Search

The Energy Exascale Earth System Model (E3SM) developed by the Department of Energy has a goal of addressing challenges in understanding the global water cycle. Success depends on correct simulation of cloud and precipitation elements. However, lack of appropriate evaluation metrics has hindered the accurate representation of these elements in general circulation models. We derive metrics from the three-dimensional data of the ground-based Next-Generation Radar (NEXRAD) network over the US to evaluate both horizontal and vertical structures of precipitation elements. We coarsened the resolution of the radar observations to be consistent with the model resolution and improved the coupling of the Cloud Feedback Model Intercomparison Project Observation Simulator Package (COSP) and E3SM Atmospheric Model Version 1 (EAMv1) to obtain the best possible model output for comparison with the observations. Three warm seasons (2014–2016) of EAMv1 simulations of 3-D radar reflectivity features at an hourly scale are evaluated. A general agreement in domain-mean radar reflectivity intensity is found between EAMv1 and NEXRAD below 4 km altitude; however, the model underestimates reflectivity over the central US, which suggests that the model does not capture the mesoscale convective systems that produce much of the precipitation in that region. The shape of the model-estimated histogram of subgrid-scale reflectivity is improved by correcting the microphysical assumptions in COSP. Different from previous studies that evaluated modeled cloud top height, we find the model severely underestimates radar reflectivity at upper levels – the simulated echo top height is about 5 km lower than in observations – and this result is not changed by tuning any single physics parameter. For more accurate model evaluation, a higher-order consistency between the COSP and the host model is warranted in future studies.

Regional seasonal forecasting requires accurate simulation of the variability of local climate drivers. The South Pacific Convergence Zone (SPCZ) is a large region of low-level convergence, clouds and precipitation in the South Pacific, whose effects extend as far as northeast Australia (NEA). The location of the SPCZ is modulated by the El Niño-Southern Oscillation (ENSO) which causes rainfall variability in the region. Correctly simulating the ENSO-SPCZ teleconnection and its interplay with local conditions is essential for improving seasonal rainfall forecasts. Here we analyse the ability of the ACCESS-S1 seasonal forecast system to predict the SPCZ’s relationship with ENSO including its latitudinal shifts, zonal slope and rainfall magnitude between 1990 and 2012 for the December–January–February (DJF) season. We found improvements in ACCESS-S1’s SPCZ prediction capability compared to its predecessor (POAMA), although prediction of the slope is still limited. The inability of ACCESS-S1 to replicate seasons with a strong anti-zonal SPCZ slope is attributed to its atmospheric model. This has implications for accurate seasonal rainfall forecasts for NEA and South Pacific Islands. Future challenges in seasonal prediction facing regional communities and developers of coupled ocean–atmosphere forecast models are discussed.

Identifying, detecting, and localizing extreme weather events is a crucial first step in understanding how they may vary under different climate change scenarios. Pattern recognition tasks such as classification, object detection, and segmentation (i.e., pixel-level classification) have remained challenging problems in the weather and climate sciences. While there exist many empirical heuristics for detecting extreme events, the disparities between the output of these different methods even for a single event are large and often difficult to reconcile. Given the success of deep learning (DL) in tackling similar problems in computer vision, we advocate a DL-based approach. DL, however, works best in the context of supervised learning – when labeled datasets are readily available. Reliable labeled training data for extreme weather and climate events is scarce. We create “ClimateNet” – an open, community-sourced human-expert-labeled curated dataset that captures tropical cyclones (TCs) and atmospheric rivers (ARs) in high-resolution climate model output from a simulation of a recent historical period. We use the curated ClimateNet dataset to train a state-of-the-art DL model for pixel-level identification – i.e., segmentation – of TCs and ARs. We then apply the trained DL model to historical and climate change scenarios simulated by the Community Atmospheric Model (CAM5.1) and show that the DL model accurately segments the data into TCs, ARs, or “the background” at a pixel level. Further, we show how the segmentation results can be used to conduct spatially and temporally precise analytics by quantifying distributions of extreme precipitation conditioned on event types (TC or AR) at regional scales. The key contribution of this work is that it paves the way for DL-based automated, high-fidelity, and highly precise analytics of climate data using a curated expert-labeled dataset – ClimateNet. ClimateNet and the DL-based segmentation method provide several unique capabilities: (i) they can be used to calculate a variety of TC and AR statistics at a fine-grained level; (ii) they can be applied to different climate scenarios and different datasets without tuning as they do not rely on threshold conditions; and (iii) the proposed DL method is suitable for rapidly analyzing large amounts of climate model output. While our study has been conducted for two important extreme weather patterns (TCs and ARs) in simulation datasets, we believe that this methodology can be applied to a much broader class of patterns and applied to observational and reanalysis data products via transfer learning.

