Monitoring Of Heavy Metal Stress In Rice Based On Remote Sensing Evaluation Of Physiological Factors And Physiological Functions

Remote sensing technology has been one of the most important methods for motoring heavy metal stress in farmland owing to its fast and continuous dynamic monitoring characteristics.Most studies focus on the unilateral effects of heavy metal stress on physiological elements or physiological functions of crops,but there is no research on monitoring of heavy metal stress of crops in coordination with physiological elements and physiological functions.From the perspective of physiological factors,heavy metal stress can reduce the chlorophyll content of crops and lead to changes in crop spectral characteristics.From the perspective of physiological function,heavy metal stress can affect photosynthesis and transpiration of crops.However,most studies focus on physiological factors or physiological functions of crops heavy metal stress,there is no research on monitoring heavy metal stress of crops in combination with physiological factors and physiological functions.The study area was selected in Zhuzhou city,Hunan province.The soil heavy metal content data,ground hyperspectral data and heavy metal content of rice were collected.The cloud-free Sentinel-2 images during the whole growth period of rice in2017 and Landsat-8 images in 2014-2017 were obtained.The study aim to monitor heavy metal stress in coordination with physiological factors and physiological functions of crops,and the research contents and main conclusions of this study are as follows:(1)From the point of view of physiological factors,a novel vegetation index called heavy metal stress sensitive index(HMSSI)utilizing red-edge bands of Sentinel-2 images was proposed.The discrimination accuracy of heavy metal stress at different growth stages of rice was compared between HMSSI and existing indices.The multitemporal monitoring model based on random forest algorithm and multitemporal HMSSI images were constructed.Results show that the performance of a multitemporal monitoring model combining the whole growth stage images was better than any other single growth stage in distinguishing heavy metal stress levels.(2)From the point of view of physiological functions,the difference of LUE under different heavy metal stress levels was estimated based on improved LUE calculation method in the LUE models for three consecutive years in the rice growing period.The relationships between LUE indices(i.e.,SIPI and RGR)and LUE were investigated.Results showed that LUE could be used as an indicator to distinguish severe and mild heavy metal stress levels,and SIPI and RGR were significantly correlated with LUE.(3)From the point of view of synergistic physiological factors and physiological functions,a two-dimensional characteristic space named HMSSI_LUE was established to distinguish the stress level of heavy metals based on HMSSI and LUE.The results showed that HMSSI_LUE could make up for the deficiency that LUE could not distinguish moderate heavy metal stress levels and improve the discrimination accuracy in monitoring heavy metal stress.