Remote Sensing of Environment最新文献

Operational canopy height mapping at high resolution remains a challenging task at country-level. Most of the existing state-of-the-art inversion methods propose physically-based schemes which are specifically tuned for local scales. Only few approaches in the literature have attempted to produce country or global scale estimates, mostly by means of data-driven approaches and multi-spectral data sources. In this paper, we propose a robust deep learning approach that exploits single-pass interferometric TanDEM-X data to generate accurate forest height estimates from a single interferometric bistatic acquisition. The model development is driven by considerations on both the final performance and the trustworthiness of the model for large-scale deployment in the context of tropical forests. We train and test our model over the five tropical sites of the AfriSAR 2016 campaign, situated in the West Central state of Gabon, performing spatial cross-validation experiments to test its generalization capability. We define a specific training dataset and input predictors to develop a robust model for country-scale inference, by finding an optimal trade-off between the model performance and the large-scale reliability. The proposed model achieves an overall estimation bias of 0.12 m, a mean absolute error of 3.90 m, a root mean squared error of 5.08 m and a coefficient of determination of 0.77. Finally, we generate a time-tagged country-scale canopy height map of Gabon at 25 m resolution, discussing the potential and challenges of these kinds of products for their application in different scenarios and for the monitoring of forest changes.

Ocean tide loading (OTL) displacements, shown as long-wavelength errors in Interferometric Synthetic Aperture Radar (InSAR), must be considered in large-scale applications. Despite efforts to explore the impacts of OTL on InSAR, most studies use individual interferograms and simple metrics, which fail to characterize the spatial structure of OTL. Moreover, the OTL contribution to InSAR time series remains relatively unexplored. The aliasing effect and related biases due to OTL, which are common to space-geodetic time series, are primarily theoretical with few practical observations for InSAR. This study comprehensively explores the statistical properties of OTL and their impacts on InSAR measurements, using the Southwest United Kingdom and Northwest France as study areas. Spatially, OTL artifacts on interferograms exhibit an escalating magnitude along the principal direction that aligns with the coastline's orientation. Temporally, the aliasing effect originating from OTL introduces periodic signals with prominent 15/64-day cycles into the Sentinel-1 InSAR time series, causing high velocity biases (up to ∼1 cm/yr) and uncertainties (up to ∼5 mm/yr) for short time spans. Applying OTL correction mitigates the noise level in the displacement time series, leading to a 16% improvement in accuracy, as validated against the Global Navigation Satellite System (GNSS). The study proposes the “overlapping effect” concept, which links InSAR tropospheric delay errors and OTL effects. It underscores the importance of accurate error assessment and removal. Neglecting this interaction may result in a 13% underestimation of the tropospheric error correction efficacy.

The utilization of satellite remote sensing images for retrieving aerosol optical parameters has been extensively discussed over the past few decades. While employing machine learning models is indeed a viable approach, a significant portion of these studies still rely on redundant data. Moreover, the discussion regarding aerosol absorption, a crucial factor for determining aerosol radiative impact and distinguishing aerosol components, is limited in current machine learning studies. In this study, we propose a random forest model to retrieve high-precision aerosol properties and their absorption over land from Himawari-8 geostationary satellite images. Remarkably, this model attains a high degree of accuracy in estimating aerosol optical depth (AOD), absorption aerosol optical depth (AAOD), and single scattering albedo (SSA) of heavy air mass using only seven primary predictors (observational radiances or their mathematical combinations, geometries, and wavelength). For AOD, the new random forest model demonstrates excellent performance on an hourly scale (R² ≥ 0.89, MAE < 0.07, RMSE <0.13), >80% of the samples fall within the expected error (EE) range. Concerning AAOD, the validation indicates that at least 65% of AAODs have a bias of ≤50%, with an R² exceeding 0.78, MAE ≤ 0.008 and RMSE ≤0.016. SSA also demonstrates a high accuracy (R² ≥ 0.57, MAE < 0.03, RMSE <0.05), with >70% of the results have an error ≤ 0.03. Through more comprehensive independent spatiotemporal cross validation, it can be determined that the model also offers reliable spatial and temporal predictions. The proposed RF model is capable of learning aerosol properties under most atmosphere scenarios, providing a reasonable conversion from predictors to AOD and AAOD/SSA under high aerosol loadings. The spatial patterns of these parameters suggest that the retrievals show considerable potential in capturing high aerosol loading in East Asia and biomass burning in Southeast Asia. The method introduced in this study offers a new approach to obtaining aerosol properties from geostationary satellite remote sensing, featuring a flexible process, simple inputs, high accuracy, and enhanced robustness. Additionally, it furnishes supplementary insights into aerosol absorption, presenting new possibilities in determining aerosol radiative impact and distinguishing aerosol components.

Reference data obtained by interpreters is a key component of sample-based estimation of area of land cover and land cover change. However, interpreters may disagree when assigning the reference class label for a given sample unit and this inconsistency between interpreters contributes to the overall uncertainty of the estimated area. Interpenetrating subsampling (IPS) offers a practical way to incorporate interpreter variability into an unbiased estimator of the total variance. This method requires partitioning the full sample into g nonoverlapping groups with the sample units in each group then evaluated by a different interpreter and each interpreter determines the reference class data for only one group. The total variance is estimated by the among group variability of the g estimates of area. IPS was applied to estimate the total variance of land cover area estimates for a sample of 300 pixels selected from the Puget Sound region of the Northwest United States. The reference land cover data were obtained by seven interpreters who each labeled all 300 pixels. These data provided a unique opportunity to explore properties of IPS such as variability over different random partitions of the sample into groups and variability over different subsets of interpreters. IPS estimates of total variance were produced for each land cover class for group sizes of g = 2 through g = 6 and all possible combinations of the seven interpreters for each group size. The estimated total variance decreased with increasing number of groups. Incorporating interpreter variance increased the estimated total variance by a factor ranging from 1.08 (agriculture) to 7.06 (grass/shrub) in simple random sampling. The total variance estimates varied substantially over the random partitions of the sample into groups, but this variability decreased as the group size increased. Compared with other total variance estimators, the IPS estimator is simpler to compute and is more cost effective because it does not require repeat interpretations

