Study On The Variation Of Crop Normalized Difference Vegetation Index

NDVI is one of the most widely used vegetation indexes.It is closely related to crop growth state and sensitive to the growth of crop,meteorological factors and soil moisture change.The monitoring of crop growth process by NDVI can provide basis for accurate crop management and technical support for crop field management.In this paper,winter wheat was taken as the research object,and the experiment was carried out in the water-saving irrigation experimental station of Northwest A&F University.Four different irrigation levels were set in the experiment.During the reviving,jointing and heading stages,the NDVI data of winter wheat were collected by Greenseeker handheld spectrometer after irrigation.Then we analyzed the daily variation of NDVI in winter wheat,and the quadratic polynomial regression,Gauss function and Sine function were used to fit the daily variation curve of normalized NDVI.The relationship between NDVI and meteorological factors was studied,and four methods,stepwise regression,principal component regression,partial least squares regression and ridge regression were used to establish the prediction model of NDVI for winter wheat.The quantitative relationship between NDVI and soil moisture content was studied,and then we established the linear regression model to analyze the relationship between NDVI and soil moisture content.The following conclusions are obtained:(1)The study constructed a diurnal variation model of NDVI of winter wheat.The study proved that the winter wheat canopy NDVI values are dynamic in different periods of a day.The NDVI data demonstrated clear parabolic shaped diurnal variations.It decreased gradually from 8:00AM,and reached to its minimum at 13:00PM or 14:00PM followed by a rapid increase in the afternoon.In order to describe the variations of the daily NDVI values,we used the quadratic polynomial regression,Gauss function and Sine function to fit the normalized NDVI daily variation curve respectively.Before fitting,a normalization processing was made to limit the data in the same range,which was convenient for the compare of different models.In significance test,the selected models were all statistically significant(P<0.01)in the three growing stages of winter wheat.And the three models have the best fitting effect in the jointing period with the coefficient of determination(R~2)all above 0.9.But the quadratic polynomial model expressed better stability compared to the other two models.Then the predicted and measured values were compared and the best fitting model was found by root mean square error(RMSE)and mean absolute error(MAE).Results showed that all three models had good fitting effects,however The prediction precision of quadratic polynomial model was the best.It can provide a reference for the establishment of NDVI diurnal variation model in the future.(2)The relationship model between NDVI and meteorological factors was established.By calculating the correlation coefficient between NDVI and meteorological factors,we can see that the tmperature,relative humidity,ground temperature,solar radiation and wind speed all have a certain correlation with NDVI of winter wheat,but wind speed has the least correlation with NDVI.Relative humidity was positively correlated with NDVI,while the other four meteorological factors were negatively correlated with NDVI.Using multiple stepwise stepwise regression,principal component regression and partial least squares regression and ridge regression to establish various models of meteorological factors and winter wheat NDVI,the results show tha the best fitting effect was the multiple stepwise regression model,the coefficient of determination(R~2)in three reproductive periods were0.613,0.827 and 0.4,the worst model was the principal component regression model,the coefficient of determination(R~2)were 0.448,0.744 and 0.394.(3)The relationship model between NDVI and soil moisture content was established.The difference between the NDVI and soil moisture content of winter wheat and the reference plot data was processed,then we can obtain CNDVI data and CSM data,and then the linear regression equation of the two variables was established.In the reviving stage and jointing stage of winter wheat,the correlation between NDVI and soil moisture content was the best at 10:00,but the correlation was better at heading stage at 14:00.Different irrigation treatments had different effects on NDVI of winter wheat,the influence of water on NDVI enhanced as the amount of water increased.At reviving stage,the correlation between winter wheat NDVI and soil moisture content was better at the depth of 30cm-40cm,and the correlation with 60cm depth of soil moisture content was relatively low.And in jointing stage and heading stage,the relationship between deep soil layer soil moisture content and winter wheat NDVI significantly increased because of exuberant growth of winter wheat in these two periods.