Various uncertainties exist in a hindcast due to the inabilities of numerical models to resolve all the complicated atmosphere-sea interactions, and the lack of certain ground truth observations. Here, a comprehensive analysis of an atmospheric model performance in hindcast mode (Hurricane Weather and Research Forecasting model—HWRF) and its 40 ensembles during severe events is conducted, evaluating the model accuracy and uncertainty for hurricane track parameters, and wind speed collected along satellite altimeter tracks and at stationary source point observations. Subsequently, the downstream spectral wave model WAVEWATCH III is forced by two sets of wind field data, each includes 40 members. The first ones are randomly extracted from original HWRF simulations and the second ones are based on spread of best track parameters. The atmospheric model spread and wave model error along satellite altimeters tracks and at stationary source point observations are estimated. The study on Hurricane Irma reveals that wind and wave observations during this extreme event are within ensemble spreads. While both Models have wide spreads over areas with landmass, maximum uncertainty in the atmospheric model is at hurricane eye in contrast to the wave model.

Snow depth over sea ice is an essential variable for understanding the Arctic energy budget. In this study, we evaluate snow depth over Arctic sea ice during 1993–2014 simulated by 31 models from phase 6 of the Coupled Model Intercomparison Project (CMIP6) against recent satellite retrievals. The CMIP6 models capture some aspects of the observed snow depth climatology and variability. The observed variability lies in the middle of the models’ simulations. All the models show negative trends of snow depth during 1993–2014. However, substantial spatiotemporal discrepancies are identified. Compared to the observation, most models have late seasonal maximum snow depth (by two months), remarkably thinner snow for the seasonal minimum, an incorrect transition from the growth to decay period, and a greatly underestimated interannual variability and thinning trend of snow depth over areas with frequent occurrence of multi-year sea ice. Most models are unable to reproduce the observed snow depth gradient from the Canadian Arctic to the outer areas and the largest thinning rate in the central Arctic. Future projections suggest that snow depth in the Arctic will continue to decrease from 2015 to 2099. Under the SSP5-8.5 scenario, the Arctic will be almost snow-free during the summer and fall and the accumulation of snow starts from January. Further investigation into the possible causes of the issues for the simulated snow depth by some models based on the same family of models suggests that resolution, the inclusion of a high-top atmospheric model, and biogeochemistry processes are important factors for snow depth simulation.

This paper presents an inter-comparison performance assessment of the newly developed Centre for Weather Forecast and Climate Studies (CPTEC) model (the Brazilian Atmospheric Model version 1.2, BAM-1.2) against four sub-seasonal to seasonal (S2S) prediction project models from: Japan Meteorological Agency (JMA), Environmental and Climate Change Canada (ECCC), European Centre for Medium-range Weather Forecasts (ECMWF) and Australian Bureau of Meteorology (BoM). The inter-comparison was performed using hindcasts of weekly precipitation anomalies and the daily evolution of Madden–Julian Oscillation (MJO) for 12 extended austral summers (November–March, 1999/2000–2010/2011), leading to a verification sample of 120 hindcasts. The deterministic assessment of the prediction of precipitation anomalies revealed ECMWF as the model presenting the highest (smallest) correlation (root mean squared error, RMSE) values among all examined models. JMA ranked as the second best performing model, followed by ECCC, CPTEC and BoM. The probabilistic assessment for the event “positive precipitation anomaly” revealed that ECMWF presented better discrimination, reliability and resolution when compared to CPTEC and BoM. However, these three models produced overconfident probabilistic predictions. For MJO predictions, CPTEC crosses the 0.5 bivariate correlation threshold at around 19 days when using the mean of 4 ensemble members, presenting similar performance to BoM, JMA and ECCC. Overall, CPTEC proved to be competitive compared to the S2S models investigated, but with respect to ECMWF there is scope to improve the prediction system, likely by a combination of including coupling to an interactive ocean, improving resolution and model parameterization schemes, and better methods for ensemble generation.

