Search for: All records

Abstract Sustainable biofuel cropping systems aim to address climate change while meeting energy needs. Understanding how soil and plant‐associated microbes respond to these different cropping systems is key to promoting agriculture sustainability and evaluating changes in ecosystem functions. Here, we leverage a long‐term biofuel cropping system field experiment to dissect soil and root microbiome changes across a soil‐depth gradient in poplar, restored prairie and switchgrass to understand their effects on the microbial communities. High throughput amplicon sequencing of the fungal internal transcribed spacer (ITS) and prokaryotic 16S DNA regions showed a common trend of root and soil microbial community richness decreasing and evenness increasing with depth. Ecological niche (root vs. soil) had the strongest effect on community structure, followed by depth, then crop. Stochastic processes dominated the structuring of fungal communities in deeper soil layers while operational taxonomic units (OTUs) in surface soil layers were more likely to co‐occur and to be enriched by plant hosts. Prokaryotic communities were dispersal limited at deeper depths. Microbial networks showed a higher density, connectedness, average degree and module size in deeper soils. We observed a decrease in fungal‐fungal links and an increase of bacteria–bacteria links with increasing depth in all crops, particularly in the root microbiome. 

Wild waterbirds, and especially wild waterfowl, are considered to be a reservoir for avian influenza viruses, with transmission likely occurring at the agricultural-wildlife interface. In the past few decades, avian influenza has repeatedly emerged in China along the East Asian-Australasian Flyway (EAAF), where extensive habitat conversion has occurred. Rapid environmental changes in the EAAF, especially distributional changes in rice paddy agriculture, have the potential to affect both the movements of wild migratory birds and the likelihood of spillover at the agricultural-wildlife interface. To begin to understand the potential implications such changes may have on waterfowl and disease transmission risk, we created dynamic Brownian Bridge Movement Models (dBBMM) based on waterfowl telemetry data. We used these dBBMM models to create hypothetical scenarios that would predict likely changes in waterfowl distribution relative to recent changes in rice distribution quantified through remote sensing. Our models examined a range of responses in which increased availability of rice paddies would drive increased use by waterfowl and decreased availability would result in decreased use, predicted from empirical data. Results from our scenarios suggested that in southeast China, relatively small decreases in rice agriculture could lead to dramatic loss of stopover habitat, and in northeast China, increases in rice paddies should provide new areas that can be used by waterfowl. Finally, we explored the implications of how such scenarios of changing waterfowl distribution may affect the potential for avian influenza transmission. Our results provide advance understanding of changing disease transmission threats by incorporating real-world data that predicts differences in habitat utilization by migratory birds over time. 

Neuromorphic computing systems promise high energy efficiency and low latency. In particular, when integrated with neuromorphic sensors, they can be used to produce intelligent systems for a broad range of applications. An event‐based camera is such a neuromorphic sensor, inspired by the sparse and asynchronous spike representation of the biological visual system. However, processing the event data requires either using expensive feature descriptors to transform spikes into frames, or using spiking neural networks (SNNs) that are expensive to train. In this work, a neural network architecture is proposed, reservoir nodes‐enabled neuromorphic vision sensing network (RN‐Net), based on dynamic temporal encoding by on‐sensor reservoirs and simple deep neural network (DNN) blocks. The reservoir nodes enable efficient temporal processing of asynchronous events by leveraging the native dynamics of the node devices, while the DNN blocks enable spatial feature processing. Combining these blocks in a hierarchical structure, the RN‐Net offers efficient processing for both local and global spatiotemporal features. RN‐Net executes dynamic vision tasks created by event‐based cameras at the highest accuracy reported to date at one order of magnitude smaller network size. The use of simple DNN and standard backpropagation‐based training rules further reduces implementation and training costs.