Application of Normalized Differential Vegetation Index and Geostatistical Techniques in Cattle Diet Quality Mapping in Ghana

Abstract

The feasibility of using Normalized Differential Vegetation Index (NDVI) and geostatistical techniques in the mapping of livestock diet quality was investigated using the northern Ghana savanna zone as a test bed. Near infrared reflectance spectroscopy (NIRS), crude protein (CP) and digestible organic matter (DOM) calibration equations developed for Sub-Saharan African cattle were used for the analysis of faecal samples collected between February and August 2000 from selected households located within the northern Ghana savanna. The predicted CP and DOM values (%) were used with Normalized Differential Vegetation Index (NDVI) data and cokriging procedures to create maps describing diet quality for March and July 2000, representing the dry and wet seasons respectively. Cross-validation results indicated an above average capability of cokriging to estimate the predicted CP% for March (r² = 0.687, SEp = 1.736), but average for July (r² = 0.513, SEp = 1.558). The cokriged-estimated DOM value for July was above average (r² = 0.584, SEp = 3.611); however, March DOM% estimation was rather poor (r² = 0.132, SEp = 3.891). The results suggested that a reasonable level of success was achieved in using NDVI and geostatistics techniques in cattle diet quality mapping; however, further research is required in this field.