Atmospheric Sciences Project Update

Transforming Satellite Data into Weather Forecasts

A NASA project spans the gap between research and operations, introducing new composites of satellite imagery to weather forecasters to prepare for the next generation of satellites.

By Emily Berndt, Andrew Molthan, , and Kevin Fuell

Satellite imagery has been an immense benefit to weather forecasters. With it, they can assess aviation hazards such as low clouds, monitor thunderstorms, and track the evolution of dust plumes.

Satellite sensors are continually evolving to provide ever greater imaging capabilities, and researchers continue to develop advanced techniques to identify hazards in satellite imagery. However, barriers can prevent experimental products from reaching forecasters in the operational environment.

A NASA project has developed an interactive process whereby weather forecasters learn to interpret the latest satellite imagery and apply it to their operations. Forecasters then provide feedback to the researchers who are developing algorithms and products to further improve future products. This effort has taken on greater importance with the anticipated launch of a new series of satellites that will provide substantially greater amounts of data than are currently available.

Visualizing a Wealth of Data

Multispectral composites combine satellite data collected at different wavelengths of the electromagnetic spectrum and present them as red-green-blue (RGB) images, which serve as enhanced representations of specific phenomena such as low clouds and fog. With the launch of the Geostationary Operational Environmental Satellite–R Series (GOES-R, now called GOES-16) on 19 November 2016 and the first of the Joint Polar Satellite System (JPSS) series in 2017, U.S. National Weather Service (NWS) forecasters will have access to a wider array of imagery than is available with the current GOES satellites.

The current GOES satellites collect data from only five ranges of wavelengths in the electromagnetic spectrum, called bands. GOES-16 will have a total of 16 bands. The increased number of bands, and the resulting increase in the data richness of imagery, creates an even greater need for RGBs, which provide the ability to synthesize information from several bands into a single image for rapid interpretation.

The NASA Short-term Prediction Research and Transition Center (SPoRT) has used a research-to-operations/operations-to-research (R2O/O2R) paradigm to introduce RGB imagery to NWS forecasters to prepare them for new capabilities that will be available in the future.

Application and Feedback

This testbed environment paradigm relies on an interchange of information between research and operational organizations.
Fig. 1. The SPoRT R2O/O2R test bed environment paradigm relies on an interchange of information between research and operational organizations. Modified from Jedlovec [2013].
Keys to this paradigm (Figure 1) are active collaboration with end users (the weather forecasters), distribution of experimental products for use in end user display systems, creation of applications-based training materials, and evaluation of experimental products to gather end user feedback. SPoRT has applied its R2O/O2R paradigm to support the introduction of RGB imagery to NWS forecast offices since the early 2000s, when SPoRT worked with partner NWS forecast offices to introduce Moderate Resolution Imaging Spectroradiometer (MODIS) true-color and false-color RGB imagery to forecasters.

As a result of close collaboration with partner NWS forecast offices, weather forecasters now routinely use various types of RGB imagery in their operations. For example, the Nighttime Microphysics (NtMicro) RGB is widely used to differentiate between fog and low clouds, which is useful for assessing aviation hazards. The Dust RGB, which monitors the evolution of dust plumes, has been adopted into daily operations at the Albuquerque, N.M., NWS forecast office [Fuell et al., 2016]. Forecasters at the NWS Ocean Prediction Center routinely use the Air Mass RGB, which highlights differences between tropical and polar air masses, to assist with forecasting strong midlatitude cyclones [Zavodsky et al., 2013; Berndt et al., 2016].

Over the past 7 years, SPoRT has played an integral role in introducing RGB imagery and applications-based training to NWS forecasters. These efforts also serve to develop innovative solutions to display imagery in the operational environment to prepare forecasters for advanced imagery available with next-generation satellites.

History and Best Practices

In 2002, the European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) led an effort to develop RGB products from the data collected by newly launched satellites. The effort produced a set of best practices that included recipes for a suite of standard RGB products [Lensky and Rosenfeld, 2008; EUMETSAT, 2009]. The World Meteorological Organization later adopted these RGB best practices to develop a set of standard products among international partners.

Best practices ensure that RGB products are developed with a defined recipe and that each product is well adapted to its intended use so that forecasters know what they are looking at in a given RGB product. Because RGBs are designed to enhance a specific phenomenon, individual spectral bands or differences between pairs of bands are chosen for an RGB on the basis of the use of specific satellite bands to detect particular features.

