Fixed-Grid Format Data flowing in AWIPS

June 19th, 2018 |

AWIPS imagery of GOES-16 Low-Level Water Vapor (7.34 µm) at 1527 and 1532 UTC on 19 June (Click to enlarge)

Until today, GOES-16 Data that flowed into AWIPS was remapped twice: First, from the observational perspective (that is, how the satellite views it) to a spherical fixed-grid projection that approximates the Earth, and then to a Lambert Conformal projection with (for infrared data) 2-km resolution over the Globe. That Lambert Conformal data was then shipped to AWIPS, where the data were again re-projected into the observational perspective desired by the meteorologist.

The 2-km resolution of the data shipped to AWIPS before today is applicable only at the sub-satellite point (nadir) for GOES-16. Thus, the second remap was suggesting better resolution than was warranted by the data. Additionally, the number of data points needed to be sent was very big.

At 1532 UTC on 19 June, the first fixed-grid format data were directly shipped to AWIPS; remapping to a Lambert Conformal projection is no longer done upstream of AWIPS. The toggle above shows the difference in the 7.34 µm “Low-Level” Infrared Water Vapor imagery over the coast of Oregon, near 46º N, 124º W (very far from the GOES-16 sub-satellite point at 0º N, 75.2º W), in the AWIPS CONUS projection.  At 1532 UTC, after the double remap is removed, the pixels are more distinct, and as expected they splay away from the sub-satellite point.

Removing a remapping in the data processing means that pixel-sized extremes — such as overshooting tops, or fires — and gradients will be better represented in the data.  Consider the Clean Window (10.3 µm) Infrared imagery below of strong convection over the Gulf of Mexico east of Texas.  Overshooting tops Brightness Temperatures are colder and the tops themselves more distinct after 1532 UTC than at 1527 UTC.

AWIPS imagery of GOES-16 Clean Window Infrared Data (10.3 µm) from 1347 to 1612 UTC on 19 June. The animation pauses on the last double-remapped image at 1527 UTC, and the first fixed-grid format image at 1532 UTC (Click to enlarge)


See also this blog postThis training also discusses the remapping.  And here (or here) is the National Weather Service announcement on the change.

The 3.9 µm channel at night over very cold cloud tops

May 17th, 2018 |

GOES-16 ABI Infrared Imagery from 3.9 µm (Upper Left), 10.3 µm (Upper Right), 8.5 µm (Lower Left) and 12.3 µm (Lower Right), 0747 – 0832 UTC on 15 May 2018 (Click to enlarge)

When cloud top temperatures are very cold, the 3.9 µm imagery will have characteristics that suggest a noisy signal.  The 45-minute animation above shows a cold cloud top east of Florida in 4 different infrared channels:  3.9 µm (Upper Left), 10.3 µm (Upper Right), 8.5 µm (Lower Left) and 12.3 µm (Lower Right).  That the 3.9 µm image shows noise is not a new problem, as it was present in legacy GOES imagery as explained here.  At very cold temperatures the relationship between radiance (detected by the satellite) and temperature is highly non-linear, because of the character of the Planck function for that wavelength, meaning a very small change in radiance — within the noise — causes a large change in temperature (Compare the first two figures at this link for legacy GOES, for example).

Examine the two figures for GOES-16 below. They show the Planck curves for Band 14 (11.2 µm) and Band 7 (3.9 µm). Two things are apparent. Band 7 (3.9 µm), by design, covers a larger range of temperatures. In addition, very small changes in detected radiance (“counts”) at cold temperatures cause very big changes in the 3.9 µm brightness temperature. The relationship between detected radiance and very cold temperatures is much smoother at 11.2 µm.  The 3.9 µm band lacks precision compared to the other window channels, such as the 11.2 µm, for very cold temperatures. 

Plot of discrete values of Radiance vs. 11.2 µm brightness temperatures (190 K to 420 K) according to the Planck Relationship (Click to enlarge)

Plot of discrete values of Radiance vs. 3.9 µm brightness temperatures (190 K to 420 K) according to the Planck Relationship (Click to enlarge)

A zoomed-in view for cold brightness temperatures between 190 and 230 K (-83.15º C to -43.15º C) is shown below. If a true temperature of 208 K is being sensed by the satellite at the two wavelengths, it will be well-resolved at 11.2 µm, but the 3.9 µm detection will jump between 205 K and 210 K: the nature of the relationship between radiance and brightness temperature is such that there is less precision at the colder end at 3.9 µm. In the 30 K range from 197-227 K, just 12 possible bits are available in the 3.9 µm band (12 out of 2^14 — 16,384; recall that Band 7 on ABI has the highest bit depth of all the channels).  A change of just one count is a large difference in 3.9 µm brightness temperature.

Users need smarter ways to enhance the coldest 3.9 µm to prevent the flashing pixels evident in common traditional color and black-and-white enhancements.  Consider creating a color enhancement that shows only one color at temperatures colder than, say, -40º C, because the detector does not precisely distinguish between the coldest temperatures.  In other words, don’t highlight the noise!  Conversely, don’t use the 3.9 µm imagery at night to discern cloud-top features.   During the day, solar radiation at 3.9 µm reflected off cloud tops causes an increase in apparent brightness temperature so this quantization noise does not occur.

Plot of discrete values of Radiance vs. 11.2 µm brightness temperatures (190 K to 230 K) according to the Planck Relationship (Click to enlarge)

Plot of discrete values of Radiance vs. 3.9 µm brightness temperatures (190 K to 230 K) according to the Planck Relationship (Click to enlarge)

As noted above, this is not a new problem. An image (produced using McIDAS-X) of an Mesoscale Complex over the Great Plains of the United States from GOES-16 is here at 10.3 µm and here at 3.9 µm; the same image from GOES-15 is shown here at 10.7 µm and here at 3.9 µm. In both shortwave images, speckling at very cold cloud top temperatures is apparent.

(Thanks to Mat Gunshor, CIMSS, and Tim Schmit, NOAA, for figures and comments on this entry)

GOES-16 ABI RGB product artifacts related to Keep Out Zones

April 6th, 2018 |

GOES-16 ABI Full Disk Imagery at 0430 UTC on 6 April. Bands 8 (6.19 µm), 10 (7.34 µm), 12 (9.6 µm) and 13 (10.3 µm) are shown (Click to enlarge)

GOES-16 ABI Full Disk Imagery at 0530 UTC on 6 April. Bands 8 (6.19 µm), 10 (7.34 µm), 12 (9.6 µm) and 13 (10.3 µm) are shown (Click to enlarge)

Eclipse season for GOES-16 occurs around the Equinoxes when the satellite can enters the Earth’s shadow. Long ago (before GOES-12), Eclipse Season meant the satellite lost power because the solar array that powers the satellite was in a shadow. GOES-16 has batteries that allow it to function when solar power is missing. However, as the satellite emerges from the Earth’s shadow, the Advanced Baseline Imager can point too close to the Sun, so portions of the image are not scanned and sent to the receiving station. The two animations above step through four different channels on ABI (Band 8 (6.19 µm), Band 10 (7.34 µm), Band 12 (9.6 µm) and Band 13 (10.3 µm), at 0430 UTC, when the Keep-Out Zone (KOZ, sometimes called the Cookie Monster effect) is northwest of the subsatellite point, and at 0530 UTC, (also Band 8 (6.19 µm), Band 10 (7.34 µm), Band 12 (9.6 µm) and Band 13 (10.3 µm)) when the KOZ is northeast of the subsatellite point.  Note in particular that the size of the KOZ is different with different channels.

If Channel Differences are used in a product, this intra-band size difference of the KOZ has an affect.  The toggle below shows the Airmass RGB (Red Component:  Split Water Vapor (6.19 µm – 7.34 µm) ; Green Component: Split Ozone, (9.6 µm – 10.3 µm); Blue Component: 6.19 µm) at 0430 UTC and 0530 UTC on 6 April 2018.  At 0430 UTC, there is a region where only the blue part of the Airmass RGB (from the 6.19 µm band) is present;  at 0530 UTC, the Keep-Out Zone border shows only magenta (i.e., a lack of Green) because the Split Ozone Channel (9.6 µm – 10.3 µm) is missing there.

AIrMass RGB (defined in Text) at 0430 and 0530 UTC on 6 April 2018 (Click to enlarge)

NOAA’s Office of Satellite and Product Operations (OSPO) maintains a website that describes Eclipses and Keep Out Zones.

Adjustments to GOES-16 Radiances in Visible and near-Infrared Channels

April 29th, 2017 |

GOES-16 Visible (0.64) Imagery at 1145 UTC and 1200 UTC on 29 April 2017, before and after, respectively, a calibration change (Click to enlarge)

Reflectance values detected near Solar Noon over thick clouds can exceed 1 (i.e., the albedo will exceed 100%) because of contributions (caused by scattering) to a pixel from neighboring pixels. A fix that allows for these larger-than-one reflectance values in the GOES-16 Processing is scheduled to take place in July. Until July the values exceeding one will be treated as missing in AWIPS. The change discussed in this blog post, below, implemented on 29 April 2017, should reduce the number of those missing pixels.

GOES-16 has on-orbit calibration of visible and near-infrared channels. This is in contrast to legacy GOES (GOES-13, GOES-14 and GOES-15); the lack of on-orbit calibration means that visible imagery from GOES-13 and GOES-15 show different reflectance values, as shown in this image from 2000 UTC on 01 May 2017: GOES-13 shows smaller reflectance (the right half of the image) than GOES-15 (the left half of the image). Such degradation should not occur with GOES-16.

On-orbit calibration is achieved by viewing the Sun for a set period and determining the photons detected. That value is then related to the amount of solar energy reflected back from the Earth from bright surfaces. It’s important that the exact amount of time for the solar view is known; otherwise the reflectance cannot be calibrated correctly.

NESDIS Engineers and the NOAA CWG (Calibration Working Group) determined in April that visible and near-infrared reflectances were too large (by up to 10%) because the amount of time viewing the Sun was different than the time-value used in the calibration. Shortly before 1200 UTC on 29 April a change in the calibration module was made to mitigate this mis-match. Reflectances dropped about 10% because of this calibration change.

The animation above shows GOES-16 0.64 imagery before and after the calibration change and it’s difficult to note a reduction in the brightness; however, a scatterplot of the reflectance over the convection that is over the tropical Atlantic east of the Amazon (click here to see the region, it is shaded in purple; this image shows the 1145 UTC image with the region toggled on and off) does show higher values at 1145 vs. 1200 UTC, at a time during the day when the rising Sun should cause increases in reflectance. (A similar plot of 1215 UTC vs. 1200 UTC shows reflectance values with no bias).

Density Diagram of GOES-16 Reflectance Values at 1145 UTC (x-axis) vs. GOES-16 Reflectance Values at 1200 UTC (y-axis) (Click to enlarge)

Imagery in this blog post was created using the Satellite Information Familiarization Tool (SIFT).

GOES-16 data posted on this page are preliminary, non-operational data and are undergoing testing.