The Study On Temperature Inversion And Chilling Damage Of Corn Based On MODIS Data In Northeast China

Chilling damage is one of the main meteorological disasters of crops in Northeast China.As the main grain crops in Northeast China,the yield of corn seriously affects the development of animal husbandry and people's living standard.With global warming,the northern boundary of grain production is moving northerliness in Northeast China.And planting area and scale also enlarge gradually.Once the chilling damage occurred,there will be a sever effect on it.Using remote sensing data to monitor maize chilling damage is very rate so far.In this study,selecting data from MODIS about 2005-2014 of two satellite platform to establish the remote sensing estimation model of air temperature with high-quality LST.The MODIS sensor is used to integrate high quality LST data to the earth 4 times a day,gaining the temperature of remote sensing data in the whole day.By time interpolation of data to get continuous time series data.In order to obtain the missing data,spatial interpolation is needed.Finally,using the meteorology occupation standard for chilling damage reference.To determine the chilling damage years of corn in Northeast China based on the later spatial interpolation data.The main conclusions of this paper are as follows:(1)Observing the daily mean temperature as the independent variable,leading into land surface temperature(LST),longitude(LON),latitude(LAT),normalized differential vegetabtion index(NDVI),solar zenith angle(SAZ),altitude(ALT),daily ordinal(N)as the model factors,the selected factors were significantly correlated with the level of 0.01 Selecting the four high quality LST data as the data sourcees,adjusted R2 reached 0.632,0.824,0.53 and 0.706 by using the multivariate linear regression method to structure the temperature remote sensing model.Validating the model by datas of 2013-2014,most samples of estimation model fall near the line(1:1).(2)Considering the integrity of the data,integrate the 4 times observation data of the double star platform,according to the degree of adjusted R2,ccording to the rules to fusion that double star platform night data priority,data in the day second,the errors before and after fusion satisfy the normal distribution.Application of adjacent time temperature interpolation method based on time fusion of time series to the missing temperature.After the interpolation of time series,the amount of data is increased by more than l times,the average errors increase are less than 0.5 ?.(3)Adopting the active accumulated temperature of 10 ? temperature in growing season anomalies index(index 1)and the the average of the sum of the mean temperature anomalies index of May to September(index2)to discriminant maize chilling damage years in Northeast of China in 2005-2014.The two discriminant indices were similar to those of chilling injury years and sites,but also slightly different.There are all cold injury in 2005,2006,2009 and 2011..And most of the site have chilling damage in 2005,concentrated in Liaoning and Jilin.The years of chilling damage of monitoring results of index 2 were better than those of index 1,index 2 monitoring less sites of cold damage than index 1.(4)Applying the agrometeorological disaster data and meteorological station data to validate chilling damage index of remote sensing estimation.The two cold damage indexes calculated by the meteorological data are highly consistent with the year and the site of chilling injury,but there are differences in the classification of chilling injury;Validation of cold damage estimation by remote sensing using meteorological indexs,some of the years and sites showed inconsistent phenomenon;Using agrometeorological disaster data to validate the remote sensing estimation of chilling damage indexs and chilling damage indexs,part of the year and site consistent.