Thermal Airborne Spectrographic Imager Research Articles

Response of Industrial Warm Drainage to Tide Revealed by Airborne and Sea Surface Observations

Maintaining the balance between power station operation and environmental carrying capacity in the process of cooling water discharge into coastal waters is an essential issue to be considered. Earth observations with airborne and sea surface sensors can efficiently estimate distribution characteristics of extensive sea surface temperature compared with traditional numerical and physical simulations. Data acquisition timing windows for those sensors are designed according to tidal data. The airborne thermal infrared data (Thermal Airborne Spectrographic Imager, TASI) is preprocessed by algorithms of atmospheric correction, geometric correction, strip brightness gradient removal, and noise reduction, and then the seawater temperature is inversed in association with sea surface synchronous temperature measurement data (Sea-Bird Electronics, SBE). Verification analyses suggested a satisfied accuracy of less than about 0.2 °C error between the predicted and the measured values in general. Multiple factors influence seawater temperature, i.e., meteorology, ocean current, runoff, water depth, seawater convection, and eddy current; tidal activity is not the only one. Environmental background temperature in different seasons is the governing factor affecting the diffusion effect of seawater temperature drainage according to analyses of the covariances and correlation coefficients of eight tidal states. The present study presents an efficient and quick seawater temperature monitoring technique owing to industrial warm drainage to sea by means of a complete set of seawater temperature inversion algorithms with multi-source thermal infrared hyperspectral data.

Assessing the effects of urban green landscape on urban thermal environment dynamic in a semiarid city by integrated use of airborne data, satellite imagery and land surface model

The response of the urban thermal environment to green space landscapes has been studied previously, while its detailed pattern is insufficiently explored owing to the constraints of spatial and temporal resolution of available datasets. This study uses integrated Thermal Airborne Spectrographic Imager (TASI) data, Landsat TM/ETM and MODIS satellite imagery, and Noah land surface model output to investigate the effects of landscape pattern on the urban thermal environment across Shijiazhuang, China. The present study not merely proposes a generalized framework for the spatiotemporal analysis of urban thermal environment, but also affords several insights into the cooling effects related to urban green space landscapes in semiarid cities. Firstly, trees and lawns show a noticeable disparity in land surface temperature (LST) response to urban green landscape metrics, primarily due to the difference in cooling efficiency via evapotranspiration. This disparity can be explained further by the radiation-shading effect of trees. Secondly, analysis confirms that the composition of urban green space has a substantial impact on LST throughout summer. This pattern is largely stable for trees owing to the constant Bowen ratio, but they are altered for lawns. Conversely, the configurations of urban green space exhibit less impact on LST. These effects vary temporally in magnitude and can be enhanced notably in humid conditions. The lower correlation between the configuration metrics of urban green and LST has to do with the surface resistance alterations and additional cooling effects. Finally, consistent patterns of the impact of urban green space landscape metrics on LST are illustrated at different spatial scales and region sizes but greater effects are revealed for smaller analytical units, further confirming the impacts of urban green landscape on urban thermal environment.

Lithology Classification Using TASI Thermal Infrared Hyperspectral Data with Convolutional Neural Networks

In recent decades, lithological mapping techniques using hyperspectral remotely sensed imagery have developed rapidly. The processing chains using visible-near infrared (VNIR) and shortwave infrared (SWIR) hyperspectral data are proven to be available in practice. The thermal infrared (TIR) portion of the electromagnetic spectrum has considerable potential for mineral and lithology mapping. In particular, the abovementioned rocks at wavelengths of 8–12 μm were found to be discriminative, which can be seen as a characteristic to apply to lithology classification. Moreover, it was found that most of the lithology mapping and classification for hyperspectral thermal infrared data are still carried out by traditional spectral matching methods, which are not very reliable due to the complex diversity of geological lithology. In recent years, deep learning has made great achievements in hyperspectral imagery classification feature extraction. It usually captures abstract features through a multilayer network, especially convolutional neural networks (CNNs), which have received more attention due to their unique advantages. Hence, in this paper, lithology classification with CNNs was tested on thermal infrared hyperspectral data using a Thermal Airborne Spectrographic Imager (TASI) at three small sites in Liuyuan, Gansu Province, China. Three different CNN algorithms, including one-dimensional CNN (1-D CNN), two-dimensional CNN (2-D CNN) and three-dimensional CNN (3-D CNN), were implemented and compared to the six relevant state-of-the-art methods. At the three sites, the maximum overall accuracy (OA) based on CNNs was 94.70%, 96.47% and 98.56%, representing improvements of 22.58%, 25.93% and 16.88% over the worst OA. Meanwhile, the average accuracy of all classes (AA) and kappa coefficient (kappa) value were consistent with the OA, which confirmed that the focal method effectively improved accuracy and outperformed other methods.

