NCCOS_SOW_RISE_PAF.pdf

Statement of Work

NOAA - NASA RISE Project - Using Synoptic- based Models, satellite ocean observations and climatology to support sub-monthly to seasonal predictions of anomalous sea levels

Introduction/Background Chronic coastal inundation and high tide floods coupled with sea level rise pose a significant risk to property, infrastructure, and environmental health (e.g. waterborne diseases), and have accelerated the need for new models and forecast tools to support enhanced preparedness and resiliency in high priority areas. To address this critical need, senior scientists from the National Ocean Service (NOS) Center for Operational Oceanographic Products and Services (CO-OPS) with participation from NASA JPL, NOAA Climate Program Office (CPO), National Center of Coastal Ocean Science (NCCOS) and other collaborators formed a joint NOAA-NASA RISE project. The project leverages NOAA OAR CPO MAPP funded projects to produce predictions of anomalous sea levels and coastal flooding in priority areas on time scales ranging from week two to month six. The long-term goal for these predictions is to inform the NOS/CO-OPS Seasonal High Tide Bulletin, Annual High Tide Flood Outlook, and other decision support services.

To support improved facilitation and transfer of NCCOS products and information into mature coastal decision- support applications for end users, this project was awarded through the NCCOS Discretionary Programmatic Approach to Funding (PAF) process: To understand and predict coastal inundation and other stressors related to rising seas, heavier precipitation, and more extreme weather variations and events and incorporate this knowledge into next generation modeling systems and forecast tool development. As a pilot project for tool development, NCCOS scientists and partners (via Contractor) shall use data and information from NOAA’s ocean and weather services, NASA and NOAA observational satellites, and advanced synoptic-scale modeling techniques to develop a regional- to local-scale understanding of the most reliable predictors of coastal inundation and high tide flood-level response. Prototype flood forecast outputs including coastal sea level anomalies, flood -risk probabilities, and other metrics will be developed for NCCOS- clients to track and monitor the occurrence, frequency and duration of coastal inundation and high tide flood hazard events for individual tide stations, and at the regional scale. Full scale development of forecast tools will be contingent on future working collaboration with NCCOS through the PAF.

This project builds upon previous NCCOS collaborations on using satellite remote sensing technology for developing new tools related to coastal hazards, smart planning and improved ecological risk awareness in vulnerable areas. A key means of analyzing and modeling coastal inundation and flooding are through small- to regional-scale atmospheric circulation patterns.

Mid-latitude cyclones, as well as smaller-scale pressure fields and wind patterns can significantly affect tidal variability as short-term fluctuations in sea level (Sheridan et al. 2017, 2019; Pirhalla et

al. 2022). One of the more common atmospheric forcings is the inverse barometer (IB) effect (Piecuch and Ponte, 2015; Pirhalla et al., 2022), whereby low (high) atmospheric pressure leads to less (more) force pressing downward on the ocean surface, thereby allowing (forcing) the water level to rise (fall). Equally important is the wind, relative to both on-shore vs. off-shore flow direction (that would work similarly to that of a weak storm surge), and/or the long-shore winds, either forcing sea-surface water onshore (resulting in anomalously high tides in areas of downwelling) or offshore (resulting in anomalously low tides in areas of upwelling) along the coast via Ekman processes (Woodworth et al., 2019; Pirhalla et al., 2022). At longer time-scales, seasonal variability can also play a role, via thermal expansion and contraction of ocean water volume (Chen et al., 2000), and the associated seasonal frequency or dearth of synoptic-scale systems that bring about the aforementioned IB-effects, wind and storm forcings. Even further out in time, interannual variability in non-tidal residuals has been linked to multiple phenomena, most prominently the El Nino Southern Oscillation (ENSO; Merrifield et al., 2012; Hamlington et al., 2015; Long et al., 2020). Synoptic climatological analysis – which specifically analyzes and applies regional modes of atmospheric circulation and their accompanying surface weather conditions (Lee and Sheridan 2015) to a variety of outcomes – is thus very well suited for examining these multivariate and multi-scale factors influencing the coastal environment. As such, this synoptic perspective has been used extensively to identify a number of environmental response variables, including hydrologic and marine responses such as chlorophyll levels (Sheridan et al. 2013), water clarity (Lee et al. 2016, Pirhalla et al. 2016), short-term sea-level fluctuations (Sheridan et al. 2017), and marine ecosystem thermal stress tolerances (Pirhalla et

al. 2014).

NCCOS has the unique capability of using this portfolio and framework to develop portable solutions for multiple coastal regions and end users, and facilitate new solutions on emerging issues that impact coastal resilience. To that end, this project will expand the use of satellite-derived data, synoptic climatological approaches and machine learning to predict coastal inundation, tidal flood hazards and impacts along vulnerable U.S. coastlines. NCCOS will benefit from this project through strengthened collaborations between the NCCOS Cooperative Oxford Lab, CO-OPS, and RISE partners.

