Human water regulation, groundwater lateral flow, and the movement of frost and thaw fronts (FTFs) affect soil water and thermal processes, as well as energy and water exchanges between the land surface and atmosphere. Reasonable representation of these processes in land surface models is very important to improving the understanding of land‐atmosphere interactions. In this study, mathematical descriptions of groundwater lateral flow, human water regulation, and FTFs were synchronously incorporated into a high‐resolution community land model, which is then named the Land Surface Model for Chinese Academy of Sciences (CAS‐LSM). With a series of atmospheric forcings and high‐resolution land surface data from the Heihe Watershed Allied Telemetry Experimental Research (HiWATER) program, numerical simulations of the period 1981–2013 using CAS‐LSM with 1‐km resolution were conducted for an endorheic basin, the Heihe River Basin in China. Compared with observations, CAS‐LSM reproduced the distributions of groundwater, evapotranspiration, and permafrost reasonably and well matched the temporal changes in ground temperature, heat fluxes, and FTFs. Results illuminate the temporal and spatial characteristics of frozen soil and the changes in the land‐atmosphere exchange of carbon, water, and energy. The permafrost and seasonally frozen soil were distinguished. In the seasonally frozen areas, the maximum soil frost depth increased by 0.65 mm/year within natural areas and decreased by 2.12 mm/year in human‐dominated areas. The active layer thickness increased 8.63 mm/year for permafrost. In the permafrost zone evapotranspiration and latent heat flux increased, and the sensible heat flux declined. In the human‐dominated areas water use raised the latent heat flux and reduced the sensible heat flux, net ecosystem exchange, and streamflow recharging to the eco‐fragile region in the lower reaches. Results suggested that the land surface model CAS‐LSM is a potential tool for studying land surface processes, especially in cold and arid regions experiencing human interventions.

The groundwater system is an essential part of Earth's systems. However, most current land surface models (LSMs) for climate modeling do not explicitly account for the lateral groundwater flow process. In this study, schemes describing LSM‐lateral groundwater flow module coupling, model resolution conversion, and parallel simulation were designed and implemented to incorporate a lateral groundwater flow module into the Community Land Model 4.5. The depth to less permeable bedrock also was included in the large‐scale groundwater flow modeling. Model validation was performed using multiple observations from the 20‐year continuous groundwater table depth measurements at 67 stations in 10 countries, 1.6 million worldwide time‐averaged groundwater table depth measurements, previous knowledge about the locations of major aquifer systems, and inversed terrestrial water storage anomalies derived from satellite data. The simulated results show that the groundwater table pattern is a combined reflection of climatic and topographic factors across a range of spatial scales. Lateral groundwater flow significantly modified the equilibrium water table patterns in North Africa, the Arabian Peninsula, central Asia, and southern Australia, deepening the water tables by more than 6 m. The trend of deepening groundwater tables observed between 1970 and 2010, which was found to be 0.025 to 0.125 m/decade, was exacerbated by the lateral flow; however, the seasonal variability of the groundwater table depth was reduced by the buffering effect of the lateral flow.

BackgroundInfluenza is an infectious respiratory disease that can cause serious public health hazard. Due to its huge threat to the society, precise real-time forecasting of influenza outbreaks is of great value to our public.ResultsIn this paper, we propose a new deep neural network structure that forecasts a real-time influenza-like illness rate (ILI%) in Guangzhou, China. Long short-term memory (LSTM) neural networks is applied to precisely forecast accurateness due to the long-term attribute and diversity of influenza epidemic data. We devise a multi-channel LSTM neural network that can draw multiple information from different types of inputs. We also add attention mechanism to improve forecasting accuracy. By using this structure, we are able to deal with relationships between multiple inputs more appropriately. Our model fully consider the information in the data set, targetedly solving practical problems of the Guangzhou influenza epidemic forecasting.ConclusionWe assess the performance of our model by comparing it with different neural network structures and other state-of-the-art methods. The experimental results indicate that our model has strong competitiveness and can provide effective real-time influenza epidemic forecasting.

No abstract

The brain activity pattern can be presented by Electroencephalogram (EEG), which is considered as an alternative to traditional biometrics. Researchers have done conducted studies on EEG-based identification, while few of them discussed the effect of time robustness which is very important for the identification system. In this study, we compared and analyzed the two runs EEG signals of restingstate of eye open/closed (REO/REC). The time intervals between two runs were at least two weeks. Here are 17 participants joined in this study. Each of them took two runs experiment. Each run contains four sessions, each session includes 150 seconds of REO/REC. Spectral and statistical analyses were used to extract feature. Three classifiers, Euclidean distance, SVM, and LDA, were used to get classification accuracies and to compare the performance between features of each run and two runs. The results of two runs PSD values of both REO and REC conditions show that there is a similarity within each subject and a difference between subjects. The classification accuracies of three methods of each run are almost 99%. The classification accuracies using two runs data as training set can also reach up to 97% while using each of two-run data as training set is nearly 80%. Thus, the features of most subjects have cross-time robustness and could be used as identification. This study will have an important role in EEG-based identification system.INDEX TERMS Electroencephalography(EEG), identification, resting-state, robustness.

No abstract

No abstract

No abstract