Wednesdays at 1:00 p.m. – 1:45 p.m.
(UNLESS OTHERWISE NOTED.)
Please contact Phillip M. Bitzer (961-7046) or Daniela Cornelius (961-7877) concerning any questions, suggestions, or contributions relative to the seminar schedule.
1/10/2018 — NO SEMINAR – AMS Annual Meeting
1/17/2018 — Cancelled – Snow
1/24/2018 – The impact of mesoscale convective systems on global precipitation: A multi-scale modeling study – Wei-kuo Tao (Goddard Space Flight Center), [V/C Chris Schultz]
1/31/2018 – Time-Resolved Observations of Precipitation structure and storm Intensity with a Constellation of Smallsats (TROPICS) – Brad Zavdosky (SPoRT), [V/C Phillip Bitzer]
The NASA Time-Resolved Observations of Precipitation structure and storm Intensity with a Constellation of Smallsats (TROPICS) mission is a constellation of state-of-the-science observing platforms that will measure temperature and humidity soundings and precipitation with spatial resolution comparable to current operational passive microwave sounders but with unprecedented temporal resolution. TROPICS is a cost-capped ($30M) Venture-class mission funded by the NASA Earth Science Division. The mission is comprised of a constellation of 3 unit (3U) SmallSats, each hosting a 12-channel passive microwave spectrometer based on the Micro-sized Microwave Atmospheric Satellite 2 (MicroMAS-2) developed at MIT LL. TROPICS will provide imagery near 91 and 205 GHz, temperature sounding near 118 GHz, and moisture sounding near 183 GHz. Spatial resolution at nadir will be around 27 km for temperature and 17 km for moisture and precipitation. The swath width is approximately 2000 km. TROPICS enables temporal resolution similar to geostationary orbit but at a much lower cost, demonstrating a technology that could impact the design of future Earth-observing missions. The TROPICS satellites for the mission are slated for delivery to NASA in 2019 with potential launch opportunities in 2020. The primary mission objective of TROPICS is to relate temperature, humidity, and precipitation structure to the evolution of tropical cyclone (TC) intensity. This presentation provides a brief overview of the TROPICS mission and summarizes the outcomes of the 1st TROPICS Applications Workshop, held from May 8-10, 2017 at the University of Miami. At this meeting, a series of presentations and breakout discussions in the topical areas of Tropical Cyclone Dynamics, Tropical Cyclone Analysis and Nowcasting, Tropical Cyclone Modeling and Data Assimilation, and Terrestrial Impacts were convened to identify applications of the mission data and to begin to establish a community of end-users who will be able to benefit from TROPICS. Key takeaways, partnerships, and applications will be highlighted.
2/07/2018 – Severe Storms Precipitation Physics Studies with Dual-Polarization Radar and Idealized Numerical Modeling – Matt Kumjian (The Pennsylvania State University), [V/C Larry Carey]
This talk will provide an overview of my group’s ongoing research into precipitation processes in severe convective storms. We make use of a variety of tools for these studies, including dual-polarization radar observations, detailed electromagnetic scattering calculations, and idealized storm-scale numerical modeling. Research topics to be discussed are (i) an investigation of dual-polarization radar precursor signatures in tornadic nonsupercell storms in the eastern and southeastern United States, (ii) polarimetric radar and environmental characteristics of storms that produce large accumulations of small hail, (iii) radar scattering characteristics of real hailstones, and (iv) idealized simulations of hail growth trajectories to elucidate the environmental controls on hail size. The overarching goal of this research is to provide better understanding of severe storm processes, ultimately leading to improvements in the anticipation and detection of high-impact weather events.
2/14/2018 – A tool for creating regionally calibrated high-resolution land cover data sets for the West African Sahel: Using machine learning to scale up hand-classified maps in a data- sparse environment – Mollie van Gordon (University of Berkeley), [V/C Africa Flores]
Using machine learning and high-accuracy hand-classified maps, we have developed an extensive land cover dataset and classification tool for the West African Sahel. Our classifier produces high- resolution and regionally-calibrated land cover maps for the West African Sahel, representing a significant contribution to the data available for this region.
Global land cover products are unreliable for the Sahel, and accurate land cover data for the region are sparse. To address this gap, the U.S. Geological Survey and the Regional Center for Agriculture, Hydrology and Meteorology (AGRHYMET) in Niger produced high-quality land cover maps for the region via hand-classification of Landsat images. This method produces highly accurate maps, but the time and labor required constrain the spatial and temporal resolution of the data products. By using these hand-classified maps alongside machine learning techniques, we successfully increase the resolution of the land cover maps by 1-2 orders of magnitude, from 2km-decadal resolution to 30m- annual resolution.