For example, the NtMicro RGB was designed to enhance low-cloud and fog features. This RGB recipe (Figure 2) incorporates the well-known brightness temperature difference (ΔBT) between 11 micrometers (µm) and 3.9 µm. This quantity represents the difference in the intensity of the long- and short-wave infrared radiation at these two wavelengths that travels upward from the atmosphere to the satellite, and it is traditionally used to differentiate between water and ice clouds. The 12- and 11-µm ΔBT helps to distinguish between high, thick clouds and high, thin clouds by delineating cloud phase (ice or liquid water) and cloud particle size (small or large). The 11-µm brightness temperature is included to enhance low, warm clouds and distinguish them from colder, higher clouds.

Red, green, and blue components for the NtMicro RGB imagery and the resulting composite image.
Fig. 2. Red, green, and blue components for the Nighttime Microphysics (NtMicro) RGB imagery and the resulting composite image. Imagery was derived from the Himawari-8 satellite using channels similar to those that will be available on GOES-16.

Once the purpose for the RGB (e.g., monitoring low clouds and fog) and the bands and/or band differences are chosen, the images are preprocessed to improve contrast and sharpen features of interest. The images are then assigned to the red, green, and blue components to create a single RGB image, as depicted in Figure 2.

From 2009 to 2011, SPoRT focused its efforts on developing a technique to display EUMETSAT RGB products derived from polar-orbiting satellites in NWS display systems. They also developed training materials and introduced the RGBs to NWS forecasters.

Because the current generation of GOES does not have enough bands to create the EUMETSAT RGBs, imagery from research instruments was used to introduce NWS forecasters to the new imagery that would be routinely available in the future with GOES-16. For example, by 2012, SPoRT was using data from MODIS and the Visible Infrared Imaging Radiometer Suite (VIIRS) to mock up the RGBs that could be created with GOES-16 bands.

Interaction with the End Users

Today, partner NWS forecast offices across the United States use the NtMicro RGB most widely. The Alaska region especially benefits from this satellite imagery. In Alaska’s vast expanse of remote, complex terrain, low clouds and fog are a frequent hazard. The NtMicro RGB has become a part of daily operations in the Alaska NWS forecast offices as a result of an interactive partnership in which SPoRT conducted targeted product assessments of a variety of RGB products with forecasters.

Alaska NWS forecast offices participated in an assessment during the 2013–2014 winter, and SPoRT obtained feedback about the RGB from individual forecasters with a short online survey. The survey results indicated that the NtMicro RGB was valuable for distinguishing fog from low clouds beyond the traditional 11- and 3.9-µm ΔBT and that use of the RGB gave them more confidence when issuing aviation forecasts [SPoRT, 2014].

With the extreme range in hours of day and night unique to the high latitudes, the NtMicro RGB is most useful during periods of extended nighttime during the Alaskan winter. Therefore, forecasters inquired about a similar product for use during the summer. SPoRT recently developed the capability to derive the EUMETSAT daytime microphysics (DtMicro) RGB [Rosenfeld and Lensky, 1998; Lensky and Rosenfeld, 2008; EUMETSAT, 2009].

VIIRS data display for nighttime and daytime low clouds and fog over the Bering Sea
Fig. 3. VIIRS data display for (top) nighttime and (bottom) daytime low clouds and fog over the Bering Sea, using NtMicro and DtMicro RGB composite images, respectively. NASA SPoRT processed satellite imagery for 12 June 2016 at 12:59 Coordinated Universal Time. Low cloud and fog features are dull aqua and gray in the NtMicro and aqua in the DtMicro images.

The DtMicro RGB is specifically designed to complement the NtMicro RGB for tracking low clouds and fog during the day (Figure 3). Alaska region partners assessed the utility of the DtMicro RGB during the 2016 summer, and they provided feedback that the imagery had a positive impact on operations and influenced aviation forecasts similar to the NtMicro RGB. In addition, forecasters found the DtMicro RGB valuable for identifying strong thunderstorms and low-level rain clouds.

This example of the success of the SPoRT R2O/O2R paradigm involves the EUMETSAT microphysics RGB products and the Alaska region partners. Additional NWS partners across the United States are interacting with SPoRT to integrate RGB imagery into operations and prepare for the wealth of new imagery available with GOES-16 in the near future.

