Analyzing Dust Particle Patterns in Animal Stables

This project aimed to use the functionalities of the Healthy Climate Monitor, a hardware device designed for integration into animal stables, to improve air quality in the animal stables. Dust accumulation in these environments presents significant health hazards for both humans and animals. Humans exposed to stable dust are at risk of developing respiratory issues, suffering allergic reactions, and encountering various other health complications. Animals, too, face numerous health challenges due to this dust; however, their shorter lifespans make the long-term impacts more acutely detrimental to humans.

The Healthy Climate Monitor operates continuously, employing a suite of sensors to gather data round-the-clock. These sensors are adept at measuring a range of environmental factors, such as temperature, relative air humidity, CO2 levels, dust particles, and many more variables. A key focus of this research is on the measurement of dust particles. Our aim is to analyze the dust particle data in conjunction with other environmental readings. By doing so, we seek to identify recurring patterns that either contribute to an increase or a decrease in dust particle concentration. For instance, if one opens a door in a stable, it could lead to a drop in temperature (assuming it's colder outside) and a reduction in CO2 levels inside the stable. However, it's uncertain whether this decrease in temperature and CO2 impacts the dust particles - whether it increases, decreases, or has no effect on them. This study is limited to the patterns in the measurements, and our conclusions will be used to subsequently associate these patterns with possible events.

To analyze patterns, we quickly turned to a correlation study. If an increase in temperature causes an increase in dust, this would result in a positive correlation. If an increase in temperature leads to a decrease in dust, this would be a negative correlation. The stronger the movement of one measurement affects the dust measurement, the higher the correlation value will be.

The continuous 24/7 collection of measurements results in a time-series data set. The simple comparison of measurement values to assess correlations might obscure underlying patterns, particularly those involving temporal dynamics. It's crucial to consider the temporal element in these patterns. For instance, a scenario where an elevation in CO2 levels leads to an increase in dust particles after an hour could remain undetected if the time factor is overlooked. To address this, we employed Time Lagged Cross Correlation analysis. This method not only reveals the correlation between two different measurements but also identifies the specific time lag at which this correlation is most pronounced. Understanding which measurements most substantially affect dust particle levels, coupled with the knowledge of the optimal time lag for the strongest impact, equips us with the tools to investigate real-life events occurring in these environments that might influence fine dust levels, and so improve animal and human welfare.