These high-resolution regionally calibrated land cover datasets, along with the classifier we developed to produce them, will be publically available, laying the foundation for major advances in studies of land surface processes in the region. These datasets will provide more accurate inputs for food security modeling, hydrologic modeling, analyses of land cover change and climate change adaptation efforts. The classification tool will be available to classify future years of remote sensing data over the Sahel, and can be adapted for use in other parts of the world.
Mollie Van Gordon is in her final year as a PhD candidate at UC Berkeley, studying in the Geography department with an emphasis in Computational Science and Engineering. She is also a third-year fellow of the NASA Earth and Space Sciences Fellowship program. Mollie’s work focuses on inductive data science methods applied in a data sparse environments. She studies climate, land surface and ecohydrological changes in the West African Sahel, and next year will be establishing a data methods group at the Institute for Disease Modeling.
2/21/2018 – Ground- and Space-based Lightning Observations: Instruments and Applications – Ken Cummins (U. of Arizona), [V/C Phillip Bitzer]
Since the establishment of the first U.S. national-scale lightning detection networks in the late 1980’s, lightning information has steadily grown in value and use in meteorological applications. Today, this is exemplified by the by the broad use of multiple lightning datasets by industry and weather-service forecasters, as well as by hundreds of research scientists. The ultimate expression of the importance of this information is the inclusion of the Geostationary Lightning Mapper (GLM) as one of the central instruments on the GOES-R series of satellites, with the second-of-four being launched next month.
This presentation will provide a brief overview of the electrical nature of the thunderstorm lifecycle from the perspective of ground- and space-based lightning observations. Lightning locating systems will be reviewed in terms of detection methodology, performance characteristics, and spatial coverage. The complimentary nature of lightning information, when coupled with other remote-sensing observations, will be illustrated through examples of industrial, meteorological and climatological applications. Recent findings about the performance of GLM on GOES-16 will also be presented. The talk will conclude with a brief overview of the thunderstorm-driven electrical environment we live in, highlighting the value of the fair-weather electric field for monitoring the global nature and strength of deep convection.
Field observations and laboratory experiments have provided much of our current understanding of thunderstorm electrification, but there is no equivalent of radar to fully sample the internal electrical structures of storms. The niche field of electrification modeling acts as a bridge from the laboratory data on the conditions of charge separation to produce self-consistent 3D charge structures and lightning behavior to compare against observations. Primary thunderstorm charging arises from rebounding collisions between ice particles, such that electrification simulations place greater demands on the realism of cloud microphysics processes such as the production of small ice crystals that may be less important for other applications. Aerosol concentration adds yet another layer of sensitivity. Thus one outgrowth of this work is the NSSL two-moment microphysics scheme in the WRF and CM1 models.
This presentation will provide an overview of electrification modeling and applications to small storms, severe supercells, and recent work on intensity changes in tropical cyclones. A major challenge is that lightning flash rates can vary from less than 1 per minute for weak storm to hundreds per minute for individual severe storms or large storm complexes (hurricanes, MCSs).
Some of the earliest satellites, starting with OSO (1965), ARIEL (1967), and RAE (1968), detected lightning using either optical and RF sensors, although that was not their intent. One of the earliest instruments designed to detect lightning was the PBE (1977). The use of space to study lightning activity has exploded since these early days.
The advent of focal-plane imaging arrays made it possible to develop high performance optical lightning sensors. Prior to the use of charged-coupled devices (CCD), most space-based lightning sensors used only a few photo-diodes, which limited the location accuracy and detection efficiency (DE) of the instruments. With CCDs, one can limit the field of view of each detector (pixel), and thus improve the signal to noise ratio over single-detectors that summed the light reflected from many clouds with the lightning produced by a single cloud. This pixelization enabled daytime DE to increase from a few percent to close to 90%.
The OTD (1995), and the LIS (1997), were the first lightning sensors to utilize focal-plane arrays. Together they detected global lightning activity for more than twenty years, providing the first detailed information on the distribution of global lightning and its variability. The FORTE satellite was launched shortly after LIS, and became the first dedicated satellite to simultaneously measure RF and optical lightning emissions. It too used a CCD focal plane to detect and locate lightning.
In November 2016, the GLM became the first lightning instrument in geostationary orbit. Shortly thereafter, China placed its GLI in orbit. Lightning sensors in geostationary orbit significantly increase the value of space-based observations. For the first time, lightning activity can be monitored continuously, over large areas of the Earth with high, uniform DE and location accuracy. In addition to observing standard lightning, a number of sensors have been placed in orbit to detect transient luminous events and tropospheric gamma-ray flashes.
A lineal history of space-based lightning observations will be presented as well as a discussion of the scientific contributions made possible by these instruments. In addition, relative merits of space versus ground measurements will be addressed, as well as an effort to demonstrate the complementary nature of the two approaches.
