冬小麦植被指数（NDVI）变化规律及其与土壤水分相关性研究

发布时间：2018-08-25 17:24

【摘要】：绿色植物对红光波段具有很强的吸收性,对近红外波段具有很强的反射性,这两种波段进行归一化差值运算可以得到归一化植被指数(NDVI),归一化植被指数对植物的生长状况及生理活性变化的反映具有很强的敏感性,利用NDVI监测冬小麦生长状况以及农田墒情,指导农业生产,对现代化冬小麦田间管理技术的发展具有重大的意义。本次研究于西北农林科技大学旱区农业节水重点实验室灌溉站内进行了大田冬小麦实验,实验灌水量60 L/m2(W1)、50 L/m2(W2)和30 L/m2(W3),0 L/m2(W4)共4个水平。分别在越冬期,拔节期,灌浆期进行一次灌溉。采用播种量控制密度,高密度(D1)播种量150kg//hm2,低密度播种量(D2)75kg/hm2。在冬小麦越冬期、拔节期、灌浆期、成熟期分别采集NDVI,田间气象数据,土壤水分数据。分析冬小麦NDVI日变化规律及其与气象因素及土壤水分的相关关系,并利用多元回归及逐步回归两种方法建立模型,为冬小麦的大田信息监测提供理论和技术支持。本论文分别分析了冬小麦NDVI日变化规律、NDVI与气象因素相关关系,以及NDVI与土壤水分的相关关系,并得出以下结论。(1)研究分析了不同生育期内冬小麦的NDVI日变化规律。得出NDVI与单日内时间序列呈现较好的二次抛物线性的相关关系,在冬小麦越冬期,NDVI拟合结果的决定系数为0.9、0.934,拔节期的决定系数为0.954、0.887,灌浆期决定系数都为0.829。但在冬小麦越冬期的NDVI总体变化幅度不大;灌浆期NDVI波动较大,是由于这一生育期内,天气变化异常,导致冬小麦NDVI波动较大。经实测数据验证得出,模型精度较好。针对冬小麦不同生育期NDVI与时间序列的二次抛物线反演模型,可为冬小麦NDVI快速获取技术提供理论支持。(2)研究分析了不同生育期内NDVI与气象因素相关性。得出冬小麦越冬期NDVI与气温以及相对湿度的相关性最高,拔节期NDVI与气温及地温的相关性最高,灌浆期地温、相对湿度与冬小麦NDVI相关性最好。对于地温,气温,相对湿度这些气象因素,冬小麦NDVI与之都存在一定的相关性,不同的气象因素与冬小麦不同生育NDVI间相关性存在一定差异。相关模型可以修正冬小麦在异常天气状况条件下的NDVI。(3)研究分析了对不同生育期冬小麦NDVI与土壤水分相关关系,并且分别采用多元回归法及逐步回归法进行模拟。结果表明,不同生育期冬小麦NDVI与不同深度的土壤水分通过差值处理后存在一定的相关关系,在越冬期,NDVI差值与20cm及60cm深度土层的土壤含水量差值相关性较好。在拔节期NDVI差值与表层10cm处以及40cm深度土壤水分差值相关性较好。在灌浆期NDVI与土壤水分的相关性与拔节期较为相似,同样也是10cm深度以及60cm深度的土壤含水量差值与NDVI差值相关性较好。三个不同生育期冬小麦都表现出表层10cm处以及60cm深度下的土壤水分差值与NDVI差值有较好的相关性。可见NDVI只适用于反演这两种深度下的土壤含水量。有关结论为NDVI反演冬小麦土壤水分提供参考。
[Abstract]:Green plants have strong absorption to red light band and strong reflectivity to near infrared band. Normalized vegetation index (NDVI) can be obtained by normalized difference calculation between the two bands. Normalized vegetation index has strong sensitivity to plant growth status and physiological activity changes. NDVI is used to monitor winter small. It is of great significance for the development of modern winter wheat field management technology to guide the agricultural production by the growth of wheat and the soil moisture in the field. 4) A total of 4 levels were conducted at overwintering stage, jointing stage and filling stage. The density was controlled by seeding rate, the density (D1) was 150 kg / / hm2, and the low density (D2) was 75 kg / hm2. NDVI, field meteorological data and soil moisture data were collected at overwintering stage, jointing stage, filling stage and maturity stage of winter wheat. The relationship between NDVI and meteorological factors and soil moisture was studied. The models were established by multiple regression and stepwise regression to provide theoretical and technical support for field information monitoring of winter wheat. The results are as follows. (1) The daily variation of NDVI of Winter Wheat in different growth stages is studied and analyzed. The results show that NDVI has a good quadratic parabolic linear correlation with the time series in one day. In winter wheat, the determination coefficient of NDVI fitting results is 0.9, 0.934, the determination coefficient of jointing stage is 0.954, 0.887, and the filling stage is decided. The definite coefficients are all 0.829. But the NDVI of Winter Wheat in overwintering period has little change, and the NDVI of grain filling period fluctuates greatly, which is due to the abnormal weather changes during this growing period, resulting in the great fluctuation of NDVI of winter wheat. The model can provide theoretical support for rapid NDVI acquisition of winter wheat. (2) The correlation between NDVI and meteorological factors in different growth stages was studied and analyzed. The correlation between NDVI and ground temperature, air temperature and relative humidity is the best. There is a certain correlation between NDVI and these meteorological factors, and there is a certain difference between different meteorological factors and different growth and development of winter wheat. The results showed that there was a certain correlation between NDVI and soil moisture in different growth stages and different depths. In overwintering stage, the difference of NDVI and soil moisture in depth of 20 cm and 60 cm was different. The correlation between NDVI and soil moisture at 10 cm and 40 cm depth at jointing stage was better than that at jointing stage. The correlation between NDVI and soil moisture at filling stage was similar to that at jointing stage. The results showed that there was a good correlation between NDVI and soil moisture difference at the depth of 60 cm and 10 cm. NDVI was only suitable for inversion of soil moisture at these two depths.
【学位授予单位】：西北农林科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：S512.11;S152.7

【参考文献】