L-band microwave-retrieved fuel temperature predicts million-hectare-scale destructive wildfires

Natural Hazards and Earth System Sciences

The Canadian Fire Weather Index (FWI) system, even though originally developed and calibrated for an upland Jack pine forest, is used globally to estimate fire danger for any fire environment. However, for some environments, such as peatlands, the applicability of the FWI in its current form, is often questioned. In this study, we replaced the original moisture codes of the FWI with hydrological estimates resulting from the assimilation of satellite-based L-band passive microwave observations into a peatland-specific land surface model. In a conservative approach that maintains the integrity of the original FWI structure, the distributions of the hydrological estimates were first matched to those of the corresponding original moisture codes before replacement. The resulting adapted FWI, hereafter called FWI, was evaluated using satellite-based information on fire presence over boreal peatlands from 2010 through 2018. Adapting the FWI with model- and satellite-based hydrological information was found to be beneficial in estimating fire danger, especially when replacing the deeper moisture codes of the FWI. For late-season fires, further adaptations of the fine fuel moisture code show even more improvement due to the fact that late-season fires are more hydrologically driven. The proposed FWI should enable improved monitoring of fire risk in boreal peatlands.

Biogeosciences

The snow cover extent across the Northern Hemisphere has diminished, while the number of lightning ignitions and amount of burned area have increased over the last 5 decades with accelerated warming. However, the effects of earlier snow disappearance on fire are largely unknown. Here, we assessed the influence of snow disappearance timing on fire ignitions across 16 ecoregions of boreal North America. We found spatially divergent trends in earlier (later) snow disappearance, which led to an increasing (decreasing) number of ignitions for the northwestern (southeastern) ecoregions between 1980 and 2019. Similar northwest–southeast divergent trends were observed in the changing length of the snow-free season and correspondingly the fire season length. We observed increases (decreases) over northwestern (southeastern) boreal North America which coincided with a continental dipole in air temperature changes between 2001 and 2019. Earlier snow disappearance induced earlier ignitions of between 0.22 and 1.43 d earlier per day of earlier snow disappearance in all ecoregions between 2001 and 2019. Early-season ignitions (defined by the 20 % earliest fire ignitions per year) developed into significantly larger fires in 8 out of 16 ecoregions, being on average 77 % larger across the whole domain. Using a piecewise structural equation model, we found that earlier snow disappearance is a good direct proxy for earlier ignitions but may also result in a cascade of effects from earlier desiccation of fuels and favorable weather conditions that lead to earlier ignitions. This indicates that snow disappearance timing is an …

Science of the Total Environment

Climate change has disproportional effects on Arctic-boreal ecosystems, as the increase of air temperatures in these northern regions is several times higher than the global average. Ongoing warming and drying have resulted in recent record-breaking fire years in Arctic-boreal ecosystems, resulting in substantial carbon emissions that might accelerate climate change. While recent trends in Arctic-boreal burned area have been well documented, it is still unclear how fire intensity has changed. Fire intensity relates to the energy release from combustion and to a large extent drives the impact of a fire on the vegetation and soils, the emission of various gasses and the combustion completeness of different fuels. Here, we used the active fire product from the Moderate Resolution Imaging Spectroradiometer (MODIS) to examine trends in fire radiative power (FRP) over the entire Arctic-boreal region. We found a …

Biogeosciences

Fire is the dominant disturbance agent in Alaskan and Canadian boreal ecosystems and releases large amounts of carbon into the atmosphere. Burned area and carbon emissions have been increasing with climate change, which have the potential to alter the carbon balance and shift the region from a historic sink to a source. It is therefore critically important to track the spatiotemporal changes in burned area and fire carbon emissions over time. Here we developed a new burned-area detection algorithm between 2001–2019 across Alaska and Canada at 500 m (meters) resolution that utilizes finer-scale 30 m Landsat imagery to account for land cover unsuitable for burning. This method strictly balances omission and commission errors at 500 m to derive accurate landscape- and regional-scale burned-area estimates. Using this new burned-area product, we developed statistical models to predict burn depth and carbon combustion for the same period within the NASA Arctic–Boreal Vulnerability Experiment (ABoVE) core and extended domain. Statistical models were constrained using a database of field observations across the domain and were related to a variety of response variables including remotely sensed indicators of fire severity, fire weather indices, local climate, soils, and topographic indicators. The burn depth and aboveground combustion models performed best, with poorer performance for belowground combustion. We estimate  ha (2.37 Mha) burned annually between 2001–2019 over the ABoVE domain (2.87 Mha across all of Alaska and Canada), emitting 79.3  27.96 Tg ( standard deviation) of carbon (C …

