A modified version of a Handix Scientific printed optical particle spectrometer is introduced. The paper presents characterization experiments, including concentration, size, and time responses. Integration of an external multichannel analyzer card removes counting limitations of the original instrument. It is shown that the high-resolution light-scattering amplitude data can be used to sense particle-phase transitions.

A novel approach is used for the determination of greenhouse gas (GHG) emissions of small rural settlements, which may significantly differ from those of urban regions and have hardly been studied yet. Among other results, it turned out that wintertime nitrous oxide emission is significantly underestimated in the official emission inventories. Given the large number of such settlements, the underestimation may also distort the national total emission values reported to international databases.

The Global Precipitation Measurement mission is an international satellite mission providing regular global rain measurements. We present two newly developed machine-learning-based implementations of one of the algorithms responsible for turning the satellite observations into rain measurements. We show that replacing the current algorithm with a neural network improves the accuracy of the measurements. A neural network that also makes use of spatial information unlocks further improvements.

Atmospheric samples can be complex, and current analysis methods often require substantial human interaction and discard potentially important information. To improve analysis accuracy and computational cost of these large datasets, we developed an automated analysis algorithm that utilizes a factor analysis approach coupled with a decision tree. We demonstrate that this algorithm cataloged approximately 10 times more analytes compared to a manual analysis and in a quarter of the analysis time.

Scene-dependent narrow-to-broadband transformations are developed to facilitate the use of observations from the Advanced Baseline Imager (ABI), the primary instrument on GOES-R, to derive surface shortwave radiative fluxes. This is a first NOAA product at the high resolution of about 5 k over the contiguous United States (CONUS) region. The product is archived and can be downloaded from the NOAA Comprehensive Large Array-data Stewardship System (CLASS).

Fog formation is highly sensitive to near-surface temperatures and humidity profiles. Passive remote sensing instruments can provide continuous measurements of the vertical temperature and humidity profiles and liquid water content, which can improve fog forecasts. Here we compare the performance of collocated infrared and microwave remote sensing instruments and demonstrate that the infrared instrument is especially sensitive to the onset of thin radiation fog.

Methane is a major greenhouse gas and contributor to climate change with various human-caused and natural sources. Currently, atmospheric methane is expensive to sense. We investigate repurposing cheap methane safety sensors for atmospheric sensing, finding several promising sensors and identifying some of the challenges in this approach. This work will help in developing inexpensive sensor networks for methane monitoring, which will aid in reducing methane leaks and emissions.

Geophysical validation of the Integrated Water Vapour (IWV) product from the Sentinel-3 Ocean and Land Colour Instrument (OLCI) was performed against reference observations from SUOMINET and IGRA databases. Results for cloud-free matchups over land show a wet bias of 7 %–10 % for OLCI, with a high correlation against the reference observations (0.98 against SUOMINET and 0.90 against IGRA). Special attention is given to validation of uncertainty estimates and cloud flagging.

Optical array probes are instruments used aboard research aircraft to capture 2D images of ice or water particles in clouds. This study presents a new tool using innovative machine learning, called convolutional neural networks, designed to identify the shape of imaged ice particles for two of these imagers, namely 2DS and PIP. Such a tool will be a very strong asset for understanding cloud microphysics. Beyond traditional evaluation metrics, human inspections were performed of unknown data.

For atmospheric science and weather prediction, it is important to make water vapor measurements in real time. A low-cost lidar instrument has been developed by Montana State University and the National Center for Atmospheric Research. We developed an advanced signal-processing method to extend the scientific capability of the lidar instrument. With the new method we show that the maximum altitude at which the MPD can make water vapor measurements can be extended up to 8 km.

A convolutional neural network (CNN) is introduced to retrieve cloud optical thickness (COT) from passive cloud imagery. The CNN, trained on large eddy simulations from the Sulu Sea, learns from spatial information at multiple scales to reduce cloud inhomogeneity effects. By considering the spatial context of a pixel, the CNN outperforms the traditional independent pixel approximation (IPA) across several cloud morphology metrics.