Study On Monitoring And Evaluating The Chilling Injury Of Rice In Northeast China By Remote-sensing And Crop Model

Some studies have stated that chilling injury is still one of the major agro-meteorological disasters affecting the growth, development and yield of paddy rice in northeast China. Global climate warming has become a consensus,in this context, the delayed-type chilling injury and sterile-type chilling injury showed a downward trend during the period of 2000-2010 in northeast China. Nevertheles, the risk of chilling injury to rice production still exists with the change of planting pattern, extreme low temperature events also happen now and then. Therefore, it is necessary to study how to quickly and effectively monitor the occurrence of chilling injury and how to evaluate the effect of chilling injury on rice yield in northeast China. In this paper, two aspects of remote sensing monitoring and model simulation are used to study the chilling injury monitoring and yield loss assessment. First, the spectral characteristics of paddy field under the influence of low temperature were analyzed, and the CVI index variable used to characterize the delay of phenological period was constructed. MODIS spectral reflectance products from 2001 to 2015 were used to extract the rice planting area in northeast China and identify the key phenological stages of rice. Remote sensing monitoring index system based on CVI index was constructed to realize remote sensing monitoring of delayed-type chilling injury. Second, based on the ORYZA2000 crop model, the rice chilling injury module and the yield components simulation module (including the tiller dynamics, the rice sterile, the blighted grain rate and the 1000-grain weight) were coupled. Meanwhile,the rice seeding experimental data were used to calibrate the ORYZA2000. Comsequently, the monitoring and yield loss assessment of chilling injury were examined based on the meteorological data and phenological data from 34 agro-meteorological stations in northeast China during the period of 2001-2014.The summarized results and conclusions are listed as follows:1. The EVI index increased gradually from transplanting stage to flowering stage(TF stages). The largest EVI appeared at flowering stage. The EVI index decreased gradually after flowering stage. During TF stages, low temperature blocked rice development process, as a result, the EVI index growth rate decreased and flowering stage was delayed. During flowering stage to mature stage (FM stages), low temperature blocked the natural decay of rice, as a result, the EVI index decline rate decreased and mature stage was delayed. In summary, the rate of EVI index change can reflect whether the phenological stage of rice is ahead or delayed.2. The CVI index was used to describe the average rate of change of EVI index.On the basis of multi-temporal MODIS remote-sensing data, the transplanting stage and flowering stage of rice in northeast China were identified and the CVI index of TF stages was calculated. The statistical results of CVI index range corresponding to different length of TF stages showed that CVI index decreases with the increase lenght of rice phenological stage. The smaller the CVI index, the greater probability of TF stages delayed. Different length of TF stages have different CVI probability distribution, that is to say, the same one CVI index could correspond to several length of TF stages.3. The results showed that the probability of sever delayed-type chilling injury in the northeastern part of the Sanjiang Plain, the eastern part of the Songnen Plain, the southwest of the Liaohe Plain and the Changbai Mountain was larger than that of the mild and moderate delayed-type chilling injury. The frequency of delayed-type chilling injury in the middle reaches of the Songhua River, the southwest of the Songnen Plain, the western Songhua River and the middle and northern part of the Liaohe River Plain was small. Large-range distribution of delayed-type chilling injury happened in 2003,2006 and 2007 in northeast China. The proportion of delayed-type chilling injury showed a decreasing trend after 2009.4. To improve simulation of rice phenological stages, the following methods were used: 1) The Beta equation was used to calculate the daily effective heat units for phenological development as the temperature response function. 2) Different three"cardinal" temperatures were used for different phenological stages of rice. 3) The parameters of ORYZA2000 were optimized by Shuffled Complex Evolution.ORYZA2000 coupled with chilling injury monitoring module and yield components module, so that crop model is able to assess the impact of cold damage on rice yield.5. The remote sensing monitoring indicators were established based on analyzing characteristics of vegetation index with the remote-sensing identification of rice growth stages and spectral observation data of paddies. Using the indicators could avoid the uncertainty caused by remote sensing inversion of surface temperature and then applied meteorological monitoring indicators. The model project not only combines improved ORYZA2000 model and the meteorological indicators, but also combines yield components module and chilling injury monitoring module. The results were conducive to evaluate the chilling injury influence.