Seminarraum SR 2.058 im Kollegiengebäude Mathemaik

Ensemble weather predictions from global forecast systems require

statistical postprocessing in order to remove systematic errors and to

obtain reliable probabilistic forecasts. Many traditional postprocessing

methods are based on statistical models that make parametric assumptions

about the forecast distribution and/or the relationship (e.g. linearity)

between predictors and predictands. A number of recent papers, however,

have demonstrated for ensemble temperature and wind speed forecasts that

more accurate predictions can be obtained using artificial neural

networks (ANNs) for statistical post-processing. Here, we propose a

statistical post-processing approach for precipitation forecasts that is

built around an artificial neural network (ANN) and addresses the

statistical peculiarities of precipitation as well as the challenges

that come with the low signal-to-noise ratio encountered at subseasonal

forecast lead times.

Our basic approach uses only precipitation forecasts from a numerical

weather prediction model and geographic information as predictors. In a

subseasonal forecast context, however, predictors like geopotential

height at 500 hPa (Z500) and total column water (TCW) may be more

useful since the NWP models may have better skill in predicting

large-scale weather patterns than surface weather variables at a

specific location. We therefore propose an extension of our basic

framework that uses a convolutional neural network to process images of

Z500 and TCW over a larger domain and uses them as predictors for

localized precipitation amounts.

DESCRIPTION: Using artificial neural networks for generating probabilistic subseasonal precipitation forecasts over California.\nSeminarraum SR 2.058 im Kollegiengebäude Mathemaik \n\n

Ensemble weather predictions from global forecast systems require

\nstatistical postprocessing in order to remove systematic errors and to

\nobtain reliable probabilistic forecasts. Many traditional postprocessing

\nmethods are based on statistical models that make parametric assumptions

\nabout the forecast distribution and/or the relationship (e.g. linearity)

\nbetween predictors and predictands. A number of recent papers, however,

\nhave demonstrated for ensemble temperature and wind speed forecasts that

\nmore accurate predictions can be obtained using artificial neural

\nnetworks (ANNs) for statistical post-processing. Here, we propose a

\nstatistical post-processing approach for precipitation forecasts that is

\nbuilt around an artificial neural network (ANN) and addresses the

\nstatistical peculiarities of precipitation as well as the challenges

\nthat come with the low signal-to-noise ratio encountered at subseasonal

\nforecast lead times.

\n

\nOur basic approach uses only precipitation forecasts from a numerical

\nweather prediction model and geographic information as predictors. In a

\nsubseasonal forecast context, however, predictors like geopotential

\nheight at 500 hPa (Z500) and total column water (TCW) may be more

\nuseful since the NWP models may have better skill in predicting

\nlarge-scale weather patterns than surface weather variables at a

\nspecific location. We therefore propose an extension of our basic

\nframework that uses a convolutional neural network to process images of

\nZ500 and TCW over a larger domain and uses them as predictors for

\nlocalized precipitation amounts.

\n

\n