Improving Summer Precipitation Prediction in China Using Deep Learning

2021 ◽  
Author(s):  
Weixin Jin ◽  
Yong Luo
Keyword(s):  
Deep Learning ◽  
General Circulation ◽  
Dynamic Models ◽  
Summer Precipitation ◽  
Seasonal Prediction ◽  
Representation Learning ◽  
Feature Representation ◽  
Model Assessment ◽  
State Variables ◽  
Precipitation Prediction

<p>Summer precipitation in China exhibits considerable spatial-temporal variation with direct social and economic impact. Yet seasonal prediction remains a long-standing challenge. The dynamical models even with a 1-month lead still shows limited forecast skill over China in summer. The present study focuses on applying deep learning to summer precipitation prediction in China. We train a convolutional neural network (CNN) on seasonal retrospective forecast from forecast centres in several European countries, and subsequently use transfer learning on reanalysis and observational data of 160 stations over China. <span>The Pearson’s correlation coefficient (PCC) and the root mean square error (RMSE) </span><span>are used to evaluate the performance of precipitation forecasts.</span> <span>The results demonstrate</span> that deep learning approach produces skillful forecast better than those of current state-of-the-art dynamical forecast systems and traditional statistical methods in downscaling, with <span>PCC increasing by 0.1–0.3, at 1–3 months leads</span>. <span>Moreover, experiments show that </span>the data-driven model is capable to learn the complex relationship of input atmospheric state variables from reanalysis data and precipitation from station observations, with PCC of about 0.69. Image-Occlusion technique are also performed to determine variables and  spatial features of the general circulation in the Northern Hemisphere which contribute maximally to the spatial distribution of summer precipitation in China <span>through the automatic feature representation learning</span>, and help evaluate the weakness of dynamic models, in order to gain a better understanding of the factors that limit the capability to seasonal prediction. It suggests that deep learning is a powerful tool suitable for both seasonal prediction and for dynamical model assessment.</p>

scholarly journals SCL-SSC: Supervised Contrastive Learning for Sleep Stage Classification

2022 ◽  
Author(s):  
Chandra Bhushan Kumar
Keyword(s):  
Deep Learning ◽  
Sleep Stage ◽  
Synthetic Data ◽  
Representation Learning ◽  
Feature Representation ◽  
Training Dataset ◽  
Sleep Stages ◽  
Eeg Signals ◽  
Stage Classification ◽  
Sleep Stage Classification

<div>In this study, we have proposed SCL-SSC(Supervised Contrastive Learning for Sleep Stage Classification), a deep learning-based framework for sleep stage classification which performs the task in two stages, 1) feature representation learning, and 2) classification. The feature learner is trained separately to represent the raw EEG signals in the feature space such that the distance between the embedding of EEG signals of the same sleep stage has less than the distance between the embedding of EEG signals of different sleep stages in the euclidean space. On top of feature learners, we have trained the classifier to perform the classification task. The distribution of sleep stages is not uniform in the PSG data, wake(W) and N2 sleep stages appear more frequently than the other sleep stages, which leads to an imbalance dataset problem. This paper addresses this issue by using weighted softmax cross-entropy loss function and also dataset oversampling technique utilized to produce synthetic data points for minority sleep stages for approximately balancing the number of sleep stages in the training dataset. The performance of our proposed model is evaluated on the publicly available Physionet datasets EDF-Sleep 2013 and 2018 versions. We have trained and evaluated our model on two EEG channels (Fpz-Cz and Pz-Oz) on these datasets separately. The evaluation result shows that the performance of SCL-SSC is the best annotation performance compared to the existing state-of art deep learning algorithms to our best of knowledge, with an overall accuracy of 94.1071% with a macro F1 score of 92.6416 and Cohen’s Kappa coefficient(κ) 0.9197. Our ablation studies on SCL-SSC shows that both triplet loss based pre-training of feature learner and oversampling of minority classes are contributing to better performance of the model(SCL-SSC).</div>

scholarly journals SCL-SSC: Supervised Contrastive Learning for Sleep Stage Classification

