overall map accuracy of 80%. DEA Land Cover provides Australia with a nationally consistent picture of land cover, with an open-source software package using readily available global coverage data and demonstrates a pathway of adoption for national implementations across the world.

Increased fire activity across the Amazon, Australia, and even the Arctic regions has received wide recognition in the global media in recent years. Large-scale, long-term analyses are required to postulate if these incidents are merely peaks within the natural oscillation, or rather the consequence of a linearly rising trend. While extensive datasets are available to facilitate the investigation of the extent and frequency of wildfires, no means has been available to also study the severity of the burnings on a comparable scale. This is now possible through a dataset recently published by the German Aerospace Center (DLR). This study exploits the possibilities of this new dataset by exemplarily analyzing fire severity trends on the Australian East coast for the past 20 years. The analyzed data is based on 3503 tiles of the ESA Sentinel-3 OLCI instrument, extended by 9612 granules of the NASA MODIS MOD09/MYD09 product. Rising trends in fire severity could be found for the states of New South Wales and Victoria, which could be attributed mainly to developments in the temperate climate zone featuring hot summers without a dry season (Cfa). Within this climate zone, the ecological units featuring needleleaf and evergreen forest are found to be mainly responsible for the increasing trend development. The results show a general, statistically significant shift of fire activity towards the affection of more woody, ecologically valuable vegetation.