Earth Science Branch

The Earth Science Branch conducts research of the Earth as a system with a focus on lightning and precipitation processes, weather and climate variability, monitoring fluxes of heat and water from the surface, and associated data management and mining activities for scientific discovery and applications for societal benefit.

Using the Geostationary Lightning Mapper (GLM) to Create Rapid-Scan Visible Imagery

NASA, Marshall Space Flight Center (MSFC), and University of Alabama at Huntsville Researchers, led by Dr. Daniel Cecil, used GOES-R Series Geostationary Lightning Mapper (GLM) products to create rapid-scan visible imagery for two test cases. GLM is normally used for mapping lightning locations, flash rates, and flash energies, not for visible imagery. But it does take visible-band (0.774 micron) pictures of its field of view before applying filters to find the lightning flashes. A visible image is transmitted from the satellite every 2.5 minutes. Those images are not calibrated or geolocated. We applied a basic calibration and geolocation for two test cases, then created animations.

The imagery from GLM has coarse (~10 km) horizontal resolution compared to ABI (~1 km), but in our test cases it does depict relevant structures in a severe thunderstorm outbreak, including overshooting cloud tops, low-level outflow boundaries, convective initiation, and atmospheric flow directions at multiple vertical levels. In another test case, it depicts rapid expansion of an ash cloud from a deadly volcanic eruption in Guatemala


Rapid-scan visible imagery from GLM can enable research and monitoring of rapidly evolving, high impact events such as severe thunderstorms and volcanic eruptions. It can also potentially enable derivation of atmospheric flow fields (wind velocity vectors) at cloud-top levels. Another GOES instrument, the Advanced Baseline Imager (ABI), is normally used for visible imagery. ABI provides scans of the Continental US and adjacent waters every five minutes, and scans of selected smaller-scale domains (up to two per satellite) every one minute. Outside of these regions, the "full-disk" scan is limited to every ten minutes (previously every 15 minutes, before April 2019). 2.5-minute imagery from GLM would provide a rapid update that is not normally available for Central and South America, or for large parts of the Atlantic and Pacific Oceans. This includes thunderstorm-prone regions in South America, volcanoes along the Andes and in Central America, and the Central Atlantic where hurricanes make their approach toward Caribbean islands, Bermuda, or eventually the US.

Advanced Microwave Precipitation Radiometer (AMPR) has excellent performance in latest deployment.

The Advanced Microwave Precipitation Radiometer (AMPR) flew approximately 140 hours in the Cloud, Aerosol, and Monsoon Processes Philippines Experiment (CAMP2Ex), with significantly improved initial data quality relative to its last deployment on the NASA P-3B in 2016. This will enable novel clear-air geophysical retrievals, such as water vapor vertical profiles, as well as ultra-high-resolution cloud microphysical retrievals in coordination with the triple-frequency radar used in CAMP2Ex.


AMPR remotely senses passive microwave signatures of geophysical parameters from an airborne platform and the instrument’s low noise system provides multi-frequency microwave imagery with high spatial and temporal resolution. The purpose of CAMP2Ex is to deconvolute tropical meteorology and aerosol science at the meso-b to cloud level. It’s best thought of as the precursor suborbital field experiment to the planned Aerosol and Cloud-Convection Precipitation (ACCP) mission. CAMP2Ex results will help define the fundamentally new science that ACCP can provide.


The AMPR team, led by Marshall’s Dr. Timothy Lang, combined a microwave active-passive remote sensing system with APR-3 and was also co-located with AVAPS. Samples were taken at a variety of altitudes. AMPR featured a new custom multifrequency radome that, when coupled with new filters for APR-3 Ka-band, provided vastly improved initial data quality from ORACLES 2016 deployment. Sampling was performed multiple times per flight during select cloud overflights. When coupled with APR-3, the instrument provided ultra-high-resolution microphysical retrievals along a nadir curtain in clouds. The instrument also enabled evaluation and improvement of calibration using air-cooled cold load.


AMPR's overall excellent performance during the campaign, coupled with the novel flight and instrument scanning maneuvers used during the campaign, will enable unique retrievals of geophysical information relevant to ACCP science, such as vertical profiles of atmospheric water vapor and combined active/passive microwave retrievals related to cloud microphysical processes.

