No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Flattening Gamma: Radiometric Terrain Correction for SAR Imagery
Abstract
Enabling intercomparison of synthetic aperture radar (SAR) imagery acquired from different sensors or acquisition modes requires accurate modeling of not only the geometry of each scene, but also of systematic influences on the radiometry of individual scenes. Terrain variations affect not only the position of a given point on the Earth's surface but also the brightness of the radar return as expressed in radar geometry. Without treatment, the hill-slope modulations of the radiometry threaten to overwhelm weaker thematic land cover induced backscatter differences, and comparison of backscatter from multiple satellites, modes, or tracks loses meaning. The ASAR & PALSAR sensors provide state vectors and timing with higher absolute accuracy than was previously available, allowing them to directly support accurate tie-point-free geolocation and radiometric normalization of their imagery. Given accurate knowledge of the acquisition geometry of a SAR image together with a digital height model (DHM) of the area imaged, radiometric image simulation is applied to estimate the local illuminated area for each point in the image. Ellipsoid-based or sigma naught (σ<sup>0</sup>) based incident angle approximations that fail to reproduce the effect of topographic variation in their sensor model are contrasted with a new method that integrates terrain variations with the concept of gamma naught (γ<sup>0</sup>) backscatter, converting directly from beta naught (β<sup>0</sup>) to a newly introduced terrain-flattened γ<sup>0</sup> normalization convention. The interpretability of imagery treated in this manner is improved in comparison to processing based on conventional ellipsoid or local incident angle based σ<sup>0</sup> normalization.
... It is important to note that although the geometry of the backscatter estimate was corrected in GRD products, the radiometry of the resulting image remained an ellipsoid model based on σ 0 [39]. ...
... Small [39] argues that such angle-based normalizations are flawed in that the attendant sensor model fails to account for numerous important properties of the radar backscatter in regions with large topographic variation. The backscatter coefficient provides a backscatter ratio estimate per given reference area. ...
Saline wetlands experience large temporal fluctuations in water supply during the year and are recharged only or mainly through precipitation, meaning they are vulnerable to climate-change-induced aridification. Most passive satellite sensors are unsuitable for continuous wetland monitoring due to cloud cover and their relatively low temporal resolution. However, active satellite sensors such as the C-band synthetic aperture radar of Sentinel-1 satellites offer free, cloud-independent data. We examined surface water cover changes from October 2014 to November 2018 in the strictly protected area (13,000 ha) of the Upper-Kiskunság Alkaline Lakes region in the Danube–Tisza Interfluve in Hungary, with the aim of helping with nature protection planning. Changes and sensitivity can be defined based on the knowledge of variability. We developed a method for water cover detection based on automatic classification, applying the so-called WEKA K-Means clustering algorithm. For satellite data processing and analysis, we used the Google Earth Engine cloud processing platform. In terms of validation, we compared our results with the multispectral Modified Normalized Difference Water Index (MNDWI) derived from Landsat 8 and Sentinel-2 top-of-atmosphere reflectance images using a threshold-based binary classifier (receiver operator characteristics) for the MNDWI data. Using two completely distinct methods operating in distinct wavelength ranges, we obtained adequately matching results, with Spearman’s correlation coefficients (ρ) ranging from 0.54 to 0.80.
... This slope correction is employed to lessen the effects of surface scattering on the backscatter signal, which is a significant issue in mountainous and complex terrain, such as that of the Brooks Range. Radiometric distortions over rugged terrain within SAR backscatter products originate from the side-looking SAR imaging geometry and are strong enough to exceed weaker differences of the signal due to variation in land cover (Small, 2011). The S1 SAR data used in this study are from the Ground Range Detected (GRD) collection in GEE, which are pre-processed by applying an orbit file, removing thermal noise, removing GRD border noise, applying a radiometric calibration to σ0, and applying a range-doppler terrain correction . ...
... The S1 SAR slope correction was employed to lessen the effects of surface scattering on the backscatter signal, which is a significant issue in mountainous and complex terrain, such as that of the Brooks Range. Radiometric distortions over rugged terrain within SAR backscatter products originate from the side-looking SAR imaging geometry and are strong enough to exceed weaker differences of the signal due to variation in land cover (Small, 2011). The S1 SAR data used in this study are from the Ground Range Detected (GRD) collection in GEE, which are preprocessed by applying an orbit file, removing thermal noise, removing GRD border noise, applying a radiometric calibration to σ0, and applying a range-doppler terrain correction . ...
Perennial snowfields, such as those found in the Brooks Range of Alaska, are a critical component of the cryosphere. They serve as habitat for an array of wildlife, some of which are crucial for rural subsistence hunters. Snowfields also influence hydrology, vegetation, permafrost, and have the potential to preserve valuable archaeological artifacts. In this study, perennial snowfield extents in the Brooks Range are derived from satellite remote sensing, field acquired data, and snowmelt modeling. The remote sensing data are used to map and quantify snow cover area changes across multiple temporal scales, spatial resolutions, and geographic sub-domains. Perennial snowfield classification techniques were developed using optical multi-spectral imagery from NASA Landsat and European Space Agency Sentinel-2 satellites. A Synthetic Aperture Radar change detection algorithm was also developed to quantify snow cover area using Sentinel-1 data. Results of the remote sensing analyses were compared to helicopter and manually collected field data. Also, a snowfield melt model was developed using an adaptation of the temperature index method to determine probability of melt via binary logistic regression in two dimensions. The logistic temperature melt model considers summer season snow cover area changes per pixel in remotely sensed products and relationships to several independent variables, including elevation-lapse-adjusted air temperature and terrain-adjusted solar radiation. Evaluations of the Synthetic Aperture Radar change detection algorithm via comparison with results from optical imagery analysis, as well as via comparison with field acquired data, indicate that the radar algorithm performs best in small, focused geographic sub-domains. The multi-spectral approach appears to perform similarly well within multiple geographic domain sizes. This may be the result of synthetic aperture radar algorithm dependency on backscatter thresholding techniques and slope corrections in mountainous complex topography. Results indicate that perennial snowfield extents in the Brooks Range are decreasing over decadal time scales, with short-lived, interannual and seasonal increases. Results also show that perennial snowfields are more persistent at higher elevations over time with notable consistency in at least one of the Brooks Range sub-domains of this study, Gates of the Arctic National Park and Preserve. Climate change may be altering the distribution, elevation, melt behavior, and overall extents of the Brooks Range perennial snowfields. Such changes could have significant implications for hydrology, wildlife, vegetation, and subsistence hunting in rural Alaska.
... After importing the original data (Sentinel-1 GRDH IW Level 1 data), a spatial s of the study area extent is processed using suitable corner coordinates. In the nex steps, the data gets radiometrically corrected to gamma naught [52]. This alternat the common calibration to sigma naught [53] additionally takes the effect of topogr variation into account, which is very important regarding the partially rugged terr the study area and forested areas in the German mountain ranges in general. ...
... The co eration of terrain effects in the radiometric calibration is essential for further themati cessing and analysis, i.e., the calculation of a consistent radar drought index ove After importing the original data (Sentinel-1 GRDH IW Level 1 data), a spatial subset of the study area extent is processed using suitable corner coordinates. In the next two steps, the data gets radiometrically corrected to gamma naught [52]. This alternative to the common calibration to sigma naught [53] additionally takes the effect of topographic variation into account, which is very important regarding the partially rugged terrain in the study area and forested areas in the German mountain ranges in general. ...
A timely and spatially high-resolution detection of drought-affected forest stands is important to assess and deal with the increasing risk of forest fires. In this paper, we present how multitemporal Sentinel-1 synthetic aperture radar (SAR) data can be used to detect drought-affected and fire-endangered forest stands in a spatially and temporally high resolution. Existing approaches for Sentinel-1 based drought detection currently do not allow to deal simultaneously with all disturbing influences of signal noise, topography and visibility geometry on the radar signal or do not produce pixel-based high-resolution drought detection maps of forest stands. Using a novel Sentinel-1 Radar Drought Index (RDI) based on temporal and spatial averaging strategies for speckle noise reduction, we present an efficient methodology to create a spatially explicit detection map of drought-affected forest stands for the year 2020 at the Donnersberg study area in Rhineland-Palatinate, Germany, keeping the Sentinel-1 maximum spatial resolution of 10 m × 10 m. The RDI showed significant (p < 0.05) drought influence for south, south-west and west-oriented slopes. Comparable spatial patterns of drought-affected forest stands are shown for the years 2018, 2019 and with a weaker intensity for 2021. In addition, the assessment for summer 2020 could also be reproduced with weekly repetition, but spatially coarser resolution and some limitations in the quality of the resulting maps. Nevertheless, the mean RDI values of temporally high-resolution drought detection maps are highly correlated (R2 = 0.9678) with the increasing monthly mean temperatures in 2020. In summary, this study demonstrates that Sentinel-1 data can play an important role for the timely detection of drought-affected and fire-prone forest areas, since availability of observations does not depend on cloud cover or time of day.
... The amplitude images are multi-looked by a factor 2 in azimuth and in range, to reduce speckle, leading to a roughly 28 × 5 m slant range resolution. Radiometric terrain correction is applied to account for the local incidence angle variating with slope angle resulting in amplitude values that are independent of slope angle (Small, 2011). ...
... Nevertheless, this trend must be considered with a certain caution: (1) the trend is dependent on the quality of terrain correction during the pre-processing step (Sect. 3.1), which should make SAR values independent of slope angle (Small, 2011); (2) a changing slope angle could influence the GH size (Chen et al., 2016); and (3) we take the average slope angle per GH. Elongated GH features (mainly the flash flood features in the GH inventories) will have an average slope angle that is not representative of every part of the GH. ...
Landslides and flash floods are geomorphic hazards (GHs) that often co-occur and interact. They generally occur very quickly, leading to catastrophic socioeconomic impacts. Understanding the temporal patterns of occurrence of GH events is essential for hazard assessment, early warning, and disaster risk reduction strategies. However, temporal information is often poorly constrained, especially in frequently cloud-covered tropical regions, where optical-based satellite data are insufficient. Here we present a regionally applicable methodology to accurately estimate GH event timing that requires no prior knowledge of the GH event timing, using synthetic aperture radar (SAR) remote sensing. SAR can penetrate through clouds and therefore provides an ideal tool for constraining GH event timing. We use the open-access Copernicus Sentinel-1 (S1) SAR satellite that provides global coverage, high spatial resolution (∼10–15 m), and a high repeat time (6–12 d) from 2016 to 2020. We investigate the amplitude, detrended amplitude, spatial amplitude correlation, coherence, and detrended coherence time series in their suitability to constrain GH event timing. We apply the methodology on four recent large GH events located in Uganda, Rwanda, Burundi, and the Democratic Republic of the Congo (DRC) containing a total of about 2500 manually mapped landslides and flash flood features located in several contrasting landscape types. The amplitude and detrended amplitude time series in our methodology do not prove to be effective in accurate GH event timing estimation, with estimated timing accuracies ranging from a 13 to 1000 d difference. A clear increase in accuracy is obtained from spatial amplitude correlation (SAC) with estimated timing accuracies ranging from a 1 to 85 d difference. However, the most accurate results are achieved with coherence and detrended coherence with estimated timing accuracies ranging from a 1 to 47 d difference. The amplitude time series reflect the influence of seasonal dynamics, which cause the timing estimations to be further away from the actual GH event occurrence compared to the other data products. Timing estimations are generally closer to the actual GH event occurrence for GH events within homogenous densely vegetated landscape and further for GH events within complex cultivated heterogenous landscapes. We believe that the complexity of the different contrasting landscapes we study is an added value for the transferability of the methodology, and together with the open-access and global coverage of S1 data it has the potential to be widely applicable.
... GRD products were preprocessed using the Sentinel-1 Toolbox (S1TBX; ESA, 2022c). The processing chain included (i) radiometric calibration to radar brightness, (ii) multi-looking to the nominal Sentinel-1 resolution (20 m square pixels), (iii) terrain-flattening correction (Small, 2011) for removing topographic effects, and (iv) orthorectification using the range Doppler method (Small & Schubert, 2008). Finally, gamma naught (γ 0 ) backscatter coefficients of VV and VH polarizations were log-transformed to dB units. ...
