scholarly journals AutoNowP: An Approach Using Deep Autoencoders for Precipitation Nowcasting Based on Weather Radar Reflectivity Prediction

Mathematics ◽  
2021 ◽  
Vol 9 (14) ◽  
pp. 1653
Author(s):  
Gabriela Czibula ◽  
Andrei Mihai ◽  
Alexandra-Ioana Albu ◽  
Istvan-Gergely Czibula ◽  
Sorin Burcea ◽  
...  
Keyword(s):  
Binary Classification ◽  
Weather Radar ◽  
Predictive Performance ◽  
Weather Conditions ◽  
Radar Data ◽  
Probability Of Detection ◽  
Radar Reflectivity ◽  
Classification Model ◽  
Precipitation Forecast ◽  
Short Term

Short-term quantitative precipitation forecast is a challenging topic in meteorology, as the number of severe meteorological phenomena is increasing in most regions of the world. Weather radar data is of utmost importance to meteorologists for issuing short-term weather forecast and warnings of severe weather phenomena. We are proposing AutoNowP, a binary classification model intended for precipitation nowcasting based on weather radar reflectivity prediction. Specifically, AutoNowP uses two convolutional autoencoders, being trained on radar data collected on both stratiform and convective weather conditions for learning to predict whether the radar reflectivity values will be above or below a certain threshold. AutoNowP is intended to be a proof of concept that autoencoders are useful in distinguishing between convective and stratiform precipitation. Real radar data provided by the Romanian National Meteorological Administration and the Norwegian Meteorological Institute is used for evaluating the effectiveness of AutoNowP. Results showed that AutoNowP surpassed other binary classifiers used in the supervised learning literature in terms of probability of detection and negative predictive value, highlighting its predictive performance.

Impact of Doppler radar reflectivity and velocity data assimilation on the quality of precipitation forecasting in Belarus in different seasons

2020 ◽  
Author(s):  
Palina Zaiko ◽  
Siarhei Barodka ◽  
Aliaksandr Krasouski
Keyword(s):  
Data Assimilation ◽  
Weather Radar ◽  
Radar Data ◽  
Radar Reflectivity ◽  
Precipitation Forecast ◽  
Doppler Weather Radar ◽  
Velocity Data ◽  
Numerical Weather ◽  
Convective Processes ◽  
Different Seasons

<p>Heavy precipitation forecast remains one of the biggest problems in numerical weather prediction. Modern remote sensing systems allow tracking of rapidly developing convective processes and provide additional data for numerical weather models practically in real time. Assimilation of Doppler weather radar data also allows to specify the position and intensity of convective processes in atmospheric numerical models.</p><p>The primary objective of this study is to evaluate the impact of Doppler  radar reflectivity and velocity assimilation in the WRF-ARW mesoscale model for the territory of Belarus in different seasons of the year. Specifically, we focus on the short-range numerical forecasting of mesoscale convective systems passage over the territory of Belarus in 2017-2019 with assimilated radar data.</p><p>Proceeding with weather radar observations available for our cases, we first perform the necessary processing of the raw radar data to eliminate noise, reflections and other kinds of clutter. For identification of non-meteorological noise fuzzy echo classification was used. Then we use the WRF-DA (3D-Var) system to assimilate the processed radar observations from 3 Belarusian Doppler weather radar in the WRF model. Assimilating both radar reflectivity and radial velocity data in the model we aim to better represent not only the distribution of clouds and their moisture content, but also the detailed dynamical aspects of convective circulation. Finally, we analyze WRF modelling output obtained with assimilated radar data and compare it with available meteorological observations and with other model runs (including control runs with no data assimilation or with assimilation of conventional weather stations data only), paying special attention to the accuracy of precipitation forecast 12 hours in advance.</p>

scholarly journals Research on Short-Term Load Prediction Based on Seq2seq Model

Energies ◽  
2019 ◽  
Vol 12 (16) ◽  
pp. 3199 ◽  
Author(s):  
Gangjun Gong ◽  
Xiaonan An ◽  
Nawaraj Kumar Mahato ◽  
Shuyan Sun ◽  
Si Chen ◽  
...  
Keyword(s):  
Power System ◽  
Prediction Model ◽  
Short Term Memory ◽  
Predictive Performance ◽  
Weather Conditions ◽  
Attention Mechanism ◽  
Load Prediction ◽  
Short Term ◽  
System Load ◽  
The Impact