Prediction of incidences of thunderstorms using different techniques is quite well established. To add to this knowledge, in this paper, a heuristic equation is proposed which relates the correlation coefficient of atmospheric parameters with the number of hourly incidences of thunderstorms. There are four ways to compute the indices values from the proposed heuristic equation. These indices values are used in the classification of the hourly incidences of thunderstorms. The proposed equation and indices work well, as tested on pre-monsoon hourly atmospheric data from 2018 and validated with hourly data from April 2019 and 2020. From the four indices, the first index value is computed with normalized average values of parameters of only hourly incidence of thunderstorms data of 2016-2017 in the month-wise method. The other three indices use optimization techniques namely, the Teaching Learning Based Optimization (TLBO) technique, Differential Evaluation (DE), and Simulated Annealing (SA). TLBO shows better classification of hourly incidences of thunderstorms for 2018 atmospheric data. TLBO is also precisely validating the hourly incidences of thunderstorms for April 2019 and April 2020 hourly atmospheric data. It performed better with 2020 data by 88%. The variations of atmospheric parameters before, after or, during the incidences of thunderstorms are also depicted.

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.

Carbon fiber composites (CFCs) are widely used in the aeronautics industry because of their excellent properties, including low weight and better corrosion and fatigue resistance compared with metals such as aluminum. However, metals are still required in primary aircraft components when, for example, damage tolerance designs are used. In this study, CFC structures used in aircraft are simulated and experimental tests are carried out. The direct and indirect effects of a high current discharge through the fuselage of an aircraft are compared with experimental data to validate the proposed simulation model.

Removing rain from images is of a great importance to various applications such as autonomous driving, drone piloting, and photo editing. Conventional methods rely on some heuristics to handcraft various priors to remove or separate rain from images. Recently, deep learning models are proposed to learn various end-to-end methods to complete this task. However, these methods might fail in obtaining satisfactory results in some real-world scenarios, especially when the captured images suffer from heavy rain that brings not only rain streaks but also a haze-like effect (caused by the accumulation of tiny raindrops). Different from most of the existing deep learning deraining methods that focus on handling rain streaks, we add a new variable to model the haze-like effect in a general model for rain, based on which a deep neural network is designed accordingly. Specifically, in our method, two branches are designed to handle rain streaks and the haze-like effect, respectively. The output of such branch structure is fed to an additional module to further enhance the performance. Three modules are trained jointly, leading to an end-to-end network, which supports a adjustment to the strength of removing the haze-like effect. Extensive experiments on several datasets show that our method outperforms several state-of-the-art methods in both objective assessment and visual quality.

Long-range transport of biomass burning (BB) aerosol from regions affected by wildfires is known to have a significant impact on the radiative balance and air quality in receptor regions. However, the changes that occur in the optical properties of BB aerosol during long-range transport events are insufficiently understood, limiting the adequacy of representations of the aerosol processes in chemistry transport and climate models. Here we introduce a framework to infer and interpret changes in the optical properties of BB aerosol from satellite observations of multiple BB plumes. Our framework includes (1) a procedure for analysis of available satellite retrievals of the absorption and extinction aerosol optical depths (AAOD and AOD) and single-scattering albedo (SSA) as a function of the BB aerosol photochemical age and (2) a representation of the AAOD and AOD evolution with a chemistry transport model (CTM) involving a simplified volatility basis set (VBS) scheme with a few adjustable parameters. We apply this framework to analyze a large-scale outflow of BB smoke plumes from Siberia toward Europe that occurred in July 2016. We use AAOD and SSA data derived from OMI (Ozone Monitoring Instrument) satellite measurements in the near-UV range along with 550 nm AOD and carbon monoxide (CO) columns retrieved from MODIS (Moderate Resolution Imaging Spectroradiometer) and IASI (Infrared Atmospheric Sounding Interferometer) satellite observations, respectively, to infer changes in the optical properties of Siberian BB aerosol due to its atmospheric aging and to get insights into the processes underlying these changes. Using the satellite data in combination with simulated data from the CHIMERE CTM, we evaluate the enhancement ratios (EnRs) that allow isolating AAOD and AOD changes due to oxidation and gas–particle partitioning processes from those due to other processes, including transport, deposition, and wet scavenging. The behavior of EnRs for AAOD and AOD is then characterized using nonlinear trend analysis. It is found that the EnR for AOD strongly increases (by about a factor of 2) during the first 20–30 h of the analyzed evolution period, whereas the EnR for AAOD does not exhibit a statistically significant increase during this period. The increase in AOD is accompanied by a statistically significant enhancement of SSA. Further BB aerosol aging (up to several days) is associated with a strong decrease in EnRs for both AAOD and AOD. Our VBS simulations constrained by the observations are found to be more consistent with satellite observations of strongly aged BB plumes than “tracer” simulations in which atmospheric transformations of BB organic aerosol were disregarded. The simulation results indicate that the upward trends in EnR for AOD and in SSA are mainly due to atmospheric processing of secondary organic aerosol (SOA), leading to an increase in the mass scattering efficiency of BB aerosol. Evaporation and chemical fragmentation of the SOA species, part of which is assumed to be absorptive (to contain brown carbon), are identified as likely reasons for the subsequent decrease in the EnR for both AAOD and AOD. Hence, our analysis reveals that the long-range transport of smoke plumes from Siberian fires is associated with major changes in BB aerosol optical properties and chemical composition. Overall, this study demonstrates the feasibility of using available satellite observations for evaluating and improving representations in atmospheric models of the BB aerosol aging processes in different regions of the world at much larger temporal scales than those typically addressed in aerosol chamber experiments.