... GRD products were preprocessed using the Sentinel-1 Toolbox (S1TBX; ESA, 2022c). The processing chain included (i) radiometric calibration to radar brightness, (ii) multi-looking to the nominal Sentinel-1 resolution (20 m square pixels), (iii) terrain-flattening correction (Small, 2011) for removing topographic effects, and (iv) orthorectification using the range Doppler method (Small & Schubert, 2008). Finally, gamma naught (γ 0 ) backscatter coefficients of VV and VH polarizations were log-transformed to dB units. ...
Abstract The structural complexity of plant communities contributes to maintaining the ecosystem functioning in fire‐prone landscapes and plays a crucial role in driving ecological resilience to fire. The objective of this study was to evaluate the resilience to fire off several plant communities with reference to the temporal evolution of their vertical structural diversity (VSD) estimated from the data fusion of C‐band synthetic aperture radar (SAR) backscatter (Sentinel‐1) and multispectral remote sensing reflectance (Sentinel‐2) in a burned landscape of the western Mediterranean Basin. We estimated VSD in the field 1 and 2 years after fire using Shannon's index as a measure of vertical heterogeneity in vegetation structure from the vegetation cover in several strata, both in burned and unburned control plots. Random forest (RF) was used to model VSD in the control (analogous to prefire scenario) and burned plots (1 year after fire) using as predictors (i) Sentinel‐1 VV and VH backscatter coefficients and (ii) surface reflectance of Sentinel‐2 bands. The transferability of the RF model from 1 to 2 years after wildfire was also evaluated. We generated VSD prediction maps across the study site for the prefire scenario and 1 to 4 years postfire. RF models accurately explained VSD in unburned control plots (R2 = 87.68; RMSE = 0.16) and burned plots 1 year after fire (R2 = 80.48; RMSE = 0.13). RF model transferability only involved a reduction in the VSD predictive capacity from 0.13 to 0.20 in terms of RMSE. The VSD of each plant community 4 years after the fire disturbance was significantly lower than in the prefire scenario. Plant communities dominated by resprouter species featured significantly higher VSD recovery values than communities dominated by facultative or obligate seeders. Our results support the applicability of SAR and multispectral data fusion for monitoring VSD as a generalizable resilience indicator in fire‐prone landscapes.
... SAR imagery is used in a wide range of applications that make use of the SAR normalized radar cross section (NRCS) that is represented in terms of a backscatter coefficient -an estimate of the backscatter per given reference area. Depending on which reference area convention is chosen for normalization, one can distinguish between three representations of the backscatter coefficient known as: i) beta naught (β 0 ), ii) sigma naught (σ 0 ) or iii) gamma naught (γ 0 ) (Small 2011). The σ 0 and γ 0 coefficients are normally calculated using an ellipsoid for the reference area computation and in this case, they have an important limitation, namely the fact that their radiometric properties are heavily distorted by topographic variations, even in only slightly undulating terrain. ...
... To overcome this limitation, several methods for the radiometric normalization of the backscatter coefficient were introduced in the literature. These include methods based on the local incidence angle (LIA) (Ulander 1996;Kellndorfer et al. 1998) or the actual ground area visible to the radar which is known as radiometric terrain flattening (RTF) (Small 2011). In recent years, the RTF became widely used, especially in snow or ice melt mapping (Scharien et al. 2017;Jewell et al. 2020) or forest monitoring (Rüetschi, Schaepman, and Small 2018;Akbari and Solberg 2020). ...
Terrain-induced variations of radar backscatter represent an important limiting factor of many Synthetic Aperture Radar (SAR)-based applications. Radiometric terrain flattening (RTF) is a well-established method that minimizes these variations in SAR imagery. To fully understand the implications of SAR RTF, validation of its impact on the derived products is needed. In this study, we quantified the influence of the RTF on a forest mapping and classification algorithm over Austria, and compared the classification results for the conventional sigma naught and radiometrically terrain-corrected gamma backscatter. The overall accuracy for forest/non-forest mapping and forest type classification improved by 2% and 4%, respectively, over the whole of Austria, with improvements of up to 16% and 20%, respectively, in regions with strong topography.
... The sigma naughts in HH and VV, σ • HH , σ • VV , are the normalized radar cross section, which are directly related to the power of the ground targets returned to SAR satellite antenna [40]. σ • HH and σ • VV can be theoretically defined as ...
Extracting meaningful attributes of radar scatterers from SAR images, PAZ in our case, facilitates a better understanding of SAR data and physical interpretation of deformation processes. The attribute categories and attribute extraction method are not yet thoroughly investigated. Therefore, this study recognizes three attribute categories: geometric, physical, and land-use attributes, and aims to design a new scheme to extract these attributes of every coherent radar scatterer. Specifically, we propose to obtain geometric information and its dynamics over time of the radar scatterers using time series InSAR (interferometric SAR) techniques, with SAR images in HH and VV separately. As all InSAR observations are relative in time and space, we convert the radar scatterers in HH and VV to a common reference system by applying a spatial reference alignment method. Regarding the physical attributes of the radar scatterers, we first employ a Random Forest classification method to categorize scatterers in terms of scattering mechanisms (including surface, low-, high-volume, and double bounce scattering), and then assign the scattering mechanism to every radar scatterer. We propose using a land-use product (i.e., TOP10NL data for our case) to create reliable labeled samples for training and validation. In addition, the radar scatterers can inherit land-use attributes from the TOP10NL data. We demonstrate this new scheme with 30 Spanish PAZ SAR images in HH and VV acquired between 2019 and 2021, covering an area in the province of Friesland, the Netherlands, and analyze the extracted attributes for data and deformation interpretation.
... Subsequently, the values were converted from unsigned 16-bit digital numbers (DN) values into gamma-naught (γ 0 ) backscatter values (unit: decibel) because less likely dependent on the incidence angle of the radar beam. This solution was adopted to basically improve the incorporation of terrain variations (Small;. The conversion was applied following the formula proposed by Rosenqvist et al. (Rosenqvist et al., 2007), for which DN is the digital number and CF is a calibration factor equal to "-83" for PALSAR/PALSAR2 data: 0 ( ) = 10 log 10 (DN) 2 + CF The NASADEM digital elevation model was also collected, and derives from an interpolation of ASTER GDEM, ICESat, and PRISM data (Crippen et al., 2016). ...
Roads are a major threat to wildlife due to induced mortality and restrictions to animal movement. A central issue in conservation biology is the accurate site identification for the implementation of multispecies mitigation measures, on roads. Those measures entail high costs and methodological challenges and their efficiency highly depend on the right location. The aim of this PhD is to inform, through remote sensing and connectivity modelling, how to increase the efficiency of planning mitigation measures to reduce roadkill and promote connectivity; and demonstrate the usefulness of remote sensing in defining suitable areas for the conservation of an endangered species that often occurs in the vicinity of roads. To do so, we first assessed whether occurrence-based strategies were able to infer functional connectivity, compared to those more complex and financially demanding based on telemetry, with respect to daily and dispersal movements. Secondly, we assessed whether remote sensing data were sufficiently informative to identify key habitats for a threatened species around road verges. Thirdly, we assessed the predictive and prioritisation ability of road mitigation units intercepting multispecies corridors to prevent vulnerability to roadkill. Findings revealed that simple models are suitable as complex ones for both daily and dispersal movements, allowing for costly-effective connectivity assessments. Results demonstrated the ability of free remote sensing data to identify microhabitat conditions in verges and surrounding landscape, for a threatened rodent, allowing for the delimitation of refugee areas and definition of monitoring strategies for the species. Undemanding data (occurrence and remote sensing) were able to describe species-specific ecological requirements for birds, bats and non-flying mammals as well as roadkill patterns, possibly due to similar overlapping corridors and habitats, despite some mismatches that occurred for highly mobile species. This framework ensured high efficiency in prioritisation of multispecies roadkill mitigation planning, resilient to long-term landscape dynamics.
... By contrast, scene-dependent radiometric distortions arise from the inherent 2D imaging process of a 3D irregular ground shape, and they are inherently local in nature [2][3][4][5][6][7][8][9][10][11]. Accordingly, the 2D imaging process of an irregular relief intrinsically implies unavoidable radiometric distortions. ...
The modeling and simulation of topography-induced imaging distortions are crucial for consistent radiometric information exploitation in current and forthcoming SAR-based Earth observation missions with a high spatial and temporal resolution, with relevance in several applications. In this paper, for the first time, we specifically investigate the compensation of topography-induced radiometric distortions affecting SAR images acquired by the L-band Argentinian satellite SAOCOM. We adopt a recently developed calibration method relying on an analytical formulation derived in the rigorous framework of the differential geometry of surfaces. We first provide an original interpretation of the analytical formulation, thus providing further insights into the relevant area-stretching-based formalism. Then, the numerical implementation of the method is specialized to systematically process the data acquired by SAOCOM sensors; hence, the resulting sensor-specific prototype solution processor is employed in this study. Finally, experiments performed over a real scenario in the southern part of Italy, characterized by large topography variations, are presented and discussed, thus elucidating the effectiveness of the adopted method applied to SAOCOM images. The adopted effective SAR calibration strategy opens up the way to its operational use in large-scale SAOCOM data processing.
... The images were then converted from ground range geometry to the backscatter coefficient sigma naught (σ 0 ), measured in decibels (dB), using a digital elevation model (DEM) from the Shuttle Radar Topography Mission (SRTM). The radiometry of the imagery remained an ellipsoid model based on σ 0 (Small, 2011). The GRD data is delivered as log-scaled (i.e., 10*log10 (x)). ...
Frequent observations of surface water at fine spatial scales will provide critical data to support the management of aquatic habitat, flood risk and water quality. Sentinel-1 and Sentinel-2 satellites can provide such observations, but algorithms are still needed that perform well across diverse climate and vegetation conditions. We developed surface inundation algorithms for Sentinel-1 and Sentinel-2, respectively, at 12 sites across the conterminous United States (CONUS), covering a total of >536,000 km2 and representing diverse hydrologic and vegetation landscapes. Each scene in the 5-year (2017-2021) time series was classified into open water, vegetated water, and non-water at 20 m resolution using variables from Sentinel-1 and Sentinel-2, as well as variables derived from topographic and weather datasets. The Sentinel-1 algorithm was developed distinct from the Sentinel-2 model to explore if and where the two time series could potentially be integrated into a single high-frequency time series. Within each model, open water and vegetated water (vegetated palustrine, lacustrine, and riverine wetlands) classes were mapped. The models were validated using imagery from WorldView and PlanetScope. Classification accuracy for open water was high across the 5-year period, with an omission and commission error of only 3.1% and 0.9% for the Sentinel-1 algorithm and 3.1% and 0.5% for the Sentinel-2 algorithm, respectively. Vegetated water accuracy was lower, as expected given that the class represents mixed pixels. The Sentinel-2 algorithm showed higher accuracy (10.7% omission and 7.9% commission error) relative to the Sentinel-1 algorithm (28.4% omission and 16.0% commission error). Patterns over time in the proportion of area mapped as open or vegetated water by the Sentinel-1 and Sentinel-2 algorithms were charted and correlated for a subset of all 12 sites. Our results showed that the Sentinel-1 and Sentinel-2 algorithm open water time series can be integrated at all 12 sites to improve the temporal resolution, but sensor-specific differences, such as sensitivity to vegetation structure versus pixel color, complicate the data integration for mixed-pixel, vegetated water. The methods developed here provide inundation at 5-day (Sentinel-2 algorithm) and 12-day (Sentinel-1 algorithm) time steps to improve our understanding of the short- and long-term response of surface water to climate and land use drivers in different ecoregions.
... A rolling mean operand was, therefore, used here for simplicity. Using default parameters, the radiometric data were normalized to radar brightness β 0 and terrain-flattened γ 0 T [66]. With the polarimetric data, the eigenproblem for [C 2 ] was solved as described by Cloude [67] using custom implementations with python packages 'numpy' [68] and 'numba' [69]. ...