Electricity load prediction is the primary basis on which power-related departments to make logical and effective generation plans and scientific scheduling plans for the most effective power utilization. The perpetual evolution of deep learning has recommended advanced and innovative concepts for short-term load prediction. Taking into consideration the time and nonlinear characteristics of power system load data and further considering the impact of historical and future information on the current state, this paper proposes a Seq2seq short-term load prediction model based on a long short-term memory network (LSTM). Firstly, the periodic fluctuation characteristics of users’ load data are analyzed, establishing a correlation of the load data so as to determine the model’s order in the time series. Secondly, the specifications of the Seq2seq model are given preference and a coalescence of the Residual mechanism (Residual) and the two Attention mechanisms (Attention) is developed. Then, comparing the predictive performance of the model under different types of Attention mechanism, this paper finally adopts the Seq2seq short-term load prediction model of Residual LSTM and the Bahdanau Attention mechanism. Eventually, the prediction model obtains better results when merging the actual power system load data of a certain place. In order to validate the developed model, the Seq2seq was compared with recurrent neural network (RNN), LSTM, and gated recurrent unit (GRU) algorithms. Last but not least, the performance indices were calculated. when training and testing the model with power system load data, it was noted that the root mean square error (RMSE) of Seq2seq was decreased by 6.61%, 16.95%, and 7.80% compared with RNN, LSTM, and GRU, respectively. In addition, a supplementary case study was carried out using data for a small power system considering different weather conditions and user behaviors in order to confirm the applicability and stability of the proposed model. The Seq2seq model for short-term load prediction can be reported to demonstrate superiority in all areas, exhibiting better prediction and stable performance.

scholarly journals Comparison of 3DVar and EnSRF Data Assimilation Using Radar Observations for the Analysis and Prediction of an MCS

2018 ◽  
Vol 2018 ◽  
pp. 1-18 ◽  
Author(s):  
Shibo Gao ◽  
Jinzhong Min
Keyword(s):  
Data Assimilation ◽  
Three Dimensional ◽  
Mesoscale Convective System ◽  
Final Analysis ◽  
Radar Data ◽  
Radar Reflectivity ◽  
Precipitation Forecast ◽  
Convective System ◽  
Wind Fields ◽  
Radar Observations

Using radar observations, the performances of the ensemble square root filter (EnSRF) and an indirect three-dimensional variational (3DVar) data assimilation method were compared for a mesoscale convective system (MCS) that occurred in the Front Range of the Rocky Mountains, Colorado (USA). The results showed that the root mean square innovations (RMSIs) of EnSRF were lower than 3DVar for radar reflectivity and radial velocity and that the spread of EnSRF was generally consistent with its RMSIs. EnSRF substantially improved the analysis of the MCS compared with an experiment without radar data assimilation, and it produced a slight but noticeable improvement over 3DVar in terms of both coverage and intensity. Forecast results initiated from the final analysis revealed that EnSRF generally produced the best prediction of the MCS, with improved quantitative reflectivity and precipitation forecast skills. EnSRF also demonstrated better performance than 3DVar in the prediction of neighborhood probability for reflectivity at thresholds of 20 and 35 dBZ, which better matched the observed radar reflectivity in terms of both shape and extension. Additionally, the humidity, temperature, and wind fields were also improved by EnSRF; the largest error reduction was found in the water vapor field near the surface and at upper levels.

scholarly journals A study on WRF radar data assimilation for hydrological rainfall prediction

2013 ◽  
Vol 17 (8) ◽  
pp. 3095-3110 ◽  
Author(s):  
J. Liu ◽  
M. Bray ◽  
D. Han
Keyword(s):  
Data Assimilation ◽  
Real Time ◽  
Meteorological Data ◽  
Weather Radar ◽  
Radar Data ◽  
Rain Gauge ◽  
Radar Reflectivity ◽  
Storm Event ◽  
Hydrological Applications ◽  
Radar Data Assimilation

