Abstract

The availability of multi-resolution spatial data and advances in modeling techniques have given an impetus to land use land cover (LULC) change analyses. Geo-visualization of possible land uses (LU) with policy decisions is vital for formulating appropriate sustainable resource management policies. For the prudent management of natural resources, LU planning has to take environmental dimensions into account. LU dynamics helps to understand the macro background of regional population growth, economic development, social progress, and changes in the natural environment. In this study, LU transitions from 1985 to 2019 were assessed through a supervised classifier based on the Gaussian maximum likelihood estimation algorithm. Geo-visualisation of landscape dynamics was implemented through a fuzzy analytical hierarchy process (AHP) with Markov cellular automata (MCA) for Karnataka state, India. It considered five policy scenarios, namely: (i) business as usual (BAU), (ii) agent-based land use transition (ALT), (iii) reserve forest protection (RFP), (iv) afforestation (AF), and (v) sustainable development plan (SDP). Prior knowledge of likely LU aids in assessing the implications of chosen policies forms a base for sustainable resource management with conservation of biological diversity. LU analyses revealed that forests in Karnataka state constituted 21% in 1985, witnessed large-scale transitions, and reduced to 15% of the geographical area in 2019. BAU depicts a likely increase in the built-up area to 11.5% from 3% (2019). The SDP scenario (with stringent policy implementation) indicates that the forest cover would remain at 11% (compared to 15% in 2019), which is the least possible loss among all considered scenarios (BAU, ALT, RFP, AF, and SDP). Modeling and visualization of landscape dynamics aids in regional LU planning as a spatial decision support system (SDSS) towards achieving sustainable development goals.

Keywords: Modeling; land use dynamics; scenarios; sustainable development; policy