De laatste jaren zijn natuurbranden zowel binnen als buiten Nederland een zeer actueel thema. De vraag is hoe, onder andere met oog op klimaatverandering, natuurbranden in Nederland aan het veranderen zijn. Dat geldt niet alleen voor het aantal natuurbranden, hun brandgedrag en grootte, maar ook voor de mogelijke impact van die branden. Met deze vraagstukken in gedachten is een consortium gevormd van experts op het gebied van natuurbranden vanuit de kennisinstituten NIPV, KNMI, WUR, VU en Deltares, die samen het voorliggende rapport hebben geschreven. Om inzicht te krijgen in de (toekomstige) ontwikkelingen is de volgende onderzoeksvraag opgesteld: Hoe ontwikkelt het natuurbrandrisico zich in Nederland? Deze hoofdvraag is onderverdeeld in een aantal deelvragen: 1. Hoe verandert de gevoeligheid voor het ontstaan van natuurbranden? 2. Hoe verandert de gevoeligheid voor het ontwikkelen van natuurbranden? 3. Hoe verandert de impact van natuurbranden? Een groter deel van Nederland zal geconfronteerd worden met meer natuurbranden. Het natuurbrandrisico manifesteerde zich in Nederland vooral in de lente, maar in de toekomst zal er daarbij ook steeds vaker in de zomer (langdurig) sprake zijn van een (hoog) risico. Natuurbranden zullen zich vaker ontwikkelen tot branden die niet meer geblust kunnen worden, maar pas stoppen als er geen brandstof meer is. De toename van het aantal branden, gepaard met een verdere verdichting van Nederland, leidt tot een hoge kans op natuurbranden met veel impact op gezondheid, welzijn, natuur en economie.

Nature Geoscience

Fires can be ignited by people or by natural causes, which are almost exclusively lightning strikes. Discriminating between lightning and anthropogenic fires is paramount when estimating impacts of changing socioeconomic and climatological conditions on fire activity. Here we use reference data of fire ignition locations, cause and burned area from seven world regions in a machine-learning approach to obtain a global attribution of lightning and anthropogenic ignitions as dominant fire ignition sources. We show that 77% (uncertainty expressed as one standard deviation = 8%) of the burned area in extratropical intact forests currently stems from lightning and that these areas will probably experience 11 to 31% more lightning per degree warming. Extratropical forests are of global importance for carbon storage. They currently experience high fire-related forest losses and have, per unit area, among the largest fire …

Nature communications

In most of the world, conditions conducive to wildfires are becoming more prevalent. Net carbon emissions from wildfires contribute to a positive climate feedback that needs to be monitored, quantified, and predicted. Here we use a causal inference approach to evaluate the influence of top-down weather and bottom-up fuel precursors on wildfires. The top-down dominance on wildfires is more widespread than bottom-up dominance, accounting for 73.3% and 26.7% of regions, respectively. The top-down precursors dominate in the tropical rainforests, mid-latitudes, and eastern Siberian boreal forests. The bottom-up precursors dominate in North American and European boreal forests, and African and Australian savannahs. Our study identifies areas where wildfires are governed by fuel conditions and hence where fuel management practices may be more effective. Moreover, our study also highlights that top-down …

Land

Fires can have an anthropogenic or natural origin. The most frequent natural fire cause is lightning. Since anthropogenic and lightning fires have different climatic and socio-economic drivers, it is important to distinguish between these different fire causes. We developed random forest models that predict the fraction of anthropogenic and lightning fire incidences, and their burned area, at the level of the Nomenclature des Unités Territoriales Statistiques level 3 (NUTS3) for Europe. The models were calibrated using the centered log-ratio of fire incidence and burned area reference data from the European Forest Fire Information System. After a correlation analysis, the population density, fractional human land impact, elevation and burned area coefficient of variation—a measure of interannual variability in burned area—were selected as predictor variables in the models. After parameter tuning and running the models with several train-validate compositions, we found that the vast majority of fires and burned area in Europe has an anthropogenic cause, while lightning plays a significant role in the remote northern regions of Scandinavia. Combining our results with burned area data from the Moderate Resolution Imaging Spectroradiometer, we estimated that 96.5 ± 0.9% of the burned area in Europe has an anthropogenic cause. Our spatially explicit fire cause attribution model demonstrates the spatial variability between anthropogenic and lightning fires and their burned area over Europe and could be used to improve predictive fire models by accounting for fire cause.

