GPM Applications: Weather

Using GPM Data for Weather, Climate, and Land Surface Modeling

Using GPM Data for Weather, Climate, and Land Surface Modeling

Variations in rain, snow, and other forms of precipitation are an integral part in everyday weather and long term climate trends. Initialization of short-term weather and long-term climate models with accurate precipitation information enhances their prediction skills and extends their skillful lead times. To get the resolution and temporal coverage to measure precipitation across the globe, we often rely on satellite information. Satellite data can play a fundamental role in our ability to monitor and predict weather systems as well as to forecast future changes to our climate and land surface. Satellite data from GPM’s suite of precipitation products are integrated into numerical weather prediction models that are operated by operational partners to provide and improve the observations from which the forecasts are then generated. Similarly, climate and land surface models use satellite precipitation observations from GPM to describe the conditions that exist today in order to project how conditions may change in the future. The Weather, Climate, and Land Surface Modeling applications area promotes the use of GPM data to help monitor existing weather activity and model future behavior of precipitation patterns and climate.

Overview

Variations in rain, snow, and other forms of precipitation are an integral part in everyday weather and long term climate trends. Initialization of short-term weather and long-term climate models with accurate precipitation information enhances their prediction skills and extends their skillful lead times. To get the resolution and temporal coverage to measure precipitation across the globe, we often rely on satellite information. Satellite data can play a fundamental role in our ability to monitor and predict weather systems as well as to forecast future changes to our climate and land surface. Satellite data from GPM’s suite of precipitation products are integrated into numerical weather prediction models that are operated by operational partners to provide and improve the observations from which the forecasts are then generated. Similarly, climate and land surface models use satellite precipitation observations from GPM to describe the conditions that exist today in order to project how conditions may change in the future. The Weather, Climate, and Land Surface Modeling applications area promotes the use of GPM data to help monitor existing weather activity and model future behavior of precipitation patterns and climate.

Sections

GPM Data for Decision Making

Image

NOAA’s Climate Prediction Center (CPC) issues extended range outlook maps for 6-10 days in the future. The above figure shows a 6-10 forecast of precipitation probability for the first week of October 2018. This product complements short-range weather forecasts issued by other components of the National Weather Service. Credit: NOAA/NCEP/CPC
 

Numerical weather prediction (NWP) is the use of computer models to predict upcoming weather. Specifically, NWP centers rely on microwave-based satellite rainfall information, such as data retrieved from GPM’s GMI, to improve short- to long-term weather forecasts and correct track forecasts for tropical cyclones. In addition, NWP centers create precipitation products for “nowcasting” weather in the immediate 1-5 hours (e.g. using near-real-time rainfall data from GPM) to meet the needs of a wider user community, including weather forecasters, hydrologists, farmers, numerical modelers, the military and the climate community. Methods for integrating rainfall data are constantly evolving and advancing, and with GPM’s advanced instruments, scientists can influence and enhance their scientific research and benefit socioeconomic activities.

Image

European Centre for Medium-Range Weather Forecast (ECMWF) Seasonal Forecast of precipitation probability. Percent probability is determined by using the predictive anomaly relative to 24 years of observed precipitation from 1993-2016. Credit: European Centre for Medium-Range Weather Forecast
 

To understand the changing climate and make future climate predictions, scientists need to use sophisticated computer models to recreate Earth’s climate conditions. Understanding current rainfall and snowfall variability, among other climate factors on regional and global scales, helps scientists model future behavior of precipitation patterns and climate. But for a system as complicated as the Earth, the models are only as good as the data provided. Satellite precipitation measurements from GPM and its predecessor TRMM provide global scale observational data sets that are comprehensive and consistent over long time periods, two characteristics scientists need to understand the relationships between different parts of the climate system. Specifically, organizations use GPM and TRMM data as input to verify and validate their seasonal and climate model simulations. The ultimate goal is to be able to predict changes in climate on time scales as short as the next hurricane season and as far into the future as changes that may occur in the coming decades or centuries. 

Image

Climate change may lead to an increase in temperatures and a decrease in snowpack within the Absaroka Range, found at the eastern edge of Yellowstone National Park. Credit: National Park Service/Neal Herbert
 

Precipitation is the fundamental driver of land surface hydrological processes and a key component of the terrestrial water cycle, which in turn affects the functioning of atmospheric and climate processes. High-resolution modeling of land surface hydrological processes requires detailed rainfall estimates as inputs to improve understanding of the state of the water cycle and impacts on land-surface processes during extreme events. Satellite precipitation data from GPM is integrated into land surface models to study surface features and how they change due to manmade and natural conditions such as urbanization and erosion. The use of GPM precipitation data together with other satellite data including soil moisture within land surface models will improve weather and hydrological prediction, which will help city planners and even decision makers save lives. 

