Source: RUTGERS, THE STATE UNIVERSITY OF NEW JERSEY submitted to NRP
DIAGNOSIS AND PREDICTION OF STORMS THAT CAN AFFECT THE NORTHEAST
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1010951
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 24, 2016
Project End Date
Sep 30, 2021
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
RUTGERS, THE STATE UNIVERSITY OF NEW JERSEY
3 RUTGERS PLZA
NEW BRUNSWICK,NJ 08901-8559
Performing Department
Environmental Sciences
Non Technical Summary
Currently, the tools that weather forecters use to forecast thunderstorms do not account for the fact that the air outside a thunderstorm can mix with the air inside a thunderstorm as the storm develops and matures. In some cases, this mixing of air could prevent a storm from advancing beyond the stage of being a small cloud. As a result, thunderstorms are occasionally forecast that do not materialize. This can then lead to lost economic opportunities (e.g., a farmer decides not to plant because of the forecast for storms that don't actually happen). This research project is investigating the best way to incorporate the mixing process into thunderstorm forecasting techniques without it getting so complicated as to be intractable. Software that is commonly used by weather forecasters will be modified to incorporate the mixing process, and particular forecasts will be re-examined with the new tools. Ultimately, forecasts of thunderstorms should improve as a result.In a similar vein, new techniques are also being researched in the realm of forecasting nor'easters. Again, software in common use is being tweaked and validated, with the overall goal being an improvement in weather forecasting. For both weather phenomena (thunderstorms and nor'easters), results will be communicated at professional conferences of meteorologists as well as in the classroom to the next generation of meteorologists.
Animal Health Component
40%
Research Effort Categories
Basic
20%
Applied
40%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
13204202070100%
Knowledge Area
132 - Weather and Climate;

Subject Of Investigation
0420 - Weather;