Biogeosciences

Boreal forests are particularly vulnerable to climate warming, which increases the occurrence of natural disturbances, such as fires and insect outbreaks. It is therefore essential to better understand climate-induced changes in boreal vegetation dynamics. This requires accurate estimates of variations in biomass across regions and time. This remains challenging in the extensive larch forests of northeast Siberia because of the paucity of allometric equations and physical properties of woody debris needed for quantifying aboveground biomass pools from field surveys. Our study is the first to present values of mean squared diameter (MSD) and specific gravity that can be used to calculate fine dead and downed woody debris loads in Cajander larch (Larix cajanderi) forests using the line-intersect sampling approach. These values were derived from field measurements collected in 25 forest stands in the Republic of Sakha, Russia, and compared with values reported for other prevalent boreal tree species. We developed allometric equations relating diameter at breast height (DBH, at 1.3 m) to stem wood, stem bark, branches, foliage, and aboveground biomass based on measurements of 63 trees retrieved from previous studies. Differences between our allometric models and existing equations were assessed in predicting larch aboveground biomass in 53 forest stands sampled in the Republic of Sakha. We found that using fine woody debris (FWD) parameters from other boreal tree species and allometric equations developed in other regions may result in significantly lower biomass estimates in larch-dominated forests of northeast Siberia. The …

Remote Sensing

We present a simple modeling technique based on linear spectral mixture analysis to assess satellite detectability of sub-pixel burned area. Pixel observations are modeled using a linear combination of pure land covers, called endmembers. We executed an experiment using spectral data from Yellowstone National Park, USA. Using endmember samples from spectral libraries, pixel samples were assessed on burn detectability using the widely used differenced Normalized Burn Ratio (dNBR). While individual samples yielded differing results for Landsat 8, Sentinel-2, and the Moderate Resolution Imaging Spectroradiometer (MODIS), the average park-wide detectability of burned area was consistent across satellites. For the commonly used dNBR threshold of 0.15, the results indicated that detectability is reached when around a quarter of a pixel’s area is burned. However, a significant percentage of the modeled burned pixels remained undetectable, especially those with low pre-fire vegetation cover. This has consequences for burned area estimates, as smaller fires in sparsely vegetated terrain may remain undetected in moderate resolution burned area products.

Environmental Research Letters

Reducing the climate change-induced risk of uncontrollable fires in landscapes under nature management, with severe impacts on landscape and society, is particularly urgent in densely-populated and fragmented areas. Reducing fire risk in such areas requires active involvement of a wide diversity of stakeholders. This research letter investigates stakeholders' needs with regard to fire risk reduction in the Veluwe area in the Netherlands. This densely populated landscape is a popular tourist attraction, and it is one of the most fire-prone landscapes of the Netherlands, with abundant fuels and human ignition sources. We draw upon seven in-depth qualitative interviews with key stakeholders in the Veluwe area, which we situate in a wider review of existing literature. Our analysis demonstrates that the rising incidence of uncontrollable fires poses four types of new challenges to these stakeholders in the Veluwe area …

Environmental Research Letters

Lightning-induced fire is the primary disturbance agent in boreal forests. Recent large fire years have been linked to anomalously high numbers of lightning-caused fire starts, yet the mechanisms regulating the probability of lightning ignition remain uncertain and limit our ability to project future changes. Here, we investigated the influence of lightning properties, landscape characteristics, and fire weather on lightning ignition efficiency—the likelihood that a lightning strike starts a fire—in Alaska, United States of America, and Northwest Territories, Canada, between 2001 and 2018. We found that short-term fuel drying associated with fire weather was the main driver of lightning ignition efficiency. Lightning was also more likely to ignite a wildfire in denser, evergreen forest areas. Under a high greenhouse gas emissions scenario, we predicted that changes in vegetation and fire weather increase lightning ignition …

Recent wildfire outbreaks around the world have prompted concern that climate change is increasing fire incidence, threatening human livelihood and biodiversity, and perpetuating climate change. Here, we review current understanding of the impacts of climate change on fire weather (weather conditions conducive to the ignition and spread of wildfires) and the consequences for regional fire activity as mediated by a range of other bioclimatic factors (including vegetation biogeography, productivity and lightning) and human factors (including ignition, suppression, and land use). Through supplemental analyses, we present a stocktake of regional trends in fire weather and burned area (BA) during recent decades, and we examine how fire activity relates to its bioclimatic and human drivers. Fire weather controls the annual timing of fires in most world regions and also drives inter‐annual variability in BA in the …

