Engineering topics

Natasa Dacic

Publications and source records attributed to Natasa Dacic.

3 records

NASA NTRS2021

Aerosol and Cloud Detection Using Machine Learning Algorithms and Space-Based Lidar Data

Clouds and aerosols play a significant role in determining the overall atmospheric radiation budget, yet remain a key uncertainty in understanding and predicting the future climate system. In addition to their impact on the Earth’s climate system, aerosols from volcanic eruptions, wildfires, man-made pollution events, and dust storms are hazardous to aviation safety and human health. Space-based lidar systems provide critical information about the vertical distributions of clouds and aerosols that greatly improve our understanding of the climate system. However, daytime data from backscatter lidars, such as the Cloud-Aerosol Transport System (CATS) on the International Space Station (ISS), must be averaged during science processing at the expense of spatial resolution to obtain sufficient signal-to-noise ratio (SNR) for accurately detecting atmospheric features. For example, 50% of all atmospheric features reported in daytime operational CATS data products require averaging to 60 km for detection. Furthermore, the single-wavelength nature of the CATS primary operation mode makes accurately typing these features challenging in complex scenes. This paper presents machine learning (ML) techniques that, when applied to CATS data, enable detection of atmospheric features during daytime operations with a horizontal resolution of 5 km compared to the 60 km horizontal resolution often required for daytime CATS data. A Convolutional Neural Network (CNN) trained using CATS standard data products also demonstrated the potential for improved cloud-aerosol discrimination compared to the operational CATS algorithms for cloud edges and complex near-surface scenes during daytime.

lidar↗

NASA NTRS2021

Aerosol and Cloud Detection Using Machine Learning Algorithms and Space-Based Lidar Data

Clouds and aerosols play a significant role in determining the overall atmospheric radiation budget, yet remain a key uncertainty in understanding and predicting the future climate system. In addition to their impact on the Earth’s climate system, aerosols from volcanic eruptions, wildfires, man-made pollution events and dust storms are hazardous to aviation safety and human health. Space-based lidar systems provide critical information about the vertical distributions of clouds and aerosols that greatly improve our understanding of the climate system. However, daytime data from backscatter lidars, such as the Cloud-Aerosol Transport System (CATS) on the International Space Station (ISS), must be averaged during science processing at the expense of spatial resolution to obtain sufficient signal-to-noise ratio (SNR) for accurately detecting atmospheric features. For example, 50% of all atmospheric features reported in daytime operational CATS data products require averaging to 60 km for detection. Furthermore, the single-wavelength nature of the CATS primary operation mode makes accurately typing these features challenging in complex scenes. This paper presents machine learning (ML) techniques that, when applied to CATS data, (1) increased the 1064 nm SNR by 75%, (2) increased the number of layers detected (any resolution) by 30%, and (3) enabled detection of 40% more atmospheric features during daytime operations at a horizontal resolution of 5 km compared to the 60 km horizontal resolution often required for daytime CATS operational data products. A Convolutional Neural Network (CNN) trained using CATS standard data products also demonstrated the potential for improved cloud-aerosol discrimination compared to the operational CATS algorithms for cloud edges and complex near-surface scenes during daytime.

lidar↗

NASA NTRS2019

Evaluation of NASA's High-Resolution Global Composition Simulations: Understanding a Pollution Event in the Chesapeake Bay During the Summer 2017 OWLETS Campaign

Recirculation of pollutants due to a bay breeze effect is a key meteorological mechanism impacting air quality near urban coastal areas, but regional and global chemical transport models have historically struggled to capture this phenomenon. We present a case study of a high ozone (O3) episode observed over the Chesapeake Bay during the NASA Ozone Water-Land Environmental Transition Study (OWLETS) in summer 2017. OWLETS included a complementary suite of ground-based and airborne observations, with which we characterize the meteorological and chemical context of this event and develop a framework to evaluate model performance. Two publicly-available NASA global high-resolution coupled chemistry-meteorology models (CCMMs) are investigated: GEOS-CF and MERRA2-GMI. The GEOS-CF R squared value for comparisons between the NASA Sherpa C-23 aircraft measurements to the GEOS-CF resulted in good agreement (R squared: 0.67) on July 19th and fair agreement (R squared: 0.55) for July 20th. Compared to surface observations, we find the GEOS-CF product with a 25 x 25 km squared grid box, at an hourly (R squared: 0.62 to 0.87) and 15-minute (R squared: 0.64 to 0.87) interval for six regional sites outperforms the hourly nominally 50 x 50 km squared gridded MERRA2-GMI (R squared: 0.53 to 0.76) for four of the six sites, suggesting it is better capable of simulating complex chemical and meteorological features associated with ozone transport within the Chesapeake Bay airshed. When the GEOS-CF product was compared to the TOLNet LiDAR observations at both NASA Langley Research Center (LaRC) and the Chesapeake Bay Bridge Tunnel (CBBT), the median differences at LaRC were -6 to 8% and at CBBT were ± 7% between 400 to 2000 m ASL. This indicates that, for this case study, the GEOS-CF is able to simulate surface level ozone diurnal cycles and vertical ozone profiles at small scales between the surface level and 2000 m ASL. Evaluating global chemical model simulations at sub-regional scales will help air quality scientists understand the complex processes occurring at small spatial and temporal scales within complex surface terrain changes, simulating nighttime chemistry and deposition, and the potential to use global chemical transport simulations in support of regional and sub-regional field campaigns.

NASA Ozone Water-Land Environmental Transition Stu↗