scholarly journals A Model Output Deep Learning Method for Grid Temperature Forecasts in Tianjin Area

2020 ◽  
Vol 10 (17) ◽  
pp. 5808 ◽  
Author(s):  
Keran Chen ◽  
Ping Wang ◽  
Xiaojun Yang ◽  
Nan Zhang ◽  
Di Wang
Keyword(s):  
Temperature Field ◽  
Deep Learning ◽  
Numerical Weather Prediction ◽  
Weather Forecasting ◽  
Weather Prediction ◽  
Physical Models ◽  
Model Output ◽  
Post Processing ◽  
Numerical Weather ◽  
Convolution Model

In weather forecasting, numerical weather prediction (NWP) that is based on physical models requires proper post-processing before it can be applied to actual operations. Therefore, research on intelligent post-processing algorithms has always been an important topic in this field. This paper proposes a model output deep learning (MODL) method for post-processing, which can improve the forecast effect of numerical weather prediction. MODL is an end-to-end post-processing method based on deep convolutional neural network, which directly learns the mapping relationship between the forecast fields output by numerical model and the observation temperature field in order to obtain more accurate temperature forecasts. MODL modifies the existing deep convolution model according to the post-processing problem’s characteristics, thereby improving the performance of the weather forecast. This paper uses The International Grand Global Ensemble (TIGGE) dataset from European Centre for Medium-Range Weather Forecasts (ECMWF) and the observed air temperature of 2 m obtained from Tianjin meteorological station in order to test the post-processing performance of MODL. The MODL method applied to temperature in post-processing is compared with the ECMWF forecast, Model Output Statistics (MOS) methods, and Model Output Machine Learning (MOML) methods. The Root Mean Square Error (RMSE) of the temperature field predicted by MODL and the observed temperature field is smaller than the other models and the accuracy of the temperature difference of 2 °C (Acc) is higher, especially where the prediction time is in the first three days. The lightweight nature of MODL also makes it suitable for most operations.

scholarly journals Post-processing numerical weather prediction ensembles for probabilistic solar irradiance forecasting

2021 ◽  
Author(s):  
Sebastian Lerch ◽  
Benedikt Schulz ◽  
Mehrez El Ayari ◽  
Sándor Baran
Keyword(s):  
Solar Energy ◽  
Numerical Weather Prediction ◽  
Solar Irradiance ◽  
Weather Prediction ◽  
Physical Models ◽  
Lead Times ◽  
Post Processing ◽  
Energy Forecasting ◽  
Numerical Weather ◽  
Geographical Regions

<p>In order to enable the transition towards renewable energy sources, probabilistic energy forecasting is of critical importance for incorporating volatile power sources such as solar energy into the electrical grid. Solar energy forecasting methods often aim to provide probabilistic predictions of solar irradiance. In particular, many hybrid approaches combine physical information from numerical weather prediction models with statistical methods. Even though the physical models can provide useful information at intra-day and day-ahead forecast horizons, ensemble weather forecasts from multiple model runs are often not calibrated and show systematic biases. We propose a post-processing model for ensemble weather predictions of solar irradiance at temporal resolutions between 30 minutes and 6 hours. The proposed models provide probabilistic forecasts in the form of a censored logistic probability distribution for lead times up to 5 days and are evaluated in two case studies covering distinct physical models, geographical regions, temporal resolutions, and types of solar irradiance. We find that post-processing consistently and significantly improves the forecast performance of the ensemble predictions for lead times up to at least 48 hours and is well able to correct the systematic lack of calibration.</p>

Assimilation of Satellite Cloud and Precipitation Observations in Numerical Weather Prediction Models: Introduction to the JAS Special Collection

2007 ◽  
Vol 64 (11) ◽  
pp. 3737-3741 ◽  
Author(s):  
Ronald M. Errico ◽  
George Ohring ◽  
Fuzhong Weng ◽  
Peter Bauer ◽  
Brad Ferrier ◽  
...  
Keyword(s):  
Numerical Weather Prediction ◽  
Prediction Models ◽  
Weather Forecasting ◽  
International Workshop ◽  
Weather Prediction ◽  
Satellite Observations ◽  
Atmospheric Sciences ◽  
Numerical Weather ◽  
Special Collection ◽  
Numerical Weather Prediction Models

Abstract To date, the assimilation of satellite measurements in numerical weather prediction (NWP) models has focused on the clear atmosphere. But satellite observations in the visible, infrared, and microwave provide a great deal of information on clouds and precipitation. This special collection describes how to use this information to initialize clouds and precipitation in models. Since clouds and precipitation often occur in sensitive regions for forecast impacts, such improvements are likely necessary for continuing to acquire significant gains in weather forecasting. This special collection of the Journal of the Atmospheric Sciences is devoted to articles based on papers presented at the International Workshop on Assimilation of Satellite Cloud and Precipitation Observations in Numerical Weather Prediction Models, in Lansdowne, Virginia, in May 2005. This introduction summarizes the findings of the workshop. The special collection includes review articles on satellite observations of clouds and precipitation (Stephens and Kummerow), parameterizations of clouds and precipitation in NWP models (Lopez), radiative transfer in cloudy/precipitating atmospheres (Weng), and assimilation of cloud and precipitation observations (Errico et al.), as well as research papers on these topics.

