Japanese Earth Research Articles

Evaluation of JERS-1 SAR Images From a Coastal Wind Retrieval Point of View

Wind retrieval from Japanese Earth Resources Satellite-1 (JERS-1) synthetic aperture radar (SAR) using an L-band model function in coastal regions is evaluated. It is known that JERS-1 SAR has excessive ambiguities. This paper also gives a quantitative evaluation of excessive ambiguities in coastal scenes of JERS-1 SAR. First, focusing on the cases where wind blows from the shore in Sagami Bay, we investigate phenomena of wind speed increase with offshore distance using European Remote sensing Satellite-1 (ERS-1) SAR-derived wind speeds. The relation between wind speed and offshore distance is well formulated, which indicates the transition of the atmospheric boundary layer from land to sea surface. Wind speeds derived from JERS-1 SAR should be overestimated due to the excessive ambiguity. Then, for observation time of each JERS-1 SAR capturing the cases that wind blows from the shore in Sagami Bay, the expected wind speed growth profile is derived from the wind speed growth formula and an in situ wind observation of Hiratsuka Experiment Station. We convert the wind-speed profile into the sigma-0 profile by an L-band model function. Finally, the profiles of JERS-1 SAR-observed and the estimated sigma-0 are compared, and the excessive ambiguity is estimated as the difference between them. As a result, the dynamic range of first azimuth ambiguity is as large as that of the wind-relating signal from the ocean surface. Moreover, higher order azimuth ambiguities and range ambiguity also may have a significant impact on near-shore wind retrieval.

Updated repeat orbit interferometry package released

RO1_PAC V2.3, a Repeat Orbit Interferometry package that allows topographic and surface change researchers to apply Interferometric Synthetic Aperture Radar (InSAR) methods, is now freely available to the community InSAR is the synthesis of conventional SAR and interferometry techniques that have been developed over several decades in radio astronomy and radar remote sensing. In recent years, it has opened entirely new application areas for radar in the Earth system sciences, including topographic mapping and geodesy.RO1_PAC, developed primarily to work with European Remote Sensing (ERS) satellite radar data, currently supports ERS‐1, ERS‐2, and Japanese Earth Resources Satellite (JERS) radar data, and is configurable to work with “strip‐mode” data from all existing satellite radar instruments. The first release of RO1_ PAC (V1.0) was made quietly in 2000, and roughly 30 groups in the academic and research community currently use it.

Illuminating earth's interior through advanced computing

Today's computational strategies for modeling Earth's interior structure and dynamics come from high-performance computing systems in the US and others such as the Japanese Earth Simulator. Modeling efforts currently underway focus on problems such as geodynamo and earthquake modeling. Space-based measurement and observational technologies are revolutionizing our understanding of the solid Earth and revealing subtle changes that occur on regional and global scales. Understanding these complex processes requires large global data sets and sophisticated computational models coupled with the necessary associated computational infrastructure. The authors discuss how the rapidly increasing availability of data and a more robust and pervasive computational infrastructure could combine to give us new opportunities to understand this complex system.

Radar remote sensing for monitoring of dynamic ecosystem processes related to biogeochemical exchanges in tropical peatlands

Peatlands are an important multipurpose ecosystem, supporting huge quantities of biomass and peat soil carbon. A time series of Japanese Earth Resource Satellite-1 (JERS-1) L-band Synthetic Aperture Radar (SAR) data was employed to monitor two dynamic ecosystem processes; deforestation and inundation patterns. Using a change detection analysis for three images acquired during dry seasons of 1994, 1997 and 1998, we detected the deforestation that has occurred in the region due to the anthropogenic and natural causes. At a threshold of ±2 dB change in backscattering response, an area of about 98 km2 of these forests was found to have been cleared during 1994–1997 for conversion to cultivatable lands. However, the agricultural crops miserably failed to grow on these cleared lands because of the adverse water and soil chemistry conditions. The deliberate draining of these lands, by laying and extension of a huge network of canals, created congenial ecological conditions for the spread of forest fires, particularly during the 1997 El Nino period. An area of 250 km2 of forests was thus detected to have been destroyed by these fires between September 1997 and January 1998. These deforested lands are rapidly regenerating since their abandonment and the regenerating carbon stocks were simulated using the CENTURY ecosystem dynamics model. Furthermore, the L-band SAR was able to detect the pixel-wise seasonal and spatial inundation information for particular forest types where the transmissivity of the L-band SAR signals was quite significant. These forest types corresponded to comparatively low biomass areas. The SAR derived information about these two important dynamic processes would be useful for improving the accuracy of modelling the spatial and temporal distribution of the carbon and other trace gases in these ecosystems.

