Treatment of the Top of Bedrock in Previously Glaciated Regions

In the previously glaciated regions across the northern tier of States, a digital dataset of the altitude of the base of Quaternary glacial deposits (Soller and Garrity, 2018; DAS13 on fig. 2 and in Sweetkind and others, 2024) was sampled to each model cell using zonal statistics sampling in a GIS (bedrock method 2 labeled “BEDRX2” in fig. 4 and table 1). As part of their investigation, Soller and Garrity (2018) released a “Map of Bedrock Topography,” which was the altitude of the top of any consolidated rock present beneath the glacial sediments. For this study, this subsurface horizon needed to be subdivided into areas where glacial sediments overlie stratified Phanerozoic consolidated rocks that are defined as “the top of bedrock” and, where bedrock units are absent because of erosion, areas where glacial sediments overlie highly deformed or crystalline rocks defined as “basement rock units.” National bedrock geologic maps that portray the subglacial geology of the region (King and Beikman, 1974; Schruben and others, 1998; Horton and others, 2017) were used to make selection sets of map polygons of geologic units that could be classified as “bedrock” (stratified Phanerozoic consolidated rocks) or “basement” (deformed crystalline rocks of mostly Precambrian age). The selection sets were then used to guide whether the maps of Quaternary sediment thickness were used to populate the model cells for the altitude of the top of bedrock or the top of basement.

In the four-State region of Illinois, Indiana, Kentucky, and Missouri, the altitude of the buried top of bedrock surface was sampled from a digital regional map that was compiled as part of the National Science Foundation-funded EarthScope OIINK experiment (Illinois State Geological Survey, 2015; DAS37 on fig. 2 and in Sweetkind and others, 2024). That compilation itself was created by merging statewide bedrock topographic contours from Missouri, Illinois, and Indiana; in Kentucky, the EarthScope OIINK experiment used topographic data because bedrock is at or very near the surface across the State (Illinois State Geological Survey, 2015). In the previously glaciated parts of Missouri, Illinois, and Indiana, these digital data were used instead of the Soller and Garrity (2018) dataset. Quaternary sediments of the Mississippi Embayment are present in a small region in southeastern Missouri, southern Illinois, and western Kentucky; in this region, the Illinois State Geological Survey (2015) dataset was not used and instead, a contoured map of the top of Cretaceous beneath the Mississippi Embayment was used to represent the base of the Cenozoic section in this area (DAS75 on fig. 2 and in Sweetkind and others, 2024).

In the northeastern part of the Williston Basin in the northeast part of North Dakota, Cenozoic sedimentary strata underlie the Quaternary glacial sediments and overlie consolidated Cretaceous rocks that are classified as “bedrock” in adjacent areas (Thamke and others, 2014; DAS06 on fig. 2 and in Sweetkind and others, 2024). Within the basin, as much as 700 m of sedimentary rocks of the Paleogene Fort Union Formation overlie Upper Cretaceous mudstones and sandstones of the Hell Creek Formation, which overlie as much as 1,000 m of Upper Cretaceous marine shale (Gill and Cobban, 1973; Thamke and others, 2014; Spangler, 2024b). In contrast to most of the area of previously glaciated States east of the Rocky Mountains, where the base of the glacial sediments defines the altitude of the top of bedrock or, locally, the top of basement, in this part of the Williston Basin the top of bedrock is defined as the base of the Cenozoic section and the top of underlying Cretaceous consolidated rocks. Using the base of the Cenozoic section as the top of bedrock within the Williston Basin is consistent with the definition of the top of bedrock beneath the High Plains aquifer and in the intermontane basins of the Intermountain West (bedrock method 4 labeled “BEDRX4” on fig. 4 and table 1). Digital data that define the base of the Fort Union Formation (Thamke and others, 2014; DAS06 on fig. 2 and in Sweetkind and others, 2024) were sampled to define the altitude of the top of bedrock in this area.

Treatment of the Top of Bedrock Overlain by Thin Alluvial Deposits in the Midcontinent

To the south of the previously glaciated regions of the United States, bedrock units are discontinuously overlain by unconsolidated surficial deposits and weathered residuum developed on bedrock. These deposits have been mapped at a national scale, and the thickness of the deposits has been estimated (Soller and Reheis, 2004; Soller and others, 2009). Across large parts of the Central United States in areas not covered by glacially deposited sediment, these unconsolidated materials represent most or all of the thickness of sediment overlying consolidated bedrock (bedrock method 3 labeled “BEDRX3” in fig. 4 and table 1). As such, estimated thickness reported on the national compilation of surficial materials was used to estimate the altitude of bedrock in these nonglaciated regions.

The national map of surficial materials (Soller and Reheis, 2004; Soller and others, 2009; DAS41|DAS42 on fig. 2 and in Sweetkind and others, 2024) used three thickness categories that were applied to various types of surficial deposits: (1) areas where the surficial material is discontinuous or patchy in distribution, (2) areas where surficial materials are continuous in distribution and are generally less than 31-m thick, and (3) areas where surficial materials exceed 31 m in thickness. For defining the altitude of the top of bedrock, we arbitrarily assigned thickness values of 0 m, 25 m, and 50 m, respectively to the three thickness categories. In practice, a national bedrock geologic map (King and Beikman, 1974; Schruben and others, 1998) was used to identify regions where bedrock was mapped without overlying thick sections of Cenozoic sedimentary rocks, eliminating, for example, the region of the Mississippi Embayment. Within the defined regions, polygons from the national map of surficial materials (Soller and Reheis, 2004; Soller and others, 2009; DAS41|DAS42 on fig. 2 and in Sweetkind and others, 2024) were selected within a GIS, and the selected cells were assigned a thickness value based on the map unit description and published thickness classes. An altitude of the top of bedrock was calculated by subtracting the estimated sediment thickness from the DEM-derived land surface elevation (Consortium for Spatial Information, 2018). In places, thickness values were modified along alluvial drainages so that the deposits underlying the trunk stream had a greater alluvial thickness than deposits underlying tributary streams. Locally beneath map polygons of large lakes and reservoirs depicted as water, a small constant sediment thickness value was arbitrarily assigned; these polygons were depicted without an interpreted sediment thickness on the national map of surficial materials (Soller and Reheis, 2004; Soller and others, 2009; DAS41|DAS42 on fig. 2 and in Sweetkind and others, 2024).

