Inferring precipitation from atmospheric general circulation model variables

2021 ◽  
Author(s):  
Philipp Hess ◽  
Niklas Boers
Keyword(s):  
General Circulation Model ◽  
General Circulation ◽  
Weather Forecasting ◽  
Extreme Values ◽  
Weather Prediction ◽  
Hybrid Approach ◽  
Circulation Model ◽  
Physical Models ◽  
Data Driven ◽  
Medium Range

<p>The accurate prediction of precipitation, in particular of extremes, remains a challenge for numerical weather prediction (NWP) models. A large source of error are subgrid-scale parameterizations of processes that play a crucial role in the complex, multi-scale dynamics of precipitation, but are not explicitly resolved in the model formulation. Recent progress in purely data-driven deep learning for regional precipitation nowcasting [1] and global medium-range forecasting [2] tasks has shown competitive results to traditional NWP models.<br>Here we follow a hybrid approach, in which explicitly resolved atmospheric variables are forecast in time by a general circulation model (GCM) ensemble and then mapped to precipitation using a deep convolutional autoencoder. A frequency-based weighting of the loss function is introduced to improve the learning with regard to extreme values.<br>Our method is validated against a state-of-the-art GCM ensemble using three-hourly high resolution data. The results show an improved representation of extreme precipitation frequencies, as well as comparable error and correlation statistics.<br>   </p><p>[1] C.K. Sønderby et al. "MetNet: A Neural Weather Model for Precipitation Forecasting." arXiv preprint arXiv:2003.12140 (2020). <br>[2] S. Rasp and N. Thuerey "Purely data-driven medium-range weather forecasting achieves comparable skill to physical models at similar resolution." arXiv preprint arXiv:2008.08626 (2020).</p>

scholarly journals Development of the GEOS-5 atmospheric general circulation model: evolution from MERRA to MERRA2

2014 ◽  
Vol 7 (6) ◽  
pp. 7575-7617 ◽  
Author(s):  
A. Molod ◽  
L. Takacs ◽  
M. Suarez ◽  
J. Bacmeister
Keyword(s):  
General Circulation Model ◽  
General Circulation ◽  
Atmospheric General Circulation Model ◽  
Weather Prediction ◽  
Circulation Model ◽  
Wave Drag ◽  
Atmospheric General Circulation ◽  
Wide Range ◽  
Simulated Climate ◽  
The Impact

Abstract. The Modern-Era Retrospective Analysis for Research and Applications-2 (MERRA2) version of the GEOS-5 Atmospheric General Circulation Model (AGCM) is currently in use in the NASA Global Modeling and Assimilation Office (GMAO) at a wide range of resolutions for a variety of applications. Details of the changes in parameterizations subsequent to the version in the original MERRA reanalysis are presented here. Results of a series of atmosphere-only sensitivity studies are shown to demonstrate changes in simulated climate associated with specific changes in physical parameterizations, and the impact of the newly implemented resolution-aware behavior on simulations at different resolutions is demonstrated. The GEOS-5 AGCM presented here is the model used as part of the GMAO's MERRA2 reanalysis, the global mesoscale "nature run", the real-time numerical weather prediction system, and for atmosphere-only, coupled ocean–atmosphere and coupled atmosphere–chemistry simulations. The seasonal mean climate of the MERRA2 version of the GEOS-5 AGCM represents a substantial improvement over the simulated climate of the MERRA version at all resolutions and for all applications. Fundamental improvements in simulated climate are associated with the increased re-evaporation of frozen precipitation and cloud condensate, resulting in a wetter atmosphere. Improvements in simulated climate are also shown to be attributable to changes in the background gravity wave drag, and to upgrades in the relationship between the ocean surface stress and the ocean roughness. The series of "resolution aware" parameters related to the moist physics were shown to result in improvements at higher resolutions, and result in AGCM simulations that exhibit seamless behavior across different resolutions and applications.

scholarly journals Statistical interpretation of general circulation model: A prospect for automation of medium range local weather forecast in India

MAUSAM ◽  
2021 ◽  
Vol 47 (3) ◽  
pp. 229-236
Author(s):  
ASHOK KUMAR ◽  
PARVINDER MAINI
Keyword(s):  
General Circulation ◽  
Weather Forecasting ◽  
Monsoon Season ◽  
Circulation Model ◽  
Weather Forecast ◽  
General Circulation Models ◽  
Statistical Interpretation ◽  
Medium Range Weather Forecast ◽  
Medium Range ◽  
Local Weather

The General Circulation Models (GCM), though able to provide reasonably good medium range weather forecast. have comparatively less skill in forecasting location-specific weather. This is mainly due to the poor representation of 16cal topography and other features in these models. Statistical interpretation (SI) of GCM is very essential in order to improve the location-specific medium range local weather forecast. An attempt has been made at the National Centre for Medium Range Weather Forecasting (NCMRWF), New Delhi to do this type of objective forecasting. Hence location-specific SI models are developed and a bias free forecast is obtained. One of the techniques for accomplishing this, is the Perfect Prog. Method (PPM). PPM models for precipitation (quantitative, probability, yes/no) and maximum minimum temperature are developed for monsoon season (June to August) for 10 stations in lndia. These PPM models and the output from the GCM (R-40) operational at NCMRWF, are then used to obtain the SI forecast. An indirect method based upon SI forecast and observed values of previous one or two seasons, for getting bias free forecast is explained. A comparative study of skill of bias free SI and final forecast, with the observed, issued from NCMRWF to 10 Agromet Field Units (AMFU) during monsoon season 1993, has indicated that automation of medium range local weather forecast can be achieved with the help of SI forecast.