scholarly journals Evaluation of IWV from the numerical weather prediction WRF model with PPP GNSS processing for Bulgaria

2016 ◽  
Author(s):  
Tzvetan Simeonov ◽  
Dmitry Sidorov ◽  
Felix Norman Teferle ◽  
Georgi Milev ◽  
Guergana Guerova
Keyword(s):  
Surface Pressure ◽  
Numerical Weather Prediction ◽  
Weather Forecasting ◽  
Weather Prediction ◽  
Wrf Model ◽  
Mean Difference ◽  
Global Navigation Satellite Systems ◽  
Numerical Weather ◽  
Precipitation Efficiency ◽  
Good Agreement

Abstract. Global Navigation Satellite Systems (GNSS) meteorology is an established operational service providing hourly updated GNSS tropospheric products to the National Meteorologic Services (NMS) in Europe. In the last decade through the ground-based GNSS network densification and new processing strategies like Precise Point Positioning (PPP) it has become possible to obtain sub-hourly tropospheric products for monitoring severe weather events. In this work one year (January–December 2013) of sub-hourly GNSS tropospheric products (Zenith Total Delay) are computed using the PPP strategy for seven stations in Bulgaria. In order to take advantage of the sub-hourly GNSS data to derive Integrated Water Vapour (IWV) surface pressure and temperature with similar temporal resolution is required. As the surface observations are on 3 hourly basis the first step is to compare the surface pressure and temperature from numerical weather prediction model Weather Forecasting and Research (WRF) with observations at three synoptic stations in Bulgaria. The mean difference between the two data-sets for 1) surface pressure is less than 0.5 hPa and the correlation is over 0.989, 2) temperature the largest mean difference is 1.1 °C and the correlation coefficient is over 0.957 and 3) IWV mean difference is in range 0.1–1.1 mm. The evaluation of WRF on annual bases shows IWV underestimation between 0.5 and 1.5 mm at five stations and overestimation at Varna and Rozhen. Varna and Rozhen have also much smaller correlation 0.9 and 0.76. The study of the monthly IWV variation shows that at those locations the GNSS IWV has unexpected drop in April and March respectively. The reason for this drop is likely problems with station raw data. At the remaining 5 stations a very good agreement between GNSS and WRF is observed with high correlation during the cold part of 2013 i.e. March, October and December (0.95) and low correlation during the warm part of 2013 i.e. April to August (below 0.9). The diurnal cycle of the WRF model shows a dry bias in the range of 0.5-1.5 mm. Between 00 and 01 UTC the GNSS IWV tends to be underestimate IWV which is likely due to the processing window used. The precipitation efficiency from GNSS and WRF show very good agreement on monthly bases with a maximum in May-June and minimum in August–September. The annual precipitation efficiency in 2013 at Lovech and Burgas is about 6 %.

scholarly journals EuLerian Identification of ascending Air Streams (ELIAS 2.0) in Numerical Weather Prediction and Climate Models. Part I: Development of deep learning model

2021 ◽  
Author(s):  
Julian Francesco Quinting ◽  
Christian M. Grams
Keyword(s):  
Logistic Regression ◽  
Deep Learning ◽  
Numerical Weather Prediction ◽  
Climate Models ◽  
Weather Prediction ◽  
Data Sets ◽  
Forecast Uncertainty ◽  
Numerical Weather ◽  
Process Oriented ◽  
Air Streams

Abstract. Physical processes on the synoptic scale are important modulators of the large-scale extratropical circulation. In particular, rapidly ascending air streams in extratropical cyclones, so-called warm conveyor belts (WCBs), modulate the upper-tropospheric Rossby wave pattern and are sources and magnifiers of forecast uncertainty. Thus, from a process-oriented perspective, numerical weather prediction (NWP) and climate models should adequately represent WCBs. The identification of WCBs usually involves Lagrangian air parcel trajectories that ascend from the lower to the upper troposphere within two days. This requires numerical data with high spatial and temporal resolution which is often not available from standard output and requires expensive computations. This study introduces a novel framework that aims to predict the footprints of the WCB inflow, ascent, and outflow stages over the Northern Hemisphere from instantaneous gridded fields using convolutional neural networks (CNNs). With its comparably low computational costs and relying on standard model output alone the new diagnostic enables the systematic investigation of WCBs in large data sets such as ensemble reforecast or climate model projections which are mostly not suited for trajectory calculations. Building on the insights from a logistic regression approach of a previous study, the CNNs are trained using a combination of meteorological parameters as predictors and trajectory-based WCB footprints as predictands. Validation of the networks against the trajectory-based data set confirms that the CNN models reliably replicate the climatological frequency of WCBs as well as their footprints at instantaneous time steps. The CNN models significantly outperform previously developed logistic regression models. Including time-lagged information on the occurrence of WCB ascent as a predictor for the inflow and outflow stages further improves the models' skill considerably. A companion study demonstrates versatile applications of the CNNs in different data sets including the verification of WCBs in ensemble forecasts. Overall, the diagnostic demonstrates how deep learning methods may be used to investigate the representation of weather systems and of their related processes in NWP and climate models in order to shed light on forecast uncertainty and systematic biases from a process-oriented perspective.

Sign in / Sign up

Export Citation Format

Share Document