Solar-induced chlorophyll fluorescence (SIF) is an indicator of vegetation photosynthesis, and multiple satellite SIF products have been generated in recent years. However, current SIF products are limited for applications toward vegetation photosynthesis monitoring because of low spatial resolution or spatial discontinuity. This study uses a spatial downscaling method to obtain a redistribution of the original TROPOspheric Monitoring Instrument (TROPOMI) SIF (OSIF). As a result, a downscaled SIF dataset (TroDSIF) with fine spatio-temporal resolutions (500 m, 16 days) was generated. Compared with a machine learning (ML) SIF product and OSIF, TroDSIF can better reproduce the OSIF signals with higher R2, lower root mean square error (RMSE), and nearly zero residuals at different latitudes. Direct validation on TroDSIF using tower-based SIF measurements demonstrated a good consistency between them. However, TroDSIF is dependent on the linear hypothesis between OSIF and the ML-predicted SIF used in the redistribution process. Nonetheless, we believe TroDSIF is anticipated to be beneficial to conducting global vegetation photosynthesis and climate change studies at precise scales.

In China, the Loess Plateau’s fragile geological structure leads to complex and variable surface subsidence in old gob areas following coal mining activities. Accurately predicting this residual subsidence remains a significant scientific challenge. In this study, a method for residual subsidence prediction using an Exponential Smoothing Long Short-Term Memory (EsLSTM) model is proposed. The investigation centers on the 18,001# old goaf area of the Yangquan Coal Mine in Shanxi Province. Using Sentinel-1A imagery, continuous SAR data from 98 periods were acquired and processed via Enhanced Distributed Scatter InSAR technology. The EsLSTM model was then developed to capture the subsidence time-series characteristics of all surface scatter points and predict future ground subsidence. The analysis reveals that the EsLSTM model delivered excellent accuracy, achieving an R2 value of 0.975. It also outperformed SVR and traditional LSTM models, with a Mean Absolute Error of 2.2 mm and a Root Mean Square Error of 7.9 mm. Predicted results indicate that by October 2023, the maximum cumulative subsidence at the 18,001# working face of the Yangquan Coal Mine will reach 204 mm. The subsidence trend is expected to become more gradual and stable, suggesting a low likelihood of geological disasters in the area.

The China Loess Plateau (CLP) is a unique geomorphological unit with abundant coal resources but a fragile ecological environment. Since the implementation of the Western Development plan in 2000, the Grain for Green Project (GGP), coal mining, and urbanization have been extensively promoted by the government in the CLP. However, research on the influence of these human projects on the ecological environment (EE) is still lacking. In this study, we investigated the spatial–temporal variation of EE in a typical CLP region using a Remote Sensing Ecological Index (RSEI) based on the Google Earth Engine (GEE). We obtained a long RSEI time series from 2002–2022, and used trend analysis and rescaled range analysis to predict changing trends in EE. Finally, we used Geodetector to verify the influence of three human projects (GGP, coal mining, and urbanization). Our results show that GGP was the major driving factor of ecological changes in the typical CLP region, while coal mining and urbanization had significant local effects on EE. Our research provides valuable support for ecological protection and sustainable social development in the relatively underdeveloped region of northwest China.

The product of near-infrared radiation reflected by vegetation (NIRv) and PAR (NIRvP) is a promising proxy for the remote estimation of gross primary production (GPP). However, the efficiency of NIRvP in estimating the GPP and its limitations across multiple biomes and climate zones remain unclear. In this study, we aimed to evaluate the performance and limitations of NIRvP in estimating the GPP in comparison to absorbed photosynthetically active radiation (APAR), solar-induced chlorophyll fluorescence (SIF), and the MOD17A2H GPP product. Overall, the correlation between NIRvP and eddy covariance (EC) GPP was stronger than that of APAR, SIF, and MOD17A2H GPP across most biomes with usually similar seasonal variations in radiation, air temperature (TA), and precipitation. The near-​infrared (NIR) reflectance (ρNIR) and light use efficiency (LUE) exhibited a covarying relationship under these environmental conditions, which suggested that the ρNIR contributed positively to the NIRvP-GPP relationship under such climatic conditions. However, the performance of NIRvP was poor in some biomes and climate zones, which exhibited different variations in the seasonal patterns of radiation, TA, and precipitation. The resulting inconsistencies between ρNIR and LUE implied that the ρNIR contributed negatively to the NIRvP-GPP relationship in these regions. Altogether, the findings demonstrated that the NIRvP-GPP relationship was robust but attained a moderate overall relationship across ecosystems (R2 < 0.50) in the majority of biomes and climate zones. In addition, this study also elucidated the limitations of NIRvP as a GPP proxy in certain climate zones, which was attributed to the synergistic contributions of APAR and ρNIR in the NIRvP-GPP relationship.

