Search

Real-time heatwave risk management with fine-grained spatial resolution is important for analysis of urban heat island (UHI) effects and local heatwaves. This study analyzed the spatio-temporal behavior of ground temperatures and developed methods for modeling them. The developed models consider two higher-order stochastic spatial properties (skewness and kurtosis), which are key to understanding and describing local temperature fluctuations and UHI effects. Application of the developed models to the greater Tokyo metropolitan area demonstrated the feasibility of statistically incorporating a variety of real datasets. Remotely sensed imagery and data from a variety of ground-based monitoring sites were used to build models linking urban covariates to air temperature. Air temperature models were used to capture high-resolution spatial emulator outputs for modeling ground surface temperatures. The main processes studied were the Tukey g-and-h processes for capturing spatial and temporal aspects of heat processes in urban environments. The main finding is that consideration of not only the mean temperature but also the variance, skewness, and kurtosis parameters can reveal hidden heatwave structures.

In this paper, we explore the problem of removing atmospheric turbulence to obtain better images of remote sidereal star systems. The imaging process is remodeled as a transmit–receive wireless communication paradigm in a novel approach for correcting space- and time-varying blur in stellar images. In particular, the effect of starlight passing through atmospheric turbulence is modeled as multipath fading communication channels. The problem is then transformed into channel equalization in space and time domains to produce a sharp stellar image. This approach involves first estimating the center of a blurred stellar image. Next, linear regression obtains an equivalent two-dimensional channel response through decomposition of the blurred image into the weighted sum of the diffraction-limited patterns in the spatial domain. Finally, an alignment algorithm is implemented for synthesis, and a final output is generated. Experiments were performed with real field-captured images; the results revealed that this approach could be used to effectively correct image blur and obtain diffraction-limited stellar images.

This article introduces a dynamic semiempirical model that predicts the degradation of a proton exchange membrane fuel cell (PEMFC) by introducing time-based terms in the model. The concentration voltage drop is calculated using a new statistical equation based on the load current and working time, whereas the ohmic and activation voltage drops are updated using time-based equations borrowed from the existing literature. Furthermore, the developed model calculates the membrane water content in the PEMFC, which indicates the membrane hydration state and indirectly diagnoses the flooding and drying faults. Moreover, the model parameters are optimized using a recently developed butterfly optimization algorithm. The model is simple and has a short runtime; therefore, it is suitable for monitoring. Voltage degradation under various loading currents was observed for long working hours. The obtained results indicate a significant degradation in PEMFC performance. Therefore, the proposed model is also useful for prognostics and fault diagnosis.

Image haze removal is essential in autonomous driving as the outdoor images captured during unfavorable weather conditions, such as haze or snow, are affected by poor visibility. Much research has been done to overcome image degradation such as low contrast and faded color due to haze. However, in the traditional model, a phenomenon is neglected that several particles simultaneously involved in light acquisition. To address this problem, we propose a novel single image dehazing method based on the spatially adaptive atmospheric point spread function (APSF). We developed a module that estimates the APSF to overcome the limitations of the spatially invariant APSF which used in existing dehazing algorithms. The key factor in the estimation is that road scenes with haze have different statistical characteristic from common hazy images in color and resolution. Furthermore, the APSF on the traffic signs or lights is estimated by generating superpixels to prevent halo artifacts around the sharp edges of the images. We adopted the total variation model as a regularization functional to reduce halo and unnatural artifacts that may occur during deconvolution. The haze-free images from the proposed method tested whether the proposed method can enhance the performance of vision algorithms for autonomous driving. The experimental results demonstrate that the proposed method outperforms state-of-the-art image dehazing methods enhancing the performance of the vision algorithms. Moreover, additional experiments demonstrated the effectiveness of the proposed method for quantitative and qualitative comparison with the state-of-the-art algorithms.

Time series interferometric synthetic aperture radar (TS-InSAR) has been a powerful tool for monitoring land surface deformation over the last two decades. Atmospheric effects cause large-scale delays in InSAR observations, which is one of the difficulties facing deformation calculations from differential InSAR and time-series InSAR. Currently, the atmospheric delay is derived mainly from auxiliary data from sources such as the global navigation satellite system (GNSS) and moderate-resolution imaging spectroradiometry (MODIS), but GNSS data are limited by the sparse distribution of observation stations. MODIS data may also not temporally match SAR image acquisition, which leads to low accuracy in atmospheric phase correction. This article presents a decomposition method to remove the atmospheric delay. We consider the atmospheric phase to be caused by the combined changes in spatial position and elevation. Therefore, quadtree segmentation is applied to divide the topographic units, and we improve the drift function of universal kriging by adding an elevation component. We then interpolate the whole atmospheric phase space from reliable sampling points estimated by the coherence coefficient. Using Sentinel-1 data, we test the proposed method in discriminating and monitoring a mining subsidence area in Shanxi Province and compare the results with the results from interferometric point target analysis and the network-based variance-covariance estimation method. The results demonstrate that the proposed method is superior to existing methods for the detection of deformation inverted from TS-InSAR.

Medium resolution spectral imager II (MERSI-2) is a payload for the Chinese meteorological satellite Feng Yun 3 (FY-3). China's satellite remote sensing observation capabilities such as climate change research can be improved during MERSI-2's operation in orbit, and the sensor is the world's first imaging instrument that can obtain a global infrared split-window data with a spatial resolution of 250 m. We developed an operational spilt-window (SW) algorithm to retrieve land surface temperature (LST) accurately from the MERSI-2 data. The SW algorithm coefficients were derived from a simulation dataset that was established with the Moderate spectral resolution atmospheric Transmittance model version 5.2 and the thermodynamic initial guess retrieval dataset. In the practical retrieval, the precise algorithm coefficients were determined by view zenith angle and atmospheric water vapor content (WVC), the atmospheric WVC were obtained from the ERA5 dataset, and the land surface emissivity was dynamically estimated using the advanced spaceborne thermal emission and reflection radiometer global emissivity dataset, considering the fractional of vegetation cover and snow cover. The retrieved LST compared with in situ LST, which was highly consistent with the in situ LST and that the root-mean-square error of the two is within 3 K. The retrieved LST was compared with the MYD11_L2 and MYD21_L2 LST products, and the results indicated that MERSI-2 LST was more consistent with the MYD21 LST. The operational SW algorithm for FY-3D MERSI-2 developed in this study could retrieve LST accurately and has a wide range of popularization and application values.

Single-layer phase screen is a classical, simple and accurate approach for simulating the optical field under turbulence, and it is generally used in free space optical communication (FSOC). However, conventional single-layer phase screen studies have rarely considered the pointing errors from FSOC, which prevents the theoretical results for mobile FSOC from being verified using simulated optical fields. This work establishes a simulation method for the single-layer phase screen with pointing errors. Statistics on simulated optical field with pointing errors indicated that the previous theoretical results for the long-term average irradiance and irradiance fluctuation probability density function (PDF) for mobile FSOC can be modified to better fit the simulation results. In addition, the received power fluctuation PDF, average bit-error-rate (BER), and the influence of non-zero mean pointing errors are investigated through the simulated optical field with pointing errors. The proposed simulation method provides a powerful approach for understanding and evaluating the comprehensive influences of atmospheric turbulence and pointing errors on the FSOC.