Dual-season mapping of wetland inundation and vegetation for the central Amazon basin

Wetland extent was mapped for the central Amazon region, using mosaicked L-band synthetic aperture radar (SAR) imagery acquired by the Japanese Earth Resources Satellite-1. For the wetland portion of the 18×8° study area, dual-season radar mosaics were used to map inundation extent and vegetation under both low-water and high-water conditions at 100-m resolution, producing the first high-resolution wetlands map for the region. Thematic accuracy of the mapping was assessed using high-resolution digital videography acquired during two aerial surveys of the Brazilian Amazon. A polygon-based segmentation and clustering was used to delineate wetland extent with an accuracy of 95%. A pixel-based classifier was used to map wetland vegetation and flooding state based on backscattering coefficients of two-season class combinations. Producer's accuracy for flooded and nonflooded forest classes ranged from 78% to 91%, with lower accuracy (63–65%) for flooded herbaceous vegetation. Seventeen percent of the study quadrat was occupied by wetlands, which were 96% inundated at high water and 26% inundated at low water. Flooded forest constituted nearly 70% of the entire wetland area at high water, but there are large regional variations in the proportions of wetland habitats. The SAR-based mapping provides a basis for improved estimates of the contribution of wetlands to biogeochemical and hydrological processes in the Amazon basin, a key question in the Large-Scale Biosphere–Atmosphere Experiment in Amazônia.

Measurements of soil compaction rate by using JERS-1 SAR and a prediction model

The soil compaction rate in a reclaimed land has been estimated by using the Japanese Earth Resources Satellite 1 (JERS-1) synthetic aperture radar (SAR) two-pass differential interferometry (DInSAR) and magnetic probe extensometer (an accuracy of /spl plusmn/1 mm) from 42 ground stations. Twelve JERS-1 SAR interferometric pairs were acquired in the study area. We applied a soil compaction prediction model based on a hyperbolic method using the ground truth data and the DInSAR measurements. The hyperbolic model fitted well to the ground measurements with an rms error of 1.65 cm. The rms error of the model driven by DInSAR measurements was 2.24 cm. The two model results agreed well, showing a difference of 8.1% (12.9 cm) in permanent settlement that is defined as the amount of vertical subsidence as time goes to infinity. A two-dimensional (2-D) subsidence map was constructed from five qualified pairs. The correlation coefficient R between the radar measurements and the ground truth data was 0.87 with an rms error of 1.44 cm. The results demonstrate that the L-band DInSAR combined with a prediction model is useful for geotechnical applications.

JERS SAR interferometry for land subsidence monitoring

In this paper, the potential of L-band repeat-pass differential synthetic aperture radar (SAR) interferometry for land subsidence monitoring is evaluated using Japanese Earth Resources Satellite (JERS) SAR data. Bologna, Mexico City, and the Ruhrgebiet are selected as application sites representing slow to fast deformation velocities. The investigation includes feasibility aspects such as data availability, the temporal decorrelation over different landcover classes and the range of useful spatial baselines, an analysis of the achieved deformation accuracy, and considerations on the complementarity to European Remote Sensing satellite (ERS) SAR interferometry and leveling surveys. In spite of the rather limited data availability, land subsidence maps could be generated for the three selected application sites. In contrast to ERS C-band SAR data, JERS L-band interferometry permitted the retrieval of subsidence values over vegetated areas and forest when using interferograms of less than one year acquisition time interval and short baseline. In addition, the longer L-band wavelength was found to be superior in the case of large deformation gradients that lead to phase-unwrapping problems in C-band interferometry.

