Research Article
Using Spatial Regression to Model Potentially Toxic Metal (PTM) Mobility Based on Physicochemical Soil Properties
Table 4
Spatial multivariate regression results.
| Diagnostic | Zn | Mn | Cu | Pb | Cd |
| Distance threshold (m) | 150 | 300 | 200 | 150 | 150 | Best spatial model | Spatial error | Spatial error | Spatial error | Spatial error | Spatial lag | R2 | OLS | 0.371 | 0.162 | 0.366 | 0.515 | 0.558 | SLag | 0.384 | 0.186 | 0.369 | 0.615 | 0.617 | SError | 0.400 | 0.212 | 0.369 | 0.694 | 0.587 | AIC | OLS | 762.255 | 827.317 | 406.895 | 779.265 | 290.25 | SLag | 763.487 | 828.315 | 408.711 | 772.458 | 286.875 | SError | 760.706 | 825.638 | 406.729 | 762.513 | 288.248 | | Soil properties retained | pH (+) CEC (−) | Moisture content (−) % silt (−) | % sand (−) % silt (−) BD (−) | pH (+) CEC (+) λ (−) | pH (+) % sand (−) % silt (−) Wy (−) |
|
|
λ is a spatial autoregressive error term; Wy is a spatially lagged dependent variable; and (+) and (−) indicate the sign of the coefficient.
|