Argo-Two Decades: Global Oceanography, Revolutionized

Argo, an international, global observational array of nearly 4,000 autonomous robotic profiling floats, each measuring ocean temperature and salinity from 0 to 2,000 m on nominal 10-day cycles, has revolutionized physical oceanography. Argo started at the turn of the millennium,growing out of advances in float technology over the previous several decades. After two decades, with well over 2 million profiles made publicly available in real time, Argo data have underpinned more than 4,000 scientific publications and improved countless nowcasts, forecasts, and projections. We review a small subset of those accomplishments, such as elucidating remarkable zonal jets spanning the deep tropical Pacific; increasing understanding of ocean eddies and the roles of mixing in shaping water masses and circulation; illuminating interannual to decadal ocean variability; quantifying, in concert with satellite data, contributions of ocean warming and ice melting to sea level rise; improving coupled numerical weather predictions; and underpinning decadal climate forecasts.

Opportunities and challenges for machine learning in weather and climate modelling: hard, medium and soft AI

In September 2019, a workshop was held to highlight the growing area of applying machine learning techniques to improve weather and climate prediction. In this introductory piece, we outline the motivations, opportunities and challenges ahead in this exciting avenue of research. This article is part of the theme issue 'Machine learning for weather and climate modelling'.

Domain-driven models yield better predictions at lower cost than reservoir computers in Lorenz systems

Recent advances in computing algorithms and hardware have rekindled interest in developing high-accuracy, low-cost surrogate models for simulating physical systems. The idea is to replace expensive numerical integration of complex coupled partial differential equations at fine time scales performed on supercomputers, with machine-learned surrogates that efficiently and accurately forecast future system states using data sampled from the underlying system. One particularly popular technique being explored within the weather and climate modelling community is the echo state network (ESN), an attractive alternative to other well-known deep learning architectures. Using the classical Lorenz 63 system, and the three tier multi-scale Lorenz 96 system (Thornes T, Duben P, Palmer T. 2017 Q. J. R. Meteorol. Soc. 143, 897-908. (doi:10.1002/qj.2974)) as benchmarks, we realize that previously studied state-of-the-art ESNs operate in two distinct regimes, corresponding to low and high spectral radius (LSR/HSR) for the sparse, randomly generated, reservoir recurrence matrix. Using knowledge of the mathematical structure of the Lorenz systems along with systematic ablation and hyperparameter sensitivity analyses, we show that state-of-the-art LSR-ESNs reduce to a polynomial regression model which we call Domain-Driven Regularized Regression (D2R2). Interestingly, D2R2 is a generalization of the well-known SINDy algorithm (Brunton SL, Proctor JL, Kutz JN. 2016 Proc. Natl Acad. Sci. USA 113, 3932-3937. (doi:10.1073/pnas.1517384113)). We also show experimentally that LSR-ESNs (Chattopadhyay A, Hassanzadeh P, Subramanian D. 2019 (http://arxiv.org/abs/1906.08829)) outperform HSR ESNs (Pathak J, Hunt B, Girvan M, Lu Z, Ott E. 2018 Phys. Rev. Lett. 120, 024102. (doi:10.1103/PhysRevLett.120.024102)) while D2R2 dominates both approaches. A significant goal in constructing surrogates is to cope with barriers to scaling in weather prediction and simulation of dynamical systems that are imposed by time and energy consumption in supercomputers. Inexact computing has emerged as a novel approach to helping with scaling. In this paper, we evaluate the performance of three models (LSR-ESN, HSR-ESN and D2R2) by varying the precision or word size of the computation as our inexactness-controlling parameter. For precisions of 64, 32 and 16 bits, we show that, surprisingly, the least expensive D2R2 method yields the most robust results and the greatest savings compared to ESNs. Specifically, D2R2 achieves 68 × in computational savings, with an additional 2 × if precision reductions are also employed, outperforming ESN variants by a large margin. This article is part of the theme issue 'Machine learning for weather and climate modelling'.

Stable reliability diagrams for probabilistic classifiers

A probability forecast or probabilistic classifier is reliable or calibrated if the predicted probabilities are matched by ex post observed frequencies, as examined visually in reliability diagrams. The classical binning and counting approach to plotting reliability diagrams has been hampered by a lack of stability under unavoidable, ad hoc implementation decisions. Here, we introduce the CORP approach, which generates provably statistically consistent, optimally binned, and reproducible reliability diagrams in an automated way. CORP is based on nonparametric isotonic regression and implemented via the pool-adjacent-violators (PAV) algorithm-essentially, the CORP reliability diagram shows the graph of the PAV-(re)calibrated forecast probabilities. The CORP approach allows for uncertainty quantification via either resampling techniques or asymptotic theory, furnishes a numerical measure of miscalibration, and provides a CORP-based Brier-score decomposition that generalizes to any proper scoring rule. We anticipate that judicious uses of the PAV algorithm yield improved tools for diagnostics and inference for a very wide range of statistical and machine learning methods.

