Spatial Resolution Satellite Data Research Articles

Spatial Distribution of Thaw Depth in Palsas Estimated From Optical Unoccupied Aerial Systems Data

ABSTRACTMaximum seasonal thaw depth, referred to as active layer thickness (ALT), is one of the key parameters used to monitor permafrost conditions. ALT maps based on interpolation of point measurements or derived from coarse or moderate spatial resolution satellite data often hide small‐scale spatial variations in thaw depth resulting from differences in surface characteristics and microtopography. To model and predict changes in hydrological and biogeochemical processes in permafrost areas accurately, high‐resolution remote sensing‐based estimations of ALT are needed. Therefore, we applied random forest (RF) regression on a set of topographical and spectral vegetation indices derived from optical unoccupied aerial systems data, Landsat 8 land surface temperature (LST) data, and field measurements to estimate thaw depths in palsas at three mires in north‐west Finland. We also analyzed differences in thaw depths between mires located at different elevations, between dome and plateau‐shaped palsas, and between different vegetation and surface cover classes. The RF models resulted in root mean square errors from 2.4 to 5.7 cm between predicted and observed thaw depths and the R2 values of 0.57–0.96. Height from the surrounding fen surface and LST were the most important variables in thaw depth models, although high‐accuracy results were also achieved without LST. The mean thaw depths did not differ between the sites with lowest and highest elevation, whereas the thaw depths were significantly deeper in dome‐shaped palsas compared to plateaus. The thaw depths were significantly different between vegetation cover classes only on plateau‐shaped palsas. The results indicate the high impact of the topography on the palsa thaw depth, thus highlighting the importance of accurate elevation models in spatial modeling of palsa ALT. The methodology presented in this study can be applied to other permafrost regions where field measurements of ALT are accompanied with high‐resolution topographical and multispectral data.

Photoperiod alone does not explain the variations of leaf senescence onset across Europe

The onset of autumnal leaf senescence (TLSO) plays a crucial role in understanding and modeling how plants prepare for winter dormancy, but it has remained under-explored. Previous research, based on limited in situ observations, have proposed that the year-to-year invariant photoperiod is the primary trigger of TLSO, indicating potential insensitivity to climate change. This study verifies this hypothesis using high spatial resolution satellite data and ground-based observations throughout the European continent. Surprisingly, our findings reveal a widespread delay in TLSO during 2001–2019 (approximately 73 %, significant in 19 % of the continent, p < 0.05), closely aligned with delayed leaf senescence end (TLSD) as reported in previous studies. Moreover, as the delaying trend was larger for TLSO than for TLSD, it is apparent that climate change does not only postpone the onset of the leaf senescence process but also shortens its duration across the continent. This finding indicates that a reduced duration for plants to get ready for dormancy could potentially undermine their resilience to cold and other winter stresses. Contrary to the hypothesized photoperiodic control, TLSO exhibited opposing inter-annual temporal trends along temperature gradients, despite similar daylengths. Ground based observations failed to support photoperiodic influence either, highlighting the substantial spatial variation in TLSO responses to climate change. Intriguingly, complex climate-TLSO interactions were observed, with radiation dominating TLSO changes in 37.0 % of the study area. This analysis underscores the necessity of considering multiple variables to comprehend the patterns of TLSO at continental and vegetation scales.

Improving spatial resolution of chlorophyll-a in the Mediterranean sea based on machine learning