Understanding forest decline under drought pressure is receiving research attention due to the increasing frequency of large-scale heat waves and massive tree mortality events. However, since assessing mortality on the ground is challenging and costly, this study explores the capability of satellite-borne Copernicus Sentinel-1 (S-1) C-band radar data for monitoring drought-induced tree canopy damage. As droughts cause water deficits in trees and eventually lead to early foliage loss, the S-1 radiometric signal and polarimetric indices are tested regarding their sensitivities to these effects, exemplified in a deciduous broadleaf forest. Due to the scattered nature of mortality in the study site, we employed a temporal-only time series filtering scheme that provides very high spatial resolution (10 m ×10 m) for measuring at the scale of single trees. Finally, the anomaly between heavily damaged and non-damaged tree canopy samples (n = 146 per class) was used to quantify the level of damage. With a maximum anomaly of −0.50 dB ± 1.38 for S-1 Span (VV+VH), a significant decline in hydrostructural scattering (moisture and geometry of scatterers as seen by SAR) was found in the second year after drought onset. By contrast, S-1 polarimetric indices (cross-ratio, RVI, Hα) showed limited capability in detecting drought effects. From our time series evaluation, we infer that damaged canopies exhibit both lower leaf-on and leaf-off backscatters compared to unaffected canopies. We further introduce an NDVI/Span hysteresis showing a lagged signal anomaly of Span behind NDVI (by ca. one year). This time-lagged correlation implies that SAR is able to add complementary information to optical remote sensing data for detecting drought damage due to its sensitivity to physiological and hydraulic tree canopy damage. Our study lays out the promising potential of SAR remote sensing information for drought impact assessment in deciduous broadleaf forests.
... We first applied absolute radiometric calibration to level 1.1 SAR products [42]. To reduce speckle noise, a moving average filter was then applied to each matrix element using a window size of 2 × 10 (range and azimuth direction, respectively). ...
The biodiversity loss in Southeast Asia indicates an urgent need for long-term monitoring, which is lacking. Much attention is being directed toward bird diversity monitoring using remote sensing, based on relation to forest structure. However, few studies have utilized space-borne active microwave remote sensing, which has considerable advantages in terms of repetitive observations over tropical areas. Here, we evaluate threatened bird occurrence from L-band satellite data explaining forest structure in Sumatra, Indonesia. First, we identified L-band parameters with strong correlations with the forest layer structure, defined as forest floor, understory, and canopy layers. Then, we analyzed the correlation between threatened bird occurrence and L-band parameters identified as explaining forest structure. The results reveal that several parameters can represent the layers of forest floor, understory, and canopy. Subsequent statistical analysis elucidated that forest-dependent and threatened bird species exhibit significant positive correlations with the selected L-band parameters explaining forest floor and understory. Our results highlight the potential of applying microwave satellite remote sensing to evaluate bird diversity through forest structure estimation, although a more comprehensive study is needed to strengthen our findings.
... The flow of the preprocessing operations performed is similar to the one suggested by Filipponi (2019). The terrain radiometric correction (Small, 2011) and the geometric correction were applied by using the Digital Elevation Model at 5 m resolution provided from the Spanish National Centre of Geographic Information. ...
A preliminary analysis based on the application of a change detection method for remote sensed soil moisture retrieval at high resolution is presented. Sentinel-1 SAR images are used for studying agricultural areas in Spain, where in situ soil moisture data are available through the International Soil Moisture Network. The total backscattered SAR signal is modelled as the sum of vegetation and soil contributions. At first, the relationship between soil moisture and the co-polarized band of Sentinel-1 was analyzed for all the measurement stations of the area, and the ones with stronger relation were selected. Time series analyses were then conducted at the field scale for studying the interactions between some SAR parameters and the in situ data. The two polarizations and the polarization ratio were analyzed with respect to in situ soil moisture observations and precipitation data in order to identify homogeneous time domains in which the method can be applied in a consistent manner. Analyses show that the main driver of wide range SAR signal variations is the presence of precipitation events. Moreover, SAR coherence and polarization rate manifest specific behaviors that can be exploited either for deepening the knowledge on the role of model parameters and identifying suitable time and space extends in which operate separate estimations of vegetation, soil moisture and soil roughness parameters. Identification and isolation of precipitation driven patterns, as long as the selection of homogeneous time spans and space regions is the basis for improving the capability of satellite based soil moisture retrieval models.
... Surface scatterers are the main variables that determine the backscatter signal of the ground surface. Slopes facing towards the satellite will appear brighter than ones facing away (due to the effect of differences in local incidence angle), although this contribution can be corrected through radiometric terrain corrections if accurate DEMs are available (Small, 2011;Meyer et al., 2015). The roughness of a surface is dependent on instrument wavelength. ...
The extrusion rate of a lava dome is a critical parameter for monitoring silicic eruptions and forecasting their development. Satellite radar backscatter can provide unique information about dome growth during a volcanic eruption when other datasets (e.g., optical, thermal, ground-based measurements, etc.) may be limited. Here, we present an approach for estimating volcanic topography from individual backscatter images. Using data from multiple SAR sensors we apply the method to the dome growth during the 2021 eruption at La Soufrière, St. Vincent. We measure an average extrusion rate of 1.8 m3s−1 between December 2020 and March 2021 before an acceleration in extrusion rate to 17.5 m3s−1 in the 2 days prior to the explosive eruption on 9 April 2021. We estimate a final dome volume of 19.4 million m3, extrapolated from the SAR sensors, with approximately 15% of the total extruded volume emplaced in the last 2 days. A possible explanation for the acceleration in extrusion rate could be the combined emptying of a conduit and reservoir of older material before the ascent of gas-rich magma in April 2021.
... The local incident angle and topography variation influence the radar-reflected signal [26]. The terrain variations affect the brightness of the radar's received energy which can reduce the accuracy of land cover classification [29]. The radiometric calibration implemented here was proposed by [30]: where I is the intensity, DN is the digital number of the image, and K is the calibration constant. ...
Flood events have become intense and more frequent due to heavy rainfall and hurricanes caused by global warming. Accurate floodwater extent maps are essential information sources for emergency management agencies and flood relief programs to direct their resources to the most affected areas. Synthetic Aperture Radar (SAR) data are superior to optical data for floodwater mapping, especially in vegetated areas and in forests that are adjacent to urban areas and critical infrastructures. Investigating floodwater mapping with various available SAR sensors and comparing their performance allows the identification of suitable SAR sensors that can be used to map inundated areas in different land covers, such as forests and vegetated areas. In this study, we investigated the performance of polarization configurations for flood boundary delineation in vegetated and open areas derived from Sentinel1b, C-band, and Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) L-band data collected during flood events resulting from Hurricane Florence in the eastern area of North Carolina. The datasets from the sensors for the flooding event collected on the same day and same study area were processed and classified for five landcover classes using a machine learning method—the Random Forest classification algorithm. We compared the classification results of linear, dual, and full polarizations of the SAR datasets. The L-band fully polarized data classification achieved the highest accuracy for flood mapping as the decomposition of fully polarized SAR data allows land cover features to be identified based on their scattering mechanisms.
... In mountainous regions the topography can affect the microwave signal (Davenport et al., 2008;Mialon et al., 2008;Naeimi et al., 2009) and this led to uncertainties in drought monitoring in several studies (Mishra et al., 2017;Paredes-Trejo and Barbosa, 2017). Potential ways to mitigate the effect of topography on the SAR backscatter signal is to use gamma • (Small, 2011;Navacchi et al., 2022). ...
Agricultural droughts are extreme events which are often a result of interplays between multiple hydro-meteorological processes. Therefore, assessing drought occurrence, extent, duration and intensity is complex and requires the combined use of multiple variables, such as temperature, rainfall, soil moisture (SM) and vegetation state. The benefit of using information on SM and vegetation state is that they integrate information on precipitation, temperature and evapotranspiration, making them direct indicators of plant available water and vegetation productivity. Microwave remote sensing enables the retrieval of both SM and vegetation information, and satellite-based SM and vegetation products are available operationally and free of charge on a regional or global scale and daily basis. As a result, microwave remote sensing products play an increasingly important role in drought monitoring applications. Here, we provide an overview of recent developments in using microwave remote sensing for large-scale agricultural drought monitoring. We focus on the intricacy of monitoring the complex process of drought development using multiple variables. First, we give a brief introduction on fundamental concepts of microwave remote sensing together with an overview of recent research, development and applications of drought indicators derived from microwave-based satellite SM and vegetation observations. This is followed by a more detailed overview of the current research gaps and challenges in combining microwave-based SM and vegetation measurements with hydro-meteorological data sets. The potential of using microwave remote sensing for drought monitoring is demonstrated through a case study over Senegal using multiple satellite- and model-based data sets on rainfall, SM, vegetation and combinations thereof. The case study demonstrates the added-value of microwave-based SM and vegetation observations for drought monitoring applications. Finally, we provide an outlook on potential developments and opportunities.
... Multilooking with a factor of 2 × 2 (range × azimuth) was performed before orthorectification. Radiometric normalisation with respect to the projected area of the scattering element was performed to eliminate the topography-induced radiometric variation [40]. In this way, a time series of coregistered "gamma-naught" backscatter images was formed, with a pixel size of 20 m by 20 m. ...
Time series of SAR imagery combined with reference ground data can be suitable for producing forest inventories. Copernicus Sentinel-1 imagery is particularly interesting for forest mapping because of its free availability to data users; however, temporal dependencies within SAR time series that can potentially improve mapping accuracy are rarely explored. In this study, we introduce a novel semi-supervised Long Short-Term Memory (LSTM) model, CrsHelix-LSTM, and demonstrate its utility for predicting forest tree height using time series of Sentinel-1 images. The model brings three important modifications to the conventional LSTM model. Firstly, it uses a Helix-Elapse (HE) projection to capture the relationship between forest temporal patterns and Sentinel-1 time series, when time intervals between datatakes are irregular. A skip-link based LSTM block is introduced and a novel backbone network, Helix-LSTM, is proposed to retrieve temporal features at different receptive scales. Finally, a novel semisupervised strategy, Cross-Pseudo Regression, is employed to achieve better model performance when reference training data are limited. CrsHelix-LSTM model is demonstrated over a representative boreal forest site located in Central Finland. A time series of 96 Sentinel-1 images are used in the study. The developed model is compared with basic LSTM model, attention-based bidirectional LSTM and several other established regression approaches used in forest variable mapping, demonstrating consistent improvement of forest height prediction accuracy. At best, the achieved accuracy of forest height mapping was 28.3% relative root mean squared error (rRMSE) for pixel-level predictions and 18.0% rRMSE on stand level. We expect that the developed model can also be used for modeling relationships between other forest variables and satellite image time series.
... The S1 pre-processing workflow commenced by applying precise orbit information and removing thermal noise. This workflow involved radiometric calibration (to beta, β0, nought backscatter standard conventions) (Small 2011) and radiometric terrain correction (RTC) processes, through which the images were radiometrically flattened and geometrically corrected using the shuttle radar topography mission (SRTM) 1 arc-second (Farr et al. 2007) digital elevation model (DEM). All S1 images were mosaicked as a function of the acquisition date, resulting in a final dataset of 15 large image mosaics covering the entire study area, and stacked, relying on the geolocation product to initialize the offset method. ...
Wildland fires are among the main factors affecting the surrounding territory in terms of ecological and socioeconomic changes at different temporal and spatial scales. In the Mediterranean environment, although fire can positively influence some bio-physical dynamics of habitats, it acts as a pressing disturbance on ecosystems when the severity, spatial scale, and/or frequency are high, thereby determining their degradation. Therefore, knowing and mapping the accurate quantitative spatial distribution of all areas affected by fire during an entire high-frequency fire season and on a relatively large scale (regional/national scale) is an essential step to initialize the numerous subsequent effect monitoring analyses that can be carried out. This work proposes a reliable and open-access workflow to map burned areas on regional and national scales during the entire fire season. To achieve this, we integrated optical (Sentinel-2, S2) and Synthetic Aperture Radar (SAR; Sentinel-1, S1) free high spatial and temporal resolution data into a multitemporal composite criterion. Open-source software and Python-based libraries were used to develop the workflow. In particular, the second-lowest near infra-red (NIR) image composite (secMinNIR) criterion, based on the retrieval of the second minimum values that the NIR values reached in each pixel during the entire time frame considered, was applied to cloud-free S2 imagery to optimize the separability between burned and unburned areas. Subsequently, a second temporal composite criterion was developed and applied to the S1 time series, relying on the SAR capacity to detect vegetation fire-induced structural and humidity changes. It was based on retrieving the S1 pixel value of the first next (or the same) date to the corresponding date of the pixel value previously found by secMinNIR. The burned area map was created using an object-based geographic analysis (GEOBIA) process, using two optical and SAR composite images as input layers. The large-scale mean-shift (LSMS) algorithm was employed to segment the image, while the random forest (RF) classifier was the machine-learning model used to perform supervised classification. GEOBIA-based burned area classification was also performed using only the optical composite. The resulting accuracy values were compared using the precision (p), recall (r), and F-score accuracy metrics. The classification achieved high accuracy levels (F-score value greater than 0.9) in both cases (S1+ S2, 0.956; S2, 0.914), highlighting the increased effectiveness of this approach in detecting burned areas, heterogeneous in terms of amplitude, and affected site-specific characteristics that occurred during the fire season. Although the use of only optical data is sufficient to map the fire-affected areas early, some commission errors, represented by small regions scattered over the entire study area, remain, proving that the integration of SAR data improves the quality of the obtained results.
