The new single-channel approaches for retrieving land surface temperature and the preliminary results

Two satellites named HJ-1A and HJ-1B were launched on 6 September 2008, which are intended for environment and disaster monitoring and forecasting. The infrared scanner (IRS) onboard HJ-1B has one thermal infrared band. Currently, for sensors with one thermal band (e.g. Landsat TM/ETM+ and HJ-1B), several empirical algorithms have been developed to estimate land surface temperature (LST). However, surface emissivity and atmospheric parameters which are not readily accessible to general users are required for these empirical methods. To resolve this problem, particularly for HJ-1B, new retrieval methodology is desired. According to proper assumptions, two approaches were proposed, which included the single-channel method based on temporal and spatial information (MTSC) and the image based single-channel method (IBSC). The newly developed methods are mainly for estimating LST accurately from one thermal band, even without any accurate information related to the atmospheric parameters and land surface emissivity. In this paper, we introduce and give preliminary assessments on the new approaches. Assessments generally show good agreement between the HJ-1B retrieved results and the MODIS references. Especially, over sea and water areas the biases were less than 1K while the root mean square errors were about 1K for both MTSC and IBSC methods. As expected, the MTSC method did superiorly to the IBSC method, owning to spatiotemporal information is incorporated into the MTSC method, although more experiments and comparisons should be conducted further.