Treatment of Top of Bedrock Beneath the High Plains Aquifer

The High Plains aquifer consists of as much as 250 m of hydraulically connected geologic units of Miocene to Quaternary age in the western part of the Great Plains in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming (Gutentag and others, 1984). The base of the High Plains aquifer has been mapped as part of USGS hydrogeologic studies (Cederstrand and Becker, 1998; Nebraska Water Science Center, 2016) and other studies (Macfarlane and Wilson, 2019). Across most of this region, the base of the High Plains aquifer is defined at the altitude where the Miocene–Pliocene Ogallala Formation and overlying Quaternary deposits are separated by a regional angular unconformity from underlying Permian to Cretaceous consolidated rocks (Gutentag and others, 1984). However, in the northwestern part of the aquifer extent in western Nebraska, northeastern Colorado, southwestern South Dakota, and southeastern Wyoming, Oligocene sediments of the White River Group and Arikaree Group are present between the Cretaceous rocks and the aquifer units (Gutentag and others, 1984).

For the large part of the region where the High Plains aquifer unconformably overlies Permian to Cretaceous consolidated rocks, the contoured base of the aquifer (Cederstrand and Becker, 1998; Nebraska Water Science Center, 2016; DAS04 and DAS05, respectively, on fig. 2 and in Sweetkind and others, 2024) was used to define the base of the Cenozoic section. For the northwestern part of the aquifer area where older Paleogene units underlie the Ogallala Formation, geologic map relations, a subcrop map at the base of the High Plains aquifer (Gutentag and others, 1984), and stratigraphic thickness estimates from Swinehart and others (1985) provided the conceptual framework to estimate the depth of the Mesozoic consolidated rocks beneath the contoured base of the aquifer. In this area, the altitude of the top of bedrock, defined as the base of the Paleogene section beneath the aquifer, was derived from data from a 3-D geologic model of the Denver-Julesberg Basin (Goldberg, 2025), a 3-D geologic model of the western half of South Dakota (Spangler, 2024a), and the top of the highest Cretaceous formations in oil and gas exploration boreholes in Nebraska (Nebraska Oil and Gas Conservation Commission, 2025).

Treatment of the Top of Bedrock in the Gulf Coast and Atlantic Coastal Plain

The top of bedrock in the Gulf Coast and the Atlantic Coastal Plain was taken as the base of the Cenozoic sedimentary section and top of Cretaceous rocks (bedrock method 4 labeled “BEDRX4” in fig. 4 and table 1). In the Gulf Coast, a contoured map of the top of Cretaceous from a regional stratigraphic study and aquifer analysis (Hosman, 1996; Sweetkind and others, 2023a; DAS75 on fig. 2 and in Sweetkind and others, 2024) was used for most of the onshore area; a second contoured map of the top of Cretaceous (Williamson, 1959; DAS54 on fig. 2 and in Sweetkind and others, 2024) was used for the Texas and Louisiana coastal areas. In southern Mississippi, southern Alabama, and the Panhandle of Florida, altitude of the top of the Cretaceous section was derived from contour maps of the southeastern Coastal Plain (Renken, 1996; Cannon and others, 2012). For this area, contours of the top of Navarroan rocks were used; where those rocks were mapped as absent, contours from the underlying top of Taylorean rocks were used (Renken, 1996; Cannon and others, 2012; DAS79 on fig. 2 and in Sweetkind and others, 2024).

In the Atlantic Coastal Plain, information on the altitude of the base of the Cenozoic sedimentary section and top of Cretaceous rocks came from structure contour maps of the entire Atlantic Coastal Plain (Pope and others, 2016; DAS11 on fig. 2 and in Sweetkind and others, 2024), and from more detailed local maps of the Coastal Plain in Georgia (Herrick and Vorhis, 1963; DAS48 on fig. 2 and in Sweetkind and others, 2024), North Carolina and South Carolina (Campbell and Coes, 2010; DAS12 on fig. 2 and in Sweetkind and others, 2024), Virginia (McFarland and Bruce, 2006; DAS50 on fig. 2 and in Sweetkind and others, 2024), and New Jersey (Volkert and others, 1996; DAS11|DAS51 on fig. 2 and in Sweetkind and others, 2024). Many of these maps were compiled in digital form by Mills and others (2020), which was the source of digital data used in this study.

