最新刊期
卷 27 ， 期 4 ， 2023
-
摘要：In the past 10 years, the basic research field of remote sensing science in China has been continuously expanding. A number of research achievements have been made, and many excellent talents have been cultivated. The steady improvement of the basic research level of remote sensing science has not only greatly promoted the comprehensive development of geographical science but has also played a significant role in serving the main battlefield and major strategies of the national economy. This study analyzes the application and funding of NSFC projects in remote sensing science from 2013 to 2022, summarizes research hotspots and major achievements, and discusses the development challenges of basic research in remote sensing science in China. The results show that in the past decade, the application codes of fund projects related to remote sensing have been continuously optimized and have gone through four versions. These versions have been updated rapidly and have played an obvious role in driving and promoting the development of the discipline. Moreover, they have better adapted to the development law of remote sensing science in China. During this period, the number of applications for fund projects in the field of remote sensing science continued to increase, the quality of applications increased tremendously, and the number of grants received also increased slightly. The research hotspots of the funded projects mainly focus on four frontier fields: hyperspectral remote sensing, multispectral remote sensing, thermal infrared remote sensing, and visible near-infrared remote sensing. With the support of the National Natural Science Foundation of China (NSFC), fruitful research achievements have been made in the field of remote sensing science. The rise and rapid development of remote sensing technology have become the most important source of big data in the current geographical science research. Cross integration with artificial intelligence has led to changes in the research paradigm of geographical science. It has created a data-driven automatic extraction of geographical knowledge and laws in the field of geographical science, promoted the interpretable causal analysis of geographical events and phenomena, and improved the human understanding and prediction ability of geographical problems. It supports the construction of the scientific paradigm and technical system for spatio-temporal big data analysis. In the follow-up research of remote sensing science, the basic research of remote sensing mechanism must be strengthened. Moreover, the ability of sensor research and development must be improved, and national major needs must be supported. In addition, research on domestic high-resolution remote sensing satellites must be improved to promote the healthy and sustainable development of remote sensing science. In this field, comprehensive, exploratory, and forward-looking project applications are encouraged. Cutting-edge cross research is encouraged around national major needs, such as “ecological civilization construction,” “the Belt and Road,” “rural revitalization,” and “land and space planning.” Proceeding from the urgent and long-term needs of the country, NSFC will continue to strengthen the basic research of remote sensing science and promote the development of remote sensing science in China during the “Fourteenth Five Year Plan” period and in the long term.关键词：National Natural Science Foundation of China;remote sensing science;funding situation;research findings;development challenges- 1224
- |
- 948
- |
- 1
发布时间：2023-05-10 -
摘要：The Global LAnd Surface Satellite (GLASS) products suite includes high-level satellite products of land surface essential variables from multiple universities and research institutes. Producing the GLASS products suite has been undertaken since 2010. The suite spans from the initial five products to the current 16 products, which are generated mostly from the Advanced Very High-Resolution Radiometer and/or Moderate Resolution Imaging Spectroradiometer data. Some of the products have been previously introduced in the literature, and this study provides an update on the algorithm developments, validation accuracies, and their typical applications in all 16 products. This study also describes the Hi-GLASS products at 30 m resolution and some perspectives for further future improvement and development of the GLASS products.Estimating land surface variables from satellite observations is an “ill-posed” inversion problem. For each pixel, the number of multispectral bands is usually smaller than the number of environmental variables, and the values of many spectral bands are highly correlated. Some novel solutions have been proposed to address the insufficient information in generating reliable GLASS products. We can identify at least four approaches. The first is based on the temporal signature of the satellite observations. A typical example is the MODIS Leaf Area Index (LAI) and the Fraction of Absorbed Photosynthetically Active Radiation (FAPAR) products generated using two-year observations simultaneously. The second uses an algorithm ensemble. A typical example is the evapotranspiration product based on integrating five estimation algorithms. The third uses multiple satellite observations. For example, the forest aboveground biomass product is based on optical, Lidar, and microwave data products. The last incorporates the physical model to generate the products, such as the gross primary production product.The GLASS products have several unique features compared with similar products on the market, including the following:(1) Several products are unique, such as the high-resolution (1 km) broadband emissivity and time-series forest aboveground biomass products.(2) Most products have long time series (i.e., over 40 years), while most other similar global products start from approximately the year 2000, with a period of approximately 20 years.(3) The radiation products, covering the world’s land and ocean surfaces, have a spatial resolution of 5 km, which is an order of magnitude higher than other such products in wide use, for example, the Global Energy and Water Exchanges, the Clouds and the Earth’s Radiant Energy System, and the International Satellite Cloud Climatology Project, which have spatial resolutions coarser than 100 km.(4) Several long-time-series global products have the highest spatial resolution in the world, such as 250 m for the LAI, FAPAR, and albedo products and 5 km for snow cover extent. Moreover, the all-weather LST and near-surface air temperature products have a 1-km resolution.(5) GLASS products are of high quality and accuracy.Over 2000 peer-reviewed papers based on the GLASS products have been published. Their applications are distributed in many scientific disciplines and societal benefits areas.We will continue to improve the quality and accuracy of the existing GLASS products and produce more GLASS products with higher spatial resolutions.关键词：satellite remote sensing;land surface;GLASS produces;radiation and energy budget;carbon cycle- 1107
- |
- 585
- |
- 0
发布时间：2023-05-10 -