The objective of this study is to increase the spatial resolution of data on the level of chlorophyll-a in the Mediterranean Sea using satellite images and ground measurements. The goal of this study is to develop an information technology based on machine learning to create chlorophyll-a concentration maps with high spatial resolution for the pilot areas of the Mediterranean Sea. Traditional ground-based methods for measuring chlorophyll-a are time-consuming, expensive, and have limited spatial and temporal coverage. Therefore, satellite observations have become an effective tool for monitoring chlorophyll-a over large areas. Low spatial resolution satellite data such as GCOM-C/SGLI and Sentinel-3 OLCI allow measurements of chlorophyll-a concentration at the sea surface. However, these data have limited accuracy and spatial resolution, which creates challenges for monitoring local changes in coastal zones and small water areas. Tasks: to analyze available satellite data and ground-based measurements of chlorophyll-a for the Mediterranean Sea; to investigate the correlation between satellite data of different spatial resolutions and ground measurements; to select informative features from satellite data for building machine learning models; and to develop models for increasing the spatial resolution of chlorophyll-a based on regression and machine learning algorithms. Obtained results: information technology combining satellite data with ground measurements in the Google Earth Engine cloud platform is proposed; correlations between satellite measurements of chlorophyll-a and ground data are investigated; models based on Random Forest and Multilayer Perceptron with coefficients of determination up to 0.36 and correlation of 0.6 with test data are built; chlorophyll-a maps with a spatial resolution of 10 m are created for the pilot area near Cyprus. Conclusions. The developed information technology allows the effective combination of satellite data of different spatial resolutions and ground measurements to increase the accuracy and detail of chlorophyll-a maps in the Mediterranean Sea. Further research involves improving the preprocessing of satellite data, using more features, involving data from other regions, and applying more sophisticated machine learning models.

Implementing a dynamic representation of fire and harvest including subgrid-scale heterogeneity in the tile-based land surface model CLASSIC v1.45

Abstract. Canada's forests play a critical role in the global carbon (C) cycle and are responding to unprecedented climate change as well as ongoing natural and anthropogenic disturbances. However, the representation of disturbance in boreal regions is limited in pre-existing land surface models (LSMs). Moreover, many LSMs do not explicitly represent subgrid-scale heterogeneity resulting from disturbance. To address these limitations, we implement harvest and wildfire forcings in the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC) land surface model alongside dynamic tiling that represents subgrid-scale heterogeneity due to disturbance. The disturbances are captured using 30 m spatial resolution satellite data (Landsat) on an annual basis for 33 years. Using the pan-Canadian domain (i.e., all of Canada south of 76° N) as our study area for demonstration, we determine the model setup that optimally balances a detailed process representation and computational efficiency. We then demonstrate the impacts of subgrid-scale heterogeneity relative to standard average individual-based representations of disturbance and explore the resultant differences between the simulations. Our results indicate that the modeling approach implemented can balance model complexity and computational cost to represent the impacts of subgrid-scale heterogeneity resulting from disturbance. Subgrid-scale heterogeneity is shown to have impacts 1.5 to 4 times the impact of disturbance alone on gross primary productivity, autotrophic respiration, and surface energy balance processes in our simulations. These impacts are a result of subgrid-scale heterogeneity slowing vegetation re-growth and affecting surface energy balance in recently disturbed, sparsely vegetated, and often snow-covered fractions of the land surface. Representing subgrid-scale heterogeneity is key to more accurately representing timber harvest, which preferentially impacts larger trees on higher quality and more accessible sites. Our results show how different discretization schemes can impact model biases resulting from the representation of disturbance. These insights, along with our implementation of dynamic tiling, may apply to other tile-based LSMs. Ultimately, our results enhance our understanding of, and ability to represent, disturbance within Canada, facilitating a comprehensive process-based assessment of Canada's terrestrial C cycle.

Analysis of 10-m Sentinel-2 imagery and a re-normalization approach reveals a declining trend in the latest rubber plantations in Xishuangbanna

Natural rubber is one of the four main raw materials of industry and a globally important strategic material. Rubber plantations with agroforestry attributes have undergone rapid growth in the northern part of mainland Southeast Asia (including Xishuangbanna) since the 1990s, leading to a series of eco-environmental effects on biological diversity, climate change, carbon stocks, and hydrological processes. Accurate, detailed, and updated spatial information on rubber plantations, therefore, is fundamental for developing efficient management strategies. Currently, the booming rubber plantations are being experienced new changes in Xishuangbanna. The commonly used coarser spatial resolution satellite data (e.g., 250-m MODIS and/or 30-m Landsat) have limited applications because most plantations are often small and scattered, especially in the mountainous areas. Here, we developed a straightforward and effective re-normalization of red green normalized difference vegetation index (ReNDI) approach based on finer resolution (10-m) Sentinel-2 imagery, and then mapped the first annual 10-m rubber plantations in the entire Xishuangbanna during 2018–2021. Interestingly, rubber plots no longer expanded as rapidly as in past decades and even decreased slightly. The latest area of rubber plantations was 2514.7 km2, a decrease of 31.9 km2 comparing since 2018, with an average overall accuracy and kappa coefficient of four-year reached up to 95.37 % and 0.90, respectively. Among them, Jinghong city and Menghai county increased by 52.4 km2 and 14.8 km2 respectively, and Mengla county decreased by 99.1 km2. The main area of rubber expansion was around the Jinghong city, where rubber was initially planted. More importantly, owing to pursuing increased economic benefits and requirements of environmental protection, a considerable portion of the rubber encroachment has disappeared, particularly near roads and rivers. Our phenology-based ReNDI algorithm not only enriches the remotely-sensed methods for other industrial plantations mapping, but also provides a new chance to understanding the current patterns of rubber plantations (i.e., increase and decrease), which will contributed to rational planting planning and agroforestry management in the future, especially in the tropics.