Field Of Science
2070 - Meteorology and climatology;
Goals / Objectives
This work has two objectives, one involving the diagnosis of convective environments, and the other involving nor'easter studies. These are further discussed below.1) Diagnosis of convective environmentsThis work will extend parcel theory by including the effects of entrainment on the calculation of standard diagnostics such as CAPE, CIN, LFC, LNB, and others. Since there is no one way to include entrainment, a variety of approaches will be tested to determine which best reflects reality for the midlatitude, continental convection this work is concerned about. Past casestudies from the literature will be reevaluated in light of the new diagnostics. Finally, the diagnostics will be used to repeat previous climatological studies. To summarize, new diagnostics of convection will allow for improved analyses and predictions of convection, and may lead to statistical analyses of environments favorable for convection that are more powerful in discriminating between those environments conducive for convection, and those environments that are not, both for current and future climates.2) Nor'easter studiesThe objective of the nor'easter studies is to improve our understanding of the dynamics and thermodynamics of nor'easters. The objective can be divided into two parts. In one part of the study, a collection of nor'easters is studied so that conclusions may be drawn about nor'easter dynamics most generally. Questions related to this objective include the following: How sensitive is the modeled nor'easter development to the assimilation of COSMIC RO data? How does nor'easter development depend on the way the microphysics is parameterized in simulations? Additionally, particular focus is made on a high-impact March 2010 storm. In that case, it is equally important to understand the source of the model errors that led to a terrible wind forecast. Would additional observations (say, from COSMIC) improve the forecast? More generally, how does the model analysis need to be changed to improve the forecast?
Project Methods
This section begins with an outline of the methods to be used for the two project objectives, followed by a general description of the efforts that will be used to affect the target audiences, as well as how the overall project will be evaluated.I) Outline of methodsA) Diagnosis of convective environmentsA number of approaches to incorporating entrainment in parcel theory will be used to determine the optimal choice. The goal is to add as little to standard parcel theory as possible to allow the new approach to be tractable in an educational setting and also to prevent overfitting various parameters associated with entrainment. The first challenge is to implement basic parceltheory. Even here, there is a wide variation in standard practice. Although highly accurate ways of computing parcel ascent paths (with no entrainment) are known, many standard tools such as N-SHARP (the National Sounding/Hodograph Analysis and Research Program) and its Python-based replica SHARPPy (https://github.com/sharppy/SHARPpy) use a highlyinaccurate method. Thus, the first decision is whether to incorporate entrainment into an existing tool despite the issues over accuracy or "reinvent the wheel" to some degree for better accuracy. A related question is whether to conserve moist static energy, the approach many studies of tropical convection take, or to conserve equivalent potential temperature, as most studies of midlatitude convection do.Studies in the literature have outlined a number of detailed methods and parameterizations for entrainment, which will be investigated, but for simplicity a constant entrainment rate may be all that is necessary for the purposes of this project. It is important to point out that entrainment is highly variable, dependent on the size of the convection, the organization of the convection, and other factors. The intent here is not to diagnose the true CAPE experienced by each individual cloud, but rather to develop a technique that allows forecasters to identify situations where neglecting entrainment could lead their forecast to go astray. If the technique indicates that neglecting entrainment does not degrade from the forecast (perhaps because the lifting mechanism is expected to be so strong that parcels will reach their LFC anyway), then all of the standard thresholds and rules of thumb applied to traditional convective indices can still be applied.These various entrainment techniques will then be tested on a variety of cases previously examined in the literature, some of which were situations where convection failed to occur when forecast, and others for which the forecast was reasonable. Through these analyses, the way to incorporate entrainment that best adds value to diagnostics such as CAPE will be determined, and ways in which the new diagnostics can be incorporated into a forecast process will be able to be determined.Finally, past climatological studies will be repeated to determine the impact the new diagnostics have on the results.B) Nor'easter studiesIn the proposed study, the latest version of the WRF-ARW Model will be used for three forecast domains. These domains (at 45-, 15- and 5-km grid spacing, respectively) will have two-way interaction to allow the air-sea exchanges modeled on the inner domains to feed back to the larger scale circulations. Seven nor'easter cases have been chosen for further study based upon three constraints: All cases must have occurred after April 2006 (when COSMIC launched), must cover a wide array of months, and must have affected the Northeast. All cases remained offshore, except some storms clipped Cape Cod, and one hugged the Long Island coast. All events were close enough to land to produce significant impacts to society.For each case, simulations will be run for 180 hours, starting roughly 72 hours prior to the first precipitation impacts in the highly populated Mid-Atlantic US and associated cyclogenesis. This time frame serves to focus attention exclusively on cyclone initiation and its later impact on this region. A 72-hour lead time will allow simulations to spin up, establish baroclinicity between the cooler eastern US and warmer Gulf Stream, and simulate surface latent heat fluxes along theexpansive (>1000 km) northern edge of the Gulf Stream, which are all vital for nor'easter simulations. Specifically, model initialization will be 72 hours prior to the first nor'easter-related 0.5 mm (~0.02 inch) precipitation reading from the New Jersey Weather and Climate Network. A New Jersey-centric approach was chosen due to its high population density (461.6 km-2), significant contribution ($473 billion) to the US gross domestic product, and importance to the mission of the New Jersey Agricultural Experiment Station. Additionally the simulation period is long enough to cover the entire event duration. The drawback of such a long forecast length is the significant forecast error that is expected to develop over the 180-h period. However, the assimilation of COSMIC data should mitigate this risk.Before running any nor'easter cases, it is necessary to choose appropriate physics parameterizations. Preliminary work to determine an optimal configuration was completed using many tens of 12-hour WRF model runs. The optimal combination of WRF physics parameterizations were determined to be those that qualitatively produced the best results, yet allowed a 180-h WRF Model run to complete in no more than 18 hours on the computational resources available. Based on this analysis, the following WRF physics parameterizations have been selected:• Microphysics: Goddard Cumulus Ensemble Model scheme,• Longwave radiation: RRTM longwave radiation scheme,• Shortwave radiation: MM5 Dudhia shortwave,• Surface layer: MM5 similarity,• Land surface: NOAH LSM,• Planetary boundary layer: BouLac PBL, and• Cumulus parameterization: Grell-Devenyi ensemble scheme.All of these parameterizations are identical for each domain. The only exception is that the innermost domain does not use a cumulus parameterization.To determine the importance of COSMIC assimilation, two WRF configurations will be used for each nor'easter case, a control run with no COSMIC assimilation, and an experimental run with COSMIC assimilation (using WRFDA). To investigate BMPS influence upon nor'easter simulations, five BMPS will be used. In particular, the three, six-class, three-ice, single-momentschemes known as the Lin, Goddard Cumulus Ensemble (GCE6), and WRF single moment (WSM6) will be used, as well as a seven-class, four-ice, single-moment scheme (GCE7), and finally, a six-class, three-ice, double-moment scheme [WRF double-moment, six class (WDM6)]. The March 2010 nor'easter will be further analyzed by varying the other parameterizations used to determine the sensitivity of the forecast to these choices. Note that, although the initial conditions certainly already include COSMIC observations indirectly, these retrospective forecasts do not include further information from COSMIC (unless assimilated via WRFDA).II) Efforts and evaluationEfforts that will be used to change the target audience will include talks at conferences including annual meetings of the American Meteorological Society, as well as lectures to meteorology undergraduate students.This project will be evaluated by peer reviewers when pieces are submitted for publication as research articles, as well as by students through teaching evaluations.

