El Niño-Southern Oscillation (ENSO) is one of the fundamental drivers of the Earth's climate variability. Thus, its skillful prediction at least a few months to years ahead is of utmost importance to society. Using both dynamical and statistical methods, several studies reported skillful ENSO predictions at various lead times. Predictions with long lead times, on the other hand, remain difficult. In this study, we propose a convolutional neural network (CNN)-based statistical ENSO prediction system with heterogeneous CNN parameters for each season with a modified loss function to predict ENSO at least 18–24 months ahead. The developed prediction system indicates that the CNN model is highly skillful in predicting ENSO at long lead times of 18–24 months with high skills in predicting extreme ENSO events compared with the Scale Interaction Experiment-Frontier ver. 2 (SINTEX-F2) dynamical system and several other statistical prediction systems. The analysis indicates that the CNN model can overcome the spring barrier, a major hindrance to dynamical prediction systems, in predicting ENSO at long lead times. The improvement in the prediction skill can partly be attributed to the heterogeneous parameters of seasonal CNN models used in this study and also to the use of a modified loss function in the CNN model. In this study, we also attempted to identify various precursors to ENSO events using CNN heatmap analysis.
Majority of CMOS image sensors in consumer market utilize a rolling shutter to increase sensitivity. However, it causes severe distortions, such as jitter, wobble, or skew. Since most of these kinds of sensors are used in hand-held devices, the approach of undistorting and generating stabilized images is restricted to resource limited systems. It has also been one of the major challenges to have a mathematical representation of CMOS rolling effect depicting the practical scenario, while keeping accuracy and stability. We propose that a CMOS sensor can be modeled by a section-wise charge-coupled devices model which has multiple homographies and exploit the observation that rolling shutter mechanism gives close relationships among them. We present a CMOS seven-parameter model, and show video stabilization algorithm by the iterative parameter estimation technique. We address four issues while accelerating our stabilization algorithm within resource limited environment: accuracy, stability, computation time, and resource utilization. We developed cache based optimization techniques to meet the requirement of the memory bandwidth and computational time for the iterative parameter estimation and final output image interpolation, and also proposed the incremental form of the seven-parameter model to greatly reduce resource consumption while maintaining the same results as the previous. The validity and effectiveness of our approach is demonstrated by experiments for different types of camera motions. The cache based optimization technique can be used to accelerate other types of iterative vision algorithms that require repetitive memory access: feature tracking, motion estimation, motion compensation, various types of image distortion correction, and also image warping and scaling.Index Terms-Cache-based optimization, CMOS sensor model, hardware interpolation technique, rolling shutter mechanism, video stabilization.
<p><strong>Abstract: </strong></p><p>The hydroclimatic teleconnections between global Sea Surface Temperature (SST) fields and monthly rainfall for the summer (June to August) and winter (December to February) seasons over east and west Japan (divided along 138&#176;E longitude) are investigated using the concept of Global Climate Pattern (GCP) (Chanda and Maity, 2015). It is established in a recent study that these teleconnections exhibit contrasting features and have different origins - rainfall anomalies over west Japan are associated with SST anomalies in the tropical Pacific and Indian Ocean, whereas those over east Japan are associated with high-latitude SST anomalies (Maity et al., 2020). Moreover, the teleconnections show inter-seasonal and intra-seasonal variations. For instance, the El Ni&#241;o Modoki (La Ni&#241;a Modoki) phenomena are found to influence the early summer (winter) rainfall over west Japan. In east Japan, early summer (June) and winter (December) rainfall is associated with positive SST anomaly differences in eastern sub-tropical Pacific and south Pacific respectively. Further, the study establishes that, beyond the traditional teleconnection patterns such as ENSO, El Ni&#241;o Modoki, other climatic precursors are also instrumental in triggering below- and above- normal monthly rainfall in east and west Japan. The predictive potential of all such identified teleconnection patterns for monthly rainfall variation is assessed using a machine learning approach, Support Vector Regression (SVR) and a hybrid Graphical Modelling/C-Vine copula (GM-Copula) approach. The later technique helps to construct a conditional independence structure among the correlated variables to prune the redundant information in the predictor pool and develop a month-wise prediction model using the pruned predictor sets only. It is observed that the complex association between the predictors and the predictand is better captured by this GM-Copula approach with slightly better prediction performance in summer (R = 0.66 to 0.70) than in winter (R = 0.45 to 0.75) for both east and west Japan. Thus, it is concluded that, establishing the conditional dependence structure of the predictor pool is an important step to resolve the complexity and dimensionality of the model and the proposed model may be recommended for operational forecast of monthly rainfall over east and west Japan. Further details can be found in Maity et al., (2020).</p><p><strong>Keywords:</strong> Rainfall prediction, Hydroclimatic teleconnection, Global climate pattern, Sea surface temperature, Machine learning, SVR, Graphical Model, Copula, Japan.</p><p><strong>References:</strong></p><p>Chanda K. and R. Maity, (2015). Uncovering Global Climate Fields Causing Local Precipitation Extremes. Hydrological Sciences Journal, Taylor and Francis. doi: 10.1080/02626667.2015.1006232.</p><p>Maity, R., K. Chanda, R. Dutta, J.V. Ratnam, M. Nonaka, S. Behera (2020), Contrasting features of hydroclimatic teleconnections and the predictability of seasonal rainfall over east and west Japan, Meteorological Applications, Royal Meteorological Society (RMetS), In Press, doi: DOI: 10.1002/met.1881.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.