Validation of satellite-derived water-leaving reflectance in contrasted French coastal waters based on HYPERNETS field measurements

Since 2021, two autonomous HYPERNETS (A new hyperspectral radiometer integrated in automated networks of water and land bidirectional reflectance measurements for satellite validation) stations are operated in contrasted French coastal waters: one in the center of an optically complex coastal lagoon and one at the mouth of a highly turbid estuary. These stations perform predefined sequences of above-water hyperspectral radiometric measurements following a strict viewing geometry. The data recorded by the ®HYPSTAR radiometer is automatically transmitted to servers for quality-controls then computation of the water-leaving reflectance signal. Numerous matchups were identified with high (Sentinel2-MSI and Landsat8/9-OLI) and medium (Sentinel3-OLCI and Aqua-MODIS) spatial resolution satellite data and are analyzed to assess the performance of different atmospheric correction algorithms (Sen2Cor, ACOLITE, POLYMER, iCOR, C2RCC, GRS, BPAC, NIR-SWIR). Considering the specifications of each site (i.e., spatial and temporal variations of water optical properties), optimized matchup protocols are first established to guaranty high quality comparisons between satellite products and field measurements. The matchup results highlight the failure and limits of several atmospheric correction algorithms in complex/turbid coastal waters. The importance of accurate sun glint corrections in low to moderately-turbid waters (with the good performances of POLYMER, C2RCC and GRS processors, e.g., errors (MAPE) lower than 25% in the green spectral region) is also shown while the use of dark targets and spectral fitting to estimate the aerosol contributions is proved to be the most accurate method in the case of turbid waters (with Sen2Cor and ACOLITE errors (MAPE) lower than 20% in the visible and near-infrared spectral regions).

Cascading Machine Learning to Monitor Volcanic Thermal Activity Using Orbital Infrared Data: From Detection to Quantitative Evaluation

Several satellite missions are currently available to provide thermal infrared data at different spatial resolutions and revisit time. Furthermore, new missions are planned thus enabling to keep a nearly continuous ‘eye’ on thermal volcanic activity around the world. This massive volume of data requires the development of artificial intelligence (AI) techniques for the automatic processing of satellite data in order to extract significant information about volcano conditions in a short time. Here, we propose a robust machine learning approach to accurately detect, recognize and quantify high-temperature volcanic features using Sentinel-2 MultiSpectral Instrument (S2-MSI) imagery. We use the entire archive of high spatial resolution satellite data containing more than 6000 S2-MSI scenes at ten different volcanoes around the world. Combining a ‘top-down’ cascading architecture, two different machine learning models, a scene classifier (SqueezeNet) and a pixel-based segmentation model (random forest), we achieved a very high accuracy, namely 95%. These results show that the cascading approach can be applied in near-real time to any available satellite image, providing a full description of the scene, with an important contribution to the monitoring, mapping and characterization of volcanic thermal features.