Abstract. Mesoscale numerical weather prediction (NWP) models are gaining more attention in providing high-resolution rainfall forecasts at the catchment scale for real-time flood forecasting. The model accuracy is however negatively affected by the "spin-up" effect and errors in the initial and lateral boundary conditions. Synoptic studies in the meteorological area have shown that the assimilation of operational observations, especially the weather radar data, can improve the reliability of the rainfall forecasts from the NWP models. This study aims at investigating the potential of radar data assimilation in improving the NWP rainfall forecasts that have direct benefits for hydrological applications. The Weather Research and Forecasting (WRF) model is adopted to generate 10 km rainfall forecasts for a 24 h storm event in the Brue catchment (135.2 km2) located in southwest England. Radar reflectivity from the lowest scan elevation of a C-band weather radar is assimilated by using the three-dimensional variational (3D-Var) data-assimilation technique. Considering the unsatisfactory quality of radar data compared to the rain gauge observations, the radar data are assimilated in both the original form and an improved form based on a real-time correction ratio developed according to the rain gauge observations. Traditional meteorological observations including the surface and upper-air measurements of pressure, temperature, humidity and wind speed are also assimilated as a bench mark to better evaluate and test the potential of radar data assimilation. Four modes of data assimilation are thus carried out on different types/combinations of observations: (1) traditional meteorological data; (2) radar reflectivity; (3) corrected radar reflectivity; (4) a combination of the original reflectivity and meteorological data; and (5) a combination of the corrected reflectivity and meteorological data. The WRF rainfall forecasts before and after different modes of data assimilation are evaluated by examining the rainfall temporal variations and total amounts which have direct impacts on rainfall–runoff transformation in hydrological applications. It is found that by solely assimilating radar data, the improvement of rainfall forecasts are not as obvious as assimilating meteorological data; whereas the positive effect of radar data can be seen when combined with the traditional meteorological data, which leads to the best rainfall forecasts among the five modes. To further improve the effect of radar data assimilation, limitations of the radar correction ratio developed in this study are discussed and suggestions are made on more efficient utilisation of radar data in NWP data assimilation.

scholarly journals The impact of data-complexity and team characteristics on performance in the classification model

2022 ◽  
Vol 9 (1) ◽  
pp. 0-0
Keyword(s):  
Machine Learning ◽  
Binary Classification ◽  
Class Imbalance ◽  
Predictive Ability ◽  
Predictive Performance ◽  
Classification Model ◽  
Data Complexity ◽  
High Performing ◽  
Machine Learning Model ◽  
The Impact

This article investigates the impact of data-complexity and team-specific characteristics on machine learning competition scores. Data from five real-world binary classification competitions hosted on Kaggle.com were analyzed. The data-complexity characteristics were measured in four aspects including standard measures, sparsity measures, class imbalance measures, and feature-based measures. The results showed that the higher the level of the data-complexity characteristics was, the lower the predictive ability of the machine learning model was as well. Our empirical evidence revealed that the imbalance ratio of the target variable was the most important factor and exhibited a nonlinear relationship with the model’s predictive abilities. The imbalance ratio adversely affected the predictive performance when it reached a certain level. However, mixed results were found for the impact of team-specific characteristics measured by team size, team expertise, and the number of submissions on team performance. For high-performing teams, these factors had no impact on team score.

scholarly journals Improving Short-Term Rainfall Forecasts by Assimilating Weather Radar Reflectivity Using Additive Ensemble Perturbations

2018 ◽  
Vol 123 (17) ◽  
pp. 9047-9062 ◽  
Author(s):  
S. Yokota ◽  
H. Seko ◽  
M. Kunii ◽  
H. Yamauchi ◽  
E. Sato
Keyword(s):  
Weather Radar ◽  
Radar Reflectivity ◽  
Short Term

scholarly journals Comparison of the GPM IMERG Final Precipitation Product to RADOLAN Weather Radar Data over the Topographically and Climatically Diverse Germany

2018 ◽  
Vol 10 (12) ◽  
pp. 2029 ◽  
Author(s):  
Thomas Ramsauer ◽  
Thomas Weiß ◽  
Philip Marzahn
Keyword(s):  
Spatial Variability ◽  
Winter Season ◽  
Weather Radar ◽  
Radar Data ◽  
Probability Of Detection ◽  
Statistical Measures ◽  
Weather Events ◽  
Precipitation Measurement ◽  
Spatio Temporal ◽  
Global Precipitation

Precipitation measurements provide crucial information for hydrometeorological applications. In regions where typical precipitation measurement gauges are sparse, gridded products aim to provide alternative data sources. This study examines the performance of NASA’s Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement Mission (IMERG, GPM) satellite precipitation dataset in capturing the spatio-temporal variability of weather events compared to the German weather radar dataset RADOLAN RW. Besides quantity, also timing of rainfall is of very high importance when modeling or monitoring the hydrologic cycle. Therefore, detection metrics are evaluated along with standard statistical measures to test both datasets. Using indices like “probability of detection” allows a binary evaluation showing the basic categorical accordance of the radar and satellite data. Furthermore, a pixel-by-pixel comparison is performed to assess the ability to represent the spatial variability of rainfall and precipitation quantity. All calculations are additionally carried out for seasonal subsets of the data to assess potentially different behavior due to differences in precipitation schemes. The results indicate significant differences between the datasets. Overall, GPM IMERG overestimates the quantity of precipitation compared to RADOLAN, especially in the winter season. Moreover, shortcomings in detection performance arise in this season with significant erroneously-detected, yet also missed precipitation events compared to the weather radar data. Additionally, along secondary mountain ranges and the Alps, topographically-induced precipitation is not represented in GPM data, which generally shows a lack of spatial variability in rainfall and snowfall estimates due to lower resolution.

