NSSL researchers lead project to evaluate experimental flash flood products

DSC_0052During July, NOAA National Weather Service forecasters from forecast offices and river forecast centers will assess emerging hydrometeorological concepts and products in the Multi-Radar / Multi-Sensor (MRMS) Hydro Experiment 2015. Their goal is to improve the accuracy, timing, and specificity of flash flood watches and warnings.

MRMS-Hydro is led by NSSL and is part of the 2015 United States Weather Research Program (USWRP) Hydrometeorological Testbed (HMT). Operational activities will take place Monday through Friday for three weeks (July 6 to 24).

During the experiment, participating forecasters will evaluate short-term predictive tools derived from MRMS quantitative precipitation estimates (QPE) and Flooded Locations and Simulated Hydrographs (FLASH) hydrologic modeling framework. Forecasters will also explore the utility of experimental watch and warning products conveying uncertainty and magnitude issued through the Hazard Services software from the Earth Systems Research Lab/Global Systems Division (GSD). Research scientists will investigate human factors to determine operationally relevant best practices for the warning decision making process and the system usability of the Hazard Services platform.

HMT-Hydro will coordinate operations with the third annual Flash Flood and Intense Rainfall experiment (FFaIR) at the NOAA/NWS Weather Prediction Center (WPC) to simulate the collaboration that occurs between the National Weather Service’s national centers, river forecast centers, and local forecast offices during flash flood events.

HMT-Hydro and FFaIR will simulate the real-time workflow from WPC 6-24 hour forecast and guidance products to experimental flash flood watches and warnings issued in the 0-6 hour period. The HMT-hydro team will shift its area of responsibility on a daily basis to where heavy precipitation events and associated flash flooding is anticipated.

Researchers will collect feedback from NWS operational forecasters through comments during their shifts, electronic surveys, de-briefings, and a webinar at the end of each week. NWS feedback is critical for future development and eventual implementation of new applications, displays, and product concepts into AWIPS2 and other operational systems.
HMT-Hydro 2015 provides a real-time environment to rapidly test the latest observational and modeling capabilities so they may be improved and optimized for transition to operational decision-making in the National Weather Service to support a Weather-Ready Nation.

Share this:

Experimental flash flood forecasting system tested by Texas rain event

Screen Shot 2015-06-04 at 10.13.41 AMRecent flooding in Texas and Oklahoma put NSSL’s experimental Multiple Radar Multiple Sensor (MRMS) Flooded Locations and Simulated Hydrograph (MRMS-FLASH) system to a rigorous test, and researchers were pleased.

The real-time MRMS-FLASH hydrologic modeling suite produces forecasts of flash flooding that are compared to historical simulations that use more than a decade of NEXRAD-based inputs at each 1km grid point. On May 25, the City of Houston, Texas, experienced deadly flash flooding. For this event, FLASH predicted extreme water flows out to six hours in advance.

The Coupled Routing and Excess Storage (CREST) distributed hydrologic model, also a part of the MRMS-FLASH modeling suite, generates maps of streamflow and unit streamflow (cubic meters per second per square kilometer) every 15 minutes. Comparisons between the observations of flash flooding in Houston and the maps of unit streamflow show a good correspondence between areas of high unit streamflows and flash flooding.

The FLASH model represents surfaces that do not absorb water, such as in urban zones, and is able to model dynamic soil moisture conditions, and how water will be routed downstream. MRMS-FLASH has run in real-time demonstration mode for several years.

NSSL’s MRMS-FLASH system provides information and services to make communities more resilient, focusing on enhanced water forecasting and delivery services, and also helps support the NWS to evolve its operations. https://inside.nssl.noaa.gov/flash/2015/05/flash-performance-with-recent-flood-events-in-oklahoma-texas/

Share this:

NOAA Scientists tackle mystery of nighttime thunderstorms

PECAN researchers will deploy an armada of instruments after dark, including weather balloons. (Credit: NOAA).
PECAN researchers will deploy an armada of instruments after dark, including weather balloons. (Credit: NOAA).

This summer, more than 20 NOAA scientists will stay up late to learn why some thunderstorms form and grow at night, without the energy from the sun’s heat. They will be participating in the Plains Elevated Convection At Night (PECAN), a large, intensive field campaign to collect data before and during nighttime thunderstorms in the western Great Plains from June 1 to July 15.