Estimating of heavy metal concentration in agricultural soils from hyperspectral satellite sensor imagery: Considering the sources and migration pathways of pollutants

The evolution of hyperspectral remote sensing and artificial intelligence technologies has led to a surge in their application for predicting Soil Heavy Metal Concentrations (SHMC). Nevertheless, the preponderance of existing research within this sphere centers around data procured from ground-based and airborne hyperspectral sources. Studies that employ satellite-based methodologies typically rely on medium spatial resolution hyperspectral or multispectral satellite data. The application of high spatial and spectral resolution satellite data, such as that obtained from GaoFen-5 (GF-5), remains conspicuously underexplored. Furthermore, the impact of geographical environmental factors (GEFs) on the accuracy of predictions has been infrequently considered. In the context of this backdrop, the present study introduces stacking models designed to estimate SHMC. This approach integrates reflectance spectral features (SFs) derived from GF-5 hyperspectral imagery and GEFs, including topography and pollution sources. The results demonstrate a notable improvement in the predictive accuracy of SHMC using our Stacking model, as compared to single models. The incorporation of GEFs into the method results in a varying degree of reduction in the Root Mean Square Error (RMSE), along with an enhancement in the R2 on the training set. The predictive performance improvement is most prominent for Cd and As, with the RMSE decreasing by 52% and 48%, respectively. Notably, apart from Pb, there is an improvement in performance for all elements within the test set. This study confirms the effectiveness of integrating GEFs into SHMC prediction models to enhance accuracy. Applying this technique to predict soil pollution at a regional scale and to demarcate heavily polluted areas can yield satisfactory results. In the future, we plan to apply this technique to other research areas or datasets to expand its universality. Furthermore, we aim to delve more deeply into the potential of GEFs to enhance the predictive capacity of SHMC.

Permanent & Non-Permanent Water Identification Based on Image Fusion of Sentinel 1 & 2 Data (Case Study: Lamongan Regency)

The Earth’s surface consists of land and water. Surface water is separated into two groups: permanent water and temporary water. Flooding is temporary water that is often experienced in many places in Indonesia, including Lamongan Regency. In this area, at the beginning of 2022, seasonal floods caused many residents’ houses, agricultural land, and access roads to be submerged in water. This is due to its topography, which is characterized by lowlands and water. Land cover monitoring in large areas can be identified and analyzed using remote sensing. In this case, remote sensing is used to obtain information about surface water for separating permanent water and temporary water for comprehensive treatment in the long term. This study uses two primary sources of data: Sentinel-1 and Sentinel-2 satellite images from five years of observation from January 2018 to December 2022. Sentinel-1 is used to exploit its Synthetic Aperture Radar (SAR) sensor, which is sensitive to dielectric properties; thus, the water objects can be identified easily. Meanwhile, the Sentinel-2 satellite provides up to 10 meters of spatial-resolution satellite data. Processing was carried out using Image Fusion with Fuzzy Logic Min to merge the advantages of different data sources. The result of processing obtained that the probability of permanent water distribution in Lamongan Regency from 2018 to 2022 has differences in the total area every year. The total area of permanent water from Image Fusion is 12.769 km2. Meanwhile, the total area for flood inundation is 57.023 km2. A comparison of the processing results with comparative data is carried out for the calculation of the accuracy test. Overall accuracy for permanent water is 91%, with a kappa coefficient of 0.82. Then an overall accuracy for flood inundation obtained 90% with a kappa coefficient is 0.79. The result of this study is a permanent water map and flood inundation map of the Lamongan Regency, which is expected to be a reference in flood disaster mitigation activities.

A comparison of moderate and high spatial resolution satellite data for modeling gross primary production and transpiration of native prairie, alfalfa, and winter wheat

Although agroecosystems have a significant potential to offset carbon dioxide (CO2), the amount of CO2 captured can vary significantly depending on management practices. Accurate estimation of gross primary production (GPP) and transpiration (T) of agroecosystems at the field scale are essential for the study of food security and water resource management. To date, the carbon and water fluxes data products for commercial agroecosystems are limited, mostly at the moderate spatial resolution (MSR, hundreds of meters), which cannot be used to assess the temporal dynamics of GPP and T at the field scale. This study used the vegetation photosynthesis model (VPM) and vegetation transpiration model (VTM) to estimate field-level daily GPP (GPPVPM) and T (TVTM), respectively, in native prairie, alfalfa (Medicago sativa L.), and winter wheat (Triticum aestivum L.) in central Oklahoma, USA. We evaluated the reliability and advantages of vegetation indices (enhanced vegetation index, EVI and land surface water index, LSWI) in monitoring the land surface phenology using moderate spatial resolution data from Moderate Resolution Imaging Spectroradiometer (MODIS) and high spatial resolution (HSR, tens of meters) data from Landsat and Sentinel-2. The accuracy of GPPVPM and TVTM estimates at different spatial scales was evaluated using GPP (GPPEC) and evapotranspiration (ETEC) from the eddy flux tower sites, respectively. Results demonstrate the capacity of VPM and VTM to estimate the field-level carbon and water flux dynamics and their responses to weather conditions. The use of HSR vegetation indices helped to address certain challenges faced by MSR indices, especially in capturing the crop phenology in smaller areas with conservation measures or disturbances. The findings highlight the importance of using HSR GPP estimates to reduce uncertainty in quantifying CO2 fluxes for croplands and grasslands. The findings also demonstrate the ability of the models to track field-level vegetation phenology, carbon uptake, and water use in agroecosystems under different management practices.

