New land surface temperature retrieval algorithm for heavy aerosol loading during nighttime from Gaofen-5 satellite...

Land surface temperature (LST) is a key variable used for studies of the water cycles and energy budgets of land-atmosphere interfaces. The Chinese Gaofen-5 (GF5) satellite, with an onboard visual and infrared multispectral imager (VIMS), is the only satellite that can capture the earth's thermal infrared information for use in the national high-resolution earth observation project of China; it can observe the earth surface at a high spatial resolution of 40 m in four thermal infrared channels and two mid-infrared channels. This article selects the optimum spectral channel combination for reducing the aerosol effect on LST retrieval with the aid of simulated data, and a new four-channel LST retrieval method from GF5 infrared data under heavy dust aerosol during nighttime is proposed. The results show that the channel combination of channels 7, 8, 9, and 10 (denoted as CC1) performed better than the combination of channels 7, 8, 11, and 12 (denoted as CC2). The root mean square errors (RMSEs) between the actual and estimated LST were 0.28 K for the CC1 group with an aerosol optical thickness (AOD) of 0.1 and 1.94 K for the CC1 group with an AOD of 1.0. The RMSEs for CC2 were 0.28 K for the group with an AOD of 0.1 and 2.54 K for the other group with an AOD of 1.0. Moreover, an error analysis for the proposed method was performed in terms of the noise equivalent temperature difference (NEΔT), the uncertainties of land surface emissivity (LSE), water vapor content (WVC) and AOD. The results show that the LST errors caused by an LSE uncertainty of 0.01, a NEΔT of 0.2 K, a WVC uncertainty of 20%, an AOD uncertainty of 0.1 were 0.31 ∼ 1.01 K, 0.4 ∼ 2.0 K, within 0.6 K, and within 0.3 K for CC1 and 0.32 ∼ 3.08 K, 0.4 ∼ 1.7 K, within 0.7 K, and within 0.3 K for CC2, respectively.