Equally likely inverse solutions to a groundwater flow problem including pattern information from remote sensing images

摘要: [1] Groundwater flow modeling for large areas in arid and semiarid regions, like the Chobe region Botswana, suffers from a severe lack of data. This study addresses usefulness remote sensing (RS) images to constrain recharge rate estimates region. The derived METEOSAT NOAA advanced very high resolution radar (AVHRR) are correlated with values estimated chloride measurements used jointly generation multiple, equally likely realizations colocated cosimulation algorithm. algorithm is suited generate stochastic parameter that includes information covariable given on regular, dense grid as RS information. These realizations, together multiple transmissivity conditioned by inversion hydraulic head data digital elevation model. For inverse conditioning an additional penalty term was added objective function, penalizing too deviations pattern image. As such, observed still honored calibrated realizations. It reduces significantly ensemble variance rates.