Mapping sugarcane residue burnt areas in smallholder farming systems using machine learning approaches

Satellite remote sensing methods have been proven to quantify the burnt areas resulting from different fire activities reliably. However, they were reported to perform poorly to identify crop residue burnt areas, particularly in smallholder systems due to lack of frequent and high spatial resolution satellite data. In this study, we used Harmonized Landsat Sentinel-2 (HLS) observations and evaluated two machine learning classifiers (i.e., Support Vector Machines (SVM) and Artificial Neural Networks (ANN)) to map sugarcane burnt areas for the 2019–20 season in a smallholder farming region in Thailand.Results showed that both classifiers performed well in identifying the spatial patterns of sugarcane burnt areas in the region. The ANN outperformed SVM at both pixel and regional scales. At pixel level, ANN accuracy was 93.4 % while SVM's best-performing Polynomial kernel accuracy was 82.7 %. The ANN estimated average percent burnt area (51.1 %) in the region was closer to reported value (48.7 %) by Thailand's Office of Cane and Sugar Board (OCSB), compared to the SVM estimate (62.9 %). The total estimated burnt areas by ANN and SVM (315 and 418 thousand ha, respectively) deviated more from OCSB's area (240 thousand ha) compared to percent burnt area. However, area estimates from classifiers had significantly better accuracy than the estimates of MODIS burnt products.Overall, this study demonstrated that HLS observations provided required spectral information to build promising models to map burnt areas in smallholder systems with higher accuracy than global products. Our mapping algorithm using the ANN classifier showed the potential to monitor sugarcane burnt areas reliably, and contribute to the successful implementation of regulatory policies in Thailand.

Kuala Lumpur Green Space Fragmentation Index Analysis and Green Corridors Suggestion using Geographic Information System

More than half of the world’s population lives in urban areas, and the United Nations has projected that nearly all future global population growth will occur in urban areas. Urbanization has caused rapid industrial, transportation and residential development at the expense of environmental deterioration. Activities involving large-scale land clearances are conducted to satisfy human needs for expansion. The clearances have caused green area fragmentation and patches, especially in the rapidly developing urban areas. This study is aimed at detecting the current coverage of green space in the metropolitan city of Kuala Lumpur (KL), the capital city of Malaysia and monitor its fragmentation using a very high spatial resolution satellite data. Green cover which includes trees, grass, and shrubs of KL was first extracted from the SPOT 6 data using Object Oriented method which can classified the green covers. The fragmented green spaces in KL were analyzed using a range of fragmentation indices namely Number of Patches, Edge Density, Class Area, Total Landscape Area, Mean Patch Size and Mean Shape Index using the Patch Analyst tool embedded in the ArcGIS software version 10.0 to study the patterns of green patches. The connectivity of the green patches was studied and proposed to overcome the fragmentation problems. Kuala Lumpur (KL) was chosen as the case study as it is a rapidly developing city with a significant loss of green space. The urban green spaces were more fragmented as there were more built-up areas. Results of the green cover extraction reveal that the KL city centre had the lowest percentage of green cover (2.09 %) area to the urban area compared to the other five zones (Sentul Manjalara 5.03 %, Wangsa Maju Maluri 6.09 %, Damansara Penchala 11.22 %, Bukit Jalil Seputeh 7.07 % and Bandar Tun Razak Sg Besi 7.42 %). Since KL’s urban growth predominantly consists of compact building blocks, the highest land use was believed to be for commercial purposes. Around 66% of industrial and commercial activities are concentrated in the KL city centre, resulting in a diverse green space structure and a growing gap between patches in the city centre zone. The green patches were then connected to create green corridors in the city centre. Several parameters were involved in the analysis, including area, distance, types of land use (green areas and place of interest for the origin and destination), slopes under 7% value (topology with flat surfaces) and road networks. The origin and destination points were analyzed using the network analysis method in Quantum GIS to produce routes with the shortest path algorithm. Results showed that six routes were suitable for designing green corridors in the KL city centre area connecting green spaces and places of interest. All six routes were analyzed every 400 meters using Google Earth Street View to see the surrounding facilities. The Green Corridor Route 6 was the most appropriate for people, as it included parks and places of interest for city dwellers. The green corridor could then be promoted as a walkable area to minimize carbon emissions in the urban area, reducing the urban heat island phenomenon and increasing green connectivity for a better urban life.

