Use temperature sensors to anticipate weather-triggered illness

Sharing a real setup with Aqara that’s been genuinely helpful at home. I run a small ecosystem with Aqara sensors/cams plus Philips Hue and a few other smart lights. The project is simple, but it’s made a real difference for my partner.

Context
My partner wasn’t just uncomfortable; she was getting truly sick under certain weather conditions—dizziness, nausea, vomiting, and general malaise. Looking back through the sensor history, a consistent pattern emerged:

  • Temperature trending upward within a short window
  • Relative humidity at or above 55%
  • A sudden drop in barometric pressure around the same time

Identifying that weather pattern from the historical data was the turning point. Building an alert around it has been essential to prevent episodes before they escalate.

What I built (step by step)

  1. Collect a few days of data
    I enabled/checked the Aqara temperature and humidity logs to spot trends rather than isolated readings.
  2. Define simple thresholds
  • Temperature “rise”: roughly +0.5 °C within 60 minutes
  • Humidity “high”: ≥ 55%
  • Pressure: track a downward trend (a few hPa over a couple of hours) using history/readouts to validate the pattern
  1. Create a combined alert
    I set an automation that notifies me when temperature is rising and humidity is ≥ 55%. When that alert fires, I also check the pressure trend, since that’s part of the trigger pattern we observed.
  2. Act before symptoms start
    When the alert triggers, we hydrate, ventilate, lower blinds, or switch on a dehumidifier if needed. I also change a Hue light to amber as a visual cue and send a push notification with the current readings.

Calibration tips

  • Seasonal tuning: in summer, humidity thresholds may be a bit higher (around 58–60%); in winter, a bit lower (50–55%).
  • Avoid false positives: require conditions to hold for 10–15 minutes before alerting.
  • Quiet hours: limit alerts to daytime (for example, 06:00–22:00).
  • Quick log: keep a simple note (date, values, symptoms rated 1–5). After a week or two, refine thresholds based on what actually correlates with symptoms.

Simple if/then example
If temp has risen ≥ 0.5 °C in 60 minutes and humidity ≥ 55% (and ideally you’re seeing pressure trending down), then send a notification and run a comfort scene (amber light, ventilation, hydrate reminder).

Results
Since using this, we’ve reduced the severe episodes by acting earlier—before the room conditions drift into a trigger zone. It’s lightweight, repeatable, and easy to tailor to your climate and home.

15 Likes

@valdebenito Cool :+1:t2: , thanks for telling us how to avoid bad weather. This is really important, because being sick is sad and not easy.

3 Likes

How are your Aqara automations structured to apply this rule? “risen ≥ 0.5 °C in 60 minutes” ? Are you doing cascading IF checks that store to an intermediate variable so you can do delays somehow?

Also, how do you encode the “require conditions to hold for X minutes” hysteresis in Aqara automation engine?

2 Likes

Thank you for sharing. This is quite an interesting read. My wife also suffers from allergies that make her miserable. I have been noticing some correlations between temperature and humidity and her symptoms. Your setup inspires me to see what I can do to improve the microclimates in our home and maybe help reduce my wife’s symptoms through automation as you have done. Again thank you.

1 Like

@AqaraOfficial

The logic described by the user is not currently possible in Aqara automation, don’t you agree?

I think the user must have used home assistant to program conditions such as “risen ≥ 0.5 °C in 60 minutes” and “Avoid false positives: require conditions to hold for 10–15 minutes before alerting”

Or was this user using the new Automations 2.0 feature?

No, this case can’t be done with Automation 2.0. Valdebenito made use of another platform that supports more advanced automations — and it’s such a clever idea! Hopefully, it can bring some inspiration to others who are also struggling with the weather.

2 Likes