To acquire daytime land surface temperature (LST) in thin cirrus cloudy skies, we have developed a three-channel LST retrieval algorithm based on a widely used two-channel LST retrieval algorithm for the clear-sky conditions. In this algorithm, the LST is expressed as a multiple linear function of MODIS channels 29, 31 and 32 with the coefficients of the linear function dependent on the cirrus optical depth (COD) and cirrus effective radius (R). The influences from land surface emissivities (LSEs) are also considered in this algorithm. The simulated dataset shows that the LST could be estimated using the proposed algorithm with the root mean squire error (RMSE) less than 2.2 K in thin cirrus cloudy skies (COD less than 0.7) when viewing zenith angle (VZA) equivalent to 0°. As VZA is equivalent to 60°, the maximum RMSE are 2.7 K. The widely used generalized split-window (GSW) algorithm proposed for clear-sky conditions are used in cirrus cloudy skies, and the RMSEs of GSW algorithm estimated LST are 16.89 K and 22.32 K for VZA =0° and VZA =60° respectively when COD is 0.7. It indicates that the proposed three-channel algorithm can significantly improve the LST retrieval accuracy using thermal infrared data in cirrus cloudy skies. To estimate the LST errors caused by the uncertainties of COD, R, LSE and instrument noise, a sensitivity analysis was performed. It shows that the accuracy of cirrus COD is more important for the retrieval of LST compared with other parameters. The maximum total LST errors, taking into account all the input parameters’ uncertainty and algorithm error itself, are 3.8 K and 4.3 K when VZA =0° and VZA =60° respectively.

The presence of haze could seriously contaminate the observations of optical satellite imagery. Haze not only significantly affects the visual interpretation but also reduces the accuracy of map products. In this article, a semiphysical approach is proposed to reduce the haze effects for Landsat image. The proposed approach is based on the physical model of radiative transfer theory and the presence of dark objects. As the depth map of satellite remotely sensed image is almost a constant value, the coarse transmission map of atmosphere is estimated by the haze thickness, other than the scene depth map. The derived coarse transmission is utilized to correct the color shift induced by airlight. For haze veiled textural information, the guided-filter based approach is adopted to refine the coarse transmission map to restore the textural information. Experiments are conducted upon the images acquired by Landsat at different dates and spatial locations. The visual interpretation upon the dehazed results suggests that the proposed approach could generate visually promising results and preserve spectral properties of land surfaces. Moreover, it also performs favorably against several state-of-the-art deep learning based methods and a classic algorithm. The results of quantitative assessments demonstrate that the developed approach could enhance the information of Landsat scene with minimum spectrum changes. With the visual and quantitative assessments, we can conclude that the proposed approach is a promising solution for Landsat image dehazing. It is expected to improve the data quality of hazy scene and expand the usability of Landsat data.

Surface soil moisture retrievals were performed by inverting physical scattering models for forests over $\text{30}^\circ$ to $\text{50}^\circ$ incidence angle range and 0.05 to 0.40 m $^3$ /m $^3$ soil moisture range using $L$ -band airborne synthetic aperture radar (SAR) data during a 28-day period. The forward models implemented single-scattering of discrete elements of trees and were validated at $F$ 2 site within 1.5 dB rmse of observation for VV-pol, which was enabled by introducing the gaps between trees. The physical forward models were inverted using the time-series SAR data to retrieve soil moisture and soil surface roughness, which were validated with in situ data at four sites $F$ 1, $F$ 2, $F$ 3, and $F$ 5. Retrievals using VV input over the full dynamic ranges of wetness are accurate to 0.044 m $^3$ /m $^3$ unbiased rmse with correlations of 0.71 to 0.84, which is very encouraging for retrieval under forest canopy. The conditions of these results are the vegetation water content varied from 7.3 to 25.6 kg/m $^2$ and the sensitivity of VV to soil moisture changes ranged from 0.64 to 2.27 dB/(0.1 m $^3$ /m $^3$ ). When both HH and VV were used as inputs to retrieval, the performance did not improve because the benefit of the multichannels was offset by the larger uncertainties in HH modeling. Due to the short duration of in situ data, the efficacy of commonly used relative change index retrieval is diminished. In comparison, the inversion of the scattering model is found to be an effective way to estimate forest soil moisture, it is capable of systematic correction of vegetation effect, and offers accurate retrieval of the dynamic ranges of soil moisture values in terms of unbiased rmse. The retrieval with the physical forward model provides a first step toward global application for the future NISAR satellite.

Near-surface relative humidity (RHns) is an essential meteorological parameter for water, carbon, and climate studies. However, spatially continuous RHns estimation is difficult due to the spatial discontinuity of in situ observations and the cloud contamination of satellite-based data. This article proposed machine learning-based models to estimate spatially continuous daily RHns at 1 km resolution over Japan and South Korea under all sky conditions and examined the spatiotemporal patterns of RHns. All sky estimation of RHns using machine learning has been rarely conducted, and it can be an alternative to the currently available RHns data mostly from numerical models, which have relatively low spatial resolution. We combined two schemes for clear sky conditions (scheme A, which uses satellite and reanalysis data) and cloudy sky conditions (scheme B, which uses reanalysis data solely). The relatively small numbers of data in extremely low and high RHns conditions (i.e., 70%, respectively) were augmented by applying an oversampling method to avoid biased training. The machine learning models based on random forest (RF) and XGBoost were trained and validated using 94 in situ observation sites from meteorological administrations of both countries from 2012 to 2017. The results showed that XGBoost produced slightly better performance than RF, and the spatially continuous RHns model combined based on XGBoost yielded the coefficient of determination of 0.72 and a root-mean-square error of 10.61%. Spatiotemporal patterns of the estimated RHns agreed with in situ observations, reflecting the effect of topography on RHns. We expect that the proposed RHns model could be used in various environmental studies that require RHns under all sky conditions as input data.

