Advanced preparation for satellite data from the next-generation GOES-R advanced baseline imager (ABI) is supported by coupling high resolution mesoscale and radiative transfer numerical models. Calculated GOES-R ABI imagery is produced in a two-step process. First, a mesoscale model is used to simulate an event over a region with 400 m horizontal grid spacings; secondly, output from the mesoscale model is used as input to a second model that calculates top of the atmosphere radiances at selected GOES-R ABI wavelengths. Such radiances or brightness temperatures are referred to as synthetic imagery. In order for the synthetic imagery to contain realistic horizontal variability of values of surface reflectance at wavelengths from 0.44 to 2.25 μm, MODIS 16-day albedos are incorporated in the radiative transfer calculations. One application of synthetic GOES-R imagery is that of algorithm development and testing. Algorithms may focus on, but are not limited to, the detection and retrieval of smoke, volcanic ash, fires, blowing dust, and the state of surface physiography. Proper identification of such features is, at times, dependent on the horizontal variability of surface reflectance values. MODIS 16-day spectrally dependent albedos are a valuable dataset in aiding the generation of synthetic GOES-R imagery.
This article (CID 063598) was originally published in Vol. 6 of the Journal of Applied Remote Sensing on 30 October 2012 with incorrect captions for Figs. 4–12. The captions have been corrected, and the paper was republished on 15 January 2013.
With the launch of GOES-R expected in 2015, research is currently under way to fully understand the characteristics of every channel on its Advanced Baseline Imager (ABI). The ABI will have two infrared (IR) window bands centered near 10.35 and 11.2 μm. Since no broad-band space-borne sensor has a channel near 10.35 μm, radiative transfer model simulations are used to study the clear-sky gaseous absorption properties in this wavelength range. It is shown that water vapor preferentially absorbs radiation at 11.2 μm compared to 10.35 μm, making the 10.35 μm a "cleaner" window IR band.
True-color imagery, which is formed via a weighted combination of red, green, and blue (RGB) spectral information, has important operational applications for qualitative environmental characterization, including the detection of smoke plumes, volcanic ash, and other aerosols that are not as easily discerned in conventional visible or infrared imagery, but may be more readily characterized via color properties. Despite its universal popularity, true-color is currently unavailable from geostationary satellites, and the next-generation GOES-R advanced baseline imager (ABI) will fall one band (green; 0.55 μm) short of doing so. However, approximations exist, and a process for simulating true-color imagery representative of capabilities anticipated from the ABI is presented and assessed here. High-resolution atmospheric model simulations are used to produce the ABI reflective band imagery required for true-color imagery. Those simulations are then rendered at ABI spatial (0.5-km visible) and temporal (5 min) resolution, to provide realistic data, long before the anticipated 2015 launch of GOES-R. An additional analysis, a color-space transformation, is used to assess the true-color (RGB) ABI images. The resulting hue images verify the less-green bias in the synthetic-green band and synthetic-RGB images created on ABI simulated data. Assessing the deficiencies in the RGB process will hopefully lead to an improved and standard means for generating an RGB product from the ABI data stream. Finally, as one of the many product applications of true-color imagery, an example of synthetic true-color imagery with added smoke is presented. The incorporation of aerosol properties into simulated imagery may help reveal the limits of detectability for atmospheric aerosols with future ABI.
Realistic simulated satellite imagery for GOES-R ABI using state of the art mesoscale modeling and accurate radiative
transfer is being produced at the Cooperative Institute for Research in the Atmosphere (CIRA) and used in developing
and testing new products.
Products which have been produced in support of the GOES-R Algorithm Working Group (AWG) include 6-hour
imagery at 5 minute intervals for 4 GOES-R ABI bands (2.25 μm, 3.9 μm, 10.35 μm, and 11.2 μm) that include fire
hotspots. The imagery was initially produced at 400 m resolution and a point-spread function applied on the data to
create ABI resolution imagery. Also created was corresponding imagery for current GOES at 2 bands (3.9 μm and 10.7
μm). These fire hotspots were simulated for 4 different cases over Kansas, Central America, and California.
Additionally, high quality imagery for 10 GOES-R ABI bands (3.9 μm and higher) were produced for 4 extreme weather
events. These simulations include a lake effect snow case, a severe weather case, Hurricane Wilma, and Hurricane Lili.
All simulations for extreme weather events were also performed for current GOES and compared with available imagery
for quality control purposes.
Future work focuses on the creation of additional fire proxy datasets including true-color imagery for 3 ABI visible
bands. This project also supports the GOES-R AWG Aviation Team in their effort to test their convective initiation
algorithm by providing simulated ABI datasets for bands between 2.25 μm and 13.3 μm for a severe weather case. In
addition, simulated ABI was generated from MSG infrared (IR) window band imagery and corresponding simulated ABI
for the 7 tropical cyclones from 2006-2008 that became hurricanes in the east Atlantic for evaluation of the GOES-R
ADT algorithm conducted by the University of Wisconsin Cooperative Institute for Meteorological Satellite Studies
In preparation for future satellites, we have developed a method to simulate satellite observations using a cloud-scale numerical model and radiative transfer models. In short, a numerical cloud model was used to simulate mesoscale weather events, including severe storms and tropical cyclones. A second model was used to calculate the brightness temperatures of the clear and cloud sky scenes from the model simulations. This procedure allows for advanced product development for severe weather (precipitation estimation, updraft diagnosis products) and tropical cyclones (intensity estimation). The development of products in advance of the satellite launch extends the useful life of the satellite system. Examples of this method for
a severe storm case will be presented.