Ecological environment quality assessment using remote sensing models and a machine learning approach for land consolidation

Kanchanamala, P
Venu Madhav, T
Kachancheeri, Muhammed Shameem
Sutar, Ajim Shabbir
Muthukumaran, N
Kannan, S.P.M.
Gupta, Rupesh

Journal of Degraded and Mining Lands Management
Vol. 12 No. 5 (2025)

The development of land consolidation (LC) in Kochi leads to ecological alterations, requiring effective methods to assess ecological environmental quality (EEQ). This study introduced a remote sensing (RS)-based framework using pre-, during-, and post-LC data. The Remote Sensing Ecological Index (RSEI), incorporating four factors wetness (LSM), greenness (NDVI), heat (LST), and dryness (NDBSI)—was applied to evaluate EEQ variations. Land Use Land Cover (LULC) was derived using Random Forest classification to highlight changes during Kochi’s smart city project. RSEI values range from 0 (low EEQ) to 1 (high EEQ). The results showed that EEQ was good before LC, declined during LC, and significantly dropped after LC, with areas such as Pazhangad, Fort Kochi, and Mattancherry experiencing severe degradation. Recovery of EEQ took over a decade. This approach offers a reliable, scientific method for long-term EEQ monitoring at the project scale and supports sustainable planning in ecologically sensitive urban developments.