Objectives For the NOAA - NASA RISE project, we require contractual labor to produce sea level anomaly hindcasts and forecasts for NOS CO-OPS priority tidal stations. While the Contractor is open to suggestions, for the scope of the immediate project, the Contractor shall examine the southeastern Atlantic Coast subregion as defined by Sheridan et al., (2019), and the 9 tidal stations therein (i.e. region A4 in Figure 1 below). A complete sea level forecast provided by observation- based statistical models shall incorporate all (i.e. multivariate) factors influencing sea level variability into a prediction product. For this scope of work, in addition to synoptic-scale atmospheric patterns, global-scale products from satellites, including sea surface height (SSH), sea surface temperature (SST), and geostrophic currents (u,v) shall be examined (Sheridan et al.

2019), to better understand the synergistic atmospheric and ocean- related processes that contribute to high water events and flood impacts for two different coastal regions.

The Contractor shall develop predictions of anomalous sea levels and other flood hazard metrics at our test sites from one week to six months in advance. The sea level predictions shall include the difference in mean sea level relative to established linear sea level trends, seasonal cycles and tide predictions (i.e. the meteorologically-forced sea-level anomaly). The flood hazard metrics shall include sets of weather and ocean risk patterns, hindcast and forecast prototypes of coastal inundation and flood hazard probabilities, and other forecast visualization tools.

Contractor Requirements Contractor must have a minimum of 10 years of extensive experience for addressing health-related impacts using synoptic climatological methods. Experience shall include expertise in understanding regional weather factors and human- health and well-being impacts. Strong technical expertise and skills related to environmental response to atmospheric patterns are required, which are acquired from long term experience and familiarity with meteorological characteristics, trends, and statistics.

Scope NOAA-managed areas along US coasts are facing multiple climate-related sea-level rise threats and impacts: coastal floods of terrestrial and oceanic origin, storm surge, and heavy precipitation events are affecting critical infrastructure and facilities, water resources, ecosystems, and habitats. Understanding long-term regional weather patterns and forcing contributing to these threats is vital for tracking and preparing for future hazard risk, maintaining healthy, resilient coastal marine habitats and nature- based shorelines. In this 1-year pilot study, the Contractor shall assess the role of several physical atmospheric and oceanic drivers that influence regional anomalous sea levels that can be scaled to local coastal community flood hazard metrics, thus informing climate adaptation and resilience efforts at these scales. Modeling and forecasting at these scales shall facilitate the development of consistent and reliable coastal resilience metrics, portable to other locations and high priority areas in future work.

Figure 1. Tide regions, sub-regions, and stations (from Sheridan et al., 2019, Figure 2), with proposed sub-region of interest (region A4) circled.

Tasks The Contractor shall undertake this work through the following Tasks/Requirements:

Task 1 – UNIVARIATE PATTERN ANALYSIS Task 1.1 Oceanic & Atmospheric Pattern Identification & Impacts on Sea- Level variability (October 1, 2022 ̶ January 31, 2023): 49 hours

Task 1.2 Effects of Oceanic-Atmospheric Coupling (October 1, 2022 ̶ January 31, 2023): 49 hours

Task 2– MULTIVARIATE MODELING ANALYSIS Task 2.1 Model Synthesis (February 1, 2023 ̶ May 31, 2023): 49 hours

Task 2.2 Product Synthesis (June 1, 2023 ̶ September 30, 2023): 49 hours

Project Milestones/Completion Dates (Estimates and as applicable)

Task Number

Work Milestones

Projected Completion Date

1.1 Interim Milestone 1: Sets of atmospheric and oceanic circulation patterns (SOMs) and time-series for the domain, and an evaluation of the SOM pattern relationships to historical anomalous sea-levels, and coastal inundation events at each of the 9 tide gauges analyzed.

2023-Q2 (January 31 2023).

1.2 Interim Milestone 2: An empirical estimate based upon 1993-2020 data) of coastal risk weather patterns showing the strongest likelihood of a coastal inundation flood event for the domain (e.g.

Sheridan et al., Figure 9).

2023-Q2 (January 2023).

2.2 Historical Model Output: A hindcast from 1993-2020 and initial skill/validation assessment of modeled mean sea level predictions at each tide gauge analyzed.

2023-Q4 (September 2023).

2.2 Forecasts/Tools: Weekly to monthly predictions of mean sea levels at the regional and local tide gauge scale, from January 2020 to the present, for a lead time of at least 6 months, and updated at least once monthly.

2023-Q4 (September 2023).

