Theses and Dissertations


Laura Wilson

Issuing Body

Mississippi State University


Ramirez-Avila, John J.

Committee Member

Martin, James L.

Committee Member

Truax, Dennis D.

Date of Degree


Document Type

Graduate Thesis - Open Access


Civil Engineering

Degree Name

Master of Science (M.S.)


James Worth Bagley College of Engineering


Department of Civil and Environmental Engineering


Tillage practices on agricultural fields have an impact on not only the amount of soil erosion from the fields, but also on the hydrologic and other environmental characteristics of the land. This erosion takes away soil that is necessary for sustainable agriculture, and the sediment and nutrient removal from the fields can pollute surrounding waterbodies. The Llanos Orientales of Colombia used to be a region of extended savannas and native fragile ecosystems dedicated to extended cattle ranch that has been transitioning to crop production. Agricultural expansion in this area, involving mechanization, could importantly accelerate the degradation of soils, limiting the development of sustainable agricultural systems. As a first step to understand long term effects of different tillage practices on new agricultural areas in the region, this study aims to evaluate the performance of the Agricultural Policy Environmental eXtender (APEX) model to simulate runoff, soil erosion and crop yield from fields under conventional tillage, reduced tillage, and no tillage in the Llanos Orientales of Colombia. Calibrated APEX model predictions were compared against measured runoff, soil loss and crop yield data from row crop plots established in the Experimental Station la Libertad in Colombia under conventional, reduced and no-tillage management. APEX satisfactorily predicted runoff (Nash Sutcliffe Efficiency NSE>0.53, Percent Bias - [PBIAS] < 21%) and crop yield for all three tillage systems (NSE>0.82, [PBIAS] <15%), but was not successful in predicting soil loss from the studied plots. Unsuccessful results were related to model limitations to predict erosion (USLE equations), but also to any uncertainty attributed to issues in the data collection. A calibrated APEX model could be used to predict runoff and crop yield responses under different management practices in the Llanos Orientales of Colombia, but needs improvements for prediction of soil erosion in tropical soils.