摘要：A global consensus has been made to promote carbon emission reduction in response to global warming caused by the increase of greenhouse gases, such as CO2 and CH4. Spaceborne observation has the characteristics of large observation space and continuous observation time, which are among the main means of observing greenhouse gases at present. The establishment of a sound carbon monitoring system and spaceborne passive remote sensing of major greenhouse gases in the atmosphere will help to evaluate the impact of the greenhouse effect and guide human greenhouse gas emission activities, which is of great significance to human society.Active satellite-borne remote sensing of greenhouse gases become successful through proper planning. Among the spaceborne passive remote sensing payloads of greenhouse gases successfully applied in orbit, three technical systems are mainly included: Michelson interference spectroscopy represented by GOSAT (Greenhouse gases Observing SATellite) and GAS (Greenhouse gases Absorption Spectrometer); grating spectroscopy represented by OCO (Orbiting Carbon Observatory) and ACGS; and spatial heterodyne interference spectroscopy represented by GMI (Greenhouse gases Monitoring Instrument). This study focuses on the analysis of these three typical technology systems and compares the advantages and disadvantages of different detection technologies. At the same time, comprehensive satellite payloads for the detection of greenhouse gases include IMG (Interferometric Monitor for Greenhouse gases), SCIAMACHY (SCanning Imaging Absorption SpectroMeter CHartographY), AIRS (Atmospheric Infrared Sounder), ACE-FTS (Atmospheric Chemistry Experiment Fourier Transform Spectrometer), IASI (Infrared Atmospheric Sounding Instrument), and CrIS (Cross-track Infrared Sounder). Moreover, projects for the spaceborne passive remote sensing of greenhouse gases that include GeoCarb (Geostationary Carbon Observatory) and Copernicus CO2 Monitoring Mission are introduced.To meet the needs of the next generation of spaceborne remote sensing of greenhouse gases, combined with the in-orbit performance of the GMI on GF-5 and the research progress of the new spatial heterodyne interference imaging spectroscopy technology, the feasibility of further achieving high spatial resolution on the basis of hyperspectral resolution and high signal-to-noise ratio is analyzed. This study proposes a payload technology scheme with high timeliness and regional carbon monitoring capability of different subdivisions, which will provide a technical basis for the development of the next generation of detection payload for greenhouse gases.Reviewing the development process of technologies for the spaceborne detection of greenhouse gases, six development trends of spaceborne passive remote sensing payloads of greenhouse gases are summarized: (1) specialization of detection load; (2) improvement of detection sensitivity; (3) wide width and high spatial resolution; (4) integration of multiple observation modes; (5) systematization of high/medium/ow orbit monitoring; (6) miniaturization of detection load.关键词：greenhouse gases;Passive remote sensing;Carbon Monitoring;satellite payload;interference imaging spectroscopy- 828
- |
- 362
- |
- 2
发布时间：2023-05-10
Review
-
摘要：The Geophysical Model Function (GMF) is indispensable in space-born scatterometer wind retrieval data processing because it is the main factor that dominates the error characteristics of retrieved winds. In this study, a new Ku-band GMF (named NSCAT-5.HY-2), which includes Sea Surface Temperature (SST) dependence, is developed for improved HY-2 satellite scatterometer (HSCAT) wind retrieval. Based on the existing NSCAT-4 GMF, the dependence of σ0 on wind speed and direction will be refined, and that of σ0 on SST is then added. The sea surface winds from the C-band advanced scatterometer (ASCAT) onboard the MetOp-B and C satellites are of high quality and have no dependence on SST. Accordingly, the ASCAT winds are used as reference for developing the NSCAT-5.HY-2 GMF. HY-2C and HY-2D are in non-sun-synchronous orbits with a 66.0° inclination, and their equator crossing times shift each orbit. The HY-2C or HY-2D scatterometer represents a unique opportunity to acquire closely collocated HSCAT and ASCAT scatterometer winds at different times of the day. In this study, the wind vector cells between HSCAT and ASCAT are matched by limiting the spatial distance within km and time differences within 45 min. The close collocations of HSCAT and ASCAT winds allow accurate attribution of the different correlated residual geophysical effects of wind speed bias and distribution, SST, sea state, rain, wind variability, and geographical sampling biases. Here, we explore the opportunity of close collocations of HSCAT-C and ASCAT and develop an SST-dependent Ku-band GMF for HY-2 scatterometers. The wind speed corrections as a function of wind speed can be calculated by using the cumulative distribution function matching technique, which aligns the Probability Density Function (PDF) of HSCAT wind speeds with the referenced PDF. During the wind direction modulations, five terms of the Fourier series are used, and the harmonic coefficients are derived utilizing ASCAT wind direction as reference. The variations of σ0 as a function of SST are given as a polynomial expansion for each given wind speed, and the polynomial coefficients are obtained from the observed and simulated radar cross-sections using closely collocated ASCAT winds of either vertical or horizontal polarizations. The validation of the new HSCAT wind products retrieved using NSCAT-5.HY-2 GMF shows clear improvements over those obtained with NSCAT-4 GMF. The consistency between HSCAT and ASCAT winds is much improved by at least 10%, and wind speed differences no longer depend on SST.关键词：Remote sensing of sea surface wind vectors;HY-2 microwave scatterometer;ASCAT scatterometer;geophysical model function;sea surface temperature- 531
- |
- 245
- |
- 1
发布时间：2023-05-10 -
摘要：The interaction between ocean waves and sea ice in the Arctic ocean has received significant attention. However, the study on this issue is significantly limited due to the lack of observation data. Synthetic Aperture Radar (SAR) plays an important role in the research of ocean wave and sea-ice interaction because of its unique capability of imaging sea surface in two dimensions. Sentinel-1 (S1), which consists of Sentinel-1A (S1A) and Sentine-1B (S1B), can cover the entire Arctic Area within 2 days. The Interferometric Wide Swath (IW) mode, one of the main imaging modes of S1 in the Arctic, has been providing SAR images with a high resolution of 10 m. The ocean wave spectra are derived using the S1 IW data. The spectra are likely to provide vital observation for studying the interaction between sea-ice and waves as they present the distribution of wave energy and their variations in different frequencies and directions. Meanwhile, the retrieved ocean wave spectra are an excellent validation data source for SWIM, which is onboard the China-France Oceanography Satellite and provides measurements of ocean wave spectra in the global ocean.In this study, we use the sub-images from the S1 IW image with a size of 1024 pixels × 1024 pixels (10.24 km × 10.24 km) to retrieve ocean wave spectra by using a nonlinear