Key Findings from a Simple Reanalysis Proxy for US Cloud-to-Ground Lightning

Researchers, including Marshall’s William Koshak, published the results of a simple reanalysis proxy for US cloud-to-ground lightning. The reanalysis showed that in the cool season (November through April), negative CG flash counts across the continental United States (CONUS) are well correlated with CONUS lightning proxy CP (CAPE X Precipitation) values, but CP proxy performance during the warm season (May through October) is worse.

The relatively strong relations between CP and CG flash counts in some regions and times of the year at daily resolution provide a benchmark for more complex proxies and suggest that proxy-based extended- and long-range prediction of lightning activity may be feasible to the extent that precipitation rate and CAPE can be predicted.

A paper by Romps et al. (2014) introduced a proxy CP (CAPE x Precipitation) for cloud-to-ground (CG) lightning over the continental US (CONUS). The paper received media attention since it pointed to an increase in CG strikes over CONUS under global warming conditions. However, the paper only conducted a one year period analysis. To more rigorously test the attributes of the CP proxy, our study analyzed CP for a 14 year period (2003-2016) using reanalysis data, thereby substantially improving the applicability and limitations of CP.

To read the full research article published in the International Journal of Climatology, click here:

Remote Sensing of Hail Damage Swaths

Intense thunderstorms can bring damaging winds and large hail that leave scars of damage to agricultural regions during the prime growing season. These hail damage swaths have frequently been observed through the use of satellite remote sensors, prominently optical instruments. A commonly used index for monitoring changes in and around suspected damaged areas is the Normalized Difference Vegetation Index (NDVI). NDVI values routinely decrease in and around the hail damage swath, sometimes significantly if the swath occurs late in the growing season when the vegetation is near maturity. The drawback to using optical data, however, is that the sensor can be restricted from viewing the surface depending on the sky and atmospheric conditions (i.e. clouds, diurnal cycle). Synthetic Aperture Radar (SAR) provides another opportunity to view the surface regardless of the sky condition and time of day.

The primary objective of this study looked to see if SAR could be used in aiding optical data in evaluating hail damage swaths. Using observations throughout the growing season, ESA's Sentinel 1A/1B SAR imagery in co- and cross-polarization is used to identify changes in backscatter of corn and soybeans damaged by hail during severe thunderstorm events in the early and late growing season across the central United States. While NDVI studies have routinely examined a decrease associated with damage, these events produced SAR signatures evident in the damage region but with direction of change believed related to vegetation structure and soil moisture conditions.

SAR Hail damage

This image shows: a) Sentinel-1 co-polarization (VV) intensity image from 12 August 2018; b) Same as a) but cross-polarization (VH); and a) Time series for NDVI for the damaged and background areas of corn crops identified by the 2018 Crop Data Layer (CDL). The black vertical line represents the approximate time the damaging thunderstorm occurred. Image b) is the same as a), but for soybeans. c) Time series of the co- and cross-polarizations for the damaged and background areas of the 2018 CDL identified corn damaged and background areas. d) Same as c) but for the 2018 CDL identified soybean areas.

In the U.S. every year, more than 10 billion dollars in insured losses is attributable to severe weather and 70% of that loss is due to hail. Hail damage and damage swaths are also not just confined to the United States. Argentina experiences intense thunderstorms and hail across their agricultural regions annually, as well. Remote sensing continues to be an technology increasingly used by all parts of the agricultural sector. During the growing season, it is not uncommon to go 5 to 7 days in between clear views of the surface, which limits the utility of optical instruments, a large percentage of the Earth Observation fleet. SAR provides another tool and can help fill in some of the information gaps caused by the limitations of optical instruments. Exploring methodologies and techniques to characterize these damage swaths and other intense thunderstorm related damage with data from NASA's upcoming L-band NISAR mission, will allow data to be easily inserted into existing workflows.

An article titled, "Complementing Optical Remote Sensing with Synthetic Aperture Radar Observations of Hail Damage Swaths to Agricultural Crops in the Central United States" has been accepted by the American Meteorological Society's Journal of Applied Meteorology and Climatology. The lead author of the article was Jordan Bell, Marshall Space Flight Center, with co-authors from the University of Alabama in Huntsville, Texas Christian University, and the University of Alaska Fairbanks.

