Integrating Satellite Imagery and IoT Field-Based Observations for In-Situ Soil Moisture Analysis

The precision and real-time monitoring of the soil moisture is the key to optimizing agricultural output and sustainable management of water resources. Traditional methods of measuring soil moisture using the gravimetric analysis is accurate but not scalable and remote sensing methods are scalable but not accurate. In this paper, a two-layer model is presented, which combines the soil moisture measurement with the help of the IoT and the Normalized Soil Moisture Index (NSMI) that is calculated using the data obtained by the sentinel-2A satellite. This IoT part of the model employs the Arduino microcontroller capable of linking the soil moisture sensor with the SIM900A GSM module that sends the data to ThingSpeak cloud service. The other variable of the model uses Google Earth Engine application (GEE) which is a cloud computing system that gathers information at the data on the soil moisture map which has been generated using the NSMI of the sentinel-2A satellite images. The comparison between the accuracy of the soil moisture level and the soil moisture level was made using the data obtained with the help of the IoT sensor at the proximity of the Poranki area of the Vijayawada area of the Republic of India and the data provided by the NSMI of sentinel-2A satellite data. The system generated a high degree of accuracy of R2= 0.9888.