retrieval method (i.e., MPI scheme). The retrieved ocean wave spectra and wave parameters are compared with SWIM measurements acquired at an incidence angle of 10°, and the Significant Wave Height (SWH) is measured at nadir. The SWIM spectrum covers a large area by approximately 70 km × 90 km. To make a comparison, all the SAR sub-image spectra within a SWIM beam coverage are averaged to calculate a new observed SAR spectrum, which is inputted into the MPI inversion scheme. Then, the retrieved SAR ocean wave spectrum is compared with the SWIM spectrum. The footprint size of the SWIM nadir beam is 18 km, which is comparable to the S1 sub-image size. Accordingly, a SAR sub-image is extracted at SWIM nadir, and the corresponding ocean wave spectrum is retrieved. The SWH is calculated by integrating the retrieved wave spectra to compare with the SWIM nadir measurements of SWH. The experiment was carried out using the data acquired in September 2020 in the Greenland Sea and Norwegian Sea, where the ocean waves generated in the North Atlantic and propagating vast distances to the ice-covered area in the Arctic ocean can be frequently observed.Fifty-four ocean wave spectra were retrieved from 25 IW data and are compared with the SWIM slope spectra. The comparison shows that the SAR-retrieved spectra are consistent with SWIM spectra in terms of structure and energy distribution. Good agreements are also found between the integral parameters of the SAR ocean wave spectra and SWIM slope spectra. The comparison yields a bias and an RMSE of 0.11 and 0.71 m for SWH and a bias and an RMSE of -0.52 s and 0.62 s for mean wave period. The comparison of the dominant wave parameters yield a bias of -7.74° and an RMSE of 15.75° for the dominant wave direction and a bias of -0.56 m and an RMSE of 52.73 m for the dominant wavelength. Furthermore, 5075 data pairs of S1-retrieved SWH and SWIM nadir SWH were collocated and compared. The comparison result yields a bias of 0.03 m, an RMSE of 0.48 m, and a correlation of 0.95.The comparison between the S1 retrieved results and the SWIM measurements suggests that ocean wave information can be effectively retrieved from S1 IW data by using the MPI method in the Arctic ocean. Although the MPI method relies on prior information, it is still an effective method for obtaining ocean wave spectra in high resolution. The spectra retrieved from S1 are likely to show the energy attenuation of ocean waves in different frequencies and directions when propagating toward an ice-covered area. This finding will be of great support for the further study on the interaction between sea ice and ocean waves.关键词：remote sensing;Arctic;ocean wave;SAR;CFOSAT- 496
- |
- 250
- |
- 0
发布时间：2023-05-10 -
摘要：Ocean wave is one of the important marine dynamic phenomenon that affect human activities. At present, the main observation means include buoy observation, marine numerical prediction model, and microwave remote sensing observation. However, we cannot conduct large-scale observation by buoy, and the marine numerical prediction model’s result is not measured data. Spectrometers and altimeters in microwave remote sensing instruments can also measure spectral parameters. However, SAR, which has a higher resolution, can provide 2D sea surface information. The Sentinel-1 satellite of ESA and GF-3 satellite independently developed by China are now in orbit, and numerous teams are working to retrieve wave parameters from SAR data of these two satellites. In this work, we compared the wave parameter inversion accuracy of Sentinel-1 SAR Interferometric Wide Swath model and GF-3 SAR strip model based on wave spectrum, which provides a reference for the wide application of GF-3 SAR data.The sea states according to the ERA-5 data of ECMWF are divided into three categories: low, moderate, and high sea states. The sea areas of Hormuz and Malacca Straits of the maritime Silk Road in the Indian Ocean and the coastal waters of the Pacific and Atlantic Ocean are selected as the study areas. Meanwhile, the SAR data of Sentinel-1 and GF-3 satellites under different sea states are selected as the data source. The MPI method is used to retrieve the wave spectrum and wave parameters using the E spectrum as the initial guess. Subsequently, the SAR data inversion results of the two satellites under different sea states are compared with the ERA-5 and buoy wave data. The inversion accuracy of the wave parameters can be verified by calculating the values of the Root Mean Square Error (RMSE) and Scatter Index (SI), and the inversion accuracy of the wave parameters of the two satellites under different sea conditions can be compared.The RMSEs of significant wave height (Hs) retrieved by GF-3 SAR under low, moderate, and high sea conditions are 0.30, 0.34, and 0.48 m, and those of mean wave period (Tm) are 1.02, 0.99, and 0.95 s, respectively, compared with the ERA-5 data. In addition, the RMSE of Hs retrieved by Sentinel-1 SAR under low, moderate, and high sea conditions are 0.30, 0.29, and 0.33 m, respectively, and the RMSEs of Tm are 0.94, 0.51, and 0.64 s, respectively. The RMSEs of Hs and Tm under different sea conditions retrieved by GF-3 SAR are 0.38 m and 0.99 s, and those of Hs and Tm retrieved by Sentinel-1 SAR are 0.31 m and 0.70 s, respectively, compared with the ERA-5 data. The RMSEs of the retrieved Hs and Tm of GF-3 satellite are 0.42 m and 0.94 s, and those of the retrieved Hs and Tm of Sentinel-1 are 0.40 m and 0.91 s, respectively, compared with the buoy data.The SAR wave parameter inversion of Sentinel-1 and GF-3 SAR based on the wave spectrum shows that the inversion results of the two satellites meet the index requirements in this field, and the accuracy of the inversion results of wave spectrum is the same. The strip mode SAR data of GF-3 satellite, China’s first self-developed SAR satellite, has broad prospects in marine research fields.关键词：SAR;GF-3;Sentinel-1;Wave spectrum inversion;Accuracy comparison- 606
- |
- 283
- |
- 1
发布时间：2023-05-10 -
摘要：Oceanic internal waves are widely presented in all levels of the water column in deep oceans and marginal areas. These waves play an important role in seawater energy exchange. The study of oceanic internal waves has important academic values and practical significance in marine resources, marine engineering, and the marine military. The oceanic internal waves are distinct bright and dark stripes in Synthetic Aperture Radar (SAR) images. Those stripes can serve as clues to efficiently identify the oceanic internal waves from SAR images. The growing popularity of computer vision has led to the wide adoption of deep learning for the detection of oceanic features in remote sensing data. In this study, we intend to apply the faster R-CNN, a state-of-the-art deep learning method, to the automatic detection of oceanic internal waves.The faster R-CNN is the most widely used version of the R-CNN family. This algorithm depends on the region proposal algorithms to hypothesize object locations. Based on the bright and dark stripes in the SAR images, a Faster R-CNN-based method is developed for oceanic internal wave detection. First, the oceanic internal waves are manually labeled in the SAR images to serve as the training set. The training