The full article can be found at

SERVIR Science Coordination Office (SCO) Supports Synthetic Aperture Radar (SAR) Training around the World

Ms. Rebekke Muench of the SERVIR SCO supported a training on "Synthetic Aperture Radar for Forest Monitoring" led by Dr. Josef Kellndorfer (EarthBigData) at the Universidad Nacional Agraria la Molina (UNLM) in Lima, Peru, November 18th through 22nd, 2019. The training was provided as part of Dr. Kellndorfer's subject matter expert work to support the expansion of SAR to existing forest monitoring platforms in SERVIR-Amazonia partner organizations. This training also built upon the previous SERVIR Global capacity building effort, the SAR Handbook, to include the newest SERVIR hub. The 15 participants included individuals fro Asociacion para la Conservacion de la Cuenca Amonzonica (ACCA), UNLM, Ministerio del Ambiente, Servicio Nacional Forestal y de Fauna Silvestre, Comision Nacional de Investigacion y Desarrollo Aeroespacial, Ministerio de Cultura, Servicio Nacional de Areas Naturales Protegidas del Peru, and the Amazon Conservation Association. Technical skills developed in this workshop will feed directly into the forest monitoring efforts of ACCA, improving the timeliness of deforestation alerts across the region.

On January 13th through 15th, 2020, Ms. Helen Baldwin and Mr. Tim Mayer from the NASA SERVIR SCO led a workshop on using L-band SAR data to estimate forest stand height, hosted by SERVIR-Mekong in Bangkok, thailand. The 20+ participants came from six countries (Myanmar, Nepal, Thailand, Cambodia, Vietnam, and Laos), and represented eight organizations, including government departments such as the Cambodia Ministry of Environment, Universities such as the National University of Laos, and institutes involved in REDD+ (Reducing Emissons fro Deforestation and Forest Decredation) reporting such as Vietnam's Forest Inventory and Planning Institute (FIPI). This activity provided participants with an understanding of how to apply the method of estimating forest stand height that relies on temporal decorrelation as described in chapter four of the SERVIR's SAR Handbook, which has been downloaded over 300,000 times. Participants left the workshop with a concrete understanding of how to apply the techniques to their own countries for MRV (Monitoring, Reporting, and Verification) and land classifications. One participant stated that, "This training revolutionized my concept of boring and complex coding to a fun and powerful way of analyzing Earth bservation data in an understandable way."

You can find the link to the SAR Handbook at:

IMPACT Team Member Presents the Resilience community at 2019 GeoPlatform Community Meeting

Interagency Implementation and Advanced Concepts Team (IMPACT) member Ms. Jeanne Le Roux presented the Resilience Community to the attendees of the 2019 GeoPlatform Community Meeting. The Resilience Community is an interactive, topically-focused web portal to share web content, datasets, services, maps, applications, and tools relevant to environmental change and climate resilience.


The teams also collaborated together to incorporate key climate relevant datasets from the Climate Data Initiative (CDI) into, including relevant Earth observation data from NASA. The Resilience Community website was built over the past year by IMPACT in collaboration with GeoPlatform. The presentation discussed the process of building the community space as well as lessons learned from the experience.

The Resilience Community is available at

Marshall Lightning Mapping Array featured at International Workshop in Argentina

In November 2019, Dr. Timothy J. Lang traveled to Buenos Aires, Argentina to present at the Remote sensing of Electrification, Lightning, and Mesoscale/microscale Processes with Adaptive Ground Observations - Clouds, Aerosols, and Complex Terrain Interactions (RELAMPAGO-CACTI) Data Analysis Workshop. Marshall provided a Lightning Mapping Array (LMA), which measures the three-dimensional structure of lightning flashes, to RELAMPAGO-CACTI in support of validation of the GOES-16 (Geostationary Operational Environmental Satellite - 16) Geostationary Lightning Mapper (GLM). Dr. Lang spoke abut the success of the LMA deployment and showed initial results, which demonstrates how GLM detection efficiency varies as functions of thunderstorm life cycle and lightning flash type. During the trip, Dr. Lang attended a RELAMPAGO-CACTI reception at the U.S. Embassy, where he met with the U.S. Ambassador to Argentina, Edward C. Prado.

RELAMPAGO-CACTI occurred in the Cordoba province of Argentina during November 2018 to April 2019, and successfully observed some of the strongest thunderstorms on Earth. The multinational project is supported by the National Science Foundation and U.S. Department of Energy, with important contributions from NASA and the National Oceanic and Atmospheric Administration. Logistical support was provided by the U.S. Department of State. Key Argentinian partners included the Servicio Meteorologico Nacional (SMN), Universidad de Buenos Aires (IBA), and Universidad Nacional de Cordoba (UNC).