Progress 10/24/16 to 09/30/21

Outputs
Target Audience:The target audiences during this project were practitioners and students in the fields of meteorology and atmospheric science, as well as colleagues in the fields of meteorology and atmospheric science who attend professional meetings and read peer-reviewed journals. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The community MetPy project continues to providing tool that are being increasingly used in meteorology classrooms nationwide. In fact, over the summer, I converted my lab assignments in the Synoptic Meteorology course I teach each fall from GEMPAK (an older analysis tool) to Python/MetPy. How have the results been disseminated to communities of interest?The results have been disseminated through a conference poster as well as classroom instruction. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? During this reporting period, the main work in service of goals (1) and (2) has been contributions to MetPy, a Python package that can be used to analyze all sorts of meteorological data. Improving this package will allow a new graduate student, who arrived in August 2021 and will begin work specifically on goal (1) after this reporting period, to pick right up where the previous graduate student left off and carry out analyses more efficiently. MetPy bug fixes have also made Python a more amenable tool to wrapping up work on goal (2), the main accomplishments for which were detailed in earlier reports. In addition, work was begun looking at not just precipitation in the Northeast, but also temperature, and in particular, Population-Weighted Degree Days, which are an important metric for anticipating energy demand for heating and cooling (and to some extent crop conditions, although without the population weighting in that case). By analyzing connections between the state of a key tropical oscilation (the Madden-Julian Oscillation) and subsequent weather patterns across North America, we have been able to make skillful predictions of population-weighted degree days, particularly of heating degree days in the cool season, at 16-20 day lead times. As the project has come to an end, here is a summary of accomplishments: We have learned that accounting for dry air aloft (particularly 2 to 4 kilometers above the ground) can be important in forecasting thunderstorm development. Accounting for this factor can reduce false alarms, meaning that there can be fewer instances where the forecast indicates a thunderstorm is possible but no thunderstorm actually occurs. We have also seen that, in a warming environment, extratropical cyclones along the East Coast (aka nor'easters) will have a tendency to track further offshore. However, their precipitation shields will expand, so that the precipitation along the East Coast associated with nor'easters may not change all that much due to these countervailing effects.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Decker, S. G., D. Margolin, and L. Trabachino, 2021: An MJO analog technique to forecast population-weighted degree days. Poster, 12th Conf. on Weather, Climate, and the New Energy Economy, 101st AMS Annual Meeting, Virtual, Amer. Meteor. Soc., 517.


Progress 10/01/19 to 09/30/20

Outputs
Target Audience:The target audiences during this reporting period were practitioners and students in the fields of meteorology and atmospheric science. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The community MetPy project is providing tools that are being increasingly used in meteorology classrooms nationwide. How have the results been disseminated to communities of interest?The results have been disseminated through publications (a master's thesis) and a conference talk. What do you plan to do during the next reporting period to accomplish the goals?COVID challenges should be overcome so that the work addressing goal (2) will be submitted to Monthly Weather Review. The remaining work addressing goal (1) is more unclear, but may be supported by a new graduate student late in 2021. In addition, contributions will continue to be made toward MetPy.