Reconstruct the AMSR-E/2 thin ice thickness algorithm to create a long-term time series of sea-ice production in Antarctic coastal polynyas

This study presented an extended time series of ice production in Antarctic coastal polynyas for 20 years of 2002–21 using high spatial resolution satellite data from passive microwave sensors: the Advanced Microwave Scanning Radiometer for EOS (AMSR-E) and the Advanced Microwave Scanning Radiometer 2 (AMSR2). For this, we reconstructed the previously presented thin ice thickness algorithms and re-estimated ice production by replacing atmospheric input data for the heat flux calculations with the newly released ECMWF Reanalysis v5 (ERA5). The consistency of ice production between AMSR-E and AMSR2, whose operation periods do not overlap, was confirmed from comparisons with ice production by the Special Sensor Microwave/Imager (SSM/I) and the Special Sensor Microwave Imager Sounder (SSMIS) with relatively coarse spatial resolution. The open ocean area, which cannot be detected from the thin ice thickness algorithm, was defined using sea-ice concentration (SIC) from AMSR-E and AMSR2. The satellite derived SIC has been suggested to be underestimated in the new thin ice (polynya) area. The effect of the underestimation on the ice production estimation was evaluated. The underestimation of SIC from three major algorithms was also investigated from comparisons with the thin ice thickness presented in this study. The coastal polynya dataset covering 20 years with the higher spatial resolution presented in this study is vital for climate-change-related studies.

Features of the joint use of visible-range satellite and satellite radar data for monitoring arctic icebergs based on data from Franz Josef Land area

Given the availability of data, in order to monitor icebergs in the vast Arctic region, it is necessary to ensure an optimal ratio of the satellite radar data cost (with a wide swath and high spatial resolution) and the reliability of the results obtained. The aim of the study was to evaluate the effectiveness of using only non-commercial satellite information for monitoring icebergs. To achieve this goal, an iceberg detection technique was applied based on statistical criteria of searching for gradient zones in analyzing two-dimensional fields of satellite images. As a source of information, we used high spatial resolution satellite data obtained from non – commercial satellites Sentinel-1 (radar data) and Landsat-8 (visible spectral range data) with a pixel size of 40 × 20 and 15 m, respectively.Through comparison with the data of the visible spectral range of ultra-high resolution, the error was estimated of measuring the size of icebergs using the visible range equipment of the Landsat-8 satellite and the radar of the Sentinel-1 satellite. Based on the analysis of satellite data of the visible spectral range and radar data, the spatial dimensions of icebergs formed by the outlet glaciers of Franz Josef Land n 2020–2022 were estimated. The materials were collected for the period August-September, corresponding to the conditions of minimum ice extent in the Franz Josef Land area and sufficient natural illumination for using the data of the visible spectral range in the Arctic. In total, about 100 satellite images were analyzed. The sources of error in the satellite measurements are discussed, as well as the reasons for the impossibility in some cases to identify Franz Josef Land icebergs on radar images. It is shown that the largest contribution to the iceberg runoff of Franz Josef Land is made by outflow glaciers located on Georg Land, Wilczek Land, Gall, Salisbury and Champa islands. According to the results of the study, it is concluded that radar data with a pixel size of 40 m allow one to obtain essentially reliable information on the spatial distribution of icebergs in the water area. However, it is not possible to obtain a reliable function of the distribution of icebergs by size since most of the icebergs in the Franz Josef Land water area have small and medium sizes and are rarely identified in images of such spatial resolution.

