Complex Meteorological Conditions Research Articles

Seasonal migration pattern of an important rice pest, Nilaparvata lugens (Hemiptera: Delphacidae), on Hainan Island, China.

The brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae) (BPH), is a major, regionally migratory pest of rice in Asia. Despite intensive migratory studies, the seasonal migratory pattern of this species in the year-round breeding region (i.e., Hainan Island) remains largely unknown. To understand the migration dynamics, we conducted relevant research based on BPH light trap catches on Hainan in 2017-2022. Results showed that the occurrence dynamics of BPH in Hainan oscillated in different years and seasons. Overall, there are 4 migration peak periods on Hainan, with outbound peak periods in April-May and August-September and inbound peak periods in June and October. Trajectory and wind fields showed that in August-September, Hainan had complex meteorological conditions, and the outbound paths of BPH were variable, mainly landing in southeastern Guangxi and southwestern Guangdong. In contrast, April-May was influenced by southeasterly winds, with a wider outbound range and a further distance, landing along northeastern Vietnam from west to east toward Guangxi and Guangdong areas with time. Wind-rain fields, trajectory, and hydrogen isotope showed that in June, southwesterly winds transported BPH northward from the Indochina peninsula, where it was hindered by rainfall in the Leizhou Peninsula area and the migrants will be "rained out" into Hainan. Identically, in October, the interactions between northeasterly winds and rainfall belts on the southwest shore of Hainan increased the opportunities for BPH to migrate southward to reach Hainan. Our results suggest that Hainan is not only an important source of BPH migrating from East Asia but also plays the role of a bridge.

Mapping and Pre- and Post-Failure Analyses of the April 2019 Kantutani Landslide in La Paz, Bolivia, Using Synthetic Aperture Radar Data

Urban landslides have brought challenges to developing countries undergoing urbanization. Rapid approaches to assess ground deformation are required when facing the challenge of insufficient geological survey methods. Additionally, it is indeed a challenge to map landslide-affected areas, especially precipitation-induced landslides, through optical remote sensing methods. This study applied SAR change detection methods to map the slope failure event of the San Jorge Kantutani landfill site in La Paz, Bolivia, which occurred in April 2019, and Multi-Temporal Synthetic Aperture Radar Interferometry (MTInSAR) methods to assess pre- and post-failure ground stability related to this event. We found that the amplitude information of high-resolution COSMO-SkyMed SAR imagery and its texture information can be very useful in landslide mapping, especially in situations in which optical images are not available because of complex meteorological conditions and the similar spectral characteristics between the original land cover and landslide deposits. The MTInSAR analyses found that there was already significant deformation of more than 50 mm/year along the slope direction over this site before the landslide, and such deformation could be clearly discriminated from the surrounding environment. After the landslide event and the remobilization of the landslide deposit, the slope still shows a deformation velocity of more than 30 mm/year. The SAR amplitude change detection and MTInSAR fully exploited the SAR data in landslide studies and were useful in back analyzing the occurred landslides; this could be a good method for monitoring the ground stability of La Paz or even on a national scale over the long term for reducing the catastrophic effects of geological hazards in this landslide-prone city.

Terahertz Transmission Characteristics for UAV Swarm on Atmosphere-Limited Line-of-sight Links Under Complex Meteorological Conditions

Terahertz Transmission Characteristics for UAV Swarm on Atmosphere-Limited Line-of-sight Links Under Complex Meteorological Conditions

Aircraft Target Detection from Remote Sensing Images under Complex Meteorological Conditions

Taking all-day, all-weather airport security protection as the application demand, and aiming at the lack of complex meteorological conditions processing capability of current remote sensing image aircraft target detection algorithms, this paper takes the YOLOX algorithm as the basis, reduces model parameters by using depth separable convolution, improves feature extraction speed and detection efficiency, and at the same time, introduces different cavity convolution in its backbone network to increase the perceptual field and improve the model’s detection accuracy. Compared with the mainstream target detection algorithms, the proposed YOLOX-DD algorithm has the highest detection accuracy under complex meteorological conditions such as nighttime and dust, and can efficiently and reliably detect the aircraft in other complex meteorological conditions including fog, rain, and snow, with good anti-interference performance.