Multitemporal jers repeat-pass coherence for growing-stock volume estimation of siberian forest

Multitemporal radar data from the Japanese Earth Resources Satellite (JERS) satellite from the period 1993 to 1998 have been used to investigate if L-band interferometric coherence with a 44-day temporal baseline is suitable for estimations of growing-stock volume in boreal forest. Two forest regions north of Krasnoyarsk in Siberia have been used as test areas. Seasonal variations in the repeat-pass coherence have been studied, and a comparison with C-band coherence from the European Remote sensing Satellite 1 and 2 (ERS-1/2) tandem missions in 1997 and 1998 has been done. JERS coherence from the winter shows a clear correlation with the forest growing-stock volume. For the summer scenes, the spread in the values is too large to give reliable results. Acquisitions from the spring and fall show large problems with decorrelation caused by temporal changes. The results indicate potential of repeat-pass interferometric L-band coherence in winter, as will be provided by the forthcoming Advanced Land Observing Satellite/Phased Array type L-band Synthetic Aperture Radar (ALOS/PALSAR) to map growing-stock volume in Siberia and boreal forests.

Multitemporal repeat-pass sar interferometry of boreal forests

Multitemporal European Remote Sensing satellites 1 and 2 (ERS-1/2) and the Japanese Earth Resources Satellite 1 (JERS-1) interferometric synthetic aperture radar (InSAR) data from a boreal forest test site in Sweden (stem volumes up to 335 m/sup 3//ha, equivalent to above-ground dry biomass of /spl sim/200 tons/ha) are studied in order to estimate stem volume using coherence and backscatter. The changes of JERS-1 backscatter and ERS-1/2 tandem coherence between images are consistent over the area studied, in contrast to ERS-1/2 backscatter. A model-based regression analysis has been performed, and the use of the model for inversion is discussed and compared with other approaches found in the literature. The model parameters are discussed in terms of their relation to wind speed and temperature. Results from the different acquisitions are combined to improve the stem volume estimation. The accuracy in terms of rms error (RMSE) for standwise estimated stem volume is /spl ap/10 m/sup 3//ha using ERS-1/2 coherence. Using backscatter and coherence from JERS-1 we obtain an RMSE of /spl ap/30-35 m/sup 3//ha. Finally, conditions for accurate retrieval of stem volume using multitemporal InSAR observations are discussed. We conclude that C- and L-band repeat-pass InSAR can provide stem volume estimates in boreal forests with accuracies similar to those of standard in situ measurements.

An L-band geophysical model function for SAR wind retrieval using JERS-1 SAR

An L-band geophysical model function is developed using Japanese Earth Resources Satellite-1 (JERS-1) synthetic aperture radar (SAR) data. First, we estimate the SAR system noise, which has been a serious problem peculiar to the JERS-1 SAR. It is found that the system noise has a feature common in all the SAR images and that the azimuth-averaged profile of noise can be expressed as a parabolic function of range. By subtracting the estimated noise from the SAR images, we can extract the relatively calibrated ocean signals. Second, using the noise-removed SAR data and wind vector data from the NASA Scatterometer and buoys operated by the Japan Meteorological Agency, we generate a match-up dataset, which consists of the SAR sigma-0, the incidence angle, the surface wind speed, and wind direction. Third, we investigate the sigma-0 dependence on incidence angle, wind speed, and wind direction. While the incidence angle dependence is negligible in the present results, we can derive distinct sigma-0 dependence on wind speed and direction. For wind speeds below 8 m/s, the wind direction dependence is not significant. However, for higher wind speeds, the upwind-downwind asymmetry becomes very large. Finally, taking into account these characteristics, a new L-band-HH geophysical model function is produced for the SAR wind retrieval using a third-order harmonics formula. Resultant estimates of SAR-derived wind speed have an rms error of 2.09 m/s with a negligible bias against the truth wind speed. This result enables us to convert JERS-1 SAR images into the reliable wind-speed maps.