PECAN researchers will deploy instrumented aircraft, ground-based instruments, mobile radars, and weather balloons to learn what triggers these storms, how the atmosphere supports their lifecycle, and how they impact lives, property, agriculture and the water budget in the region. Meteorologists believe these targeted observations will build understanding and ultimately improve forecasts of these sometimes damaging storms.

A nighttime thunderstorm near Scottsbluff, NE. Photo credit: Chris Spannagle.
A nighttime thunderstorm near Scottsbluff, NE. Photo credit: Chris Spannagle.

“Large nighttime thunderstorms are an essential source of summer rain for crops, but also produce widespread and potentially hazardous severe weather, excessive rainfall, flash flooding, and unusually frequent cloud-to-ground lightning,” said Conrad Ziegler, a research meteorologist at the NOAA National Severe Storms Laboratory and principal scientist for PECAN.  “Weather forecast models often struggle to accurately account for these. The PECAN field campaign will provide us with valuable insights—and improve our ability to save lives and property through more accurate forecasts.”

The PECAN field campaign will involve scientists, students, and support staff from eight research laboratories and 14 universities. The $13.5 million project is largely funded by the National Science Foundation (NSF), which contributed $10.6 million. Additional support is provided by NOAA, NASA, and the U.S. Department of Energy.

Nighttime storm triggers
Once the sun goes down, the Earth and its lower atmosphere usually loses heat and becomes more stable, an environment not so favorable for supporting thunderstorms.  In the Great Plains, however, many summer storms form after sunset, and sometimes without an obvious trigger.

PECAN scientists are interested in large complexes of thunderstorms called Mesoscale Convective Systems that can grow overnight, last for hours and often produce severe and hazardous weather. They will investigate how a low-level river of air triggers thunderstorms and supports storm evolution, what causes storms to grow into MCSs, and how MCSs respond to the surrounding environment.

In addition, PECAN researchers will test their hypotheses about how deep waves in the atmosphere form and ripple across the plains, like what happens with water when a stone is thrown in a pond, causing new storms to form after sunset. One type of atmospheric ripple is called a “bore.” Thunderstorms can create bores, but bores can also cause a thunderstorm to suddenly intensify. PECAN is the first modern campaign to study the role of bores and how they trigger and support Mesoscale Convective Systems.

Armada of instruments
More than 20 NOAA researchers and students will be responsible for gathering data with multiple instruments including the NOAA-X-Pol, a dual-pol mobile radar, two mobile balloon launch vehicles, and two “mobile mesonet” vehicles equipped with weather instruments. New to the fleet is the Collaborative Lower Atmosphere Mobile Profiling System (CLAMPS) designed by NSSL researchers to meet many of NOAA’s and its National Weather Service’s needs for lower atmosphere temperature, humidity and wind profiles. Additionally, one of the three aircraft participating in PECAN will be a NOAA Lockheed WP-3D Orion aircraft, best known for its hurricane hunting missions.

Unique to the experiment is an observation strategy that uses PECAN Integrated Sounding Array (PISA) stations to provide temperature, humidity, and wind profiles about every five minutes. The Department of Energy will provide six out of the eight ground-based upward-looking infrared spectrometer instruments. Dave Turner, NSSL scientist and PECAN steering committee member, will coordinate their operation.

Deploying in the dark
The campaign is based in Hays, Kansas, and will begin each day at 8 a.m. CDT.  A team of meteorologists, including retired forecasters from NOAA’s Storm Prediction Center, will work on a forecast for the upcoming night. At 3 p.m., scientists will use the forecast to determine where across northern Oklahoma, central Kansas, or south-central Nebraska to deploy their mobile resources. Teams will then ferry the instruments to the target area, set up, and collect data from dusk until after midnight. When the observation period is complete, they will ferry the instruments back to the base in Hays. A better understanding of these storms will have relevance for areas beyond the Great Plains, because clustered nighttime thunderstorms are common in various regions scattered across the globe.

Share this:

Scientists gather to talk about using Unmanned Aerial Systems for weather research

UAS Scottsbluff, NE
University of Colorado’s Brian Argrow (with back to the camera) and graduate assistant Jason Roadman assemble the Tempest UAS prior to launch into a supercell near Scottsbluff, NE, June 2010.

Unmanned Aerial Systems (UASs) are becoming increasingly important as instrument platforms for remote and in-situ observations of the atmosphere just above the ground. Their adaptability, potential ease of deployment, and low cost make them an attractive research option. NSSL scientists will participate in the annual meeting of the International Society on Atmospheric Research using Remotely-piloted Aircraft (ISARRA) in Norman, Oklahoma, May 20 to 22 to share knowledge about using these aircraft systems to observe and monitor the atmosphere.