Improved reinforcement learning for collision-free local path planning of dynamic obstacle

The rapid development of artificial intelligence has driven the transformation of traditional industry. The proposal for an intelligent ship reflects the development of the shipping industry in the direction of intelligence. Safe maritime transportation is an essential branch of the intelligent ship industry. In this context, applying various intelligent algorithms in ship dynamic obstacle avoidance has attracted the attention and discussion of scholars. Due to the complexity and diversity of ship navigation, environments cannot be described by a definite mathematical model. Reinforcement learning has certain advantages in solving path planning in a complex environment. Traditional mathematical algorithms and swarm intelligence algorithms need mathematical models to constrain boundary conditions when solving complex problems. However, some complex dynamic environments in the real world cannot define mathematical models. This paper optimizes the parameter update and exploration strategies of the actor–critic algorithm in reinforcement learning to improve the algorithm’s performance. Under complex meteorological conditions, improved reinforcement learning is applied to the dynamic ship offshore channel simulation environment to verify the algorithm’s performance.

Artificial Intelligent Power Forecasting for Wind Farm Based on Multi-Source Data Fusion

Wind power forecasting is a typical high-dimensional and multi-step time series prediction problem. Data-driven prediction methods using machine learning show advantages over traditional physical or statistical methods, especially for wind farms with complex meteorological conditions. Thus, effective use of different data sources and data types will help improve power forecasting accuracy. In this paper, a multi-source data fusion method is proposed, which integrates the static information of the wind turbine with observational and forecasting meteorological information together to further improve the power forecasting accuracy. Firstly, the characteristics of each time step are re-characterized by using the self-attention mechanism to integrate the global information of multi-source data, and the Res-CNN network is used to fuse multi-source data to improve the prediction ability of input variables. Secondly, static variable encoding and feature selection are carried out, and the time-varying variables are combined with static variables for collaborative feature selection, so as to effectively eliminate redundant information. A forecasting model based on the Encoder–Decoder framework is constructed with LSTM as the basic unit, and the Add&Norm mechanism is introduced to further enhance the input variable information. In addition, the self-attention mechanism is used to integrate the global time information of the decoded results, and the Time Distributed mechanism is used to carry out multi-step prediction. Our training and testing data are obtained from an operating wind farm in northwestern China. Results show that the proposed method outperforms a classic AI forecasting method such as that using the Seq2Seq+attention model in terms of prediction accuracy, thus providing an effective solution for multi-step forecasting of wind power in wind farms.

The mixed layer modified radionuclide atmospheric diffusion based on Gaussian model.

Atmospheric diffusion is often accompanied by complex meteorological conditions of inversion temperature. In response to the emergency needs for rapid consequence assessment of nuclear accidents under these complex meteorological conditions, a Gaussian diffusion-based model of radionuclide is developed with mixed layer modification. The inhibition effect of the inversion temperature on the diffusion of radionuclides is modified in the vertical direction. The intensity of the radionuclide source is modified by the decay constant. The results indicate that the enhancement effect of the mixed layer on the concentration of radionuclides is reflected. The shorter the half-life of the radionuclide, the greater the effect of reducing the diffusion concentration. The Kincaid dataset validation in the Model Validation Kit (MVK) shows that, compared to the non-modified model, predictions of the modified model have an enhancement effect beyond 5 km, modulating the prediction values to be closer to the observation values. This development is consistent with the modification effects of the mixed layer. The statistical indicators show that the criteria of the modified model meet the criteria of the recommended model.