In recent years, human health across the world is becoming concerned by a constant threat of air pollution, which causes many chronic diseases and premature mortalities. Poor air quality does not have only serious adverse effects on human health and vegetation but also some major negative political, societal, and economic impacts. Hence, it is essential to invest more effort on accurate forecasting of ambient air pollution to provide practical and relevant solutions, achieve acceptable air quality, and plan for prevention. In this work, we propose a flexible and efficient deep learning-driven model to forecast concentrations of ambient pollutants. This article introduces first the traditional variational autoencoder (VAE) and the attention mechanism to develop the forecasting modeling strategy based on the innovative integrated multiple directed attention (IMDA) deep learning architecture. To assess the performance of the proposed forecasting methodology, experimental validation is then performed using air pollution data from four US states. Six statistical indicators have been used to evaluate the forecasting accuracy. A discussion of the results obtained finally demonstrates the satisfying performance of integrated multiple directed attention variational autoencoder (IMDA-VAE) methods to forecast different pollutants in different locations. Furthermore, results indicate that the proposed IMDA-VAE model can effectively improve air pollution forecasting performance and outperforms the deep learning models, namely VAE, long short-term memory (LSTM), gated recurrent units (GRU), bidirectional LSTM, bidirectional GRU, and ConvLSTM. We also showed that the forecasting results of the proposed model surpass the performance of LSTM and GRU with the attention mechanism.

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.

Energy submetering at the equipment level provides a tool to identify energy use anomalies and detect operational inefficiencies. While physical meters can be costly and difficult to install, virtual meters (VMs) overcome practical issues associated with physical meters and provide insights into critical unmeasured quantities. This article introduces an inverse greybox model-based virtual metering method to estimate the energy in an air handling unit (AHU). Models that represent the components typically found in AHUs are formulated using a data set from a highly instrumented AHU and combined into an integrated greybox model. The use of the integrated model to create VMs is demonstrated by using a data set from an independent AHU located in a large office building in Ottawa, ON, Canada. Model parameters are estimated by using the genetic algorithm and used in creating VMs that can estimate the heat supplied/extracted at the AHU level. In addition, the model is used to estimate a monthly average outdoor air fraction used by the AHU. The potential of the component models and VMs to detect operational inefficiencies and support operational decisions is demonstrated through illustrative examples. Note to Practitioners —This article presents a novel virtual metering algorithm to estimate the heating and cooling energy at the air handling unit (AHU) level. This virtual metering algorithm fills a gap in the literature and provides a tool that will help detect and interpret energy use anomalies, identify operational inefficiencies, and guide on-going commissioning of building energy systems. Facility managers, operators, and other stakeholders can use the insights gained from virtual metering to improve building operational performance. Future planned research includes developing virtual meters to characterize energy flows at the zone level and visualization methods to make inverse modeling results more accessible to different building stakeholders.

Uncertainty in the representation of biomass burning (BB) aerosol composition and optical properties in climate models contributes to a range in modeled aerosol effects on incoming solar radiation. Depending on the model, the top-of-the-atmosphere BB aerosol effect can range from cooling to warming. By relating aerosol absorption relative to extinction and carbonaceous aerosol composition from 12 observational datasets to nine state-of-the-art Earth system models/chemical transport models, we identify varying degrees of overestimation in BB aerosol absorptivity by these models. Modifications to BB aerosol refractive index, size, and mixing state improve the Community Atmosphere Model version 5 (CAM5) agreement with observations, leading to a global change in BB direct radiative effect of −0.07 W m−2, and regional changes of −2 W m−2 (Africa) and −0.5 W m−2 (South America/Temperate). Our findings suggest that current modeled BB contributes less to warming than previously thought, largely due to treatments of aerosol mixing state., Wildfires produce aerosols known to impact the climate, but the wider-reaching effects of this biomass burning are poorly constrained in models. Here the authors use a suite of observations from 12 campaigns around the globe to determine that the values used by most climate models overestimate the contribution of biomass burning aerosols.