Improved Temperature and Emissivity Separation Algorithm for Multispectral and Hyperspectral Sensors

The Temperature and Emissivity Separation (TES) algorithm was originally developed for the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). This paper focuses on improving the TES algorithm. The main modification is the replacement of the normalized emissivity module with a new module, which is based on the smoothing of spectral radiance signatures. Smoothing is performed by estimating emissivity using an optimized approximation of the relationship between brightness temperature and emissivity. The improved TES algorithm, which is called Optimized Smoothing for Temperature Emissivity Separation (OSTES), was first tested on simulated data from three different sensors, namely, ASTER, Airborne Hyperspectral Scanner, and Thermal Airborne Spectrographic Imager. Processing simulated data with OSTES shows that it produces more accurate and precise temperature and emissivity retrievals than TES for samples with low spectral contrast. These results also show that the OSTES is less sensitive to variations in atmosphere and sample temperatures. The OSTES performance was secondly tested and compared with ASTER standard products. Testing showed that the OSTES temperature retrievals agree with AST_08 (kinetic temperature) but the emissivity retrievals differ substantially from AST_05 (surface emissivity). This implies that the ASTER standard products, AST_08 and AST_05, are not consistent with each other. Emissivities that are derived using temperatures from AST_08 and ASTER thermal surface radiances from AST_09T perform similar or worse when compared with OSTES emissivities. Emissivities delivered by AST_05 also exhibit step discontinuities, whereas the OSTES emissivitites are smoother over homogeneous surfaces. However, all emissivity spectra are affected by imperfect atmospheric corrections.

Analysis of the Urban Heat Island Effect in Shijiazhuang, China Using Satellite and Airborne Data

The urban heat island (UHI) effect resulting from rapid urbanization generally has a negative impact on urban residents. Shijiazhuang, the capital of Hebei Province in China, was selected to assess surface thermal patterns and its correlation with Land Cover Types (LCTs). This study was conducted using Landsat TM images on the mesoscale level and airborne hyperspectral thermal images on the microscale level. Land surface temperature (LST) was retrieved from four scenes of Landsat TM data in the summer days to analyze the thermal spatial patterns and intensity of surface UHI (SUHI). Surface thermal characteristics were further examined by relating LST to percentage of imperious surface area (ISA%) and four remote sensing indices (RSIs), the Normalized Difference Vegetation Index (NDVI), Universal Pattern Decomposition method (VIUPD), Normalized Difference Built-up Index (NDBI) and Biophysical Composition Index (BCI). On the other hand, fives scenes of airborne TASI (Thermal Airborne Spectrographic Imager sensor) images were utilized to describe more detailed urban thermal characteristics of the downtown of Shijiazhuang city. Our results show that an obvious surface heat island effect existed in the study area during summer days, with a SUHI intensity of 2–4 °C. The analyses reveal that ISA% can provide an additional metric for the study of SUHI, yet its association with LST is not straightforward and this should a focus in future work. It was also found that two physically based indices, VIUPD and BCI, have the potential to account for the variation in urban LST. The results concerning on TASI indicate that diversity of impervious surfaces (rooftops, concrete, and mixed asphalt) contribute most to the SUHI, among all of the land cover features. Moreover, the effect of impervious surfaces on LST is complicated, and the composition and arrangement of land cover features may play an important role in determining the magnitude and intensity of SUHI. Overall, the analysis of urban thermal signatures at two spatial scales complement each other and the use of airborne imagery data with higher spatial resolution is helpful in revealing more details for understanding urban thermal environments.

Temperature and emissivity separation and mineral mapping based on airborne TASI hyperspectral thermal infrared data

Thermal infrared remote sensing (8–12μm) (TIR) has great potential for geologic remote sensing studies. TIR has been successfully used for terrestrial and planetary geologic studies to map surface materials. However, the complexity of the physics and the lack of hyperspectral data make the studies under-investigated. A new generation of commercial hyperspectral infrared sensors, known as Thermal Airborne Spectrographic Imager (TASI), was used for image analysis and mineral mapping in this study. In this paper, a combined method integrating normalized emissivity method (NEM), ratio algorithm (RATIO) and maximum–minimum apparent emissivity difference (MMD), being applied in multispectral data, has been modified and used to determine whether this method is suitable for retrieving emissivity from TASI hyperspectral data. MODTRAN 4 has been used for the atmospheric correction. The retrieved emissivity spectra matched well with the field measured spectra except for bands 1, 2, and 32. Quartz, calcite, diopside/hedenbergite, hornblende and microcline have been mapped by the emissivity image. Mineral mapping results agree with the dominant minerals identified by laboratory X-ray powder diffraction and spectroscopic analyses of field samples. Both of the results indicated that the atmospheric correction method and the combined temperature–emissivitiy method are suitable for TASI image. Carbonate skarnization was first found in the study area by the spatial extent of diopside. Chemical analyses of the skarn samples determined that the Au content was 0.32–1.74g/t, with an average Au content of 0.73g/t. This information provides an important resource for prospecting for skarn type gold deposits. It is also suggested that TASI is suitable for prospect and deposit scale exploration.