A pixel-wise calculation of soil evaporative efficiency with thermal/optical remote sensing and meteorological reanalysis data for downscaling microwave soil moisture

Soil evaporative efficiency (SEE) was estimated as the relative difference of soil temperature (Ts) to its maximum (Tsmax) and minimum (Tsmin) values at ‘minimum and maximum’ soil moisture (SM), i.e., SEE= (Tsmax−Ts)/(Tsmax−Tsmin). This thermal indicator of SM (SMI) has been proven effective in downscaling satellite microwave SM in some local areas. However, the Tsmax and Tsmin were usually determined in an empirical way which is not conducive to a wider range of applications or global downscaling. In addition to the Ts-based SMI (SEE), solo-LST (Land Surface Temperature)-based SMI and composite index of LST and optical SMIs have also been widely used in downscaling microwave SM. However, they have rarely been evaluated with spatially distributed SM at a similar spatial resolution, leading to a vague understanding of their pros and cons in the downscaling. In this study, we first implemented a pixel-wise and theoretically calculating of SEE with LST/FVC space and meteorological reanalysis data. Then, a critical evaluation with satellite, aircraft, and in-situ SM observations was made via the SMPAVEX15 and SMAPVEX16 aircraft-based field experiments. Results indicated that the thermal SMIs presented a better performance than the optical SMIs for depicting the variations of SM. Among the thermal SMIs, Ts-based SEE showed better performance than the solo-LST-based index and the composite index temperature-vegetation-soil moisture dryness index (TVMDI). The results demonstrate the validity and necessity of obtaining the Ts-based SEE for downscaling satellite microwave SM with thermal/optical data. The provided pixel-wise method was validated in this study for calculating the Ts-based SEE, which is suggested for future downscaling of satellite microwave SM with microwave and thermal/optical fusion. However, the disadvantages of cloud contamination, the decoupling effect, and uncertainties brought by complex meteorological conditions and reanalysis data urgently need more research efforts.

Prediction of apple first flowering date using daily land surface temperature spatio-temporal reconstruction and machine learning

The first flowering date (FFD) is a critical phenological parameter closely related to the apple yield, so the accurate prediction of the FFD is important for precise orchard production management. Existing methods to predict the FFD are mostly based on air temperature (Ta) measured at meteorological stations, but to great differences in meteorological variations and the ecological conditions, these methods cannot present the differences of FFD under complex meteorological conditions and provide spatially continuous FFD information at the level of a region. Therefore, we propose a method to predict spatially continuous apple FFD from remote sensing land surface temperature (LST) based on flowering prediction model. Firstly, the missing LST data were reconstructed by using spatio-temporal reconstruction (STR) approach developed. Next, new air temperature (NAT) data were generated by using the daily Ta estimation (DTE) model and the reconstructed LST. Finally, apple FFD was predicted by the NAT data and the apple flowering prediction model established based on random forest (RF) algorithm and the phenology sequential model, and the prediction accuracy was verified by comparison with the independently measured apple FFD. The LST reconstructed by using the STR approach has mean absolute error (MAE) ranging from 0.51 to 0.68 °C, and root mean square error (RMSE) ranging from 1.07 to 1.21 °C. The MAE between the NAT data and the High-Resolution Land Surface Data Assimilation System (HR-CLDAS) meteorological data ranges from 2.15 to 3.23 °C, and the RMSE ranges from 2.81 to 4.27 °C. In addition, the determination coefficient (R2) and RMSE between the predicted and measured FFD is 0.72 and 2.96 days, respectively. These results demonstrate that the developed method maximizes the potential of MODIS LST in predicting spatially continuous apple FFD, which is valuable for flower and fruit thinning, to defend against frost disasters, and in general for refined orchard production management.