scholarly journals Operational Implementation of the 1D+3D-Var Assimilation Method of Radar Reflectivity Data in the AROME Model

2014 ◽  
Vol 142 (5) ◽  
pp. 1852-1873 ◽  
Author(s):  
Eric Wattrelot ◽  
Olivier Caumont ◽  
Jean-François Mahfouf
Keyword(s):  
Relative Humidity ◽  
Data Assimilation ◽  
Three Dimensional ◽  
Radar Reflectivity ◽  
Precipitation Forecast ◽  
Limited Area ◽  
Short Term ◽  
Background Error ◽  
Reflectivity Data

AbstractThis paper presents results from radar reflectivity data assimilation experiments with the nonhydrostatic limited-area model Application of Research to Operations at Mesoscale (AROME) in an operational context. A one-dimensional (1D) Bayesian retrieval of relative humidity profiles followed by a three-dimensional variational data assimilation (3D-Var) technique is adopted. Several preprocessing procedures of raw reflectivity data are presented and the use of the nonrainy signal in the assimilation is widely discussed and illustrated. This two-step methodology allows the authors to build up a screening procedure that takes into account the evaluation of the results from the 1D Bayesian retrieval. In particular, the 1D retrieval is checked by comparing a pseudoanalyzed reflectivity to the observed reflectivity. Additionally, a physical consistency between the reflectivity innovations and the 1D relative humidity increments is imposed before assimilating relative humidity pseudo-observations with other observations. This allows the authors to counteract the difficulty of the current 3D-Var system to correct strong differences between model and observed clouds from the crude specification of background-error covariances. Assimilation experiments of radar reflectivity data in a preoperational configuration are first performed over a 1-month period. Positive impacts on short-term precipitation forecast scores are systematically found. The evaluation shows improvements on the analysis and also on objective conventional forecast scores, in particular for the model wind field up to 12 h. A case study for a specific precipitating system demonstrates the capacity of the method for improving significantly short-term forecasts of organized convection.

scholarly journals Improving QPF by blending techniques at the Meteorological Service of Catalonia

2010 ◽  
Vol 10 (7) ◽  
pp. 1443-1455 ◽  
Author(s):  
A. Atencia ◽  
T. Rigo ◽  
A. Sairouni ◽  
J. Moré ◽  
J. Bech ◽  
...  
Keyword(s):  
Lead Time ◽  
Weather Prediction ◽  
Radar Reflectivity ◽  
Precipitation Forecast ◽  
Short Term ◽  
Convective Structures ◽  
Prediction Horizon ◽  
Quantitative Precipitation Forecast ◽  
Correction Technique ◽  
Nwp Model

Abstract. The current operational very short-term and short-term quantitative precipitation forecast (QPF) at the Meteorological Service of Catalonia (SMC) is made by three different methodologies: Advection of the radar reflectivity field (ADV), Identification, tracking and forecasting of convective structures (CST) and numerical weather prediction (NWP) models using observational data assimilation (radar, satellite, etc.). These precipitation forecasts have different characteristics, lead time and spatial resolutions. The objective of this study is to combine these methods in order to obtain a single and optimized QPF at each lead time. This combination (blending) of the radar forecast (ADV and CST) and precipitation forecast from NWP model is carried out by means of different methodologies according to the prediction horizon. Firstly, in order to take advantage of the rainfall location and intensity from radar observations, a phase correction technique is applied to the NWP output to derive an additional corrected forecast (MCO). To select the best precipitation estimation in the first and second hour (t+1 h and t+2 h), the information from radar advection (ADV) and the corrected outputs from the model (MCO) are mixed by using different weights, which vary dynamically, according to indexes that quantify the quality of these predictions. This procedure has the ability to integrate the skill of rainfall location and patterns that are given by the advection of radar reflectivity field with the capacity of generating new precipitation areas from the NWP models. From the third hour (t+3 h), as radar-based forecasting has generally low skills, only the quantitative precipitation forecast from model is used. This blending of different sources of prediction is verified for different types of episodes (convective, moderately convective and stratiform) to obtain a robust methodology for implementing it in an operational and dynamic way.

Sign in / Sign up

Export Citation Format

Share Document