Topics presented by NSSL include using UASs as part of a composite observing system for predicting the formation and evolution of severe convective storms, roles for UAS in the 2016 VORTEX-Southeast project, and ground radar support of UAS operations with Multi-function Phased Array Radar (MPAR).

Using UASs for research is a developing endeavor. A University of Colorado (CU) UAS team successfully probed the rear-flank downdraft of a tornadic supercell in northeast Colorado during the second Verification of the Origins of Rotation in Tornadoes Experiment in 2009. With NSSL support, in June 2013, a CU, University of Nebraska-Lincoln, and NSSL team flew a UAS in coordination with an NSSL mobile mesonet (vehicle with atmospheric instruments) to sample outflows from several supercells in northeast Colorado.
These interactions support the NOAA goal of investing in observational infrastructure, and NOAA’s science mission to understand and predict changes in climate, weather, oceans and coasts.

Share this:

Forecasters to evaluate sets of advanced weather models that can depict thunderstorms

13985530779_b86a7f9932_kExperiments designed to improve National Weather Service severe weather forecasts will be conducted in the 2015 Spring Forecasting Experiment from May 4 through June 5, part of the NOAA Hazardous Weather Testbed (HWT) Experimental Forecast Program. The effort is a joint project between the Storm Prediction Center (SPC) and NSSL/CIMMS to support NOAA’s goal to evolve the National Weather Service and build a Weather-Ready Nation. The interactions between operational forecasters, model developers, and research scientists are critical for the effective transfer of operationally relevant guidance, tools, and techniques to operational forecasters.

Forecast teams will work with a number of sets of advanced weather computer models, called ensembles, that can depict thunderstorms (4km grid or less) to create experimental severe weather hazard outlooks valid over shorter periods (1-hr and 4-hr periods) than current SPC operational products. The outlooks will define the probability of a hazard occurring, the confidence in its path, and adjust to trends in the threat level based on new observations. These activities are foundational to the emerging FACETs vision and designed to link with initial Warn on Forecast activities conducted by the Experimental Warning Program. In addition, the predictability of severe weather hazards into Days 2 and 3 will be explored using these ultra-high-resolution forecast systems.

The HWT experiments support the NOAA mission of Science, Service, and Stewardship, in addition to providing information that will help communities be more resilient. HWT research will improve the nation’s forecasting and numerical weather prediction capabilities through collaborative efforts between the academic community and NSSL. The effort is highly relevant to NOAA’s goal to evolve the National Weather Service.

Share this:

2015 Spring Warning Project will look at new severe weather warning guidance

Researchers and forecasters work side by side in the Hazardous Weather Testbed.

Several experiments to improve National Weather Service severe weather warnings will be conducted this spring in the NOAA Hazardous Weather Testbed (HWT) as part of the annual Experimental Warning Program, a joint project of the National Weather Service and NSSL/CIMMS to support NOAA’s goal to evolve the National Weather Service and build a Weather-Ready Nation. The EWP’s Spring Warning Project will run from May 4 through June 12, and provides a conceptual framework and a physical space to foster collaboration between research and operations to test and evaluate emerging technologies and science.

Forecasters will evaluate an updated Lightning Jump Algorithm (LJA), based on the GOES-R Geostationary Lightning Mapper, that was enhanced based on feedback from forecasters participating in the 2014 program. In severe storms, rapid increases in lightning flash rate, or “lightning jumps,” typically precede severe weather such as tornadoes, hail, and straight line winds at the surface by tens of minutes.  These evaluations will help prepare for possible operational implementation in 2016 following the launch of GOES-R.

Earth Networks’ total lightning and total lightning derived products, including storm-based flash rates tracks, time-series, and three levels of thunderstorm alerts will be evaluated in real time, building upon the initial evaluation in 2014. The 2015 evaluation will test the feasibility of use and performance under the stress of real-time warning operations.

A new set of high-resolution Weather Research and Forecasting (WRF) models will serve as a prototype for developing the “Warn-on-Forecast” warning paradigm. Feedback from this project will go into developing new model tools capable of managing the large amounts of model information associated with future forecast systems.

