The measurement of environmental variables has become a daily problem in recent years. However, the equipment commonly used for these measurements is expensive and bulky, and therefore, it is not possible to have enough spatial resolution. In addition, many of the measurement methods do not provide real-time information to deliver to citizens in a timely manner. In some works, these issues have been handled through the deployment of wireless sensor networks based on low cost technologies. The improvement of the spatial and temporal resolution implies the increase in the amount of information to be transmitted and stored. For this reason, this paper presented a method for data reduction through a dynamic subsampling of the measured variable, data fusion from several sensors for the same variable, and data scaling taking into account the variables range. The reduction of data is implemented to save energy, reduce the transmission time, keep the channel available, and save storage space. The method is validated using a low-cost monitoring station that combines environmental, particulate matter (PM), gas, electromagnetic radiation, and inertial sensors to be transmitted in a 50-byte reduced packet using an LoRa network. The subsampling adjustment was developed for the PM signal. The results show a reduction in the volume of stored data and the relevant information is not affected. The transmitted data packet can be reduced from 96 to 50 bytes, and sampling can be reduced to 4% of the original sampling without affecting the trend of the PM information.