Environmental monitoring has gained a lot of traction in the past decade. High-quality data is required to underpin accurate models of prediction over long periods of time. Spatial and temporal granularity are important features of such data. Long-Range Wide Area Network (LoRaWAN) networks are increasingly used for communicating sensor information from the edge nodes to the cloud, ensuring both characteristics. This paper details the experimental procedure for developing a protocol for increased efficiency of a LoRaWAN application by using environmental data to predict the effectiveness of communication at a given point in time. This paper presents a platform for environmental data sensing and derives a LoRaWAN application protocol using environmental data. The main aim of the study is to improve the battery life of the edge device while at the same time improving the quality of service (QoS) in these networks. By taking into account information from the environmental sensors, the amount of power that is used while transmitting is predicted to be reduced by 74.88% while maintaining the QoS. Simulations also show that both the QoS and battery life can be improved by using temporary storage of sensed data in the accurate short-term weather forecast.