【作者】 薛辉；

【导师】 倪绍祥； 韦玉春；

【作者基本信息】 南京师范大学 ， 地图学与地理信息系统， 2006， 硕士

【摘要】 土壤水分是植物生长和发育的必要条件，也是指示土壤墒情、确定春耕生产时间及补偿灌溉管理和评价产量高低等的重要参数。对于东亚飞蝗来说，土壤含水量对飞蝗产卵地选择、蝗卵的发育和孵化等有直接影响；同时，还通过影响植被对飞蝗的蝗蝻和成虫取食、活动产生间接影响。 目前，热红外遥感监测土壤水分的方法主要有热红外法、热惯量法和温度植被指数法，它们都是基于温度与土壤含水量之间关系建立模型，而对热辐射的另一个重要参数——发射率与土壤含水量的关系的研究还非常少。 本文在辐射模型算法的基础上设计了一套简单实用的土壤发射率野外测量方法。用此方法在河北省黄骅市实地测量了不同类型土壤的发射光谱。对这些土壤光谱进行分析时发现，土壤表层覆盖严重干扰了土壤光谱信息。地表平坦、影响因素较少的自然裸地，在8-9.5μm波长范围内，土壤发射率随土壤含水量的增加而增大。 对比分析昆山和黄骅两类有机质含量不同的土壤的载玻片发射率光谱发现，有机质含量较低的土壤在reststrahlen双谷中长波谷(9-9.48μm)的发射与短波谷(8.13-8.43μm)非常接近，有机质含量高的土壤长波谷则明显要低于短波谷。同时，在8.19-8.60μm区间，有机质含量低的土壤发射率光谱增长迅速；而有机质含量高的土壤则增长缓慢。这三段正好对应于ASTER的第10、11和12波段。研究发现，有机质含量在2％以上的土壤样品，波段10／11都大于0.99，波段10／12都大于1.01。因此对于风干土壤而言，波段10／11＞0.99或波段10／12＞1.01可以作为区别有机质含量高低的预滤器。 通过分析野外实测土壤发射率的光谱特征参数与含水量的关系，实测光谱在各波段的发射率值的均方差，以及发射率值与土壤含水量之间的相关系数和诊断指数，发现，波长为9.076-9.368μm的土壤发射率对含水量较为敏感，较适于反演研究区土壤含水量。利用模拟计算获得的ASTER第12波段的发射率和12波段与14波段发射率的比值与土壤含水量做回归分析，发现，b12／b14模型的精度更高，分析认为，这是因为发射率比值数据能够部分地消除土壤温度对发射率的影响。最终，建立了最适合研究区的热红外发射率土壤含水量反演线性模型。更多还原

【Abstract】 Soil moisture content is indispensable factor for growing and development of plants, and a vital parameter which indicates soil drought, determines timing of spring cultivation, conduct the management of compensating irrigation and estimates crop yield. For Oriental migratory locust soil moisture content has direct impact to site selection for egg lay of the locusts, and their egg development and hatching. In addition, soil moisture content has indirect effects to the food taking and activities of both nymphs and adults of Oriental migratory locust.At present, the thermal IR remote sensing assisted methods for monitoring soil moisture content includes mainly thermal infrared method, thermal inertia method and temperature/vegetation index method. All these methods are models relying on analysis of the relationship between temperature difference and soil moisture content. Yet, as another vital parameter of thermal radiation, the relationship between emission rate and soil moisture content has been rarely studied.In this thesis, a suit of easy and useful way to measure the soil emissivity in the field was designed on the basis of radiant model calculation algorithm. By this approach, the emissive spectrum of different kinds of soil was measured in the field in Huanghua city in Hebei province. While analyzing these soil spectrums, it was found that the spectrum information is interfered seriously by the cover types on soil surface. The results show that for the flat and bare lands the soil emissivity in the 8-9.5μ m region increases gradually when soil moisture content increases.By comparing and analyzing the emissivity spectrums of two kinds of soil that contain different content of organic matter and collected from Huanghua city and Kunshan city, it was found that the emission of the long wavelength lobe of the quartz reststrahlen doublet of the soil, which contains a relatively small amount of organic matter, is approaching to those of shorter wavelength lobe. Moreover, in the range of 8.19-8.60μm, the emissive spectra of those soil containing a relatively small amount of organic matter increase rapidly, whereas those with large amount of更多还原