[目的]研究红枣生长模型模拟输入参数的敏感性和产量预测不确定性,为红枣生长模拟模型的本地化和区域化参数调整优化提供依据,以提高模型模拟预测精度和效率.[方法]以新疆昆玉市现代农业示范区为研究区,应用可扩展傅里叶振幅敏感分析法(EFAST)和蒙特卡罗法分析基于DNDC模型系统新构建的红枣生长模型的输入参数敏感特性和产量预测不确定性.[结果]作物参数中全株生物量中果实比例(Gfra)、最大作物产量(MaxY)、生长积温(TDD)和需水量(WaterR)等指标敏感度最高,土壤参数中田间持水率(FC)和孔隙度(Por)等指标敏感度最高,田间管理参数中灌溉量(IrrAm)、施肥量(FerAm)和有机肥施肥量(ManAm)等指标敏感度最高;随着参数的波动范围由±5％增大到±10％,红枣预测产量正态分布的相关一致性系数增大,模型的平稳性增加.[结论]调整参数优化模型,并对2015～2019年各年份进行产量模拟测试验证,预测产量结果相对误差控制在±8％以内(最小误差为-1.99％),调整红枣产量预测模型参数,提高了模型预测产量的精度,优化趋于合理.

The accurate prediction of surface subsidence induced by coal mining is critical to safeguarding the environment and resources. However, the precision of current prediction models is often restricted by the lack of pertinent data or imprecise model parameters. To overcome these limitations, this study proposes an approach to predicting mine subsidence that leverages Interferometric Synthetic Aperture Radar (InSAR) technology and the long short-term memory network (LSTM). The proposed approach utilizes small baseline multiple-master high-coherent target (SBMHCT) interferometric synthetic aperture radar technology to monitor the mine surface and applies the long short-term memory (LSTM) algorithm to construct the prediction model. The Shigouyi coalfield in Ningxia Province, China was chosen as a study area, and time series ground subsidence data were obtained based on Sentinel-1A data from 9 March 2015 to 7 June 2016. To evaluate the proposed approach, the prediction accuracies of LSTM and Support Vector Regression (SVR) were compared. The results show that the proposed approach could accurately predict mine subsidence, with maximum absolute errors of less than 2 cm and maximum relative errors of less than 6%. The findings demonstrate that combining InSAR technology with the LSTM algorithm is an effective and robust approach for predicting mine subsidence.

Non-photosynthetic components within a forest ecosystem account for a large proportion of the canopy but are not involved in photosynthesis. Therefore, the accuracy of gross primary production (GPP) estimates is expected to improve by removing these components. However, their influence in GPP estimations has not been quantitatively evaluated for deciduous forests. Several vegetation indices have been used recently to estimate the fraction of photosynthetically active radiation absorbed by photosynthetic components (FAPAR(green)) for partitioning APAR(green) (photosynthetically active radiation absorbed by photosynthetic components). In this study, the enhanced vegetation index (EVI) estimated FAPAR(green) and to separate the photosynthetically active radiation absorbed by photosynthetic components (APAR(green)) from total APAR observations (APAR(total)) at two deciduous forest sites. The eddy covariance-light use efficiency (EC-LUE) algorithm was employed to evaluate the influence of non-photosynthetic components and to test the performance of APAR(green) in GPP estimation. The results show that the influence of non-photosynthetic components have a seasonal pattern at deciduous forest sites, large differences are observed with normalized root mean square error (RMSE*) values of APAR(green)-based GPP and APAR(total)-based GPP between tower-based GPP during the early and end stages, while slight differences occurred during peak growth seasons. In addition, daily GPP estimation was significantly improved using the APAR(green)-based method, giving a higher coefficient of determination and lower normalized root mean square error against the GPP estimated by the APAR(total)-based method. The results demonstrate the significance of partitioning APAR(green) from APAR(total) for accurate GPP estimation in deciduous forests.

