The Global Precipitation Measurement (GPM) Mission Applications Examples

Dalia Kirschbaum GPM Deputy Project Scientist for Applications Dalia.b.Kirschbaum@nasa.gov

www.nasa.gov/gpm Twitter: NASA_Rain Facebook: NASA.Rain

1 Applications Overview The new generation of GPM precipitation products advance the societal applications of the data to better address the needs of the end users and their applications areas. The demonstration of value of NASA Earth science data through applications activities has rapidly become an integral piece in translating satellite data into actionable information and knowledge used to inform policy and enhance decision-making at local to global scales. TRMM and GPM precipitation observations can be quickly and easily accessed via various data portals. This PowerPoint provides examples of how GPM is being applied routinely and operationally across a range of societal benefit areas.

2 Societal Benefit Areas

Extreme Events and Disasters • Landslides • Floods • Tropical cyclones • Re-insurance

Water Resources and Agriculture • Famine Early Warning System • Drought • Water Resource management • Agriculture

Weather, Climate & Land Surface Modeling • Numerical Weather Prediction • Land System Modeling • Global Climate Modeling

Public Health and Ecology • Disease tracking • Animal migration

• Food Security 3 Numerical Weather Prediction

• Air Force Weather Agency (AFWA) (557th Weather Wing) incorporates GMI data into their Weather Research and Forecasting (WRF) Model, delivering operational worldwide weather products to Air Force and Army units, unified commands, National Programs, and the National Command Authorities.

• Joint Center for Satellite Data Assimilation (JCSDA/NOAA): brings GMI data into their Global Data Assimilation System (GDAS), which is used by the Global Forecast System (GFS) model to initialize weather forecasts with observational data.

• Naval Research Lab (NRL): Uses GMI and other sensors in their Automated Tropical Cyclone Forecasting System (ATCF) for improved track prediction.

• European Centre for Medium-Range Weather Forecasts (ECMWF): uses GPM data operationally within their global numerical weather prediction model focusing on medium range (up to two weeks ahead) forecasts. 4 4 GPM Data used for Operational Tropical Cyclone Tracking The Naval Research Lab (NRL) routinely uses GPM Microwave Imager (GMI) data along with other sensors in their Automated Tropical Cyclone Forecasting System for improved storm track prediction. The NRL’s forecasts are used by weather prediction and disaster response organizations around the world.

Hurricane Matthew affecting Nassau in the Bahamas as a Category 4 storm on 10/6/2016

http://www.nrlmry.navy.mil 5 GPM for NHC Operational Decision Support

NASA MSFC Short-term Prediction Research and Transition (SPoRT) Center is working with the National Weather Service to transition GPM observations into their decision support systems. SPoRT provides GPM L1B single channel and multispectral (i.e., RGB) imagery and

L2 rain rate products from all of the 89 GHz RGB image of Hurricane satellites in the GPM Constellation to the Joaquin in N- AWIPS (1 Oct National Hurricane Center formatted for 2015 2101 UTC) their operational N-AWIPS decision support system. This allows the NHC to seamlessly obtain products and display in N-AWIPS alongside all of their other operational data sets.

“Integration of the suite of passive microwave L2 GPROF Rain imagery into N-AWIPS has been one of the most Rate image of significant dataset additions in recent years.” Hurricane Joaquin in Image and Caption Credit: Bradley N-AWIPS (4 Oct - Chris Landsea, National Hurricane Center Zavodsky/MSFC, Andrew 2015 2001 UTC) 6 Molthan/MSFC, Jordan Bell/UAH Assimilation of GPM data in NASA Unified WRF model forecasts GPM constellation data assimilation improves model simulation of West Africa Monsoon, an important role in Atlantic hurricane formulation and global climate Assimilate radiance correct microphysics improve monsoon rain band simulation

uOBS (mm h-1) uDA uNo-DA GMI data assimilation Improves rainfall forecasts for flood prediction in mountainous region: 35% error reduction in 24-hour rain accumulation verified by data from local ground observation network. (May 2014) 250 uOBS (mm) 200 uDA uNo-DA 150

100

50

0 GSMRGN (20) ECONet (1) HADS (6) Zheng/NASA SSAI 7 JCSDA Assimilation of GMI data for Julio, August 2014

Joint Center for Satellite Data Assimilation (JCSDA) is testing GMI data within their Numerical Weather Prediction models.