The F2-layer peak parameters of the International Reference Ionosphere (IRI) model are critical to its subsequent applications because of the vital role of the peak parameters in determining other IRI model outputs. Therefore, we analyzed the statistical deviation and ionospheric climatology of NmF2/hmF2 between the most current version of the IRI model, namely IRI-2016, and ionospheric radio occultation (IRO) data, including Constellation Observing System for Meteorology Ionosphere and Climate (COSMIC) and FengYun-3C (FY3C), during the recent solar minimum in 2017–2019. The International Union of Radio Science (URSI) model for F2-layer peak electron density (NmF2) and the Bilitza–Sheikh–Eyfrig (BSE), Altadill-Magdaleno–Torta–Blanch (AMTB), and SHUbin (SHU) models for maximum ionization height (hmF2) in IRI-2016 are compared with the IRO dataset. A statistical analysis shows a systematic offset of less than 10% in NmF2 between IRI-2016 URSI and IRO data, and the SHU option for hmF2 shows a better fit to IRO data than BSE and AMTB. A climatology comparison shows that NmF2 estimated by IRI-2016 URSI is consistent with that of the IRO dataset in terms of the general trend in the typical ionospheric characteristics; except for some discrepancies in the marine area, the IRO data may provide a feasible reference for IRI estimation in marine regions. The hmF2 options of BSE/AMTB/SHU show hemispheric asymmetry like the IRO dataset. Moreover, hmF2 of BSE/AMTB and IRO/SHU show some discrepancies in fine structures, such as the day–night reversal phenomenon.

The atmosphere-land exchange inverse disaggregation (DisALEXI) algorithm is a multi-scale energy balance model that estimates evapotranspiration (ET) using land-surface temperature (LST) as a driving remote sensing input. Using LST products from ECOSTRESS, a thermal radiometer mounted on the International Space Station, DisALEXI ET products have been produced over the contiguous United States (CONUS) at 70 m resolution. The goal of this study is to demonstrate the accuracy of the CONUS-wide ET produced by the Jet Propulsion Laboratory (JPL) and to compare the results with the original DisALEXI ET produced by researchers at the United States Department of Agriculture (USDA). DisALEXI-USDA has been produced ad-hoc using Landsat LST, and is routinely produced over six target sites using ECOSTRESS LST. DisALEXI-JPL was implemented in order to expand the spatial coverage. DisALEXI-JPL was evaluated at 26 CONUS eddy covariance sites, showing good correlation, with R2 = 0.80 and RMSE = 0.81 mm/day, which is comparable to previous DisALEXI validation studies (RMSE ∼1 mm/day). The two DisALEXI implementations compared well, with R2 = 0.92. This article evaluates DisALEXI-JPL and shows that the algorithm is valid over a larger segment of CONUS. We also show the impact of quality flags, as pixels with high view zenith angles or high aerosol optical depth showed greater deviation from field measurements. As a product demonstration, we show a regional map of fine-scale ET, where the fine-scale variation over wider areas can detect small areas of stress much sooner than products with coarse resolution representing average conditions.

Protected crop production is rapidly expanding in the Mediterranean Basin, and particularly in Morocco. Increased local and overseas demand for these products led to a rapid development in greenhouse usage encouraged by government policies. The aim of this study is to investigate key design parameters that affect the thermal behavior and the heating/cooling energy need of a greenhouse situated in Agadir (Morocco). The parameters include the cladding material characteristics, shape, orientation, and air change rate. The greenhouse is modeled by a developed thermal model using TRNSYS software. The model considers the presence of the plants inside the greenhouse by adding the heat and humidity gain into the heat and water balance of the greenhouse using an evapotranspiration sub-model. The effect of evapotranspiration on the greenhouse thermal behavior was also examined in this study. A validation of the current TRNSYS simulation and evapotranspiration model was made using previous studies from the literature, and the comparison showed fair agreement. The relative error of the annually heating demand obtained by this model is 1.66%, and the evapotranspiration model used in this study shows relative deviation less than 6.5%. The results of this study indicate that the East-West greenhouse orientation is the optimum orientation as it can reduce the annual cost of air-conditioning of the greenhouse by 9.28% compared to North-South orientation. Quonset shape is the optimum greenhouse shape in Morocco as it can save 14.44% of annual cost of air-conditioning instead of the Even-span shape.

In this article, a water vapor density inversion model based on the long short-term memory network is proposed for E-band commercial microwave links (CMLs). A full-duplex E-band microwave link located in Prague (two sublinks with frequencies of 73.5 and 83.5 GHz, both vertically polarized) was used to verify the performance of the model. The results show that the model inversion results are in good agreement with the water vapor density calculated by temperature and humidity sensors. Compared with previous water vapor inversion methods based on CMLs, this model has a higher temporal resolution and can realize real-time monitoring.

Kernel-driven models provide an effective way for correcting the thermal radiation directionality effect. Under a general kernel-driven modeling framework proposed by Cao et al., by using three fixed-width hotspot kernels, and considering whether to combine two existing base shape kernels, this article proposed nine kernel-driven models with different coefficient requirements. Specifically, the three hotspot kernels are four-component kernel (KGO4), three-component kernel (KGO3), and sunlit soil kernel (KGOg), and the two base shape kernels are temperature difference kernel (KLSF) and emissivity kernel (KEmi). Based on discrete anisotropic radiative transfer model, a comprehensive simulation data set, including three fraction of vegetation covers (FVC) and 36 component temperature profiles, was generated for evaluating the performances of novel models. First, the order of model performances was determined, and the result is basically independent of FVC and component temperature difference. In the case of five-parameter, four-parameter, three-parameter, and two-parameter, the best models are GO4_LSF model or GO4_Emi model, GO3_LSF model or GO3_Emi model, GOg_LSF model or GOg_Emi model, and GOg model, respectively. In addition, for three-parameter models, the comparison with two current linear kernel-driven models, i.e., LSF_Li model and Vinnikov model, was discussed in detail, and it is revealed that newly proposed GOg_LSF and GOg_Emi model in this article have a better performance.

The Southern Ocean comprises 25% of the global ocean surface area, accounts for nearly half of the total carbon sink of the global oceans, and is a place that significantly reduces the impacts of anthropogenic CO2 emissions. Due to the sparsity of observational data, the changes in Southern Ocean carbon sinks over time remain uncertain. In this study, we integrated correlation analysis and a feedforward neural network to improve the accuracy of carbon flux estimations in the Southern Ocean. Based on observation data from 1998–2018, we reconstructed the Southern Ocean's pCO2 grid data during this period. The root-mean-square error obtained by fitting the observation data was 8.86 μatm, indicating that the results were better than those of the two primary statistically based models in the Surface Ocean pCO2 mapping intercomparison. The results also showed that the Southern Ocean's capacity to act as a carbon sink has gradually increased since 2000; it reduced during 2010–2013 but increased significantly after that. The Southern Ocean's seasonality is characterized by minimum carbon uptake in winter due to increased upwelling; this is followed by a rapid increase toward maximum uptake in summer, which is mainly biologically driven. There is an apparent double-ring structure in the Southern Ocean, as noted in other studies. This study confirms that the inner ring (50–70°S) is a carbon source area gradually transforming into a carbon sink, while the outer ring (35–50°S) continues to serve as a carbon sink.

