Search for: All records

Abstract Climate change can alter wetland extent and function, but such impacts are perplexing. Here, changes in wetland characteristics over North America from 25° to 53° North are projected under two climate scenarios using a state-of-the-science Earth system model. At the continental scale, annual wetland area decreases by ~10% (6%-14%) under the high emission scenario, but spatiotemporal changes vary, reaching up to ±50%. As the dominant driver of these changes shifts from precipitation to temperature in the higher emission scenario, wetlands undergo substantial drying during summer season when biotic processes peak. The projected disruptions to wetland seasonality cycles imply further impacts on biodiversity in major wetland habitats of upper Mississippi, Southeast Canada, and the Everglades. Furthermore, wetlands are projected to significantly shrink in cold regions due to the increased infiltration as warmer temperature reduces soil ice. The large dependence of the projections on climate change scenarios underscores the importance of emission mitigation to sustaining wetland ecosystems in the future. 

Abstract Agricultural irrigation has experienced rapid expansion, and its growing freshwater consumption is potentially exacerbating water scarcity issues. Previous studies predominantly relied on observations or land-only simulations, often neglecting land–atmosphere interactions or failing to capture long-term evolution. We therefore analyse the effects of historical irrigation expansion on water fluxes and resources using seven Earth system models. Here we show that irrigation expansion in many regions substantially decreases the net water influx from the atmosphere to land, further aggravating the existing drying trends caused by climate change. For example, irrigation expansion changed the trend of this net influx from −0.664 ( ± 0.283) to −1.461 ( ± 0.261) mm yr⁻²in South Asia after 1960. Consequently, the local terrestrial water storage depletion rate is substantially enlarged by irrigation expansion (for example, from −2.559 ( ± 0.094) to −16.008 ( ± 0.557) mm yr⁻¹). Our results attribute the land water loss to irrigation expansion and climate change, calling for immediate solutions to tackle the negative trends. 

Attention-based models have achieved many remarkable breakthroughs in numerous applications. However, the quadratic complexity of Attention makes the vanilla Attentionbased models hard to apply to long sequence tasks. Various improved Attention structures are proposed to reduce the computation cost by inducing low rankness and approximating the whole sequence by sub-sequences. The most challenging part of those approaches is maintaining the proper balance between information preservation and computation reduction: the longer sub-sequences used, the better information is preserved, but at the price of introducing more noise and computational costs. In this paper, we propose a smoothed skeleton sketching based Attention structure, coined S3Attention, which significantly improves upon the previous attempts to negotiate this trade-off. S3Attention has two mechanisms to effectively minimize the impact of noise while keeping the linear complexity to the sequence length: a smoothing block to mix information over long sequences and a matrix sketching method that simultaneously selects columns and rows from the input matrix. We verify the effectiveness of S3Attention both theoretically and empirically. Extensive studies over Long Range Arena (LRA) datasets and six time-series forecasting show that S3Attention significantly outperforms both vanilla Attention and other state-of-the-art variants of Attention structures. 

With increasingly deployed cameras and the rapid advances of Computer Vision, large-scale live video analytics becomes feasible. However, analyzing videos is compute-intensive. In addition, live video analytics needs to be performed in real time. In this paper, we design an edge server system for live video analytics. We propose to perform configuration adaptation without profiling video online. We select configurations with a prediction model based on object movement features. In addition, we reduce the latency through resource orchestration on video analytics servers. The key idea of resource orchestration is to batch inference tasks that use the same CNN model, and schedule tasks based on a priority value that estimates their impact on the total latency. We evaluate our system with two video analytic applications, road traffic monitoring and pose detection. The experimental results show that our profiling-free adaptation reduces the workload by 80% of the state-of-the-art adaptation without lowering the accuracy. The average serving latency is reduced by up to 95% comparing with the profiling-based adaptation. 

Graph convolutional network (GCN) has been shown effective in many applications with graph structures. However, training a large-scale GCN is still challenging due to the high computation cost that grows with the size of the graph. In this paper, we propose CM-GCN, a distributed GCN framework using cohesive mini-batches to accelerate large-scale GCN training. The cohesive mini-batches group nodes that are tightly connected in the graph. As a result, CM-GCN can reduce the computation required to train a GCN. We propose a computation cost function to quantify the computation required for mini-batches. By exploring the submodular property of the computation cost function, we develop an efficient algorithm to partition nodes into tightly coupled mini-batches. Based on the computation cost function, we evenly distribute the workloads of mini-batches to workers. We design asynchronous computations between GCN layers to further eliminating the waiting among workers. We implement a CM-GCN framework and evaluate its performance with graphs that contain millions of nodes. Our evaluation shows that CM-GCN can achieve up to 3X speedup without compromising the training accuracy.