set for the detection method contains 5480 SAR images, which are in multi-band, multi-polarization mode, and multi-spatial scale. These images are collected in the South China Sea region from 2001 to 2020. Then, the Faster R-CNN network is trained based on the obtained training set. Meanwhile, the parameters (such as training epochs) are optimized. The transfer learning technic is applied in the training process to transfer information from the previously learned tasks for detecting oceanic internal waves to accelerate the training process and avoid overfitting. The well-trained Faster R-CNN network can be applied by a sliding window on the SAR images to detect the oceanic internal waves. When the boundaries are obscured, the waves may be detected multiple times. In this case, the detection results will be grouped and merged. Finally, the detection results of the oceanic internal waves are acquired and recorded.The evaluations are conducted on multi-source SAR data, showing that the accuracy rate (AP) and recall rate (AR) of the developed method are up to 95.7% and 92.3%, respectively. This method can achieve high accuracy while keeping the false alarm rate relatively low. The experiments on the SAR images with complex ocean conditions also show favorable results.This work developed the approach for the detection of oceanic internal waves. This approach successfully transferred techniques developed in the computer vision field to solve the issues in remote sensing problems. The establishment of this method provides a technical basis for the detection of oceanic internal waves from a large amount of SAR data and further promotes the research of internal wave parameter inversion and dynamic processes. In addition, the proposed method was initially designed for lunar research, but it could be applied to the detection of other oceanic features.关键词：oceanic internal waves;automatic detection;SAR images;deep learning;Faster R-CNN- 657
- |
- 286
- |
- 0
发布时间：2023-05-10 -
摘要：The Surface Water and Ocean Topography (SWOT) mission, which is scheduled to launch in 2022, will carry a Ka-band Radar Interferometer to characterize the ocean mesoscale and submesoscale circulation. The observation errors must be reduced to realize the observation target of the SWOT mission. Among the errors, the baseline inclination error caused by the inaccuracy of the baseline inclination measurement and the phase error induced by the phase mismatch in the process of interferometric imaging are difficult to eliminate. These two errors make the sea surface height measurement in the swath inaccurate and identification of the characteristics of mesoscale phenomena challenging. Therefore, the purpose of this article is to correct phase error and baseline roll error to achieve the observation goal.Both errors are linear with across track distance from nadir. Assuming that all errors, except the phase and roll errors, have met the error budget requirements, a bilateral independent cross-calibration method is proposed to detect and mitigate the two spatial coherence errors. First, the errors of the left and right swaths in the experiment are calculated. Second, the observed values in the cross region are subtracted to reduce the influence of the ocean signal. Finally, the errors are estimated based on their linear correlation with the across track distance. The orbit parameters of SWOT and Jason-2 (before October 2016) provided by Archiving, Verification, and Interpretation of Satellite Oceanographic data are used for along-track sampling. The errors in the 25-day SWOT sample data are estimated by using the self-crossover and crossover (with Jason-2) methods.In the SWOT self-crossover, more than 90% of the error was corrected from 8 cm to 4 cm, and approximately 73% of the intersection-region standard deviation was adjusted from 6 cm to 2 cm.In the SWOT crossover with Jason-2, approximately 86% of the error was corrected from 8 cm to 4 cm, and approximately 55% of the intersection-region standard deviation was adjusted from 6 cm to 2 cm. Based on the absolute value and standard deviation, the error inversion effect of the self-crossover and crossover (with Jason-2) methods on the intersection point is good. Results show that the bilateral independent cross-calibration method effectively estimates the overall error and significantly reduces the error level in the case of low instrument accuracy.The crossover method weakens the influence of a larger portion of the ocean signal, increases the weight of the error in the calculation process, and calculates the error through the matrix. Given that the characteristics that the phase error is different from the baseline inclination error, the left and right swaths are cross-calibrated separately to calculate the gradient slope value in the cross-track direction of the unilateral swath.The calculated gradient slope value can be directly back-calculated to obtain the overall error and eliminate the error. The bilateral independent cross-calibration method not only inherits the ability of the cross-calibration algorithm to correct the baseline inclination error but also scientifically and effectively corrects the phase error. The experiment also suggests that the cross-calibration formula can be optimized based on the error characteristics to estimate other errors with gradient changes in the across track and contributes to error reduction.关键词：cross-calibration;Error-correction;phase error;Baseline roll error;Interferometric imaging altimeter- 337
- |
- 224
- |
- 0
发布时间：2023-05-10 -
摘要：Eddies play an important role in water transport and energy balance in the ocean. Ocean observations show that eddies are not isolated, and cyclonic eddies and anti-cyclonic eddies always form more stable structures: dipole eddies. Dipole eddies have a more evident dynamic modulation compared with monopole eddies. Moreover, dipole eddies enhance the vertical movement of water to ensure that they promote the propagation and distribution of heat, energy, and organic matter in the ocean, which affects the global biochemical process. The parameters’ variations of dipole eddies during propagation must be examined to extensively understand the coupling state of eddies. The quantitative analysis of the modulation effect can provide a reference for exploring the dynamic mechanism of dipole eddies.In this work, the Archiving, Validation, and Interpretation of Satellite Oceanographic (AVISO) merged data from a combination of T/P, Jason-1, Jason-2, Jason-3, and Envisat missions are used to identify and track eddies in the global ocean during 1993—2020. These data have a daily temporal resolution and a 1/4°×1/4° spatial resolution. We established criteria for extracting dipole eddies based on eddy identification and track data: the distance of eddy cores is less than twice the sum of their radii, and their concomitant time is more than 60 days. We analyze the variations of eddies’ parameters to reveal the role of the dipole in modulating eddies based on over 67,500 dipole eddies that we found.The results show that dipoles are distributed in 10°N—60°N and 11°S—66°S. The dipole eddies are composed of eddies at a probability of over 15% and in the strong currents region, the frequency