A global spatial-temporal land use regression model for nitrogen dioxide air pollution

Introduction: The World Health Organization (WHO) recently revised its health guidelines for Nitrogen dioxide (NO2) air pollution, reducing the annual mean NO2 level to 10 μg/m3 (5.3 ppb) and the 24-h mean to 25 μg/m3 (13.3 ppb). NO2 is a pollutant of global concern, but it is also a criteria air pollutant that varies spatiotemporally at fine resolutions due to its relatively short lifetime (~hours). Current models have limited ability to capture both temporal and spatial NO2 variation and none are available with global coverage. Land use regression (LUR) models that incorporate timevarying predictors (e.g., meteorology and satellite NO2 measures) and land use characteristics (e.g., road density, emission sources) have significant potential to address this need.Methods: We created a daily Land use regression model with 50 × 50 m2 spatial resolution using 5.7 million daily air monitor averages collected from 8,250 monitor locations.Results: In cross-validation, the model captured 47%, 59%, and 63% of daily, monthly, and annual global NO2 variation. Daily, monthly, and annual root mean square error were 6.8, 5.0, and 4.4 ppb and absolute bias were 46%, 30%, and 21%, respectively. The final model has 11 variables, including road density and built environments with fine (30 m or less) spatial resolution and meteorological and satellite data with daily temporal resolution. Major roads and satellite-based estimates of NO2 were consistently the strongest predictors of NO2 measurements in all regions.Discussion: Daily model estimates from 2005–2019 are available and can be used for global risk assessments and health studies, particularly in countries without NO2 monitoring.

Forest status assessment in China with SDG indicators based on high-resolution satellite data

ABSTRACT To assess the status and change trend of forest in China, an indicator framework was developed using SDG sub-indicators. In this paper, we propose an improved methodology and a set of workflows for calculating SDG indicators. The main modification include the use of moderate and high spatial resolution satellite data, as well as state-of-the-art machine learning techniques for forest cover classification and estimation of forest above-ground biomass (AGB). This research employs GF-1 and GF-2 data with enhanced texture information to map forest cover, while time series Landsat data is used to estimate forest AGB across the whole territory of China. The study calculate two SDG sub-indicators: SDG15.1.1 for forest area and SDG15.2.1 for sustainable forest management. The evaluation results showed that the total forest area in China was approximately 219 million hectares at the end of 2021, accounting for about 23.51% of the land area. The average annual forest AGB from 2015 to 2021 was estimated to be 105.01Mg/ha, and the overall trend of forest AGB change in China was positive, albeit with some spatial differences.

Ecosystems threatened by intensified drought with divergent vulnerability

Global climate change is projected to increase the severity and duration of droughts across many regions of the world, which could push ecosystems across their tipping points. To protect the most vulnerable ecosystems, it is critical to understand if ecosystems are likely to collapse rapidly or have a certain buffer capacity with aggravating drought (as characterized by their onset, duration and severity). However, such ecosystem vulnerability has been rarely characterized. Here, we present a novel, multi-faceted approach to quantify the ecosystem vulnerability and associated impacts of drought characteristics across different ecosystems at the continental-scale in Europe using high spatial and temporal resolution satellite data. We observed substantially different vulnerabilities across ecosystems where the vegetation damage increases with earlier, longer and more intense droughts. In particular, irrigated croplands are under prominently high risk when facing intensified droughts with the shortest delay in response to drought (0.67 times that of other ecosystems) and a fast increase in damage to drought (1.27 times that of other ecosystems). Mixed forests have a prominently low vulnerability in terms of a notably slow increase in damage to drought (0.67 times that of other ecosystems) and a relatively long response time (longer than over half ecosystems). Moreover, the vulnerability of most ecosystems largely increases with increasing drought severity. Consequently, vegetation tends to succumb more quickly to intensified droughts. Our multi-faceted approach to ecosystem vulnerability characterization indicates that under future intensified droughts in the 21st century, the functioning of a vast range of ecosystems will be threatened.

