Chapter 13
Sea Surface Temperature Measurements
from Thermal Infrared Satellite Instruments:
Status and Outlook
Craig J. Donlon
13.1 Introduction
Thermal Infrared (TIR) sensors have been deployed on earth observing satellites for
over 30 years providing measurements of Sea Surface Temperature (SST), clouds
and many other products. Developed initially for meteorology and now used widely
by the oceanographic and climate communities, TIR derived SST measurements are
available in an operational context in Near Real Time (NRT) from a wide variety
of satellite missions. TIR sensors have a characteristically high spatial resolution
of 0.5–1.1 km (at nadir) with quasi global coverage on a daily basis (using two
operational wide swath TIR missions). TIR sensors are typically calibrated using
on-board reference blackbody systems alone (e.g. Corlett et al., 2006) or a combination of blackbody and deep-space “cold” views to an accuracy of 0.1–0.2 K (e.g.
Robinson, 2004). On-board calibration is sometimes supplemented with vicarious
calibration adjustments implicit in some Level-2 SST retrieval algorithms that compensate for the atmospheric attenuation of water leaving radiances using in-situ SST
measurements (e.g. Kilpatrick et al., 2001; Zhang et al., 2009). Other approaches to
atmospheric correction rely on the use of radiative transfer modes to derive look-uptables that can be applied to brightness temperature measurements using a suitable
SST retrieval algorithm (e.g. Merchant and Le Borgne, 2004).
This approach has the benefit of releasing in-situ observations for use in on-going
verification and validation work and for a more detailed investigation of sensor and
algorithm biases, essential activities for the production of fundamental climate data
records (e.g. Merchant et al., 2008b). Most importantly, well defined and error quantified measurements of SST are required for climate time series (in the form of
Fundamental Climate Data Records, or FCDR) that can be analyzed to reveal the
role of the ocean in short and long term climate variability.
This chapter first presents a summary of key TIR satellite sensors from 2000
to 2020. In Section 13.3 it outlines the primary on-going challenges and issues
C.J. Donlon (B)
ESTEC (EOP-SME), European Space Agency, Noordwijk, 2201 AZ, The Netherlands
e-mail: Craig.Donlon@esa.int
211
V. Barale et al. (eds.), Oceanography from Space,
DOI 10.1007/978-90-481-8681-5_13, C
Springer Science+Business Media B.V. 2010
Sea Surface Temperature Measurements
from Thermal Infrared Satellite Instruments:
Status and Outlook
Craig J. Donlon
13.1 Introduction
Thermal Infrared (TIR) sensors have been deployed on earth observing satellites for
over 30 years providing measurements of Sea Surface Temperature (SST), clouds
and many other products. Developed initially for meteorology and now used widely
by the oceanographic and climate communities, TIR derived SST measurements are
available in an operational context in Near Real Time (NRT) from a wide variety
of satellite missions. TIR sensors have a characteristically high spatial resolution
of 0.5–1.1 km (at nadir) with quasi global coverage on a daily basis (using two
operational wide swath TIR missions). TIR sensors are typically calibrated using
on-board reference blackbody systems alone (e.g. Corlett et al., 2006) or a combination of blackbody and deep-space “cold” views to an accuracy of 0.1–0.2 K (e.g.
Robinson, 2004). On-board calibration is sometimes supplemented with vicarious
calibration adjustments implicit in some Level-2 SST retrieval algorithms that compensate for the atmospheric attenuation of water leaving radiances using in-situ SST
measurements (e.g. Kilpatrick et al., 2001; Zhang et al., 2009). Other approaches to
atmospheric correction rely on the use of radiative transfer modes to derive look-uptables that can be applied to brightness temperature measurements using a suitable
SST retrieval algorithm (e.g. Merchant and Le Borgne, 2004).
This approach has the benefit of releasing in-situ observations for use in on-going
verification and validation work and for a more detailed investigation of sensor and
algorithm biases, essential activities for the production of fundamental climate data
records (e.g. Merchant et al., 2008b). Most importantly, well defined and error quantified measurements of SST are required for climate time series (in the form of
Fundamental Climate Data Records, or FCDR) that can be analyzed to reveal the
role of the ocean in short and long term climate variability.
This chapter first presents a summary of key TIR satellite sensors from 2000
to 2020. In Section 13.3 it outlines the primary on-going challenges and issues
C.J. Donlon (B)
ESTEC (EOP-SME), European Space Agency, Noordwijk, 2201 AZ, The Netherlands
e-mail: Craig.Donlon@esa.int
211
V. Barale et al. (eds.), Oceanography from Space,
DOI 10.1007/978-90-481-8681-5_13, C
Springer Science+Business Media B.V. 2010