| The rain and heat in northeast China are the same period,and the land is fertile,which is a suitable planting area for spring maize.Spring maize is very sensitive to water in the whole growing period,and the spatial and temporal distribution of precipitation in Northeast China is not uniform,which leads to the spring maize is prone to drought.At present,the widely used sectional drought indexes based on crop development stage have limitations in practical application.Therefore,scientifically selecting the indexes sensitive to spring maize drought in northeast China,constructing drought grade index model,and determining the dynamic critical threshold of each drought grade are of great importance for improving the meteorological service quality of spring corn drought disaster in Northeast China.In this paper,spring maize in Northeast China was taken as the research object,a typical drought index was constructed based on remote sensing data,an optimal drought index model was screened,and a sample set of time series was constructed combined with actual disaster data,and then a dynamic threshold of spring maize drought grade was established and verified.Then the optimal drought index and the constructed dynamic threshold were used to analyze the temporal and spatial characteristics of the study area.The main results are as follows:(1)Drought indexes NDVI,NDWI and SIF constructed based on remote sensing data were used to analyze the accuracy and sensitivity of spring maize drought identification in Northeast China by using box line method.It was found that the accuracy of the three indexes in corn drought identification was more than80%,among which the accuracy of SIF index was the highest,reaching 89.27%.The monitoring accuracy of the three indexes for severe drought was higher than that of mild and moderate drought,reaching more than 90%,and the monitoring accuracy of SIF index for severe drought was more than 95%.From the perspective of different development stages of maize,the three indexes had the highest drought monitoring accuracy of more than 90% at seedling stage,and the worst at jointing-booting stage and filling-maturity stage.The drought monitoring accuracy of SIF index at the four development stages of maize was better than that of NDWI and NDVI.The sensitivity to maize drought identification of NDVI,NDWI and SIF index was different.The sensitivity of SIF index to maize drought identification was the highest,followed by NDWI index,and NDVI was slightly lower.In terms of drought level,the sensitivity of the three indexes to severe drought was higher than that of mild and moderate drought.(2)Based on the SIF data and the actual drought disaster data of spring corn in Northeast China from 2000 to 2020,the sample set of time series was constructed.Polynomial and Gaussian fitting methods were used to construct different drought grade curves of spring corn in Northeast China,and the optimal fitting model was obtained.The average of the fitting curves of adjacent drought grade was taken as the dynamic critical threshold of each drought grade in the whole growth period of corn.Independent samples and typical drought events were used to verify the dynamic critical thresholds.The results showed that the Gaussian fitting model was more effective than the polynomial fitting in describing the SIF values of the time series of different drought grades of spring maize in Northeast China.The dynamic thresholds of different drought grades of spring corn in Northeast China established in this paper could well reflect the drought situation of spring corn in Northeast China.The results of drought grade identification were fully consistent with the actual disaster grade in82.76% of cases,and basically consistent with the actual disaster grade in 91.03% of cases.The verification accuracy was high.The spatio-temporal verification results based on typical drought events also showed that the dynamic threshold of drought grade could better reflect the spatio-temporal evolution characteristics of spring maize drought and the occurrence and development dynamics of drought disaster in Northeast China.(3)SIF index and drought dynamic threshold were used to analyze the spatiotemporal characteristics of drought in the Northeast spring corn growing area from2000 to 2020.The results showed that the frequency of drought in the west of the Northeast spring corn growing area was higher than that in the east during the study period.The average drought-affected area ratio of different drought classes ranged from 9.54% to 20.39%,the area ratio of light drought was the largest(20.39%),and the area ratio of severe drought was the smallest.The average drought-affected area ratio at the development stage was 37.93% to 47.22%.The drought-affected area ratio at the seedling stage was the largest(47.22%),and the minimum was at the heading to flowering stage.The interannual growth rate of SIF value in Northeast China was0.0023/a,and the proportion of area with increasing trend was 95.38%,which indicated that the drought in Northeast China would be alleviated during 2000-2020. |
PDF Full Text Request