Authors: Prince Pawar, Sujal Sisodiya, Associate Professor Pradeep Patel
Abstract: Rapid microclimatic fluctuations often evade detection by sparse, conventional meteorological networks, necessitating hyper-local, real-time monitoring solutions. This paper presents an analysis of an 8-hour diurnal environmental dataset (10:00 AM to 5:00 PM) captured via an Internet of Things (IoT)-based monitoring node. The system integrates low-cost sensors to continuously log ambient temperature, relative humidity, air quality (particulate/gas concentrations in ppm), light intensity, and precipitation. The empirical data reveals distinct thermodynamic transitions and strong inter-parameter correlations. Specifically, the dataset captures a textbook meteorological shift: midday solar heating—evidenced by a peak temperature of 34∘C, peak light intensity (100%), and a concurrent relative humidity drop to 45%—followed by a sudden convective afternoon rain shower. The onset of precipitation at 3:00 PM triggered an immediate environmental inversion, characterized by a 3∘C drop in temperature, a sharp moisture surge to 68% relative humidity by 5:00 PM, and a significant reduction in airborne pollutants (from a peak of 180 ppm down to 125 ppm) due to the atmospheric scrubbing effect of the rain. These findings demonstrate that high-frequency IoT sensor networks provide highly reliable, granular data essential for unmasking the velocity and impact of localized weather fronts. The proposed approach offers scalable, actionable insights applicable to urban climate mapping, smart agriculture, and industrial environmental compliance.
