Short-term predictions of land use fragmentation in Panchnoi River Basin, Assam, India using artificial neural networks

Kalita, Nilotpal and Bhattacharjee, Niranjan and Nath, Manash Jyoti (2024) Short-term predictions of land use fragmentation in Panchnoi River Basin, Assam, India using artificial neural networks. ECOCYCLES, 10 (2). pp. 70-79. ISSN 24162140

Preview

Text Kalita_GALLEY.pdf - Published Version Available under License Creative Commons Attribution. Download (654kB) | Preview

Abstract

Land use and land cover monitoring, management and prediction are critical aspects of overseeing specific Earth’s surface segments or river basins. These changes often influence man and environment relationships. Escalating population pressure and demand for land resources induce substantial alterations which are predominantly driven by human activities. The 'Panchnoi' river basin exemplifies similar changes. This study evaluates the rate of land use fragmentation and land cover changes utilizing satellite imagery and GIS. Different key metrics, such as the number of patches (NP) and mean patch size (MPS), reveal insights into land fragmentation patterns. The impact is evident in the rising NP and declining MPS, indicating significant fragmentation in the Panchnoi River Basin between 2008 and 2019 across multiple land-use and land-cover categories. Anticipating future changes, a Land Use and Land Cover Prediction Map for 2026, generated using MOLUSCE Plugins (specifically, ANN-Multi Layer Perception) within QGIS Version 2.8.2, aims to predict shifts and discern the evolution in land use patterns from 2008 to 2019 and into 2026. The ANN-Multi Layer Perception is a quick way of projecting future transitions of land use and land cover of the Panchnoi River basin. As the forest and grassland areas are declining rapidly, such type of prediction will help policymakers for effective management of ecosystems of the river basin. The grassland area is projected to experience the most significant reduction, decrease from 6421.860 hectares in 2008 to just 0.0224 hectares by 2026 over a span of 18 years. The study not only focuses on land use fragmentation in the Panchnoi River basin but also predicts the changing trend of land use using the Artificial Neural Network (CA-ANN) technique.

Actions (login required)

Edit Item