Large-scale mapping of boreal forest in SIBERIA using ERS tandem coherence and JERS backscatter data

Siberia's boreal forests represent an economically and ecologically precious resource, a significant part of which is not monitored on a regular basis. Synthetic aperture radars (SARs), with their sensitivity to forest biomass, offer mapping capabilities that could provide valuable up-to-date information, for example about fire damage or logging activity. The European Commission SIBERIA project had the aim of mapping an area of approximately 1 million km 2 in Siberia using SAR data from two satellite sources: the tandem mission of the European Remote Sensing Satellites ERS-1/2 and the Japanese Earth Resource Satellite JERS-1. Mosaics of ERS tandem interferometric coherence and JERS backscattering coefficient show the wealth of information contained in these data but they also show large differences in radar response between neighbouring images. To create one homogeneous forest map, adaptive methods which are able to account for brightness changes due to environmental effects were required. In this paper an adaptive empirical model to determine growing stock volume classes using the ERS tandem coherence and the JERS backscatter data is described. For growing stock volume classes up to 80 m 3/ha, accuracies of over 80% are achieved for over a hundred ERS frames at a spatial resolution of 50 m.

A physical method to analyse scattered waves from burnt coal seam and its application to estimate thickness of fire scars in central Borneo using L-Band SAR data

A physical method was conducted to analyse scattered waves from burnt coal seam in order to estimate the thickness of fire scars. The model was composed of three media namely; air, burnt coal seam and peat. For computation purposes, the equivalent circuit of this model was conducted using a classical transmission line circuit method. The relationship between backscattering coefficient and thickness of burnt coal seam was defined in terms of the logarithm of the reflection coefficient (in power). The analysis result was confirmed by simulation using a Finite Difference Time Domain (FDTD) method. The simulation was performed using a two-dimensional (2-D) finite-difference model for scattered waves from a burnt coal seam. The model used the equations of scattered electromagnetic fields that were derived from Maxwell's equations. A Mur method was used to surround the simulation space and absorb the outward travelling waves. Analysis and simulation results were similar. Subsequently, the model was applied to estimate the thickness of burnt coal seam in central Borneo fire events, that occurred in 1997, using a Japanese Earth Resources Satellite (JERS-1) SAR data. Results showed that fire scars in the study area reached 0.52 m in depth (thickness). This agrees with ground measurements.

Disturbance recognition in the boreal forest using radar and Landsat-7

As part of a Siberian mapping project supported by the National Aeronautics and Space Administration (NASA), this study evaluated the capabilities of radars flown on the European Remote Sensing Satellite (ERS), Japanese Earth Resources Satellite (JERS), and Radarsat spacecraft and an optical sensor enhanced thematic mapper plus (ETM+) on-board Landsat-7 to detect fire scars, logging, and insect damage in the boreal forest. Using images from each sensor individually and combined, an assessment of the utility of using these sensors was developed. Transformed divergence analysis revealed that Landsat ETM+ images were the single best data type for this purpose. However, the combined use of the three radar and optical sensors did improve the results of discriminating these disturbances.

Estimation of burnt coal seam thickness in central Borneo using JERS-1 SAR data

We analysed the interaction of microwaves with a burnt coal seam. This analysis approximated a burnt coal seam as a microwave absorber. The impedance of an incident wave with horizontal polarization (transverse magnetic mode) was derived in order to calculate the relationship between burnt coal seam thickness and backscattering coefficient. The result was confirmed by simulating scattering on a burnt coal seam using the Finite Difference Time Domain (FDTD) method. Both were similar. This relationship was used to estimate burnt coal seam thickness in central Borneo using Japanese Earth Resources Satellite (JERS-1) SAR data. Estimation results and ground data were similar.

