Abstract
The objective of this study was to test the efficiency of the Hydraulic Pedotransfer Functions (PTFs) employed in the Decision Support System for Agrotechnology Transfer – Crop Simulation Model (DSSAT-CSM) in modeling topsoil WHC in Northern Guinea Savanna (NGS) and Sudan Savanna (SS) of Kano State in Nigeria. Coefficient of determination (R2), Root Mean Squared Error (RMSE), and Index of Agreement (d-index) were the three statistical methods used to test the fitness between predicted, and laboratory observed WHC of disturbed, auger sampled topsoil. Findings of the study established that the PTFs fitted in the algorithm of DSSAT-CSM soil water sub module made a significant topsoil WHC estimation in NGS with statistics R² = 0.352, RMSE = 0.03, and d-Index = 0.71. However, the model did not estimate the WHC validly in Sudan Savanna, with insignificant statistics of R² = 0.031, RMSE of 0.10, and 0.44 as the index of agreement. The conclusion drawn was that DSSAT made fair and poor predictions of topsoil WHC in NGS and SS soils respectively, irrespective of texture and other intrinsic properties. Based on the findings above, we recommend the development of local PTFs alternatives to be used with DSSAT’s algorithm for Nigerian Savanna soils.
Keywords:
(rmse)
and index of agreement (d-index) were the three statistical methods
Full Text
The full text of this article is available as a PDF
You can download the PDF version of this article for easier reading and printing.
Download Full Text PDF