Weather & Climate Featured Resources

Jump to a Year

2020 | 2019 | 2018 | 2017 | 2016

2015 | 2014 | 2013 | 2012 | 2011

2010 | 2009 | 2008 | 2007 | 2006

2005 | 2004 | 2003 | 2002

Hurricane Isaias Impacts the US East Coast
Hurricane Isaias Precipitation, Sea Surface Temperature, and Wind History, August 5, 2020 From July 29 to August 5, 2020, NASA’s IMERG algorithm observed tropical storm Isaias’ rainfall over the Caribbean and large parts of the Eastern US. This animation shows the IMERG rain rates in green shading as Isaias tracked from the tropical Atlantic into the Caribbean, then northward along the Atlantic coast and into New England. The yellow line shows the location of Isaias' low-pressure center, as tracked by the National Hurricane Center and smoothed in time here for the animation. Several factors
IMERG Hurricane Hanna 7-27-20 cropped
Hanna formed from a westward propagating tropical easterly wave that entered the southeast corner of the Gulf of Mexico on Tuesday July 21st. The wave provided a focus for shower and thunderstorm activity, which then led to the formation of an area of low pressure over the central Gulf of Mexico. The National Hurricane Center (NHC) found that this low had developed a closed circulation by the evening of Wednesday July 22nd, making it Tropical Depression #8. Over the next 24 hours, the depression slowly organized and intensified over the central Gulf before reaching tropical storm intensity on
GPM Hurricane Douglas 7-25-20
Hurricane Douglas continued to approach the Hawaiian islands during this GPM overpass early in the morning (02:11 UTC) of July 25, 2020. Douglas had previously strengthened to a Category 4 hurricane the day before, but had substantially weakened over cooler waters throughout the day. Regardless, the GMI and DPR instruments recorded rain rates near 50 millimeters/hour (~2 inches/hour) near Douglas` center. The Central Pacific Hurricane Center advised residents of Hawaii to expect hurricane-strength winds and rainfall starting Saturday evening and lasting through Monday. View fullscreen in STORM
IMERG rainfall totals from Japan, July 3 - 9 2020
From July 3-9, 2020, NASA’s IMERG algorithm continued to observe the heavy precipitation that fell as part of the seasonal Meiyu-Baiu rains (“plum rains”) in east Asia. Weekly totals reached their regional maxima over the island of Kyushu in southern Japan. About half of the island of Kyushu received over 45 cm (~18 inches) of rain. The majority of Honshu, Japan’s main island, as well as Shikoku to its south, were also impacted by the rains, receiving from 10-25 cm, depending on the location. Additionally, large areas of eastern China were also covered by the plum rains during this weekly
IMERG rainfall totals from Japan, June 29 - July 5, 2020
This animation shows NASA IMERG rain rates (blue shading) and accumulations (green shading) near Kyushu island, in the southwest of Japan from June 29 - July 5, 2020. Devastating floods and landslides swept through parts of Kyushu on July 4, 2020, resulting in over 40 deaths and orders for hundreds of thousands of people to evacuate their homes according to media reports. Download video (right-click -> Save As) The rains that triggered the flooding occurred in the context of the Meiyu-Baiu rainy season, which arrives in east Asia every year from June to mid-July. “Meiyu” and “Baiu” are the
Hurricane Isaias Impacts the US East Coast
Hurricane Isaias Precipitation, Sea Surface Temperature, and Wind History, August 5, 2020 From July 29 to August 5, 2020, NASA’s IMERG algorithm observed tropical storm Isaias’ rainfall over the Caribbean and large parts of the Eastern US. This animation shows the IMERG rain rates in green shading as Isaias tracked from the tropical Atlantic into the Caribbean, then northward along the Atlantic coast and into New England. The yellow line shows the location of Isaias' low-pressure center, as tracked by the National Hurricane Center and smoothed in time here for the animation. Several factors
IMERG Hurricane Hanna 7-27-20 cropped
Hanna formed from a westward propagating tropical easterly wave that entered the southeast corner of the Gulf of Mexico on Tuesday July 21st. The wave provided a focus for shower and thunderstorm activity, which then led to the formation of an area of low pressure over the central Gulf of Mexico. The National Hurricane Center (NHC) found that this low had developed a closed circulation by the evening of Wednesday July 22nd, making it Tropical Depression #8. Over the next 24 hours, the depression slowly organized and intensified over the central Gulf before reaching tropical storm intensity on
GPM Hurricane Douglas 7-25-20
Hurricane Douglas continued to approach the Hawaiian islands during this GPM overpass early in the morning (02:11 UTC) of July 25, 2020. Douglas had previously strengthened to a Category 4 hurricane the day before, but had substantially weakened over cooler waters throughout the day. Regardless, the GMI and DPR instruments recorded rain rates near 50 millimeters/hour (~2 inches/hour) near Douglas` center. The Central Pacific Hurricane Center advised residents of Hawaii to expect hurricane-strength winds and rainfall starting Saturday evening and lasting through Monday. View fullscreen in STORM
IMERG rainfall totals from Japan, July 3 - 9 2020
From July 3-9, 2020, NASA’s IMERG algorithm continued to observe the heavy precipitation that fell as part of the seasonal Meiyu-Baiu rains (“plum rains”) in east Asia. Weekly totals reached their regional maxima over the island of Kyushu in southern Japan. About half of the island of Kyushu received over 45 cm (~18 inches) of rain. The majority of Honshu, Japan’s main island, as well as Shikoku to its south, were also impacted by the rains, receiving from 10-25 cm, depending on the location. Additionally, large areas of eastern China were also covered by the plum rains during this weekly
IMERG rainfall totals from Japan, June 29 - July 5, 2020
This animation shows NASA IMERG rain rates (blue shading) and accumulations (green shading) near Kyushu island, in the southwest of Japan from June 29 - July 5, 2020. Devastating floods and landslides swept through parts of Kyushu on July 4, 2020, resulting in over 40 deaths and orders for hundreds of thousands of people to evacuate their homes according to media reports. Download video (right-click -> Save As) The rains that triggered the flooding occurred in the context of the Meiyu-Baiu rainy season, which arrives in east Asia every year from June to mid-July. “Meiyu” and “Baiu” are the