Impacts
What was accomplished under these goals? Due to a graduate student being on leave, not much has occurred regarding goal (1) during this reporting period. Regarding goal (2), a Master's thesis was formally published in October 2019 on that topic. The description of this work was provided in the previous progress report. Unfortunately, disruptions due to COVID have delayed the production of a related submission of this work to Monthly Weather Review. Issues and pull requests in support of the development of MetPy continue to occur, as was reported previously. In addition, this and related work is making its way into the classroom through forecast contests, one of which was reported on at the most recent Annual Meeting of the American Meteorological Society. These contests provide a context in which techniques used to analyze and forecast storms (and in this particular case their impact on energy markets) can be taught to and used by students.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Decker, S. G. A medium-range forecast contest to bridge the gap between academia and the private sector. 29th Conf. on Education, Boston, MA, Amer. Meteor. Soc., 5.4, https://ams.confex.com/ams/2020Annual/meetingapp.cgi/Paper/366011.
  • Type: Theses/Dissertations Status: Published Year Published: 2019 Citation: Yeh, P. The sensitivity of the structure and evolution of the 2010 Boxing Day blizzard in the U.S. to environmental temperature. M.S. Thesis, Rutgers University, 108pp.


Progress 10/01/18 to 09/30/19

Outputs
Target Audience:The target audiences during this reporting period were practicioners and students in the fields of meteorology and atmospheric science. Changes/Problems:The graduate student that had been working on goal (1) has taken a leave of absence from the university, so no progress has been made during this reporting period. The project director may end up finalizing this work for publication myself during the current or subsequent reporting periods. What opportunities for training and professional development has the project provided?One graduate student has been able to develop and apply his academic knowledge in support of this project, culminating in the production of a Master's Thesis and awarding of an M.S. degree. In addition, the community MetPy project is providing tools that are being increasing used in meteorology classrooms nationwide. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?The work addressing goal (2) will be submitted to Monthly Weather Review, and hopefully the work addressing goal (1) will be finalized as it heads toward submission. In addition, contributions will continue to be made toward MetPy.

Impacts
What was accomplished under these goals? Regarding goal (2), a graduate student completed a Master's thesis on that topic right at the end of this reporting period. In the thesis work, the student under my direction investigate a nor'easter from 2010 that had been an outlier in previous studies on the impact of climate change on nor'easters. Most storms in previous studies were shown to be more intense in a warmer climate, but this particular storm was less intense. The thesis work found that the nor'easter in 2010 likely developed through the generation of a mesoscale structure called a diabatic Rossby vortex, and that subsequent vortices that formed in the future climate simulations prevented a single circulation from being well organized. This result has some support from theoretical studies that have shown this tendency for smaller-scale (and less predictable) circulations to become more pronounced if the climate warms enough, rather than the large-scale circulations that typify nor'easters in the current climate. The 2010 storm appears to be closer to that tipping point than the others in previous studies, which explains why it was an outlier. The thesis is currently being adapted for publication in Monthly Weather Review, but this is happening after the reporting period this progress report is covering. We continue to rely more and more on the open-source meteorological package MetPy, and continue to contribute bug fixes and new features in support of the research goals of this project. These contributions can be found under the PI's GitHub handle (@sgdecker) at https://github.com/Unidata/MetPy/releases/tag/v0.11.0

Publications


    Progress 10/01/17 to 09/30/18

    Outputs
    Target Audience:The target audience during this reporter were practitioners in the fields of meteorology and atmospheric science. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students have been able to apply their academic knowledge in support of thisprogram. One graduate student has focused on the "Diagnosis of convective environments" goal as part of his Master's thesisresearch, and the other is focused on the "Nor'easter studies" goal as part of his Master's thesis research. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?Research on both goals should reach publication stage during the next reporting period, and we will continue to contribute to MetPy.