"Subtropical South American Hailstorm Characteristics and Environments" Article Published in Monthly Weather Review

An article titled "Subtropical South American Hailstorm Characteristics and Environments" has been published in Monthly Weather Review, with authors Zachary Bruick (Colorado State University), Kristen Rasmussen (Colorado State University), and Daniel Cecil (MSFC/ST11).  The article uses methods developed by Cecil to identify potential hailstorms based on NASA Satellite data and to assess the likelihood of large hail from each storm. Those methods were applied to parts of Argentina, Paraguay, Brazil, and Uruguay, where intense thunderstorms are common, but ground-based observational data are limited.  A broad diurnal cycle was found, with hailstorms extending later into the night than in other regions.  Environmental atmospheric conditions accompanying the hailstorms were documented, with anomalously warm and moist low levels contributing to thermodynamic instability, together with strong upper- and lower-level jets providing dynamic support for the storms.

hailstorms in South America

Anomaly fields for the atmospheric environments on days likely to have severe hailstorms in subtropical South America, compared to other days with significant convection, but not likely to include severe hailstorms. The image shows: (a) Surface pressure, with lower pressure in blue; 500-hPa geopotential height contoured. (b) Surface temperature, with warmer temperature in red; surface dew point contoured. (c) 850-hPa wind vectors and specific humidity, with moisture in green; meridional wind speed contoured (dashed contours are wind from the north). (d) 250-hPa wind speed, with orange for winds from the east and purple for winds from west; 250-hPa geopotential height contoured. All fields plotted are anomalies, for hail-days minus non-hail days.

These findings are significant because the Northern Argentina-Southern Brazil-Paraguay-Uruguay region has some of the most intense thunderstorms on Earth, as indicated by multiple NASA satellite datasets.  Ground-based observational networks are more limited there than in the United States, where intense thunderstorms also occur.  Understanding both similarities and differences between severe thunderstorm behavior and environments in different parts of the world can lead to a better understanding (and ultimately, better prediction) of severe thunderstorms and associated weather hazards (hail, damaging wind, tornadoes).

To read the article, go to:

Science Coordination Office (SCO) Team Members Meet with King of Bhutan

On November 1, 2018, Mr. Tony Kim (SERVIR Project Manager) and Dr. Ashutosh Limaye (SERVIR Chief Scientist) participated in a meeting with the King of Bhutan and Dr. Eugene Tu, NASA Ames Research Center (ARC) Director, when His Majesty, Jigme Khesar Namgyel Wangchuck visited ARC.

Mr. Kim and Dr. Limaye were given the opportunity to share an overview of NASA's SERVIR, Develop, and Global Learning and Observation to Benefit the Environment (GLOBE) projects, and how the funding from the new Inter-Agency Agreement with the DoS can help capacity building in the Kingdom of Bhutan.

Bhutan 1
Bhutan 2

Southern Africa Development Community's (SADC) Climate Service Center Using SERVIR Products for Improved Meteorological Analysis

The Southern Africa Development Community's (SADC's) Climate Service Center (CSC) is now using Early Warning Explorer (EWX) and Climate Hazard InfraRed Precipitation with Stations - Global Ensemble Forecast System (CHIRPS-GEFS) to help produce their meteorological, environmental, hydro-met, and historical analysis products for risk management and mitigating climate extremes for end users, including national meteorological services in the regions and for partners such as the United Nations World Food Programme (WFP), the Food and Agriculture Organization (FAO), and the International Red Cross.

A series of workshops held in 2018 and 2019 focused on training potential users to access, validate, and apply EWX data. After the September 2018 training in Lusaka, Zambia, the country's Meteorological Department now uses EWS to produce official weather bulletins for farmers and other regional users.

EWX Training in Lukasa, Zambia
EWX Training in Lukasa, Zambia

EWX, developed by Dr. Shraddhan Shukla's Applied Sciences Team Project, has been fully transitioned to SERVIR - Eastern and Southern Africa at the Regional Centre for Mapping of Resources for Development (RCMRD) and CHIRPS-GEFS forecasts were integrated into EWX. Through the EWX portal, the CSC is able to access the CHIRPS data to generate monthly and seasonal rainfall assessments and historical climate extreme indices.