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.

Errors in soil moisture adversely impact the modeling of land–atmosphere water and energy fluxes and, consequently, near-surface atmospheric conditions in atmospheric data assimilation systems (ADAS). To mitigate such errors, a land surface analysis is included in many such systems, although not yet in the currently operational NASA Goddard Earth Observing System (GEOS) ADAS. This article investigates the assimilation of L-band brightness temperature (Tb) observations from the Soil Moisture Active Passive (SMAP) mission in the GEOS weakly coupled land–atmosphere data assimilation system (LADAS) during boreal summer 2017. The SMAP Tb analysis improves the correlation of LADAS surface and root-zone soil moisture versus in situ measurements by ∼0.1–0.26 over that of ADAS estimates; the unbiased root-mean-square error of LADAS soil moisture is reduced by 0.002–0.008 m3/m3 from that of ADAS. Furthermore, the global land average RMSE versus in situ measurements of screen-level air specific humidity (q2m) and daily maximum temperature (T2mmax) is reduced by 0.05 g/kg and 0.04 K, respectively, for LADAS compared to ADAS estimates. Regionally, the RMSE of LADAS q2m and T2mmax is improved by up to 0.4 g/kg and 0.3 K, respectively. Improvement in LADAS specific humidity extends into the lower troposphere (below ∼700 mb), with relative improvements in bias of 15–25%, although LADAS air temperature bias slightly increases relative to that of ADAS. Finally, the root mean square of the LADAS Tb observation-minus-forecast residuals is smaller by up to ∼0.1 K than in a land-only assimilation system, corroborating the positive impact of the Tb analysis on the modeled land–atmosphere coupling.

Optical wireless networks realized by means of gigabit optical wireless communication (OWC) systems are becoming, in a variety of applications, an important alternative, or a complementary solution, to their fiber-based counterparts. However, performance of the OWC systems can be considerably degraded in periods of unfavorable weather conditions, such as heavy fog, which temporarily reduce the effective capacity of the network. In this paper, we focus on optical wireless mesh networks that use terrestrial links (called FSO – free-space optical – links) composed of several parallel full-duplex FSO systems, and present two complementary solutions that together provide a means to maximize network traffic performance in various weather conditions encountered during network operation. The first solution is a method for estimating the degradation of the effective FSO link capacity in adverse weather conditions such as fog, rain and snow (called the weather states in this paper). The second solution is an optimization model aiming at maximizing the network traffic throughput for a given list of weather states, derived from the conducted measurements. The model assumes the so-called affine flow thinning (AFT) traffic routing and protection mechanism capable of controlling the end-to-end traffic flows in response to fluctuations of capacity available on FSO links caused by changing weather conditions. The proposed link capacity modeling approach and the elaborated optimization model are verified through an exhaustive numerical study, illustrating the trade-off between the increase of traffic performance of the FSO networks and the corresponding cost of additional OWC systems.

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.

Although advanced single image deraining methods have been proposed, their generalization ability to real-world images is usually limited, especially when dealing with rain patterns of different densities, shapes, and directions. In order to improve the robustness and generalization of these deraining methods, we propose a novel density-aware single image deraining method with gated multi-scale feature fusion, which consists of two stages. In the first stage, a sophisticated physics model is leveraged for initial deraining and a network branch is utilized for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), attempting to eliminate artifacts and improve the restoration quality. Extensive experiments have been conducted on the representative synthetic rain datasets and real rain scenes, demonstrating the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-arts.

A sensor for monitoring the dynamics of changes in the concentration of carbon dioxide in humans during respiration is presented. The sensor is based on a matched pair of LED-emitter-receiver. It was revealed that the shape of the nasal cavity affects the hydrodynamics and heat transfer in the respiratory system. It has been shown that dynamic parameter analysis is necessary for the diagnosis of respiratory diseases.