Real-Time Weather Monitoring and Prediction Using City Buses and Machine Learning

Accurate weather data are important for planning our day-to-day activities. In order to monitor and predict weather information, a two-phase weather management system is proposed, which combines information processing, bus mobility, sensors, and deep learning technologies to provide real-time weather monitoring in buses and stations and achieve weather forecasts through predictive models. Based on the sensing measurements from buses, this work incorporates the strengths of local information processing and moving buses for increasing the measurement coverage and supplying new sensing data. In Phase I, given the weather sensing data, the long short-term memory (LSTM) model and the multilayer perceptron (MLP) model are trained and verified using the data of temperature, humidity, and air pressure of the test environment. In Phase II, the trained learning model is applied to predict the time series of weather information. In order to assess the system performance, we compare the predicted weather data with the actual sensing measurements from the Environment Protection Administration (EPA) and Central Weather Bureau (CWB) of Taichung observation station to evaluate the prediction accuracy. The results show that the proposed system has reliable performance at weather monitoring and a good forecast for one-day weather prediction via the trained models.

Strategy use in probabilistic categorization by rhesus macaques (Macaca mulatta) and capuchin monkeys (Cebus [Sapajus] apella)

Probabilistic categorization tasks present the learner with a set of possible responses and imperfect cue evidence of which response will be rewarded. A single, optimal integration of all available cues into an optimal response is possible given any set of evidence. In contrast, there are many possible uses of the cues that offer the learner suboptimal (but better than chance) responding. We presented a classic probabilistic categorization task to 3 rhesus macaques (Macaca mulatta) and 13 capuchin monkeys (Cebus [Sapajus] apella) to explore what strategies for integration of imperfectly predictive stimulus information would be used by the animals. Using the strategy analysis models that have been previously used to describe human strategy use in probabilistic categorization tasks, we fit each of thousands of blocks of responses to 25 types of response strategies ranging from complex cognitive strategies (e.g., optimal integration of all evidence) to heuristic strategies (e.g., identify a highly predictive cue and respond based only on its presence or absence) to rote behavior (e.g., choosing the same response every trial). Inferences about strategy use were highly stable within animals and were heterogeneous across animals, with some animals never using cue information and others using it fruitfully. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Analog Forecasting of Extreme-Causing Weather Patterns Using Deep Learning

Numerical weather prediction models require ever-growing computing time and resources but, still, have sometimes difficulties with predicting weather extremes. We introduce a data-driven framework that is based on analog forecasting (prediction using past similar patterns) and employs a novel deep learning pattern-recognition technique (capsule neural networks, CapsNets) and an impact-based autolabeling strategy. Using data from a large-ensemble fully coupled Earth system model, CapsNets are trained on midtropospheric large-scale circulation patterns (Z500) labeled 0-4 depending on the existence and geographical region of surface temperature extremes over North America several days ahead. The trained networks predict the occurrence/region of cold or heat waves, only using Z500, with accuracies (recalls) of 69-45% (77-48%) or 62-41% (73-47%) 1-5 days ahead. Using both surface temperature and Z500, accuracies (recalls) with CapsNets increase to ∼ 80% (88%). In both cases, CapsNets outperform simpler techniques such as convolutional neural networks and logistic regression, and their accuracy is least affected as the size of the training set is reduced. The results show the promises of multivariate data-driven frameworks for accurate and fast extreme weather predictions, which can potentially augment numerical weather prediction efforts in providing early warnings.

Crop Yield Prediction Using Deep Neural Networks

Crop yield is a highly complex trait determined by multiple factors such as genotype, environment, and their interactions. Accurate yield prediction requires fundamental understanding of the functional relationship between yield and these interactive factors, and to reveal such relationship requires both comprehensive datasets and powerful algorithms. In the 2018 Syngenta Crop Challenge, Syngenta released several large datasets that recorded the genotype and yield performances of 2,267 maize hybrids planted in 2,247 locations between 2008 and 2016 and asked participants to predict the yield performance in 2017. As one of the winning teams, we designed a deep neural network (DNN) approach that took advantage of state-of-the-art modeling and solution techniques. Our model was found to have a superior prediction accuracy, with a root-mean-square-error (RMSE) being 12% of the average yield and 50% of the standard deviation for the validation dataset using predicted weather data. With perfect weather data, the RMSE would be reduced to 11% of the average yield and 46% of the standard deviation. We also performed feature selection based on the trained DNN model, which successfully decreased the dimension of the input space without significant drop in the prediction accuracy. Our computational results suggested that this model significantly outperformed other popular methods such as Lasso, shallow neural networks (SNN), and regression tree (RT). The results also revealed that environmental factors had a greater effect on the crop yield than genotype.

More reliable forecasts with less precise computations: a fast-track route to cloud-resolved weather and climate simulators?

This paper sets out a new methodological approach to solving the equations for simulating and predicting weather and climate. In this approach, the conventionally hard boundary between the dynamical core and the sub-grid parametrizations is blurred. This approach is motivated by the relatively shallow power-law spectrum for atmospheric energy on scales of hundreds of kilometres and less. It is first argued that, because of this, the closure schemes for weather and climate simulators should be based on stochastic-dynamic systems rather than deterministic formulae. Second, as high-wavenumber elements of the dynamical core will necessarily inherit this stochasticity during time integration, it is argued that the dynamical core will be significantly over-engineered if all computations, regardless of scale, are performed completely deterministically and if all variables are represented with maximum numerical precision (in practice using double-precision floating-point numbers). As the era of exascale computing is approached, an energy- and computationally efficient approach to cloud-resolved weather and climate simulation is described where determinism and numerical precision are focused on the largest scales only.