A Physics-Based Unmixing Method to Estimate Subpixel Temperatures on Mixed Pixels

This paper presents a new algorithm for the analysis of linear spectral mixtures in the thermal infrared domain, with the goal to jointly estimate the abundance and the subpixel temperature in a mixed pixel, i.e., to estimate the relative proportion and the temperature of each material composing the mixed pixel. This novel approach is a two-step procedure. First, it estimates the emissivity and the temperature over pure pixels using the standard temperature and emissivity separation (TES) algorithm. Second, it estimates the abundance and the subpixel temperature using a new unmixing physics-based model, called Thermal Remote sensing Unmixing for Subpixel Temperature (TRUST). This model is based on an estimator of the subpixel temperature obtained by linearizing the black body law around the mean temperature of each material. The abundance is then retrieved by minimizing the reconstruction error with the estimation of the subpixel temperatures. The TRUST method is benchmarked on simulated scenes against the fully constrained least squares unmixing applied on the radiance and on the estimation of surface emissivity using the TES algorithm. The TRUST method shows better results on pure and mixed pixels composed of two materials. TRUST also shows promising results when applied on thermal hyperspectral data acquired with the Thermal Airborne Spectrographic Imager during the Detection in Urban scenario using Combined Airborne imaging Sensors campaign and estimates coherent localization of mixed-pixel areas.

Scale Mismatch Between <italic>In Situ</italic> and Remote Sensing Observations of Land Surface Temperature: Implications for the Validation of Remote Sensing LST Products

The validation of remote sensing land surface temperature (LST) products is vital for their broad application. Conventional validation methods use ground-based measurements to evaluate the LSTs retrieved from remote sensing data. However, it is difficult to directly validate low or medium spatial resolution LST products because of the scale mismatch between in situ and remote sensing observations. In this letter, we compare two ground-based in situ observations with Moderate Resolution Imaging Spectroradiometer (MODIS) 1-km-resolution LST products (MOD11A1). We focus on the differences in scale between the ground-based measurements and the MODIS observations. A comparison of the results obtained during the daytime with those obtained at night indicates that stronger heterogeneity leads to a greater scale-mismatch effect. Because the LST heterogeneity influences the mismatch effect, semivariance is employed to analyze the heterogeneity of the MODIS 1-km mixed-pixel data using the LSTs retrieved from the high-resolution Thermal Airborne Spectrographic Imager and the Advance Spaceborne Thermal Emission and Reflection Radiometer data.

Estimation of soil organic carbon from airborne hyperspectral thermal infrared data: a case study

SummaryAirborne hyperspectral imagery in the thermal infrared region (8.0–11.5 µm) was acquired with a thermal airborne spectrographic imager (TASI‐600) on bare soils in agricultural fields in Pontecagnano (southern Italy). These data were related to ground sampling in order to assess the capability of this technology to predict topsoil properties at the field scale. Emissivity spectra were used to calibrate prediction models for the prediction of clay, sand and soil organic carbon (SOC) by partial least squares regression (PLSR) and Cubist regression techniques. The TASI‐600 predictive models were validated by leave‐one‐out cross‐validation. The results were compared with those obtained under laboratory conditions using both Fourier transform infrared (FT‐IR) and visible and near‐infrared (VNIR) spectroscopy. The accuracy of the predictive models was assessed in terms of R2, root mean square error (RMSE), the ratio of the performance to deviation (RPD) and the ratio of performance to interquartile range (RPIQ). In the laboratory, results obtained from the Fourier transform infrared were better than those obtained from visible and near‐infrared (VNIR) for the prediction of specific soil characteristics (sand, clay and SOC). The use of airborne data resulted in less accurate predictions than the FT‐IR data obtained in the laboratory (resampled to the TASI‐600 spectral characteristics). For TASI‐600 airborne data, SOC was predicted more accurately (RPIQ = 1.96; RMSE = 0.26%) than clay and sand content. The results obtained in this study demonstrate the good potential of long‐wave infrared (LWIR) remote sensing data for the quantitative estimation of the SOC content of topsoil.

Multiscale analysis of urban thermal characteristics: case study of Shijiazhuang, China

The urban heat island (UHI) effect caused by urbanization generally leads to adverse economic and environmental impacts. Thus, a detailed assessment of the thermal properties over individual land cover types at different spatial resolutions is required to better understand the establishment of UHI. Two scale levels are utilized to examine surface thermal characteristics in the case of Shijiazhuang, China. At the regional level, Landsat TM data are used to extract subpixel impervious surface area (ISA) and to inverse land surface temperature (LST) by means of multiple endmember spectral mixture analysis. Urban thermal characteristics are analyzed by relating the LST to normalized difference vegetation index (NDVI) and ISA. On the other hand, thermal airborne spectrographic imager data, with a high spatial resolution, are employed to describe the spatial distribution of urban thermal patterns at the local level. Results indicate that there is an approximate linear relationship among LST, NDVI, and ISA. In addition, the thermal characteristics over each land cover type are consistent at both levels, suggesting UHI is evident at Shijiazhuang and impervious surface is contributing most to this phenomenon. This confirms that different spatial scales are requested in UHI studies.