SERVIR-West Africa Hosts Cloud Computing and Big Data Forum in Accra, Ghana


SERVIR-Hindu Kush Himalaya Hosts Regional Knowledge Forum on Early Warning Systems


On October 21-25, SERVIR-West Africa hosted a Cloud-Computing and Big Data Forum as a part of the Africa Geospatial Data and internet Conference in Accra, Ghana. The week was book-ended with opening remarks from Ambassador Stephanie Sullivan, the US Ambassador to Ghana, and a keynote address from Marshall Space Flight Center's Dan Irwin, Program Manager of SERVIR.  The event consisted of presentations from a wide range of representatives from academia to the private sector all focused around how cloud-computing can be beneficial to Africa, as well as the challenges and how to overcome them.  The event ended with clearly identified takeaways and next steps for cloud computing and big data in West Africa.

A press release on the event can be found at


SERVIR-Hindu Kush Himalaya (SERVIR-HKH) hosted the Regional knowledge Forum on Early Warning for Floods and High-Impact Weather Events on October 22 and 23, at the International Centre for Integrated Mountain Development (ICIMOD), the hub host consultation workshop, as well as a day of internal SERVIR conversations.  The Department of Hydrology and Meteorology (DHM, Nepal), the Flood Forecasting and Warning Center (FFWC, Bangladesh), North Eastern Space Applications Centre (NESAC, India), and the National Center for Hydrology and Meteorology (NHCM, Bhutan) discussed challenges to Early Warning Systems (EWS) in the region, including specific recommendations for improving the utility of SERVIR's High Impact Weather Assessment Toolkit (HIWAT) system, co-developed by SERVIR-HKH and an applied Sciences Team led by Dr. Patrick Gatlin (MSFC). In his presentation during the stakeholder workshop, the representative from NCHM (Bhutan) expressed high interest in being more involved with SERVIR, specifically extending HIWAT coverage and other existing services to include Bhutan.  During the forum, it was suggested that the way forward for implementing SERVIR EWS includes collaboration between all countries on validation of the products, and the use of the SERVIR-provided Hydrostat tool.


Global Hydrology Resource Center (GHRC) Successfully Migrates its Operations to the Commercial Cloud

To address the challenge of managing data from several upcoming large missions, NASA's Earth Observing System Data and Information System (EOSDIS) is seeking to utilize the commercial cloud.  Toward that end, GHRC was selected as the first Distributed Active Archive Center (DAAC) to fully migrate to the commercial cloud.  One of the reasons GHRC was selected as the first DAAC to fully migrate to the cloud was due to its history of managing data from a wide variety of sources including satellite missions, field campaigns, and science investigators.

In 2017, GHRC provided requirements and DAAC expertise for Cumulus, an EOSDIS developed re-usable, open source, cloud native framework for data ingest, archive, and processing.  After a yearlong successful prototype of Cumulus, GHRC was selected to fully migrate to cloud and serve as a pathfinder DAAC.  GHRC was involved in providing requirements, planning new features, deploying features, and collaboratively solving issues with the EOSDIS team using the scaled agile framework (SAF3) methodology.

A simplified GHRC data publication architecture in the NASA-compliant cloud platform.
A simplified GHRC data publication architecture in the NASA-compliant cloud platform.

At the end of 2019, GHRC successfully completed migration of its operations to the cloud on time and on budget without disrupting its existing day-to-day operations.  As a result, all of GHRC's data is now being distributed from cloud.  GHRC deployed its data publication workflows for backup restoration, on-going datasets, and new datasets to the NASA-compliant General Application Platform (NGAP), a NASA security compliant platform used for hosting EOSDIS applications. Several members of the GHRC team are now trained to operate Cumulus hosted in NGAP. Lessons learned, as part of the migration, have been documented for future migration of other DAACs to the commercial cloud.

disasters logo white

MSFC Disasters Team led response efforts to wet spring and summer across Midwest.

The MSFC Disasters Team led response efforts for the broader NASA Earth Science Disasters Program throughout the spring and summer as floodwaters impacted portions of the Midwest multiple times.

MSFC Disasters coordinators worked with multiple federal agencies such as the Federal Emergency Management Agency (FEMA), the National Guard Bureau (NGB), and the United States Department of Agriculture (USDA) to understand their needs as the floods evolved and worked with them to provide derived products that could help them with situational awareness. In addition to coordinating with these agencies and other NASA Disaster coordinators across various centers, the MSFC team was able to produce satellite derived water extents from multiple sensors.

