Reservoir and seal units are predicted for use in petroleum exploration at global scale by estimating palaeosediment yield and grain size from drainage basins analysis. Sediment yield is governed partially by topography and climate, therefore it can also be predicted using these attributes. Palaeomorphological and climate attribute information was obtained from an Earth Systems Model (ESM). These models form two classes; Modern (MESM) and Palaeo (PESM). Existing sediment yield equations do not account for resolution and data limitations of the PESM. Therefore a need exists for an equation which considers these limitations. This was accomplished by formulating an equation in a MESM that has the same data limitations as a PESM. A predictive equation was produced by extracting basin attributes from the MESM and applying regression techniques against modern sediment yield measurements. The equation was then applied to the PESM to predict sediment yield and location of deposition in the palaeoenvironment. The resulting regression equation accounts for 71% of the sediment yield variance. Remaining variance could be explained by geological, tectonic and anthropogenic factors, which are not accounted for in the PESM. However, it is likely that geology and tectonic factors partially contribute to the equation due to their effects on topographic attributes. The MESM sum of global sediment yield is verified by independent estimations to be 22 GT yr1. PESM predictions correlate with independent calculations of seismic based sediment yield estimates. The case study of Cotton Valley demonstrates the predictive power of the data. It illustrates the palaeogeography of the region with accuracy. However the sediment yield value is low due to the inaccuracy of drainage analysis in this particular region. Outlet location in the Zambezi study is shown to vary by 75 km, which at a regional exploration scale is relatively minor.
