Bryan Roscoe

The University of Texas at Dallas, Goddard Space Flight Center

Papers

Total Citations

281

H-Index

About

Bryan Roscoe’s research lies at the intersection of atmospheric science, optical sensing, and machine learning, with a focus on improving how we measure and understand air quality and greenhouse gas emissions. His most influential work pioneered the development of compact, low-power laser-based sensors for unmanned aerial vehicles (UAVs), enabling the first practical measurements of trace greenhouse gases from drone platforms—a breakthrough for remote and hard-to-reach environments. His 2012 paper on this topic has garnered 134 citations, underscoring its importance in advancing lightweight atmospheric sensing. Roscoe is equally recognized for his innovative use of machine learning to correct systematic biases in satellite-derived aerosol optical depth (AOD). His 2009 study, with 122 citations, demonstrated how neural networks and support vector machines could dramatically improve the accuracy of MODIS aerosol data by aligning it with ground-based Aerosol Robotic Network measurements. This work has direct implications for public health, as accurate AOD estimates are critical for assessing harmful particulate matter exposure. Together, Roscoe’s contributions—spanning sensor engineering and data-driven bias correction—have provided essential tools for environmental monitoring and climate research.