The most detailed view of our daily weather has been created using NASA's newest extended precipitation record known as the Integrated Multi-satellitE Retrievals for GPM, or IMERG analysis. The IMERG analysis combines almost 20 years of rain and snow data from the Tropical Rainfall Measuring Mission (TRMM) and the joint NASA-JAXA Global Precipitation Measurement mission (GPM). The daily cycle of weather, also known as the diurnal cycle, shapes how and when our weather develops and is fundamental to regulating our climate.

Music Credits: "Battle For Our Future" and "Wonderful Orbit" by Tom...

NASA engineer Manuel Vega can see one of the Olympic ski jump towers from the rooftop of the South Korean weather office where he is stationed. Vega is not watching skiers take flight, preparing for the 2018 PyeongChang Winter Olympics and Paralympic games. Instead, he’s inspecting the SUV-sized radar beside him. The instrument is one 11 NASA instruments specially transported to the Olympics to measure the quantity and type of snow falling on the slopes, tracks and halfpipes. NASA will make these observations as one of 20 agencies from eleven countries in the Republic of Korea as participants...

NASA researchers now can use a combination of satellite observations to re-create multi-dimensional pictures of hurricanes and other major storms in order to study complex atmospheric interactions. In this video, they applied those techniques to Hurricane Matthew. When it occurred in the fall of 2016, Matthew was the first Category 5 Atlantic hurricane in almost ten years. Its torrential rains and winds caused significant damage and loss of life as it coursed through the Caribbean and up along the southern U.S. coast. 

Music: "Buoys," Donn Wilkerson, Killer Tracks; "Late Night Drive," Donn...

NASA scientists can measure the size and shape distribution of snow particles, layer by layer, in a storm. The Global Precipitation Measurement mission is an international satellite project that provides next-generation observations of rain and snow worldwide every three hours.

The Global Precipitation Measurement (GPM) Core Satellite captured a 3-D image of a winter storm on February 17, 2015, that left six to 12 inches of snow over much of Kentucky, southwestern West Virginia, and northwestern North Carolina. The shades of blue in the 3-D image indicate rates of snowfall with more intense snowfall shown in darker blue. Underneath where it melts into rain, the most intense rainfall is shown in red. You can see a lot of variation in precipitation types over the southeastern portion of the United States.

The GPM Core Observatory carries two instruments that show the...

Hide Body

Hide Date