The behaviors of thermospheric nitric oxide (NO) cooling during the 15 May 2005 intense geomagnetic storm are studied using measurements by the Sounding of the Atmosphere using Broadband Emission Radiometry instrument on board the Thermosphere‐Ionosphere‐Mesosphere Energetics and Dynamics satellite and simulations by the Thermosphere‐Ionosphere‐Electrodynamics General Circulation Model. The geomagnetic storm was the most intense (Dst = −247 nT) of 2005 with a short and rapid main phase and long‐lasting recovery (more than 3 days). NO cooling responded globally to the geomagnetic storm within 2 hr after the onset of storm main phase. The most significant NO cooling increases occurred at middle and low latitudes in the Northern Hemisphere and at middle latitudes in the Southern Hemisphere. The model outputs agree with observations in general but overestimate the NO cooling at high latitudes and underestimate the NO cooling elsewhere. Furthermore, observations show a significant upward shifting of the NO cooling peak altitude in the storm main phase and a significant downward shifting of the NO cooling peak altitude during the storm recovery phase at low latitudes. An unusual double‐peak structure in the NO cooling rate appeared during storm main phase and recovery phase. By investigating the NO cooling vertical profiles, we suggest that the horizontal equatorward transport plays an important role in inducing these significant variations of the NO cooling peak altitude.

Observations from lidars and satellites have shown that large neutral temperature increases and decreases occur in the middle and low latitudes of the mesosphere and lower thermosphere region during geomagnetic storms. Here we undertake first‐principles simulations of mesosphere and lower thermosphere temperature responses to storms using the Thermosphere Ionosphere Mesosphere Electrodynamics General Circulation Model to elucidate the nature and causes of these changes. Temperature variations were not uniform; instead, nighttime temperatures changed earlier than daytime temperatures, and temperatures changed earlier at high latitudes than at low ones. Furthermore, temperatures increased in some places/times and decreased in others. As the simulation behaves similar to observations, it provides an opportunity to understand physical processes that drive the observed changes. Our analysis has shown that they were produced mainly by adiabatic heating/cooling that was associated with vertical winds resulting from general circulation changes, with additional contributions from vertical heat advection.

We investigate the latitudinal and longitudinal variations of thermospheric nitric oxide (NO) cooling rate from 2005 to 2016. We compare the maximum of the NO cooling rate and its altitude from measurements by the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument on National Aeronautics and Space Administration (NASA)'s TIMED (Thermosphere‐Ionosphere‐Mesosphere Energetics and Dynamics) satellite and from simulations by the Thermosphere‐Ionosphere‐Electrodynamics General Circulation Model (TIEGCM). In both data and model, the maximum NO cooling rate varies with ~11‐year solar cycle, while the altitude of the peak value does not. At local noon, the maximum NO cooling rate from TIEGCM is significantly larger than that derived from SABER, especially in solar minimum. There are differences in the altitudes of the maximum NO cooling rates between the observations and simulations. The largest difference occurs at noon at low latitudes, with SABER values exceeding TIEGCM values by more than 10 km with a maximum difference of 16 km near solar maximum. Model sensitivity tests suggest that the biases are caused by a model temperature overestimation in the low thermosphere, and tides in the TIEGCM lower boundary play an important role in inducing the low‐latitude temperature overestimation by the model. In addition, SABER's height‐integrated cooling rate (NO flux) exceeds that from TIEGCM, primarily due to the lower cooling rates at 140–200 km in the model.

As an important index of solar activity, the 10.7-cm solar radio flux (F_10.7) can indicate changes in the solar EUV radiation, which plays an important role in the relationship between the Sun and the Earth. Therefore, it is valuable to study and forecast F_10.7. In this study, the long short-term memory (LSTM) method in machine learning is used to predict the daily value of F_10.7. The F_10.7 series from 1947 to 2019 are used. Among them, the data during 1947–1995 are adopted as the training dataset, and the data during 1996–2019 (solar cycles 23 and 24) are adopted as the test dataset. The fourfold cross validation method is used to group the training set for multiple validations. We find that the root mean square error (RMSE) of the prediction results is only 6.20~6.35 sfu, and the correlation coefficient (R) is as high as 0.9883~0.9889. The overall prediction accuracy of the LSTM method is equivalent to those of the widely used autoregressive (AR) and backpropagation neural network (BP) models. Especially for 2-day and 3-day forecasts, the LSTM model is slightly better. All this demonstrates the potentiality of the LSTM method in the real-time forecasting of F_10.7 in future.

In this study, the Empirical Mode Decomposition algorithm (EMD) and the Long Short Term Memory neural network (LSTM) are combined into an EMD‐LSTM model, to predict the variation of the >2 MeV electron fluxes 1 day ahead. Input parameters include the Pc5 power, AP, AE, Kp, >0.6 MeV, and historical electron flux values, are used for predictions. All the time resolution of parameters are daily integral values. As compared the prediction results of the EMD‐LSTM model with other classical prediction models, the results show that the 1 day ahead prediction efficiency of the >2 MeV electron fluxes possesses a prediction efficiency of 0.80, and the highest prediction efficiency can reach 0.93. These results are superior to the prediction accuracy of more previous models. Using two high‐energy electron flux storm events for validation, the results indicate that the performance of the EMD‐LSTM model in the period of the high‐energy electron flux storm is also relatively good, especially for the prediction of high‐energy electron fluxes at extreme points, and the predictions are closer to actual observations.