We propose a novel method to calculate the electromagnetic (EM) wave propagation from low-earth orbit satellite (LEO) to a ground station based on the physical optics (PO), ray tracing technique, and geometric optics (GO) considering interpolated atmospheric environments. Our method includes the reflector antenna analysis using PO, the interpolation of the meteorological data using PCHIP and Kriging interpolation, transmission analysis using ray tracing and geometrical optics. Tropospheric and stratospheric environments are modeled using meteorological data—air pressure and temperature, relative humidity, and rain rate—measured at 9 different radiosonde observatories in and around South Korea. Furthermore, we utilize Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) and Kriging-exponential methods for vertical and horizontal interpolations of the raw meteorological data, respectively. Hence, the interpolated atmospheric environments are amenable to the best use of ray tracing technique and GO. Subsequently, effective refractive indices of the stratified media can be extracted via millimeter-propagation-model93. The simplified Appleton-Hartree equation characterizes the ionospheric environment. Considering a sun-synchronous orbit satellite passing through South Korea, we calculate atmospheric attenuation, boresight error, received power, and compensation angle of satellite antenna for various conditions.

The rising temperature is one of the key indicators of a warming climate, capable of causing extensive stress to biological systems as well as built structures.Ambient temperature collected at ground level can have higher variability than regional weather forecasts, which fail to capture local dynamics. There remains a clear need for accurate air temperature prediction at the suburban scale at high temporal and spatial resolutions. This research proposed a framework based on a long short-term memory (LSTM) deep learning network to generate day-ahead hourly temperature forecasts with high spatial resolution. Air temperature observations are collected at a very fine scale (~150m) along major roads of New York City (NYC) through the Internet of Things (IoT) data for 2019-2020. The network is a stacked two layer LSTM network, which is able to process the measurements from all sensor locations at the same time and is able to produce predictions for multiple future time steps simultaneously. Experiments showed that the LSTM network outperformed other traditional time series forecasting techniques, such as the persistence model, historical average, AutoRegressive Integrated Moving Average (ARIMA), and feedforward neural networks (FNN). In addition, historical weather observations are collected from in situ weather sensors (i.e., Weather Underground, WU) within the region for the past five years. Experiments were conducted to compare the performance of the LSTM network with different training datasets: 1) IoT data alone, or 2) IoT data with the historical five years of WU data. By leveraging the historical air temperature from WU, the LSTM model achieved a generally increased accuracy by being exposed to more historical patterns that might not be present in the IoT observations. Meanwhile, by using IoT observations, the spatial resolution of air temperature predictions is significantly improved.

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.

Global pressure and temperature (GPT) series models can provide the underlying meteorological parameters for tropospheric corrections without any other meteorological observations, which allows them to be widely used for a series of geodetic as well as meteorological and climatological purposes. Due to the height difference between the empirical model height and user location, a vertical correction of meteorological parameters is inevitable, particularly for airborne users. Unfortunately, the GPT series models have limitations on the vertical correction. We explored the temperature lapse rate for the vertical adjustment using 10 years of reanalysis data provided by the National Centers for Environmental Prediction (NCEP), and extended the GPT models to improved global pressure and temperature (IGPT) series models by introducing a new temperature lapse rate model and a new formulation of pressure reduction. An evaluation of the IGPT models expression determines that the IGPT models have better accuracy than the GPT models, particularly under large height differences, which is attributed to their ability to consider the real behavior of temperature in the atmosphere and adiabatic effects on air pressure. The performance of the IGPT models in zenith tropospheric delay (ZTD) estimations was also evaluated by comparison with the fifth-generation European Centers for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA5) data and International GNSS Service (IGS) data. The results confirm that our new models can effectively improve the accuracy of ZTDs, particularly at larger altitude differences between the target height and the corresponding four grid points of the model, not only enhancing the performance of the model in complex terrain but also extending the feasibility of the IGPT models from the Earth’s surface to higher altitudes.

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.

Images acquired from the outdoors are often degraded owing to bad weather conditions, such as haze, fog, and rain. Various algorithms for dehazing daytime images have been introduced to eliminate the effects of haze. However, unlike daytime conditions, where sunlight is the main source of light, there are multiple regional light sources in nighttime conditions. Thus, it is difficult to effectively eliminate the effects of haze at night using dehazing algorithms that primarily target daytime hazy images. In addition, most nighttime dehazing algorithms adopt the dark channel prior (DCP) to estimate the transmission, but this approach has problems due to the spatially variant illuminations. In this paper, we propose a nighttime single image dehazing algorithm based on the structural patch decomposition. First, atmospheric light is estimated by separating the intensity and color of light. To estimate the transmission, the candidates for the transmission values are set, and the final transmission value is obtained based on a weighted summation of these candidates for each image pixel. The weights of the candidates are determined according to the structural patch information of the scene radiances estimated with these candidates. Using this approach, an appropriate transmission map can be obtained, and the dehazing procedure is shown to be robust to the mixed illumination conditions observed in the nighttime conditions. The experimental results show that the proposed algorithm effectively eliminate the effect of haze in images based on quantitative and qualitative evaluations.