The tropospheric delay is a major error source for Global Navigation Satellite System (GNSS) positioning and navigation, and usually can be corrected by using an empirical model. Due to the small number of parameters and simplified algorithm, the UNB3m model has been widely hardwired in GNSS receivers for real-time applications. However, many studies have noted that the UNB3m model has significant systematic errors in the correction of tropospheric delays, mainly due to the assumption of north-south symmetry. Therefore, considering the realistic atmospheric behavior, we proposed a new tabular zenith tropospheric delay (TZTD) model using 10 years of NCEP (National Centre for Environmental Prediction) data. The performance of the TZTD model was assessed along with the GPT3 and UNB3m models by comparison with ZTDs derived from the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA5) data and GNSS observations. The results show that the TZTD model (with a RMSE of 41mm) significantly outperforms the UNB3m model (with a RMSE of 49 mm). Furthermore, the accuracy of the new model is better than that of the GPT3 model in terms of the zenith tropospheric hydrostatic delay (ZHD) prediction, particularly at high altitudes. The TZTD model characterized by simplicity and accuracy, is expected to be a substitute for the UNB3m model in real-time GNSS applications.

A multimodal generative modeling approach combined with permutation-invariant set attention is investigated in this article to support long-wave infrared (LWIR) in-scene atmospheric compensation. The generative model can produce realistic atmospheric state vectors ( $\text{T}, \text{H}_2\text{O}, \text{O}_3$ ) and their corresponding transmittance, upwelling radiance, and downwelling radiance (TUD) vectors by sampling a low-dimensional space. Variational loss, LWIR radiative transfer loss, and atmospheric state loss constrain the low-dimensional space, resulting in lower reconstruction error compared to standard mean-squared error approaches. A permutation-invariant network predicts the generative model low-dimensional components from in-scene data, allowing for simultaneous estimates of the atmospheric state and TUD vector. Forward modeling the predicted atmospheric state vector results in a second atmospheric compensation estimate. Results are reported for collected LWIR data and compared against fast line-of-sight atmospheric analysis of hypercubes-infrared (FLAASH-IR), demonstrating commensurate performance when applied to a target detection scenario. Additionally, an approximate eight times reduction in detection time is realized using this neural network-based algorithm compared to FLAASH-IR. Accelerating the target detection pipeline while providing multiple atmospheric estimates is necessary for many real world, time sensitive tasks.

The impact of fine particulate matter on health has captured attention worldwide. Many studies have proven that fine particulate matter harms the respiratory system and the cardiovascular system. To prevent people from being harmed, many scientific research studies on PM2.5 prediction have been conducted in recent years. Accurate PM2.5 forecasting can not only alert people to stay away from concentrated areas but also provide the government with environmental policies in the future. In this paper, we propose a hybrid time-series prediction framework for daily-based PM2.5 forecasting. The proposed framework consists of three components: the autoencoder, the dilated convolutional neural network, and the gated recurrent unit. The experimental dataset with 76 monitoring stations from the Taiwan Environmental Protection Administration is applied for comparison of the baseline and the proposed models. The proposed model is not only for the specified city-/county-wide region but also for the particular monitoring station/site to predict PM2.5 concentration. By considering air quality data, meteorological data, and geographical data simultaneously, the proposed model can increase the accuracy of PM2.5 prediction. In addition, the proposed PM2.5 forecasting model can learn the location-centric spatial features and the daily-based temporal features simultaneously. The experimental results show that the prediction accuracy of the proposed model is superior to those of the baseline models.

Lake bathymetry plays a pivotal role in environmental monitoring, ecological management, water quality protection, etc. Hyperspectral remote sensing technology can provide large-scale coverage and more detailed spectral information for bathymetry estimation than traditional measurements or multispectral imagery techniques. In this study, a multiband linear model with successive projections algorithm (SPA-MLM) was developed to retrieve the bathymetry of Qinghai Lake, which is the largest inland saltwater lake in China. The three most sensitive spectral bands were first selected by the SPA, and a multiband linear model was established by the least squares method combined with the in situ measured water depth. Zhuhai-1 hyperspectral remotely sensed imagery is employed as the data source. In all, 98 in situ bathymetry measurements matched with the obtained images were obtained during three surveys performed in May, September, and October 2020. The results demonstrated that the established retrieval model can be used to accurately estimate the water depth in the study area, with an accuracy exceeding approximately 90%, which suggests that the proposed model performs better than those used in previous studies employing hyperspectral imagery. The correlation coefficient reaches 0.92, and the root-mean-square error is approximately 1.26 m. This demonstrates that bathymetry estimation obtained using remotely sensed hyperspectral data is an effective detection method and can provide large-scale, rapid monitoring data to the relevant decision-making departments.

For a pulsed laser transmitted in the raised dust from battlefield vehicle movement, the premise of obtaining its transmission properties and anti-jamming method is to acquire the spatial distribution rules of dust mass concentration. In this study, the emission mechanism for vehicle-raised dust in the battlefield environment is analyzed, as well as various influencing factors of the dust emission velocity. A finite volume method (FVM)-based model for simulating the spatial dispersion of dust raised by a certain moving tank is built, and the effects of mass flow rate, traveling speed, wind speed and wind direction on the spatial distribution of dust mass concentration are explored in the battlefield environment. The results show that the raised dust is plume-shaped within the computational domain, and the spatial mass concentration of plume is positively correlated with the mass flow rate. Besides, it is necessary to determine the impact of tank travelling speed on the spatial mass concentration of dust based on the relationship between the travelling speed and the dust mass flow rate. Although front wind reduces the spatial mass concentration of dust plumes, the reduction rate decelerates when the wind speed exceeds 4 m/s. Crosswind leads to the deflection and decreased mass concentration of dust plumes. Under varying wind directions, the variation of crosswind component exerts a more pronounced influence on the spatial mass concentration of dust than that of the front wind component. This study provides a technical support for the further study of near-ground laser weapon system against vehicle dust interference in the battle field.

Research on forecasting methods of time series data has become one of the hot spots. More and more time series data are produced in various fields. It provides data for the research of time series analysis method, and promotes the development of time series research. Due to the generation of highly complex and large-scale time series data, the construction of forecasting models for time series data brings greater challenges. The main challenges of time series modeling are high complexity of time series data, low accuracy and poor generalization ability of prediction model. This paper attempts to cover the existing modeling methods for time series data and classify them. In addition, we make comparisons between different methods and list some potential directions for time series forecasting.