Accuracy assessment of a large-scale forest cover map of central Siberia from synthetic aperture radar

Russia's boreal forests host 11% of the world's live forest biomass. They play a critical role in Russia's economy and in stabilizing the global climate. The boreal forests of central and western Siberia represent the largest unbroken tracts of forest in the world. The European Commission funded SIBERIA project aimed at producing a forest map covering an area of 1.2 million square kilometres. Three synthetic aperture radars (SAR) on board the European remote sensing satellites ERS-1 and ERS-2 and the Japanese Earth resources satellite JERS-1 were used to collect remote sensing data. Radar is the only sensor capable of penetrating cloud cover and imaging at night. An adaptive, model-based, contextual classification to derive ranked total growing stock volume classes suitable for large-scale mapping is described. The accuracy assessment of the Siberian forest cover map is presented. The weighted coefficient of agreement κw is calculated to quantify the agreement between the classified map and the reference data. First, the classified map is compared with Russian forest inventory data (κw = 0.72). The inherent uncertainty in the forest inventory data is simulated by allowing for fuzziness. The effect of uncertainty on the unweighted coefficient of agreement κ is stronger than that on the weighted coefficient of agreement κw. Second, the map is compared with a more reliable, independent posterior ground survey by Russian forestry experts (κw = 0.94). The follow-on project SIBERIA-II started in January 2002 and is striving to develop multisensor concepts for greenhouse gas accounting (www.siberia2.uni-jena.de).

Estimating forest biomass using satellite radar: an exploratory study in a temperate Australian Eucalyptus forest

A study was undertaken to explore the relationship between backscattering coefficients from a Japanese Earth Resources Satellite synthetic aperture radar (JERS-1 SAR) image and aboveground biomass sampled at 12 field plots located in Murramarang National Park, New South Wales, Australia. This is the first such investigation in Australian Eucalyptus forests. From the field survey we obtained tree and coarse woody debris (CWD) measurements for eight forested, two paddock tree and two grass (effectively zero biomass) plots. Total aboveground and component biomass were estimated for all plots using allometric equations. Live aboveground woody biomass ranged from 0 to 610 t ha −1. The mean JERS-1 SAR backscattering coefficients for the field plot areas ranged from −12.4 to −7.0 dB. The results show positive linear trends between backscattering coefficients and the biomass components of dry Eucalyptus forest. The strongest trend was produced with small branch (2 cm) biomass estimates ( r 2=0.84). The biomass saturation level for the JERS-1 SAR data may be higher than estimated by other studies (possibly up to 600 t ha −1), although this trend was not statistically confirmed due to the small sample size ( n=8). The results suggest that estimation of forest biomass for biomass inventories in Australia might be possible using satellite radar data when landscape characteristics such as topography, surface water, and forest structure are taken into account.

JERS-1 SAR mosaics of Southeast Asia using calibrated path images

We have produced a continent-scale, high-resolution radar mosaic of Southeast Asia using L-band Synthetic Aperture Radar (SAR) data collected by Japanese Earth Resources Satellite-1 (JERS-1). This 100 m resolution, 16-bit mosaic dataset of the backscattering coefficients has a calibration accuracy of 1.1 dB, geometric accuracy of 406 m, and an image quality similar to those of Amazon and central Africa previously produced by NASA Jet Propulsion Laboratory (JPL) and the Joint Research Center (JRC) of the European Commission (EC). This dataset may be beneficial in Earth observation research, especially for forest monitoring, because the L-band signal is sensitive to forest characteristics. This paper introduces a new mosaicking method that was developed to assemble the SAR path images into a large-scale mosaic, a way to recalibrate the radiometric imbalance across paths, and the validation.