PPNet: A more effective method of precipitation prediction

AbstractPrecipitation nowcasting plays an important role in the early warning of disasters and many other aspects of people's lives. In this study, we address the problem of radar reflectivity image extrapolation, which has great significance for precipitation near‐range forecasting. In recent years, the related achievements of nowcasting indicate that deep learning‐based methods have been far ahead of traditional ones. However, most deep learning methods focus on spatial appearance but cannot characterize the motion information well. To solve this complex problem, in addition to completing pixel‐level predictions by McNet, we introduce FlowNet and optical flow loss into generative adversarial networks (GAN) to express motion information more effectively. The reflectance image changes significantly due to complex meteorological conditions. Only by adding motion features can the law of change be determined. Thus, we can retain more details in the image. To the best of our knowledge, this is the first time FlowNet has been combined with GAN to fulfil the task of precipitation prediction. Extensive experiments on the dataset provided by the Shenzhen Meteorological Bureau demonstrate that our network performs favourably against other state‐of‐the‐art methods, which presents great guiding significance for precipitation nowcasting and possesses broad application prospects.

Weather Route Optimization Method of Unmanned Ship Based on Continuous Dynamic Optimal Control

Intelligent weather route optimization method was an essential guarantee for safe and efficient navigation of ships. In traditional methods, the multi-stage decision-making method was generally used to solve route dynamic optimization problems, in which it was impossible to achieve dynamic optimal routes, due to the temporal and spatial complexity of route design. In this paper, an unmanned ship weather route optimization model was established based on a method of continuous dynamic optimal control. Marine meteorological information was analyzed. Navigation safety, energy consumption, and sailing time were integrated considered. Dual-target route evaluation function of energy consumption and sailing time was established. The problem about the multi-stage decision was transformed into that of one-step optimal control, and an improved ant colony algorithm was adopted. Simulation results showed that compared with some traditional methods, the proposed method was better performed, which can be applied to dynamic route optimization of the unmanned ship in a large marine area under complex meteorological conditions.

Study on the BL Early Warning Model of Transmission Line Tower Based on KNN-SVM Algorithm

With the expansion of China’s power grid construction scale, the transmission line span are gradually improved, which also increases the risk of BL stroke on the transmission line. However, the traditional passive BL protection has many problems, such as weak pertinence and high investment cost, which can not meet the needs of social development. KNN can well describe the similarity measure between the two, which can effectively reduce the training samples. SVM can find the best compromise between model complexity and learning ability in small samples, which is a good sample training method. Through KNN - in-depth learning of the historical data of BL activities accumulated in the power grid, a supervised BL early warning model (hereinafter referred to as EWM) of transmission line can be trained. At the same time, the BL strike of transmission line tower (hereinafter referred to as TLT) has complex meteorological conditions, which requires comprehensive confirmation of various monitoring point parameters. Therefore, it is of great significance to study the BL EWM of TLT based on KNN-SVM algorithm. Firstly, this paper analyzes the KNN-SVM algorithm. Then, this paper establishes an EWM. Finally, this paper is verified.

A global observational analysis to understand changes in air quality during exceptionally low anthropogenic emission conditions.