Earth System Science Data Discussions

Holdover fires are usually associated with lightning-ignited wildfires (LIWs), which can experience a smouldering phase or go undetected for several hours to days and weeks before being reported. Since the existence and duration of the smouldering combustion in LIWs is usually unknown, holdover time is conventionally defined as the time between the lightning event that ignited the fire and the time the fire is detected. Therefore, all LIWs have an associated holdover time, which may range from a few minutes to several days. However, we lack a comprehensive understanding of holdover times. Here, we introduce a global database on holdover times of LIWs. We have collected holdover time data from 29 different studies across the world through a literature review and datasets assembled by authors of the original studies. The database is composed of three data files (censored data, non-censored data, ancillary data) and three metadata files (description of database variables, list of references, reproducible examples). Censored data are the core of the database and consist of different frequency distributions reporting the number or relative frequency of LIWs per interval of holdover time. In addition, ancillary data provide further information to understand the methods and contexts in which the data were generated in the original studies. The first version of the database contains 42 frequency distributions of holdover time built with data on more than 152,375 LIWs from 13 countries in five continents covering a time span from 1921 to 2020. This database is the first freely available, harmonized, and ready-to-use global source of holdover time data …

Environmental Research Letters

Overwintering fires are a historically rare phenomenon but may become more prevalent in the warming boreal region. Overwintering fires have been studied to a limited extent in boreal North America; however, their role and contribution to fire regimes in Siberia are still largely unknown. Here, for the first time, we quantified the proportion of overwintering fires and their burned areas in Yakutia, eastern Siberia, using fire, lightning, and infrastructure data. Our results demonstrate that overwintering fires contributed to 3.2±0.6% of the total burned area during 2012–2020 over Yakutia, compared to 31.4±6.8% from lightning ignitions and 51.0±6.9% from anthropogenic ignitions (14.4% of the burned area had unknown cause), but they accounted for 7.5±0.7% of the burned area in the extreme fire season of 2020. In addition, overwintering fires have different spatiotemporal characteristics than lightning and anthropogenic …

Geoscientific Model Development

In fire emission models, the spatial resolution of both the modelling framework and the satellite data used to quantify burned area can have considerable impact on emission estimates. Consideration of this sensitivity is especially important in areas with heterogeneous land cover and fire regimes, and when constraining model output with field measurements. We developed a global fire emissions model with a spatial resolution of 500 m using MODerate resolution Imaging Spectroradiometer (MODIS) data. To accommodate this spatial resolution, our model is based on a simplified version of the Global Fire Emissions Database (GFED) modelling framework. Tree mortality as a result of fire, i.e. fire-related forest loss, was modelled based on the overlap between 30-m forest loss data and MODIS burned area and active fire detections. Using this new 500-m model, we calculated global average carbon emissions from fire of 2.1 ± 0.2 (±1σ interannual variability; IAV) Pg C yr–1 during 2002–2019. Fire-related forest loss accounted for 2.5 ± 0.9 % (uncertainty range = 1.9–3.2 %) of global burned area and 25 ± 6 % (uncertainty range = 18–32 %) of emissions, indicating that fuel consumption in forest fires is an order of magnitude higher than the global average. Emissions from the combustion of soil organic carbon in the boreal region and tropical peatlands accounted for 14 ± 4 % of global emissions. Our global fire emissions estimate was higher than the 1.5 Pg C yr–1 from GFED4 and similar to 2.1 Pg C yr–1 from GFED4s. Even though GFED4s included more burned area by accounting for small fires undetected by the MODIS burned area mapping …

Science

The summers of 2019, 2020, and 2021 experienced unprecedented fire activity in northeastern Siberia, driven by record high spring and summer temperatures. Many of these fires burned in permafrost peatlands within the Arctic Circle. We show that early snowmelt together with an anomalous Arctic front jet over northeastern Siberia promoted unusually warm and dry surface conditions, followed by anomalously high lightning and fire activity. Since 1966, spring snowmelt has started 1.7 days earlier each decade. Moreover, Arctic front jet occurrences in summer have more than tripled in frequency over the last 40 years. These interconnected climatological drivers promote extreme fire activity in eastern Siberia, including a northward shift of fires, which may accelerate the degradation of carbon-rich permafrost peatlands.