The land surface temperature (LST) images obtained by thermal infrared remote sensing sensors are of great significance for numerous fields of research. However, the low spatial resolution is a drawback of LST images. Downscaling is an effective way to solve this problem. The traditional downscaling methods, however, have various drawbacks, including their low temporal and spectral resolutions, difficult processes, numerous errors, and single downscaling factor. They also rely on two or more separate satellite platforms. These drawbacks can be partially compensated for by the Sentinel-3 satellite’s ability to acquire LST and multispectral images simultaneously. This paper proposes a downscaling model based on Sentinel-3 satellite and ASTER GDEM images—D-DisTrad—and compares the effects of the D-DisTrad model with DisTrad model and TsHARP model over four sites and four seasons. The mean bias (MB) range of the D-DisTrad model is −0.001–0.017 K, the mean absolute error (MAE) range is 0.103–0.891 K, and the root mean square error (RMSE) range is 0.220–1.235 K. The Pearson correlation coefficient (PCC) and R2 ranges are 0.938–0.994 and 0.889–0.989, respectively. The D-DisTrad model has the smallest error, the highest correlation, and the best visual effect, and can eliminate some “mosaic” effects in the original image. This paper shows that the D-DisTrad model can improve the spatial resolution and visual effects of LST images while maintaining high temporal resolution, and discusses the influence of the terrain and land cover on LST data.

Synthetic Aperture Radar (SAR) is one of the most widely utilized methods to extract elevation information and identify large-scale deformations in mountainous areas. Homologous points in stereo SAR image pairs are difficult to identify due to complex geometric and radiometric distortions. In this paper, a new approach for mountainous area images is suggested. Firstly, a simulated SAR image and a look-up table based on DEM data are generated by a range-Doppler model and an empirical formula. Then, a point matching RPM-L2E algorithm is used to match images obtained by the simulation and in real-time to indirectly obtain the feature points of the real SAR images. Finally, the accurate registration of mountainous areas in the SAR images is achieved by a polynomial transform. Experimental verification is performed by using the data of mountainous SAR images from the same sensor and different sensors. When the registration accuracy of the method is compared with that of two state-of-the-art image registration algorithms, better outcomes are experimentally shown. The suggested approach can effectively solve the registration problem of SAR images of mountainous areas, and can overcome the disadvantages of poor adaptability and low accuracy of traditional SAR image registration methods for mountainous areas.

针对传统多光谱与全色影像融合方法容易产生畸变、忽略多光谱影像本身的空间细节特征等问题,提出了一种基于超分辨率卷积神经网络与Curvelet变换的影像融合方法,以提升多光谱影像的空间细节,加强其与全色影像的相关性,减少融合产生的畸变.该方法首先利用高分辨率全色影像进行超分辨率重建学习,利用学习得到的网络参数对多光谱影像进行超分辨率卷积神经网络重建,提升其空间细节特征;其次,在Gram-Schmid变换融合基础上,根据Curvelet变换具有保持影像空间细节的特点,将全色影像与替换分量进行融合;最后,通过逆变换得到高分辨率遥感影像.实验结果表明,该算法在影像光谱信息和空间细节表达能力上,整体优于其他传统算法,且对不同数据具有很好的适应性.

由于农村建筑物结构多样、空间分布复杂等特征,自动提取面临较多困难.针对该问题,本文提出采用膨胀卷积和金字塔池化表达的神经网络模型用于遥感影像中农村建筑物自动提取.在膨胀卷积神经网络模块中,通过改变孔尺寸的大小,获取不同感受野的特征信息;在金字塔表达方面,每个模块输入不同尺度的信息,且同时下采样的倍率也不同,获取多维的金字塔尺度特征;最终将提取的浅层及深层尺度特征信息进行融合,构建一个改进的适用于农村建筑物目标自动提取的深度学习模型.试验结果表明,与FCN-8s和DeepLab模型提取的结果相比,本文方法在农村建筑物提取中表现较好的性能,提取精度明显提高,且更好保留了目标边界细节信息,减少了噪声.

合成孔径雷达(SAR,synthetic aperture radar)卫星轨道参数是干涉测量技术中影像配准、基线估算、平地相位去除等环节的重要参数,但部分SAR卫星轨道参数采样间隔较大,导致干涉测量过程中产生残余相位,发生较大的系统误差;利用编程工具,对卫星原始轨道状态矢量进行了埃尔米特插值法拟合,等距插值计算后,发现可以缩小轨道参数采样间隔,提高干涉测量精度;以覆盖巴姆地区的Envisat卫星为例,分别获取了基于粗轨、埃尔米特插值轨道参数和代尔夫特精密轨道参数得到的干涉测量图,定性判断出埃尔米特插值法可有效提高SAR卫星轨道精度;再以覆盖陕西地区的AOLS卫星为例,插值轨道矢量采样间隔分别为10秒、5秒、2秒,发现间隔为5秒时相干性最优;结果表明:采用埃尔米特插值法可有效增加SAR卫星轨道状态矢量数量,消除系统误差,提高干涉测量精度.

