Short-range ensemble forecasts from the Storm and Mesoscale Ensemble Experiment (SAMEX) are examined to explore the importance of model diversity in short-range ensemble forecasting systems. Two basic techniques from multivariate data analysis are used: cluster analysis and principal component analysis. This 25-member ensemble is constructed of 36-h forecasts from four different numerical weather prediction models, including the Eta Model, the Regional Spectral Model (RSM), the Advanced Regional Prediction System (ARPS), and the Pennsylvania State University–National Center for Atmospheric Research fifth-generation Mesoscale Model (MM5). The Eta Model and RSM forecasts are initialized using the breeding of growing modes approach, the ARPS model forecasts are initialized using a scaled lagged average forecasting approach, and the MM5 forecasts are initialized using a random coherent structures approach. The MM5 forecasts also include different model physical parameterization schemes, allowing us to examine the role of intramodel physics differences in the ensemble forecasting process. Cluster analyses of the 3-h accumulated precipitation, mean sea level pressure, convective available potential energy, 500-hPa geopotential height, and 250-hPa wind speed forecasts started at 0000 UTC 29 May 1998 indicate that the forecasts cluster largely by model, with few intermodel clusters found. This clustering occurs within the first few hours of the forecast and persists throughout the entire forecast period, even though the perturbed initial conditions from some of the models are very similar. This result further highlights the important role played by model physics in determining the resulting forecasts and the need for model diversity in short-range ensemble forecasting systems.

The least squares fit of observations with known error covariance to a strong‐constraint dynamical model has been developed through use of the time evolution of sensitivity functions—the derivatives of model output with respect to the elements of control (initial conditions, boundary conditions, and physical/empirical parameters). Model error is assumed to stem from incorrect specification of the control elements. The optimal corrections to control are found through solution to an inverse problem. Duality between this method and the standard 4D‐Var assimilation using adjoint equations has been proved. The paper ends with an illustrative example based on a simplified version of turbulent heat transfer at the sea/air interface.

Machine learning (ML) algorithms, such as principal component analysis, independent component analysis, self-organizing maps, and artificial neural networks, have been used by geoscientists to not only accelerate the interpretation of their data, but also to provide a more quantitative estimate of the likelihood that any voxel belongs to a given facies. Identifying the best combination of attributes needed to perform either supervised or unsupervised ML tasks continues to be the most-asked question by interpreters. In the past decades, stepwise regression and genetic algorithms have been used together with supervised learning algorithms to select the best number and combination of attributes. For reasons of computational efficiency, these techniques do not test all of the seismic attribute combinations, potentially leading to a suboptimal classification. In this study, we have developed an exhaustive probabilistic neural network (PNN) algorithm that exploits the PNN’s capacity in exploring nonlinear relationships to obtain the optimal attribute subset that best differentiates target seismic facies of interest. We determine the efficacy of our proposed workflow in differentiating salt from nonsalt seismic facies in a Eugene Island seismic survey, offshore Louisiana. We find that from seven input candidate attributes, the exhaustive PNN is capable of removing irrelevant attributes by selecting a smaller subset of four seismic attributes. The enhanced classification using fewer attributes also reduces the computational cost. We then use the resulting facies probability volumes to construct the 3D distribution of the salt diapir geobodies embedded in a stratigraphic matrix.