This directory contains some (but not all) of the data files and estimated topography fields described in Lowry, Ribe and Smith (2000), Dynamic elevation of the Cordillera, western United States, J. Geophys. Res., 105, 23,371-23,390.

All files are ascii in a three-column (longitude, latitude, field) format. Included here:

• WUStopo.llz contains the raw topography (derived from an old USGS data set described in Simpson et al. (1986), J. Geophys. Res., 91, 8348-8372). These data were used to estimate effective elastic thickness Te and as raw elevation for the estimate of "dynamic topography".
  
  
• WUSgrav.llz contains Bouguer gravity data (derived from an old USGS data set described in Simpson et al. (1986), J. Geophys. Res., 91, 8348-8372). These data were used to estimate effective elastic thickness Te.
  
  
• WUSq_s.llz contains surface heat flow data (optimally interpolated from Dave Blackwell's compilation of measurements, circa 1998). These data were used to estimate the conductive thermal contribution to elevation and the variable thickness of mechanical lithosphere.
  
  
• WUSTe.llz contains estimates of effective elastic thickness Te, using the methodology described in Lowry and Smith, J. Geophys. Res., 100, 17,947-17,963, 1995. These Te estimates were used to calculate surface load contributions to elevation, for isostatic filtering of estimated internal contributions to topography from crustal mass and conductive thermal variations, and in the estimation of variable thickness of mechanical lithosphere.
  Caveats:
  (1) The estimation windows used here are small (200 to 400 km). Subsequent research on synthetic data has shown that such small estimation windows can yield large variance in Te estimates.
  (2) These estimates used an old approach to load deconvolution that neglected the higher (1000 kg/m^3) density of surficial fluid in ocean regions, which can bias Te toward lower values in oceanic lithosphere.
  (3) The old style of load deconvolution also used an a priori fixed surface-to-internal load ratio of -1 at long spatial wavelengths prone to singularity. This approach can bias toward higher or lower values of Te depending on the relationship of the assumed load ratio to the true average.
  (4) Elevation and Bouguer gravity data grids were extrapolated up to 50 km past the raw data constraints in some parts of the Pacific ocean, which can also introduce errors near the coast.
  
  
• surf_h.llz contains estimates of surface load contributions to total elevation, derived from isostatic deconvolution of the surface and internal loading. This was subtracted from the raw elevation as the first step toward estimating "dynamic topography".
  Caveats:
  (1) Load estimates are very sensitive to assumed Te (see caveats above).
  (2) At long wavelengths prone to singularity (for which load ratios were assumed fixed in original Te estimation), all elevation was assumed to result from internal loading. Subsequent studies have shown that surface loading can have substantial amplitude at long wavelengths (e.g. Lowry and Zhong (2003) J. Geophys. Res., 108(E9), 10.1029/2003JE002111, #5099).
  
  
• crust_h.llz contains estimates of crustal mass contributions to total elevation (incorporating variations in both thickness and density estimated from seismic refraction data). This was also subtracted from raw elevation as a step toward estimating "dynamic topography".
  Caveats:
  (1) Regression of seismic velocity to density has very large uncertainty, resulting in elevation uncertainties of order 600 to 1000 m (one-sigma!).
  (2) The seismic refraction data used to constrain crustal mass are very sparsely sample, have highly variable quality of original data and used various different modeling/inversion approaches to arrive at velocity structure. (This portion of the analysis really should be updated with more recent data). The crustal mass estimate is consequently the largest source of error in the estimate of "dynamic topography".
  
  
• therm_h.llz contains estimates of (mantle lithospheric) conductive thermal contributions to total elevation (estimated from a locally one-dimensional approximation of conductive geothermal variation given surface heat flow and crustal thickness). This was also subtracted from raw elevation as a step toward estimating "dynamic topography".
  Caveats:
  (1) Mantle temperatures depend on poorly-known crustal thermal conductivity and radiogenic heating.
  (2) Modeling assumes the geotherm is in steady-state. Consequently large spurious negative buoyancy is modeled in Cascadia and California (where mining of heat by ancient and/or ongoing subduction yields artificially low estimates of deep temperature). Negative buoyancy may also be overestimated in areas heavily influenced by Pleistocene glaciation, as surface temperatures may not have had time to fully re-equilibrate.
  (3) Geotherm was referenced to a potential temperature but never grafted to an adiabat, which may introduce errors due to variable thickness of the thermal boundary layer. (This was corrected in later versions of the codes but the calculation was not redone).
  
  
• dynamic.llz contains the estimate of dynamic topography after subtracting surface load, crustal mass and mantle conductive thermal mass contributions to raw elevation.
  Caveats:
  (1) Uncertainties in the fields subtracted from raw elevation are large (800+ m at one-sigma, see above) and high-frequency (less than 1000 km wavelength) variations are highly suspect owing to modeling assumptions and data sampling problems.
  (2) This is not true "dynamic elevation" but rather a deep mantle contribution to elevation incorporating BOTH sublithospheric thermal contributions AND mantle contributions relating to variable mineralogy (particularly variable garnet/pyroxene concentrations resulting from melt removal) and variable water content; the latter contribution may be as large or larger than the former.
  
  

This material is based upon work supported by NASA under SENH grant number NAG5-7619. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of NASA.