Treatment of the Top of Bedrock in Florida

USGS hydrogeologic studies of the Floridan aquifer system that occurs in the subsurface in Florida and parts of Georgia, Alabama, and South Carolina have established the relations between regionally correlated time-stratigraphic and rock-stratigraphic units and the corresponding position of the aquifer system (Williams and Dixon, 2015; Williams and Kuniansky, 2016). The base of the Floridan aquifer system is marked by low-permeability rocks that range in age from late Eocene to late Paleocene, depending on the area considered. Throughout the Florida Panhandle, the base of the Floridan aquifer system is defined within the upper third of the Paleocene Cedar Keys Formation (Williams and Dixon, 2015; Williams and Kuniansky, 2016). The contoured base of the Floridan aquifer system (Williams and Dixon, 2015; DAS53 on fig. 2 and in Sweetkind and others, 2024) was used to approximate the base of Cenozoic surface in Florida because of the ready availability of the digital dataset and the general consistency of this horizon with adjacent base-of-Cenozoic datasets from the Atlantic and Gulf Coastal Plains.

Treatment of the Top of Bedrock in the Western United States

Regional gravity studies have long been used in the Great Basin to estimate the shape and extent of Cenozoic basins in three dimensions (Jachens and Moring, 1990; Saltus and Jachens, 1995; Blakely and others, 1999). The large density contrast between pre-Cenozoic consolidated rocks in the Great Basin and the overlying low-density Cenozoic sedimentary basin fill is used to estimate the depth of pre-Cenozoic rocks. An iterative technique separates the isostatic residual gravity anomaly into basin and basement components. Basin thickness is estimated from the basin gravity component using a density-depth relation within the basin fill (Jachens and Moring, 1990; Saltus and Jachens, 1995); the method was later modified to incorporate well and other constraints to improve the thickness estimates (for example, Blakely and Ponce, 2001).

For part of the Western United States, the altitude of the base of the Cenozoic section was derived from a gravity-based basin depth map of the Basin and Range Province (Saltus and Jachens, 1995) augmented by subsequent gravity-based studies (bedrock method 1 labeled “BEDRX1” in fig. 4 and table 1). For this study, the original basin-depth map of the Basin and Range Province (Saltus and Jachens, 1995) was modified by the results of more detailed gravity-based studies of the northern Rocky Mountains (Mankinen and others, 2004; DAS82 on fig. 2 and in Sweetkind and others, 2024), in the Death Valley region of Nevada and California (Blakely and Ponce, 2001), in eastern Nevada (Watt and Ponce, 2007), and in west-central Nevada (Ponce and Glen, 2008). The last three of these detailed datasets had been subsequently combined to support the construction of a hydrogeologic framework model of the eastern Great Basin (Cederberg and others, 2011; Sweetkind and others, 2026; DAS82 on fig. 2 and in Sweetkind and others, 2024).

One challenge of estimating the top of bedrock from gravity studies is accounting for the significant but sometimes poorly known effect that thick sections of volcanic rocks have on the density‐depth structure of the subsurface. Volcanic rocks can produce density variations that can be abrupt and unpredictable as a result of variable degree of welding in ash-flow tuffs, effects of alteration, such as zeolitization, or the presence of thick accumulations of pumiceous or nonwelded tuff. Because previous gravity-based studies may have accounted for the presence of volcanic rocks in varying ways, there is uncertainty whether the gravity defined surface within the three-layer model reliably estimates the altitude of the base of the Cenozoic section. For the National Crustal Model, Boyd (2019) suggested that the higher-density rocks that lie beneath the gravity-defined surface might include volcanic rocks older than Miocene in addition to pre-Cenozoic consolidated sedimentary rocks. This suggestion is generally consistent with the timing of eruption of widespread Oligocene to Miocene ignimbrites across the northern and central Great Basin (Best and others, 2016); these volcanic rocks would presumably be present in the deeper parts of many basins. Where a significant thickness of volcanic rocks is present within a basin, the modeled top of bedrock, defined in the model as the base of Cenozoic, might more closely represent the base of Miocene rocks, such that true bedrock could be even deeper. Another complicating factor can be the presence of basalt flows younger than Miocene within the low-density sedimentary basin fill; such basalt can cause the gravity-derived surface to be too shallow.

In parts of the Western United States outside of the extent of the gravity-based basin depth map of the Basin and Range Province (Saltus and Jachens, 1995), the altitude of the base of the Cenozoic section was derived from an elevation grid created for the crustal model of the Western United States (Shah and Boyd, 2018; Boyd, 2019; bedrock method 6 “labeled” BEDRX6 in fig. 4 and table 1). In this area, the altitude of the top of bedrock is defined by a mix of methods and is not solely gravity-based (Shah and Boyd, 2018). The base case in this area was a model of combined sediment and soil thickness derived by Pelletier and others (2016) that attempted to model the thickness of unconsolidated sediments of Miocene age or younger on a regional scale. The surface defined by Pelletier and others (2016) was modified locally by Shah and others (2018; DAS03 on fig. 2 and in Sweetkind and others, 2024) through the addition of results from well data and gravity-based surveys. However, in much of northern California, the eastern parts of Oregon and Washington, and western Idaho the altitude grid of Shah and others (2018) was unmodified from the original sediment thickness model (Pelletier and others, 2016). In these regions, the sediment thickness estimate is not based on gravity data, in contrast to the area defined as “bedrock method 1” (labeled “BEDRX1” in fig. 4 and table 1).

One challenge in the use of the elevation grid created for the crustal model of the Western United States (Shah and Boyd, 2018; Shah and others, 2018), especially for the areas where the sediment thickness model of Pelletier and others (2016) was unmodified by other data, is that a primary source for this grid, the national-scale sediment thickness map of Frezon and others (1983), did not explicitly consider or contour thickness of Cenozoic volcanic rocks in the Western United States. As a result, in areas of the Western United States where thick sections of volcanic rocks are exposed, bedrock is calculated to be at land surface. Geologically, the volcanic rocks would simply be a consolidated rock layer that would be local bedrock as distinguished from unconsolidated sediment. However, using the definitions within the three-layer model, “bedrock” is defined as the base of Cenozoic rocks, forcing the altitude of pre-Cenozoic rocks to be at land surface in areas where thick sections of Cenozoic volcanic rocks are known to exist. This is a known deficiency in the three-layer model that is not completely solved and includes areas in northeastern California, the volcanic plateaus of eastern Oregon, and parts of the northern Great Basin.