2022 ◽  
Author(s):  
Chandra Bhushan Kumar
Keyword(s):  
Deep Learning ◽  
Sleep Stage ◽  
Synthetic Data ◽  
Representation Learning ◽  
Feature Representation ◽  
Training Dataset ◽  
Sleep Stages ◽  
Eeg Signals ◽  
Stage Classification ◽  
Sleep Stage Classification

<div>In this study, we have proposed SCL-SSC(Supervised Contrastive Learning for Sleep Stage Classification), a deep learning-based framework for sleep stage classification which performs the task in two stages, 1) feature representation learning, and 2) classification. The feature learner is trained separately to represent the raw EEG signals in the feature space such that the distance between the embedding of EEG signals of the same sleep stage has less than the distance between the embedding of EEG signals of different sleep stages in the euclidean space. On top of feature learners, we have trained the classifier to perform the classification task. The distribution of sleep stages is not uniform in the PSG data, wake(W) and N2 sleep stages appear more frequently than the other sleep stages, which leads to an imbalance dataset problem. This paper addresses this issue by using weighted softmax cross-entropy loss function and also dataset oversampling technique utilized to produce synthetic data points for minority sleep stages for approximately balancing the number of sleep stages in the training dataset. The performance of our proposed model is evaluated on the publicly available Physionet datasets EDF-Sleep 2013 and 2018 versions. We have trained and evaluated our model on two EEG channels (Fpz-Cz and Pz-Oz) on these datasets separately. The evaluation result shows that the performance of SCL-SSC is the best annotation performance compared to the existing state-of art deep learning algorithms to our best of knowledge, with an overall accuracy of 94.1071% with a macro F1 score of 92.6416 and Cohen’s Kappa coefficient(κ) 0.9197. Our ablation studies on SCL-SSC shows that both triplet loss based pre-training of feature learner and oversampling of minority classes are contributing to better performance of the model(SCL-SSC).</div>

scholarly journals A novel framework for spatio-temporal prediction of environmental data using deep learning

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Federico Amato ◽  
Fabian Guignard ◽  
Sylvain Robert ◽  
Mikhail Kanevski
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Nonlinear Function ◽  
Representation Learning ◽  
Machine Learning Algorithms ◽  
Feature Representation ◽  
Environmental Data ◽  
Real World Data ◽  
Temporal Prediction ◽  
Spatio Temporal

AbstractAs the role played by statistical and computational sciences in climate and environmental modelling and prediction becomes more important, Machine Learning researchers are becoming more aware of the relevance of their work to help tackle the climate crisis. Indeed, being universal nonlinear function approximation tools, Machine Learning algorithms are efficient in analysing and modelling spatially and temporally variable environmental data. While Deep Learning models have proved to be able to capture spatial, temporal, and spatio-temporal dependencies through their automatic feature representation learning, the problem of the interpolation of continuous spatio-temporal fields measured on a set of irregular points in space is still under-investigated. To fill this gap, we introduce here a framework for spatio-temporal prediction of climate and environmental data using deep learning. Specifically, we show how spatio-temporal processes can be decomposed in terms of a sum of products of temporally referenced basis functions, and of stochastic spatial coefficients which can be spatially modelled and mapped on a regular grid, allowing the reconstruction of the complete spatio-temporal signal. Applications on two case studies based on simulated and real-world data will show the effectiveness of the proposed framework in modelling coherent spatio-temporal fields.