In addition to global radiation (R_g), direct radiation (R_dir) and diffuse radiation (R_dif) are important fundamental data urgently needed in scientific and industrial fields. However, compared with R_g, R_dir and R_dif have received little attention in the past, either in observations or in satellite retrievals, mainly due to the high cost of their observations and the difficulty of retrieving them effectively from satellites. In this study, a long-term global gridded dataset of R_dir and R_dif was constructed by separating from a high-precision satellite-based product of R_g using the Light Gradient Boosting Machine (LightGBM) model, trained with in-situ observations measured at the Baseline Surface Radiation Network (BSRN). The inputs to construct the dataset are the four variables of R_g, the cloud transmittance for R_g, the ratio of R_dif to R_g under clear sky condition (call the clear diffuse ratio), and the cosine of the solar zenith angle. The developed dataset was validated against in-situ observations and compared with other satellite-based products. Evaluations against the BSRN observations indicated that our proposed method has good generality and outperforms the machine learning-based direct estimation method of Hao et al. (2020). Independent validations were further performed against the observations measured at 17 China Meteorological Administration (CMA) radiation stations and the estimation based on sunshine duration observations at >2400 CMA routine meteorological stations, respectively. It was found that the accuracies of our estimates for both R_dir and R_dif were improved when upscaled to ≥ 30 km. Comparisons with three other satellite-based products indicate that our developed dataset of both R_dir and R_dif was generally more accurate than the global products of the Earth's Radiant Energy System (CERES) and Hao et al. (2020) based on the Deep Space Climate Observatory/Earth Polychromatic Imaging Camera (EPIC) (DSCOVER/EPIC) satellite, and the regional gridded product (JIEA) of Jiang et al. (2020a). The dataset developed in this study will contribute to ecological research and solar engineering applications.

Achieving the Global Climate Observing System goal of 10 m resolution leaf area index (LAI) maps is critical for applications related to climate adaptation, sustainable agriculture, and ecosystem monitoring. Five strategies for producing 10 m LAI maps from Sentinel-2 (S2) imagery are evaluated: i. bi-cubic interpolation of 20 m resolution S2 LAI maps from the Simplified Level 2 Prototype Processor Version 1 (SL2PV1) as currently performed by the Sentinel Applications Platform (SNAP), ii. applying SL2PV1 to S2 reflectance bands spatially downscaled to 10 m using bi-cubic interpolation (BICUBIC), iii. Applying SL2PV1 to S2 reflectance bands spatially downscaled to 10 m using Area to Point Regression Kriging (ATPRK), iv. using a recalibrated version of SL2PV1 (SL2PV2) requiring only three S2 10m bands, and iv) a novel use of the previously developed Active Learning Regularization (ALR) approach to locally approximate the SL2PV1 algorithm using only 10 m bands.

Algorithms were assessed in terms of per-pixel accuracy and spatial metrics when comparing 10 m LAI maps produced using either actual S2 imagery or S2 imagery synthesized from airborne hyperspectral imagery to reference 10 m LAI maps traceable to in-situ fiducial reference measurements at 10 sites across the continental US. ATPRK and ALR algorithms had the lowest precision error of ∼0.15 LAI, compared to 0.19 LAI for SNAP and BICUBIC and 0.35 LAI for SL2PV2, and ranked highest in terms of local correlation and Structural Similarity Index measure as well as qualitative agreement with reference maps. SL2PV2 LAI showed evidence of saturation over forests related to decreased sensitivity of input visible reflectance. All algorithms had a similar uncertainty of ∼0.55 LAI compared to traceable reference maps, due to the trade-off between bias and precision. However, ATPRK and ALR uncertainty reduced to 0.11 LAI and 0.16 LAI, respectively, when compared to reference maps that ignored canopy clumping. These results suggest that both ATPRK and ALR are suitable for producing 10 m S2 LAI maps assuming bias due to local clumping can be corrected in the underlying SL2PV1 algorithm.

Shoreline variations, triggered by climate change, eustatism, and tectonic, drive the coastal landscape evolution over multiple spatial and temporal scales. Among the many different existing coast types, sandy coasts are the most sensitive to coastal erosion and accretion processes and, at the same time, often host valuable anthropogenic assets. The rapid and ongoing evolution of these coastal environments poses challenges for their management, necessitating cost-effective and highly reliable methods for measuring these changes. Many remotely sensed shoreline extraction methods have been proposed in the literature, providing valuable tools for improving coastal management. Even if these methodologies allow the demarcation of the shoreline, its pixelated shape usually requires refinement through subsequent smoothing or vector generalization processes. It is important to note that the position of the thus extracted coastline is not a direct result of a measured physical quantity but rather a product of these refinement techniques. To address this problem, we developed a sub-pixel resolution method for extracting shorelines from remotely sensed images of sandy beaches, leveraging the radiometric signature of the shoreline. Validated through precise Global Navigation Satellite System field surveys for positioning the beach foreshore, this method was successfully applied to three beaches in Sicily, in the central Mediterranean, all exhibiting similar microtidal conditions. Its robust design allows for application across various satellite images, employing a straightforward radiometric interpolation method adaptable to different spatial resolutions. This method would be a valuable tool for coastal managers in detecting and mitigating coastal erosion and developing and maintaining anthropogenic coastal assets.