Based On The TVDI Model, The Drought Monitoring Of Key Wheat Phenology In Guanzhong Area Was Studied

Affected by global warming,drought has become one of the worldwide most extensive coverage、the highest frequency and the most serious influence natural disasters.Guanzhong area is located in China’s western and Shaanxi’s middle part,whose special climate,topography,soil,economic and other natural human environment caused frequent drought in this area.Guanzhong area is rich in winter wheat and summer maize,frequent drought is very detrimental to the growth of winter wheat and summer maize,especially in spring,winter wheat critical growth period of rainfall,to the region’s food production and economic development caused a huge threat.Therefore,it is necessary to carry out the drought monitoring work in the area,and to grasp the depth and breadth of the drought in time and provide the decision-making basis for the drought prevention and relief measures.Remote sensing technology is widely used in drought monitoring research because of its large scale of monitoring,data acquisition time and high spatial resolution,data accuracy and so on.Based on the analysis of the status quo of remote sensing and drought monitoring at home and abroad,this paper chooses the TVDI model with good drought monitoring effect to analyze the temporal and spatial changes of drought and the main influencing factors in key growth period of winter wheat in Guanzhong region from 2004 to 2013,Draw the following main conclusions:(1)The MODIS NDVI data of the study area were reconstructed by Savitzky-Golay filtering method,and the anomalous noise caused by atmospheric conditions and data transmission errors was removed.The filtered vegetation index curve is more smooth and more realistic on the basis of maintaining the original shape.The characteristics of the NDVI curve of winter wheat during the growth period of winter wheat are consistent with the growth and development of winter wheat.The filtered image And the outliers are also effectively corrected,so that the pixel value is closer to the true value of the ground,laying the foundation for further extraction of winter wheat area and phenology monitoring.(2)According to the characteristics of NDVI curve of four typical species types in the winter wheat growth period,the decision tree classification model was extracted from the winter wheat planting area in Guanzhong area.From the spatial position,the winter wheat was mainly distributed in Fuping County,Pucheng County,Weinan City,Huaxian County,Huaxian County,Xianyang County,dry county,martial arts county,Xingping City,Jingyang County,Baoji City,Fengxiang County,Qishan County,Meixian County and Fufeng County,Xian County,Chang’an County,Gaoling County,Yanliang District and Tongchuan City Yao County.On the whole,the winter wheat planting area in Guanzhong area was mainly concentrated in the central part of Guanzhong,and distributed from west to east.Comparing the results with the classification of TM image supervised classification and unsupervised classification,it is found that the winter wheat pixel extracted by the decision tree model is more accurate.Finally,the area of the extracted winter wheat is calculated and compared with the statistical data of the same year The accuracy of the extraction is 94.29%compared with the statistical yearbook.The results show that it is feasible to use the decision tree model to extract winter wheat in Guanzhong area.(3)The NDVI time series curve was fitted by using the double Logistic function to extract the key phenophase of winter wheat in Guanzhong area in 2013.It was found that the turn-back period began generally before day 57,before February 26,and the heading period generally began on the 105th day Day 112,that is,April 15 to April 22,mature period generally began in the first 145 days to 160 days,that is,May 25 to June 9.In the area earlier,there are districts in Huxian County,Wugong County,Xingping City,Sanyuan County and Jingyang County,which include Yongshou County,Chunhua County,Yao County and Baishui County.The earlier areas of heading period include the martial arts County,Huaxian County,Huayin County,Dali County,the time delay of the region Yongshou County,Chunhua County;mature period ahead of Zhou to County,Pucheng County,Huaxian County,Huayin County,Fengxiang County,Yongshou County,Chunhua County,the overall performance of the early Qingqing area,heading and maturity will follow the advance,and later into the back to the green period of the regional heading and maturity will be delayed.(4)The Ts-NDVI feature space constructed by winter wheat NDVI and surface temperature is triangular or trapezoidal.The highest surface temperature,ie,the dry surface of NDVI,increases first and then decreases,and the overall performance is decreasing.NDVI corresponds to The minimum surface temperature,ie,the wet side,decreases first and then increases,and the overall trend increases.The results show that the fitting effect of dry side is better than that of wet edge,and the average value of R2 reaches 0.73.From the fitting of each stage,the whole fitting effect of turning green and heading is more mature Well,TVDI is more suitable for monitoring the drought in the Guanzhong area.The correlation between TVDI and 10cm,20cm,50cm soil moisture was found.In the F test,the correlation coefficient between TVDI and 20cn soil moisture was the highest,and all of them were tested by significant difference(P<0.05).A = 0.05),followed by 10cm,and 50cm soil moisture correlation coefficient did not pass the significant test(a = 0.05),indicating that TVDI can reflect and indicate the surface 10～20cm soil moisture status.(5)The TVDI model was used to monitor the drought in the three key growth stages of winter wheat in Guanzhong area from 2004 to 2013.The drought was the most serious in the mature period in 2005 and the drought in 2012.TVDI showed a trend of"rising-falling-rising" in winter wheat in three key growth stages in ten years.Among them,TVDI showed a sharp upward trend from 2004 to 2005,and the trend of decreasing trend in 2005-2010 was 2010-2013 But also showed a sharp upward trend,the overall performance of the rising trend,but the increase is not large,indicating that the 10 years in the Guanzhong area,although the spring drought has increased,but the change is not obvious.TVDI showed the upward trend in the three key stages of winter wheat growth in winter wheat,and the most serious drought was the greening period,with the main drought area,which accounted for 59.58%of the winter wheat planting area.The drought was dominated by drought and drought,and the area was 54.56%and 65.21%respectively.The drought at all stages is presented as a spatial distribution of the north.