For the compilation in this study, the regional elevation surface of Shah and others (2018) was replaced locally by datasets that explicitly defined the presence and thickness of Cenozoic volcanic rocks. On the Columbia Plateau in eastern Washington, a gridded surface representing the altitude of the top of bedrock that underlies the Miocene Columbia River Basalt Group (Burns and others, 2011; DAS83 on fig. 2 and in Sweetkind and others, 2024) was used to represent the top of the pre-Cenozoic section; this forced the bedrock to a depth below the known volcanic section. In a similar way, in the western half of the Snake River Plain of southwestern Idaho, contours defining the base of basaltic volcanic rocks (Whitehead, 1992; DAS71 on fig. 2 and in Sweetkind and others, 2024) were used to approximate the base of the Cenozoic section, pushing bedrock down and allowing some part of the volcanic rock section to exist above the base-of-Cenozoic bedrock interface.

Elsewhere in the Pacific Northwest, geologic outcrop data (Colgan and others, 2011) and analysis of gravity and seismic data (Zucca and others, 1986; Khatiwada and Keller, 2015) suggest that the Cenozoic volcanic rock section is between 2–4-km thick in the region, but these topical studies provide thickness data only at points or along profiles. There is no spatially distributed dataset of the thickness of the Cenozoic section to sample as an update to the previously modeled result. Therefore, certain parts of the 3-D model are left showing pre-Cenozoic bedrock at land surface in places where geologic maps clearly show thick sections of Cenozoic volcanic rocks.

For comparison, the National Crustal Model (Boyd, 2019) adjusted for the presence of volcanic rocks in the Western United States in two ways. First, bedrock was defined as the depth to the base of unconsolidated sediment and a sediment thickness model (Pelletier and others, 2016) was modified to estimate depth to pre-Miocene strata. Second, the National Crustal Model (Boyd, 2019) estimated the top of pre-Cenozoic rocks using a map-based nearest neighbor approach that used distance to pre-Cenozoic rocks shown on regional maps. The three-layer model of this report differs substantially from the National Crustal Model (Boyd, 2019) in this area because of the differences in approach.

Top of Bedrock Data in Basins of the Rocky Mountain Region

Intermountain sedimentary basins of the Rocky Mountain region developed adjacent to basement-cored uplifts during the Laramide orogeny from latest Cretaceous through the Eocene (Baars and others, 1988; Lawton, 2008). Marine and marginal marine sedimentation associated with the intracratonic Cretaceous seaway was supplanted by deposition in continental sedimentary environments as subsiding basins received sediment shed from fault-block uplifts. Basin deposition during the Cenozoic took place in alluvial fan, fluvial, deltaic, and lacustrine settings (Baars and others, 1988; Lawton, 2008). Each of the basins of the Rocky Mountain region has a unique tectonic and basin-filling history (Baars and others, 1988; Dickinson and others, 1988), such that defining the top of “bedrock” as the base of the Cenozoic section can serve as a useful starting point to compare basin evolution in the region.

For each basin in the Rocky Mountain region (fig. 7; table 3), a published structure contour map defining the base of the Cenozoic section was georeferenced and digitized in a GIS, converted to a raster grid and elevation values were sampled to the 2.5-km array of cells (bedrock method 4 labeled “BEDRX4” in fig. 5 and table 2). For the Raton, San Juan, Greater Green River, Powder River, and Bighorn Basins, a mapped horizon that corresponded to the Cretaceous–Paleogene boundary was digitized and sampled (fig. 2; table 3; Parker, 1986; Martin, 1996; Cather, 2004; Topper and others, 2011; Thamke and others, 2014; Lynds and Carroll, 2015). For the Uinta, Piceance, Hannah, Shirley, and Laramie Basins, no published contoured horizon was available for the Cretaceous–Paleogene boundary; instead, a contoured horizon or data picked from oil and gas wells within the Upper Cretaceous section was chosen as a proxy for the base of Cenozoic (fig. 2; table 3; Larsen, 1983; Love and others, 1993; Johnson and Roberts, 2003; Wyoming Oil and Gas Conservation Commission, undated). In these basins, the vertical offset from the true elevation of the Cretaceous–Paleogene boundary was estimated from stratigraphic studies and tabulated (fig. 7; table 3). For the Wind River Basin, the best available published contour data were for a horizon within the lower part of the Paleogene section above the Cretaceous–Paleogene boundary (fig. 2; table 3; Roberts and others, 2007; DAS52). In the Denver Basin of Colorado, the Cretaceous–Paleogene boundary is exposed in outcrop at several locations around the margins of the basin (fig. 2; Dechesne and others, 2011; DAS68); to capture more of the basin geometry, the top of a deeper horizon, the Upper Cretaceous Laramie Formation (fig. 2; Dechesne and others, 2011; DAS68), was digitized (table 3).

Treatment of the Top of Bedrock Where Basement Rocks Are Exposed at Land Surface

The altitude of the top of bedrock was assigned a null value wherever rocks defined as “basement” cropped out at land surface (bedrock method 5 labeled “BEDRX5” in table 1). National-scale geologic maps (King and Beikman, 1974; Schruben and others, 1998; Horton and others, 2017) were used to make selection sets of map polygons of geologic units that could be classified as “basement”; cells whose centroids fell within basement map polygons were assigned a null elevation value for bedrock altitude, implying that bedrock is absent in that cell.