2.2 Forecasts/Tools: Probability estimates of coastal flood hazard risk by site/region and forecast lead time, to estimate the frequency and duration of high tide flood events.

2023-Q4 (September 2023).

2.2 Visual Tools: Production of visual charts (e.g. Figure 2) for inclusion within the NOAA Seasonal High Tide Bulletin and Annual Outlook products on an experimental (non-public) platform (e.g., shared pilot network).

2023-Q4 (September 2023)

1 Task 1.1 - Oceanic & Atmospheric Pattern Identification & Impacts on Sea-Level variability

Satellite observations of SSH, SST and ocean currents will be used to classify oceanic circulation patterns for the spatial domains surrounding the southeastern Atlantic coast, using a neural network-based self-organizing map (SOM) technique.

Patterns will be analyzed for seasonal and interannual cycles and longer-term trends.

Each of the ocean patterns developed will be individually assessed for their utility in understanding short-term and seasonal-term changes in anomalous sea levels and flood events for each station. During this task, the Contractor shall also use SOMs to classify and validate atmospheric circulation patterns for the meteorological variables that have been found to impact marine environments in our previous research (e.g.

sea-level pressure, near-surface winds, and 850-mb temperatures). These atmospheric patterns will also be evaluated for seasonal, interannual, and secular trends, before quantifying their relationship with tide gauge anomalies and extreme events.

1 Task 1.2 - Effects of Oceanic-Atmospheric Coupling Task 1.2 will evaluate the effects of atmospheric-oceanic pattern coupling during historical flood events for our chosen tidal stations. First, the Contractor shall identify the effect of co-occurring atmospheric and oceanic patterns in terms of their anomalous sea-levels. The Contractor shall also assess the relative increased risk of inundation/flood events arising from the synergy of atmospheric an oceanic forcing.

The Contractor will investigate the time-lagged inter-relationships between ocean patterns and atmospheric patterns using cross-correlations. The Contractor will also evaluate these results against those obtained during Task 1.1.

2 Task 2.1 - Model Synthesis Based upon the univariate results from Task 1, the Contractor shall synthesize this information into multivariate models linking daily- to seasonal- scale modulations in the atmospheric and oceanic patterns to anomalous sea level changes for the set of tidal stations. Prior research has trained Non-linear Autoregressive Neural Network time-series models with exogenous input (NARX models) in order to conduct similar modeling, as such models are able to incorporate all of the categorical patterns (SOMs) developed in Task 1, along with temporal/lagged autocorrelation of tides, and any non-linearity in the relationship between sea-levels and atmospheric/oceanic patterns (Sheridan et al. 2019; Lee et al., 2016). This said, other modeling techniques will also be tested.

2 Task 2.2 – Product Synthesis Task 2.2 will be the formulation of coastal high tide flood risk products addressing individual site–based water level variability, and the enumeration of hazard events and flood probabilities into prototype products and tools. A suite of experimental hindcast and forecast products will be evaluated for inclusion in the flood hazard monitoring and tracking community, and for ecosystem and habitat vulnerability and resilience assessments. The Contractor shall work with NOAA RISE partners on fine tuning experimental products for dissemination in multiple decision support tools for application in future work. The Contractor shall summarize results of modeling, and make available the following deliverables (and their diagnostics) for delivery to

NOAA/NOS.

Figure 2. Example of visual chart of forecasted sea-level anomaly.

Delivery Deliverable Schedule

Deliver able

Description Quantity/Media Date Completed

1 Historical Model Output: A hindcast from 1993-2020 and initial skill/validation assessment of modeled mean sea level predictions at each tide gauge analyzed.

Digital Transfer to

NOAA/NOS

2023-Q4 (September 2023).

2 Forecasts/Tools: Weekly to monthly predictions of mean sea levels at the regional and local tide gauge scale, from January 2020 to the present, for a lead time of at least 6 months, and updated at least once monthly.

Digital Transfer to

NOAA/NOS

2023-Q4 (September 2023).

3 Forecasts/Tools: Probability estimates of coastal flood hazard risk by site/region and forecast lead time, to estimate the frequency and duration of high tide flood events.

Digital Transfer to

NOAA/NOS

2023-Q4 (September 2023).

4 Visual Tools: Production of visual charts (e.g. Figure 2) for inclusion within the NOAA Seasonal High Tide Bulletin and Annual Outlook products on an experimental (non-public) platform (e.g., shared pilot network).

Digital Transfer to

NOAA/NOS

2023-Q4 (September 2023).

Environmental Compliance Categorical Exclusion Status Applicable

Place of Performance Performance will be at the Contractor’s facility

Period of Performance October 1, 2022- September 30, 2023

Government POC:

Doug Pirhalla | doug.pirhalla@noaa.gov mailto:doug.pirhalla@noaa.gov