During three weeks of the experiment, forecasters will assess a new tool using rapidly-updating high-resolution gridded Probabilistic Hazard Information (PHI) as the basis for next-generation severe weather warnings. This experiment is part of a broad effort to revitalize the NWS watch/warning paradigm known as Forecasting a Continuum of Environmental Threats (FACETs). The major emphasis of the HWT PHI experiment will be on initial testing of concepts related to human-computer interaction while generating short-fused high-impact Probabilistic Hazard Information for severe weather. The long-term goal of this effort is to migrate the refined concepts and methodologies that result from this experiment into Hazard Services, the next generation warning tool for the NWS, for further testing and evaluation in the HWT prior to operational deployment.

This year will mark the inaugural HWT Experiment with Emergency Managers. The EMs will be provide feedback on their interpretation of experimental probabilistic forecasts generated in the HWT from the PHI experiment and the Experimental Forecast Program (EFP). This feedback will be used in conjunction with feedback from forecasters to refine how the uncertainty information is generated and disseminated.

Share this:

Significant papers

Significant Papers reported to HQ for the week ending April 17. The links to each are in blue.

1. Carlin, J. 2015. Weather radar polarimetry:  Dual-pol radar promises to improve the modeling of convective storms. Physics Today. March, 2015.

Significance: This is a concise and very accurate description of the basics of weather radar polarimetry and its potential benefits for convective-storm modeling.

2. Johnson, A., Wang, X., Carley, J., Wicker, L., Karstens, C. (2015). A comparison of multi-scale GSI-based EnKF and 3DVar data assimilation using radar and conventional observations for mid-latitude convective-scale precipitation forecasts. Monthly Weather Review 2015.

Significance: Implication is that the current state of 3DVar within the GSI is not well suited for the assimilation of high-resolution meoscale and storm-scale data. The results supports NCEP developing ensemble-based data assimilation methods (potentially including hybrid methods) for meso- and storm-scale prediction.

3. Potvin, C., Flora, M. (2015). Sensitivity of idealized supercell simulations to horizontal grid spacing: Implications for Warn-on-Forecast. Monthly Weather Review 2015.

Significance: Improved understanding of grid spacing dependence of simulated convection will be needed to properly interpret and calibrate ensemble output, and to optimize tradeoffs between model resolution and other computationally constrained parameters like ensemble size and forecast lead time.

4. Yussouf, N., Dowell, D., Wicker, L., Knopfmeier, K., Wheatley, D. (2015). Storm-scale Data Assimilation and Ensemble Forecasts for the 27 April 2011 Severe Weather Outbreak in Alabama. Monthly Weather Review 2015.

Significance: The short-range ensemble probabilistic forecasts obtained from this study demonstrate the potential of a frequently-updated, high-resolution NWP systems that could be used to extend severe weather warning lead times (WoF).

Share this:
Tags: None

CIMMS/NSSL researchers work to get West Texas lightning data in AWIPS in time for severe weather

For the past few weeks, Darrel Kingfield and Dr. Kristin Calhoun from CIMMS/NSSL and Dr. Eric Bruning from Texas Tech University have been working alongside forecasters from the National Weather Service (NWS) Forecast Office in Lubbock, Texas, to generate and integrate real-time 1km lightning products into the Advanced Weather Interactive Processing System-2 (AWIPS-2). AWIPS-2 is the weather forecasting, display and analysis package currently being used by the NWS. The team pushed hard to get these products functional for potential severe weather on April 16 and were successful.

Two products — Flash Extent Density and Flash Initiation Density — are gridded visualizations of total lightning data from Lightning Mapping Arrays (LMA). There are several arrays across the U.S. that are able to map the three-dimensional shape, extent and development of branched lightning channels. These data are an essential component of modern lightning detection and physics studies, because they reliably map the extent of the in-cloud charge reservoirs tapped by each lightning flash. Both of these products have been successfully evaluated in the Hazardous Weather Testbed.

A comparison of the 2000 UTC 1 min. flash extent density (left) and mean flash area (right) for a supercell over Kingfisher County, Oklahoma on 16 May 2010. This storm produced a wide swath of giant hail (>2" in diameter), causing severe damage to buildings and vehicles in it path.
An example of the lightning products installed at the NWSFO in Lubbock, TX. This is a comparison of the 2000 UTC 1 min. flash extent density (left) and mean flash area (right) for a supercell over Kingfisher County, Oklahoma on 16 May 2010. This storm produced a wide swath of giant hail (>2″ in diameter), causing severe damage to buildings and vehicles in its path.