Environmental Research Letters

Forest characteristics, structure, and dynamics within the North American boreal region are heavily influenced by wildfire intensity, severity, and frequency. Increasing temperatures are likely to result in drier conditions and longer fire seasons, potentially leading to more intense and frequent fires. However, an increase in deciduous forest cover is also predicted across the region, potentially decreasing flammability. In this study, we use an individual tree-based forest model to test bottom-up (i.e. fuels) vs top-down (i.e. climate) controls on fire activity and project future forest and wildfire dynamics. The University of Virginia Forest Model Enhanced is an individual tree-based forest model that has been successfully updated and validated within the North American boreal zone. We updated the model to better characterize fire ignition and behavior in relation to litter and fire weather conditions, allowing for further …

Extreme climatic events are on the rise in Arctic regions and also outside the main growing season where they directly impact ecosystem functioning but also leave legacies for following seasons. These extreme events challenge organisms depending on strong predictable seasonal patterns. To understand the consequences of multiple seasonal extremes on Arctic ecosystem functioning there is a need for a better fundamental understanding of organisms and ecosystem processes outside the growing season.

International Journal of Applied Earth Observation and Geoinformation

Green Plastic Cover (GPC) has been widely used in construction sites of China as the main measure for dust suppression. As a unique land cover in China, GPC could be viewed as a sign for evaluating the intensity of urban construction. Therefore, time series GPC maps could provide a new perspective to understand the pattern of urban development, including both urban expansion and urban shrinkage. The objective of this study is to present a novel framework for detection of urban development patterns based on high resolution GPC maps generated from Google Earth Engine (GEE) cloud platform. Specifically, we first generated the GPC maps of 137 districts and counties in Shandong Province of China from 2018 to 2022 by combining time series spectral features and textural features derived from Sentinel-1 and Sentinel-2 images provided by GEE cloud platform. Afterwards, we designed a GPC-based urban …

International Journal of Applied Earth Observation and Geoinformation

Corn is the dominant crop planted in Northeast China, and its accurate and timely mapping is important for food security and agricultural management in China. However, the absence of enough labels is challenging for corn accurate mapping in a regional area using machine learning methods or deep learning methods. In this study, an efficient way of automatic labelling and mapping of corn planted areas by combining Global Ecosystem Dynamics Investigation (GEDI) data and Sentinel-2 images is proposed. We explore the height and vertical structure differences between corn and other crops derived from GEDI features and generate labels automatically by referencing crop type products and transferring models from historical years. The trained learning networks of automatic labelling from GEDI points and the trained decision trees of the Random Forest (RF) classifier can be transferred to corn mapping in …

International Journal of Applied Earth Observation and Geoinformation

The carbon sinks of North American boreal forests have been threatened by global warming and forest disturbances in recent decades, but knowledge about the carbon balance of these forests in recent years remains unknown. We tracked annual aboveground carbon (AGC) changes from 2016 to 2021 across the forest regions of NASA’s Arctic Boreal Vulnerability Experiment (ABoVE) core study domain, using Vegetation Optical Depth derived from low-frequency passive microwave observations. The results showed that these forests showed a net AGC increase of + 28.49 Tg C/yr during the study period, with total AGC gains of + 219.34 Tg C/yr counteracting total AGC losses of −190.86 Tg C/yr. Forest degradation (-162.21 Tg C/yr), defined as a reduction in the capacity of forest to provide goods and services, contributes 5 times more to the total AGC loss than forest cover loss (-28.65 Tg C/yr), defined as the …

International Journal of Applied Earth Observation and Geoinformation

Primary forests in Indonesia are in high decline. The increase in population exacerbates forest reductions, as forests are changed to settlements. This impacts the loss of biodiversity and animals in Indonesia. An important way to prevent the loss of biodiversity and animals is to conserve crucial areas. Spatial modeling can be done to detect conservation areas that are important for the future. This study analyzes the index of forest ecosystem services and habitat suitability of extinct animals to determine priority conservation areas. Here we predict forest areas that are indicated to be lost in the future. We use integrated spatial data using GIS (Geography Information System) to determine priority conservation areas. From the results, we predict that there will be a change in land cover to settlement by 2030. This causes the ecosystem service index (ESI) value to decrease from an initially high value each year. In terms of …

