The lower this measure is the better, so being below the target is exceeded and above the target is not met. Visibility and cloud ceiling forecasts are critical for aircraft safety and efficient operations. When visibility or cloud ceilings are low, pilots rely on instruments to navigate instead of visual reconnaissance. The Federal Aviation Administration establishes Instrument Flight Rule (IFR) thresholds—visibility less than three statute miles and/or cloud ceilings at, or below, 1000 feet—for safety. NWS assesses the quality of IFR threshold forecasts in response to these requirements. Fundamental statistical metrics, specifically Probability of Detection (POD) and False Alarm Ratio (FAR), are used to track IFR forecast performance. Probability of Detection (POD), also known as Accuracy, is a ratio that describes the number of times IFR is correctly forecasted compared to the total number of IFR occurrences. FAR is a ratio that describes the number of IFR forecasts when IFR was not observed compared to the total number of forecast attempts. These two metrics must always be used in conjunction, as one can be improved at the expense of the other. Greater accuracy and a minimized FAR result in safer flights and fewer flight delays; conversely, poorer accuracy and an increased FAR result in a greater incidence of unnecessary flight delays.

*Please also note that a FAR is is the number of false alarms per the total number of warnings. So if a FAR actual is higher than the target, then the target has not been met. If the actual is the same number or less, then it has been met or exceeded.

Weather information users are surveyed continuously by means of a web-based, pop-up survey on NWS web pages throughout the Nation. A sample size of approximately 6,000 responses is collected quarterly for a maximum of 24,000 annual responses. The Customer Satisfaction Index (CSI) score is calculated as a weighted average of three survey questions that measure different facets of satisfaction with NWS services. American Customer Satisfaction Index (ACSI) researchers use proprietary software technology to estimate the weighting. The three questions include the overall satisfaction of NWS services, expectations of service, and a comparison to an ideal organization. Indexes are reported on a 0 to 100 scale. The CSI was started in the United States in 1994 by researchers at the University of Michigan, in conjunction with the American Society for Quality in Milwaukee, Wisconsin, and CFI Group in Ann Arbor, Michigan. The Index was developed to provide information on satisfaction with the quality of products and services available to consumers. The survey data serve as inputs to an econometric model that benchmarks customer satisfaction with more than 300 companies in 43 industries and 10 economic sectors, as well as various services of federal and local government agencies.

The measure is a proxy for the useful forecast lead time of the Global Ensemble Forecast System (GEFS) beyond the current forecasts skill (9.5 days) and is computed over the range of forecast days. Extending lead time in prediction of future weather events allows people to make informed choices and enables protection of life, property, and enhancement of the National economy. The 500 hPA Anomaly Correlation will be used.

This performance measure will assess how the UxS Research Transition Office (UxSRTO) investments are maturing the readiness levels of UxS observing system applications.

The annual number of peer reviewed publications is an indicator of productivity and relevance and is tracked using online resources. Peer review is one of the important procedures used to ensure that the quality of published information meets the standards of the scientific and technical community. The Science Council method to determine the outyear targets is based on an average of the previous 3 years. * FY 2019 is the first year that we are reporting on both NOAA-authored publications (previously the only type of publication measured) and NOAA-funded publications.

The Arctic is a vast, remote region that is critical to the United States economy, weather and sea ice forecasting, our understanding of ecosystem health (including marine mammals and fisheries), and our understanding of climate change. Due to the presence of low temperatures, harsh sea conditions, limited telecommunications infrastructure, and sea ice, maintaining a network of sea ice buoys, Air-Launched Autonomous Micro-Observers, and Bering & Chukchi moorings is difficult, necessitating constant deployment of a combination of expendable technologies and retrieval and redeployment of a large network of moorings in the Bering and Chukchi Seas.

This metric shows the current number of sites where GHG measurements are made with the goal of building the number of sites per year. The Paris Accord and other resolutions to mitigate climate change rely on accurately taking stock of both natural and anthropogenic emissions of greenhouse gases. These observing network sites enable OAR to accurately track the net global increase in GHGs and provide insights on where the bulk of the emissions are coming from. With an increased number of sites, OAR can provide more specificity and direct feedback to mitigation of emissions and the changing natural system that is so valuable to controlling the atmospheric GHG burden. (Quarterly number)

The measure is the cumulative number of instances of NOAA datasets made openly available via partners' cloud platforms through collaborations with selected industry partners.

Seasonal outlooks are used by sectors of the U.S. economy, such as energy, agriculture, transportation, etc. as one factor in resource decision making. Seasonal outlooks are reported as the probability of temperature being above normal, near normal, below normal or, where no definite seasonal guidance can be provided, equal chances. The seasonal outlooks are the cumulative skill calculated for regions where predictions are made. These forecasts are verified using a 48 month running mean of Heidke Skill score computed for seasonal outlooks for each 3-month seasonal mean (e.g., January-February-March mean; February-March-April mean; March-April-May mean; and so on).

New or improved PBL parameterizations will lead to improving the accuracy of atmospheric models and forecasts such as for precipitation lead time.