A third product, the experimental Flash Area product from Texas Tech University, was also integrated into AWIPS-2. Forecasters have seen in the LMA data that small, compact flashes are mostly associated with robust or developing convection. As thunderstorms mature and reach the subsequent dissipation stage, the flash area starts to increase. The transfer of these products to operations will provide the developers, researchers and forecasters with the opportunity to learn more about how total lightning products can be utilized in the forecast and warning decision process.

This work benefits operational forecasters, highlights research to operations transitions, and supports NOAA’s work to evolve the National Weather Service. It also supports NSSL’s Grand Scientific Challenge to predict useful warnings of lightning activity one hour in advance.

Share this:

NSSL/CIMMS scientist to brief NWS forecasters

Screen Shot 2014-05-15 at 10.48.39 AMKristin Calhoun (NSSL/CIMMS) will give an invited webinar to National Weather Service meteorologists and hydrologists on April 1, 2015, about current lightning prediction products in research and development at NSSL. Calhoun will also discuss the NOAA Hazardous Weather Testbed (HWT) and how forecaster feedback is used to evaluate and refine the products.The webinar is part of a monthly series for NWS Science Operations Officers and Development and Operations Hydrologists that benefits operational forecasters, highlights research to operations transitions, and supports NOAA’s work to evolve the National Weather Service.

It has long been theorized, and in limited studies demonstrated, that the use of total lightning detections and associated derivative products could have positive impacts on the warning process for thunderstorm events. Two total lightning algorithms to potentially improve short-term prediction and warnings of severe storms have been evaluated in the Hazardous Weather Testbed (HWT) in Norman, Oklahoma.

In severe storms, rapid increases in lightning flash rate, or “lightning jumps,” typically precede severe weather, such as tornadoes, hail and straight line winds, at the surface by tens of minutes. The GOES-R Geostationary Lightning Mapper (GLM) will allow the use of continuous total lightning observations and the lightning jump concept operationally throughout the United States. A total lightning jump algorithm (LJA) that can be used by NWS forecasters to enhance situational awareness and diagnose convective trends was evaluated in the HWT as part of the experimental warning program in 2014 and will be evaluated again in 2015.

Earth Networks (ENI), a private company that provides lightning data and products, has indicated the potential for their total lightning data and “Dangerous Thunderstorm Alerts” to increase lead-time over current National Weather Service (NWS) severe weather and tornado warnings, while maintaining a similar probability of detection and false alarm ratio. This project integrates the ENI total lightning data and products into the NWS operational software and tests the feasibility of use and performance under the stress of real time warning operations.

In 2014, 18 NWS forecasters visited the HWT during a period of six weeks, 21 July-29 August, for a full product evaluation. The forecasters completed a series of six two hour weather-warning simulations of marginally severe storms to high-impact tornadic events throughout the United States. The 2015 HWT experiment will build upon the initial evaluation in 2014, including enhancements based on forecaster feedback.

Share this:

Impacts of Phased Array Radar Data on Forecaster Performance during Severe Hail and Wind Events

Early online release 1/13/15
Journal: Weather and Forecasting
Impacts of Phased Array Radar Data on Forecaster Performance during Severe Hail and Wind Events

Katie A. Bowden, Pamela L. Heinselman, Darrel M. Kingfield, and Rick P. Thomas

Summary:  Twelve National Weather Service (NWS) forecasters participated in the Phased Array Innovative Sensing Experiment (PARISE) 2013 and were assigned to either a control (5-min radar data updates) or experimental (1-min radar data updates) group. Each group worked a marginally severe hail event and a severe hail and wind event in simulated real time. While working each event, participants made warning decisions regarding the detection, identification, and re-indentification of severe weather, now known as “the compound warning decision process.”

Important conclusions:  The experimental group’s performance exceeded that of the control group’s, as demonstrated through their significantly longer median warning lead time, as well as superior probability of detection and false alarm ratio scores. The experimental group also had a larger proportion of mastery decisions (i.e., confident and correct) than the control group, possibly because of their enhanced ability to observe and track individual storm characteristics through the use of 1-min updates.

Significance:  This work furthers efforts that have already been made to understand the impact of higher-temporal resolution radar data, as provided by PAR, on the warning decision process of NWS forecasters. The research questions, methodology, and analysis presented in this paper build upon the findings presented from earlier PARISE work, while also sharing findings that are of a new nature.

Share this: