+0.73 degrees C After NOAA-19 satellite record was truncated – did the watts increase?

From Dr. Spencer's Global Warming Blog

Author: Dr. Roy W. Spencer

The global mean lower tropospheric temperature (LT) anomaly in version 6.1 for October 2024 is +0.73 degrees. C deviates from the 1991-2020 average and is +0.80 degrees below the September 2024 anomaly. C.

The new (version 6.1) global regional average temperature trend (January 1979 to October 2024) is now +0.15 °C/decade (+0.21 C/decade over land, +0.13 C/decade over ocean).

Previous (version 6.0) trends up to September 2024 were +0.16 C/decade (global), +0.21 C/decade (land) and +0.14 C/decade (ocean).

The following provides background on the changes leading to the new version (v6.1) of the UAH data set.

focus

• The older NOAA-19 satellites now drift too far into the day-night cycle for our drift correction methods to provide useful adjustments. Therefore, we have decided to truncate NOAA-19 data processing starting in 2021. This truncation is consistent with truncation made on previous satellites after orbital drift began to affect temperature measurements.
• This change only slightly reduces recent global warming records, making our calculated global temperatures more consistent with RSS and NOAA satellite data sets over the past 2-3 years.
• Despite the recent decrease in temperatures, the 1979-2024 trend decreases by only 0.01°C/decade, from +0.16 C/decade to +0.15 C/decade. Recent temperatures between 2023 and 2024 remain satellite-era records, with every month since October 2023 setting a record for that calendar month.

background

Monitoring global deep atmospheric temperatures using satellite microwave radiometers (systems originally designed for routine global weather monitoring) has always required corrections and adjustments to calibration data to enable long-term trend detection. The most important of these fixes/adjustments are:

1. satellite calibration biasrequiring mutual calibration between consecutively launched satellites during overlapping service coverage periods. These adjustments are typically tenths of a degree Celsius.
2. The drift of an orbit away from its nominal sun-synchronous observation timeneed to be based on comparisons of drifting and non-drifting satellites (UAH method) or climate models (remote sensing systems [RSS] method, which I believe is also used by the NOAA data set). These corrections can go up to 1 degree. C or higher tropospheric lower temperature (LT) products, especially over land and in summer.
3. Correcting the influence of body temperature on the instrument Calibrate temperature (only older MSU instruments have this problem, it produces spurious temperature rises).
4. Track height attenuation adjustment For the multi-view version of the Lower Troposphere (LT) product (as of V6.0, the UAH data set is no longer required, which uses multiple channels instead of multiple angles from a single channel.)

The second adjustment (day and night drift) is the change theme from UAH v6.0 to v6.1. The graph below shows the equatorial crossing times (local solar time) of the individual satellites that make up the satellite temperature record. As afternoon observations transition from afternoon to evening, the drift of the satellites (except for the non-drifting Aqua and MetOp satellites, which have fuel on board to allow orbital maintenance) brings cooling to the LT measurements of the afternoon satellites. As nighttime observations transition into the late afternoon, the drift of morning satellites increases their low temperatures.

The red vertical line indicates the date when satellite data is no longer included in v6.0 (UAH) processing, with the addition of NOAA-19 truncation for v6.1. Note that the NOAA-19 satellite drifted further during the local observing time than any previous afternoon satellite. The NOAA-19 local observation period is beyond our training dataset, which includes the assumption that diurnal temperatures drift linearly through time. Therefore, we have decided that it is now necessary to truncate the NOAA-19 data starting in 2021, which we are now doing with the October 2024 update.

As a result, version 6.1 of our dataset has begun, with a name change intended to reduce confusion and signal a significant change in the way we approach it. As shown in the figure above, 2020, the last year for NOAA-19 data inclusion, is roughly consistent with the v6.0 cutoff for the NOAA-18 and NOAA-14 (PM) satellites.

This type of change in our process is similar to changes we made in previous years, after several years of data collection to determine the existence of a problem. The time delay is necessary because we previously found that the processing temperatures of two operating satellites in different orbits can differ but converge again later. As shown below, we now have sufficient reason to truncate the NOAA-19 data record starting in 2021.

If the satellite drifts during the local observation time, why do we include it?

There are three reasons for including daily drifting satellites (imperfectly adjusted): (1) Most satellites during the 1979-2024 record period drifted, so including them is essential for producing a complete, intercalibrated satellite records are necessary temperatures; (2) two operational satellites (usually one with much larger drifts than the other) provide a more complete sampling of our gridded dataset during the month, which The error is 2.5 degrees. Latitude/longitude resolution; (3) Having two (sometimes 3) satellites allows monitoring of potential drift, i.e. the time series of differences between the 2 satellite measurements should remain relatively stable over time.

Version 6.1 brings UAH data closer to RSS and NOAA from past years

Some people have noticed that our temperature anomalies are higher than those of the RSS or NOAA satellite products. It now appears that this is due to NOAA-19's orbit drifting beyond the useful range of our drift corrections. The plot below (preliminary, provided to me by John Christy) shows that the truncation of NOAA-19 records now brings UAH anomalies more in line with RSS and NOAA products.

As can be seen, this change has significantly reduced recent global average temperatures. For example, without the NOAA-19 truncation, the October anomaly would be +0.94 degrees. C, but only MetOp-B after 2020, which is now +0.73 degrees. C.

The following table lists the deviation of version 6.1 LT by region from the 30-year (1991-2020) average over the past 22 months (all-time highs are shown in red):

The full UAH global temperature report, as well as LT global grid point anomaly images for October 2024, along with more detailed analysis by John Christy, should be available here in the next few days. Since the data set has changed from v6.0 to v6.1, it may take longer this time.

Monthly anomalies for each of the four deep areas we monitor via satellite will be available over the next few days (with possible delays):

Lower troposphere:

http://vortex.nsstc.uah.edu/data/msu/v6.1/tlt/uahncdc_lt_6.1.txt

Middle troposphere:

http://vortex.nsstc.uah.edu/data/msu/v6.1/tmt/uahncdc_mt_6.1.txt

Tropopause:

http://vortex.nsstc.uah.edu/data/msu/v6.1/ttp/uahncdc_tp_6.1.txt

Lower stratosphere:

http://vortex.nsstc.uah.edu/data/msu/v6.1/tls/uahncdc_ls_6.1.txt

Relevant