**4. Conclusion and Recommendations**

**Figure 7.** Discretized erosion probability maps for the logistic regression analysis of the LiDAR, RTK, and USGS datasets

D8 Intercept -3.08 87 \*\*

D∞ Intercept -7.19 157 \*\*

FD8 Intercept -7.63 200 \*\*

DEMON Intercept -8.23 123 \*\*

**Table 5.** Logistic regression parameters for the USGS dataset using the D8, D∞, FD8, and DEMON flow direction

Topographic Wetness Index 0.351 56 \*\* Length-Slope 0.323 17 \*\* Plan Curvature -4.46 99 \*\*

Topographic Wetness Index 0.829 123 \*\* Length-Slope 0.790 70 \*\* Plan Curvature -2.11 19 \*\*

Topographic Wetness Index 0.827 147 \*\* Length-Slope 1.02 124 \*\* Plan Curvature -0.565 2 ns

Topographic Wetness Index 0.919 96 \*\* Length-Slope 0.895 72 \*\* Plan Curvature -5.69 60 \*\*

**Parameter Estimate**

**Wald Chi Square**

**Variable**

using the D∞ flow direction algorithm.

58 Research on Soil Erosion Soil Erosion

models.

**Flow Direction Method**

> The findings of this study indicate that LiDAR data can be used to clearly identify eroded features in agricultural landscapes with a level of accuracy that is similar to RTK GPS and better than USGS DEMs. It is critical that LiDAR data are smoothed prior to modeling ero‐

sion channels. Smoothing removes artifacts resulting from differences in plant residue heights perpendicular to the direction of travel by farm machinery. These differences can ac‐ tually cause the terrain analysis flow models to incorrectly rout water along the direction of travel. While smoothing produced better results in this chapter, the method of smoothing with ArcGIS TopoToRaster was not efficient and cannot be used over large areas. More work is needed to determine computationally efficient smoothing algorithms for terrain analysis that minimize artifacts.

**Author details**

Tasos Karathanasis2

, Tom Mueller2\*, Eduardo Rienzi2

\*Address all correspondence to: mueller@uky.edu

3 Univ. Estadual Paulista (UNESP), Jaboticabal, Brazil

United States: Gov. Print. Office.

*the ASABE*, 31(4), 1098-1107.

*vation*, 60(6), 363-370.

280-288.

1 Photo Science, Lexington, KY, USA

and Marcos Rodrigues3

2 Department of Plant and Soil Sciences, University of Kentucky, Lexington, KY, USA

[1] Thorne, C. R., Zezenbergen, L. W., Grissinger, E. H., & Murphey, J. B. (1986). Ephem‐ eral Gullies as Sources of Sediment. *In: Proceeding of the 4th Federal Interagency Sedi‐ mentation Conference*, 24-27 March 1986, Las Vegas, United States. Washington, DC,

[2] Moore, I. D., Burch, G. J., & Mackenzie, D. H. (1988). Topographic Effects on the Dis‐ tribution of Surface Soil Water and the Location of Ephemeral Gullies. *Transactions of*

[3] Srivastava, K. P., & Moore, I. D. (1989). Application of Terrain Analysis to Land Re‐ source Investigations of Small Catchments in the Caribbean. *In: Proceeding of the 20th Int. Conf. of the Erosion Control Association*, 15-18 Feb. 1989, Vancouver, BC, Canada.

[4] Berry, J. K., Delgado, J. A., Pierce, F. J., & Khosla, R. (2005). Applying Spatial Analy‐ sis for Precision Conservation across the Landscape. *Journal of Soil and Water Conser‐*

[5] Pike, A. C., Mueller, T. G., Schörgendorfer, A., Shearer, S. A., & Karathanasis, A. D. (2009). Erosion Index Derived from Terrain Attributes using Logistic Regression and

[6] Pike, A. C., Mueller, T. G., Schörgendorfer, A., Luck, J. D., Shearer, S. A., & Karatha‐ nasis, A. D. (2010). Locating Eroded Waterways with United States Geologic Survey

[7] Luck, J. D., Mueller, T. G., Shearer, S. A., & Pike, A. C. (2010). Grassed Waterway Planning Model Evaluated for Agricultural Fields in the Western Kentucky Coal Field Physiographic Region of Kentucky. *Journal of Soil and Water Conservation*, 65(5),

Streamboat Springs: International Erosion Control Association.

Neural Networks. *Agronomy Journal*, 101(5), 1068-1079.

Elevation Data. *Agronomy Journal*, 102(4), 1269-1273.

, Surendran Neelakantan2

Terrain Analysis for Locating Erosion Channels: Assessing LiDAR Data and Flow Direction Algorithm

, Blazan Mijatovic2

http://dx.doi.org/10.5772/51526

,

61

Adam Pike1

**References**

Conservation planners and GIS analysts should be able to accurately identify erosion fea‐ tures with the D8, D∞, FD8, and DEMON flow direction algorithm. This is important be‐ cause previous work was based on TAPES G which is no longer being supported. Further TauDEM can utilize very large blocks of memory to cover extensive land areas, and can also operate on high performance computers. It is important to note that the choice flow algo‐ rithm will change the model parameters so it is important that they use the correct model. We recommend the TauDEM software program which uses the D8 and D∞ procedures be‐ cause it works on 64 bit machines, allows the use of multiple core processer, and works with DEMs up to 4 GB in size.

All analyses performed in this study were based on with 4-m DEMs. Efforts are necessary to better understand the impact of the scale of terrain models on the quality of erosion model predictions. It may also be possible to expand the inference space of these models by includ‐ ing erosion parameters in the analyses obtained from soil surveys.