Julio track forecasts initialized at 00 UTC between August 4 – 12, 2014 for the CNTL (left) and GMI assimilated (right) experiments. The official “best” track is shown by the black line with hurricane symbols. Each point represents the forecast location at 24- hour intervals starting with the analysis time and ending with the forecast hour valid at the last best-track data point. Kevin Garrett/JCSDA 8 South Carolina Storms, October 2015

NASA’s Land Information System routinely assimilates GPM data as a forcing input for their regional and global instances. The example shows heavy rainfall from IMERG (left) and extreme “relative” soil moisture maximums in the areas of heavy rain for a major South Carolina rainfall event in October, 2015. Forecasters are provided this data in near real- time by SPoRT at MSFC and the data are actively being assessed with feedback provided by partners in Raleigh, among other Eastern Region NWS forecast offices. In the top right image, dark blues and purples suggest that these soils are holding 70-95+% of their water capacity, hence significant and immediate runoff that contributes to flash flooding. Both GPM and modeled products like soil moisture are also being used during disasters by

FEMA and other response agencies to improve situational awareness. 9 GPM used for Flood Estimation for Oroville Dam, February 2017

IMERG Rainfall (7-day accum.) 21 Feb 2017 GPM IMERG precipitation is used by the Global Flood Monitoring System (GFMS) to detect potential flooding conditions and estimate intensity. This system also uses GEOS- X Oroville 5 forecast to estimate streamflow within affected areas. Top left shows the 7-day IMERG rainfall totals over California ending on 21 Feb. 2017. Top middle plot shows forecasted 3- day rainfall from the GEOS-5 model near the Oroville Dam area. Bottom left plot shows the forecasted flood detection/intensity for 22 Feb. 2017, forecasts over northern GEOS-5 Rainfall Forecast (3-day accum.) 22 Feb 2017 California are estimated to be over 200 mm for the 22 Feb. 2017 (bottom). This information is valuable for improving situational awareness of floods. This capability can be applied anywhere globally, especially where conventional data and methods are not available. Forecast Event Estimated Water Volume into Oroville Dam

Flood Detection/Intensity (depth above threshold [mm]) Forecast for 22 Feb 2017

10 Adler/Wu U. of Maryland flood.umd.edu GPM Observes Pineapple Express rainfall, causing flooding in California January, 2017

Rainfall anomalies, Jan 10th, 2017 An atmospheric river (“Pineapple Express”) delivered over 5 inches of rainfall in parts of California in early January, 2017 (bottom) as viewed by GPM’s IMERG data. The 30-day rainfall anomalies ending Jan. 10th show TRMM Multi- satellite Precipitation Analysis from 2017 (top right) and 2016 (bottom, right).

Rainfall anomalies, Jan 10th, 2016

Image credit: Hal Pierce, SSAI/GSFC 11 Rainfall Analysis for Oso, Washington Landslide – March 2014 TRMM precipitation data was used to estimate monthly rainfall over Oso Washington, where a major landslide that killed 49 people occurred in March, 2014. Analysis of the TRMM seasonal rainfall compared to previous years on record suggested that the 45 days leading up to the event ranked 3rd highest on record since 2000. GPM data is vital for improving this long record to evaluate seasonal variability of landslide triggering. 12 Global Landslide Nowcasting for California, February 2017 A global landslide nowcast model provides situational awareness of landslide hazards for a wide range of users. The model uses IMERG near real-time data with a global susceptibility map to identify locations with landslide potential. 1-day IMERG rainfall accum Landslide Nowcast for Feb 21st, 2017 Feb 21st, 2017 Alameda County, CA Landslide 2/20/2017

Credit: Alameda County

1-day IMERG rainfall accumulation (left) for the U.S. West Coast and corresponding landslide nowcasts (right) are shown for Feb. 21st, 2017 results are updated every 30 minutes. NASA landslide susceptibility, hazard, and rainfall data are available globally in near real-time and have been used by many international and domestic organizations, such as the World Bank, World Food Programme, Pacific Disaster Center, FEMA, and the US Army Corps of Engineers. 13 Kirschbaum/Stanley https://pmm.nasa.gov/precip-apps Global Fire Weather Data and Forecasting

The Fire Weather Index System is the most widely used fire danger rating system in the world. The Global Fire WEather Database (GFWED) developed at NASA GISS integrates different weather factors influencing the likelihood of a vegetation fire starting and spreading. Calculations require measurements of temperature, relative humidity, wind speed, daily snow-depth, and precipitation totaled over the previous 24 hours. GPM IMERG data, as well as GPCP, TRMM, MERRA-2 and GEOS-5 are incorporated in different versions of the GFWED and are used by fire management agencies around the world.