International Journal of Applied Earth Observation and Geoinformation

Content produced within the scope of the Agenda “NEXUS–Pacto de Inovação–Transição Verde e Digital para Transportes, Logística e Mobilidade”, financed by the Portuguese Recovery and Resilience Plan (PRR), with no. C645112083-00000059 (investment project no. 53). The research was also funded by the INOVC+ program (The InovC+ Intelligent Innovation Ecosystem of the Central Region project is co-funded by Centro 2020, Portugal 2020, through FEDER, through project BATDRON (“Aquisição de dados batimétricos por drone durante eventos de agitação marítima intensa”). Thanks are due FCT/MCTES for the financial support to CESAM,(UIDP/50017/2020+ UIDB/50017/2020+ LA/P/0094/2020), through national funds.

International Journal of Applied Earth Observation and Geoinformation

Forest Change Detection (FCD) is a critical component of natural resource monitoring and conservation strategies, enabling informed decision-making. Various methods utilizing the power of artificial intelligence (AI) have been developed for detecting and categorizing changes in forest cover using remote sensing (RS) data. One prominent AI-powered approach is the U-Net, a deep learning (DL) architecture famous for its segmentation proficiency. However, the standard U-Net architecture fails to effectively capture intricate spatial dependencies and long-range contextual information present in remote sensing imagery. To address this research gap, we introduce an attention-residual-based novel DL model which leverages the U-Net architecture and Sentinel-2 satellite images to map alterations in forest vegetation cover in the tropical region. Our novel model enhances the U-Net architecture by seamlessly integrating the strengths of the U-Net, harnessing attention mechanisms strategically to amplify crucial features, and leveraging cutting-edge residual connections to facilitate the smooth flow of information and gradient propagation. These meticulous design choices enabled the precise feature extraction, resulting in improved computational performance of the proposed method compared to the Standard U-Net, Deeplabv3+, Deep Res-U-Net, and Attention U-Net. The classification results demonstrate the enhanced efficiency of our model, achieving a Mean Intersection over Union (MIoU) of 0.9330 on our test dataset. This performance surpasses the Attention U-Net (0.9146), Standard U-Net (0.9029), Deeplabv3+(0.9247), and Deep Res-U-Net …

International Journal of Applied Earth Observation and Geoinformation

Burn severity has been widely studied. Typical approaches use spectral differencing indices from remotely sensed data to extrapolate in-situ severity assessments. Next generation geostationary data offer near-continuous fire behaviour information, which has been used for fire detection and monitoring but remains underutilized for fire impact estimation. Here, we explore the association between remotely sensed fire intensity metrics and spectral differencing severity indices to understand whether and where they describe similar wildfire effects. The commonly used Differenced Normalised Burn Ratio (dNBR) severity index was calculated for Advanced Himawari Imager (AHI − 2 km) and Sentinel-2 (20 m) data and compared to different Fire Radiative Power (FRP) metrics derived from fire hotspot detections from AHI data across Australia. The comparison was implemented through different stratifications based on …

International Journal of Applied Earth Observation and Geoinformation

Hyperspectral Image Classification (HSIC) is a challenging task due to the high-dimensional nature of Hyperspectral Imaging (HSI) data and the complex relationships between spectral and spatial information. This paper proposes a Graph-Infused Hybrid spatial–spectral Transformer (GFormer) for HSIC. The GFormer combines the power of graph and spatial–spectral transformer to capture both spectral relationships and spatial context. We represent the HSI data as a graph, where nodes represent pixels and edges capture spectral similarities. By incorporating an attention mechanism, the GFormer learns spatial–spectral fusion representations, allowing it to effectively discriminate between different classes. The model can capture long-range dependencies among spectral bands, enabling it to understand complex interactions in the HSI data. Moreover, the GFormer adapts to different spectral resolutions by …

International Journal of Applied Earth Observation and Geoinformation

The accurate estimation of biomass burning emissions has played a crucial role in air quality and climate forecast modeling. Satellite-based fire radiative power (FRP) has proven effective for calculating biomass burning emissions. However, FRP-based emission estimations in East Asia often rely on polar-orbiting satellites owing to the unstable performance of Japan Aerospace Exploration Agency Advanced Himawari Imager (JAXA AHI) from poor detection capability and unproper FRP retrieval method. To address this, we improve the FRP by machine learning based on mid-infrared (MIR) radiance method, leveraging the superior fire detection model developed in our previous study. In addition, we propose a multi-satellite distance-based weighted ensemble FRP estimation method. Compared to traditional MIR radiance methods, the machine learning-based FRP estimation model exhibited promising …