Thickness of the Sediment Layer Between Land Surface and Top of Bedrock

The modeled thickness of the unconsolidated to weakly consolidated sediment layer was calculated as the distance between land surface and the top of bedrock (fig. 8). In areas of the conterminous United States where the bedrock unit has thickness in the model, thickness of the overlying unconsolidated sediments is calculated as the elevation of land surface minus the altitude of the modeled top of the bedrock layer. In areas where bedrock has been removed by erosion and unconsolidated deposits overlie basement rocks, the thickness of the unconsolidated sediments is calculated as the elevation of land surface minus the altitude of the modeled top of the basement layer.

Two first-order thickness trends are evident on the thickness of sediment map (fig. 8). One notable change in the modeled sediment thickness is caused by the change in the definition of the top of bedrock from the base of Quaternary deposits to the base of the Cenozoic. In the northern tier of previously glaciated States, where glacial deposits overlie bedrock, and in the central mid-Atlantic States and the southern midcontinent where patchy alluvial or colluvial materials overlie bedrock, the thickness of unconsolidated deposits is less than 500 m, and in many places, less than 50 m (fig. 8). In the western part of the United States, to the west of the 99th meridian (west of the north arrow; fig. 8), and in the Gulf Coastal Plain and Mississippi Embayment, the thickness of sediment calculated to the base of Cenozoic is far greater, locally resulting in a thickness of several kilometers (fig. 8). A second change in the modeled sediment thickness occurs in the Western United States at the point where the top of bedrock defined by gravity-based depth to consolidated rocks changes to a more general sediment thickness model augmented locally by other data (fig. 8; transition from the northwest corner of Nevada to eastern Oregon and northeastern California).

Treatment of the Top of Basement for the Craton in the Midcontinent

The digital elevation surface of the top of crystalline Precambrian rocks on the cratonic platform (Domrois, 2013; Marshak and others, 2017a; fig. 3A) was the starting point for the compilation of basement rock data for this study. The digital raster data and table of map sources were downloaded (Marshak and others, 2017a), data were reprojected to a preferred coordinate system, and altitude values of the Precambrian surface were converted from feet to meters. The published top of Precambrian surface was locally edited to remove minor elevation errors, resampled to the 2.5-km grid used in this study, and, where geologic maps showed Archean or Proterozoic rocks at land surface (Schruben and others, 1998; Horton and others, 2017), the altitude values of the sampled surface were adjusted to match the land surface elevation grid used in this study. The digital elevation surface of Marshak and others (2017a; fig. 3A) was compared to statewide maps showing the altitude of the top of Precambrian rocks that were sources for the Marshak and others (2017a) compilation, including Wyoming (Wyoming State Geological Survey, 2022; DAS19 on fig. 3B and in Sweetkind and others, 2024), Colorado (Hemborg, 1996; DAS27 on fig. 3B and in Sweetkind and others, 2024), Kansas (Cole, 1976; DAS21 on fig. 3B and in Sweetkind and others, 2024), Iowa (Anderson, 1995; DAS14 on fig. 3B and in Sweetkind and others, 2024), and Ohio (Baranoski, 2013; DAS16 on fig. 3B and in Sweetkind and others, 2024). In all these cases, the regional “basement” compilation (Marshak and others, 2017a; DAS02 on fig. 3B and in Sweetkind and others, 2024) correctly represented the data shown on the source maps. As compilation of altitude data from areas surrounding the cratonic platform progressed, some of the outlying regions of the Marshak and others (2017a) elevation surface were edited for consistency with other input data discussed in this report.

The compiled top of crystalline Precambrian rocks of the continental interior (Domrois, 2013; Marshak and others, 2017a, b; fig. 3A) was modified and locally extended for the current (2025) compilation using cross section data or contoured data from numerous published reports (fig. 3B). A raster surface for the top of Precambrian was created for the four-State region of Missouri, Illinois, Indiana, and Kentucky as part of regional EarthScope OIINK research on the tectonics of the midcontinent (Illinois State Geological Survey, 2015; DAS37 on fig. 3B and in Sweetkind and others, 2024). The source maps used for the four-State compilation are the same as those maps used by Domrois (2013) and subsequent reports for the craton-wide compilation. For this report, the data from the four-State compilation (Illinois State Geological Survey, 2015) were used in this region because these data were created with direct involvement of the four State geologic surveys and are presumed to be the authoritative dataset for this region. A structure contour map of the top of Precambrian rocks in the upper Midwest developed by the Midwest Regional Carbon Initiative helped to refine basement elevations and the presence of the Rome Trough beneath the Appalachian Basin (Battelle, 2005; DAS84 on fig. 3B and in Sweetkind and others, 2024). To the west and southwest of the Permian Basin in west Texas, published contour maps of the top of Precambrian rocks (Hills, 1984; Bureau of Economic Geology, 2009; both are shown as DAS34 on fig. 3B and in Sweetkind and others, 2024 and Shuster and others, 2021; DAS85 on fig. 3B and in Sweetkind and others, 2024) were digitized and sampled to extend the compiled top of crystalline Precambrian rocks of the continental interior (Domrois, 2013; Marshak and others, 2017a, b) farther to the west.

