Inconsistency among Landsat Sensors in Land Surface Mapping: A Comprehensive Investigation Based on Simulation

Comprehensive investigations on the between-sensor comparability among Landsat sensors have been relatively limited compared with increasing use of multi-temporal records in time series analyses. More seriously, sensor-related difference has not always considered applications. Accordingly, comparisons were conducted all available currently, including Multispectral Scanner (MSS), Thematic Mappers (TM), Enhanced (ETM+), and Operational Land Imager (OLI)) land cover mapping, based a collection synthesized, multispectral data. Compared to TM, OLI showed obvious differences channel reflectance, especially over near infrared (NIR) shortwave (SWIR) channels, presented positive bias vegetation spectral indices. did outperform TM ETM+ classification, which related methods used. Furthermore, channels SWIR its successors contributed largely enhancement inter-class separability improvement classification. Currently, inclusion MSS data is confronted significant challenges regarding consistency surface mapping. Considering inconsistency sensors, it applicable generate consistent indices through proper transformation models. Meanwhile, suggests generation specific class(es) interest instead classes simultaneously.