The Future of Multispectral Imagery

SPoRT continues to apply its successful research-to-operations paradigm to prepare forecasters for the capabilities that will be available from next-generation satellites. Forecasters soon will have access to all 16 spectral bands observed by the new GOES-16 satellite via their display system, known as the Advanced Weather Interactive Processing System (AWIPS). SPoRT worked with a team of AWIPS programmers to develop the capability to generate RGB products directly within the AWIPS workstation, which provides greater flexibility in product generation and improved color quality.

Forecasters can access training material while viewing and analyzing imagery during the demonstration.
Fig. 4. An NWS Operations Proving Ground exercise compared (top) the three water vapor bands from the Himawari-8 satellite that are similar to imagery that will be available on GOES-16 with (bottom left) the Air Mass RGB. (bottom right) Forecasters can access training material though the AWIPS quick-reference plug-in while viewing and analyzing imagery during the demonstration.

In addition, SPoRT developed a plug-in for AWIPS that allows forecasters to review quick-reference material (Figure 4). This is ideal for accessing short refresher training for moments when a forecaster needs to quickly look up information about a product. This new capability allows forecasters to view reference material and real-time imagery at the same time to reinforce concepts without interrupting workflow.

The NWS Operations Proving Ground evaluated these new innovative capabilities during March and April of 2016. In particular, forecasters were excited about the possibility of accessing training material directly in AWIPS.

SPoRT continues to lead the community in providing multispectral composites to NWS partners though interactive partnerships, innovative training approaches, and targeted assessments tailored to specific forecasting challenges. More information and updates on this project are available on the blog The Wide World of SPoRT, the NASA SPoRT Center Facebook page, and Twitter at @NASA_SPoRT.


SPoRT works in close collaboration with the GOES-R and JPSS Proving Grounds, National Oceanic and Atmospheric Administration (NOAA) NWS Operations Proving Ground, and NOAA Cooperative Institutes.


Berndt, E. B., B. T. Zavodsky, and M. J. Folmer (2016), Development and application of Atmospheric Infrared Sounder ozone retrieval products for operational meteorology, IEEE Trans. Geosci. Remote Sens., 54(2), 958–967, doi:10.1109/TGRS.2015.2471259.

European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) (2009), Best practices for RGB compositing of multi-spectral imagery, 8 pp., User Serv. Div., Darmstadt, Germany,

Fuell, K. K., B. J. Guyer, D. Kann, A. L. Molthan, and N. Elmer (2016), Next generation satellite RGB dust imagery leads to operational changes at NWS Albuquerque, J. Oper. Meteorol., 4(6), 75–91.

Jedlovec, G. (2013), Transitioning research satellite data to the operational weather community: The SPoRT paradigm, IEEE Geosci. Remote Sens. Mag., 1(1), 62–66, doi:10.1109/MGRS.2013.2244704.

Lensky, I. M., and D. Rosenfeld (2008), Clouds-Aerosols-Precipitation Satellite Analysis Tool (CAPSAT), Atmos. Chem. Phys., 8, 6739–6753, doi:10.5194/acp-8-6739-2008.

Rosenfeld, D., and I. M. Lensky (1998), Satellite-based insights into precipitation formation processes in continental and maritime convective clouds, Bull. Am. Meteorol. Soc., 79(11), 2457–2476, doi:10.1175/1520-0477(1998)079<2457:SBIIPF>2.0.CO;2.

Short-term Prediction Research and Transition Center (SPoRT) (2014), VIIRS and MODIS multi-spectral imagery assessment for aviation weather and cloud analysis at high latitude—Winter 2013–14, assessment report, NASA, Huntsville, Ala.,

Zavodsky, B. T., A. L. Molthan, and M. J. Folmer (2013), Multispectral imagery for detecting stratospheric air intrusions associated with mid-latitude cyclones, J. Oper. Meteorol., 1(7), 71–83.

Author Information

Emily Berndt, Andrew Molthan, and William W. Vaughan (email: [email protected]), NASA Marshall Space Flight Center, Huntsville, Ala.; and Kevin Fuell, Earth System Science Center, University of Alabama in Huntsville

Citation: Berndt, E., A. Molthan, W. W. Vaughan, and K. Fuell (2017), Transforming satellite data into weather forecasts, Eos, 98, doi:10.1029/2017EO064449. Published on 05 January 2017.
Not subject to copyright.