International Journal of Applied Earth Observation and Geoinformation

Topography determines run-off direction, redistributes groundwater, and affects land surface solar radiation loads and the associated evaporative forcing, consequently, topography can modulate the impact of drought and heat waves on ecosystems. This topographic modulation effect, which typically occurs at the local scale, is often overlooked when assessing ecosystem drought responses using moderate-to-coarse spatial resolution satellite observations, such as the Moderate-resolution Imaging Spectroradiometer (MODIS) imagery. Sentinel-2 and Landsat imagery with finer resolution are suitable for monitoring changes at the local scale, however, studies relying on single vegetation metrics may fail to get a holistic picture of vegetation drought responses, particularly for forests that have complex physiological mechanisms. Here, we performed a comprehensive assessment of the topographic effects on …

International journal of applied earth observation and geoinformation

A semi-analytical model is developed for estimating the spectral diffuse attenuation coefficient of downwelling irradiance (Kd(λ)) in inland and coastal waters. The model works as a function of the inherent optical properties (absorption and backscattering), depth, and solar zenith angle. Results of this model are validated using a large number of in-situ measurements of Kd(λ) in clear oceanic, turbid coastal and productive lagoon waters. To further evaluate its relative performance, Kd(λ) values obtained from this model are compared with results from three existing models. Validation results show that the present model is a better descriptor of Kd(λ) and shows an overall better performance compared to the existing models. The applicability of the present model is further tested on two Hyperspectral Imager for the Coastal Ocean (HICO) remote sensing images acquired simultaneously with our field measurements. The …

International Journal of Applied Earth Observation and Geoinformation

Detailed maps on the spatial and temporal distribution of crops are key for a better understanding of agricultural practices and for food security management. Multi-temporal remote sensing data and deep learning (DL) have been extensively studied for deriving accurate crop maps. However, strategies to solve the problem of transferring crop classification models over time, e.g., training the model with data for a recent year and mapping back to the past, have not been fully explored. This is due to the lack of a generalized method for aggregating optical data with regard to the irregularity in annual clear sky observations and the scarcity of multi-annual crop reference data to support a more generalized DL model. In this study, we tackled these challenges by introducing a method namely Temporal Encoding (TE) to capture the irregular phenological information. Subsequently, we adapted and integrated two methods, i …

International Journal of Applied Earth Observation and Geoinformation

Accurate and up-to-date mapping of the human population is fundamental for a wide range of disciplines, from effective governance and establishing policies to disaster management and crisis dilution. The traditional method of gathering population data through census is costly and time-consuming. Recently, with the availability of large amounts of Earth observation data sets, deep learning methods have been explored for population estimation; however, they are either limited by census data availability, inter-regional evaluations, or transparency. In this paper, we present an end-to-end interpretable deep learning framework for large-scale population estimation at a resolution of 1 km that uses only the publicly available data sets and does not rely on census data for inference. The architecture is based on a modification of the common ResNet-50 architecture tailored to analyze both image-like and vector-like data …

International Journal of Applied Earth Observation and Geoinformation

Meteorological drought precedes the agricultural drought and studying the propagation time from meteorological to agricultural drought can substantially reduce agricultural losses. To find this propagation time between the meteorological and agricultural drought, this study analyzed the copula-based conditional probability between the Standardized Precipitation Index at 12 timescales (SPI-1 to12, meteorological drought) and Standardized Soil Moisture Index at 1 month timescale (SSI-1, agricultural drought), over the fifteen Agro-Climatic Zones (ACZs) of India. SSI is computed using a total column Soil Moisture (SM) derived from ESACCI SM using the Statistical Soil Moisture Profile (SSMP) model. The SSMP-based ESACCI SM is positively correlated (Correlation Coefficient of 0.871) with ERA5 Land SM. To compute the conditional probability, three copulas namely Frank, Clayton, and Gumbel copulas are fitted …

International Journal of Applied Earth Observation and Geoinformation

Spatial scene similarity plays a crucial role in spatial cognition, as it enables us to understand and compare different spatial scenes and their relationships. However, understanding spatial scenes is a complex task. While existing literature has contributed to spatial scene representation learning, these methods primarily focus on comprehending the spatial relationships among objects, often neglecting their semantic features. Furthermore, there is a lack of scene representation learning methods that can seamlessly handle different types of spatial objects (e.g., points, polylines, and polygons) in a scene. Moreover, since expert knowledge is required for the annotation process of spatial scene understanding, publicly available high-quality annotation data has a limited size which usually leads to suboptimal results. To address these issues, we propose a novel multi-scale spatial scene encoding model called …

