Crossplot

Key Takeaways

• The porosity-permeability crossplot (a log-log or semi-log plot of core measured permeability on the y-axis versus core measured porosity on the x-axis) is the most widely used crossplot in reservoir characterization because it captures the empirical relationship between two properties that are both essential for reservoir modeling but can only be directly measured at the scale of a core plug; the permeability-porosity trend on a crossplot is typically fit with a power law or exponential regression that is then used to predict permeability from log-derived porosity throughout the wellbore and between wells where core data are not available; the scatter around the regression line on a porosity-permeability crossplot reflects variability in pore geometry, cementation, clay content, and grain size distribution that is not captured by porosity alone, and this scatter is the principal justification for more sophisticated permeability prediction methods (capillary pressure-based models, nuclear magnetic resonance permeability models, neural network predictions from multi-variable log inputs) that attempt to honor the additional variability rather than treating it as random noise.
• AVO (amplitude versus offset) crossplots of intercept versus gradient are the primary tool for classifying seismic amplitude anomalies in terms of their fluid and lithology origin: the intercept (the P-wave reflection amplitude at zero offset) and gradient (the rate of amplitude change with offset) of seismic reflections from different lithological boundaries plot in characteristic regions of the intercept-gradient crossplot that correspond to the Rutherford and Williams AVO classification system; a gas sand below a shale overlay plots in a specific quadrant with a negative intercept and a negative gradient (Class III AVO), while brine sand plots near the origin with a near-zero intercept and a small positive or negative gradient; the separation between the gas and brine sand populations in the intercept-gradient crossplot is the quantitative basis for seismic fluid discrimination, and the crossplot's power lies in combining two measurements that individually have ambiguous lithology-fluid content into a joint display where the geologically distinct populations separate visually; the background trend (the distribution of reflections from normally compacting water-saturated shale-sandstone sequences) is a reference line on the crossplot that gas sands deviate from in predictable ways, enabling anomaly identification.
• The M-N crossplot and MID (matrix identification) crossplot are specialized petrophysical crossplots designed to identify complex lithologies in carbonate and evaporite sequences where the standard neutron-density crossplot cannot resolve mineral mixtures involving more than two components simultaneously: the M-N crossplot uses the M parameter (combining sonic and density measurements) and the N parameter (combining neutron and density measurements) to define a crossplot space where different minerals occupy unique positions that are better separated than in the simple neutron-density space; the MID crossplot uses the apparent matrix density and the apparent matrix neutron porosity (both calculated from the actual log readings) to define a crossplot space where limestone, dolomite, anhydrite, salt, quartz, and mixtures thereof plot in positions that allow multi-component mineral identification; these specialized crossplots are most commonly used in the evaluation of complex carbonate reservoirs in the Middle East, the Permian basin, and other carbonate-dominated provinces where two-mineral assumptions lead to significant errors in porosity and lithology interpretation.
• The crossplot of core permeability versus log-derived permeability is the definitive calibration diagnostic for any log-based permeability prediction: when the predicted permeability from a log analysis model is plotted against the directly measured core permeability for the same depth intervals, the scatter and systematic bias in this crossplot reveal the accuracy and precision of the prediction method; a well-calibrated model produces points that cluster tightly around the 45-degree (1:1) line with no systematic offset, indicating both good precision (tight clustering) and good accuracy (no bias); a model that consistently over-predicts permeability in tight zones and under-predicts in high-permeability zones is revealed by a characteristic S-shaped deviation from the 1:1 line; the permeability crossplot is the industry standard diagnostic that must be shown in any petrophysical study claiming to quantify reservoir permeability distribution, and its absence from a study report should prompt questions about the calibration rigor of the underlying analysis.
• Crossplots of well performance parameters (initial production rate, estimated ultimate recovery) versus completion variables (proppant intensity in pounds per foot, fluid intensity in barrels per foot, cluster spacing, lateral length) are the primary data-mining tools that unconventional operators use to optimize completion design by learning from historical performance across their acreage; when EUR is crossplotted against proppant loading for 200-500 wells from the same formation, the resulting scatter plot and trend line reveal whether more proppant systematically improves recovery (positive slope), at what loading level diminishing returns become apparent (the slope flattens), and how much scatter there is in the relationship (how much of the performance variation is explained by proppant loading versus other factors like geology); the limitation of these empirical crossplots is that they conflate the effects of multiple correlated variables (operators that pump more proppant also tend to use more fluid, tighter cluster spacing, and longer laterals), making it difficult to isolate the independent contribution of any single completion variable without proper multivariate statistical analysis or controlled completion design experiments.