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.

Image dehazing algorithms based on dark channel prior principle have achieved good results for most scenes. However, the popular dark channel prior tends to underestimate transmissions of bright areas or objects, such as the skies, white areas and self-luminous bodies, which may cause color distortions during dehazing. A complementary prior called the extreme reflectance channel prior (ERC), which combines the dark channel prior with the bright channel prior, is proposed to estimate the transmission map. The extreme reflectance channel is the union of dark and bright channel’s pixels which satisfy the corresponding channel. Based on the scattering analysis results that the intensities of pixels in ERC are often close to 0 or 1 for the natural haze-free images or close to global atmospheric light if hazes occur in the air, the pixels in a hazy image can be recovered according to ERC to calculate the transmission map and then solve the haze imaging mode. Experiments show that ERC method outperforms state-of-the-art methods in PSNR, SSIM and visual perception effects.

High spatiotemporal resolution land surface temperature (LST) plays an important role in various environment applications. However, the limitation of thermal infrared sensors and the effect of clouds and other atmospheric conditions result in discontinuous daily thermal observations of the Moderate Resolution Imaging Spectroradiometer (MODIS). Annual temperature cycle (ATC) models can help to supply daily continuous LSTs via limited observations, but these ATC models seldom consider the disturbance of weather conditions or the land cover change. On the other hand, spatial interpolation techniques also limit in implementation when available data in one day or several days are not able to obtain enough spatiotemporal information for LST reconstruction. The objective of this study is to propose a phenology-based ATC model (termed PATC), which takes the phenology change and local weather change into account, to reconstruct daily unscanned LSTs at an annual scale. Daily MODIS LSTs collected in 2015 were utilized to analyze the performance of PATC compared with other ATC models. Results show that PATC improved the accuracy by 1.6 and 0.5 K compared to the classic ATC model in the daytime and nighttime, respectively. Compared to the enhanced ATC model, PATC also shows better performance with higher accuracies, especially during the growing season of vegetation in the daytime. Future research may focus on an incorporation with Landsat observations and diurnal temperature cycle models to implement LST reconstruction at a diurnal scale.

The simulation and prediction of winter Northern Hemisphere atmospheric blocking in the seasonal prediction systems from the European Centre for Medium‐Range Weather Forecasts (ECMWF) is analysed. Blocking statistics from the operational November‐initialised seasonal hindcasts are evaluated in three generations of models: System3, System4, and System5 (SEAS5). Improvements in the climatological representation of blocking are observed in the most recent model configurations, with reduced bias over North Pacific and Greenland. Minor progress is seen over the European sector, where SEAS5 still underestimates the observed blocking frequency. SEAS5 blocking interannual variability is underestimated too and is proportional to the climatological frequency, highlighting that a negative bias in the blocking frequency implies an underestimation of the interannual variance. SEAS5 predictive skill and signal‐to‐noise ratio remain low, but interesting positive results are found over Western and Central Europe. Improved forecasts with reduced ensemble spread are obtained during El Niño years, especially at low latitudes. Complementary experiments show that the statistics of blocking are improved following atmospheric and oceanic resolution increase. Conversely, they remain largely insensitive to coupled model sea‐surface temperature (SST) errors. On the other hand, the implementation of stochastic parameterisations tends to displace blocking activity equatorward. Finally, by comparing seasonal hindcasts with climate runs using the same model, we highlight that the largest contributors to the chronic underestimation of blocking are persistent errors in the atmospheric model. It is also shown that SST errors have a larger impact on blocking bias in climate runs than in seasonal runs, and that increased ocean model resolution contributes to improved blocking more effectively in climate runs. Seasonal forecasts can thus be considered a suitable test‐bed for model development targeting blocking improvement in climate models.

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.

As ground-level O3 has become one of the major pollutants affecting air quality in recent years, monitoring and revealing the spatial distribution pattern of O3 are of great significance to study air-pollution characteristics. Based on the multilayer backpropagation neural network, one O3 estimation model is proposed to obtain the continuous spatial distribution of O3 concentrations, where the Landsat 8 images, meteorological parameters, and air quality data have been integrated together as the input for the model training and validation. In order to enhance the estimation accuracy, the proposed model has been optimized with respect to the influencing factors and spatial extent. In the test areas of Beijing, Tianjin, and Shijiazhuang of China, the proposed O3 estimation model has demonstrated quite satisfactory performance—with the average coefficient of determination (R2) larger than 0.90 and root-mean-square error smaller than 19.0 μg/m3. It is worth mentioning that all the data employed in this research are freely available and can be applied nationwide in the mainland of China. Taking advantage of the generic nature and the positive O3 estimation results with high accuracy and spatial resolution, the proposed model can be expected to be a new way for studying air-pollution characteristics and, thus, support the decision making for environmental governance.