International Journal of Applied Earth Observation and Geoinformation

Urban trees provide important ecosystem services to improve cities’ liveability and sustainability. Leaf chlorophyll content (C ab) estimation by remote sensing can help monitor tree health efficiently. However, the C ab retrieval of urban trees is challenging because of the complex canopy structures, backgrounds, and illuminations conditions. This paper proposed an automatic method for partitioning sunlit/shaded pixels and removal of bright-specular/dark-hole and background pixels. In addition, we proposed a new index, the Urban Tree Chlorophyll Index (UTCI), defined as UTCI=(ρ 709-ρ 697)/(ρ 709-ρ 686), based on the simulated hyperspectral images of urban tree using radiative transfer model. This proposed UTCI index outperforms existing narrow-band indices (NBIs) for estimating C ab for complex canopy structures, backgrounds, and illumination conditions evaluated using simulated hyperspectral images …

International Journal of Applied Earth Observation and Geoinformation

The inversion of inherent optical properties (IOPs) and chlorophyll a (Chla) is one of the key objectives in water color remote sensing, and hyperspectral remote sensing with rich spectral information makes precise inversion possible. In this study, we developed a semi-analytical estimation method for inland water IOPs based on the quasi-analytical algorithm (QAA). Considering the complex optical characteristics of inland waters, empirical parameter regional optimization was conducted. Furthermore, a dual-band joint inversion strategy and Gaussian function fitting method were utilized to optimize the solution processes for the backscattering coefficient of particles (bbp) and the absorption coefficient of phytoplankton pigments (aph), respectively. This approach overcomes the limitations of single-band bbp inversion in inland waters. In addition, it directly decomposes the absorption coefficient to obtain aph using a …

International Journal of Applied Earth Observation and Geoinformation

Satellite-derived bathymetry plays a significant role in characterizing river systems, furnishing invaluable insights for applications such as flood risk management. In this paper, we introduce a bathymetry inversion framework, namely reflectance transformation - convolutional neural network (RT-CNN), geared towards long river segments, exemplified by a 210-kilometer stretch of the Bei River's in South China. The framework hinges on a neural network driven by a reflectance transformation model, an optical construct that harnesses radiance data from visible, near-infrared, and shortwave infrared bands. This modified input undergoes spatial–temporal fusion and convolutional neural network processing, culminating in high-resolution (10-meter) bathymetric estimations. The RT-CNN approach adeptly blends Landsat and Sentinel imagery with a substantial dataset of over 18,000 SONAR measurements, integrating …

International Journal of Applied Earth Observation and Geoinformation

The 2014 Northwest Territories fires are one of the largest wildfires in history. However, it is difficult to explain what caused such devastating wildfires simply with meteorological conditions and hydrological drought. There is a lack of large-scale Near-Real-Time (NRT) observations that characterize fuel conditions. To fill this research gap, we provide the new earth observations that the meso-scale vegetation heat represented by L-band microwave-retrieved fuel (or canopy) temperature serves as a predictor of fire spread and lightning. We studied two million-ha-scale extreme fire events in the Northwest Territories in 2014 and British Columbia in 2018 to demonstrate that preheated endothermic vegetation condition (canopy temperature>295 K) ahead of flaming is a prerequisite for mega-fires. Canopy temperature is thus proposed as an indicator to modulate convective heating ahead of combustion, and fire spread …

International Journal of Applied Earth Observation and Geoinformation

The success of transformer networks in the natural language processing and 2D vision domains has encouraged the adaptation of transformers to 3D computer vision tasks. However, most of the existing approaches employ standard backpropagation (SBP). SBP requires the storage of model activations on a forward pass for use during the backward pass, making their memory complexity linearly proportional to model depth, hence, inefficient. Furthermore, most 3D point transformers use the classic QKV matrix multiplication design which comes with a memory bottleneck. To address these issues, we propose a memory-efficient point transformer that makes use of reversible functions and linearized self-attention to minimize SBP and transformer memory complexities, respectively. Additionally, rather than the usual UNet architectural design for segmentation, we adopt a ∇-shaped design to capture multi-size …