Urban microclimates show a lot of variation in the environment. This makes regional weather forecasts inadequate for determining personal ultraviolet (UV) radiation exposure. To tackle this issue, this paper discusses the design, implementation, and validation of a portable, low-cost Internet of Things (IoT) system for real-time environmental monitoring. The prototype includes a dual-mode feedback system. It gives immediate on-site data through an integrated LCD and allows for remote, long-term analysis using the ThingSpeak cloud platform. We conducted a month-long data collection campaign by deploying five units in different locations within the Dhaka division. This created a detailed dataset that captures data over time and space. The main contribution of this work is a quantitative analysis of this data. It confirms that while meteorological conditions are strong predictors (sunny: +0.39, rainy: -0.40), local micro-environmental features significantly influence personal safety. Importantly, we measure the protective effect of urban tree shade, which has a notable negative correlation (-0.21) with the UV index. The analysis also shows that cloudy weather does not reliably reduce UV radiation. It presents a weak positive correlation (+0.14), suggesting the influence of scattering effects like the broken cloud phenomenon. This is important for public health awareness. This study validates a practical solution for personal environmental monitoring and provides a solid dataset for developing future predictive models based on machine learning.