Intensification and spatial homogenization of coastal upwelling under climate change.

The timing and strength of wind-driven coastal upwelling along the eastern margins of major ocean basins regulate the productivity of critical fisheries and marine ecosystems by bringing deep and nutrient-rich waters to the sunlit surface, where photosynthesis can occur. How coastal upwelling regimes might change in a warming climate is therefore a question of vital importance. Although enhanced land-ocean differential heating due to greenhouse warming has been proposed to intensify coastal upwelling by strengthening alongshore winds, analyses of observations and previous climate models have provided little consensus on historical and projected trends in coastal upwelling. Here we show that there are strong and consistent changes in the timing, intensity and spatial heterogeneity of coastal upwelling in response to future warming in most Eastern Boundary Upwelling Systems (EBUSs). An ensemble of climate models shows that by the end of the twenty-first century the upwelling season will start earlier, end later and become more intense at high but not low latitudes. This projected increase in upwelling intensity and duration at high latitudes will result in a substantial reduction of the existing latitudinal variation in coastal upwelling. These patterns are consistent across three of the four EBUSs (Canary, Benguela and Humboldt, but not California). The lack of upwelling intensification and greater uncertainty associated with the California EBUS may reflect regional controls associated with the atmospheric response to climate change. Given the strong linkages between upwelling and marine ecosystems, the projected changes in the intensity, timing and spatial structure of coastal upwelling may influence the geographical distribution of marine biodiversity.

Using deep transfer learning and satellite imagery to estimate urban air quality in data-poor regions.

Urban air pollution is a critical public health challenge in low-and-middle-income countries (LMICs). At the same time, LMICs tend to be data-poor, lacking adequate infrastructure to monitor air quality (AQ). As LMICs undergo rapid urbanization, the socio-economic burden of poor AQ will be immense. Here we present a globally scalable two-step deep learning (DL) based approach for AQ estimation in LMIC cities that mitigates the need for extensive AQ infrastructure on the ground. We train a DL model that can map satellite imagery to AQ in high-income countries (HICs) with sufficient ground data, and then adapt the model to learn meaningful AQ estimates in LMIC cities using transfer learning. The trained model can explain up to 54% of the variation in the AQ distribution of the target LMIC city without the need for target labels. The approach is demonstrated for Accra in Ghana, Africa, with AQ patterns learned and adapted from two HIC cities, specifically Los Angeles and New York.

Equitable infrastructure: Achieving resilient systems and restorative justice through policy and research innovation.

Recent major investments in infrastructure in the United States and globally present a crucial opportunity to embed equity within the heart of resilient infrastructure decision-making. Yet there is a notable absence of frameworks within the engineering and scientific fields for integrating equity into planning, design, and maintenance of infrastructure. Additionally, whole-of-government approaches to infrastructure, including the Justice40 Initiative, mimic elements of process management that support exploitative rather than exploratory innovation. These and other policies risk creating innovation traps that limit analytical and engineering advances necessary to prioritize equity in decision-making, identification and disruption of mechanisms that cause or contribute to inequities, and remediation of historic harms. Here, we propose a three-tiered framework toward equitable and resilient infrastructure through restorative justice, incremental policy innovation, and exploratory research innovation. This framework aims to ensure equitable access and benefits of infrastructure, minimize risk disparities, and embrace restorative justice to repair historical and systemic inequities. We outline incremental policy innovation and exploratory research action items to address and mitigate risk disparities, emphasizing the need for community-engaged research and the development of equity metrics. Among other action items, we recommend a certification system-referred to as Social, Environmental, and Economic Development (SEED)-to train infrastructure engineers and planners and ensure attentiveness to gaps that exist within and dynamically interact across each tier of the proposed framework. Through the framework and proposed actions, we advocate for a transformative vision for equitable infrastructure that emphasizes the interconnectedness of social, environmental, and technical dimensions in infrastructure planning, design, and maintenance.

Explainable deep learning for insights in El Niño and river flows.

The El Niño Southern Oscillation (ENSO) is a semi-periodic fluctuation in sea surface temperature (SST) over the tropical central and eastern Pacific Ocean that influences interannual variability in regional hydrology across the world through long-range dependence or teleconnections. Recent research has demonstrated the value of Deep Learning (DL) methods for improving ENSO prediction as well as Complex Networks (CN) for understanding teleconnections. However, gaps in predictive understanding of ENSO-driven river flows include the black box nature of DL, the use of simple ENSO indices to describe a complex phenomenon and translating DL-based ENSO predictions to river flow predictions. Here we show that eXplainable DL (XDL) methods, based on saliency maps, can extract interpretable predictive information contained in global SST and discover SST information regions and dependence structures relevant for river flows which, in tandem with climate network constructions, enable improved predictive understanding. Our results reveal additional information content in global SST beyond ENSO indices, develop understanding of how SSTs influence river flows, and generate improved river flow prediction, including uncertainty estimation. Observations, reanalysis data, and earth system model simulations are used to demonstrate the value of the XDL-CN based methods for future interannual and decadal scale climate projections.

