Kayode Adepoju

We formalize Triangle-based Geometric Semantic Modeling (TGSM), a novel approach to time-series analysis that encodes temporal transitions as geometric primitives. Consecutive observations are mapped to right-angled triangles, linking time span and change magnitude to a unified measure of movement strength. By organizing these primitives into directional classes and aggregating them across scales, TGSM provides a transparent bridge between raw signals and semantic structure. This framework offers a new lens for interpreting dynamic behavior and establishes an audit-ready foundation for semantic compression, model transparency, and robust feature design in explainable AI. The paper highlights the formulation, key derivations, and potential applications of this approach.

Accurate information on the distribution of invasive native species could provide important and effective procedures for managing savannah environment, especially in sensitive mountainous grasslands. The study detected and mapped Seriphium plumosum within a mountainous landscape and linked the georeferenced occurrence data with the corresponding site-specific environmental factors to predict the locations of unknown populations using a MaxEnt niche model. We also explored the relative contribution in terms of species interaction with its surrounding biophysical environment. The AUC value of 0.876 estimated for the species distribution is an indication of a good model fit. Our findings indicated that Seriphium plumosum preferred areas with higher temperature associated with recurrence fire events and limited soil moisture. It was concluded that the projected conditions of increasing temperature and fire events could promote widespread gain of niche space for Seriphium plumosum while at the same time altering community structure and composition, hydrological properties, and other vital ecosystem services in the study area.

The influence of global change on vegetation cover and processes has drawn increasing attention in the past few decades. In this study, we used remotely sensed rainfall and land surface temperature to investigate the spatiotemporal pattern and trend in vegetation condition using NDVI as proxy from 2001 to 2017 in a humid and dry tropical region. We also determined the partial correlation coefficient of temperature and rainfall with NDVI and the response of NDVI to changes in landcover categories due to human activities. We found that the mean annual maximum NDVI was 0.42, decreasing at a rate of 0.06 per decade. About 27.4% of the area was found to have experienced a significant negative trend in vegetation cover, while only 0.34% exhibited significant increasing vegetation vigour. Land surface temperature increased at a mean rate of 0.75°C/decade, with higher rates in agriculture, savanna, settlements, woodlands, and riparian vegetation than in forest and mangrove vegetations. Precipitation also reduced at a mean rate of 58.69 mm/decade, with higher rates in agriculture savanna and riparian vegetation than in sahelian grasslands, mangrove, forest, and woodlands. NDVI was negatively correlated with temperature in savanna, settlements, degraded forest, and sahelian grasslands providing confirmation of ongoing land degradation. It was concluded that vegetation vigour will continue to decline under rainfall and increasing temperature conditions especially in dryer regions. The use of land surface temperature in this study is particularly valuable in highlighting areas where changes in NDVI occurred as a result of synergistic action between climate and human-induced landcover changes. Our findings underscore the importance of landuse policies that account for spatial variation in synergistic relationships between the nexus of climate and land conversion processes that influence vegetation cover change in different landcover types in tropical regions.

Land use and land cover (LULC) mapping is very important for evaluation and management of natural resources. Classification of remote sensing data in mountainous and other heterogeneous environments are particularly problematic because of their high spectral heterogeneity. The aim of this study is to conduct accuracy analyses of LULC classification of Landsat-8 OLI composite images and multisource ancillary data using four machine learning classifiers on the Google Earth Engine (GEE) platform in the Golden Gate Highland National Park (GGHNP) and its near surrounding. Our results indicate that the inclusion of short-wave infrared bands increased accuracy of image composites by more than 10% in landcover types with wider range of spectral characteristics. Best performance was achieved by MaxEnt with overall accuracy and kappa statistics of 0.846 and 0.775, respectively. Support Vector Machine (SVM), however, produces relatively higher accuracy for handling different characteristics of multisource data. In elevated areas, aspect was found to outperform other topographic indices in reducing misclassification errors associated with confusion of grasslands with bare lands and water. The LULC produced from image composites can provide some reliable information on vegetation, land and water characteristics and reduce confusion associated with complex biophysical environments on vegetation spectral signatures in sub-tropical mountainous grasslands.