Tracing vulnerable communities to ambient air pollution exposure: A geodemographic and remote sensing approach.

Most studies analyzing the effects of air pollution on disadvantaged populations use ground air quality measurements. However, ground stations are generally limited, with nearly 40% of countries having no official PM2.5 stations, not allowing air quality analysis for a significantly large share of the world's population. Furthermore, limited studies analyze community data from a geodemographic perspective, in other words, to delineate the sociodemographic profiles and geographically locate the socioeconomic groups more exposed to ambient air pollution. Therefore, a significant question arises: How can we trace vulnerable communities to air pollution in areas lacking air-quality ground data? Here, we propose a novel methodology to respond to this question. We use NO2, SO2, CO, and HCHO tropospheric column air-quality data from Sentinel-5P, a satellite that quantifies concentrations of atmospheric species from space operationally. We integrate them with census and environmental data and apply the local fuzzy geographically weighted clustering spatial machine learning method for segmentation analysis. Our findings for Bali, Indonesia, provide quantitative evidence for the benefits of this methodology in tracing and delineating the profiles of the communities most exposed to air pollution. For example, results show that communities with highly disadvantaged populations, such as unemployed (over 27.8%), low educated (over 27.9%), and children (over 22.1%) (mainly located around Bali's south and north coast touristic areas), exhibit very high values (over the 75th quartile) across the pollutants studied. The proposed method is reproducible easily, quickly, and at low cost, as it is based on freely available satellite data and not on costly ground station measurements. This will hopefully assist decision-makers in tracing the most vulnerable subpopulations, even in areas with inadequate air-quality monitoring networks, thus allowing local governments around the globe (even those that are financially weak) to achieve environmental justice and their sustainable development goals.

Information Extraction and Population Estimates of Settlements from Historic Corona Satellite Imagery in the 1960s.

The Corona satellite program was a historic reconnaissance mission which provided high spatial resolution panchromatic images during the Cold War era. Nevertheless, and despite the historic uniqueness and importance of the dataset, efforts to extract tangible information from this dataset have primarily focused on visual interpretation. More sophisticated approaches have been either hampered or unrealized, often justified by the primitive quality of this early satellite product. In the current study we attempt to showcase the usability of Corona imagery outside the context of visual interpretation. Using a 1968 Corona image acquired over the city municipality of Plovdiv, Bulgaria, we reconstruct a panchromatic 1.8 m spatial resolution georegistered image with a relative displacement Root Mean Square Error (RMSE) of 6.616 (for x dimension) and 1.886 (for y dimension) and employ segmentation and texture analysis to discern agricultural parcels and settlements' footprints. Population statistics of this past era are retrieved from national census and related to settlements' footprints. An exponential relationship between the two variables was identified by applying a semi-log regression. The high adjusted R2 value found (76.54%) indicates that Corona images offer a unique opportunity for population data analysis of the past. Overall, we showcase that the Corona images' usability extends beyond the visual interpretation, and features of interest extracted through image analysis can be subsequently used for further geographical and historical research.

Air quality development during the COVID-19 pandemic over a medium-sized urban area in Thailand.

The COVID-19 pandemic has triggered an industrial and financial slowdown due to unprecedented regulations imposed with the purpose to contain the spread of the virus. Consequently, the positive effect on the environment has been witnessed. One of the most prominent evidences has been the drastic air quality improvement, as a direct consequence of lower emissions from reduced industrial activity. While several studies have demonstrated the validity of this hypothesis in mega-cities worldwide, it is still an unsubstantiated fact whether the same holds true for cities with a smaller urban extent and population. In the present study we investigate the temporal development of atmospheric constituent concentrations as retrieved concurrently from the Sentinel-5P satellite and a ground meteorological station. We focus on the period before and during the COVID-19 pandemic over the city of Hat Yai, Thailand and present the effect of the lockdown on the atmospheric quality over this average populated city (156,000 inhabitants). NO2, PM2.5 and PM10 concentrations decreased by 33.7%, 21.8% and 22.9% respectively in the first 3 weeks of the lockdown compared to the respective pre-lockdown period; O3 also decreased by 12.5% and contrary to similar studies. Monthly averages of NO2, CO and PM2.5 for the month April exhibit in 2020 the lowest values in the last decade. Sentinel-5P retrieved NO2 tropospheric concentrations, both locally over the ground station and the spatial average over the urban extent of the city, are in agreement with the reduction observed from the ground station. Numerous studies have already presented evidence of the bettering of the air quality over large metropolitan areas during the COVID-19 pandemic. In the current study we demonstrate that this holds true for Hat Yai, Thailand; we propound that the environmental benefits documented in major urban agglomerations during the lockdown may extend to medium-sized urban areas as well.

A Workflow for Automated Satellite Image Processing: from Raw VHSR Data to Object-Based Spectral Information for Smallholder Agriculture.

Earth Observation has become a progressively important source of information for land use and land cover services over the past decades. At the same time, an increasing number of reconnaissance satellites have been set in orbit with ever increasing spatial, temporal, spectral, and radiometric resolutions. The available bulk of data, fostered by open access policies adopted by several agencies, is setting a new landscape in remote sensing in which timeliness and efficiency are important aspects of data processing. This study presents a fully automated workflow able to process a large collection of very high spatial resolution satellite images to produce actionable information in the application framework of smallholder farming. The workflow applies sequential image processing, extracts meaningful statistical information from agricultural parcels, and stores them in a crop spectrotemporal signature library. An important objective is to follow crop development through the season by analyzing multi-temporal and multi-sensor images. The workflow is based on free and open-source software, namely R, Python, Linux shell scripts, the Geospatial Data Abstraction Library, custom FORTRAN, C++, and the GNU Make utilities. We tested and applied this workflow on a multi-sensor image archive of over 270 VHSR WorldView-2, -3, QuickBird, GeoEye, and RapidEye images acquired over five different study areas where smallholder agriculture prevails.

Evaluating Sentinel-2 for Lakeshore Habitat Mapping Based on Airborne Hyperspectral Data.

Monitoring of lakeshore ecosystems requires fine-scale information to account for the high biodiversity typically encountered in the land-water ecotone. Sentinel-2 is a satellite with high spatial and spectral resolution and improved revisiting frequency and is expected to have significant potential for habitat mapping and classification of complex lakeshore ecosystems. In this context, investigations of the capabilities of Sentinel-2 in regard to the spatial and spectral dimensions are needed to assess its potential and the quality of the expected output. This study presents the first simulation of the high spatial resolution (i.e., 10 m and 20 m) bands of Sentinel-2 for lakeshore mapping, based on the satellite's Spectral Response Function and hyperspectral airborne data collected over Lake Balaton, Hungary in August 2010. A comparison of supervised classifications of the simulated products is presented and the information loss from spectral aggregation and spatial upscaling in the context of lakeshore vegetation classification is discussed. We conclude that Sentinel-2 imagery has a strong potential for monitoring fine-scale habitats, such as reed beds.