2021 ACTIVITIES - OSTI R2O

2021

December 9, 2021  The session 4 of Showcasing Leading Practices in Climate Adaptation hosted by NOAA, EPA, Water Utility Climate Alliance (WUCA), and the Water Research Foundation (WRF) was held with three speakers from US Water Alliance, US EPA, and Anthropocene Alliance; highlighting how equity can influence community goals and build partnerships with lessons learned and solutions to build equity while adapting to climate change.

December 8, 2021  Dr. Santos Da Silva gave a seminar in George Mason University on the contributions of Integrated Assessment Models (IAMs) to the climate science. Those models represent the most relevant interactions between human and Earth systems, allowing the examination of questions at the intersection of energy, water, land, socioeconomics, and climate. Examples were illustrated, helping to understand the future evolution of energy, water, land and climate systems under a wide range of conditions.

December 6, 2021  The WPO-OSTI webinar for Weeks 3-4 and S2S was held with two speakers. Dr. Dongmin Kim of NOAA/AOML presented his research on a seasonal probabilistic forecast system for U.S. regional precipitation based on the tropical Pacific and Atlantic SSTAs. Following his talk, Jon Gottschalck of NOAA/CPC gave an update on the development, evaluation and applications of CPC week 2-4 excessive heat forecast tools and services. Both presentations were recorded and are available at the address https://vlab.noaa.gov/web/weeks-3-4-s2s-webinar-series.

November 22, 2021  Dr. Yongzhen Fan of UMD gave a seminar in ESSIC on enhancement of NOAA snowfall product through machine learning, which extends snowfall detection to extremely cold conditions and improves its overall performance in global validation. Snowfall, an important indicator of climate change, accounts for a large fraction of winter precipitation in mid- and high- latitudes.

November 18, 2021  The session 2 of Showcasing Leading Practices in Climate Adaptation hosted by NOAA, EPA, Water Utility Climate Alliance (WUCA), and the Water Research Foundation (WRF) was held with three speakers. First, Kavita Heyn, WUCA and Heidi Roop, University of Minnesota gave a joint presentation on “Climate Adaptation Engineering Case Studies”, showing water sector climate adaptation in practice. They exhibited how engineers and water managers were applying climate change data to design infrastructure. Such practice enabled engineers in the water sector to learn from peers and inspired new approaches to engineering projects. Next, Jason Giovannettone of Sisters of Mercy of the Americas demonstrated how to prioritize adaptation actions based on sector impacts/vulnerability and interdependencies, and set forth his plan to foster creative thinking informed by a multi-disciplinary and socially diverse group, and bridge the gap between research and practical application.

November 1 & October 4, 2021  The OSTI-WPO Weeks 3-4/S2S Webinar Series continued.  The presentation topics were 1) using simple, explainable neural networks to predict MJO, 2) improving S2S precipitation forecasts in UFS through tropical nudging and explainable machine learning, 3) CPC Global Tropics Hazards Outlook: background, current operational products and work to transition to a probabilistic format targeting the week 2-3 period, and 4) predicting Indian monsoon onset in S2S scale: a machine learning framework.  The Q&A discussion forum document can be accessed at 
https://docs.google.com/document/d/1chJ4Dbb5x aotIEhMm31Fnz2d0nTwaG7D4JvDi4qtinQ/edit

October 25-28, 2021  In this year's NOAA Emerging Technologies Workshop, the emerging technologies of interest were focused on monitoring and observing of Climate, Extreme Weather, Fire Weather, Great Lakes and Oceans, aiming to improve the impact of emerging technologies through the end-to-end processes of Discovery, Development and Deployment/Transition. The post-workshop report will be available via the NOAA Observing System Council webpage at nosc.noaa.gov.