... Multilooking with factor of 2 × 2 (range × azimuth) was performed before the orthorectification. Radiometric normalization with respect to the projected area of the scattering element was performed to eliminate the topography-induced radiometric variation [32]. This way, a time series of co-registered "gamma-naught" backscatter images were formed, with a pixel size of 20 m by 20 m. ...
Here, a novel semi-supervised Long Short-Term Memory (LSTM) model is developed and demonstrated for predicting forest tree height using time series of Sentinel-1 images. The model uses a Helix-Elapse (HE) projection approach to capture relationship between forest temporal patterns and Sentinel-1 time series, when the acquisition time intervals are irregular. A skip-link based LSTM block is introduced and a novel backbone network, Helix-LSTM, is proposed to retrieve temporal features at different receptive scales. Additionally, a novel semi-supervised strategy, Cross-Pseudo Regression, is employed to achieve better model performance. The developed model is compared versus basic LSTM model, attention-based bidirectional LSTM and several other established regression approaches used in forest variable mapping, demonstrating consistent improvement of forest height prediction accuracy. The study site is located in Central Finland and represents boreal forestland. At best, the achieved accuracy of forest height mapping was 28.3% rRMSE for pixel-level predictions, and 18.0% rRMSE on stand level. We expect that the developed model can also be used for modeling relationships between other forest variables and satellite image time series.
... The ALOS-2 images were also filtered with the GammaMap filter to reduce speckle [47,48] using a 5 × 5 window size. The geometric correction was performed through geocoding, using the digital elevation model (DEM) produced by the Shuttle Radar Topography Mission (SRTM) and made available at the 90 m spatial resolution. ...
The re-suppression of secondary vegetation (SV) in the Brazilian Amazon for agriculture or land speculation occurs mostly in the rainy season. The use of optical images to monitor such re-suppression during the rainy season is limited because of the persistent cloud cover. This study aimed to evaluate the potential of C- and L-band SAR data acquired in the rainy season to discriminate SV in an area of new hotspot of deforestation in the municipality of Colniza, northwestern of Mato Grosso State, Brazil. This is the first time that the potential of dual-frequency SAR data was analyzed to discriminate SV, with an emphasis on data acquired during the rainy season. The L-band ALOS/PALSAR-2 and the C-band Sentinel-1 data acquired in March 2018 were processed to obtain backscattering coefficients and nine textural attributes were derived from the gray level co-occurrence matrix method (GLCM). Then, we classified the images based on the non-parametric Random Forest (RF) and Support Vector Machine (SVM) algorithms. The use of SAR textural attributes improved the discrimination capability of different LULC classes found in the study area. The results showed the best performance of ALOS/PALSAR-2 data classified by the RF algorithm to discriminate the following representative land use and land cover classes of the study area: primary forest, secondary forest, shrubby pasture, clean pasture, and bare soil, with an overall accuracy and Kappa coefficient of 84% and 0.78, respectively. The RF outperformed the SVM classifier to discriminate these five LULC classes in 14% of overall accuracy for both ALOS-2 and Sentinel-1 data sets. This study also showed that the textural attributes derived from the GLCM method are highly sensitive to the moving window size to be applied to the GLCM method. The results of this study can assist the future development of an operation system based on dual-frequency SAR data to monitor re-suppression of SV in the Brazilian Amazon or in other tropical rainforests.
... Winter reference images were created as a composite of ascending and descending orbit scenes from January of the same year as the summer scene. An angular-based radiometric slope correction was also applied (Vollrath et al., 2020) to lessen the effects of surface scattering on the backscatter signal, which is a significant issue in mountainous and complex terrain, such as that of the Brooks Range (Small, 2011). The S1 SAR data are from the pre-processed Ground Range Detected (GRD) collection in GEE, which does not include a radiometric slope correction. ...
Perennial snowfields are a critical part of the alpine ecosystem, serving as habitat for an array of wildlife species, and influencing downslope hydrology, vegetation, geology, and permafrost. In this study, perennial snowfield extents in the Brooks Range of Arctic Alaska are derived from Synthetic Aperture Radar (SAR) and multi-spectral satellite remote sensing via the Sentinel-1 (S1) and Sentinel-2 (S2) constellations. Snow cover area (SCA) is mapped using multi-spectral analysis in S2 and via the creation of a SAR backscatter change detection algorithm with S1. Results of the remote sensing techniques are evaluated by comparison with field data acquired across multiple spatial resolutions and geographic domains, including helicopter points and manual, on the-ground collected SCA. Evaluations of the SAR change detection algorithm via comparison with results from multi-spectral imagery analysis, and field acquired data, indicate that the SAR algorithm performs best in small, focused geographic sub-domains. This may be the result of SAR algorithm dependency on thresholding and slope corrections in mountainous terrain. An alternative approach to mapping the perennial snowfields is also presented, as a synthesis of the S1 and S2 results, wherein S1 results are used to fill voids left in the S2 data from cloud masking processes.
... S1 image processing workflows can be viewed in Figure 2A, adapted from [38] using ESA open-source SNAP software [58]. The output of SAR processing is radiometric terrain corrected [61] gamma nought (γ 0 ) backscatter image for a given date, at 30 m resolution. Backscatter values for two dominant land cover classes on the Mesa, grassland/herbaceous and evergreen forest, are explored from 2019-2020. ...
The transition of a cold winter snowpack to one that is ripe and contributing to runoff is crucial to gauge for water resource management, but is highly variable in space and time. Snow surface melt/freeze cycles, associated with diurnal fluctuations in radiative inputs, are hallmarks of this transition. C-band synthetic aperture radar (SAR) reliably detects meltwater in the snowpack. Sentinel-1 (S1) C-band SAR offers consistent acquisition patterns that allow for diurnal investigations of melting snow. We used over 50 snow pit observations from 2020 in Grand Mesa, Colorado, USA, to track temperature and wetness in the snowpack as a function of depth and time during snowpack phases of warming, ripening, and runoff. We also ran the physically-based SnowModel, which provided a spatially and temporally continuous independent indication of snowpack conditions. Snowpack phases were identified and corroborated by comparing field measurements with SnowModel outputs. Knowledge of snowpack warming, ripening, and runoff phases was used to interpret diurnal changes in S1 backscatter values. Both field measurements and SnowModel simulations suggested that S1 SAR was not sensitive to the initial snowpack warming phase on Grand Mesa. In the ripening and runoff phases, the diurnal cycle in S1 SAR co-polarized backscatter was affected by both surface melt/freeze as well as the conditions of the snowpack underneath (ripening or ripe). The ripening phase was associated with significant increases in morning backscatter values, likely due to volume scattering from surface melt/freeze crusts, as well as significant decreases in evening backscatter values associated with snowmelt. During the runoff phase, both morning and evening backscatter decreased compared to reference values. These unique S1 diurnal signatures, and their interpretations using field measurements and SnowModel outputs, highlight the capacities and limitations of S1 SAR to understand snow surface states and bulk phases, which may offer runoff forecasting or energy balance model validation or parameterization, especially useful in remote or sparsely-gauged alpine basins.
... (2) multilooking with three looks in range and two looks in azimuth; (3) gamma speckle filtering (Baraldi and Parmiggiani, 1995;Tso and Mather, 2016); (4) geocoding, radiometric calibration, and terrain correction using the 12.5 m/ pixel ALOS DEM (https://search.asf.alaska.edu/ #/); (5) outputting backscatter intensity in σ 0 [dB] format with a spatial resolution of 10 m/pixel (Equation (1)) (Small, 2011); and (6) converting to GeoTIFF format for subsequent processing in the ArcGIS platform. Thirdly, according to the different re-vectorized morphostratigraphic units at each site, the backscatter intensity (mean value and one standard deviation) was calculated by every 10 m/pixel of SAR data using the zonal statistics tool in the ArcGIS platform, which is used for the calculation of the NBI value. ...
Despite recent advances in mapping and dating alluvial fans, due to the availability of high-resolution remote sensing data and Quaternary dating techniques, quantifying surface features in remote sensing data remains a challenge. Surface roughness is a time-dependent feature under stable conditions, which indicates the relative age of alluvial fans in a hyperarid environment. Although surface roughness can be quantitatively inferred from remote sensing data, determining surface roughness in a uniform index remains a complex problem. Here, we used the normalized backscatter intensity (NBI) from high-resolution ALOS PALSAR data to quantify alluvial fan surface roughness, which is further used to quantitatively map alluvial fans. We established a robust power-law relation between the NBI value (R) and the in-situ age (T) as measured with independently dated alluvial fans. Based on the R-T relation, it can be further to apply the R measurement to T estimates as old as ∼540 ka with an average uncertainty of ∼25% on a regional scale. The NBI value, independent of atmospheric conditions and sensitive to surface roughness variability, is an effective criterion for quickly distinguishing alluvial fans and performing age estimation. We propose that insolation weathering is an important physical weathering pattern in the Dead Sea area, which mainly controls the surface roughness in this hyperarid region.
... The purpose of the study is to delineate the flooded areas of Odisha along Subarnarekha River Basin due to the recent natural event "Cyclone Yaas". The GRD products consist of focused SAR data that was detected, multi-looked, and projected to the ground range using the WGS-84 Earth ellipsoid model (Uddin, et al., 2019) through some pre-processing steps including data import, radiometric calibration, speckle filtering, radiometric terrain correction, linear-to-backscattering coefficient decibel scaling (dB) transformation, and data export using ESA's Sentinel Application Platform (SNAP) (Small, 2011) (Ajadi, et al., 2016). Firstly, two products from different dates were first imported into "SNAP Desktop" tool and therefore cropped into the particular extent of Odisha, India by creating a raster subset for two images. ...
Floods in India are extremely frequent owing to their geographical position. Recently, the formation of Cyclone Yaas over the Bay of Bengal struck the Indian state of Odisha, causing severe devastation to a larger extent. This study aimed at developing a methodology for rapid mapping of potential flood-affected areas exploiting Sentinel-1 synthetic aperture radar data through the usage of ground range detected images to mitigate the extent of flood damage and make a quick response. Sentinel-1 images from Pre-flood date: May 17, 2021 (The Archive Image) and post-flood date: May 29, 2021 (The crisis Image) were used from pre-processed products made available by the European Space Agency (ESA), which can be quickly treated for information extraction through different algorithms. Flooded areas appeared in red as there was a high response in the red channel but a low response in both the green and blue channels. Uniformly dark return given that will have a low backscatter return both in the archive and crisis image for which in red, green, and blue have a lower response. Finally, flood maps could be distributed to the local community to expedite aid in flood-prone areas. The data and methodology of the study could be replicated from time to time for flood mapping with high precision of flood-affected areas as the European Space Agency (ESA) offers free products i.e., systematically archived data for 24h with a high repetition rate (six days) to the public for flood analysis.
... Multi-temporal PolSAR has very rich characteristics [31]. In order to improve the accuracy of crop classification, people have proposed various methods to use the rich information in PolSAR, such as support vector machine [32] and random forest (RF) [33]. SVM and RF have attracted much attention because of their good performance and ease of use in limited training samples, but the feature extraction of these methods is still limited [34]. ...
