i
Global ESTOFS, STOFS-3D-Atlantic and OFS Water Level Skill Assessment Comparisons
-
2023
[PDF-1.34 MB]
Details:
-
Personal Author:
-
Corporate Authors:
-
NOAA Program & Office:
-
Description:Over the years, the NOAA National Ocean Service (NOS) has developed about 20 operational nowcast and forecast systems (OFS) for U.S. coastal waters as well as some deep ocean areas. The OFS support marine navigation, emergency response, search and rescue, offshore oil/gas operations, and the environmental management communities. The OFS perform nowcasts and short to long term (0 hr. - 180 hrs.) forecast guidance of pertinent parameters such as water levels, three-dimensional (3-D) water currents, salinity, water temperature. The OFS consist of the automated integration of observing system data streams, hydrodynamic model predictions, product dissemination and continuous quality-control monitoring. State-of-the-art numerical hydrodynamic models driven by real-time data and meteorological, oceanographic, and/or river flow rate forecasts form the core of these end-to-end systems.
The present study represents the first step toward gaining thorough and comprehensive insight into the relative performance between/among various NOS OFS. We focused on investigating five OFS among a total of about twenty operational OFS. The five OFS are, respectively, the San Francisco Bay OFS (SFBOFS), the Chesapeake Bay OFS (CBOFS), the Northern Gulf of Mexico OFS (NGOFS2), the Global Extratropical Surge & Tide OFS (hereafter referred to as ESTOFS), and the 3-D Surge and Tide OFS for the Atlantic Basin (STOFS-3D-Atlantic). These OFSs represent both the NOS port-based OFS (SFBOFS and CBOFS) and bay, region, or global based, larger domain OFS (NGOFS2, ESTOFS and STOFS-3D-Atlantic). Hopefully, the findings may provide technical guidance to various levels of OFS management and/or stakeholders in the planning for the development of the next generation, high-performance OFS.
We calculated the bias and RMSE of the nowcast water levels by comparing the model time series with the observed data at various NOSNational Water Level Observation Network ( NWLON) stations during a time span of one to three months periods. Some of the areas are covered by more than one OFS. The ESTOFS and STOFS-3D-Atlantic domains cover broader areas that overlapped both with each other and even encompass the domains of some other OFS. In cases of overlapped domains, the water level bias and RMSE between the concerned OFS are compared and contrasted so as to gain insight into the relative skills. The results may help identify the merits and disadvantages of each OFS.
The study concluded that these five OFS demonstrated similar levels of model skill in terms of bias and RMSE. The model skill represents an integrated balance of multiple factors, such as the system configuration, the model numerical schemes, model grid resolution, the accuracy of the forcing data, etc. The present study reveals that the five OFS demonstrated similar degrees of model performance in terms of the bias and RMSE of the nowcast water level. In certain areas, some OFS may exhibit slightly better skill, i.e., smaller bias or RMSE. However, none of the five OFS demonstrated statistically significantly better overall skill than the others.
This report is organized as follows. Following up to a brief introduction in Section 1, Section 2 describes the configurations of the concerned NOAA/NOS OFS with respect to the employed core hydrodynamic models, forcing data flow, etc. Section 3 is about the project design that details the methods for data processing and statistics analysis. Section 4 shows the model-data comparison results in terms of the model water level bias and RMSE. Section 5 summarizes the study and recommends future work.
-
Keywords:
-
Format:
-
Document Type:
-
Place as Subject:
-
Rights Information:CC0 Public Domain
-
Compliance:Submitted
-
Main Document Checksum:
-
File Type:
Supporting Files
-
No Additional Files
More +