October 13, 2021  There were two local seminar on understanding of the regional impact of climate change. Dr. Malte Stuecker from University of Hawaii gave a talk at GMU to address the drivers of regional climate change. His innovative approach allow further exploring the potential applications to investigate climate change in different regions for a wide range of variables, and to quantify uncertainty range of local feedbacks and remote drivers. Another talk on the challenge of hydrological cycle response to climate change was given by Prof. Paul O’gorman of MIT for an ESSIC seminar on 10/12. He discussed the limitation imposed by inadequate representation of convection in global models and the potential benefit of using machine learning parameterization to improve climate change prediction of precipitation.

August 2, 2021  Two presentations were given in the OSTI-WPO monthly webinar for Weeks 3-4 and S2S.  First, Dr. Maria Gehne from NOAA/PSL and CIRES/CU Boulder gave a talk on Tropical Dynamics Diagnostics for NWP to identify forecast error sources in the tropics related to moisture-convection coupling .  A comparison of FV3GFS model v15 and v16 forecasts showed the performance improvement with respect to traditional metrics.  She further demonstrated that the process and phenomena focused diagnostics can help to better understand NWP model performance regarding coupling between moist convective processes and tropical synoptic to planetary scale circulations. Next, Dr. Tara Jensen of NCAR/Research Applications Lab focused on the model evaluation tools, providing an overview on the METplus system and showing how it can be used for evaluation and diagnosis of week 3-4 subseasonal prediction.  The METplus system has been adopted by several organizations within NOAA, DOD, and international partnerships for both operation and research purposes.  The video record of this webinar is available online at vlab.

July 30, 2021  The CPC project managers reported promising results in the third quarterly review of the FY21 CPC-STI projects.  The blocking predictor evaluation project showed week 3-4 temperature forecasts had potential improvement from a merged-Multiple Linear Regression (MLR) scheme.   The skill scores can be improved by ~44% over the entire CONUS/AK domain for the November-April period when the NAO and PNA are active.  The flash drought project reported the tool they developed successfully predicted the 2012 Central Great Plains flash drought and verified well with USDM.  The machine learning project showed encouraging improvement of GEFSv12 week 1-5 precipitation and 2m temperature by NN (using EMC code) dependent forecasts.  Applying Tensorflow deep learning NN showed further improvement.

July 28, 2021  Drs. Tomas Fredrikse and Tong Lee of NASA/JPL gave a talk on multi-model seasonal sea level prediction in the NMME teleconference.  Because of coastal flooding events occurring more frequently, the sea level prediction becomes increasingly more societal relevant.  Since local storm surges are primarily weather-related and thus difficult to predict far ahead of time.  However, larger-scale background sea level variations on seasonal time scales that can affect flood risks, may have some predictability.  They used a hybrid dynamical approach by combining pre-computed sea-level sensitivities to forcings with predicted forcing anomalies and showed encouraging skill for seasonal sea-level prediction in Charleston, SC.  The skill beat the mean seasonal cycle up to 6 months lead time.  Their analysis further identified the sources of predictability.

June 7, 2021  The OSTI-WPO monthly webinar for Weeks 3-4 and S2S was held with two speakers from NCEP/EMC.  Dr. Avichal Mehra’s presentation entitled Development of Fully Coupled UFS-based Seasonal-to-Subseasonal Prototypes discussed the scientific formulation of the S2S UFS model prototypes, which had been developed towards the operational implementation of GFS v17 and GEFS v13 in FY2024.  Dr. Arun Chawla’s talk focused on the open community development of using the UFS weather model.  He gave an overview on the GitHub repository, the code management and regression test systems that had been put in place for developers to use and contribute to the development of the UFS weather model.  The video record of this webinar is available online at VLab

https://vlab.noaa.gov/web/weeks-3-4-s2s-webinar-series

May 21, 2021  The second quarterly review of the FY21 CPC-STI projects was held virtually.  The CPC project managers reported what had been accomplished and the issues/challenges that need to be overcome.  Among the promising results,  the week 3-4 temperature forecasts showed potential improvement from the original Multiple Linear Regression model can be obtained by ~7% when adding Day +7 NAO values as a predictor (by ~24% when only considering forecast initializations when the Day +7 NAO is amplified). In real-time forecasts, the Day+7 NAO value needs to be derived from dynamical model forecasts, the predictability of which would affect the result.