of dipole eddies formation is higher, even more than 35%. Given that the dipole structure influences the movement and propagation of eddies, we demonstrate it by using four dominating parameters of eddies, including amplitude, radius, EKE, and vorticity. Result showed that the dipole structures increase the amplitude by 5%—14%, radius by 2%—9%, and EKE by 4%—13% but suppress the vorticity of eddies by less than 3%. Furthermore, the various ratios of the parameters reach the peak at the middle of eddy normalized life. The dipole structures also promote the geostrophic and propagation velocities. The geostrophic velocity of eddies is significantly enhanced in strong current areas. In addition, the enhancement of geostrophic velocity of eddy is zonally distributed and gradually decreases from the equator to the poles.We conclude that dipole structures change the initial state of the eddies, causing some parameter variations: amplitude, radius, and EKE of eddies are increased by less than 13%, while vorticity is reduced. The dipole structures also have a certain acceleration effect on eddies at propagation and geostrophic velocities. Futhermore, the coupling of eddies with opposite polarity enhances their fluctuation and causes them to interact. The dynamic mechanism and ecological effect of eddies coupling based on better data sets are also the focus of future work.关键词：remote sensing;satellite altimeter;mesoscale eddy;dipole eddy;parametric statistics;modulation effect- 434
- |
- 230
- |
- 0
发布时间：2023-05-10 -
摘要：The Chinese Ocean Color and Temperature Scanner (COCTS) onboard HY-1D satellite (COCTS HY-1D) was launched on June 11, 2020. However, the performance of COCTS HY-1D has not yet been completely evaluated. In this study, the performance of COCTS HY-1D was first evaluated by comparing satellite derived remote sensing reflectance (Rrs) with in situ measurements collected at four AERONET-OC sites and two Chinese long-term platforms.Initially, the in situ data at four AERONET-OC sites were acquired to evaluate the performance of COCTS HY-1D in the global coastal waters. AERONET-OC is an ocean color component of the AERONET and provides long-term high-quality in situ normalized water leaving radiance (Lwn) measured by an autonomous radiometer system on an offshore fixed platform to support the calibration and validation of satellite ocean color sensors in coastal waters. Muping and Dong’ou sites were constructed by the China National Satellite Ocean Administration Service (NSOAS), and the data were processed following the same procedure as that of the AERONET-OC data processing scheme. The COCTS HY-1D Level 1B data covering AERONET-OC sites and two long-term platforms between 1 August 1 2020 and 31 January 31 2021 in cloud-free days were acquired from NSOAS and processed to Level 2 Rrs and Chl-a concentration products. Furthermore, Rrs and Chl-a concentration comparison with two well-calibrated ocean color sensors (i.e., MODIS Aqua and VIIRS-SNPP) were made to evaluate the performance of COCTS HY-1D on the global scale. Additionally, the COCTS HY-1D Level 1B daily global dataset between December 7 and 14, 2020 were also required from NSOAS, processed to Level 2, and binned to Level 3 daily and 8-day 9-km data products by using the spatial-temporal binning algorithms developed by NASA. MODIS Aqua and VIIRS-SNPP Level 3 global binned daily and 8-day 9-km Rrs and Chl-a concentration data collected between December 7 and 14, 2020 were acquired from NASA GSFC. The statistics used in this study included correlation coefficient (r), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and mean bias (mBias).Results demonstrated that COCTS HY-1D-derived Rrs agreed well with the in situ data at all wavelengths with the correlation coefficient r of visible bands between 0.91 and 0.98 and up to 0.98 and Mean Absolute Percentage Error (MAPE) of 22.9%. The product’s accuracy is comparable to the average MAPE of 20.5% between MODIS Aqua and in situ data. At the global scale, the COCTS HY-1D-derived Rrs and chlorophyll concentration were consistent with MODIS Aqua products with a mean correlation coefficient ranging from 0.84 and to 0.95. The correlation coefficient of Chl-a is 0.85,which is higher than 0.76 between MODIS Aqua and VIIRS-SNPP. Nevertheless, the satisfactory Rrs was derived from COCTS HY-1D at the global scale compared with the in situ measurements or well-calibrated MODIS Aqua and VIIRS-SNPP products.COCTS HY-1D can provide high quality ocean color products comparable with the international mainstream ocean color satellite sensors, and therefore can carry out stable and accurate ocean color remote sensing observation.关键词：in situ observation;ocean color;HY-1D;COCTS;remote sensing refcectance- 481
- |
- 250
- |
- 1
发布时间：2023-05-10 -
摘要：HY-1D was launched in June 2020 as the first operational generation of Chinese marine satellite constellation with the launched HY-1C satellite for networking in the morning and afternoon. The Chinese Ocean Color and Temperature Scanner (COCTS) has two thermal infrared channels (10.30—11.40 and 11.40—12.50 µm) for observing Sea Surface Temperature (SST). In this work, the Bayesian cloud detection and optimal estimation algorithm are utilized for HY-1D COCTS SST retrieval in the Northwest Pacific based on the atmospheric radiative transfer model.COCTS is a whiskbroom scanner with eight parallel detectors along-track. The different spectral responses of these eight parallel detectors caused the sharp striped noise across the scan lines. The de-striping is carried out based on the unidirectional variational model. De-striping can be viewed as an optimization problem based on the minimization of a unidirectional variational model because striping can be assumed to be unidirectional noise because it does not affect the image horizontal gradient. The solution of the Euler-Lagrange equation is obtained based on a Gauss-Seidel fixed-point iterative scheme. The de-striped analysis show that the de-striping algorithm is successfully utilized in the HY-1D COCTS radiance data.Based on the simulated brightness temperature using the moderate resolution atmospheric transmission model, a Bayesian approach is utilized for the cloud detection of COCTS infrared brightness temperatures. Bayesian cloud detection is based on Bayes’ theorem, which determines a clear-sky probability given the satellite observations and prior background information. The COCTS brightness temperature images and the retrieved SST validation distributions show that the cloud detection is effective for SST retrieval.The optimal estimation algorithm is used for COCTS SST retrieval, based on the COCTS simulated brightness temperature and ERA5 SST as the prior SST. The HY-1D COCTS-retrieved SSTs are compared with the in situ SST and Visible Infrared Imaging Radiometer Suite (VIIRS) SST. The bias of the comparisons between the COCTS-retrieved SST and the in situ measurement is -0.04 ℃, and the standard deviation is 0.45 ℃. The bias and standard deviation of the COCTS newly retrieved SST minus SST from VIIRS are -0.05 ℃ and 0.49 ℃, respectively. Validation result shows that the accuracy of HY-1D COCTS SST in the Northwest Pacific reaches the equivalent level with international operational SST.关键词：remote sensing;HY-1D;COCTS;sea surface temperature;cloud detection;Retrieval;validation- 482