This global study, which has been coordinated by the World Meteorological Organization Global Atmospheric Watch (WMO/GAW) programme, aims to understand the behaviour of key air pollutant species during the COVID-19 pandemic period of exceptionally low emissions across the globe. We investigated the effects of the differences in both emissions and regional and local meteorology in 2020 compared with the period 2015-2019. By adopting a globally consistent approach, this comprehensive observational analysis focuses on changes in air quality in and around cities across the globe for the following air pollutants PM2.5, PM10, PMC (coarse fraction of PM), NO2, SO2, NOx, CO, O3 and the total gaseous oxidant (OX=NO2+O3) during the pre-lockdown, partial lockdown, full lockdown and two relaxation periods spanning from January to September 2020. The analysis is based on in situ ground-based air quality observations at over 540 traffic, background and rural stations, from 63 cities and covering 25 countries over seven geographical regions of the world. Anomalies in the air pollutant concentrations (increases or decreases during 2020 periods compared to equivalent 2015-2019 periods) were calculated and the possible effects of meteorological conditions were analysed by computing anomalies from ERA5 reanalyses and local observations for these periods. We observed a positive correlation between the reductions in NO2 and NOx concentrations and peoples' mobility for most cities. A correlation between PMC and mobility changes was also seen for some Asian and South American cities. A clear signal was not observed for other pollutants, suggesting that sources besides vehicular emissions also substantially contributed to the change in air quality. As a global and regional overview of the changes in ambient concentrations of key air quality species, we observed decreases of up to about 70% in mean NO2 and between 30% and 40% in mean PM2.5 concentrations over 2020 full lockdown compared to the same period in 2015-2019. However, PM2.5 exhibited complex signals, even within the same region, with increases in some Spanish cities, attributed mainly to the long-range transport of African dust and/or biomass burning (corroborated with the analysis of NO2/CO ratio). Some Chinese cities showed similar increases in PM2.5 during the lockdown periods, but in this case, it was likely due to secondary PM formation. Changes in O3 concentrations were highly heterogeneous, with no overall change or small increases (as in the case of Europe), and positive anomalies of 25% and 30% in East Asia and South America, respectively, with Colombia showing the largest positive anomaly of ~70%. The SO2 anomalies were negative for 2020 compared to 2015-2019 (between ~25 to 60%) for all regions. For CO, negative anomalies were observed for all regions with the largest decrease for South America of up to ~40%. The NO2/CO ratio indicated that specific sites (such as those in Spanish cities) were affected by biomass burning plumes, which outweighed the NO2 decrease due to the general reduction in mobility (ratio of ~60%). Analysis of the total oxidant (OX=NO2+O3) showed that primary NO2 emissions at urban locations were greater than the O3 production, whereas at background sites, OX was mostly driven by the regional contributions rather than local NO2 and O3 concentrations. The present study clearly highlights the importance of meteorology and episodic contributions (e.g., from dust, domestic, agricultural biomass burning and crop fertilizing) when analysing air quality in and around cities even during large emissions reductions. There is still the need to better understand how the chemical responses of secondary pollutants to emission change under complex meteorological conditions, along with climate change and socio-economic drivers may affect future air quality. The implications for regional and global policies are also significant, as our study clearly indicates that PM2.5 concentrations would not likely meet the World Health Organization guidelines in many parts of the world, despite the drastic reductions in mobility. Consequently, revisions of air quality regulation (e.g., the Gothenburg Protocol) with more ambitious targets that are specific to the different regions of the world may well be required.

A Multi‐Dimensional Decomposition Method of the Meteorology‐Driven and Emission‐Driven Effects on Year‐to‐Year Air Quality Variations

AbstractThe annual assessment of emission control effects on air quality is essential for policy adjustments. However, this assessment is difficult as the inter‐annual changes in pollution are impacted by complex meteorological conditions. In this study, based on our wind‐pollution decomposition (WPD) method, which decomposes wind effects (wind‐driven) and nonwind effects, we established the meteorology‐pollution decomposition (MPD) method, which separates meteorological effects (met‐driven) and nonmeteorological effects by importing other meteorological parameters, to approximate the emission change effects (emission‐driven) by the nonmeteorological effects. The performance of the MPD method was assessed by comparing the results of the WPD and MPD methods at multiple representative stations in Hong Kong with the longest continuous data available in China from 2000 to 2018. The decomposed emission‐driven impact is also validated by using the gross domestic product growth rates data. The emission‐driven contributions determined by the WPD and MPD methods generally show agreement in trend, while the met‐driven effects are generally larger than the wind‐driven effects with the same or opposite trend. The results showed that the met‐driven effect from the MPD method is more reasonable than the wind‐driven effect from the WPD method in representing complex meteorological influences, which indicates a better representation of emission‐driven impacts by the nonmeteorological effects during the unusual meteorological years, such as El Niño‐Southern Oscillation events. The MPD method represents an independent approach, which is based only on regular observations, to quantify the long‐term trend in emission‐driven impacts as well as the integrated meteorological impacts on the atmospheric environment under the background of climate change.

