Monitoring Of Wheat Powdery Mildew By Using Remote Sensing And Spatiotemporal Dynamics Of Airborne Conidia Of Blumeria Graminis F. Sp. Tritici

Wheat powdery mildew is a destructive foliar disease of world-wide distributed in all major wheat producing countries. In this study, hyperspectral canopy reflectance and remote sensing of unmanned air vehicle (UAV) was used to detect wheat powdery mildew. also studied the loss ratio estimation of yield,1000-kernel weight and crude protein content caused by Blumeria graminis f. sp. tritici (Bgt). Dynamics in concentrations of Bgt conidia and its relationship to local weather conditions and disease index in wheat was analyzed, and disease prediction models based on airborne spores of Bgt or weather conditions was constructed. The main results were as follows:1. The occurrence of wheat powdery mildew caused significantly loss to wheat yield,1000-kernel weight and protein content. Loss ratio estimation models of yield, 1000-kernel weight and protein content losses were constructed based on disease index (DI) at the critical point (CP) of growth stages and area under disease progress curve (AUDPC),Every estimation model achieves a better predicted effect.2. Disease indexes of powdery mildew in flowering, early and late filling stages have significantly negative correlations with red edge slope (dkred) and the area of the red edge peak (SDr).All vegetations index of DVI, RVI, SAVI and NDVI were significantly negative correlated with disease indexes in all three growth stages. According to the results of correlation analyses, several spectrum parameters (NIR, dλred, SDr, DVI,RVI,NDVI & SAVI) were choosed to construct disease detection models at different growth stages, and estimation models of dλred, DVI, NDVI, SAVI achieve a better Predicted effect.dλred, SAVI in flowering stage and early filling stage, dλred, NDVI in late filling stage were used to construct the loss estimation models of wheat yield,1000-kernel weight and protein content, Models using dλred gave more universal predictions.3.Digital Images from 50 m,100 m,200 m,300 m and 400 m above the ground were acquired from UAV. Then the relationships between disease indexes of wheat powdery mildew and color features which extracted from digital images were analyzed. The results showed that correlations different between color features extracted from digital image 50m,100m,200m,300m and 400m above the ground and disease index, RGB values of the image have significant or extremely significant positive correlations with disease indexes, UAV digital image parameters under 400 is more suitable for predicting disease indexes. Disease indexes, yield,1000-kernel weight and protein content estimation models based on I were constructed after correlation analyses, This indicated that the use of digital image can estimate yield, 1000-kernel weight and protein content of wheat when powdery mildew occurred.4. Conidia of wheat powdery mildew in the air were trapped using volumetric spore samplers, The concentrations of wheat powdery mildew conidia in the air increased with the disease index and reached the top at filling stage, then they began to decline and finally disappeared. Time series analysis showed that the season’s data can be fitted with simple ARIMA (1,1,0) models. And one model was made based on the significant correlations between concentrations of wheat powdery mildew conidia in the air with temperature. Two sets of models were derived by inoculum only, and by both inoculum and weather variables to the disease index. Models using inoculum only gave more universal predictions than models using both inoculum and weather variables, It can be used for predicting disease indexes.