Environmental Science is experiencing new and motivating challenges, especially when it comes to the Internet of Things (IoT) and Big Data analysis. Each day, an increasing number of devices are connected to the Internet and send information minute by minute in real-time, which provides a massive growth of many databases worldwide. In atmospheric science in particular, the combination of IoT with low-cost air quality sensors drives an increasing volume of research in public institutions and regular communities, by way of citizen science projects. There is no doubt that future cities will produce substantial amounts of information provided by IoT sensors and are expected to deal with the quick management of databases. As we are currently experiencing the beginning of the era of Big Data, increasing usage of tools such as Matlab, Python, and R, allow us to extract, structure, visualize, and analyze air quality data in ways never before possible.
Low-cost sensors for measuring air quality have received extensive scientific and commercial attention in recent years. There is a consensus that the current generation still needs to be improved in order to produce data with an accuracy level equivalent to a reference instrument. However, a growing number of investigations show the possibility of “calibrating” the sensors using statistical methods based on co-located monitoring with reference instruments. Statistical techniques range from multiple linear regression to more complex methods involving machine learning algorithms, such as random forest and artificial neural networks. The goal of this Research Topic is to generate studies that allow deepening the knowledge about the variations introduced by environmental parameters and advance in the applicability of low-cost sensors in topics such as methods of calibrations and data quality assurance.
The Research Topic welcomes papers on air quality, focusing on low-cost sensor technologies and IoT designs. We welcome papers presenting novel data, processing methods, and advanced techniques in big data such as artificial intelligence and/or machine learning algorithms. Works using remote sensing or GIS information to construct sensor nodes and predictive models are also welcome. We encourage authors to submit work from databases representing real-world measurements based on field campaigns or citizen science research projects. We expect that papers collected in this Research Topic make progress in managing new and emerging tools, such as low-cost sensors, which can help a better application on future air quality monitoring networks.
Cover image made by Magdalena Ahumada Varela.
Environmental Science is experiencing new and motivating challenges, especially when it comes to the Internet of Things (IoT) and Big Data analysis. Each day, an increasing number of devices are connected to the Internet and send information minute by minute in real-time, which provides a massive growth of many databases worldwide. In atmospheric science in particular, the combination of IoT with low-cost air quality sensors drives an increasing volume of research in public institutions and regular communities, by way of citizen science projects. There is no doubt that future cities will produce substantial amounts of information provided by IoT sensors and are expected to deal with the quick management of databases. As we are currently experiencing the beginning of the era of Big Data, increasing usage of tools such as Matlab, Python, and R, allow us to extract, structure, visualize, and analyze air quality data in ways never before possible.
Low-cost sensors for measuring air quality have received extensive scientific and commercial attention in recent years. There is a consensus that the current generation still needs to be improved in order to produce data with an accuracy level equivalent to a reference instrument. However, a growing number of investigations show the possibility of “calibrating” the sensors using statistical methods based on co-located monitoring with reference instruments. Statistical techniques range from multiple linear regression to more complex methods involving machine learning algorithms, such as random forest and artificial neural networks. The goal of this Research Topic is to generate studies that allow deepening the knowledge about the variations introduced by environmental parameters and advance in the applicability of low-cost sensors in topics such as methods of calibrations and data quality assurance.
The Research Topic welcomes papers on air quality, focusing on low-cost sensor technologies and IoT designs. We welcome papers presenting novel data, processing methods, and advanced techniques in big data such as artificial intelligence and/or machine learning algorithms. Works using remote sensing or GIS information to construct sensor nodes and predictive models are also welcome. We encourage authors to submit work from databases representing real-world measurements based on field campaigns or citizen science research projects. We expect that papers collected in this Research Topic make progress in managing new and emerging tools, such as low-cost sensors, which can help a better application on future air quality monitoring networks.
Cover image made by Magdalena Ahumada Varela.