Crop classification is an important part of crop management and yield estimation. In recent years, neural networks have made great progress in synthetic aperture radar (SAR) crop classification. However, the insufficient number of labeled samples limits the classification performance of neural networks. In order to solve this problem, a new crop classification method combining geodesic distance spectral similarity measurement and a one-dimensional convolutional neural network (GDSSM-CNN) is proposed in this study. The method consisted of: (1) the geodesic distance spectral similarity method (GDSSM) for obtaining similarity and (2) the one-dimensional convolu-tional neural network model for crop classification. Thereinto, a large number of training data are extracted by GDSSM and the generalized volume scattering model which is based on radar vegetation index (GRVI), and then classified by 1D-CNN. In order to prove the effectiveness of the GDSSM-CNN method, the GDSSM method and 1D-CNN method are compared in the case of a limited sample. In terms of evaluation and verification of methods, the GDSSM-CNN method has the highest accuracy, with an accuracy rate of 91.2%, which is 19.94% and 23.91% higher than the GDSSM method and the 1D-CNN method, respectively. In general, the GDSSM-CNN method uses a small number of ground measurement samples, and it uses the rich polarity information in multi-temporal fully polarized SAR data to obtain a large number of training samples, which can quickly improve the accuracy of classification in a short time, which has more new inspiration for crop classification.
... Radiometric distortions over rugged terrain within the backscatter products on GEE originate from the side-looking SAR imaging geometry. Such distortions are strong enough to exceed weaker differences of the signal due to variation in land cover [48]. It is, therefore, necessary to account for these effects during the generation of higher-level backscatter products to enable a variety of land applications [49]. ...
Abandoned cropland may lead to a series of issues regarding the environment, ecology, and food security. In hilly areas, cropland is prone to be abandoned due to scattered planting, relatively fewer sunlight hours, and a lower agricultural input–output ratio. Furthermore, the impact of abandoned rainfed cropland differs from abandoned irrigated cropland; thus, the corresponding land strategies vary accordingly. Unfortunately, monitoring abandoned cropland is still an enormous challenge in hilly areas. In this study, a new approach was proposed by (1) improving the availability of Sentinel-1 and Sentinel-2 images by a series of processes, (2) obtaining training samples from multisource data overlay analysis and timeseries viewer tool, (3) mapping annual land cover from all available Sentinel-1 and Sentinel-2 images, training samples, and the random forest classifier, and (4) mapping the spatiotemporal distribution of abandoned rainfed cropland and irrigated cropland in hilly areas by assessing land-cover trajectories along with time. The result showed that rainfed cropland had lower F1 scores (0.759 to 0.8) compared to that irrigated cropland (0.836 to 0.879). High overall accuracies of around 0.90 were achieved, with the kappa values ranging from 0.851 to 0.862, which outperformed the existing products in accuracy and spatial detail. Our study provides a reference for extracting the spatiotemporal distribution of abandoned rainfed cropland and irrigated cropland in hilly areas.
... Temporal multi-looked (amplitude) maps and Coherence Changes Indexes (CCDIs) extracted from the available SAR images were used to detect changes due to the wildfire and flooding events characterizing the selected AOIs. The SAR images were preliminarily radiometrically calibrated [152] to extract from the digital data the maps of the radar backscatter, i.e., the sigma naught maps [153]. We assumed the 24 July 2021 date as that related to the primary fire event. ...
This work aims to clarify the potential of incoherent and coherent change detection (CD) approaches for detecting and monitoring ground surface changes using sequences of synthetic aperture radar (SAR) images. Nowadays, the growing availability of remotely sensed data collected by the twin Sentinel-1A/B sensors of the European (EU) Copernicus constellation allows fast mapping of damage after a disastrous event using radar data. In this research, we address the role of SAR (amplitude) backscattered signal variations for CD analyses when a natural (e.g., a fire, a flash flood, etc.) or a human-induced (disastrous) event occurs. Then, we consider the additional pieces of information that can be recovered by comparing interferometric coherence maps related to couples of SAR images collected between a principal disastrous event date. This work is mainly concerned with investigating the capability of different coherent/incoherent change detection indices (CDIs) and their mutual interactions for the rapid mapping of “changed” areas. In this context, artificial intelligence (AI) algorithms have been demonstrated to be beneficial for handling the different information coming from coherent/incoherent CDIs in a unique corpus. Specifically, we used CDIs that synthetically describe ground surface changes associated with a disaster event (i.e., the pre-, cross-, and post-disaster phases), based on the generation of sigma nought and InSAR coherence maps. Then, we trained a random forest (RF) to produce CD maps and study the impact on the final binary decision (changed/unchanged) of the different layers representing the available synthetic CDIs. The proposed strategy was effective for quickly assessing damage using SAR data and can be applied in several contexts. Experiments were conducted to monitor wildfire’s effects in the 2021 summer season in Italy, considering two case studies in Sardinia and Sicily. Another experiment was also carried out on the coastal city of Houston, Texas, the US, which was affected by a large flood in 2017; thus, demonstrating the validity of the proposed integrated method for fast mapping of flooded zones using SAR data.
... The pre-processing of the S1 dataset was carried out using the Sentinel-1 Toolboxes, implemented in the SNAP v.8.0.3 open-source software (ESA SNAP Homepage 2022), and performed via the SNAP-Python interface (Snappy), the access provider to SNAP Java API (ESA SNAP Cookbook 2022). Starting by applying the auto-downloaded orbit information file and the removal of the thermal noise, the SAR data pre-processing involved the radiometric calibration to beta (b0) noughts backscatter standard conventions (Small 2011) and the radiometric terrain correction (RTC) process. RTC consists of the radiometric terrain flattening and the geometric terrain correction of the images using a digital elevation model (DEM) to reduce the geometric and radiometric distortions due to the rough surface topography. ...
In the present study, the temporal and spatial dynamics of the post-fire recovery of different Mediterranean vegetation types during the three years after the fire event were analyzed, according to different fire severity categories, integrating the use of Synthetic Aperture Satellite Radar (SAR) (Sentinel-1) and optical (Sentinel-2) image time series. The results showed that Mediterranean forest species and shrub/herbaceous species are adapted to fire, with high efficiency in restoring the vegetation cover. Differently, the ecological vulnerability of non-native eucalyptus plantations was found in a lower recovery trend during the observation period. The use of optical short-wave infrared (SWIR) and SAR C-band-based data revealed that some ecological characteristics, such as the woody biomass and structure, recovered at slower rates, comparing to those suggested by using near-infrared (NIR) and red-edge data. An optimized burn recovery ratio (BRR) was proposed to estimate and map the spatial distribution of the degree of vegetation recovery.
... The images were radiometrically terrain-flattened and orthorectified with in-house software using local digital elevation model available from National Land Survey of Finland [41], [42]. Final preprocessed images were in gamma-naught format [43]. ...
In this study, we introduce an improved semisupervised deep learning approach, and demonstrate its suitability for modeling the relationship between forest structural parameters and satellite remote sensing imagery and producing forest maps. The improved approach is based on a popular UNet model, modified and fine-tuned to improve the forest parameter prediction performance. Within the improved model, squeeze-and-excitation blocks are embedded to recalibrate the multisource features via retrieved channel-wise self-attention and a novel cross-pseudo regression strategy is implemented to train the model in a semisupervised way. The improvement imposes consistency learning on two perturbed network branches: 1) generating regression pseudo-reference; 2) expanding the dataset size. For demonstration, we used satellite synthetic aperture radar (SAR) Sentinel-1 and multispectral optical Sentinel-2 images as remote sensing data, complemented with reference data represented by forest tree height as one of the key forest structural variables. The study area is located in a boreal forestland in Central Finland. Proposed approach showed larger accuracy compared to traditional machine learning methods such as random forests and boosting trees, and baseline UNet model. Best accuracy figures for forest tree height were achieved with combined SAR and optical imagery and were as small as 24.1% root-mean-square error (RMSE) on pixel-level and 15.4% RMSE on forest stand level.
... The oblique imaging geometry makes a precise orthorectification crucial to avoid image artefacts. For precise orthorectification, a SAR image simulated from an external DEM (Small et al., 1998;Small, 2011) can be used as a reference to align the reference SAR image with the DEM used for orthorectification (Wegmüller, 1999). DEM-based coregistration of a SAR image stack to a common reference scene is of advantage, in addition to coregistration of the reference scene to the simulated reference. ...
... Masking the sea off the terrestrial part of the study area was performed to understand the connotation of the island's displacement and the total area changes during the study period. Finally, the terrain of the procced data was corrected following Small (2011) to be written as the World Geodetic System (WGS_84), where the area of the islands can be monitored. Figure 2 shows the methodological framework of the current study. ...
Tiran and Sanafir are coastal ecosystem islands located in the fragile environment of the Red Sea. The importance of the islands has always been bordering security for Egypt and Saudi Arabia. In the last decade, an economic value has been added to the border islands and has grown a significant importance to be the focal point between the two countries’ trade lines. The study is aiming at the temporal analysis of the SAR data phase estimation to map the island’s vertical displacement for the last 5 years. Temporal datasets of SAR images were collected from the European Space Agency, and dataset collection started from November 2016 until May 2020 to comprehend the objectives of the research study. The SAR interferometry techniques were implemented to estimate the phase displacement for 15 SAR data sets of the border islands. Consequently, the temporal analysis demonstrated the trend of the phase displacement during the study timeframe. Results indicated that the maximum island subsidence (−0.417 m) was estimated from the scene acquired on 21 Feb 2018, while the maximum uplift (0.708 m) value was conducted from the scene acquired on 17 Nov 2017. Moreover, the trend analysis showed an instability behavior among the successive SAR datasets. The current study’s findings are crucial to the decision-makers from Egypt and Saudi Arabia in realizing the predictability of the island’s displacement mitigations to ensure the sustainability of the incoming socioeconomic activities.
... For preparation, we were looking for the signs of these CRs in the Sentinel-1 series (Figure 9). Their radar brigntnesses are presented by gamma naught (Small, 2011). Gamma naught is defined as the pure intensity of a complex signal reflected from ground. ...
Monitoring mining impact has become increasingly important as the awareness of safety and environmental protection is rising. Our project Integrated Mining Impact Monitoring (i2Mon), funded by European Commission – Research Fund for Coal and Steel, intends to monitor the mining-induced impact, in particular, ground movement. The monitoring system comprises terrestrial measurement and remote sensing: levelling, GPS, LiDAR scanning, UAV survey, and SAR interferometry. The aim is to launch an interactive GIS-based platform as an early warning and decision making service for mining industry. This study has developed a scheme based on advanced SAR interferometry to monitor the ground movement over an extensive area at millimetre level. The first test site is a deactivated open-pit mine in Cottbus, Germany owned by Lausitz Energie Bergbau AG. The whole area was reclaimed into a post-mining lake and must be monitored for the safety. The second test site is located in Poland, where the underground mining operated by POLSKA GRUPA GÓRNICZA began in June 2021. We have monitored the in-situ ground movement carefully as part of the influenced area covers the human settlement. The ground movement of our test sites was analysed from Sentinel-1 images. The crucial parameters include stepwise movement series, instantaneous velocities and accelerations, and significance index. In addition, six corner reflectors along with sensors like GPS were installed across the region in Cottbus. They were observed in the Sentinel-1 series and the GPS readings will be used for validation. Finally, all the data will be integrated into DMT’s platform – SAFEGUARD.
... Instead, we used a terrain flattening operation that takes into account the local incidence angles (denoted as θ l in Figure 2.5), since the analyzed area is wide and parcel features are compared to each other (the interest of using the local incidence is illustrated in Figure 2.5). This operation uses the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) to produce γ 0 backscattering coefficients (Small, 2011). ...