The published basement compilation of Marshak and others (2017a, b) extends only to the eastern front of the Idaho-Wyoming thrust belt in western Wyoming (fig. 3A), although the top of Precambrian rocks is interpreted as dipping westward beneath the thrust belt in relatively undeformed fashion (Dixon, 1982; Wyoming State Geological Survey, 2022). Data on the altitude of the Precambrian surface beneath the Idaho-Wyoming thrust belt in western Wyoming came from a digital version of the structure contours shown on the Precambrian basement map of Wyoming (Wyoming State Geological Survey, 2022; DAS19 on fig. 3B and in Sweetkind and others, 2024) and from altitude points measured from published serial geologic cross sections drawn across the thrust belt (Dixon, 1982; DAS58 on fig. 3B and in Sweetkind and others, 2024). Structure contour and point data were combined in a GIS, interpolated to a raster surface, and the result was sampled to the 2.5-km cells for this study.

The published basement compilation of Marshak and others (2017a, b; fig. 3A) did not include elevation data from within the Rio Grande Rift in New Mexico or Colorado, although contour maps of the top of Precambrian in New Mexico (Broadhead and others, 2009; DAS61 on fig. 3B and in Sweetkind and others, 2024) and Colorado (Hemborg, 1996; DAS27 on fig. 3B and in Sweetkind and others, 2024) show contours for some areas within the rift. The Colorado map (Hemborg, 1996) was georeferenced in a GIS and the contours digitized and combined with digital contour data of the top of Precambrian in New Mexico (Broadhead and others, 2009); contours were interpolated to a raster surface, and the result was sampled to the 2.5-km cells for this study.

The compiled top of crystalline Precambrian rocks of the continental interior (Domrois, 2013; Marshak and others, 2017a, b; fig. 3A) was modified using the modeled top of Precambrian rocks from several published 3-D geologic models. The modeled top of Precambrian surface from digital 3-D geologic models of the Williston Basin in Montana, North Dakota, and South Dakota (Gelman and Johnson, 2023; Spangler, 2024b; DAS64 on fig. 3B and in Sweetkind and others, 2024) and the Powder River Basin in Wyoming and Montana (Spangler and others, 2023; Spangler, 2024b; DAS65 and DAS91, respectively, on fig. 3B and in Sweetkind and others, 2024) was extracted from the published datasets, manipulated in a GIS, and sampled to the 2.5-km cells. Elevation data for the top of Precambrian rocks beneath the Colorado Plateau were sampled from the top of a hydrogeologic unit contained within a 3-D hydrogeologic model of the Upper Colorado River Basin in New Mexico, Arizona, Utah, Colorado, and Wyoming (Sweetkind and others, 2023b). In the published hydrogeologic model, the deepest hydrogeologic unit combined altitude data from Precambrian basement rocks and Cenozoic intrusive rocks to model the crystalline rocks at the base of the aquifer system (Sweetkind and others, 2023b; DAS63 on fig. 3B and in Sweetkind and others, 2024). This horizon was extracted from the published dataset, and model cells associated with Cenozoic intrusive rocks were identified by comparison with regional geologic maps (Schruben and others, 1998; Horton and others, 2017). Model cell altitude values were manually edited to remove land surface elevations or altitudes in the shallow subsurface that represented outcropping or near-surface intrusive bodies; altitude values were replaced by values that represented a local average altitude for Precambrian rocks adjacent to the modeled intrusive body. These revised data were then sampled to the 2.5-km cells for this study. A digital hydrogeologic model of the Death Valley regional flow system in Nevada and California (Faunt, 2006, Faunt and others, 2010; DAS67 on fig. 3B and in Sweetkind and others, 2024) included a model unit called the crystalline confining unit that represented the altitude of crystalline Precambrian rocks at the base of the groundwater flow system. Altitude data for the crystalline confining hydrogeologic unit were selected from the western parts of the model, near and to the west of the Death Valley fault zone, where Precambrian rocks crop out or are inferred in the subsurface in the Funeral Mountains, the Panamint Range, and surrounding region (Sweetkind and others, 2001; Workman and others, 2002). The selected model data were manipulated in a GIS and sampled to the 2.5-km cells.

Treatment of the Top of Basement Surface in the Ouachita-Marathon Orogenic Belt and Gulf Coast Region

Basement rocks of the Gulf Coast region lie to the south of the Ouachita-Marathon orogenic belt that forms the northern and northwestern flank of the Gulf of Mexico Basin (Viele and Thomas, 1989; Poole and others, 2005; Snedden and Galloway, 2019). In general, Proterozoic basement rocks of the Gulf Coast region are considered to have Gondwanan affinities (basement type 2 labeled “BSMT2” in fig. 5 and table 2) and were added to the margin of the continent during a late Paleozoic collisional-subductional event (Viele and Thomas, 1989; Lund and others, 2015; Ewing and others, 2019). These rocks were subsequently subjected to rifting and crustal attenuation (basement type 4 labeled “BSMT4” in fig. 5 and table 2) during the Late Triassic to Early Jurassic during the breakup of the supercontinent Pangea, resulting in a south-to-southeastward transition from crust with continental affinities to oceanic crust (Worrall and Snelson, 1989; Sawyer and others, 1991; Snedden and Galloway, 2019) that is now deeply buried by coastal plain sediments and rock strata.

National-scale tectonic and basement maps of the Gulf Coast region show the structural configuration of basement rocks at the base of the postrift Mesozoic marine section (Sawyer and others, 1991; Muehlberger, 199290). Knowledge of the depth to basement in the Gulf Coast region come from onshore wells around the margins of the basin that penetrate, and sometimes sample, the basement rocks and from wells that penetrate deeply into the sedimentary section without encountering basement, which provide a minimum constraint on basement depth (Sawyer and others, 1991; Snedden and Galloway, 2019). Most of the knowledge of depth to basement from the water-covered areas of the central and southeastern Gulf of Mexico Basin comes from seismic-reflection profiles that image basement (Sawyer and others, 1991; Snedden and Galloway, 2019).