- |
- 327
- |
- 0
发布时间：2023-05-10 -
摘要：Ship detection by satellite remote sensing is of great significance for the safety of maritime navigation and the maintenance of maritime rights and interests. The traditional ship detection based on high spatial resolution Synthetic Aperture Radar (SAR) and optical satellite images cannot easily realize high-frequency monitoring application due to the long revisit period. The medium resolution Coastal Zone Imager (CZI) carried by China’s “Ocean-1” series satellites (HY-1) has a relatively low spatial resolution (50 m). However, HY-1C and HY-1D form a double satellite network observation in the morning and afternoon, which has the advantage of short revisit period and is of great value for marine vessel monitoring. We attempt to realize the ship automatic detection and orientation technology of medium-resolution CZI images, which will be of great value to the monitoring of ships at sea. In this study, a convolutional neural network is used for feature learning and target extraction, and an automatic ship detection method of HY-1/CZI image is established. Verification results show that this method has the advantages of not requiring threshold adjustment and strong adaptability, and the detection accuracy reaches 77.71%, which can be applied to the automatic monitoring of marine moving ships in the HY-1/CZI image. The algorithm in this work can directly detect the position and motion information of marine moving ships from the medium-resolution HY-1/CZI image without manual screening, realize the automatic extraction of wake, and overcome the problem of insufficient resolution of the medium-resolution optical image. Based on the detection results, this work further quantitatively describes the wake and obtains the information of the ship's position and movement direction.关键词：Coastal Zone Imager;Vessel inspection;convolutional neural network;satellite remote sensing- 479
- |
- 263
- |
- 0
发布时间：2023-05-10 -
摘要：As an essential part of the global climate system, sea ice affects the atmosphere and ocean circulation. It is also an important indicator of climate change. Sea ice concentration is one of the most important geophysical parameters for describing polar sea ice. We conduct an inversion study of Arctic sea ice concentration based on a Microwave Radiation Imager (MWRI) carried by FY3 series satellites. The daily dynamic tie point of the brightness temperature is determined by linear regression and the threshold method. The influence of weather and land pollution on sea ice concentration retrieval is eliminated using a weather filter and land pollution correction methods. The trend of sea ice extent and sea ice area calculated from 2019 to 2020 has a strong correlation with the sea ice concentration products released by NSIDC. The mean differences in the sea ice extent and sea ice area are -0.052 ± 0.015 × 106 km2 and -0.401 ± 0.093 × 106 km2, respectively. The sea ice concentrations have negative differences, approximately -3% in winter with a mean absolute deviation of 2%—4% and negative deviations of approximately -8% in summer with a mean absolute deviation of approximately 10%. The accuracy of sea ice concentration datasets based on different algorithms of MWRI is evaluated using SAR data. Results show that the retrieval result of the bootstrap algorithm is better than that of the NASA team algorithm. The accuracy is improved by approximately 1% in winter and approximately 4% in summer. The dynamic tie points of the brightness temperature effectively reflect the seasonal variation of sea ice radiative characteristics. This research has laid a foundation for the business release of sea ice intensive products of China’s autonomous satellites, thereby guaranteeing the continuity of sea ice records in polar regions facing interruptions for more than 40 years.关键词：remote sensing;Microwave radiometer;brightness temperature;sea ice concentration;FY-3;Arctic;bootstrap algorithm- 329
- |
- 227
- |
- 0
发布时间：2023-05-10 -
摘要：Snow depth on sea ice is an important part of the cryosphere and global climate system because it is essential in the energy transfer of the ocean, sea ice, and atmosphere. Monitoring and understanding the change of snow depth on the Antarctic sea ice is beneficial to sea ice research and global climate change. Given the lack of Chinese satellite data products for snow depth on the Antarctic sea ice, we explored the application of FY-3B MWRI passive microwave brightness temperature data to retrieve the snow depth on the Antarctic sea ice.We used FY-3B MWRI 18.7 GHz, 36.5 GHz vertical polarization brightness temperature, and sea ice concentration data to retrieve the snow depth on the Antarctic sea ice by using Comiso03 model. The accuracy of snow depth on the Antarctic sea ice based on Markus98 and Comiso03 models was evaluated and compared with GCOM-W1 AMSR-2 snow depth product. The influence of brightness temperature on the retrieval of snow depth on Antarctic sea ice from FY-3B MWRI was discussed.Snow depth on the Antarctic sea ice retrieved by using Comiso03 model than Markus98 model is better based on FY-3B MWRI 18.7 GHz, 36.5 GHz vertical polarization brightness temperature, and sea ice concentration data in 2016, and the average deviation against with ice mass balance buoy measurements (2016S31, 2016S37, and 2016S40) in the Weddell Sea in 2016 is -1.72 cm. The snow depth retrieved on the Antarctic sea ice in 2016 form FY-3B MWRI, which is consistent with that of GCOM-W1 AMSR-2 released data product by NSIDC (the average deviation is -0.11 cm, and the correlation coefficient is 0.90). The difference between the FY-3B-derived and the NSIDC-released AMSR-2 snow depths is small (the spatial-temporal average deviation is -0.80 cm, and the correlation coefficient is 0.93) in the snow accumulation and stable period (from April to October), and relative large (the spatial–temporal average deviation is 2.76 cm, and the correlation coefficient is 0.85) in the snow melting period (from November to March). These differences mainly distributed in the northern Weddell Sea and margin ice zones.Snow depth on the Antarctic sea ice retrieved by Comiso03 model based on FY-3B MWRI satellite data is in good agreement with the GCOM-W1 AMSR-2 snow depth product released by NSIDC. The study on the retrieval of snow depth on the Antarctic sea ice using FY-3B MWRI satellite data can provide scientific data and technical support for monitoring polar ice and snow environment and evaluation of Antarctic sea ice change with its global effects.关键词：snow depth;Antarctic sea ice;passive microwave remote sensing;satellite remote sensing- 344
- |
- 421
- |
- 0
发布时间：2023-05-10
Progress of Satellite Ocean Remote Sensing in China
-
