Investigating the pilot point ensemble Kalman filter for geostatistical inversion and data assimilation

Parameter estimation has a high importance in the geosciences. The ensemble Kalman filter (EnKF) allows parameter for large, time-dependent systems. For large systems, EnKF is applied using small ensembles, which may lead to spurious correlations and, ultimately, divergence. We present thorough evaluation of pilot point (PP-EnKF), variant estimation. In this evaluation, we explicitly state update equations PP-EnKF, discuss differences equation compared similar methods, and perform an extensive performance comparison. PP-EnKF tested seven other methods two model setups, tracer setup well setup. both performs well, ranking better than classical EnKF. setup, ranks third out eight methods. At same time, yields estimates variance that are close results from very large-ensemble reference, suggesting it not affected by underestimation variance. comparison variances, first Additionally, size 50, correlation structures significantly closer reference EnKF, indication method's skill suppress sizes.