Analysis of Oil and Gas Ethane and Methane Emissions in the Southcentral and Eastern United States Using Four Seasons of Continuous Aircraft Ethane Measurements

AbstractIn the last decade, much work has been done to better understand methane (CH4) emissions from the oil and gas (O&G) industry in the United States. Ethane (C2H6), a gas that is co‐emitted with thermogenic sources of CH4, is emitted in the US predominantly by the O&G sector. In this study, we perform an inverse analysis on 200 h of atmospheric boundary layer C2H6 measurements to estimate C2H6 emissions from the US O&G sector. Measurements were collected from 2017 to 2019 as part of the Atmospheric Carbon and Transport (ACT) America aircraft campaign and encompass much of the central and eastern United States. We find that for the fall, winter, and spring campaigns, C2H6 data consistently exceeds values that would be expected based on EPA O&G leak rate estimates by more than 50%. C2H6 observations from the summer 2019 data set show significantly lower C2H6 enhancements in the southcentral region that cannot be reconciled with data from the other three seasons, either due to complex meteorological conditions or a temporal shift in the emissions. Combining the fall, winter, and spring C2H6 posterior emissions estimate to an inventory of O&G CH4 emissions, we estimate that O&G CH4 emissions are larger than EPA inventory values by 48%–76%. Uncertainties in the gas composition data limit the accuracy of using C2H6 as a proxy for O&G CH4 emissions. These limits could be resolved retroactively by increasing the availability of industry‐collected gas composition data.

Dynamic Optimization Method for Unmanned Ship Weather Route Based on Multi-stage Inverse Reasoning

Route is an important factor for safe and efficient navigation of ships. In this paper, unmanned ship navigation environment under complex marine meteorological conditions was analyzed. Wind, wave conditions were integrated considered. Dynamic optimization model of unmanned ship weather route was established, in which meteorological and hydrological information was considered, unfavorable influences of which on ship navigation were reduced, energy and sailing time were saved, and route was found. Simulation results showed that initial fixed weather route could be optimized, and the model could be applied to dynamic optimization of weather routes of unmanned ships in large areas under complex meteorological conditions.

Analysis of Accidents Resulting from the Interaction of Air and Ground Vehicles at Airports

The analysis of the causes and consequences of accidents resulting from the interaction of air and ground vehicles at airports is presented in this paper. The data on accidents involving the following aircrafts were taken into account: airplanes, helicopters and unmanned aircrafts; the ground vehicles included: fuel tankers, cars, snow plows, pushback tractors and other vehicles operating at the airport. As a basis for the analysis, data on eleven such accidents that occurred at the airports of a number of countries in the period from 2000 to 2021 were used. The analysis of what types of ground and aircrafts were involved in the accidents is made. When studying the main and associated causes of accidents, the actions of all participants in traffic control at the airport (operators of ground vehicles, aircraft pilots and air traffic control officers) were analyzed. In addition, the impact of complex meteorological conditions (snow, fog, heavy rain, etc.) limiting visibility was analyzed. The classification of accidents is carried out in accordance with the level of danger of their consequences. At the final stage, possible ways to reduce the risk of such events in the future are discussed.