摘要：Urban building extraction is an important research direction for the understanding and target recognition of high-resolution optical remote sensing images. Realizing accurate automatic building extraction has important application value and practical significance for the acquisition and update of basic urban geographic information. Given the complexity of urban scenes and the diversity of building forms, the characteristics of urban buildings are difficult to express fully, and the generalization ability of samples is insufficient, thus becoming a bottleneck problem for the automatic extraction of urban buildings. In this study, a multi-modal morphological-sequence-feature synergy method is proposed to utilize fully the advantages of each morphological sequence feature from different modes and mine the high-dimensional spatial information of urban buildings jointly. On this basis, we introduce multi-source a priori information and develop an adaptive segmentation model method based on multi-source a priori information to achieve the automatic recognition of urban buildings. This method can help avoid the limitations, such as errors and low efficiency, brought by manual threshold selection.The process of the method for urban building extraction proposed in this study is mainly divided into four steps. First, the differential morphological structure sequence features and differential morphological attribute sequence features of remote sensing images are calculated on the basis of high-resolution remote sensing images. Second, the feature selection model is constructed to optimize the differential morphological structure sequence features and differential morphological attribute sequence features. Then, the adaptive segmentation model is constructed on the basis of the multi-source a priori information products. The adaptive segmentation of the preferred features is performed to obtain the initial information of urban buildings. Finally, the voting method is used to fuse the initial information of urban buildings at the decision level to obtain the final urban building extraction results.The performance of the proposed method shows that the average extraction accuracy and kappa coefficient of the research method in this study are 91.3% and 0.87, which are 7.8% and 5.5% and 0.1 and 0.07 higher than the 85.7% and 83.5% and 0.81 and 0.78 of DMPs and DAPs extraction methods, respectively. Thus, the results demonstrate the effectiveness of the method in the automatic extraction of urban buildings in this study.The method in this research achieves rapid, automated, and high-precision urban building information acquisition and update. Furthermore, it provides a method reference template for rapid building detection and update in more cities. In the subsequent research, further quantitative evaluation of each type of a priori information product is needed to clarify the role of different information products in automatic building extraction, as it can improve the accuracy and automation of building extraction further.关键词：Morphological structural sequence features;morphological attribute sequence features;feature significance level model;adaptive segmentation model;decision-level information fusion;a priori information- 514
- |
- 382
- |
- 0
发布时间：2023-05-10 -
摘要：Land surface reflection depends on land surface albedo and interferes with the retrieval of cloud geometrical thickness from OCO-2 oxygen A-band observations due to its second-strongest reflection after the cloud. However, no product can provide the land surface albedo of the OCO-2 oxygen A-band required for the retrieval. Therefore, the accurate estimation of land surface albedo is necessary and beneficial to the retrieval quality.In this study, we proposed a method for estimating land surface albedo in the oxygen A-band from multichannel black/white albedos from MODIS/MCD43C3 products. Although the estimation (MODIS→OCO-2) is related to land cover type, the comparison based on Shannon entropy proved that the multichannel albedo data contains the type information and is sufficient to achieve the same accuracy as land cover-type estimation. In addition, we implement the estimation model by BP neural network, and the accuracy is consistent with that of the analysis based on the Shannon entropy.We verified the multichannel-based estimation model by tests in different times and spaces. Its correlation coefficients were all over 0.93, and the root-mean-squared errors were 0.026. In addition, the multichannel-based model was always superior to the single-channel linear model on all land cover types, whether applied to the best-performing type of barren or sparsely vegetated land or the worse-performing type of snow and ice. The quality of MODIS albedo data is the most important for the accuracy of estimation. The root-mean-squared error with the best inputs was slightly better than 0.02 and increased to more than 0.05 as the quality of the inputs decreased.The method of estimating the land surface albedo in the OCO-2 oxygen A-band from MODIS multichannel black/white albedo data is feasible and can resist the disturbance caused by unknown land cover type. The estimation accuracy depends on the quality of the input MODIS albedo data.关键词：remote sensing;land surface albedo;land cover type;OCO-2;oxygen A-band;cloud retrieval- 371
- |
- 329
- |
- 0
发布时间：2023-05-10 -
摘要：Building extraction is important for urban planning, land management, three-dimensional (3D) reconstruction, and other fields. Among various remote sensing techniques, 3D point cloud acquisition technique represented by Airborne Laser Scanning (ALS) system provides an efficient and convenient way for building extraction and modeling. Currently, deep learning-based building extraction methods are widely used. Compared with them, unsupervised building extraction methods do not need to train with a large amount of manually labeled data and do not require powerful computation equipment. Thus, developing unsupervised building extraction methods that do not rely on manual labeling is greatly meaningful.In this paper, based on the supervoxels of ALS point clouds, a high-order Markov network guided by the analysis of variance (ANOVA) is proposed for building extraction. This method first uses supervoxels as the nodes of the undirected graph, and then constructs high-order factors based on the principle of ANOVA and the local geometric features of 3D neighborhood. After that, the features are transformed into node and edge potential functions with a better expression ability. Finally, the belief propagation algorithm is used to make inference on the high-order Markov network, and an unsupervised building extraction framework is constructed.In the experiments, two groups of ALS point cloud datasets with ground-truths are employed, and four commonly used metrics are utilized to evaluate the accuracy of the results. Visual analysis shows that our method extracts buildings with complete interiors and clear boundaries. Hence, this method can be used to provide reliable data for the 3D reconstruction of buildings. According to quantitative analysis, in residential areas dominated by low-rise buildings, the averaged accuracy (the projection-area-based and object-based F1 indexes) of our method reaches 95.4% and 91.5% which are higher than that of existing supervised and unsupervised methods. In downtown areas dominated by high-rise buildings, the averaged object-based F1 score of our method reaches 93.5% which is higher than that of existing methods; and its averaged projection-area-based F1 score gets 92.9%, which is more than sufficient.This paper includes three innovative points. Firstly, an unsupervised building extraction framework from coarse to fine is constructed. Specifically, the Bayesian Gaussian Mixture Model is first used to capture the initial state of the node under the independent assumption, and then the high-order Markov network is designed to model the correlation of the 3D neighborhood for extracting buildings. Secondly, a potential function calculation method based on P-value test of ANOVA is constructed, which enhances the expression of interaction among supervoxels. Thirdly, higher-order factors are introduced into the supervoxel Markov network model, and the belief propagation algorithm is used for the inference of the network, thus the accurate identification of buildings is achieved. The results show that our method can effectively extract ALS point cloud buildings in different study areas, and the ablation study validates the positive effect of each module.关键词：ALS;Gaussian Mixture Model;building extraction;ANOVA;high-order Markov network- 372
