Significance of porosity for stratigraphy and textural composition in subsurface, coarse fluvial deposits: Boise Hydrogeophysical Research Site

Recognition and quantitative characterization of subsurface stratigraphic units in coarse unconsolidated fluvial deposits are difficult because large grain size and the large scale of sedimentary structures make direct interpretation from core difficult or impossible. In this paper, we use porosity data from well logs and grain-size distribution (GSD) data from core to investigate four pebble- and cobble-dominated units that have been identified in porosity logs in deposits at a research well field (Boise Hydrogeophysical Research Site). Lacking direct observation at an appropriate scale, questions about distribution of parameters and textural composition in these units are analyzed with statistical tests. The four pebble- and cobble-dominated "porosity stratigraphic" units may be grouped into two types: (1) Units I and 3 have low porosity (mean similar to0.17-0.18) and low porosity variance; and (2) Units 2 and 4 have higher porosity (mean similar to 0.23-0.24) and higher porosity variance. Based on GSD data, core samples are subdivided into five lithotypes. The five lithotypes occur in different proportions and have different vertical transition probability characteristics in the two types of units: (1) Units 1 and 3 have only framework-gravel-dominated lithotypes and have random vertical transition probability between these two lithotypes; and (2) Units 2 and 4 consist of both framework-gravel-dominated and sand- or matrix-dominated lithotypes and have structured vertical transition probability. The two framework-gravel-dominated lithotypes occur in all four stratigraphic units but have distinctly lower porosity in Units 1 and 3 (i.e., tighter packing) than in Units 2 and 4 (looser packing). Considering the repeated stratigraphic occurrence of (and the statistical significance of differences between) the two types of units, both the individual unit distinctions and the two unit groupings appear to be valid. It is reasonable to interpret that the observed packing differences associated with Units I and 3 compared with Units 2 and 4 are related to different sedimentary processes that produce different bedforms or grain fabrics, perhaps under different bedload transport rates.