Modeling soil depth from topographic and land cover attributes

Abstract

[1] Soil depth is an important input parameter in hydrological and ecological modeling. Presently, the soil depth data available in national soil databases (STATSGO and SSURGO) from the Natural Resources Conservation Service are provided as averages within generalized land units (map units). Spatial uncertainty within these units limits their applicability for distributed modeling in complex terrain. This work reports statistical models for prediction of soil depth in a semiarid mountainous watershed that are based upon the relationship between soil depth and topographic and land cover attributes. Soil depth was surveyed by driving a rod into the ground until refusal at locations selected to represent the topographic and land cover variation in the Dry Creek Experimental Watershed near Boise, Idaho. The soil depth survey consisted of a model calibration set, measured at 819 locations over 8 subwatersheds representing topographic and land cover variability and a model testing set, measured at 130 more broadly distributed locations in the watershed. Many model input variables were developed for regression to the field data. Topographic attributes were derived from a digital elevation model. Land cover attributes were derived from Landsat remote sensing images and high-resolution aerial photographs. Generalized additive and random forests models were developed to predict soil depth over the watershed. They were able to explain about 50% of the soil depth spatial variation, which is an important improvement over the soil depth extracted from the SSURGO national soil database.

Soil depth modelling using terrain analysis and satellite imagery: the case study of Qeshlaq mountainous watershed (Kurdistan, Iran)

“…Mean error in both model calibration and model test of RF model is less than with the other two models. This is in line with the result of the study by Tesfa et al (2009). Statistical prediction models were applied to predict soil depth over a mountainous watershed using different environmental variables derived from DEM and satellite image.…”

“…These indices describe the spatial variability Article of specific processes, which are occurring in the watershed, such as soil water content or the potential for erosion (Moore et al, 1993). Some of the secondary categories used in this study are stream power index (Florinsky et al, 2002), wetness index and sediment transport index (Moore et al, 1993;Wilson and Gallant, 2000), the D8 (deterministic eight-node) algorithm and its derivatives (D8 contributing area, D8 slope, D8 distance to stream, D8 longest upslope length, D8 total upslope length, D8 slope averaged and D8 flow direction grid), and finally D∞ algorithm and its derivatives (D∞ flow direction and D∞ slope) (Tarboton, 1997;Tesfa et al, 2009;Hass, 2010) (Table 1).…”

“…Principal component analysis (pca) was used to identify orthogonal components from the Landsat8 input bands that explain significant variance (Pearson, 1901). Different vegetation indices such as soil adjust vegetation index (SAVI) (Huete, 1988), normalised difference vegetation index (NDVI) (Rouse et al, 1973), soil moisture index (SMI) (Lingli and Qu, 2009), vegetation index (VI) and canopy cover (CC) (Tesfa et al, 2009), were used and analysed to reveal the relationship between these indices and soil depth. The Tasseled Cap transformation (tc) was used to convert the Landsat8 bands into three components (tc1, tc2, tc3) designated as brightness, greenness and wetness (Tesfa et al, 2009) (Table 2).…”

“…Statistical approach is one of the most important methods for predicting soil depth using different environmental variables. Tesfa et al (2009) applied two statistical prediction models, random forest (RF) and general additive model to identify the relationship between soil depth and landscape topographic variables derived from a digital elevation model (DEM) and land cover attributes extracted from satellite imagery. Results showed that these models could explain about 50% of the observed soil depth variability in an out of sample data on study area.…”

Implementing and Evaluating Variable Soil Thickness in the Community Land Model, Version 4.5 (CLM4.5)

“…Instead, localized estimates have been more prevalent. For instance, Dietrich et al (1995), Roering (2008), Rasmussen (2009), andTesfa et al (2009) developed geomorphically based soil depth models, applying them to upland basins.…”