Network-based restoration strategies maximize ecosystem recovery.

Redressing global patterns of biodiversity loss requires quantitative frameworks that can predict ecosystem collapse and inform restoration strategies. By applying a network-based dynamical approach to synthetic and real-world mutualistic ecosystems, we show that biodiversity recovery following collapse is maximized when extirpated species are reintroduced based solely on their total number of connections in the original interaction network. More complex network-based strategies that prioritize the reintroduction of species that improve 'higher order' topological features such as compartmentalization do not provide meaningful performance improvements. These results suggest that it is possible to design nearly optimal restoration strategies that maximize biodiversity recovery for data-poor ecosystems in order to ensure the delivery of critical natural services that fuel economic development, food security, and human health around the globe.

A Framework for Deep Learning Emulation of Numerical Models With a Case Study in Satellite Remote Sensing.

Numerical models based on physics represent the state of the art in Earth system modeling and comprise our best tools for generating insights and predictions. Despite rapid growth in computational power, the perceived need for higher model resolutions overwhelms the latest generation computers, reducing the ability of modelers to generate simulations for understanding parameter sensitivities and characterizing variability and uncertainty. Thus, surrogate models are often developed to capture the essential attributes of the full-blown numerical models. Recent successes of machine learning methods, especially deep learning (DL), across many disciplines offer the possibility that complex nonlinear connectionist representations may be able to capture the underlying complex structures and nonlinear processes in Earth systems. A difficult test for DL-based emulation, which refers to function approximation of numerical models, is to understand whether they can be comparable to traditional forms of surrogate models in terms of computational efficiency while simultaneously reproducing model results in a credible manner. A DL emulation that passes this test may be expected to perform even better than simple models with respect to capturing complex processes and spatiotemporal dependencies. Here, we examine, with a case study in satellite-based remote sensing, the hypothesis that DL approaches can credibly represent the simulations from a surrogate model with comparable computational efficiency. Our results are encouraging in that the DL emulation reproduces the results with acceptable accuracy and often even faster performance. We discuss the broader implications of our results in light of the pace of improvements in high-performance implementations of DL and the growing desire for higher resolution simulations in the Earth sciences.

Higher trends but larger uncertainty and geographic variability in 21st century temperature and heat waves.

Generating credible climate change and extremes projections remains a high-priority challenge, especially since recent observed emissions are above the worst-case scenario. Bias and uncertainty analyses of ensemble simulations from a global earth systems model show increased warming and more intense heat waves combined with greater uncertainty and large regional variability in the 21st century. Global warming trends are statistically validated across ensembles and investigated at regional scales. Observed heat wave intensities in the current decade are larger than worst-case projections. Model projections are relatively insensitive to initial conditions, while uncertainty bounds obtained by comparison with recent observations are wider than ensemble ranges. Increased trends in temperature and heat waves, concurrent with larger uncertainty and variability, suggest greater urgency and complexity of adaptation or mitigation decisions.

Resilience of Urban Transport Network-of-Networks under Intense Flood Hazards Exacerbated by Targeted Attacks.

Natural hazards including floods can trigger catastrophic failures in interdependent urban transport network-of-networks (NoNs). Population growth has enhanced transportation demand while urbanization and climate change have intensified urban floods. However, despite the clear need to develop actionable insights for improving the resilience of critical urban lifelines, the theory and methods remain underdeveloped. Furthermore, as infrastructure systems become more intelligent, security experts point to the growing threat of targeted cyber-physical attacks during natural hazards. Here we develop a hypothesis-driven resilience framework for urban transport NoNs, which we demonstrate on the London Rail Network (LRN). We find that topological attributes designed for maximizing efficiency rather than robustness render the network more vulnerable to compound natural-targeted disruptions including cascading failures. Our results suggest that an organizing principle for post-disruption recovery may be developed with network science principles. Our findings and frameworks can generalize to urban lifelines and more generally to real-world spatial networks.

DeepEmSat: Deep Emulation for Satellite Data Mining.

The growing volume of Earth science data available from climate simulations and satellite remote sensing offers unprecedented opportunity for scientific insight, while also presenting computational challenges. One potential area of impact is atmospheric correction, where physics-based numerical models retrieve surface reflectance information from top of atmosphere observations, and are too computationally intensive to be run in real time. Machine learning methods have demonstrated potential as fast statistical models for expensive simulations and for extracting credible insights from complex datasets. Here, we develop DeepEmSat: a deep learning emulator approach for atmospheric correction, and offer comparison against physics-based models to support the hypothesis that deep learning can make a contribution to the efficient processing of satellite images.

Author Correction: US Power Production at Risk from Water Stress in a Changing Climate.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

Relative performance of mutual information estimation methods for quantifying the dependence among short and noisy data.