Monitoring pollution of PM2.5 on walkable roads is important for resident health in high-density cities. Due to the spatiotemporal resolution limitations of aerosol optical depth observation, fixed-point monitoring, or traditional mobile measurement instruments, the microscale PM2.5 distribution in the walking environment cannot be fully estimated at the fine scale. In this article, by the integration of mobile measurement data, OpenStreetMap (OSM) data, Landsat images, and other multisource data in land-use regression (LUR) models, a novel framework is proposed to estimate and map PM2.5 distribution in a typical microscale walkable environment of the high-density city Hong Kong. First, the PM2.5 data on the typical walking paths were collected by the handheld mobile measuring instruments, to be selected as the dependent variables. Second, geographic prediction factors calculated by Google Street View, OSM data, Landsat images, and other multisource data were further selected as independent variables. Then, these dependent and independent variables were put into the LUR models to estimate the PM2.5 concentration on sidewalks, footbridges, and footpaths in the microscale walkable environment. The proposed models showed high performance relative to those in similar studies (adj R2, 0.593 to 0.615 [sidewalks]; 0.641 to 0.682 [footpaths]; 0.783 to 0.797 [footbridges]). This article is beneficial for mapping PM2.5 concentration in the microscale walking environment and the identification of hot spots of air pollution, thereby helping people avoid the PM2.5 hotspots and indicating a healthier walking path.

Air quality prediction is an important reference for meteorological forecast and air controlling, but over fitting often occurs in prediction algorithms based on a single model. Aiming at the complexity of air quality prediction, a prediction method based on integrated dual LSTM (Long Short-Term Memory) model was proposed in this paper. Firstly, the Seq2Seq (Sequence to Sequence) technology is used to establish a single-factor prediction model which can obtain the predicted value of each component in air quality data, independently. Each component of air quality is regarded as time series data in the forecasting process. Then, the LSTM model with attention mechanism is used as the multi-factor prediction model. The influencing factors of air quality, like the data of neighboring stations and weather data, are considered in the model. Finally, XGBoosting (eXtreme Gradient Boosting) tree is used to integrate two models. The final prediction results can be obtained by accumulating the predicted values of the optimal subtree nodes. Through evaluation and analysis using five evaluation methods, the proposed method has better performance in terms of error and model expression power. Compared with other various models, the precision of prediction data has been greatly improved in our model.

A special subclass of solutions of the three-dimensional system of ideal polytropic gas equations corresponding to an atmospheric model is considered. The properties of these solutions are completely characterized by a high-order nonlinear system of ordinary differential equations. Unlike the corresponding two-dimensional model, all singular points of this system have been found to be unstable. Some first integrals of this system have been found. In the case of axial symmetry, the system can be reduced to a single equation. If the adiabatic exponent is equal to 2, the system is integrable.

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.

Near-surface air temperature (NSAT) is a key parameter in climate changes, environmental ecosystem monitoring, and human settlement issues. As it is difficult for in situ observations to capture the spatial distribution characteristics of NSAT in great detail, various methods have been developed to use remotely sensed land surface temperature and other auxiliary variables to estimate the NSAT. Among them, machine learning turns out to be an exhilarating choice due to its superior performance. However, for long-term dataset estimation, when using machine learning methods with abundant related variables, computation loads and data storage cannot be ignored. Fortunately, Google Earth Engine (GEE) provides a wealth of geospatial datasets and powerful parallel computation capability. In this article, GEE has been utilized to demonstrate the feasibility of estimating long-term 1-km NSAT with two machine-learning models (random forest and deep neural network). After testing the effects of the input variables and the model portability, the 20-year monthly mean 1 km × 1 km NSAT over the Yellow River basin was generated and analyzed. Compared with in situ observations, the overall RMSE, MAE, R 2, and R of the NSAT product are 0.429, 0.302, 0.998, and 0.999, respectively. Specifically, for each observation station, R and R 2 are greater than 0.997, and RMSE and MAE are smaller than 0.5 and 0.4, respectively. The comparisons with three existing NSAT products show that our product outperforms the other three products. Once again, it is demonstrated in this study that GEE is a powerful platform to generate valuable products with ready-to-use datasets and computation resources.

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.