Understanding bush encroachment in Bisley Valley Nature Reserve: the use of intensity analysis and cellular automata model on RapidEye and PlanetScope image data.

Abstract

The encroachment of woody vegetation into grasslands is a worldwide phenomenon that progressively causes serious disturbances to plant and animal biodiversity, ecosystem function, and recreational activities. This phenomenon has become a major concern in nature reserves due to its adverse impact on woody vegetation and grass ratio causing various disturbances such as decline of palatable grasslands and effects on soil properties. Hence, a comprehensive analysis of bush encroachment is paramount for understanding the past, present and future distribution of the encroachment and to inform effective management schemes. This study assessed and predicted bush encroachment dynamics in Bisley Valley Nature Reserve using high spatial resolution remotely sensed multi-temporal image data. The first objective focused on monitoring bush encroachment and associated land use-land cover types in the nature reserve using RapidEye and PlanetScope data within the Google Earth Engine (GEE) platform. Using RapidEye and PlanetScope image data spanning the period from 2009 to 2023, the study estimated the changing extent of woody vegetation, grassland cover and bare areas, providing a comprehensive analysis of their dynamics over the 14- year study period. Over the study period, results show that approximately 130.69 ha (ha) of grassland was converted to woody vegetation, while approximately 2.78 ha of woody vegetation was transformed into grassland. Moreover, the study established a net increase of 127.91 ha in the total area covered by woody vegetation. The second objective sought to compute and analyze past, current, and future (2009-2033) bush encroachment trends and intensity of land cover transitions using intensity analysis and the Cellular Automata (CA) models. The findings revealed a steady increase in woody encroachment on other land cover types. Moreover, there was an intensive change of land cover in the first period (2009-2014) compared to the other periods. Additionally, the prediction of future bush encroachment demonstrated an increasing trend of woody vegetation in the next decade. This study provides valuable insights on the threat of bush encroachment in the study area and demonstrates the value of various approaches such as change detection, intensity analysis and prediction of future encroachment for spatially explicit and detailed analysis of bush encroachment. The study also revealed that there is a pressing need for evaluation and improvement of management schemes in the study area and other encroached landscapes.