The use of spaceborne radar data to model inundation patterns and trace gas emissions in the central Amazon floodplain

River floodplains are the dominant wetland habitat in the Amazon river basin, providing important habitation for aquatic flora and fauna, and playing a key role in sustaining regional fish production. The annual inundation pulse has been identified as the dominant environmental factor affecting aquatic biota on the floodplain, and the characteristics of this pulse, in terms of timing, duration and amplitude, vary spatially on the floodplain as a function of fluctuations in river stage height and topography. River floodplains are furthermore globally significant sources of methane (CH 4 ) and other trace gases essential to climate regulation. Refined information on wetland distributions and dynamics are currently needed to improve estimates of habitat availability and to calculate regional contributions of trace gases, especially CH 4, to the troposphere. This paper describes how multitemporal time series of spaceborne L-band Synthetic Aperture Radar (SAR) data from the Japanese Earth Resource Satellite 1 (JERS-1) were used to generate a model of the spatial and temporal variation of inundation on the floodplain of a typical black water river in the Central Brazilian Amazon and how this model was utilized, together with in situ measurements of river stage heights and CH4 fluxes, to model regional estimates of CH 4 emissions. We also demonstrate how a JERS-1 SAR time series can be used to map the spatial variation of flood duration on the floodplain, a key factor controlling local variations in plant biodiversity. For both applications, the availability of adequate time series of satellite sensor data is the prime factor affecting the reliability and accuracy of the flood models and the spatial details of the flood duration map. The availability of in situ data, especially daily river height measurements, was also critical for the development of the flooding model and for the subsequent decoupling of the model from the satellite sensor data.

Vegetation mapping using JERS-1 SAR mosaic for northern Australia

The Japanese Earth Resources Satellite (JERS-1) Synthetic Aperture Radar (SAR) L-band HH polarization data over northern Australia were acquired in 1996 as part of the Global Rainforest Mapping (GRFM) experiment by the National Space Development Agency of Japan (NASDA). The data were mosaiced by NASDA and re-sampled to a resolution of 100 m 2 100 m per pixel. As the mosaiced data is a preliminary product, the problem of calibration, including the removal of effects of look angle variations within each scene and the relative calibration among multiple scenes due to different acquisition dates, has not been dealt with sufficiently, which greatly degrades the quality of the product. It is understood that precise calibration in mosaiced data is of necessity in order to retrieve accurate geophysical parameters. Nevertheless, the preliminary mosaiced data is used in the paper to investigate the potential of using SAR data for vegetation mapping in a regional or global scale. Pixel-based mapping using SAR data has shown limited success due to speckle. However, area analysis techniques allow the computation of reliable statistical and textural measures, thus providing a base for pursuing radar classification. In this paper, an image is first segmented using a Gaussian Markov random field model, so that area analysis can be pursued. An unsupervised classification method is developed in which as many as 12 statistical and textural measures are used to group segments with similar textural measures into the same class in a fashion of maximum likelihood. For a pre-given number of 50 as the final number of classes in the unsupervised classification iteration, the average B- distance among these 50 classes is 1.96 when all 12 textural measures are used to compute the distance.

The JERS-1 Amazon Multi-season Mapping Study (JAMMS): Science objectives and implications for future missions

In late September 1995, the National Space Development Agency of Japan (NASDA) began a new phase of operations for the Japanese Earth Remote Sensing satellite (JERS-1) Synthetic Aperture Radar (SAR)--the Global Rain Forest Mapping (GRFM) project. The first rainforest area to be mapped was the Amazon basin, between September and November of that year (the low flood season for much of the region), in support of the JERS-1 Amazon Multi-season Study (JAMMS), sponsored by NASA. This data acquisition was repeated 6 months later to acquire a second map of the Amazon, during the high flood season in May/June of 1996. The main objective of the JAMMS project was to generate a map of inundation over the Amazon basin by comparing data from the high- and low-flood seasons. Most of the data collected during these two phases of the JAMMS project, a total of ∼5000 frames of data, was received and processed by the Alaska SAR Facility (ASF), then sent to the Jet Propulsion Laboratory (JPL) and NASDA for post-processing and analysis. The quality of the data processed by ASF for the JAMMS project has proved to be exceptional. This paper is a summary of the JAMMS project, which has resulted in a scientific dataset of very high value--a multi-season snapshot of one of the most difficult areas on Earth to monitor, even from space.