Accurate estimation of gross primary productivity (GPP) is necessary to better understand the interaction of global terrestrial ecosystems with climate change and human activities. Light use efficiency (LUE)-based GPP models are widely used for retrieving several GPP products with various temporal and spatial resolutions. However, most LUE-based models assume a clear-sky condition, and the influence of diffuse radiation on GPP estimations has not been well considered. In this paper, a diffuse and direct (DDA) absorbed photosynthetically active radiation (APAR)-based method is proposed for better estimation of half-hourly GPP, which partitions APAR under diffuse and direct radiation conditions. Firstly, energy balance residual (EBR) FAPAR, moderate resolution imaging spectroradiometer (MODIS) leaf area index (LAI) (MCD15A2H) and clumping index (CI) products, as well as solar radiation records supplied by FLUXNET2015 were used to calculate diffuse and direct APAR at a half-hourly scale. Then, an eddy covariance-LUE (EC-LUE) model and meteorological observations from FLUXNET2015 data sets were used for obtaining corresponding LUE values. A co-variation relationship between LUE and diffuse fraction was observed, and the LUE was higher under more diffuse radiation conditions. Finally, the DDA-based method was tested using the half-hourly FLUXNET GPP and compared with half-hourly GPP calculated using total APAR (GPP_TA). The results indicated that the half-hourly GPP estimated using the DDA-based method (GPP_DDA) was more accurate, giving higher R2 values, lower RMSE and RMSE* values (R2 varied from 0.565 to 0.682, RMSE ranged from 3.219 to 12.405 and RMSE* were within the range of 2.785 to 8.395) than the GPP_TA (R2 varied from 0.558 to 0.653, RMSE ranged from 3.407 to 13.081 and RMSE* were within the range of 3.321 to 9.625) across FLUXNET sites within different vegetation types. This study explored the effects of partitioning the diffuse and direct APAR on half-hourly GPP estimations, which demonstrates a higher agreement with FLUXNET GPP than total APAR-based GPP.

北斗卫星导航系统(BDS)在实现高精度导航定位的同时必须实现对BDS精密星历内插,基于滑动式拉格朗日(Lagrange)插值和切比雪夫(Chebyshev)插值方法,对BDS三种类型轨道卫星进行精密星历插值分析,实验结果表明:当使用10阶滑动式Lagrange插值时,地球静止轨道(GEO)卫星精密星历三维坐标分量精度达到最高,均方根误差(RMS)分别为0.92、0.84、0.52 mm;当使用9阶Chebyshev插值时,GEO卫星精密星历三维坐标分量精度达到最高,RMS值分别为0.45、0.85、0.62 mm;当使用14阶滑动式Lagrange插值时,倾斜地球同步轨道(IGSO)卫星精密星历三维坐标分量精度达到最高,RMS值分别为0.85、1.28、1.38 mm;当使用14阶Chebyshev插值时,IGSO卫星精密星历X方向坐标分量、Y方向坐标分量精度达到最高,RMS值分别为0.79、1.30 mm,使用15阶Chebyshev插值时,IGSO卫星精密星历Z方向坐标分量精度达到最高,RMS值为0.62 mm;当使用14阶滑动式Lagrange插值时,中圆地球轨道(MEO)卫星精密星历三维坐标分量精度达到最高,RMS值分别为0.79、1.41、1.15 mm;当使用11阶Chebyshev插值时,MEO卫星精密星历Z坐标分量精度达到最高,RMS值为3.22 mm,当使用13阶Chebyshev插值时,MEO卫星精密星历Y坐标分量精度达到最高,RMS值为3.44 mm,当使用14阶Chebyshev插值时,MEO卫星精密星历Z坐标分量精度达到最高,RMS值为3.34 mm;滑动式Lagrange插值和Chebyshev插值方法适用于BDS不同轨道卫星精密星历内插,并且不同轨道卫星达到最佳精度时插值所取阶数不同,内插时应针对不同插值方法选择不同阶数以达到最佳结果.

针对现有道路提取算法中难以大规模人工标注样本类别标签的问题,提出了一种基于自适应标注样本提取遥感影像道路的方法.首先,通过改进的模糊C均值聚类算法提取道路区域,进行初步的样本标注;其次,利用基于二次投票的集成去噪算法定位标签噪声样本,更新样本数据集;再次,将更新后的样本集投入随机森林训练并预测影像的分类结果;最后,对道路提取结果进行多方向形态学滤波去除非道路区域,得到精确的道路提取结果.通过不同分辨率、不同场景、不同方法的实验结果表明,所提方法可以自主选择并标注样本,相比传统算法具有较高的提取精度,对于高分辨率遥感影像中直线型、曲线型道路均有较好的道路提取效果.