Crop monitoring will become a major challenge in the coming years. Under the pressure of climate change on the one hand, and the increase of the world population on the other hand, food supply chains are likely to be strongly constrained, impacting food security in many areas of the planet. In this context, using remote sensing to acquire information on vegetation status will be a key asset. One of the areas directly concerned is precision agriculture, which consists in optimizing yields and agricultural practices. With the arrival of the Copernicus mission satellites, Sentinel-1 (synthetic aperture radar) and Sentinel-2 (multispectral imagery), the possibilities of applications in this area have increased drastically. Indeed, Sentinel data are freely available, with a temporal and spatial resolution adapted to crop monitoring at the parcel level. The main objective of this thesis is to propose a strategy to automatically detect agricultural parcels with abnormal agronomic development. Special attention was given to the joint use of Sentinel-1 and Sentinel-2 data. Moreover, in order to be easily deployed in an operational context, a constraint is to have a method able to analyzing a single growth cycle (or a part of it). To meet the objectives of the thesis, we first propose a processing chain allowing the extraction of agronomic indicators at the parcel-level. These indicators are calculated in two steps: 1) calculation of agronomic indicators at the pixel level and 2) calculation of spatial statistics at the plot level. Then, these indicators are used to detect parcels with abnormal phenological behavior. The detection is unsupervised and performed using an anomaly detection algorithm. A comparison of several approaches was made to find the most suitable method for our problem. Among the different algorithms tested, the most efficient method is the isolation forest, which also has the advantage of being fast and not very sensitive to the choice of its parameters. Thanks to the proposed method, it is possible to detect plots with abnormal behavior with a high accuracy. The results obtained were validated on two different types of crops, wheat and rapeseed. In a second step, we addressed the problem of anomaly detection in the presence of missing data. This problem is fundamental in remote sensing, in particular for multispectral data because they are sensitive to cloud cover. To solve this problem, we propose to reconstruct the missing data (at the parcel-level) using Gaussian mixture models. This approach has been found to be significantly better than the other tested approaches for reconstructing missing data and for detecting anomalies on parcels with incomplete time series. In addition, we also have proposed a method for estimating Gaussian mixture models that are robust to the presence of outliers in the data. This method is particularly useful in the presence of strong outlier values, for example in the presence of parcels coming from a different crop type than the one analyzed. Finally, we explore in this thesis anomaly detection approaches that take into account the temporal structure of the data. In particular, we propose a method based on an ensemble of hidden Markov models. One of the interests of this approach is to be able to localize the anomalies in time.
Mapping abandoned land is very important for accurate agricultural management. However, in karst mountainous areas, continuous high-resolution optical images are difficult to obtain in rainy weather, and the land is fragmented, which poses a great challenge for remote sensing monitoring of agriculture activities. In this study, a new method for identifying abandoned land is proposed: firstly, a few Google Earth images are used to transform arable land into accurate vectorized geo-parcels; secondly, a time-series data set was constructed using Sentinel-1A Alpha parameters for 2020 on each farmland geoparcel; thirdly, the semi-variation function (SVF) was used to analyze the spatial-temporal characteristics, then identify abandoned land. The results show: (1) On the basis of accurate spatial information and boundary of farmland land, the SAR time-series dataset reflects the structure and time-series response. The method eventually extracted abandoned land with an accuracy of 80.25%. The problem of remote sensing monitoring in rainy regions and complex surface areas is well-resolved. (2) The spatial heterogeneity of abandoned land is more obvious than that of cultivated land within geo-parcels. The step size for significant changes in the SVF of abandoned land is shorter than that of cultivated land. (3) The SVF time sequence curve presented a strong peak feature when farmland was abandoned. This reveals that the internal spatial structure of abandoned land is more disordered and complex. It showed that time-series variations of spatial structure within cultivated land have broader applications in remote sensing monitoring of agriculture in complex imaging environments.
Accurate extraction of urban impervious surface (UIS) is essential for urban planning and environmental monitoring. However, multispectral remote sensing data for UIS extraction suffers from the inter-class spectral confusions, e.g. UIS and bare soil, and intra-class variations of sub-class UIS. Hyperspectral and full/dual-polarization synthetic aperture radar (full/dual PolSAR) data provide opportunities for reducing such confusions and have potential for fine UIS mapping, i.e., roads, buildings, and grounds. In this study, we first investigated the hyperspectral data (Gaofen-5) capability to reduce the intra/inter-class misclassification in comparison with multispectral data (Landsat-8). Then, we explored contributions of synergistically using full and dual PolSAR (ALOS-2 and Sentinel-1) with hyperspectral and multispectral data using optical-SAR sparse representation classification (OSSRC). Results showed that both the hyperspectral and the SAR polarization features helped better delineation between UIS and bare soil, and sub-class UIS (roads and buildings). The relative contribution of PolSAR was higher in multispectral data than in hyperspectral data, with full PolSAR contributed significantly. The combined hyperspectral and full PolSAR data using OSSRC delivered the best result, with an overall accuracy higher than 90%. The results indicate the promising capability of synergizing hyperspectral and full/dual PolSAR data for improving UIS extraction from advanced satellite data.
In cotton, an optimal balance between vegetative and reproductive growth will lead to high yields and water-use efficiency. Remote sensing estimations of vegetation variables such as crop coefficient (Kc), Leaf Area Index (LAI), and crop height during plant development can improve irrigation management. Optical and Synthetic Aperture Radar (SAR) satellite imagery can be a useful data source since they provide synoptic cover at fixed time intervals. Furthermore, they can better capture the spatial variability in the field compared to point measurements. Since clouds limit optical observations at times, the combination with SAR can provide information during cloudy periods. This study utilized optical imagery acquired by Sentinel-2 and SAR imagery acquired by Sentinel-1 over cotton fields in Israel. The Sentinel-2-based vegetation indices that are best suited for cotton monitoring were identified, and the most robust Sentinel-2 models for Kc, LAI, and height estimation achieved R² = 0.879, RMSE= 0.0645 (MERIS Terrestrial Chlorophyll Index, MTCI); R² = 0.9535, RMSE= 0.8 (MTCI); and R² = 0.8883, RMSE= 10 cm (Enhanced Vegetation Index, EVI), respectively. Additionally, a model based on the output of the SNAP Biophysical Processor LAI estimation algorithm was superior to the empirical LAI models of the best-performing vegetation indices (R² =0.9717, RMSE=0.6). The most robust Sentinel-1 models were obtained by applying an innovative local incidence angle normalization method with R² = 0.7913, RMSE= 0.0925; R² = 0.6699, RMSE= 2.3; R² = 0.6586, RMSE= 18 cm for the Kc, LAI, and height estimation, respectively. This work paves the way for future studies to design decision support systems for better irrigation management in cotton, even at the sub-plot level, by monitoring the heterogeneous development of the crop from space and adapting the irrigation accordingly to reach the target development at different growth stages during the season.
Synthetic aperture radar (SAR) data provides an appealing opportunity for all-weather day or night Earth surface monitoring. The European constellation Sentinel-1 (S1) consisting of S1-A and S1-B satellites offers a suitable revisit time and spatial resolution for the observation of croplands from space. The C-band radar backscatter is sensitive to vegetation structure changes and phenology as well as soil moisture and roughness. It also varies depending on the local incidence angle (LIA) of the SAR acquisition’s geometry. The LIA backscatter dependency could therefore be exploited to improve the retrieval of the crop biophysical variables. The availability of S1 radar time-series data at distinct observation angles holds the feasibility to retrieve leaf area index (LAI) evolution considering spatiotemporal coverage of intensively cultivated areas. Accordingly, this research presents a workflow merging multi-date S1 smoothed data acquired at distinct LIA with a Gaussian processes regression (GPR) and a cross-validation (CV) strategy to estimate cropland LAI of irrigated winter wheat. The GPR-S1-LAI model was tested against in situ data of the 2020 winter wheat campaign in the irrigated valley of Colorador river, South of Buenos Aires Province, Argentina. We achieved adequate validation results for LAI with RCV2 = 0.67 and RMSECV = 0.88 m2 m−2. The trained model was further applied to a series of S1 stacked images, generating temporal LAI maps that well reflect the crop growth cycle. The robustness of the retrieval workflow is supported by the associated uncertainties along with the obtained maps. We found that processing S1 smoothed imagery with distinct acquisition geometries permits accurate radar-based LAI modeling throughout large irrigated areas and in consequence can support agricultural management practices in cloud-prone agri-environments.
Despite impressive progress in the past decade, accurate and efficient multi-view SAR image registration remains a challenging task due to complex imaging mechanisms and various imaging conditions. Especially for rugged areas, SAR images obtained from the opposite-side view reflect different characteristics, making popular SAR image registration methods no longer applicable. To this end, we propose a geometry-aware image registration method by extracting inherent orientation features and concentrating on geometry-invariant areas. First, slant range images are terrain-corrected using Digital Elevation Model (DEM) to reduce large relative positioning errors caused by elevation. Second, the Gabor-ratio detector is introduced to obtain multi-scale orientation features, which are more robust under various imaging conditions. Then, a geometry-aware mask is produced by intersecting the 3D space ray with DEM, and thus SAR images can be divided into three categories, layover, shadow, and geometry-invariant areas. The geometry-aware matching method, which focuses on geometry-invariant areas and masks out misleading caused by geometric and radiometric distortions, is proposed to realize accurate matching. The rational polynomial coefficients (RPCs) are refined to achieve relative correction. Extensive results on dozens of SAR images demonstrate the effectiveness and universality of the proposed algorithm by quantitative evaluation using man-made and natural corner reflectors. An analysis of the factors affecting registration accuracy is also discussed.
The role of synthetic aperture radar (SAR) data for flood area mapping is well established. The backscatter of different polarizations of SAR data interacts in varying ways over the same region and assists in differentiating land and flood pixels. This study brings out the significance of comparing and analysing different polarizations. There were insignificant differences was found in flood extent maps, and it has also been identified that both methods were capable of rapid flood mapping. Multi-year flood extent maps depicted that the information on recurrent flood extents was useful for delineating flood hazard zonation and also for prioritizing flood risk and mitigation measures.
Global warming and human activities continue to affect the dynamic balance of the Qinghai-Tibet Plateau (QTP) frozen soil environment, affecting the safety and stability of its infrastructure. Spaceborn synthetic aperture radar interferometry (InSAR) is a well developed earth observation technology in the last two decades. This study investigated the feasibility of extracting surface deformation in complex and diverse geomorphic landscape regions using ascending-descending orbit combination SAR interferometry. The temporal and spatial distribution characteristics of frozen soil deformation and its relationship with climate change were further analyzed. A total of 110 Sentinel-1A ascending and descending SLC images from October 2017 to May 2021 were employed to cover the Qinghai-Tibet engineering corridor. A time series InSAR (TS-InSAR) method based on small baseline strategy and backscattering characteristics analysis was developed to estimate the surface deformation in the LOS directions. Then, combined with the geometric parameters of the SAR system, the 2D deformation results of frozen soil were calculated. Global Navigation Satellite System (GNSS) data gained from the QHTT Crustal Movement Observation Network of China (CMONOC) reference station was used for validation, demonstrating good agreement in displacement rates and time series. Our study has observed the seasonal components of the GNSS and TS-InSAR results based on a time series decomposition model. It was revealed that both sets of results are consistent with the existing climate change law, which can reflect the freeze-thaw cycle of frozen soil.
For already more than seven years, the Sentinel-1 C-band Synthetic Aperture Radar (SAR) mission has been providing indispensable information for monitoring bio-geophysical parameters at fine temporal and spatial scales. As many applications require backscatter datacubes as input, enormous amounts of data have to be radiometrically and geometrically corrected to be in a common, Earth-fixed reference system. Pre-processing workflows accomplishing this task have already been established and are implemented in several software suites. However, typically, these workflows are computationally expensive which may lead to prohibitively large costs when generating multi-year Sentinel-1 datacubes for whole continents or the world. In this paper, we discuss existing approaches for generating sigma nought and projected local incidence angle (PLIA) data and present simplifications of the overall workflow relying on the unprecedented orbital stability of Sentinel-1. Propagating orbital deviations through the complete Sentinel-1 pre-processing pipeline helped us to simulate and identify PLIA as a static layer per relative orbit. The outcome of these simulations also provided the necessary information to replace iterative root-finding algorithms for determining the time of closest approach (TCA), i.o.w. the azimuth index, with a linear one — at no expense of radiometric accuracy. All experiments were performed using an in-house developed toolbox named wizsard, which made it possible to speed up Sentinel-1 pre-processing by approximately 4–5 times with respect to the Sentinel Application Platform (SNAP). This could pave the way for producing quality-curated, large-scale backscatter datacubes at continental and global scales in acceptable time frames.
Coastal mapping with satellite imagery is broadly used to calculate shoreline positions due to its high ecological and socioeconomic value in the context of coastal conservation and management strategies. We show the applicability of the Sentinel-1 to monitor large-scale shoreline at a country level. The present study develops a novel shoreline extraction method based on C band from SAR missions, that improves coastal ocean/land discrimination. The method considers an automated processing chain using the incorporation of GLCM-mean texture information to increase improvements in image binarization by Sauvola thresholding. Results show that the proposed method may be used for shoreline monitoring of different types of geomorphology along the Mexican coastline, thus guaranteeing its applicability in different geographic surroundings. For the six specific areas of validation, the overall agreement between binarization ranges is from 90% to 100% and Sentinel-2 images are used to evaluate the VV and VH shorelines from Sentinel-1.
