Géoinformatique et géostatistique : un aperçu

Classification of Land-Use/Land-Cover in Kaduna Metropolis, Kaduna State, Nigeria

Nantip Thomas Goselle*, Obot Akpan Ibanga and Peter AO Odjugo

This study aimed at classifying the various land use/land cover in the Kaduna metropolis of Nigeria. This study relied primarily on remotely sensed data which were acquired using landsat 5 thematic mapper, landsat 7 enhanced thematic mapper and landsat 8 operational land imager thermal infrared sensors for 1990, 2001, 2010 and 2018. Forest, grass/sparse vegetation, water and built-up/bare surfaces were the four land-use/land-cover types classified. The overall image classification accuracy of the land use and land cover was 95% for 2018 image with Kappa Coefficients (KC) of 0.96, 91% (KC=0.89) for 2010, 89% (KC=0.96) for 2001 and 91% (KC=0.91) for 1990. The fact that both the overall Image Classification Accuracy (ICA) and Kappa Coefficients (KC) were above 85% and 0.85 showed that all the remotely-sense products were satisfactorily classified. A gradual increase in built up/bare surface land use/land cover class from 5.38% of Kaduna metropolis in 1990 to 15.12% in 2018 and a decline in forest land use/land cover class from 76.08% in 1990 to 47.88% in 2018. This understanding of the dynamics observed in the land use/land cover pattern can be instrumental in reducing the negative effects of urban infrastructure development and practices on the climate of the Kaduna metropolis with improve legislative approach, thereby improving living conditions.