The Northern Hemisphere experienced dramatic changes during the last glacial, featuring vast ice sheets and abrupt climate events, while high northern latitudes during the last interglacial (Eemian) were warmer than today. Here we use high-resolution aerosol records from the Greenland NEEM ice core to reconstruct the environmental alterations in aerosol source regions accompanying these changes. Separating source and transport effects, we find strongly reduced terrestrial biogenic emissions during glacial times reflecting net loss of vegetated area in North America. Rapid climate changes during the glacial have little effect on terrestrial biogenic aerosol emissions. A strong increase in terrestrial dust emissions during the coldest intervals indicates higher aridity and dust storm activity in East Asian deserts. Glacial sea salt aerosol emissions in the North Atlantic region increase only moderately (50%), likely due to sea ice expansion. Lower aerosol concentrations in Eemian ice compared to the Holocene are mainly due to shortened atmospheric residence time, while emissions changed little.

This letter studies deep learning (DL) approaches to optimize beamforming vectors in downlink multiuser multiantenna systems that can be universally applied to arbitrarily given transmit power limitation at a base station. We exploit the sum power budget as side information so that deep neural networks (DNNs) can effectively learn the impact of the power constraint in the beamforming optimization. Consequently, a single training process is sufficient for the proposed universal DL approach, whereas conventional methods need to train multiple DNNs for all possible power budget levels. Numerical results demonstrate the effectiveness of the proposed DL methods over existing schemes.

No abstract

No abstract

No abstract