The air pollution caused by particulate matter (PM) has become a public health issue due to the risks to human life and the environment. The PM concentration in the air causes haze and affects the lungs and the heart, leading to reduced visibility, allergic reactions, pneumonia, asthma, cardiopulmonary diseases, lung cancer, and even death. In this context, the development of systems for monitoring, forecasting, and controlling emissions plays an important role. The literature about forecasting systems based on Artificial Neural Networks (ANNs) ensembles has been highlighted regarding statistical accuracy and efficiency. In this article, trainable and non-trainable combination methods are used for PM10 and PM2.5 (particles with an aerodynamic diameter less than 10 and 2.5 micrometers, respectively) time series forecasting for eight different locations, in Finland and Brazil, for different periods. Trainable ensembles based on ANNs, linear regression, and Copulas are compared with non-trainable combinations (mean and median), single ANNs, and linear statistical approaches. Different models are considered so far, including Autoregressive model (AR), Autoregressive and Moving Average Model (ARMA), Infinite Impulse Response Filters (IIR), Multilayer Perceptron (MLP), Radial Basis Function Networks (RBF), Extreme Learning Machines (ELM), Echo State Networks (ESN), and Adaptive Network Fuzzy Inference System (ANFIS). The use of ANNs ensembles, mainly combined with MLP, leads to a better one step ahead forecasting performance. The use of robust air pollution forecasting tools is prime to assist governments in managing air pollution issues like hospital collapse during adverse air quality situations. In this sense, our study is indirectly related to the following United Nations sustainable development goals: SDG 3 - good health and well-being and SDG 11 - sustainable cities and communities.

Synthetic aperture radar (SAR) has been proven to be a useful tool in monitoring hurricane structure and intensity. By far, SAR is the most promising spaceborne sensor to obtain high-resolution hurricane wind field on the ocean surface. In this article, an improved asymmetric hurricane parametric (IMAHP) model has been proposed to reconstruct the asymmetric wind speed, where the high-resolution cross-polarization SAR imagery is used to determine the value of model parameters. Compared with other models, the new model can better reconstruct hurricane wind speed with a more concise model function. For verification, taking SAR-retrieved wind speed as a reference, the root-mean-square error and bias of the wind speed estimated by the IMAHP model are 1.86 m/s, 1.89 m/s for Hurricane Arthur (2014), 2.01 m/s, 1.77 m/s for Iselle (2014), and 1.99 m/s, 1.74 m/s for Norbert (2014), respectively. Finally, comparisons with airborne stepped-frequency microwave radiometer and dropwindsondes measurements show that the wind speed simulated by the IMAHP model is close to these measurements.

The targeted compounds included Polychlorinated Biphenyls (PCBs), Pesticides (PESTs), Polycyclic Aromatic Hydrocarbons (PAHs) and so on in the Great Lakes Integrated Atmospheric Deposition Network (IADN), which is a platform based on the IoT (Internet of Things) technology to collect environmental pollutants data. While previous studies usually employed traditional statistical approaches to analyze the IADN results, we performed a complete modeling workflow of the total concentrations of PCBs, PESTs, and PAHs (which is referred to as $\sum $ PCBs, $\sum $ PEST s and $\sum $ PAHs orderly) in 1990–2016 samples by using a machine learning algorithm combined with data-driven research method, which lets the model fit the data, so as to change the model to achieve the effect. The main results of this article are as follows, 1) identifying the spatial and temporal trends of POPs (Persistent Organic Pollutants) in the air of the Great Lakes; 2) An appropriate data-driven intelligent model was constructed for the data at EH (Eagle Harbor) and STP(Sturgeon Point) sampling sites, via which we estimated their $\sum $ PCBs, $\sum $ PESTs, and $\sum $ PAHs in the following 4–5 years, showing the concentrations will continue declining with slight fluctuations; 3) The important role which IoT played in smart environmental protection was pointed out.

With the exploitation of mineral resources, pollution of the ecological environment in mines has garnered public attention. Particularly,erosion of the surrounding ecological environment re-sulting from heavy metals in tailings pond could be highly concerning. Instead of traditional field sampling and laboratory analysis method, remote sensing can be used to high-precisi es-timation soil heavy metal with less time and effort. soil heavy metal content is generally low, the spectral sensitivities of various heavy metals are insignificant, and the surface landscape is complex, there exist difficulties associated with heavy metal content estimation. Therefore, herein, we propose optimization of the commonly used partial least-square regression (PLS) method. In the optimized method, a variety of remote sensing indices and the modeled heavy metals were added as modeling factors to indirect estimation soil heavy metal. The method was validated via inversion experiments of heavy metals (Ni, Cu, and Zn) in the tailing pond and its surrounding environment,it improve the goodness-of-fit of Ni, Cu, and Zn by 0.0852,0.2291, and 0.2919 compared with traditional PLS. Spatia l analysis was then conducted on the entire studied area using the estimation model of the three heavy metals. It was shown that the results were essentially consistent with the actual heavy metal distribution in the area. Therefore, the indirect PLS model with multiple factors proves effective for the estimation of soil heavy metals. It also provides technical support for treatment and evaluation of ecological environments in mining areas.

Potential evapotranspiration (PET) is a key parameter for calculating drought monitoring index that is generally difficult to obtain. In addition, PET has low spatial resolution and can only be obtained at a site-based point. Therefore, retrieving PET with high precision, high spatial resolution, and less meteorological data becomes the focus of this article. In this article, a high-precision and high-spatial-resolution drought monitoring (HDM) model was established to accurately calculate PET by using the zenith troposphere delay (ZTD) derived from global navigation satellite system and temperature (T). The initial PET value was calculated by using the PET periodical model based on Penman–Monteith (PM) derived PET. The PET difference (DET) between the PM and periodic model was then calculated, and a multiple linear regression model was established to fit DET by using the ZTD and T differences at meteorological stations. To improve the spatial resolution of the calculated PET, a spherical harmonic function was applied to fit the coefficients of these stations. The HDM-derived PET at grid points was eventually obtained by using the fitted coefficients and ZTD/T. The HDM-derived PET and standardized precipitation evapotranspiration index (SPEI) were compared with those from the Thornthwaite (TH) and revised TH models over the loess plateau area with the PM-derived PET and SPEI as references. Comparison results highlight the excellent performance of the proposed HDM model and the Pearson's correlations of SPEI between the HDM and PM models all exceeded 0.96 under different month scales.

The study of climate change has become an important topic because of its negative impact on human life. The North-East African part lacks the studies for climate change detection, despite it being one of the most affected parts worldwide. The relationship between the emission of greenhouse gases (GHGs) and climate change is an important factor to understand. To investigate this linkage, we used machine-learning (ML) models based on essential climate variables (ECVs) to investigate the relationship between the GHGs and the rhythm of climate variable change. The article investigates how ML techniques can be applied to climatic data to build an ML model that is able to predict the state of climate variables for the short and long term. By selecting a candidate model, it will help in climate adaptation and mitigation, also determine at what level GHGs should be kept and their corresponding concentrations in order to avoid climate events and crises. The used models are long short-term memory, autoencoders, and convolutional neural network (CNN). Alternatively, the dataset has been selected from U.K. National Centre for Earth Observation and Copernicus Climate Change Services. We compared the performance of these techniques and the best candidate was the Head—CNN; based on performance metrics such as root-mean-squared-error: 5.378, 2.395, and 15.923, mean-absolute-error: 4.157, 1.928, and11.672, Pearson: 0.368, 0.649, and 0.291, and R2 coefficient: 0.607, 0.806, and 0.539 for the ECVs temperature, CO2, and CH4, respectively. We were able to link the GHG emission to ECVs with high accuracy based on the reading of this geographic area.