May 3, 2021  The OSTI-WPO monthly webinar for weeks 3-4 and S2S was held with two speakers. The first topic is System design and basic skill of a subseasonal Earth system prediction framework with the Community Earth System Model, version 2 (CESM2), and its application to explore the contribution of the stratosphere to the surface impacts in the month following the January 2021 sudden stratospheric warming, by Dr. Jadwiga Richter from NCAR; and 2) Understanding coupled vs uncoupled GEFS subseasonal forecast skill differences by Prof. Xin-Zhong Liang of UMD, who presented the importance of coupling in air-sea interactions with cloud-radiation feedback that affects subseasonal forecast skills. The video record of the webinar is available online at VLab.

April 5, 2021  In the OSTI-WPO monthly webinar for weeks 3-4 and S2S, Dr. Benjamin Green, NOAA/OAR/GSL-CIRES/CU reported working progress of the project to test the impact of incorporating three new parameterizations of atmospheric subgrid-scale physical processes into the existing physics suite used for NOAA’s Unified Forecast System (UFS) in order to advance S2S prediction capabilities.  Next, Dr. Lucas Harris from GFDL talked on “Coupled and Convective-scale Prediction”, showing GFDL’s unified modeling suite, SPEAR and SHiELD, gives two routes to make S2S prediction. From the climate side, the SPEAR (a coupled climate system of strength on 2 months and beyond) shows great simulations of SSTs and coupled variability; and from the weather side, SHiELD (a nonhydrostatic weather prediction system of strength on timescales out to 10 days) demonstrates excellent diurnal cycle and weather-scale variability.  Experiments revealed promising approaches to fill the S2S “predictability” gap from both sides by leveraging the two model systems capabilities.  The video record of the webinar is available online at VLab.

March 1, 2021  This month's weeks 3-4 and S2S Webinar co-hosted by OSTI and WPO was held with two speakers.  First, Prof. Michael DeFlorio, University of California San Diego reported on his research progress in S2S prediction of atmospheric rivers, ridging events, and precipitation over the Western U.S. to benefit water management.  Next, Prof. Ben Kirtman, University of Miami – CIMAS presented a set of seasonal prediction experiments with ocean eddy resolving model, revealing how the ocean mesoscale features feedback onto representation of the large-scale mean state, the simulation of S2S variability, the local imprint of large scale-climatic features (associated extreme events) and retrospective forecast skill.

February 25, 2021  The 2nd NOAA Workshop on Leveraging AI in Environmental Sciences was successfully ended after seven months running in slow motion of 33 sessions and follow-up panel discussions with 93 oral and 24 poster presentations and 6 tutorials given by industrial leads and academic experts.  In view of the next step, the emphasis is put on the mutual benefits of the government-academia-private partnership.  The NOAA-Google collaborative project, Large-scale Enhancement through AI Processing (LEAP), and NOAA partnering in the multi-sector AI2ES (NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography) were demonstrated.

February 24, 2021  The 2021 DTC UFS Evaluation Metrics Workshop was held online.  NWS OSTI, SRH, NCEP/EMC and SPC are the co-organizers with NCAR, NOAA Labs and universities.  Approximately 300 people from federal and state governments, academia, and International organizations joined the workshop to discuss and prioritize key verification and validation metrics, which were identified in a series of pre-workshop surveys, for evaluating UFS research products to guide the transition from research to operation.

February 18, 2021  The Intelligent Data Summit, a showcase for multi-vendors, was held online.  The presentations were given by experts from leading companies of Red Hat, Outsystems, Kove, Explorium, influxdata, Alation, and ScienceLogic on a broad range of application topics, such as finding and using the data to fuel better decisions, time series machine learning, data trust and accessibility, and behavioral correlation etc.  Emergent technical issues of how to design and deliver a modern digital enterprise powered by analytics, data and AI/ML were also explored, e.g. accelerating AI/ML lifecycle in the cloud-native era, building apps faster and smarter, and defining memory by software, just to name a few.