Assessing the reliability of RS data to predict crop disease
UK Partners: Prof Xiangming XU and Dr Bo LI, NIAB EMR
Chinese Partners: Dr Yilin ZHOU, Chinese Academy of Agricultural Sciences
Various remote sensing tools have been proposed to obtain accurate and efficient quantification/estimation of diseases and yield. To use remote sensing tools in commercial agriculture and policy-making requires a high degree of accuracy and consistency in predicting diseases and yield from remote sensing data.
Most published studies have focused on the feasibility (i.e. accuracy) of using remote sensing tools to quantify plant diseases and to predict yield. The prediction consistency (across multiple sites and time) has been, however, ignored so far. Unpublished data from the Chinese partner suggest large variability in the relationship of wheat diseases and yield with remote sensing data. If this is confirmed, it will raise an important question of whether a specific prediction model derived from a specific remote sensing tool can be applied to larger regions over time.
The project aims to assess the magnitude of variability in the disease and yield prediction models in relation to time (between and within year) and remote sensing tools (e.g. different cameras). The proposed work will use wheat as a model and combine published and unpublished data with new data to quantify the uncertainty in using remote sensing data to predict diseases and grain yield.