Aqua/Terra MODIS Active Fires

The Fire Weather Index tracks the potential for The Fine Fuel Moisture Code tracks the potential extreme fire behavior, seen here with Aqua & for fire starts, such as these predominantly Terra MODIS active fires, using GPM IMERG for agricultural and forest plantation prescribed fires Aug 2015 record-breaking Pacific Northwest in the southeast US. wildfires (Field, Engel Marlier, Lettenmaier) R. Field/NASA GISS 14 NASA Precipitation Data used by the re/insurance industry to assess for Extreme Rainfall impacts Reinsurance company Swiss Re is using NASA’s Tropical Rainfall Measuring Mission (TRMM) coupled to a portfolio of country-wide exposure to estimate potential loss from heavy rainfall events. This program supports the Caribbean Catastrophe Risk Insurance Facility. In 2014, the program paid out four times (twice to Anguilla, once to St. Kitts and Nevis and once to Barbados). Report from CCRIF showing rainfall accumulation for Tropical Cyclone Gonzalo, 2014 (right).

SwissRe’s newest microinsurance program, launched in Guatemala in “We are relying on the TRMM, any of its successors 2016, uses TRMM data coupled to a loan (GPM) and other satellite-based rainfall to make initial pitches and explain parametric excess rainfall products portfolio to determine micro-loan conceptually throughout the world… the availability of recipients affected by heavy rainfall. The satellite rainfall allows us to easily develop a conceptual overview for almost any client” hope is for this program to be replicated - Megan Linkin, Swiss Re in other parts of Central America. 15 GPM data used by World Food Programme for Flood Situational Awareness in Pakistan

GPM IMERG Data was used to raise awareness for Pakistan flooding in early April, 2016 (right) by the World Food Programme. GPM IMERG data is routinely ingested into their newly developed Spatial Risk Calendar (SPARC). Heavy rainfall caused flooding that killed more than 50 people in northwestern Pakistan. Heavy pre-monsoon rainfall moved through the area resulting in extensive damage in the Swat Valley northwest of Islamabad the capital of Pakistan. Up to 308 mm (12.1 inches) estimated by IMERG to have fallen during this period (below)

Source: Fabio Lana, World Food Programme Screen capture from WFP SPARC System 16 Agricultural Monitoring

IMERG Estimates are compared with routine rainfall sources for Remote Agricultural Drought Monitoring within the Famine Early Warning System Data Assimilation System

Start-of-season (SOS) for the 2015-16 Southern Africa growing season, computed with three different satellite derived rainfall estimates Contribution from McNally et al. Kirschbaum et al. 2016, BAMS “While the IMERG product’s spatial resolution, temporal latency and, for this example, agreement with the CHIRPS shows great promise, its utility for FEWS NET will be better assessed when a longer time series is available.” 17 Food-Water-Energy Nexus

Global 2030 projections estimate increased demands for food (50%), water (30%) and energy (40%). Global observations of precipitation from GPM, along with global population estimates can begin to provide metrics for assessing human impacts of this distribution, including precipitation per capita or individual. These estimates will help better quantify the vulnerability of urban areas to these systemic changes and better characterize hydroclimate variability of extreme events.

M. Shepherd/C. Liu 18 Tracking Plague Outbreaks using TRMM Precipitation Data

Monthly TRMM data was used to determine the trends of black plague outbreaks in Uganda, finding that more plague cases were associated with higher precipitation totals. Results were compared across other rainfall products.

adapted from MacMillan et al. 2012; Monaghan et al. 2012; Moore et al. 2012 19 GPM used to estimate Cholera Risk after Hurricane Matthew, October 2016

GPM IMERG data was used along with air temperature anomalies and population to compute maps of estimated cholera risk in Haiti following the passage of Hurricane Matthew 1-2 October, 2016. Plots show a) IMERG precipitation anomalies prior to and b) following Hurricane Matthew; c) track forecast for Matthew over Haiti, d) shows a Cholera risk map based on pre-hurricane hydroclimatic conditions, e) updated Cholera risk map 2 weeks after Hurricane Matthew, and f) reported cases of Cholera as of 10 Oct 2016. 20 GPM Data Publisher

A new publisher provides IMERG rainfall, Landslide, and Flood Nowcasts via Applications Programming Interface. Data are available in a variety of different formats and end users can access products directly from the website or via the API publisher to automate data access. • Export file formats include geoJSON, topoJSON, arcJSON, TIF, SHP

NOW AVAILABLE: https://pmm.nasa.gov/precip-apps 21 For more information on the TRMM and GPM Missions: http://gpm.nasa.gov www.nasa.gov/gpm Twitter: NASA_Rain Facebook: NASA.Rain

22