CIMMS research scientists from the MRMS QPE group shared their hard work and research findings with the weather community at the 2019 AMS Annual Meeting in Phoenix, AZ from 7-10 January 2019. For more information on the various topics presented, please visit our 2019 conference presentations page to see all of the MRMS QPE-related abstracts.
One of the bigger challenges with estimating precipitation is when hail is present. The presence of hail gives a different signal in many radar products, which impacts the ability to use radar to accurately provide a liquid accumulation value. In the operational version of MRMS, a hail cap is used in identified hail cores (i.e., where the Maximum Expected Size of Hail, or MESH, product is non-zero) by capping the precipitation rate a 2.12 inches per hour. While the hail cap was designed to prevent some overestimations of QPE, it has also created some significant underestimations as well.
In the upcoming build of MRMS, two changes will be made to mitigate the issues of hail contamination. The first is in the creation of the new MRMS dual-pol synthetic QPE. Instead of using reflectivity to estimate precipitation, hail cores will utilize specific differential phase (KDP) to better estimate rainfall in these convective storms. The image below shows how the hourly rainfall estimation is improved using this technique. The second is to reduce the impacts of the hail cap for MESH with the reflectivity-based approach by linearly increasing the influence of the hail cap from MESH = 0.01 inch per hour to MESH = 1.00 inch per hour. This would be applied in areas where radar coverage is degraded or if there are technical issues impacting the dual-polarization products from a radar.
There are two big rainfall impacts from tropical cyclones that will occur over the next few days in the continental United States. The first and greatest impact will be from Hurricane Florence with the forecast of 20+ inches of rain in the Carolinas. There is a also a tropical disturbance moving towards southern Texas with the forecast of 7+ inches of rain on top of already saturated soils.
The MRMS group at NSSL has developed a web page displaying the storm total accumulations for select experimental MRMS v12.0 precipitation products. The rainfall accumulations provided on the web site starts at 0000 UTC 13 September 2018 (8:00PM EDT 12 September). These products will continue to accumulate on this page until further notice.
The week of June 18, 2018 had series of high-impact flash flood events across the United States. One of the more prominent events occurred in southern Texas as a result of a nearly stationary tropical wave that moved in from the western Gulf of Mexico.
Events like these will undergo detailed analysis by the NSSL and CIMMS researchers in the MRMS group to determine the performance of multiple QPE sources and how MRMS can improved upon existing platforms. The focus of the evaluation shown here is within a 230 km range of the Corpus Christi, TX NEXRAD radar (KCRP) ending 1200 UTC 20 June 2018. The bubble plot shown below compares the daily gauge observations versus 24-hour accumulations of the Z-only (PPS) and dual-pol (DPR) single radar accumulations along with the operational Z-only MRMS radar-only QPE (Q3RAD) and the experimental MRMS dual-pol radar QPE (Q3DP) with evaporation correction. Warmer colors of the gauge bubble plots indicate an underestimation of the QPE product at that point, cooler colors indicate an overestimation of the QPE, and white mean the gauge and QPE are well correlated.
A statistical analysis of the various QPE sources characterized how the operational MRMS Q3RAD product showed some improvement over the two single-radar QPE accumulations; however, it still produced an underestimation bias, especially with rainfall accumulations exceeding 2-3 inches. The experimental MRMS dual-pol QPE (Q3DP) and its incorporation of various dual-polarization fields provided a much better depiction of the QPE across all accumulation ranges, especially the higher-end totals, while the evaporation correction algorithm reduced some of the overestimation bias seen with the lighter totals on the outer edge of the precipitation area.
The table below shows the average mean bias ratio (QPE over gauge), mean absolute error (MAE; in inches), and the correlation coefficient (CC) for the associated scatter plots. It clearly shows that the experimental MRMS Q3DP QPE with evaporation correction clearly outperformed the rest of the operational single-radar and MRMS QPE products.
We will continue to evaluate cases and the overall performance of our latest QPE algorithms. The MRMS dual-pol QPE with evaporation correction will become operational in the MRMS v12.0 build, currently scheduled for January 2019.
Less than two years removed from the devastating flash floods that impacted Ellicott City, MD in 2016, the same area experienced another catastrophic flash flood on Sunday, May 27th. The work conducted by NSSL scientists with the recently updated Multi-Radar Multi-Sensor (MRMS) system along with the hydrologic modeling from the Flooded Locations and Simulated Hydrographs (FLASH) project was put to the test during this event.
One of the key benefits of quantitative precipitation estimations (QPEs) from the MRMS system is the high spatio-temporal resolution of the products. The comparison of the operational MRMS radar-only QPE product at the 1-km Cartesian grid space versus QPE generated at a coarser 4-km polar stereographic grid space displays the ability for MRMS to capture the greater precipitation totals because of its finer resolution.
