Abstract. Sea ice thickness evolution within the Canadian Arctic Archipelago (CAA) is of great interest to science, as well as local communities and their economy. In this study, based on the NEMO numerical framework including the LIM2 sea ice module, simulations at both 1/4 and 1/12 • horizontal resolution were conducted from 2002 to 2016. The model captures well the general spatial distribution of ice thickness in the CAA region, with very thick sea ice (∼ 4 m and thicker) in the northern CAA, thick sea ice (2.5 to 3 m) in the west-central Parry Channel and M'Clintock Channel, and thin (< 2 m) ice (in winter months) on the east side of CAA (e.g., eastern Parry Channel, Baffin Island coast) and in the channels in southern areas. Even though the configurations still have resolution limitations in resolving the exact observation sites, simulated ice thickness compares reasonably (seasonal cycle and amplitudes) with weekly Environment and Climate Change Canada (ECCC) New Ice Thickness Program data at first-year landfast ice sites except at the northern sites with high concentration of old ice. At 1/4 to 1/12 • scale, model resolution does not play a significant role in the sea ice simulation except to improve local dynamics because of better coastline representation. Sea ice growth is decomposed into thermodynamic and dynamic (including all non-thermodynamic processes in the model) contributions to study the ice thickness evolution. Relatively smaller thermodynamic contribution to ice growth between December and the following April is found in the thick and very thick ice regions, with larger contributions in the thin ice-covered region. No significant trend in winter maximum ice volume is found in the northern CAA and Baffin Bay while a decline (r 2 ≈ 0.6, p < 0.01) is simulated in Parry Channel region. The two main contributors (thermodynamic growth and lateral transport) have high interannual variabilities which largely balance each other, so that maximum ice volume can vary interannually by ±12 % in the northern CAA, ±15 % in Parry Channel, and ±9 % in Baffin Bay. Further quantitative evaluation is required.

Photosynthetic rates in different development stages were carefully investigated in 18 cultivars of winter wheat released in the period between 1945 and 1995 in the area of Beijing, China. During this period, the recorded grain yield has increased eightfold. However, when those cultivars were planted and managed in the same environment, the difference was reduced to only 36%, indicating that agronomic practices are the most important factors for grain yield. Agronomic features have changed greatly in the past 50 years, through increasing the harvest index (R2 = 0.89, P < 0.05), shortening plant height (R2 = 0.77, P < 0.05) and slightly increasing flag leaf areas (R2 = 0.45, P < 0.05), which is mostly in agreement with many other researchers. In contrast to many reports, however, this study found a genetic increase in the rate of photosynthesis per unit leaf area. From the mid-stem elongation to soft dough stages, the average photosynthetic rates at saturated photosynthetic photon flux density (P(sat)) increased by 44%. In the process, the stomatal conductance (g(s)) also increased by 122%. Grain yield was positively related to the mean values of P(sat) (R2 = 0.61, P < 0.01) and g(s) (R2 = 0.67, P < 0.01) in the six development stages. Our experiment may suggest that increase in grain yield was associated with the elevation of leaf photosynthetic rate and stomatal conductance over the past 50 years.

No abstract

No abstract

No abstract

Background: Parkinson's disease (PD) gradually degrades the functionality of the brain. Because of its relevance to the abnormality of the brain, electroencephalogram (EEG) signal is used for the early detection of this disease. This paper introduces a novel computer-aided diagnosis method to detect PD, which is an efficient deep learning method based on a pooling-based deep recurrent neural network (PDRNN). Therefore, the purpose of this study is to detect Parkinson's disease based on deep recurrent neural network of EEG signal Methods: The EEG signals of 20 patients with Parkinson's disease and 20 healthy people in Henan ProvincialPeople's Hospital (People's Hospital of Zhengzhou University) were examined, and a PDRNN learning method was applied on the dataset for managing the demand of the traditional feature presentation step. Results:The suggested DPRNN network gives the precision, sensitivity and specificity of 88.31%, 84.84% and 91.81%, respectively. Nevertheless, 11.28% of the healthy cases are wrongly categorized in Parkinson class. Also, 11.49% percent of Parkinson cases are classified wrongly in the healthy class. Conclusions:The experimental model has high efficiency and can be used as a reliable tool for clinical PD detection. In future research, more cases should be used to test and develop the proposed model.

No abstract

No abstract