Commonly used dependence measures, such as linear correlation, cross-correlogram, or Kendall's tau , cannot capture the complete dependence structure in data unless the structure is restricted to linear, periodic, or monotonic. Mutual information (MI) has been frequently utilized for capturing the complete dependence structure including nonlinear dependence. Recently, several methods have been proposed for the MI estimation, such as kernel density estimators (KDEs), k -nearest neighbors (KNNs), Edgeworth approximation of differential entropy, and adaptive partitioning of the XY plane. However, outstanding gaps in the current literature have precluded the ability to effectively automate these methods, which, in turn, have caused limited adoptions by the application communities. This study attempts to address a key gap in the literature-specifically, the evaluation of the above methods to choose the best method, particularly in terms of their robustness for short and noisy data, based on comparisons with the theoretical MI estimates, which can be computed analytically, as well with linear correlation and Kendall's tau . Here we consider smaller data sizes, such as 50, 100, and 1000, and within this study we characterize 50 and 100 data points as very short and 1000 as short. We consider a broader class of functions, specifically linear, quadratic, periodic, and chaotic, contaminated with artificial noise with varying noise-to-signal ratios. Our results indicate KDEs as the best choice for very short data at relatively high noise-to-signal levels whereas the performance of KNNs is the best for very short data at relatively low noise levels as well as for short data consistently across noise levels. In addition, the optimal smoothing parameter of a Gaussian kernel appears to be the best choice for KDEs while three nearest neighbors appear optimal for KNNs. Thus, in situations where the approximate data sizes are known in advance and exploratory data analysis and/or domain knowledge can be used to provide a priori insights into the noise-to-signal ratios, the results in the paper point to a way forward for automating the process of MI estimation.

US Power Production at Risk from Water Stress in a Changing Climate.

Thermoelectric power production in the United States primarily relies on wet-cooled plants, which in turn require water below prescribed design temperatures, both for cooling and operational efficiency. Thus, power production in US remains particularly vulnerable to water scarcity and rising stream temperatures under climate change and variability. Previous studies on the climate-water-energy nexus have primarily focused on mid- to end-century horizons and have not considered the full range of uncertainty in climate projections. Technology managers and energy policy makers are increasingly interested in the decadal time scales to understand adaptation challenges and investment strategies. Here we develop a new approach that relies on a novel multivariate water stress index, which considers the joint probability of warmer and scarcer water, and computes uncertainties arising from climate model imperfections and intrinsic variability. Our assessments over contiguous US suggest consistent increase in water stress for power production with about 27% of the production severely impacted by 2030s.

Network Science Based Quantification of Resilience Demonstrated on the Indian Railways Network.

The structure, interdependence, and fragility of systems ranging from power-grids and transportation to ecology, climate, biology and even human communities and the Internet have been examined through network science. While response to perturbations has been quantified, recovery strategies for perturbed networks have usually been either discussed conceptually or through anecdotal case studies. Here we develop a network science based quantitative framework for measuring, comparing and interpreting hazard responses as well as recovery strategies. The framework, motivated by the recently proposed temporal resilience paradigm, is demonstrated with the Indian Railways Network. Simulations inspired by the 2004 Indian Ocean Tsunami and the 2012 North Indian blackout as well as a cyber-physical attack scenario illustrate hazard responses and effectiveness of proposed recovery strategies. Multiple metrics are used to generate various recovery strategies, which are simply sequences in which system components should be recovered after a disruption. Quantitative evaluation of these strategies suggests that faster and more efficient recovery is possible through network centrality measures. Optimal recovery strategies may be different per hazard, per community within a network, and for different measures of partial recovery. In addition, topological characterization provides a means for interpreting the comparative performance of proposed recovery strategies. The methods can be directly extended to other Large-Scale Critical Lifeline Infrastructure Networks including transportation, water, energy and communications systems that are threatened by natural or human-induced hazards, including cascading failures. Furthermore, the quantitative framework developed here can generalize across natural, engineered and human systems, offering an actionable and generalizable approach for emergency management in particular as well as for network resilience in general.

Asymmetry of projected increases in extreme temperature distributions.

A statistical analysis reveals projections of consistently larger increases in the highest percentiles of summer and winter temperature maxima and minima versus the respective lowest percentiles, resulting in a wider range of temperature extremes in the future. These asymmetric changes in tail distributions of temperature appear robust when explored through 14 CMIP5 climate models and three reanalysis datasets. Asymmetry of projected increases in temperature extremes generalizes widely. Magnitude of the projected asymmetry depends significantly on region, season, land-ocean contrast, and climate model variability as well as whether the extremes of consideration are seasonal minima or maxima events. An assessment of potential physical mechanisms provides support for asymmetric tail increases and hence wider temperature extremes ranges, especially for northern winter extremes. These results offer statistically grounded perspectives on projected changes in the IPCC-recommended extremes indices relevant for impacts and adaptation studies.