Abstract
42-years of Arctic and Antarctic Surface Temperatures from thermal Infrared satellite radiometers (AASTI) are
presented as the Copernicus Climate Change Service Ice surface temperature record v1.1 dataset (C3S IST).
It covers snow, ice and ocean surfaces with mean and max–min daily temperatures poleward of 50 degrees
North and South, for the period 1982–2023. The C3S IST is provided as a Level 3 (L3) dataset in a polar
0.25 ◦ latitude and longitude grid. It consists of two parts: (1) the C3S IST climate data record (ISTCDR v1.1),
covering the period 1. January 1982 to 30. June 2019, and (2) the C3S IST Interim CDR version 1.1 (ICDR
v1.1) covering 1. July 2019 to 31. December 2023. The surface temperatures (STs) are calculated from satellite
thermal infrared Brightness Temperature (𝑇𝐵 ) measurements from the Global Area Coverage – Advanced Very
High Resolution Radiometer (GAC - AVHRR) data, creating a comprehensive data set based solely on a single
sensor type. The underlying AASTI algorithm is a combination of algorithms specifically tuned for sea ice,
marginal ice zone, land ice and high latitude open water. In addition, each of the algorithm coefficients are
tuned specifically for each of the AVHRR instruments, using simulated Top of the atmosphere (TOA) 𝑇𝐵 s and
ERA-Interim reanalysis surface and atmosphere data. Simulated TOA 𝑇𝐵 ’s are computed using the community
radiative transfer model, RTTOV v12.3. Spatially and temporally varying uncertainties are computed for each
data-point. The C3S IST surface temperatures were validated against different in situ observation types, where
comparison against radiometric temperatures from flight campaigns and ice sheet station data resulted in
smaller mean differences of 0.20◦C and -1.84◦C in the Northern Hemisphere (NH) than for validations against
met station air temperatures, which were usually 1–3◦C higher. Surface temperature climatology and trends
have been computed for sea ice and ice sheets, showing large regional differences in surface temperature
trends within the NH. For the entire dataset, the average trend is +1.11 ◦C/decade for the NH sea ice, +0.16
◦C/decade for the Southern Hemisphere (SH) sea ice, +0.38 ◦C/decade for the Greenland ice sheet and -0.13
◦C/decade for the Antarctic ice sheet. The positive trends are typically small during summer and larger during
winter, e.g. in the Barents Sea, where trends exceed +0.3 ◦C/year in winter. Negative temperature trends are
observed in some regions such as the Bering Strait’s ice edge. For Antarctic sea ice, the trends are generally
smaller and statistically less significant than those in the NH. The trends of the ice sheet surface show a
similar pattern, where the Greenland Ice Sheet has positive trends, which both are largest and most significant
in the coastal areas. The Antarctic Ice Sheet has diverse trends and large areas have statistically non-significant
trends, except for the negative temperature trend in the central, high altitude part of east Antarctica exceeding
-0.5 ◦C/year.
presented as the Copernicus Climate Change Service Ice surface temperature record v1.1 dataset (C3S IST).
It covers snow, ice and ocean surfaces with mean and max–min daily temperatures poleward of 50 degrees
North and South, for the period 1982–2023. The C3S IST is provided as a Level 3 (L3) dataset in a polar
0.25 ◦ latitude and longitude grid. It consists of two parts: (1) the C3S IST climate data record (ISTCDR v1.1),
covering the period 1. January 1982 to 30. June 2019, and (2) the C3S IST Interim CDR version 1.1 (ICDR
v1.1) covering 1. July 2019 to 31. December 2023. The surface temperatures (STs) are calculated from satellite
thermal infrared Brightness Temperature (𝑇𝐵 ) measurements from the Global Area Coverage – Advanced Very
High Resolution Radiometer (GAC - AVHRR) data, creating a comprehensive data set based solely on a single
sensor type. The underlying AASTI algorithm is a combination of algorithms specifically tuned for sea ice,
marginal ice zone, land ice and high latitude open water. In addition, each of the algorithm coefficients are
tuned specifically for each of the AVHRR instruments, using simulated Top of the atmosphere (TOA) 𝑇𝐵 s and
ERA-Interim reanalysis surface and atmosphere data. Simulated TOA 𝑇𝐵 ’s are computed using the community
radiative transfer model, RTTOV v12.3. Spatially and temporally varying uncertainties are computed for each
data-point. The C3S IST surface temperatures were validated against different in situ observation types, where
comparison against radiometric temperatures from flight campaigns and ice sheet station data resulted in
smaller mean differences of 0.20◦C and -1.84◦C in the Northern Hemisphere (NH) than for validations against
met station air temperatures, which were usually 1–3◦C higher. Surface temperature climatology and trends
have been computed for sea ice and ice sheets, showing large regional differences in surface temperature
trends within the NH. For the entire dataset, the average trend is +1.11 ◦C/decade for the NH sea ice, +0.16
◦C/decade for the Southern Hemisphere (SH) sea ice, +0.38 ◦C/decade for the Greenland ice sheet and -0.13
◦C/decade for the Antarctic ice sheet. The positive trends are typically small during summer and larger during
winter, e.g. in the Barents Sea, where trends exceed +0.3 ◦C/year in winter. Negative temperature trends are
observed in some regions such as the Bering Strait’s ice edge. For Antarctic sea ice, the trends are generally
smaller and statistically less significant than those in the NH. The trends of the ice sheet surface show a
similar pattern, where the Greenland Ice Sheet has positive trends, which both are largest and most significant
in the coastal areas. The Antarctic Ice Sheet has diverse trends and large areas have statistically non-significant
trends, except for the negative temperature trend in the central, high altitude part of east Antarctica exceeding
-0.5 ◦C/year.
Original language | English |
---|---|
Article number | 114816 |
Journal | Remote Sensing of Environment |
Volume | 328 |
Number of pages | 18 |
ISSN | 0034-4257 |
DOIs | |
Publication status | Published - 2025 |
Keywords
- AASTI
- AVHRR
- Antarctic
- Arctic
- Ice surface temperature (IST)
- Infrared satellite observations