Synthetic Aperture Radar (SAR) data handling, processing, and interpretation are barriers preventing a rapid uptake of SAR data by application specialists and non-expert domain users in the field of agricultural monitoring. To improve the accessibility of Sentinel-1 data, we have generated a reduced-volume, multi-year Sentinel-1 SAR database. It includes mean and standard deviation of VV, VH and VH/VV backscatter, pixel counts, geometry, crop type, local incidence angle and azimuth angle at parcel-level. The database uses around 3100 Sentinel-1 images (5 TB) to produce a 12 GB time series database for approximately 770,000 crop parcels over the Netherlands for a period of three years. The database can be queried by Sentinel-1 system parameters (e.g. relative orbit) or user application-specific parameters (e.g. crop type, spatial extent, time period) for parcel level assessment. The database can be used to accelerate the development of new tools, applications and methodologies for agricultural and water related applications, such as parcel-level crop bio-geophysical parameter estimation, inter-annual variability analysis, drought monitoring, grassland monitoring and agricultural management decision-support. Measurement(s)parcel-level backscatter • cross-ratioTechnology Type(s)SAR remote sensing • Sentinel-1 • Google Earth EngineSample Characteristic - LocationThe Netherlands Measurement(s) parcel-level backscatter • cross-ratio Technology Type(s) SAR remote sensing • Sentinel-1 • Google Earth Engine Sample Characteristic - Location The Netherlands
Remote sensing techniques are of particular interest for monitoring wildfire effects on soil properties, which may be highly context-dependent in large and heterogeneous burned landscapes. Despite the physical sense of synthetic aperture radar (SAR) backscatter data for characterizing soil spatial variability in burned areas, this approach remains completely unexplored. This study aimed to evaluate the performance of SAR backscatter data in C-band (Sentinel-1) and L-band (ALOS-2) for monitoring fire effects on soil organic carbon and nutrients (total nitrogen and available phosphorous) at short term in a heterogeneous Mediterranean landscape mosaic made of shrublands and forests that was affected by a large wildfire. The ability of SAR backscatter coefficients and several band transformations of both sensors for retrieving soil properties measured in the field in immediate post-fire situation (one month after fire) was tested through a model averaging approach. The temporal transferability of SAR-based models from one month to one year after wildfire was also evaluated, which allowed to assess short-term changes in soil properties at large scale as a function of pre-fire plant community type. The retrieval of soil properties in immediate post-fire conditions featured a higher overall fit and predictive capacity from ALOS-2 L-band SAR backscatter data than from Sentinel-1 C-band SAR data, with the absence of noticeable under and overestimation effects. The transferability of the ALOS-2 based model to one year after wildfire exhibited similar performance to that of the model calibration scenario (immediate post-fire conditions). Soil organic carbon and available phosphorous content was significantly higher one year after wildfire than immediately after the fire disturbance. Conversely, the short-term change in soil total nitrogen was ecosystem-dependent. Our results support the applicability of L-band SAR backscatter data for monitoring short-term variability of fire effects on soil properties, reducing data gathering costs within large and heterogeneous burned landscapes.
This chapter provides a brief introduction to the synthetic aperture radar (SAR) imaging geometry and the imaging method and a description of radar speckle in order to provide the basics for the different image correlation methods used for SAR offset tracking. SAR systems measure complex‐valued backscatter amplitudes containing phase information, whereas optical systems can only measure intensity. There are various methods to generate a multi‐looked SAR image, but the most common method is spatially averaging the backscatter intensity, assuming that adjacent pixels represent similar types of scattering. For SAR offset tracking, the intensity distribution of measured backscatter values is relevant. Offset tracking originates from image registration methods. For SAR imagery, offset tracking originates from SAR interferometry, which requires the precise coregistration of two SAR images to obtain interferometric coherence and fringe visibility. To make tracking more robust, stacks of cross‐correlation functions or motion‐compensated image stacks can be evaluated.
The lack of data on disaster-related loss and damage, especially in developing countries, obstructs the implementation of the Sendai Framework for Disaster Risk Reduction (SFDRR). This weakens the accuracy, timeliness, and quality of the SFDRR monitoring process at the national level. This is the case for the country of Ecuador, who has not reported on the progress of SFDRR Target B, indicator B-5a, the number of workers in agriculture with crops damaged or destroyed. Consequently, this research developed a geospatial model approach to model indicator B-5a in the context of flooding. The model uses Sentinel-1 synthetic aperture radar and other spatial data to quantify indicator B-5a at a cantonal spatial level across three ecological regions in Ecuador. The results of the model were validated against country provided in-situ measured loss and damage reference data by combining elements of flood exposure and vulnerability. A statistical analysis was used to assess the agreement of the models with reference data and the models' ability to reproduce the reference data. The validation procedure produced models that had high agreement with the reference data, but the models were sensitive to the different ecological regions. This validated geospatial model approach is, to the best of the authors' knowledge, the first attempt to validate geospatially measured Sendai indicators against reference data, and provides an opportunity to support countries without information on disaster-related loss and damage in monitoring indicator B-5a of the SFDRR.
The National Imagery Interpretability Rating Scale (NIIRS) is a numeric scale of semantic criteria used to quantify the visual interpretability of imagery products. The Radar National Imagery Interpretability Rating Scale (RNIIRS) is the version used to rate imaging radar products, specifically Synthetic Aperture Radar (SAR). RNIIRS is specified by imagery users to describe the visual interpretability requirements for the task at hand and is a subjective, scene dependent estimate. A predictive equation called an Image Quality Equation (IQE) is employed based on sensor collection parameters such as Impulse Response (IPR) and Signal to Noise Ratio (SNR) to predict the RNIIRS value of a scene. This paper proposes a quantitative and objective measure to predict SAR image quality to replace these subjective measures.
The aim of this document is to describe the absolute calibration of high rate ASAR Level 1 products generated by ESA using the ASAR processing Facility (PF-ASAR).
In regions with significant terrain variations, the modulation of SAR backscatter by mountain slopes can dominate interpretation of the radar imagery unless effective countermeasures are first applied. We first demonstrate deficiencies in conventional radiometric treatments. Geocoded-terrain-corrected (GTC) products assume an ellipsoid-model for the radiometry, even if they improve upon geocoded-ellipsoid-corrected (GEC) imagery by properly compensating for the effects of terrain variations on image geometry. Both the sigma nought and gamma nought radiometric normalisation conventions as applied to distributed targets have an ellipsoidal Earth assumption at their core.
Simply using a local-incidence-angle-mask (LIM) to normalise image radiometry fails to adequately model the image formation process. We prefer to use instead a product that we refer to as terrain-corrected gamma for backscatter analysis. The product makes use of SAR image simulation, incorporating shadow checks and proper accounting of local illuminated area in foreshortened and even layover areas: the result is a substantially improved sensor model in comparison to LIM-based backscatter retrieval. Use of terrain- corrected gamma in a radiometrically terrain-corrected (RTC) product enables multi-track and even multi- sensor image overlays, as terrain-induced backscatter variations are normalised using the available DEM. By properly normalising the hills and mountains, the growing availability of SAR images from diverse sensors can be compared on a “level playing field”.
Time series analysis of hundreds of multi-track ASAR wide swath images covering Switzerland is shown to benefit when comparisons are made using terrain-corrected gamma rather than GTC or LIM- normalised SAR backscatter retrievals. We show how the spring snow melt period can be followed well using multi-track ASAR WS data only if terrain-corrected gamma backscatter values are used as the basis for comparison. Finally, we recommend improved standard backscatter retrieval from land surfaces in future SAR missions such as Biomass or CoReH2O, and Sentinel-1.
Radiometric terrain correction consists of normalising a SAR image for well-understood backscatter contribu-tions in order to amplify less easily apparent influences (e.g. thematic land cover variance). A rigorous model-ling of the SAR image formation process includes consideration of how foreshortening and layover create ambi guity when connecting map geometry grid points to and from counterparts in radar geometry (slant or ground range vs. azimuth). A radar amplitude image simulation is formed by iterating through a facetted DEM, calcu lating the accumulated illuminated area at every range and azimuth coordinate in radar geometry. We show how DEM-based image simulations gain further realism by incorporating knowledge of the SAR antenna's elevation an-tenna gain pattern (AGP). Although typical AGP corrections assume an ellipsoidal Earth, the AGP is actually draped upon the Earth's terrain. We quantify differences between estimates of local antenna gain and illuminated area performed using (a) the typical ellipsoid assumption, (b) a DEM. We demonstrate application of local an-tenna gain knowledge within the image simulation process using ENVISAT ASAR images acquired over Swit-zerland. We introduce a weighted resolution approach for robust combination of multiple radiometrically normal-ised terrain geocoded backscatter maps.
This paper summarizes the results obtained from geometric and radiometric calibrations of the Phased-Array L-Band Synthetic Aperture Radar (PALSAR) on the Advanced Land Observing Satellite, which has been in space for three years. All of the imaging modes of the PALSAR, i.e., single, dual, and full polarimetric strip modes and scanning synthetic aperture radar (SCANSAR), were calibrated and validated using a total of 572 calibration points collected worldwide and distributed targets selected primarily from the Amazon forest. Through raw-data characterization, antenna-pattern estimation using the distributed target data, and polarimetric calibration using the Faraday rotation-free area in the Amazon, we performed the PALSAR radiometric and geometric calibrations and confirmed that the geometric accuracy of the strip mode is 9.7-m root mean square (rms), the geometric accuracy of SCANSAR is 70 m, and the radiometric accuracy is 0.76 dB from a corner-reflector analysis and 0.22 dB from the Amazon data analysis (standard deviation). Polarimetric calibration was successful, resulting in a VV/HH amplitude balance of 1.013 (0.0561 dB) with a standard deviation of 0.062 and a phase balance of 0.612deg with a standard deviation of 2.66deg .
Improved geometric accuracy in SAR sensors implies that more
complex models of the Earth may be used not only to geometrically rectify imagery, but also to more robustly calibrate their radiometry. Current beta, sigma, and gamma nought SAR radiometry conventions all assume a simple “flat as Kansas” Earth ellipsoid model. We complement these simple models with improved radiometric calibration that accounts for local terrain variations. In the era of ERS-1 and RADARSAT-1, image geolocation accuracy was in the order of multiple samples, and tiepointfree establishment of the relationship between radar and map geometries was not possible. Newer sensors such as ASAR, PALSAR, and TerraSAR-X all support accurate geolocation based on product annotations alone. We show that high geolocation accuracy, combined with availability of high-resolution accurate elevation models, enables a more robust radiometric calibration standard for modern SAR sensors that is based on gamma nought normalised using an Earth terrain-model.
The nominal radar geometry available from header information accompanying a satellite radar image is often accurate enough to generate an image simulation based on topographic information, given that an accurate DHM is available.
En raison des caractéristiques relatives à la géométrie de visée des radars à antenne synthétique (RAAS) montés à bord des satellites SEASAT, ERS-1, JERS-1 et RADARSAT, les images acquises par ces capteurs présenteront une sensibilité radiométrique et géométrique à la topographie locale nettement supérieure aux images optiques correspondantes obtenues à partir des satellites Landsat et SPOT. Il est absolument essentiel, par conséquent, que des modèles numériques d'élévation fassent partie intégrante de toute analyse des capacités des radars. Cet article présente une formulation analytique unifiée d'une méthode d'analyse par simulation des RAAS — incluant le géocodage des données et la génération de produits à valeur ajoutée — dont le but est de faciliter l'interprétation de la scène. Les auteurs proposent le concept de facteur d'échelle topographique pour quantifier le taux moyen de distorsion géométrique dans une scène RAAS, supposé que l'on connaisse la variabilité de l'élévation d'un site. Un prototype de progiciel fondé sur ces algorithmes a été mis au point à peu de frais à l'aide d'un ordinateur personnel et a été utilisé pour géocoder deux scènes obtenues à partir du radar à antenne synthétique monté à bord du satellite SEASAT, qui présentent des distorsions dues au terrain à l'échelle comprise entre 1 et 1,5 km. Des précisions de l'ordre de 75 à 150 m ont été obtenues à partir de modèles numériques d'élévation à l'échelle du 1: 250 000.
