Land surface temperature measurements from the split window channels of the NOAA 7 Advanced Very High Resolution Radiometer

Abstract

The Advanced Very High Resolution Radiometer (AVHRR) on the NOAA 7 satellite acquires 1‐km spatial resolution data in “split window” channels at 10.8 and 11.9 μm. Data from these spectral channels may be used to estimate surface temperature and the atmospheric correction to radiation from the earth's surface. Analysis of a data set from July 1981 shows that (1) there is satisfactory agreement between the equation resulting from radiative transfer theory and the atmospheric correction algorithm as obtained by analysis of an area of incipient cloud street formation; (2) agreement is also satisfactory between this algorithm and the statistically derived NOAA algorithm used to obtain sea surface temperatures from the satellite data (However, the comparison assumes the NOAA algorithm is valid outside its range of derivation.); (3) in areas of cloud street formation, variations of atmospheric moisture produce radiance temperature differences of order 2–3°C, which if neglected would cause errors in the derivation of surface thermal characteristics. This meteorological variation over distances of 5–10 km would not be inferred from conventional radiosonde measurements or from lower‐resolution satellite soundings.

Validation of the land-surface temperature products retrieved from Terra Moderate Resolution Imaging Spectroradiometer data

“…LST can be retrieved from a single infrared channel through an accurate radiative transfer model if surface emissivity is known and temperature/water vapor profile is given by either satellite soundings or conventional radiosonde data (Price, 1983;Susskind, Rosenfield, Reuter, & Chahine, 1984;Chedin, Scott, Wahiche, & Moulinier, 1985;Ottlé & Vidal-Madjar, 1992). Split-window LST methods require known surface emissivities to make corrections for the atmospheric and surface emissivity effects based on the differential atmospheric absorption in the 10-13 Am split window without knowledge of the atmospheric temperature/water vapor profile although column water vapor is used in some split-window LST algorithms to improve the accuracy of LST retrieval (Price, 1984;Becker, 1987;Wan & Dozier, 1989;Becker & Li, 1990;Sobrino, Coll, & Caselles, 1991;Vidal, 1991;Kerr, Lagouarde, & Imbernon, 1992;Ottlé & Stoll, 1993;Prata, 1994;Wan & Dozier, 1996). Because the accuracy of LST retrieved by single channel methods and split-window methods depends on the accuracy of surface emissivity, these methods do not work well in semi-arid and arid regions, where surface emissivity may vary significantly with location and time.…”

“…In comparison, there is no standard global LST data product derived from satellite remote sensing data even though the use of thermal -infrared (TIR) measurements for analysis of land biophysical conditions has been under investigation for more than three decades (Fuchs & Tanner, 1966) and the AVHRR data have been used to produce LST data in the development of LST algorithms for two decades. It is well known that simple extension of the SST methods to LST for AVHRR data would lead to unacceptable errors (Price, 1984;Becker, 1987) because of the difficulty in cloud detection with AVHRR data over land (especially for thin cirrus) and the intrinsic difficulties in the LST retrieval (Wan & Dozier, 1989).…”

Introducing multisensor satellite radiance-based evaluation for regional Earth System modeling

“…Among satellite raw data, the IR window region (8μm to 14μm) is the most familiar "raw radiance" data widely used in the meteorological community. IR Tb emitted from the Earth's surface or from a cloud are used to estimate cloud top temperature [Inoue, 1987;Rossow and Lacis, 1990;King et al, 1992;Menzel et al, 2008] or land surface skin temperature [Price, 1984;Wan and Dozier, 1996;Mao et al, 2005] and for evaluation and assimilation of clouds [Chaboureau et al, 2002;Vukicevic et al, 2006;Zupanski et al, 2011b]. IR Tb is available for both daytime and nighttime, while the visible channel is generally available only for daytime.…”

Detection of Sangeang Api Volcanic Ash Cloud Based on VBICA-SVM method

“…Add noise information in the standard ICA model, making it the linear mixed ICA model with noise [9,10], and the formula is shown below: ( ) ( ) ( ) x t As t t ε = +…”