Evaluation of the nitrogen oxide emission inventory with TROPOMI observations

To compensate for the emissions missed or underestimated in the national bottom-up emission inventories, we apply the high spatial resolution satellite data from TROPOspheric Monitoring Instrument (TROPOMI) to estimate the top-down nitrogen oxide (NOx) emissions in regional scales. The NOx chemical lifetime is derived based on ground-based measurements of ozone photolysis rate, ultra-violet (UV) index, and temperature. For five designated regions of western Taiwan, the derived lifetime is about 1–2 h in summer and 2–4 h in winter. The retrieved 2021 annual emissions for regions near two major pollution sources, Taichung thermal power plant and Mailiao Industrial Zone, are comparable with the emission from the Continuous Emission Monitoring System (CEMS), with a difference of 6% and −12%, respectively. After validating the data and methods, the NOx emissions for five regions of western Taiwan are derived and applied to evaluate the bottom-up inventories. For northern and southern Taiwan, the top-down emissions agree well with emission inventories. The top-down emissions are 12%, 23%, and 16% higher than emission inventories for north-central, central, and south-central Taiwan, respectively. This indicates that the bottom-up inventories are underestimated from north-central to south-central Taiwan, which may be associated with the uncertainties from traffic sources. Given the various complex pollution sources, deriving NOx emissions from space allows us to acquire a better understanding of emissions on urban scales and improve the bottom-up emission inventories.

Using Sentinel-1, Sentinel-2, and Planet satellite data to map field-level tillage practices in smallholder systems.

Remote sensing can be used to map tillage practices at large spatial and temporal scales. However, detecting such management practices in smallholder systems is challenging given that the size of fields is smaller than historical readily-available satellite imagery. In this study we used newer, higher-resolution satellite data from Sentinel-1, Sentinel-2, and Planet to map tillage practices in the Eastern Indo-Gangetic Plains in India. We specifically tested the classification performance of single sensor and multiple sensor random forest models, and the impact of spatial, temporal, or spectral resolution on classification accuracy. We found that when considering a single sensor, the model that used Planet imagery (3 m) had the highest classification accuracy (86.55%) while the model that used Sentinel-1 data (10 m) had the lowest classification accuracy (62.28%). When considering sensor combinations, the model that used data from all three sensors achieved the highest classification accuracy (87.71%), though this model was not statistically different from the Planet only model when considering 95% confidence intervals from bootstrap analyses. We also found that high levels of accuracy could be achieved by only using imagery from the sowing period. Considering the impact of spatial, temporal, and spectral resolution on classification accuracy, we found that improved spatial resolution from Planet contributed the most to improved classification accuracy. Overall, it is possible to use readily-available, high spatial resolution satellite data to map tillage practices of smallholder farms, even in heterogeneous systems with small field sizes.

Global volcano monitoring through the Normalized Hotspot Indices (NHI) system

The Normalized Hotspot Indices (NHI) is a multi-channel algorithm developed to map thermal anomalies through the Multispectral Instrument onboard the Sentinel-2 satellite and the Operational Land Imager onboard the Landsat-8 satellite. The algorithm runs operationally under the Google Earth Engine platform and allows the analysis of volcanic thermal features (e.g. lava flows/lakes) through plots of the number of hot pixels, the total shortwave infrared radiance and the area of the hotspot. We present here the automated module of this tool: the NHI system. This system provides automated notifications about volcanic thermal anomalies detected at the global scale over the previous 48 h whenever the NHI web site ( https://sites.google.com/view/nhi-tool ) is accessed. The results of the first six months of operation are assessed through the analysis of satellite imagery and comparison with well-established programmes for global volcano monitoring. The low false positive rate (around 15%, including vegetation fires and data issues) and the successful identification of small, high-temperature features show that the NHI system may successfully integrate information from high temporal/low spatial resolution satellite data, despite some limitations (e.g. temporal sampling of the combined Sentinel-2 and Landsat-8 observations; delay of data ingestion in the Google Earth Engine platform). The recent ingestion of Landsat-9 data within the system has further extended the performance of the NHI system in supporting the surveillance of active volcanoes from space.