During normal SAR processing, a flat earth is assumed when performing
radiometric corrections such as antenna pattern and scattering element
size removal. Here we examine the effects of topographic variations on
these corrections. Local slopes will cause the actual scattering element
size to be different from that calculated using the flat earth
assumption. It is shown that this effect, which is present for both
airborne and spaceborne SAR data, may easily cause calibration errors on
the order of a dB. In the case of airborne systems, the errors
introduced by assuming a flat earth during antenna pattern removal are
also significant; errors of several dB can easily result. The effect of
topography on antenna pattern removal is expected to be negligible for
spaceborne SARs.
Inclined surface topography diminishes the geometric and radiometric quality of synthetic aperture imagery. The correction of these effects becomes indispensable when quantitative image analysis is performed with respect to the derivation of geo- and biophysical parameters. Due to their spatial extent and frequent availability, ScanSAR image products extend the operative range of microwave imagery and have a high potential for numerous operational applications over larger areas. The study presents a procedure for a pre-operational terrain correction of ScanSAR imagery as acquired by RADARSAT and ENVISAT ASAR.
Analyses and quantifies the topographic effects on the antenna
gain pattern: correction of existing spaceborne synthetic aperture radar
systems, namely ERS-1, JERS-1, SIR-C, and X-SAR. Simulations and real
SAR data of a test site are used. The corrections are carried out taking
into account the local surface topography and compared with the standard
method based on a reference ellipsoid. Results show that elevation
variations in the ERS-1 and JERS-1 cases do not affect significantly the
antenna gain pattern correction. For extreme topographic differences,
greater than 3000 m, a reference altitude or the radiometric calibration
is suggested. On the other hand, for the low-orbit SRL-1/2 terrain
information is strongly recommended, particularly, if relief differences
within the image are significant, namely greater than 1000 m.
Furthermore, it is shown that in the SIR-C case, even if the
polarizations of the antenna gain patterns are slightly different,
polarimetric calibration errors due to relief variations can be
neglected. Finally, implications for forthcoming spaceborne SAR systems,
i.e. ERS-2 and RADARSAT, are discussed
Terrain undulations affect the geometric and radiometric quality of synthetic aperture radar images. The correction of these effects becomes indispensable when quantitative image analysis is performed with respect to the derivation of geo- and biophysical parameters. The paper presents a rigorous approach for geometric and radiometric correction of SAR images. Using a digital elevation model, the imaging geometry is reconstructed and is used to perform geometric and radiometric correction of terrain induced distortions. The importance of a stringent radiometric correction based on the integration of the image brightness is emphasized. The approach guarantees that the energy contained in the image data is preserved throughout the geocoding process. The resulting backscattering images are fully terrain corrected and can be used for further quantitative investigations and may also improve qualitative studies as e.g. land cover classifications. The technique is applicable for different sensor types and image products, including already geocoded SAR images. The effect of different resolutions of digital elevation models used for the correction of the backscattering coefficient is investigated.
The fundamental principles of radar backscattering measurements are presented, including measurement statistics, Doppler and pulse discrimination techniques, and associated ambiguity functions. The operation of real and synthetic aperture sidelooking airborne radar systems is described, along with the internal and external calibration techniques employed in scattering measurements. Attention is given to the physical mechanisms responsible for the scattering emission behavior of homogeneous and inhomogeneous media, through a discussion of surface roughness, dielectric properties and inhomogeneity, and penetration depth. Simple semiempirical models are presented. Theoretical models involving greater mathematical sophistication are also given for extended ocean and bare soil surfaces, and the more general case of a vegetation canopy over a rough surface.
The radar reflectivity coefficient of a distributed scatterer, expressed either as σ<sup>0</sup> or γ, depends on local incidence angle. Prior to incidence angle projection, the reflectivity coefficient may be called "radar brightness", denoted by "beta nought" β<sup>0</sup>. In most practical situations, the local incidence angle is not known, so the image radiometrics cannot be corrected for it. It follows that β<sup>0</sup> is almost always the more appropriate radiometric attribute of radar imagery, whether calibrated or uncalibrated It is recommended that σ<sup>0</sup> (or γ) be reserved for situations in which both the local terrain slope and the illumination incidence angles are known, and the image file has been adjusted to take both angles into account. Otherwise, radar brightness is correct, which should be adopted as the standard both as concept and as working terminology. This has been endorsed by the CEOS SAR Calibration Working Group.
This paper evaluates a method to obtain the slope-corrected
normalized radar cross section (σ<sup>0</sup>) using SAR
interferometry. An error model on a slope correction factor has been
derived as a function of a scene coherence and a baseline distance. The
model has been verified using JERS-1 SAR data and a ground terrain
model. Analyses have shown that the proposed method can correct the
radiometric slope effect within an accuracy of 0.3 dB if the
interference condition is satisfied
A study was conducted to assess the potential of combined imagery
from the existing European and Japanese orbitar synthetic aperture radar
(SAR) systems, ERS-1 (C-hand, VV-polarization) and JERS-1 (L-band,
HH-palarization), for regional-to-global-scale vegetation
classification. For seven test sites from various ecoregions in North
and South America, ERS-1/JERS-1 composites were generated using
high-resolution digital elevation model (DEM) data for terrain
correction of geometric and radiometric distortions. An edge-preserving
speckle reduction process was applied to reduce the fading variance and
prepare the data for an unsupervised clustering of the two-dimensional
(2D) SAR feature space. Signature-based classification of the clusters
was performed for all test sites with the same set of radar backscatter
signatures, which were measured from well-defined polygons throughout
all test sites. While trained on one-half of the polygons, the
classification result was tested against the other half of the total
sample population. The multisite study was followed by a multitemporal
study in one test site, clearly showing the necessity of including
multitemporal data beyond a level 1 (woody, herbaceous, mixed)
vegetation characterization. Finally, classifications with simulation of
backscatter variations shows the dependence of the classification
results on calibration accuracy and on naturally occurring backscatter
changes of natural surfaces. Overall, it is demonstrated that the
combination of existing orbital L- and C-band SAR data is quite powerful
for structural vegetation characterization
An L-band SAR image from the Japanese JERS-1 satellite has been
analyzed for the effects of local surface orientation relative to the
radar illumination direction. Orthorectification of the SAR imagery and
determination of the local surface orientation is achieved with the aid
of a high resolution digital terrain model. An improved method of
determining scatterer density for a surface in three dimensions is
introduced and used to correct radiometrically the image for terrain
variation. Residual radiometric effects due to surface orientation are
shown to be dependent on the ground cover class. Backscatter from the
indigenous forest was more isotropic than that from the farmland. As
accurate registration was required for this study, a method for
identifying control points in the rectified imagery is described which
alleviated the difficulty of identifying them in the raw image
The brightness in a SAR image is affected by topographic height
variations due to (1) the projection between ground and image
coordinates, and (2) variations in backscattering coefficient with the
local scattering geometry. This paper derives a new equation for (1),
i.e. the radiometric slope correction, based on a calibration equation
which is invariant under a coordinate transformation. An algorithm is
described to obtain the slope correction from a SAR interferogram, which
also enables retrieval of the full scattering geometry. Since the SAR
image and interferogram are derived from the same data set, there is no
need to match the image with the calibration data. There is also no need
for phase unwrapping since the algorithm only uses the fringe
frequencies. A maximum-likelihood estimator for the fringe frequency is
analyzed and the algorithm is illustrated by processing ERS-1 SAR data.
The example demonstrates that the spatial resolution and calibration
error are adequate for most applications
Synthetic aperture radar (SAR) images of the Earth's terrestrial
surface contain geometric and radiometric image effects which are caused
by varying terrain elevation and slope. The radiometric effects tend to
mask signal variations caused by other physical variables such as soil
moisture and surface vegetation type, which are known to influence SAR
backscatter signals. As a result, raw SAR images are of limited use in
classifying surface vegetation type or quantifying the spatial
distribution of soil moisture in regions of terrain relief, The authors
present a technique for removing radiometric terrain effects from SAR
images. Image correction was carried out in two steps. First, an
existing modeling package was used in combination with digital elevation
data in order to map the raw image pixels onto a geodetic coordinate
system, thereby removing the geometric portion of the image distortion.
Radiometric effects were then removed with the aid of a backscatter
model which treats the reflected radiation as a combination of
diffuse-Lambertian and specular components. Parameters in the
backscatter model were determined by comparing two C-band SAR images of
a test area in a region of Arctic tundra which were taken from ascending
and descending orbit tracks of the ERS-1 satellite. The ascending and
descending images displayed reductions in pixel value variance of 30%
and 13%, respectively, after processing. Direct comparison of the two
test area images reveals a dramatic improvement in image similarity
after processing
During normal synthetic aperture radar (SAR) processing, a flat
Earth is assumed when performing radiometric corrections such as antenna
pattern and scattering area removal. The authors examine the effects of
topographic variations on these corrections. Local slopes will cause the
actual scattering area to be different from that calculated using the
flat Earth assumption. It is shown that this effect may easily cause
calibration errors larger than a decibel. Ignoring the topography during
antenna pattern removal may also introduce errors of several decibels in
the case of airborne systems. The effect of topography on antenna
pattern removal is expected to be negligible for spaceborne SARs. The
authors show how these effects can be taken into account if a digital
elevation model is available for the imaged area. The errors are
quantified for two different types of terrain, a moderate relief area
near Tombstone, AZ, and a high relief area near Oetztal in the Austrian
Alps. The authors show errors for two well-known radar systems, the
C-band ERS-1 spaceborne radar system and the three frequency NASA/JPL
airborne SAR system (AIRSAR)
Synthetic aperture radar (SAR) images reveal radiometric image
distortions that are caused by terrain undulations. The authors present
the results of a study extracting and investigating the various
components of these terrain influences. An imaging model, is set up for
the geometric rectification of the SAR image and for a reconstruction of
the imaging geometry. A prerequisite for the setup of this model is the
use of a digital elevation model. Eight different geometric parameters
are derived and investigated for their influence on grey-value
variations in the geocoded SAR image. Image grey-value variations of
three major land-use classes-forest, agricultural land, and
urban/suburban areas-are examined. Empirical models of the
SAR-backscatter variations are used to describe the relations between
image grey values and various geometric parameters
The combined effects of topography, slope, look angle, and aspect
on C -band synthetic-aperture radar (SAR) data on the
radiometric quality of SAR images in a region of moderate relief are
examined. A correction method was used to attenuate the change of
illumination across the swath due to the antenna pattern. Ground data
were integrated into the analysis using a digital terrain model (DTM).
Correction functions based on the cosine of the incidence angle were
applied to the thematic classes and to the grouped classes in order to
reduce the effects related to topography. It was found that it is
possible to eliminate part of the radiometric variations created by
moderate topographic relief. After the corrections were applied, a
reduction was noted of the variance in the radiometric values of the
spectral signatures of the cover types, which ranged between 3.03% and
9.47%, depending on the correction function used
Terrain influences on SAR backscatter around Mt A plea for radar brightness
- D Pairman
- S Belliss
- S Mcneill
- R Bamler
D. Pairman, S. Belliss, and S. McNeill, " Terrain influences on SAR
backscatter around Mt. Taranaki, New Zealand, " IEEE Trans. Geosci.
Remote Sens., vol. 35, no. 4, pp. 924–932, Jul. 1997.
[11] K. Raney, A. Freeman, B. Hawkins, and R. Bamler, " A plea for radar
brightness, " in Proc. IGARSS, Pasadena, CA, 1994, pp. 1090–1092.
Automated tie point retrieval through heteromorphic image simulation for spaceborne SAR sensors
- D Small
- S Biegger
- D Nesch
Radiometric terrain correction incorporating local antenna gain
- D Small
- M Jehle
- E Meier
- D Nesch
Terrain-corrected Gamma: Improved thematic land-cover retrieval for SAR with robust radiometric terrain correction
- D Small
- N Miranda
- L Zuberbühler
- A Schubert
- E Meier
D. Small, N. Miranda, L. Zuberbühler, A. Schubert, and E. Meier,
"Terrain-corrected Gamma: Improved thematic land-cover retrieval for
SAR with robust radiometric terrain correction," in Proc. ESA Living
Planet Symp., Bergen, Norway, Jul. 2010, ESA Special Publication
SP-686, 8 p.
Information on ALOS PALSAR Products for ADEN Users
Information on ALOS PALSAR Products for ADEN Users, ESA,
Frascati, Italy, Apr. 5, 2007, ALOS-GSEG-EOPG-TN-07-0001.