Radiological consequence assessments using time-varying source terms in ONERS- decision support system for nuclear emergency response

Online Nuclear Emergency Response System (ONERS) is a Decision Support System (DSS) developed for the management of Off-Site Nuclear Emergencies for Nuclear Power Plant sites of India. In this work, ONERS is incorporated with a module ‘Time-Dependent Source term (TDS)' for real-time online plume dispersion and radiological consequence assessments using time-varying source terms which is likely during an accidental release. TDS facilitates online dispersion and dose calculations by initiating the FLEXPART particle dispersion model with WRF predicted real-time meteorological forecasts and inputs on hourly release rates of 25 significant radionuclides of both ground level and elevated releases for any reactor accident over a 72-h meteorological forecast domain. The TDS is tested against the existing uniform source term approach for a hypothetical accident for two types of frequent weather conditions over Kalpakkam nuclear site in southern India. Simulations for a complex weather condition (monsoon & local sea breeze) during 26–29 September 2019 indicate that though the simulated plume area is similar in both the cases, the area affected is different as governed by the meteorological condition during the period of releases. It is found that the area and direction of potentially affected zone is determined by most frequent wind flow in the case of uniform source term and the flow direction during the period of maximum release in the case of time-dependent source term. Simulations for winter weather condition during 10–13 January 2020 with nearly uniform flow indicate similar results in both uniform and time-varying release cases. Results indicate that while the time-dependent source terms provide more realistic projection of radiological impact under complex meteorological conditions over uniform source term, both approaches would nearly converge when the maximum release occurs under the most frequent windflow condition. The two approaches are complementary to each other for response actions given the requirements of rapid assessments with gross source terms in the initial phases and more refined assessments with detailed source terms in the later phases of emergencies.

Daily reference evapotranspiration prediction of Tieguanyin tea plants based on mathematical morphology clustering and improved generalized regression neural network

Tieguanyin tea plant is the most important tea cultivar in Fujian Province, China. It has suffered great economic losses due to high temperature and dry weather in recently years. This study proposed a prediction model for daily reference evapotranspiration (ET0) of Tieguanyin based on the combination of mathematical morphology clustering (MMC) and generalized regression neural networks (GRNN). Average air temperature, sunshine hours and relative humidity were chosen as the input factors of GRNN after a correlation analysis of the microclimate factors of the tea garden. The MMC was adopted to cluster the historical meteorological data to find the similar class as the training dataset of GRNN. The fruit fly optimization algorithm (FOA) was used to optimize the smoothing factor of GRNN. The meteorological data from JAN. 2018 to OCT. 2019 measured in Dabaofeng tea garden, Anxi County, Fujian Province, China were used to train and test the proposed model, and the performances was assessed with RMSE, MAE and model validity coefficient. The prediction results of different seasons show that the proposed model is efficient with high accuracy and has good adaptability under complex meteorological conditions.

Retrieving Volcanic Ash Top Height through Combined Polar Orbit Active and Geostationary Passive Remote Sensing Data

Taking advantage of both the polar orbit active remote sensing data (from the Cloud-Aerosol Lidar with Orthogonal Polarization—CALIOP) and vertical information and the geostationary passive remote sensing measurements (from the Spinning Enhanced Visible and Infrared Imager) with large coverage, a methodology is developed for retrieving the volcanic ash cloud top height (VTH) from combined CALIOP and Spinning Enhanced Visible and Infrared Imager (SEVIRI) data. This methodology is a deep-learning-based algorithm through hybrid use of Stacked Denoising AutoEncoder (SDA), the Genetic Algorithm (GA), and the Least Squares Support Vector Regression (LSSVR). A series of eruptions over Iceland’s Eyjafjallajökull volcano from April to May 2010 and the Puyehue-Cordón Caulle volcanic complex eruptions in Chilean Andes in June 2011 were selected as typical cases for independent validation of the VTH retrievals under various meteorological backgrounds. It is demonstrated that using the hybrid deep learning algorithm, the nonlinear relationship between satellite-based infrared (IR) radiance measurements and the VTH can be well established. The hybrid deep learning algorithm not only performs well under a relatively simple meteorological background but also is robust under more complex meteorological conditions. Adding atmospheric temperature vertical profile as additional information further improves the accuracy of VTH retrievals. The methodology and approaches can be applied to the measurements from the advanced imagers onboard the new generation of international geostationary (GEO) weather satellites for retrieving the VTH science product.