Using the 1-km resolution MRMS QPE, the FLASH system performs QPE comparisons and runs hydrologic models to determine areas that could experience flash flooding and their potential severity. The rainfall shown exceeded the 200 year return period color scale and was likely a 1000 year return period. The unit streamflow map highlights the areas of the flash flood threat. The blue colors representing unit streamflow values exceeding 10 cms/square km are generally correlated with major flash flooding and “Flash Flood Emergency” events.
NSSL scientists will analyze this particular case and other flash flood events to continue improving upon the research and algorithms that drive the MRMS and FLASH systems. While events like these are tragic for those impacted by the power of flood water, the findings and continued research will allow for even better products to assist forecasters with flash flood prediction and their operational warning decisions.
Operational MRMS Product Viewer: https://mrms.nssl.noaa.gov/
FLASH Product Page: http://flash.ou.edu/
On 17 May 2018, MRMS Version 11.5 became operational in the National Weather Service (NWS). This build contains a number of updates that impact the generation and delivery of MRMS QPE products. These changes include the ability to handle new volume coverage patterns from the NWS fleet of WSR-88D radars, new radar quality index logic, more hourly gauges, and reduced product latency.
The next version of MRMS (v12.0) is tentatively scheduled to become operational in January 2019. More details will be provided soon on MRMS v12.0, which contains the biggest advancements in the generation of QPE in MRMS since the initial operational release back in 2014. Stay tuned…
As part of the efforts to improve MRMS precipitation estimations, NSSL scientists with the MRMS group visit National Weather Service (NWS) River Forecast Centers (RFCs) to discuss ongoing challenges with estimating precipitation. The month of April consisted of trips to the North Central RFC in Chanhassen, MN and the West Gulf RFC in Fort Worth, TX. Specific topics included the following:
- The current performance of the MRMS QPE product suite,
- Shadowing forecaster shifts to better understand the use of QPE products in developing the NWS Multisensor Precipitation Estimator (MPE) product,
- Upgrades to the MRMS system for both v11.5 and v12.0, and
- Future work regarding the next steps in improving precipitation estimation for the NWS.
The ongoing collaboration between NSSL and the NWS RFCs have yielded valuable insight into the challenges that operational forecasters have with precipitation estimation. This has allowed for MRMS scientists to develop new products and techniques, which in turn helps the forecasters create better products for their users. More site visits will follow over the next several months.
Members from Taiwan’s Central Weather Bureau and Soil and Water Conservation Bureau met with NSSL researchers in March and April to collaborate on precipitation estimation and hydrology. Collaboration between NSSL scientists and Taiwan meteorologists began over 16 years ago with the development of a MRMS domain over Taiwan.
More information about this collaboration can be found here:
While gauges provide a surface observation about how much precipitation has fallen, they are point observation that do not represent the spatial distribution of precipitation. This is why forecasters and scientists have relied on radar coverage to estimate precipitation across a large area; however, what is detected by radar at beam level does not equate to what is observed at the surface.
One of the primary contributors to changes in precipitation estimations between the radar beam level aloft and the surface is evaporation. Evaporation can greatly reduce or even completely remove light precipitation before it reaches the ground. Using radar-based observations without accounting for evaporation has shown a systematic overestimation bias of precipitation in semi-arid environments and the production of “false light precipitation” (areas where the radar detects light precipitation but gauges record nothing).
In the upcoming MRMS v12.0 update, evaporation will be accounted for in the generation of radar-based QPE. Using three-dimensional model atmospheric parameters, radar precipitation estimates are modified based on the model environment it falls through at each grid cell. This has shown to improve precipitation coverage and accuracy as well as improve gauge quality control. More information can be found in the recently published article by Martinaitis et al. (2018).
Every product that is operationally implemented within MRMS undergoes extensive testing and evaluation. This testing is conducted through numerous case study analysis and real-time analysis on the MRMS QPE Verification System (QVS). This system allows MRMS researchers to compare multiple experimental and operational QPE datasets against each other and against various gauge observations. Case studies provides MRMS researches a way to go more in depth into different events or products to determine how they are performing and what could be done to further enhance QPE accuracy. Publications like the recent one by Cocks et al. (2017) show how these evaluations influence the design and evolution of MRMS QPE.
While MRMS v11.5 is going through its final tests at NCEP for operational use, the updates of MRMS v12.0 are being evaluated now for future release. The constant testing of our latest products allows us to continually refine our schemes and promote more accurate QPEs for the field. MRMS v12.0 is by far the biggest update to how we generate precipitation estimations since the initial operating capabilities four years ago. More on what is going into MRMS v12.0 will be discussed soon…