Abstract

Researchers have described the difficulty of studying the effects of cultivation on soil properties unless native (uncultivated) areas can be used as a baseline for comparison. Even when suitable areas are located the problems of quantifying and comparing the horizontal and vertical distributions of soil properties remain. Areal distributions of soil surface, subsurface, and upper subsoil properties were compared in cultivated and forested deep loess soils on similar geomorphic surfaces and hillslope positions in two counties adjacent to the Missouri River. A 3-dimensional grid was used to sample selected soils from the summit to the lower backs lope on 3 forested and 3 cultivated hillslopes . Multiple regress ion analysis revealed that a second order response surface equation adequately modeled the depths to maximum clay percentage and thickness of "A" horizon and a third-order response surface adequately modeled thickness of the subsurface horizon and depth to top of "B" horizon, based on the independent variables percent slope, distance from hillslope summit, and site management. Further analysis showed that each site had a significantly different response surface which could be related to site characteristics and current cover (forest or cultivation). The use of analysis of variance within multiple regression allowed for the systematic description and comparison of soil properties on forested and cultivated sites, and for the determination of the relationship between the selected soil properties and slope and site characteristics.

Keywords

hillslope modeling, response surface methodology, soil variability.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Download

Share

COinS
 
Apr 30th, 12:00 PM

MULTIPLE REGRESSION ANALYSIS OF SOIL PROPERTIES ON ERODED AND NATIVE DEEP LOESS MISSOURI SOILS

Researchers have described the difficulty of studying the effects of cultivation on soil properties unless native (uncultivated) areas can be used as a baseline for comparison. Even when suitable areas are located the problems of quantifying and comparing the horizontal and vertical distributions of soil properties remain. Areal distributions of soil surface, subsurface, and upper subsoil properties were compared in cultivated and forested deep loess soils on similar geomorphic surfaces and hillslope positions in two counties adjacent to the Missouri River. A 3-dimensional grid was used to sample selected soils from the summit to the lower backs lope on 3 forested and 3 cultivated hillslopes . Multiple regress ion analysis revealed that a second order response surface equation adequately modeled the depths to maximum clay percentage and thickness of "A" horizon and a third-order response surface adequately modeled thickness of the subsurface horizon and depth to top of "B" horizon, based on the independent variables percent slope, distance from hillslope summit, and site management. Further analysis showed that each site had a significantly different response surface which could be related to site characteristics and current cover (forest or cultivation). The use of analysis of variance within multiple regression allowed for the systematic description and comparison of soil properties on forested and cultivated sites, and for the determination of the relationship between the selected soil properties and slope and site characteristics.