Phase delay caused by atmospheric effects due to spatial and temporal variations of pressure, temperature, and water vapor content is one of the major error sources in estimation of ground deformation by interferometric synthetic aperture radar (InSAR>). Therefore, accuracy of ground deformation measurement is highly contingent on the robustness of the atmospheric correction techniques. These techniques rely either on auxiliary data such as numerical weather models (NWMs) or on the analysis of the interferometric phase itself. The accuracy in phase delays estimation of mixing effects of turbulent delay in atmosphere and stratified delay in lower troposphere is a key factor in determination of performance of each technique. Hence, the performance evaluation of the techniques is required in order to assess their potentials, robustness, and limitations. This article analyzes and evaluates the performance of four NWMs (i.e., ERA-Interim, ERA5, MERRA2, and WRF) and two phase-based techniques (i.e., linear and power law) to estimate phase delay using Sentinel-1A/B data over the Corvara landslide located in the Alps. The GPS data and GACOS product were used to validate the results. We generally found that ERA5 outperformed among other weather models with a phase standard deviation reduction of 77.7% (with respect to the InSAR phase), a correlation coefficient of 0.86 (between InSAR phase and estimated tropospheric delay) and a less significant error in the velocity estimation of the landslide.

Polarized reflectance ( $R_{\mathrm {p}}$ ) and degree of linear polarization (DOLP) provide essential information about polarized characteristics of land surfaces. For a given target, DOLP determines the magnitude of $R_{\mathrm {p}}$ . It has been proved that DOLP can be used for some remote monitoring cases that cannot be well detected with either non-polarized or polarized reflectance. Several bidirectional polarization distribution function (BPDF) models have been proposed in the last several decades to reproduce the angular distribution of $R_{\mathrm {p}}$ , but much less attention has been devoted to modeling and analyzing of DOLP. In this study, the Nadal–Breon BPDF model was transferred for calculating the DOLP of earth targets, and characteristics of DOLP were analyzed based on the modeling results. To evaluate the model’s feasibility, two experiments were executed: a fitting and a a priori modeling. The results showed good correlations ( $r>0.9$ ) between estimated and measured DOLP when the model was fitted with POLDER/PARASOL (a space-borne multi-angle multi-spectral polarimetric sensor) measurements. An increase of accuracy from 490 nm to 865 nm for fitting modeling was achieved and the highest accuracy was found at 865 nm for both experiments, with overall relative root mean square errors of 1.1 and 1.3 for fitting and a priori modeling, respectively. Class-based free parameters can be used for the a priori model of DOLP. The dispersion of the target-based free parameters controls the correlation of the a priori modeling results. Moreover, the maximum DOLP was found to be strongly determined by the corresponding bidirectional reflectance factor for every surface type ( $R^{2}=0.86$ ). This study provides an additional approach for obtaining DOLP from remote sensing platform and is helpful for studies of typical land surfaces.

This study presents two dynamic models, namely recurrent neural network and long short-term memory (LSTM) models, for predicting PM2.5 concentrations in Taiwan by using PM2.5 time series obtained at air quality monitoring stations and weather information obtained at neighboring weather stations. The proposed models can efficiently predict PM2.5 by incorporating a learned memory structure with a forgetting gate. To evaluate the predictive performance of the proposed models, large-scale databases established by Taiwan’s Environmental Protection Administration, and Central Weather Bureau were used; these databases include hourly data from 77 air quality monitoring stations and 580 weather stations over a 1-year period. The results demonstrated that the proposed models outperformed three traditional machine learning methods (gradient boosting, support vector machine, and classification and regression tree models) by 27.12% and 33.69% on average in terms of the coefficient of determination and root mean square error, respectively. A geographical divergence analysis was conducted to compare predictive performance in different regions. The results revealed that the most significant improvement in predictive performance was achieved in central Taiwan. The seasonal and pollution effect on predictive performance were reduced by the LSTM and the source distribution of PM2.5 emission in Taiwan was also analyzed.

The paper presents a control algorithm for a helicopter automatic approach and landing on a moving confined platform. It discusses landing on a sea vessel deck as a representative case for a mobile confined area. The dynamic model of a single rotor helicopter with a control system, developed in the FLIGHTLAB environment, and validated against flight tests data, is used to investigate control efficiency. The developed control method is based on the Linear Quadratic Regulator combined with prediction of motion of the landing area. An important part of the research was analysis of availability of the data needed for controlling the rotorcraft. The simulations of approach and landing on a moving vessel in various environmental conditions confirmed the efficiency of the developed control methodology.

Visibility is a measure of the transparency of the atmosphere, which is an important factor for road, air, and water transportation safety. Recently, features extracted from convolutional neural networks (CNNs) have obtained state-of-the-art results for the estimation of the visibility range for images of foggy weather. However, existing CNN-based approaches have only adopted visible images as observational data. Unlike these previous studies, in this paper, visible–infrared image pairs are used to estimate the visibility range. A novel multimodal deep fusion architecture based on a CNN is then proposed to learn the robust joint features of the two sensor modalities. Our network architecture is composed of two integrated residual network processing streams and one CNN stream, which are connected in parallel. In addition, we construct a visible–infrared multimodal dataset for various fog densities and label the visibility range. We then compare our proposed method with conventional deep-learning-based approaches and analyze the contributions of various observational and classical deep fusion models to the classification of the visibility range. The experimental results demonstrate that both accuracy and robustness can be strongly enhanced using the proposed method, especially for small training datasets.

The global horizontal irradiance data recorded at the earth's horizontal surface is a mixture of deterministic (extra-terrestrially) and stochastic components due to the ever-changing atmospheric conditions. The Box-Jenkins short memory stochastic models and their hybrid versions have been used successfully to forecast solar irradiance data. However, these models lack robustness and faulter for distant horizon forecasting such as more than 2 hours in the hourly case. Using a quantile regression model as both a robust benchmark and robustification model, we remedy this drawback by using long memory models and their hybrid versions in 24 hour ahead irradiance forecasting at three sites in South Africa (RVD, UPR and SUN). Results show that the inherent long memory is anti-persistent in all models but one that shows persistence. The forecasts from these long memory-based models are generally within the prediction interval of the benchmark model forecasts for two sites. Based on the RMSE, rRMSE, MAE, rMAE and the pinball loss function at the central quantile level and two extreme quantile levels, forecasts obtained through robustification via forecasts combinations with the quantile regression yield the best performance. The predictive ability of the best model before forecasts combination against the quantile regression model combined (robustified) model based Diebold-Mariano test rejected the null hypothesis of the same accuracy. As a follow up to distinguish which has a superior predictive accuracy, “murphy diagrams” were used. For the RVD and UPR sites, the model based on forecast combination has the best predictive accuracy while the quantile regression based benchmark model has the best accuracy for the SUN data.