- |
- 179
- |
- 0
发布时间：2023-05-10 -
摘要：Canopy height is one of important indicators of carbon storage in forest ecosystems. Previous studies have demonstrated that optical stereo imagery held great potential for deriving forest canopy height. However, optical images are easily affected by cloud coverage and rains. The regional mapping of forest canopy height has to be achieved by the synthesis of multi-sensor and multi-temporal imagery. The spatial resolutions and viewing angles of existing spaceborne stereoscopic systems are quite different. It is essential to make a systematic investigation about the impact of image spatial resolutions and viewing angles on the vertical distribution of stereoscopic point clouds, which is the basis for the synergy of images acquired at different seasons even by different cameras. The theoretical model for the simulation of optical stereo imagery over forested areas is necessary for such studies. The LandStereo is such kind of model, which can simulate the along-track(only In-track viewing) stereoscopic imagery like ALOS/PRISM and ZY-3. However, the current version of LandStereo has no mode to simulate the along-track(including In-track and Cross-track) stereoscopic imagery like Worldview-1/2/3.Therefore, this study reported the modification of the LandStereo model to have it being able to simulate images acquired by any possible viewing directions in forest areas. Firstly, the calculation method of linear array projection center coordinates is improved, from originally only considering the change of observation altitude angle to considering the change of azimuth angle and altitude angle. Secondly, Rigorous imaging geometric model is improved to obtain RPC of images acquired by any possible viewing directions. Based on the improved LandStereo model, the bare and mountainous forest images with altitude angle of 75° and azimuth angle of 0°, 90° , and 225° are simulated to verify the accuracy and extract forest canopy height.The surface elevation extracted by the improved LandStereo model is consistent with the input DTM with high accuracy (r=0.99, RMSE=1.507), which proves the geometric accuracy of the improved LandStereo model. There are certain differences in the extraction accuracy of forest canopy height between stereo images of different angles.The results showed that the modified model could correctly simulate the stereoscopic features of forest canopy with given view direction, also initially demonstrated that view angle was an important factor affecting the estimation accuracy of forest canopy height by stereoscopic images.关键词：forest height;multi-view;stereoscopic observations;imagery simulation;LandStereo;POV-ray;vegetation optical model- 428
- |
- 1183
- |
- 0
发布时间：2023-05-10 -
摘要：Alpine wetlands are an important surface cover type on the Qinghai―Tibet Plateau because they play a key role in water conservation, climate regulation, and biodiversity maintenance. Accurate and timely knowledge of the temporal and spatial distribution of alpine wetlands is necessary for wetland protection and management. The selection of remote sensing classification features is crucial in wetland mapping. Although spectral, texture, and topographic features have been investigated, studies focusing on spectral index features and their mathematical statistical features and feature selection methods are limited. This study aims to classify alpine wetlands from the aspects of mathematical statistical features, alpine wetland types, feature selection methods, and selected feature sets combined with random forest classification algorithm using Sentinel-2 image data and taking the Shouqu Alpine Wetland Reserve as the research site. An in-depth and comprehensive analysis on the spectral index characteristics of alpine wetlands is performed to optimize the classification characteristics of alpine wetlands.The Gansu Shouqu Alpine Wetland Reserve was used as the research area, and classification characteristics (spectrum, vegetation index, red edge index, and water body index) were obtained on the basis of Sentinel-2 data. Filter and wrapper feature selection methods, including Jeffries–Matusita distance, Spectral Angular Distance (SAD), Euclidean Distance (ED), RF-RFE algorithm, and Relief-F algorithm are utilized to optimize these features. Meanwhile, Z test is applied for quantitative evaluation.The following conclusions can be drawn from this study. (1) Among the categories of alpine wetlands involved in the classification, rivers and bare land are the easiest to distinguish, followed by grasslands and swamps and then swampy meadows and meadows. MCARI2, NDWI, DVI, EVI, EWI, IRECI, MCARI, TCARI, and UGWI indices can be used to differentiate among adjacent swamps, swampy meadows, meadows, and grasslands. (2) The order of contribution of different index characteristics to wetland information extraction in terms of degree is water body index characteristics > vegetation index characteristics > red edge index characteristics. (3) ED and Relief-F algorithms in the filter method demonstrate excellent performance from the perspective of feature optimization methods. (4) A suitable alpine wetland information extraction method is selected using the indices RDVI, NDVI, MSR, RVI, VIgreen, RNDWI, NDWI, NDWI_B, MNDWI, EWI, and CIre. (5) The mathematical statistics of different classification features indicated that the median feature obtains the best classification result, followed by the average value feature.We provide detailed results from feature optimization methods, wetland classification optimization index, statistical feature evaluation, and categories involved in alpine wetland classification using multi-dimensional analysis. To the best of our knowledge, this study provides a novel transferable and universal method for the selection of characteristic variables for wetland information extraction.关键词：remote sensing;wetland classification;alpine wetland;feature selection;Qinghai-Tibet Plateau;Sentinel-2- 592
- |
- 400
- |
- 0
发布时间：2023-05-10
Models and Methods
- Postal code：100190
- Tel：010-58887052 Email：nrsb@aircas.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备20021838号-9
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0