MSFC derived water extents across Midwest from Sentinel Satellites
MSFC derived water extents across Midwest from Sentinel Satellites

The water extents are derived from satellite imagery and an ancillary dataset, such as land-classification.  MSFC's methodology calculates statistics on known water in the images and then uses those statistics to classify all water in the image.  The land classification is then used again to identify which areas should be water and which areas are now anomalous water (i.e. flooded).  The picture above shows an aggregate of multiple days of acquisitions of the MSFC derived water extents across parts of the Midwest from Sentinel-1 and Sentinel-2 satellites in late March 2019.  The USDA provided feedback to the program stating that the "water extents helped them with their creation of quantitative and qualitative products for a near real-time response at the request of the NASS Nebraska Regional Field Office as well as the NASS Agricultural Statistics Board in preparation for the March Prospective Plantings Report." These areas experienced several more flooding events and severely impacted the 2019 growing season.  The MSFC Disasters team continues to follow up with USDA to understand the limitations of the water extents to improve the product for the next time flooding occurs.

More about the NASA Disasters Program can be found here:

Artificial Intelligence developed to detect smoke plumes in geostationary satellite imagery.

The Interagency Implementation and Advanced Concepts Team (IMPACT) housed within Marshall's Earth Science Branch is applying artificial intelligence (AI) to devise new satellite data analysis techniques.  A recent example is the use of deep learning to enhance the detection of smoke plumes from geostationary satellite measurements.

Traditional satellite-based smoke detection relies upon multispectral techniques that can be clouded by features with similar spectral characteristics (e.g., in the visible range, clouds, dust, pollution and smoke). Additionally, this spectral analysis requires large data volumes and often subjective and time-consuming manual evaluation, which is not readily scalable. To address these weaknesses, an automated, deep learning based detection model capable of identifying smoke plumes using shortwave reflectance from the Geostationary Operational Environmental Satellite (GOES) Series R Advanced Baseline Imager (ABI).

In a recent study, IMPACT data scientists used a large number of satellite images with known smoke plumes obtained from a subset of GOES-16 observations between 2017-2019 to train a convolutional neural network (i.e., machine learning algorithm) to predict the probability that GOES-R ABI shortwave reflectance pixel contains smoke. The image on the right gives two examples of GOES-16 satellite pseudo true color images (top: 2345 UTC 19 May 2018: bottom: 2230 UTC 2 March 2018) and regions where the algorithm identified smoke plumes (yellow shaded regions). The model is able to detect smoke over open ocean and most land surfaces, including snow and urban areas, as well as distinguish the smoke plumes from most types of clouds. Compared with the non-training sample of GOES-16 images, accuracy of the smoke detection model is 92%. This GOES-based smoke detection is being integrated by IMPACT into an online web portal for operational detection and analysis of smoke, among other Earth Science phenomena.

Click on the link below to find out more about the IMPACT Project Office at Marshall Space Flight Center.


Impact Logo 2
GOES-R w Logo
Smoke Plumes

Marshall Earth Science Research & Analysis tapped for innovative extreme weather capacity building solutions.

A SERVIR Applied Science Team project led by Dr. Patrick Gatlin at Marshall Space Flight Center (MSFC) has brought together numerical weather prediction from the Short-Term Prediction Research and Transition Center (SPoRT), satellite-based storm assessment from the Global Precipitation Measurement (GPM) mission, and sattelite-based storm damage mapping from MSFC's Disasters Team into one toolkit being used in extreme weather services in Nepal and Bangladesh.

A demonstration of this service took place during the pre-monsoon and monsoon seasons of 2018 and 2019. A significantly damaging hailstorm that occurred on 30 March 2018 over Bangladesh was captured by the High Impact Weather Assessment Toolkit (HIWAT).

The top image on the right shows an example of the brightness temperature measured at 37 GHz with GPM's microwave imager (GMI) used by HIWAT to determine the probability of damaging hail. The two highlighted storms in the next picture have a >95% chance of damaging hail and the final picture shows the giant hailstones observed with these Bangladesh storms. This satellite-based hail probability is a result of recent Passive Microwave (PMW) research at MSFC by Dr. Daniel Cecil.

Find out more about SERVIR's Extreme Weather Service in the Hindu Kush-Himalyan (HKH) region at

Observing severe storms with NASA's GPM constellation of satellites for capacity building in South Asia.
GPM Hail Probability in Bangladesh Storms
Scroll to Top