Remotely sensed, spatially continuous and high spatiotemporal resolution (hereafter referred to as high resolution) land surface temperature (LST) is a key parameter for studying the thermal environment and has important applications in many fields. However, difficult atmospheric conditions, sensor malfunctioning and scanning gaps between orbits frequently introduce spatial discontinuities into satellite-retri1eved LST products. For a single sensor, there is also a trade-off between temporal and spatial resolution and, therefore, it is impossible to obtain high temporal and spatial resolution simultaneously. In recent years the reconstruction and spatiotemporal fusion of LST products have become active research topics that aim at overcoming this limitation. They are two of most investigated approaches in thermal remote sensing and attract increasing attention, which has resulted in a number of different algorithms. However, to the best of our knowledge, currently no review exists that expatiates and summarizes the available LST reconstruction and spatiotemporal fusion methods and algorithms. This paper introduces the principles and theories behind LST reconstruction and spatiotemporal fusion and provides an overview of the published research and algorithms. We summarized three kinds of reconstruction methods for missing pixels (spatial, temporal and spatiotemporal methods), two kinds of reconstruction methods for cloudy pixels (Satellite Passive Microwave (PMW)-based and Surface Energy Balance (SEB)-based methods) and three kinds of spatiotemporal fusion methods (weighted function-based, unmixing-based and hybrid methods). The review concludes by summarizing validation methods and by identifying some promising future research directions for generating spatially continuous and high resolution LST products.
Deep Dive into Spatially Continuous and High-resolution Land Surface Temperature: A Review of Reconstruction and Spatiotemporal Fusion Techniques.
Remotely sensed, spatially continuous and high spatiotemporal resolution (hereafter referred to as high resolution) land surface temperature (LST) is a key parameter for studying the thermal environment and has important applications in many fields. However, difficult atmospheric conditions, sensor malfunctioning and scanning gaps between orbits frequently introduce spatial discontinuities into satellite-retri1eved LST products. For a single sensor, there is also a trade-off between temporal and spatial resolution and, therefore, it is impossible to obtain high temporal and spatial resolution simultaneously. In recent years the reconstruction and spatiotemporal fusion of LST products have become active research topics that aim at overcoming this limitation. They are two of most investigated approaches in thermal remote sensing and attract increasing attention, which has resulted in a number of different algorithms. However, to the best of our knowledge, currently no review exists that
Land surface temperature (LST) is a crucial parameter in investigating environmental and ecological processes (Hansen et al. 2010;Tierney et al. 2008), and is also valuable in studies of evapotranspiration, soil moisture conditions, heat-related health issues and urban heat islands (Anderson et al. 2012;Sellers et al. 1997). From a climate perspective, LST is important for evaluating land surface and land-atmosphere exchange processes, constraining surface energy budgets and model parameters, and providing observations of surface temperature change both globally and in key regions (Guillevic et al. 2017). Satellite remote sensing offers the only possibility to measure LST over extended regions with acceptable temporal resolution and complete spatial coverage (Li et al. 2013b;Wan et al. 2004).
Satellite-derived thermal infrared (TIR) data has a relatively high spatial resolution with acceptable accuracy (Wan et al. 2004). Various algorithms (e.g., single-channel, split-window, and temperature and emissivity separation) have been devised to derive operational LST products (Li et al. 2013b). For example, the Moderate Resolution Imaging Spectroradiometer (MODIS) (Wan et al. 2004), the FengYun-2/3 (FY-2/3) Visible Infrared Scanning Radiometer (VIRR) and the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) LST products (Trigo et al. 2008) are available for public use. However, TIR-based retrieval algorithms only work well for data acquired under clearsky conditions and without any instrument faults (Duan et al. 2017). The spatial continuity of LST is strongly affected by pixel with invalid or missing values caused by clouds or cloud shadow (hereafter referred to as cloudy pixels). On average cloudy-sky conditions account for more than half of the day-to-day weather around the globe (Jin 2000). For example, more than 60% of MODIS LST are contaminated by clouds (Cornette and Shanks 1993). Furthermore, some cloud-free but naturally bright pixels are frequently classified as cloud covered and their corresponding LST are set to the missing pixel value (Yang et al. 2019). For another, for a single sensor, the trade-off between temporal and spatial resolution often makes it impossible to obtain LST with the high temporal and spatial resolutions required by some applications (Wu et al. 2015c). Generally, LST retrieved from sensors with fine spatial resolution have poor temporal resolution, which leads to temporal discontinuities, as shown in Figure 1(a).
Satellite passive microwave (PMW) measurements are attractive for retrieving (sub-)surface temperature, especially under cloudy conditions, because they are much less affected by clouds and water vapor than TIR measurements (Holmes et al. 2016;Shwetha and Kumar 2016). However, the spatial resolution of PMW measurements (e.g., 25 km for AMSR-E) is much lower than that of TIR measurements. Besides, missing information caused by defective sensors (e.g., Landsat ETM+ SLC-off data) (Shen et al. 2016a) and scanning gap between orbits (e.g., Auqa/AMSR-E 、 GCOM/AMSR2 data) also introduce spatial discontinuities into LST products (Duan et al. 2017).
The above-mentioned spatial discontinuities and the restrictions on simultaneous spatial and temporal resolution seriously hinder applications of LST products in many fields. For instance, urban heat islands (UHI) can be continuously observed with AMSR-E, FY-2/3 and MODIS etc., but their associated spatial resolutions are too coarse to reveal detailed UHI spatial patterns. More spatial details can be observed in Landsat, ASTER and HJ images; however, due to their long revisit cycles (more than 15 days) and frequent cloud contamination, different LST scenes acquired by these sensors differ considerably in their space-time observation conditions (Shen et al. 2016a). As a result, for practical applications there is an increasingly urgent demand for spatially continuous high-resolution LST products (SCHR-LST).
Because of the growing number of available satellite LST products, many different approaches for generating SCHR-LST have been proposed, resulting in numerous publications on SCHR-LST algorithms and methods. Therefore, it is important and timely to present an overview of the state of the art in SCHR-LST methods. Although there have been earlier a review on the disaggregation of LST to finer temporal and spatial resolutions by Zhan et al. (2013), it reviewed from the perspective of thermal sharpening and temperature unmixing. However, some recently proposed methods (e.g., spatiotemporal fusion methods) to obtain LST with finer temporal and spatial resolutions, also should give a survey. Furthermore, to our best knowledge, a thorough review of methods and algorithms for deriving spatially continuous LST has not been performed.
The objective of this paper is to review methods for generating spatially continuous highresolution LST, describe the state-of-the-art, and identify the most promising research fields, thereby ultimat
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
