Application of an unmanned aerial system for monitoring paddy productivity using the GRAMI-rice model

• Published in: International journal of remote sensing Vol. 39; no. 8; pp. 2441 - 2462
• Main Authors: Jeong, Seungtaek, Ko, Jonghan, Choi, Jinsil, Xue, Wei, Yeom, Jong-min
• Format: Journal Article
• Language: English
• Published: London Taylor & Francis 18.04.2018; Taylor & Francis Ltd
• Subjects: Agricultural economics; Agriculture; Computer simulation; Costs; Crop yield; Crops; Data; Detection; Diagnostic systems; Educational institutions; Evaluation; Farm management; Fields; Land use; Modelling; Nitrogen; Oryza sativa; Precision farming; Remote monitoring; Remote sensing; Remote sensing systems; Remote sensing techniques; Rice; Sensing techniques; Vegetation index
• ISSN: 0143-1161; 1366-5901
• DOI: 10.1080/01431161.2018.1425567

Recent developments in unmanned aerial system (UAS) require an urgent introduction to monitoring technologies of crop diagnostic information because of their advantage in manoeuvering tasks at a high-spatial resolutions and low costs in a user-friendly manner. In this study, an advanced application method of an UAS remote sensing system was performed using the grid GRAMI-rice model such that it can be driven using weather and remote sensing data to monitor the spatiotemporal productivities of rice (Oryza sativa). Remotely sensed data for the model were supplied, along with normalized difference vegetation index images obtained using the UAS remote sensing system. The model was first evaluated using paddy data from experimental fields (treated with two nitrogen (N) applications) at Chonnam National University, Gwangju, Republic of Korea (ROK). Practical application was then performed using paddy data from farm fields under conventional farm management practices at the Gimje plain in ROK. The grid GRAMI-rice model statistically well reproduces the field conditions of spatiotemporal rice productivities, showing an acceptable statistical accuracy in the comparison of growth between the simulated and observed values, using a Nash-Sutcliffe efficiency range of 0.113-0.955. According to t-tests (α = 0.05), there were no significant differences between the simulated and observed grain yields from both the evaluation and practical applications. The scientific approach adopted here is unique, advanced, and practical, in a way that UAS remote sensing methods were effectively incorporated with crop modelling techniques. Therefore, it was concluded that the UAS-based remote sensing techniques proposed in this study could represent an innovative way of projecting reliable spatiotemporal crop productivities for precision agriculture.