Table 2.
Sensitivity of curvature-thickness of mobile regolith (TMR) relationship to resolution of digital elevation model (DEM) for the Johnston Draw data set (N = 38)
| DEM resolution (m2) | Catchment curvature standard deviation (m−1) | TMR uncertainty (m) | Curvature uncertainty (m−1) | Slope (m2) | Intercept (m) | RMSE (m) | r 2 | p-Value |
|---|---|---|---|---|---|---|---|---|
| 1 | 0.126 | 0.13 | 0.1521 | 0.41 | 1.05 | 0.54 | 0.02 | <0.0001 |
| 3 | 0.036 | 0.13 | 0.0169 | 22.80 | 1.01 | 0.20 | 0.86 | <0.0001 |
| 5 | 0.021 | 0.13 | 0.0061 | 21.58 | 1.04 | 0.40 | 0.44 | <0.0001 |
| 10 | 0.018 | 0.13 | 0.0015 | 20.56 | 1.00 | 0.45 | 0.30 | 0.0004 |
| 20 | 0.011 | 0.13 | 0.0004 | 24.76 | 0.98 | 0.48 | 0.21 | 0.0039 |
| 30 | 0.007 | 0.13 | 0.0002 | 37.90 | 0.96 | 0.47 | 0.25 | 0.0013 |
| 50 | 0.006 | 0.13 | 0.0001 | 59.73 | 0.95 | 0.47 | 0.23 | 0.0022 |
TMR uncertainty reported as standard error is based on propagation of error of the average observed TMR (see Methods). Horizontal and vertical uncertainty in DEM were obtained through metadata associated with 2007 Light and Ranging (LiDAR) dataset. Curvature uncertainty as measured by standard error was calculated by the Method of Moments assuming correlation between uncertainty of neighbor and center cell points are 0 (r = 0). Slope of curvature-TMR, intercept value, root-mean-squared error (RMSE), coefficient of determination (r2), and p-value based on linear regression