Cyber-physical systems (CPS) are engineered systems that are built from, and depend upon, the seamless integration of computational algorithms and physical components (NSF, 2016). Remote sensing can be used to engineer complex cyber-physical systems that people can use or interact with and depend upon to solve environmental problems such as urban heat hazards and water resources. For example, water resources, the major driving force on our planet, support numerous ecosystems and cultural services from maintaining biodiversity, nutrient cycling, and enhancing primary productivity; to recreation, ecotourism, transport, and other cultural uses. The pressure on water resources has been on the rise and will continue to increase in the coming years because of increased frequency of drought, urbanization, urban population growth, deforestation, increased use of fertilizers and pesticides, and spread of invasive species. Therefore, citizen science and big data driven technologies are needed for timely monitoring and implementation of mitigation and restoration measures in problematic areas. Remote Sensing and Spectroscopy Lab (RSSL) at the University of Georgia (UGA), is mainly focused on developing cyber- physical system based infrastructure for solving acute environmental problems in southeast U.S. such as urban heat hazards, wetland dieback, and cyanobacterial harmful algal blooms (CyanoHABs). This presentation discusses the principle, architecture, and implementation of three cyber-physical system research projects to monitor and mitigate the aforementioned environmental problems.
Atlantic is a remote sensing, surveying and consulting firm headquartered in Huntsville, AL. Brian Mayfield, Atlantic’s President & CEO, will present an overview of his company, its clients and its capabilities. He will also discuss two programs being developed by Atlantic for UAH students who are interested in careers in the geospatial private-sector.
3/28/2018 – SPRING BREAK
Advisor: Dan Cecil
This study further investigated how the El Niño/Southern Oscillation (ENSO) phase affects lightning production on the annual time scale in the tropics and subtropics using the Tropical Rainfall Measuring Mission (TRMM) Lightning Imaging Sensor (LIS) and the Oceanic Niño Index (ONI) for ENSO phase. The lightning data were averaged into mean annual warm, cold, and neutral ‘years’ as determined by ONI for analysis of the different phases. Regional sensitivities were identified from these mean years, and NCEP/NCAR model reanalysis data were then used to determine the leading convective mechanisms and dependent atmospheric variables at each location. The processes and variables were studied for inter-annual variance and subsequent correlation to ENSO to best describe the observed lightning deviations from year to year at the locations. This talk contains a discussion of the regional sensitivities as well as a preliminary analysis at three of the locations.
What happened to the atmosphere of Mars?
It has long been theorized in the space community that Mars had lost its atmosphere because of the lack of a magnetic field and the influence of the solar wind. Thanks to the MAVEN, Mars Atmosphere and Volatile Evolution, mission scientists have been able to gain in-situ measurements of the solar wind and magnetic fields at and around Mars. The data collected have given further evidence to support the theory that the solar wind is the culprit of the loss of the atmosphere.
Advisor: Dr. Sundar Christopher
In 2016, hail was responsible for $3.5 billion and $23 million in damage to property and crops, respectively, making it the third costliest weather phenomenon in the United States. The destructive nature of hail storms has driven research into the development of more accurate hail-prediction algorithms in an effort to improve hail-detection techniques and the emergency response to impacted areas. Recently, weather forecasting efforts have found success with deep learning neural networks that can more effectively model complex, nonlinear, dynamical phenomenon, within large datasets, through multiple stages of transformation and representation. In this study, we describe a novel approach for hail detection, via deep learning. Our deep network leverages GOES channel 3 and 4 satellite imagery, atmospheric parameters derived from MERRA-2 reanalysis data, and NCEI Storm Events Database hail reports for hail storm detection. The network is applied from 2011 to 2017 for the contiguous United States and achieved a precision of 0.732. We also assess the benefits and challenges of deep learning for hail detection and the lessons learned from this experiment.
Effects of Dam Construction on Sedimentation in the Lower Mekong Basin
Advisor: Dr. Robert Griffin and Eric Anderson
Dam construction in the Mekong Basin has many cascading effects on the ecology, economy, and hydrology of the surrounding region, yet information is relatively limited. The focus of this study is to utilize the Soil Water Assessment Tool (SWAT), a rainfall-runoff hydrologic model developed at Texas A & M, combined with remote sensing and other techniques to determine sediment trapping efficiencies of reservoirs in the Lower Mekong Basin. This study uses surface water area, land use/land cover data, and reservoir datasets created by the SERVIR-Mekong Surface Water Mapping Tool, Regional Land Cover Monitoring System, and Dam Inundation Mapping Tool (respectively) as inputs into the model. The modified sediment trapping efficiency equation (Kummu & Varis (2007) is used to calculate sediment trapping efficiency of dams commissioned after 2000. The outputs from this study can be used to inform dam operation policies, study the correlation between dams and delta subsidence, and study the impact of dams on river fisheries, which are all pressing issues in the Mekong region.
4/11/2018 – Student Seminars?
4/18/2018 – No Title – Sasha Madronich, [V/C Shanhu Lee]
4/25/2018 – Student Seminars? ATS