For this study, the altitude of basement rocks to the south of the Ouachita-Marathon orogenic belt in the Gulf Coast region was compiled from digital contour data served by the American Association of Petroleum Geologists for the Tectonic Map of North America (Muehlberger, 1992; DAS32 on fig. 3B and in Sweetkind and others, 2024). This map included data from seismic-reflection profiles and well data to constrain the location of contours on the basement surface and was used in preference to older national-scale tectonic maps (Basement Rock Project Committee, 1967; Bayley and Muehlberger, 1968; King, 1969). For the immediate onshore part of the Gulf Coast, the contoured base of the Mesozoic marine section, (Sawyer and others, 1991; DAS73 on fig. 3B and in Sweetkind and others, 2024) were used to augment the basement contours shown on the Tectonic Map of North America (Muehlberger, 1992). Basement contours shown on the King (1969) tectonic map were notably shallower than other basement maps of the region. For example, near Houston, Texas, the King (1969) map shows the altitude of basement at about −3,000 m, whereas the map of Sawyer and others (1991) shows the altitude of basement at about −11,000 m and Muehlberger (1992) maps the top of basement at about −9,750 m. Along the Texas coastline, King (1969) maps the top of basement at −3,800 m, whereas Sawyer and others (1991) map the altitude of basement close to −11,000 m and Muehlberger (1992) maps the top of basement at about −12,000 m. The basement structure contours from King (1969) were not used in this study because the more recent tectonic maps (Sawyer and others, 1991; Muehlberger, 1992) are generally much deeper and are consistent with each other.

Along the north edge of the Ouachita-Marathon orogenic belt, the Proterozoic crust of the North American craton is obscured by, and possibly involved in, Pennsylvanian to Permian thrust sheets in a broad, arcuate belt that extends from the southeastern end of the Appalachian Mountains to the Marathon region of west Texas (Viele and Thomas, 1989; Ewing and others, 2019). Where subsurface data are plentiful, Proterozoic crystalline rocks may be projected beneath the allochthonous rocks of the orogenic belt (basement type 8 labeled “BSMT8” in fig. 5 and table 2). For the Ouachita Mountains, Arbenz (2008) created a suite of serial cross sections that portrayed the subthrust geometry and interpreted altitude of Precambrian basement beneath the thrusts. The change in basement definition from Precambrian basement of the craton to the basement rocks of the Gulf Coast region was arbitrarily taken as the point at which Precambrian basement of the craton was overlapped by the first thrust faults that involved highly deformed Cambrian forearc rocks (Arbenz, 2008). This created an abrupt elevation discontinuity across the orogenic front where two differently defined basement blocks are juxtaposed (fig. 9A, B).

Treatment of the Top of Basement in the Appalachians and Atlantic Coastal Plain

Metamorphic rocks of various grades form the crystalline, interior parts of the Appalachian orogenic belt (Hatcher, 1989); these rocks are defined as “basement” for the purposes of this study (basement type 3 labeled “BSMT3” in fig. 5 and table 2). These rocks crop out in northern Georgia, western North Carolina and South Carolina, western Virginia, and western Maryland (Hatcher, 1989; Horton and others, 2017). Basement rocks in this region were selected using the lithologic descriptor attributes in the Horton and others (2017) digital geologic map dataset; metamorphic rock names and metasedimentary and metaigneous lithologic descriptors were selected from the attribute table of the geologic polygons. The selection set from the geologic map was compared to the type and thickness of mapped surficial cover (Soller and Reheis, 2004) to confirm that the metamorphosed rocks were at or very near land surface. The top of basement altitude was then calculated for each 2.5-km cell using land surface elevation derived from a DEM.

The top of the basement surface beneath the southern Appalachian foreland fold and thrust belt has been constructed by combining data from seismic-reflection profiles, geologic maps, and well data (Hatcher and others, 2007; basement type 8 labeled “BSMT8” in fig. 5 and table 2). The eastward change in basement definition from the subthrust Archean to Proterozoic basement of the craton (basement type 8 labeled “BSMT8” in figs. 7B and 8) to crystalline rocks of the Blue Ridge–Piedmont crystalline sheet (basement type 3 labeled “BSMT3” in figs. 7B and 8; Hatcher and others, 2007) was arbitrarily chosen at the point where cratonic crust of the continental interior is overlapped in map view by younger, continent-vergent basement rocks. As a result of the change in definition, the mapped basement surface has a large and abrupt change in elevation at the orogenic front (fig. 9A, B).

The altitude of the top of the basement rocks of the Appalachian orogen that are buried beneath the Atlantic Coastal Plain (basement type 3 labeled “BSMT3” in fig. 5 and table 2) initially came from digital structure contours shown on the “Tectonic Map of North America” (Muehlberger, 1992; DAS32 on fig. 3B and in Sweetkind and others, 2024). These data were subsequently modified by data from regional studies that mapped the sub-Coastal Plain basement rocks in the Southeast Coastal Plain in northern Florida, southeastern Georgia, southeastern South Carolina (Murray, 1961; Mills and others, 2020; DAS80 on fig. 3B and in Sweetkind and others, 2024), and on the northern Atlantic Coastal Plain (Pope and others, 2016; DAS11 on fig. 3B and in Sweetkind and others, 2024).