Developing a data-driven ocean model

2020 ◽  
Author(s):  
Rachel Furner ◽  
Peter Haynes ◽  
Dan Jones ◽  
Dave Munday ◽  
Brooks Paige ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
General Circulation Model ◽  
General Circulation ◽  
Climate Models ◽  
Circulation Model ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Geophysical Research ◽  
Weather And Climate

<p>The recent boom in machine learning and data science has led to a number of new opportunities in the environmental sciences. In particular, climate models represent the best tools we have to predict, understand and potentially mitigate climate change, however these process-based models are incredibly complex and require huge amounts of high-performance computing resources. Machine learning offers opportunities to greatly improve the computational efficiency of these models.</p><p>Here we discuss our recent efforts to reduce the computational cost associated with running a process-based model of the physical ocean by developing an analogous data-driven model. We train statistical and machine learning algorithms using the outputs from a highly idealised sector configuration of general circulation model (MITgcm). Our aim is to develop an algorithm which is able to predict the future state of the general circulation model to a similar level of accuracy in a more computationally efficient manner.</p><p>We first develop a linear regression model to investigate the sensitivity of data-driven approaches to various inputs, e.g. temperature on different spatial and temporal scales, and meta-variables such as location information. Following this, we develop a neural network model to replicate the general circulation model, as in the work of Dueben and Bauer 2018, and Scher 2018.</p><p>We present a discussion on the sensitivity of data-driven models and preliminary results from the neural network based model.</p><p> </p><p><em>Dueben, P. D., & Bauer, P. (2018). Challenges and design choices for global weather and climate models based on machine learning. Geoscientific Model Development, 11(10), 3999-4009.</em></p><p><em>Scher, S. (2018). Toward Data‐Driven Weather and Climate Forecasting: Approximating a Simple General Circulation Model With Deep Learning. Geophysical Research Letters, 45(22), 12-616.</em></p>

Medium Range (10 Day) Forecasts with the UCLA General Circulation Model.

1986 ◽  
Author(s):  
C. R. Mechoso ◽  
M. J. Suarez ◽  
K. Yamazaki ◽  
A. Kitoh ◽  
J. A. Spahr
Keyword(s):  
General Circulation Model ◽  
General Circulation ◽  
Circulation Model ◽  
Medium Range

scholarly journals Development of the GEOS-5 atmospheric general circulation model: evolution from MERRA to MERRA2

2015 ◽  
Vol 8 (5) ◽  
pp. 1339-1356 ◽  
Author(s):  
A. Molod ◽  
L. Takacs ◽  
M. Suarez ◽  
J. Bacmeister
Keyword(s):  
General Circulation Model ◽  
General Circulation ◽  
Atmospheric General Circulation Model ◽  
Weather Prediction ◽  
Circulation Model ◽  
Wave Drag ◽  
Atmospheric General Circulation ◽  
Wide Range ◽  
Simulated Climate ◽  
The Impact

Abstract. The Modern-Era Retrospective Analysis for Research and Applications-2 (MERRA2) version of the Goddard Earth Observing System-5 (GEOS-5) atmospheric general circulation model (AGCM) is currently in use in the NASA Global Modeling and Assimilation Office (GMAO) at a wide range of resolutions for a variety of applications. Details of the changes in parameterizations subsequent to the version in the original MERRA reanalysis are presented here. Results of a series of atmosphere-only sensitivity studies are shown to demonstrate changes in simulated climate associated with specific changes in physical parameterizations, and the impact of the newly implemented resolution-aware behavior on simulations at different resolutions is demonstrated. The GEOS-5 AGCM presented here is the model used as part of the GMAO MERRA2 reanalysis, global mesoscale simulations at 10 km resolution through 1.5 km resolution, the real-time numerical weather prediction system, and for atmosphere-only, coupled ocean-atmosphere and coupled atmosphere-chemistry simulations. The seasonal mean climate of the MERRA2 version of the GEOS-5 AGCM represents a substantial improvement over the simulated climate of the MERRA version at all resolutions and for all applications. Fundamental improvements in simulated climate are associated with the increased re-evaporation of frozen precipitation and cloud condensate, resulting in a wetter atmosphere. Improvements in simulated climate are also shown to be attributable to changes in the background gravity wave drag, and to upgrades in the relationship between the ocean surface stress and the ocean roughness. The series of resolution-aware parameters related to the moist physics was shown to result in improvements at higher resolutions and result in AGCM simulations that exhibit seamless behavior across different resolutions and applications.

Sign in / Sign up

Export Citation Format

Share Document