The agricultural sector is the largest consumer of the world’s available fresh water resources. With fresh water scarcity increasing worldwide, more efficient use for irrigation water is necessary. Precision irrigation is described as the application of water to meet crop needs of a specific area, at the right amount and at the time that is optimum for crop health and management objectives. Irrigation becomes increasingly efficient through the use of precision irrigation tools. However, to maximize this efficiency, additional technologies can be applied. The main purpose of artificial intelligence (AI) is to learn from past experiences and data to perform an assigned task to solve a particular problem with efficiency and accuracy. AI is becoming pervasive in agriculture due to its ability to solve complex and unique problems. The application of AI in agriculture has the potential to greatly increase efficiency by improving our ability to manage crop inputs. When AI is utilized to implement precision irrigation, the results are economically and environmentally beneficial. Soil moisture plays a key role in crop health and productivity. In the study presented here, an AI model uses automated measurements of precipitation, estimates of evapotranspiration, and surficial soil moisture measured from satellite platforms to estimate daily crop water use (DWU) and provide irrigation scheduling recommendations. Other parameters such as soil temperature, solar radiation, and wind trends are also considered in this analysis. In this study, the machine learning model is trained and validated using in situ data collected from three farmer-managed sweet corn (Zea mays subsp. mays) farms located in Mitchell and Decatur Counties, Georgia, United States of America. Sentek™ soil moisture sensing probes are used to measure volumetric water content (VWC) of the soil. Precipitation, solar radiation, and wind speed is measured with Davis Instruments™ Vantage Pro2 weather stations installed at each field location. This poster presents the results from the first field season.
