Based on a set of deep learning and mode decomposition methods, a short-term prediction model for PM_2.5 concentration for Beijing city is established in this paper. An ensemble empirical mode decomposition (EEMD) algorithm is first used to decompose the original PM_2.5 timeseries to several high- to low-frequency intrinsic mode functions (IMFs). Each IMF component is then trained and predicted by a combination of three neural networks: back propagation network (BP), long short-term memory network (LSTM), and a hybrid network of a convolutional neural network (CNN) + LSTM. The results showed that both BP and LSTM are able to fit the low-frequency IMFs very well, and the total prediction errors of the summation of all IMFs are remarkably reduced from 21 g/m³ in the single BP model to 4.8 g/m³ in the EEMD + BP model. Spatial information from 143 stations surrounding Beijing city is extracted by CNN, which is then used to train the CNN+LSTM. It is found that, under extreme weather conditions of PM_2.5 < 35 g/m³ and PM_2.5 > 150 g/m³, the prediction errors of the CNN + LSTM model are improved by ~30% compared to the single LSTM model. However, the prediction of the very high-frequency IMF mode (IMF-1) remains a challenge for all neural networks, which might be due to microphysical turbulences and chaotic processes that cannot be resolved by the above-mentioned neural networks based on variable–variable relationship.

High-frequency (HF) radar data, derived from a pair of newly developed radar stations in the Pearl River Estuary (PRE) of China, were validated through comparison with in situ surface buoys, ADCP measurements, and model simulations in this study. Since no in situ observations are available in the radar observing domain, a regional high-resolution ocean model covering the entire PRE and its adjacent seas was first established and validated with in situ measurements, and then the HF radar data quality was examined against the model simulations. The results show that mean flows and tidal ellipses derived from the in situ buoys and ADCP were in very good agreement with the model. The model–radar data comparison indicated that the radar obtained the best data quality within the central overlapping area between the two radar stations, with the errors increasing toward the coast and the open ocean. Near the coast, the radar data quality was affected by coastlines and islands that prevent HF radar from delivering high-quality information for determining surface currents. This is one of the major drawbacks of the HF radar technique. Toward the open ocean, where the wind is the only dominant forcing on the tidal currents, we found that the poor data quality was most likely contaminated by data inversion algorithms from the Shangchuan radar station. A hybrid machine-learning-based inversion algorithm including traditional electromagnetic analysis and physical oceanography factors is needed to develop and improve radar data quality. A new radar observing network with about 10 radar stations is developing in the PRE and its adjacent shelf, this work assesses the data quality of the existing radars and identifies the error sources, serving as the first step toward the full deployment of the entire radar network.

A machine learning neural network-based design for shipborne ADCP navigation is proposed to improve the quality of high-frequency radar measurements. In traditional inversion algorithms for HF radars, sea surface velocity is directly extracted from electromagnetic echoes without constraints from oceanographic processes. Hence, we incorporated oceanographic information from observational data into seabed radar inversion results via an LSTM neural network model to enhance data accuracy. Through a series of numerical simulation experiments, we showed improved data accuracy and feasibility by incorporating both fixed-point and navigation observational data. The results indicate a significant reduction in (related) errors. This study has implications for guiding future navigation observations.

We propose a quality control method based on machine learning neural networks to enhance the quality of high-frequency (HF) radar data. Unlike traditional quality control methods that rely on radar signals as indicators and involve extensive data manipulation in specialized software, our approach employs a Bi-LSTM neural network model. This method aims to improve data quality and streamline the quality control process. Through a series of analyses, we demonstrate the feasibility of using machine learning techniques to enhance radar data quality.

An algorithm based on a long short-term memory (LSTM) network is proposed to reduce errors from high-frequency surface wave radar current measurements. In traditional inversion algorithms, the radar velocities are derived from electromagnetic echo signals, with no constraints imposed by physical oceanographic processes. In this study, sea surface winds and tides are included in the LSTM algorithm to improve radar data. These physical factors provide the LSTM network with more oceanic information by which to constrain and improve its training efficiency. The results show that the domain-averaged root-mean-square errors of the radar-derived velocities are reduced from 0.22 to 0.09 m/s for the whole radar observation area. The overall correlation coefficient increases from 0.37 to 0.88. To provide a practical strategy for future field work, we conduct a set of sensitivity experiments, showing that the LSTM network based on one single point can be applied to other data points within a sub-domain.

This paper proposes an algorithm based on the long short-term memory (LSTM) network to improve the quality of high-frequency surface wave radar current measurements. In order to address the limitations of traditional high-frequency radar inversion algorithms, which solely rely on electromagnetic inversion and disregard physical oceanography, this study incorporates a bottom-mounted acoustic Doppler current profiler (ADCP) and towed ADCP into LSTM training. Additionally, wind and tidal oceanography data were included as inputs. This study compared high-frequency radar current data before and after calibration. The results indicated that both towed and bottom-mounted ADCP enhanced the quality of HF radar monitoring data. By comparing two methods of calibrating radar, we found that less towed ADCP data input is required for the same high-frequency radar data calibration effect. Furthermore, towed ADCP has a significant advantage in calibrating high-frequency radar data due to its low cost and wide calibration range. However, as the location of the calibrated high-frequency radar data moves further away from the towing position, the calibration error also increases. This article conducted sensitivity studies on the times and different positions of using towed ADCP to calibrate high-frequency radar data, providing reference for the optimal towing path and towing time for future corrections of high-frequency radar data.