Deformed and metamorphosed basement rocks of the Appalachian chain are disrupted by Late Triassic to Early Jurassic rift basins, which contain locally thick accumulations of continental siliciclastic, lacustrine, and basaltic rocks. The altitude of the underlying basement may be as much as several kilometers deeper than similar rocks on the margins of the rift basins (Manspeizer and others, 1989; Schlische, 1993; Withjack and others, 2012). Faults and geologic polygons from the geologic compilation of Horton and others (2017) were used to locate the basins in map space; basement rock altitudes from the basement rock map of the United States (Bayley and Muehlberger, 1968; DAS08 on fig. 3B and in Sweetkind and others, 2024) were digitized to initially construct basement depth with the rift basins. Data from geologic cross sections and seismic-reflection profiles were used to refine basin depth and polarity (Manspeizer and others, 1989; Schlische, 1993; Withjack and others, 2012; DAS76 and DAS77 on fig. 3B and in Sweetkind and others, 2024). Basement rock altitudes within the South Georgia Rift basins were derived from a contoured figure in Heffner (2013; DAS78 on fig. 3B and in Sweetkind and others, 2024).

Use of Geologic Maps to Add the Top of Basement from Outcrops in the Western United States

Basement maps of the conterminous United States include Mesozoic and Cenozoic batholithic intrusive bodies of the Western United States cordillera. Basement Rock Project Committee (1967) map these rocks as “acid and intermediate plutonic rocks”; Bayley and Muehlberger (1968) subdivided major intrusive rock bodies into broad compositional and age classes, and, using color tones on the map, distinguishing between exposed rocks and subsurface rocks that were known or inferred from drill holes or projection of outcrop.

Major intrusive rock bodies in the Western United States were selected using the lithologic descriptor attributes in the Horton and others (2017) digital geologic map dataset. Igneous rock names and lithologic descriptors, such as granite, granodiorite, tonalite, quartz diorite, quartz monzonite, and others, were selected from the attribute table of the geologic polygons (basement type labeled “BSMT5” in fig. 5 and table 2). Smaller intrusive bodies, generally smaller than 250 square kilometers in area, were removed from the selection set; these small intrusive bodies were not shown on previous national-scale basement maps and are considered local intrusions that are not representative of the regional basement. The top of basement altitude was then calculated for each 2.5-km cell using land surface elevation derived from a DEM (basement type labeled “BSMT5” in fig. 5 and table 2).

In the Pacific Northwest, national bedrock geologic maps were used to make selection sets of map polygons of geologic units that could be classified as “basement,” based on a combination of age and principal lithology (King and Beikman, 1974; Schruben and others, 1998; Horton and others, 2017). Selections included highly deformed rocks (basement type labeled “BSMT6” in fig. 5 and table 2) and structurally intermingled rocks of various types, typically with some igneous or metamorphic component (basement type labeled “BSMT7” in fig. 5 and table 2).

Treatment of the Top of Basement in California

Basement maps of the conterminous United States contour the altitude of crystalline rocks in the subsurface beneath California’s Central Valley. Basement Rock Project Committee (1967) contoured the top of basement in the subsurface for the eastern half of the Central Valley, describing the mapped horizon only as including various ages of crystalline rocks. Bayley and Muehlberger (1968) showed similar subsurface contours beneath California’s Central Valley, mapping crystalline rocks in the subsurface as Paleozoic and Mesozoic metasedimentary and volcanic rocks. Frezon and others (1983) contoured the total thickness of sedimentary rocks in the Central Valley, and although the map did not include an explanation of what rocks were being contoured, the total thickness of sedimentary rocks contoured by Frezon should be equivalent to the depth to the top of basement rocks in the Central Valley.

The altitude of crystalline rocks in the subsurface of California’s Central Valley has also been contoured on regional-scale maps. Harwood and Helley (1987) contoured the top of basement rocks beneath Sacramento Valley (the northern half of the Central Valley) based on the presence of metamorphic and plutonic rocks identified in well intercepts. Basement units are described as predating the Cretaceous Great Valley sequence and consisting of metamorphic and plutonic rocks, including granitic plutons of the Sierran basement, undivided Mesozoic and Paleozoic rocks of the Sierran foothills metamorphic belt, and the Coast Range ophiolite (Harwood and Helley, 1987). Wentworth and others (1995; DAS62 on fig. 3B and in Sweetkind and others, 2024) contoured the altitude of the top of crystalline basement for a large part of central California, including the entire Central Valley. The study defined a steadily westward-dipping basement surface that was based on three sets of data: in the Sierra Nevada from the elevation of exposed granitic rocks at land surface, beneath the eastern two thirds of the Central Valley from oil and gas well data, and beneath the western part of the Central Valley from seismic-reflection and refraction profiles (Wentworth and others, 1995; DAS62 on fig. 3B and in Sweetkind and others, 2024).

For this study, the digital contour lines of Wentworth and others (1995) were used to define the altitude of crystalline rocks (DAS62 on fig. 3B). Contours were projected, attributed with altitude values in meters, and a raster surface created using a topo-to-raster tool in a GIS. Raster values were applied to those 2.5-km cells not already populated using geologic map polygon data from the State Geologic Map Compilation geodatabase (Horton and others, 2017). No attempt was made to expand the contoured data of Wentworth and others (1995) to other parts of California.

Thickness of the Bedrock Layer

The modeled thickness of the bedrock layer was generally calculated as the distance between the top of bedrock and the top of basement (fig. 10). In areas where bedrock has been removed by erosion and basement rocks either crop out at land surface or directly underlie unconsolidated deposits, thickness of bedrock was assigned as zero thickness. In areas of the Western United States where the top of basement rocks was not mapped, the thickness of the bedrock model unit could not be calculated.