Imaging of unmanned aerial vehicle easily suffer from haze, resulting in decline in the quality of required remote sensing images. The influence brings great challenges in later analysis and process. Although dark channel prior has acquired substantial achievements, some limitations, including imprecise estimation of atmospheric light, color distortion, and lower brightness of defogging image, still exist. In this article, to target these drawbacks, a novel defogging method for single image is proposed. First, a novel atmospheric scattering model is proposed to define the more accurate atmospheric light by introducing an adaptive variable strategy. Next, unlike traditional dark channel prior, a novel estimation method is presented by fusing dark and light channels to estimate more precise atmospheric light and transmittance. Then, we adopt the gray image corresponding to color image as a guidance image to refine the transmittance to further decrease the time complexity. Aiming at the region of low transmittance, a novel compensation function is created to improve the region of low transmittance and avoid color distortion. Moreover, a simple and effective calculation method is proposed to determine parameters in compensation function. Finally, the clear remote sensing image is established by an improved atmospheric scattering model. Extensive experiments on real-world datasets demonstrate that the proposed method outperforms several other state-of-the-art approaches both on subjective and objective quality evaluations.

Drones can play a game-changing role in reducing both cost and time in the context of last-mile deliveries. This paper addresses the last-mile delivery problem from a complex system viewpoint, where the collective performance of the drones is investigated. We consider a last-mile delivery system with a tradable permit model (TPM) for airspace use. Typically, in other research works regarding last-mile delivery drones, a fully cooperative centralized scenario is contemplated. In our approach, due to the TPM, the agents (i.e. drones) need to compete for airspace permits in a distributed manner. We simulate the system and evaluate how different parameters, such as the arrival rate and airspace dimensions, impact the system behavior in terms of the cost and time needed by the drones to acquire flight permits, and the airspace utilization. We use a simplified simulation model, where the agents' strategies are naïve, and the drones' flight dynamics are not accounted for. Nevertheless, the simulation's level of detail is adequate for capturing interesting properties from the agents' collective behavior, as our results support. The obtained results show that the system's performance is satisfactory, even with naïve agents and under high traffic conditions. Moreover, a real-world implementation of our competitive decentralized approach would lead to advantages, such as fast permit transactions, simple computational infrastructures, and error resilience.

This paper studies three-dimensional transient flow characteristics of a medical air-compressing nebulizer which is widely used to treat respiratory diseases in the medical field. The transient atomization process of air and water was numerically simulated and reproduced by solving Navier-Stokes equations, turbulent transport equations, the volume of fluid (VOF) model, and discrete phase model (DPM). The water volume fraction inside the nebulizer was simulated and demonstrated under different time. Besides, to better the working performance of the nebulizer, the effects of different working parameters such as different orifice diameters, length of the outflow pipe, and air velocity on the secondary atomization were also studied and discussed. It was found that the most suitable diameter of the orifice for this air-compressing nebulizer is 1.0 mm. Most suitable length of the outflow pipe and the air velocity for producing small air-water bubbles are 10 mm and 4 m/s, respectively.

Online aerodynamic parameter estimation plays an important role in compensating control system of aircraft under parameter uncertainties and unknown disturbance. In this paper, stability and control derivatives of aircraft are estimated online via support vector regression-numerical differential(SVR-ND) method. Small-sample real-time flight data reflecting real-time aerodynamic characteristics of aircraft is processed as training samples. For the small-size training samples, SVR technique is used for aerodynamic modeling. To pursue good performance in both computation efficiency and estimation accuracy, offline parameter estimation simulations are performed to select training sample size. It is observed that parameter estimation accuracy is related to the number of training samples and the noise level of samples. After that, an empirical formula is proposed to select training sample size according to results of simulations. To adapt the variation of samples, empirical formulas to tune hyper-parameters of SVR are presented based on the estimation of noise variance of samples. Finally, aerodynamic parameters are obtained by numerical differential in real-time. In a simulated maneuver, the proposed method is applied to online aerodynamic parameter estimation, and a Monte Carlo simulation is carried out to validate the robustness of SVR-ND method. Results indicate that the proposed method could realize accurate and robust estimation of aerodynamic parameters online.

Semiempirical, kernel-driven linear bidirectional reflectance distribution function (BRDF) models are widely used to characterize vegetation reflectance anisotropy and provide land surface bidirectional reflectance factor (BRF) products at the regional and global scales. However, these models usually imply an assumption of spherical leaf inclination. The effects of such an ideal assumption on simulating surface BRF remain inadequately quantified. In this article, we first evaluated the effects of leaf inclination on the most commonly used kernel-driven RossThick-LiSparse-Reciprocal (RTLSR) model by using the reflectance benchmark simulated by the mature PROSAIL (PROSAIL+SAIL) radiative transfer model. Subsequently, we improved the RTLSR model into a four-parameter version (RTLSRV4p) with a new volumetric scattering kernel derived from the assumption of vertical leaf inclination. Finally, the proposed RTLSRV4p model was validated by PROSAIL canopy BRF simulations, in situ canopy BRF measurements, and wide-angle infrared dual-mode line/area Aarray scanner (WIDAS) airborne observations. Validation results demonstrate that RTLSRV4p improves vegetation reflectance characterization for large leaf inclinations compared to the original RTLSR model, especially for the near-infrared (NIR) spectral domain. When validated against the simulated canopy BRFs, the mean root-mean-square error (RMSE), mean absolute percentage error (MAPE), bias, and coefficient of determination (R2) were improved from 0.0810, 31.63%, 0.0651, 0.6578 to 0.0453, 13.38%, 0.0326, 0.8734. Using the in situ BRF measurement, the fitted RMSE, MAPE, bias, and R2 were improved from 0.0917, 14.31%, 0.0728, and 0.5776 to 0.0226, 3.35%, 0.0166, and 0.9744. These validation metrics were improved from 0.0423, 10.85%, 0.0347, and 0.6598 to 0.0258, 5.85%, 0.0181, and 0.8732 when compared against the WIDAS observations. The RTLSRV4p model also shows good performance in calculating the reflectance of vegetation. These extensive validations suggest that RTLSRV4p is promising for capturing vegetation reflectance under large leaf inclinations.

This paper investigates a distributed event-triggered formation tracking problem of networked three-dimensional uncertain nonlinear stratospheric airships under directed networks. It is assumed that the nonlinearities of airship followers are unknown and the leader information can be obtained by only a subset of the airship followers. Approximation-based local adaptive tracking controllers with asynchronous event-triggering laws are developed to achieve the desired formations for both the positions and attitudes of uncertain stratospheric airship followers. We theoretically show that the stability and formation tracking performance of event-triggered closed-loop systems are ensured and Zeno behavior is excluded in the proposed asynchronous event-triggering mechanism. Finally, simulations illustrate the effectiveness of the proposed formation control protocol.