    Impacts
    What was accomplished under these goals? This was a year of transition. The research under goal 1 is currently on hiatus as the graduate student working on that project found full-time employment before completing his thesis and degree. He is still an enrolled student, so we have hope that we will get around to publishing this work, which was detailed in last year's progress report, albeit in a delayed fashion. The research under goal 2 has been reinvigorated by the arrival of a new graduate student, and it is currently in a data-collection phase. A number of experimental simulations are being carried out of a high-impact and poorly forecasted nor'easter, in an attempt to understand why it behaved differently than a typical nor'easter in some respects, and to understand how these systems will behave under a changing climate. This work had not yet reached publication stage during this reporting period. Carrying out this research has relied to an increasing degree on the open-source meteorological package MetPy. As a result, we have begun contributing bug fixes and new features to this software to facilitate the research goals of this project. These contributions can be found under the PI's GitHub handle (@sgdecker) at https://github.com/Unidata/MetPy/releases/tag/v0.9.0

    Publications


      Progress 10/24/16 to 09/30/17

      Outputs
      Target Audience:The target audience during this reporting period were colleagues in the field of atmospheric science who attend professional meetings and read peer-reviewed journals in the field. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One graduate student and one undergraduate student have been able to apply their academic knowledge in support of this program. The graduate student has focused on the "Diagnosis of convective environments" goal as part of his Master's thesis research. The undergraduate student focused on the "Nor'easter studies" goal as part of his senior honors thesis. How have the results been disseminated to communities of interest?Results have been disseminated through a talk at a professional conference and a journal publication. What do you plan to do during the next reporting period to accomplish the goals?The work that was presented at a professional conference will be written up and submitted to a journal publication, and a new graduate student will begin research to further our understanding of nor'easters.

      Impacts
      What was accomplished under these goals? Our research has developed new diagnostics for forecasting thunderstorms that take into account the cooling effect that evaporation of dry air into clouds can have on thunderstorm development. The findings show that, under certain conditions, the new diagnostics can clue forecasters into situations where thunderstorms will struggle to develop, and become less severe than standard diagnostics would suggest. Research into nor'easters, including one study that is now published, has shown that model simulations of these winter storms is highly sensitive to the approximations that are used to account for cloud and precipitation processes in numerical models. These processes must be approximated because they take place on scales much smaller than the scales that regional or global-scale models can directly simulate. This implies that using a variety of these approximations can be an effective way to generate probabalistic forecasts (e.g., there is a 60% chance of getting at least 3 inches of snow). Going into more detail, all severe weather events from 2015 and 2016 as reported to the Storm Prediction Center (SPC) were analyzed when the SPC had forecast at least a slight risk of severe weather. Reanalysis data was used to construct profiles of atmospheric fields like temperature and humidty in the vicinity of each severe weather event, so that a large statistical sample of events and corresponding atmospheric states could be constructed. From that sample, relationships between severe weather occurrence and various diagnostics of severe weather (including our new diagnostics incorporating dry air) have been investigated. Early results from this research were presented at the 2017 Annual Meeting of the American Meteorological Society in Seattle, WA. Regarding nor'easters, two studies have been performed examining the evolution of multiple nor'easters. One of these studies has been published, while the other served as a focal point for a student's senior honors thesis. These studies involve creating multiple simulations of these storms, each identical except for the approximations that are used to incorporate cloud and precipitation processes in the model, and then analyzing the output of these simulations both in a statistical sense as well as in ways that attempt to shed light on the ways these approximations are affecting the development of nor'easters in the model.

      Publications

      • Type: Journal Articles Status: Published Year Published: 2017 Citation: Nicholls, S. D., S. G. Decker, W.-K. Tao, S. E. Lang, J. J. Shi, and K. Mohr, 2017: Influence of bulk microphysics schemes upon Weather Research and Forecasting (WRF) version 3.6.1 nor'easter simulations. Geosci. Model Dev., 10, 1033-1049.
      • Type: Conference Papers and Presentations Status: Other Year Published: 2017 Citation: Slezak, J., and S. G. Decker, 2017: The impact of entrainment on the likelihood of convective initiation: Dilute CAPE as a forecasting tool. Preprints, 28th Conf. on Weather Analysis and Forecasting/24th Conf. on Numerical Weather Prediction, Seattle, WA, Amer. Meteor. Soc., 5A.4.