The resettlement of residents within the construction area of large projects is an important task related to people’s welfare. Livability is often used as an evaluation indicator when selecting resettlement areas. According to the results of the China Development Plan and 300 questionnaires, the human settlement factors that constitute livability include the living environment, ecological health, infrastructure, public facilities, and economic development, data on which can only be obtained from existing villages, and therefore cannot be used to directly assess the livability of potential resettlement areas. In fact, these human settlement factors are formed by the complex influences of numerous geographical factors (e.g., slope, slope orientation, accessibility, etc.), and it is scientific and reliable to use these geographical factors, which can be determined for each location, to carry out the livability assessment of potential resettlement areas. To this end, this paper takes the village resettlement project in the Dafosi coal mining area on the Loess Plateau of China as an example, calculates the livability scores of the existing villages around the coal mine using the entropy weighting method, and quantitatively analyzes the relationship between the livability scores and the selected geographic factors using a spatial correlations analysis method named Geodetector. It further uses the weighted overlayed function to superimpose the main geographic factors in order to obtain a livability grading map of the potential resettlement area. The results were successfully applied to the above resettlement project. We also verified the accuracy of this paper’s assessment method by adding 184 natural villages, and the method can be applied to other types of resettlement area livability assessment.

In various applications of airborne laser scanning (ALS), the classification of the point cloud is a basic and key step. It requires assigning category labels to each point, such as ground, building or vegetation. Convolutional neural networks have achieved great success in image classification and semantic segmentation, but they cannot be directly applied to point cloud classification because of the disordered and unstructured characteristics of point clouds. In this paper, we design a novel convolution operator to extract local features directly from unstructured points. Based on this convolution operator, we define the convolution layer, construct a convolution neural network to learn multi-level features from the point cloud, and obtain the category label of each point in an end-to-end manner. The proposed method is evaluated on two ALS datasets: the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen 3D Labeling benchmark and the 2019 IEEE Geoscience and Remote Sensing Society (GRSS) Data Fusion Contest (DFC) 3D dataset. The results show that our method achieves state-of-the-art performance for ALS point cloud classification, especially for the larger dataset DFC: we get an overall accuracy of 97.74% and a mean intersection over union (mIoU) of 0.9202, ranking in first place on the contest website.

Various means of extracting road boundary from mobile laser scanning data based on vehicle trajectories have been investigated. Independent of positioning and navigation data, this study estimated the scanner ground track from the spatial distribution of the point cloud as an indicator of road location. We defined a typical edge block consisting of multiple continuous upward fluctuating points by abrupt changes in elevation, upward slope, and road horizontal slope. Subsequently, such edge blocks were searched for on both sides of the estimated track. A pseudo-mileage spacing map was constructed to reflect the variation in spacing between the track and edge blocks over distance, within which road boundary points were detected using a simple linear tracking model. Experimental results demonstrate that the ground trajectory of the extracted scanner forms a smooth and continuous string just on the road; this can serve as the basis for defining edge block and road boundary tracking algorithms. The defined edge block has been experimentally verified as highly accurate and strongly noise resistant, while the boundary tracking algorithm is simple, fast, and independent of the road boundary model used. The correct detection rate of the road boundary in two experimental data is more than 99.2%.

High spatial resolution hyperspectral images (HR-HSIs) have shown considerable potential in urban green infrastructure monitoring. A prevalent scheme to overcome spatial resolution limitations in HSIs is by fusing low-resolution hyperspectral images (LR-HSIs) and high-resolution multispectral images (HR-MSIs). Existing methods considering the spectral dictionary or spatial dictionary can only reflect the unilateral characteristics of the HSI and cannot completely restore full information in the latent HSI. To overcome this issue, we propose a novel HSI-MSI fusion method, named DDSSLR, which joins spatial-spectral dual-dictionary and structured sparse low-rank representation. The spectral dictionary characterizing generalized spectra and the corresponding spectral sparse coefficients are extracted from LR-HSI and HR-MSI, while sparse low-rank priors of the local structure are imposed on the spectral pixels within the same superpixel in HR-MSI. Additionally, in the spatial domain, we exploit the remaining high-frequency components to learn the spatial dictionary and use the unitary transformation to factorize the spatial sparse coefficient into the sparse low-rank matrix in subspace, establishing the relationship between low-rank and sparse. We formulate the two fusion models as variational optimization problems, which are effectively solved by the alternating direction methods of multipliers (ADMM). Experiments on three HSI datasets show that DDSSLR achieves state-of-the-art performance.