Hierarchical Quality-Relevant Feature Representation for Soft Sensor Modeling: A Novel Deep Learning Strategy

2020 ◽  
Vol 16 (6) ◽  
pp. 3721-3730 ◽  
Author(s):  
Xiaofeng Yuan ◽  
Jiao Zhou ◽  
Biao Huang ◽  
Yalin Wang ◽  
Chunhua Yang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Learning Strategy ◽  
Soft Sensor ◽  
Feature Representation ◽  
Relevant Feature ◽  
Sensor Modeling

scholarly journals Representation Learning for Fine-Grained Change Detection

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4486
Author(s):  
Niall O’Mahony ◽  
Sean Campbell ◽  
Lenka Krpalkova ◽  
Anderson Carvalho ◽  
Joseph Walsh ◽  
...  
Keyword(s):  
Deep Learning ◽  
Change Detection ◽  
Model Calibration ◽  
State Of The Art ◽  
Representation Learning ◽  
Machine Intelligence ◽  
The State ◽  
Sensor Data ◽  
Fine Grained ◽  
Learning Techniques

Fine-grained change detection in sensor data is very challenging for artificial intelligence though it is critically important in practice. It is the process of identifying differences in the state of an object or phenomenon where the differences are class-specific and are difficult to generalise. As a result, many recent technologies that leverage big data and deep learning struggle with this task. This review focuses on the state-of-the-art methods, applications, and challenges of representation learning for fine-grained change detection. Our research focuses on methods of harnessing the latent metric space of representation learning techniques as an interim output for hybrid human-machine intelligence. We review methods for transforming and projecting embedding space such that significant changes can be communicated more effectively and a more comprehensive interpretation of underlying relationships in sensor data is facilitated. We conduct this research in our work towards developing a method for aligning the axes of latent embedding space with meaningful real-world metrics so that the reasoning behind the detection of change in relation to past observations may be revealed and adjusted. This is an important topic in many fields concerned with producing more meaningful and explainable outputs from deep learning and also for providing means for knowledge injection and model calibration in order to maintain user confidence.

scholarly journals Simulation of Extreme Precipitation in Four Climate Regions in China by General Circulation Models (GCMs): Performance and Projections

Water ◽  
2021 ◽  
Vol 13 (11) ◽  
pp. 1509
Author(s):  
Mengru Zhang ◽  
Xiaoli Yang ◽  
Liliang Ren ◽  
Ming Pan ◽  
Shanhu Jiang ◽  
...  
Keyword(s):  
Extreme Precipitation ◽  
General Circulation ◽  
Global Climate ◽  
Summer Precipitation ◽  
General Circulation Models ◽  
Cumulative Distribution ◽  
Extreme Precipitation Events ◽  
Increasing Trend ◽  
The Future ◽  
Semi Arid

In the context of global climate change, it is important to monitor abnormal changes in extreme precipitation events that lead to frequent floods. This research used precipitation indices to describe variations in extreme precipitation and analyzed the characteristics of extreme precipitation in four climatic (arid, semi-arid, semi-humid and humid) regions across China. The equidistant cumulative distribution function (EDCDF) method was used to downscale and bias-correct daily precipitation in eight Coupled Model Intercomparison Project Phase 5 (CMIP5) general circulation models (GCMs). From 1961 to 2005, the humid region had stronger and longer extreme precipitation compared with the other regions. In the future, the projected extreme precipitation is mainly concentrated in summer, and there will be large areas with substantial changes in maximum consecutive 5-day precipitation (Rx5) and precipitation intensity (SDII). The greatest differences between two scenarios (RCP4.5 and RCP8.5) are in semi-arid and semi-humid areas for summer precipitation anomalies. However, the area of the four regions with an increasing trend of extreme precipitation is larger under the RCP8.5 scenario than that under the RCP4.5 scenario. The increasing trend of extreme precipitation in the future is relatively pronounced, especially in humid areas, implying a potential heightened flood risk in these areas.

scholarly journals A Climate Downscaling Deep Learning Model considering the Multiscale Spatial Correlations and Chaos of Meteorological Events