Fine-grained air pollution monitoring has attracted increasing attention worldwide. Even with an increasing amount of both static and mobile sensing systems, an inference algorithm is still essential to achieve a comprehensive understanding of the urban atmospheric environment. Conventional physical model-based methods are unable to involve all the influencing factors with limited prior knowledge, and data-driven methods lacking physical interpretation may result in bad generalization ability. This paper presents a multi-task learning scheme, which combines the physical model and the data-driven model with both merits. It enhances the data learning of a neural network with the aid of prior knowledge on atmospheric dispersion, and also controls the impact of the knowledge with a tunable weighting coefficient. Evaluations over a real-world deployment in Foshan, China show that, with the resolution of 500m $\times 500\text{m}\times 15$ min, the proposed method outperforms the state-of-the-art ones with 7.9% error reduction and 6.2% correlation increase. Benefited from the physical knowledge, the neural network obtains stable performance with lower variance, as well as higher robustness against negative background conditions.

The strip floating height is a key factor affecting production efficiency and quality of the air cushion furnace. At present, the air cushion furnace with hybrid nozzles is the most typical heating and drying equipment. In order to predict the floating height of the strip in this equipment, a hard-division and multi-model based floating height prediction method is proposed. In hard division method, the process is divided into several stable and vibration stages. On the one hand, time interval based novel density clustering algorithm is proposed so as to avoid the wrong division caused by noise at hard division stage. On the other hand, covariance matrices based local density is presented to better capture dynamic characteristics of the process. In the multi-model stage, a hybrid prediction model under stable state and XGBoost-based model under vibration state are established to predict the strip floating height. The hybrid prediction model is composed of mechanism model and data driven model. The mechanism model is proposed by combining the force balance equation and inviscid theory which can predict the major information of the floating height and the data driven model compensates the error of the mechanism model. A mount of experiments have been carried out on the existing air cushion experimental platform and show desirable prediction effect.

Geological hazard monitoring plays a pivotal role in preventing geological hazards. The main challenge faced by many experiments is to quantitatively monitor and analyze geological hazards. This paper proposes a quantitative monitoring indicator, called the ground surface stability anomaly (GSSA), for the first time and gives the identification criterion of the GSSA based on three stability factors. The features of the three stability factors, including geodetic height, ground gravity, and vertical deviation, reflect the relationship between load-induced changes and geological hazards. To verify the effectiveness and applicability of the GSSA quantitative monitoring method, a regional GSSA model was constructed based on the continuously operating reference station (CORS) network data and load impact data in southeastern Zhejiang, China. The larger the value of the GSSA is, the more likely geological hazards are to occur, especially when an abnormal dynamic environment (such as heavy rainfall, sea level anomaly, atmospheric pressure anomaly, or spring tide) is encountered. By comparing the GSSA with the potential trouble points of geological hazards and 40 geological hazard events that have occurred, the results show that the method can quantitatively, accurately, and continuously monitor the GSSA in an area covered by the CORS network, and the method has the ability to capture the precursors to geological hazards.

The accurate prediction of short-term power load is of great significance for the smooth operation of the power market. LightGBM model is optimized by cuckoo search algorithm, and a new short-term power load forecasting model is established. The model selects nine factors that may cause short-term power load changes, including average temperature, humidity, and air pressure in many places in Spain, and uses Grey Relation Analysis to verify the correlation between each factor and short-term power load changes. Taking nine factors as inputs, the cuckoo algorithm optimized LightGBM (CS-LightgBM) model was verified. The results show that the MAPE (mean absolute percentage error) and RMSE (root mean square error) of the CS-LightGBM model are 4.0519% and 532.7607MW, respectively. Compared with other traditional machine learning models, it has higher prediction accuracy and provides a new idea for power load forecasting.

In September 2019, the German research icebreaker Polarstern started the largest multidisciplinary Arctic expedition, the MOSAiC (Multidisciplinary drifting Observatory for the Study of Arctic Climate) drift experiment. Being moored to ice floes at high Arctic for a whole year, thus including the winter season, the main goal of the expedition is to better understand and quantify relevant processes within the atmosphere-ice-ocean system that impact the sea ice, ultimately leading to improved climate models. Satellite remote sensing, specially multi-frequency synthetic aperture radar (SAR) plays a major role to achieve this goal. Two major objectives in SAR based remote sensing of sea ice is on the one hand to have a large coverage, and on the other hand to obtain a radar response that carries as much information as possible. A comprehensive set of C- and L- band SAR images were acquired during the course of MOSAiC. In this initial study we evaluate the effects of seasonal changes on C- and L-band backscatter in respect to three different sea ice types, i.e., Young Ice, Smooth Ice and Rough/Deformed Ice along with the performance of sea ice type retrieval of a established algorithm. Areas of deformed, smooth and young sea ice were observed in the vicinity of R/V Polarstern and these areas are included whenever possible in the yearlong time series. For both frequencies a change in all backscatter channels values can be observed during the early melt season. This is first noticeable in the C-band images and later followed by a change in the L-band images, probably caused by their different penetration depth and volume scattering sensitivities.

The proposal of this this study is to develop an innovative strategy to achieve energy saving for air conditioning (AC). In this research, an eco-wall and cooling model of building has been developed to consider the outside temperature (Tout) and thermal characteristics of a building. To evaluate the electrical energy cost (EC), discrete time optimization was applied to find the minimum electrical EC for the air conditioning (AC). As a result, the electrical EC of AC can be minimized to be IDR 25,200. In addition, this model has been tested on a sample building in Barru, South Sulawesi.

Taxi-out time prediction is one of the key elements required for successful implementation of Airport Collaborative Decision Making (A-CDM) system. In this paper, we discuss an application of a Machine Learning technique called Random Forest algorithm to predict taxi-out time. Using information retrieved from Incheon International airport’s A-CDM and METAR databases, the model is trained and tested. This paper presents the process of selecting explanatory variables (i.e., features) using feature importance and demonstrates the prediction accuracy of the resulting model.

Estimating airspace capacity under the impact of severe weather is a crucial method to guarantee efficiency and safety of air traffic operation. It is also of vital importance to understand the spatial and temporal dependencies of adjacent airspace in the process of estimation. Therefore, in this paper, considering this spatio-temporal dependencies, a two-step airspace capacity estimation method is proposed to evaluate terminal areas’ capacity including airport and terminal sectors. The first step is to translate different types of meteorological conditions into operation features depending on the type of target airspace. Maxflow/Mincut theorem and Gradient Boosting Regression are adopted for sectors and airports, respectively. Based on the extracted features, an improved Spatio-temporal Graph Convolution Networks via Initial residual (STGCNI) is developed aiming to fit the characteristics in air traffic field. Finally, through experiments with real data of Chengdu Terminal Area (ZUUUAP), both the validity and enhanced estimation accuracy of the method are verified by comparisons with other models, which exclude spatial and temporal characteristics.