2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Bin Mu ◽  
Bo Qin ◽  
Shijin Yuan ◽  
Xiaoyun Qin
Keyword(s):  
Deep Learning ◽  
General Circulation ◽  
Dynamical Downscaling ◽  
General Circulation Models ◽  
Learning Model ◽  
Spatial Correlations ◽  
Reconstruction Process ◽  
Daily Precipitation Data ◽  
Climate Downscaling ◽  
Deep Learning Model

Climate downscaling is a way to provide finer resolution data at local scales, which has been widely used in meteorological research. The two main approaches for climate downscaling are dynamical and statistical. The traditional dynamical downscaling methods are quite time- and resource-consuming based on general circulation models (GCMs). Recently, more and more researchers construct a statistical deep learning model for climate downscaling motivated by the single-image superresolution (SISR) process in computer vision (CV). This is an approach that uses historical climate observations to learn a low-resolution to high-resolution mapping and produces great enhancements in terms of efficiency and effectiveness. Therefore, it has provided an appreciable new insight and successful downscaling solution to multiple climate phenomena. However, most existing models only make a simple analogy between climate downscaling and SISR and ignore the underlying dynamical mechanisms, which leads to the overaveraged downscaling results lacking crucial physical details. In this paper, we incorporate the a priori meteorological knowledge into a deep learning formalization for climate downscaling. More specifically, we consider the multiscale spatial correlations and the chaos in multiple climate events. Depending on two characteristics, we build up a two-stage deep learning model containing a stepwise reconstruction process and ensemble inference, which is named climate downscaling network (CDN). It can extract more local/remote spatial dependencies and provide more comprehensive captures of extreme conditions. We evaluate our model based on two datasets: climate science dataset (CSD) and benchmark image dataset (BID). The results demonstrate that our model shows the effectiveness and superiority in downscaling daily precipitation data from 2.5 degrees to 0.5 degrees over Asia and Europe. In addition, our model exhibits better performance than the other traditional approaches and state-of-the-art deep learning models.

scholarly journals Analog Downscaling of Seasonal Rainfall Forecasts in the Murray Darling Basin

2013 ◽  
Vol 141 (3) ◽  
pp. 1099-1117 ◽  
Author(s):  
Andrew Charles ◽  
Bertrand Timbal ◽  
Elodie Fernandez ◽  
Harry Hendon
Keyword(s):  
General Circulation ◽  
Large Scale ◽  
Seasonal Prediction ◽  
General Circulation Models ◽  
Meridional Wind ◽  
Seasonal Forecasts ◽  
Local Climate ◽  
Continental Scale ◽  
Predictive Skill ◽  
Murray Darling Basin

Abstract Seasonal predictions based on coupled atmosphere–ocean general circulation models (GCMs) provide useful predictions of large-scale circulation but lack the conditioning on topography required for locally relevant prediction. In this study a statistical downscaling model based on meteorological analogs was applied to continental-scale GCM-based seasonal forecasts and high quality historical site observations to generate a set of downscaled precipitation hindcasts at 160 sites in the South Murray Darling Basin region of Australia. Large-scale fields from the Predictive Ocean–Atmosphere Model for Australia (POAMA) 1.5b GCM-based seasonal prediction system are used for analog selection. Correlation analysis indicates modest levels of predictability in the target region for the selected predictor fields. A single best-match analog was found using model sea level pressure, meridional wind, and rainfall fields, with the procedure applied to 3-month-long reforecasts, initialized on the first day of each month from 1980 to 2006, for each model day of 10 ensemble members. Assessment of the total accumulated rainfall and number of rainy days in the 3-month reforecasts shows that the downscaling procedure corrects the local climate variability with no mean effect on predictive skill, resulting in a smaller magnitude error. The amount of total rainfall and number of rain days in the downscaled output is significantly improved over the direct GCM output as measured by the difference in median and tercile thresholds between station observations and downscaled rainfall. Confidence in the downscaled output is enhanced by strong consistency between the large-scale mean of the downscaled and direct GCM precipitation.

Sign in / Sign up

Export Citation Format

Share Document