【作者】 徐霞；

【作者基本信息】 新疆大学 ， 地图学与地理信息系统， 2009， 硕士

【摘要】 干旱是区域水分收支或供需不平衡形成的水分短缺现象。农业干旱是因土壤供水不足导致植被正常生长育受阻,是世界各地主要自然灾害之一。农业是我国的第一产业,农业干旱将对国民生产产生很大的影响。探索一些可行的方法监测旱情,对工农业生产和人民生活都有非常重要的意义。利用传统的气象数据统计监测旱情不具有及时行和精确性,遥感具有宏观、综合、动态和快速等特点,能及时、准确、全面地获得旱情信息,对促进农业生产、保障粮食安全和区域可持续发展具有重要的现实意义。植被指数(NDVI)和地表温度(LST)两个参数可以通过表征绿色植被对干旱胁迫生境的反应揭示土壤水分信息,反映植被受旱状况,但两者单独使用时均存在局限性。基于植被指数和地表温度的二维特征空间构建的TVDI指数综合了两个参数特有的生理生态意义,不仅可以指示植被受旱时的水热胁迫环境,同时揭示了植被在这种胁迫环境下表现出的症状,可有效提高干旱监测的精度和效率。新疆地处干旱、半干旱气候区,干旱、半干旱气候加剧了盐渍化的发展,阻碍了农业的发展。因此,农业干旱监测成为急待解决的重要研究课题之一。本文首先以热红外遥感理论为基础,利用ETM影像结合气象数据获取地表温度反演关键参数:亮度温度、地表比辐射率、大气透射率、大气平均作用温度,并应用覃志豪的单窗算法反演研究区地表温度。然后,以植被指数-地表温度特征空间原理为指导,建立研究区LST-NDVI特征空间,拟合特征空间干湿边,建立TVDI指数,获取研究区旱情分布图,并分等定级,揭示研究区旱情分布。接着,利用最大似然法对研究区的影像进行分类,提取土地利用类型信息。在土地利用/土地覆盖的时空动态数据库和旱情监测的基础上,分析TVDI指数与地表温度和植被指数的关系,探讨TVDI指数与土地利用类型的相关性。研究表明,于田地区的地表温度低值区主要集中在水体,高值区集中在戈壁和沙漠,以绿洲为中心从里到外逐步升高。旱情也是由绿洲中心向绿洲外围,旱情逐步严重,靠近水源的地方以及植被覆盖良好的地方旱情较轻,干旱和重旱都分布在戈壁和沙漠地带;2002年10月3日的旱情比1999年9月13日严重。分析TVDI指数与地表温度和植被指数的关系发现,旱情指数与地表温度的相关系数高达0.957,而旱情指数与植被指数的相关系数仅达-0.636,植被指数反映旱情的能力相当有限,依据LST来预测旱情存在可取性。通过分析不同土地利用类型旱情分布发现,耕地、林地分布与TVDI值较低的区域,对地表水分状况要求高,高草的TVDI值分布较广,中草和低草分布与干旱和重旱区域,对水分要求不高。对地物不同时间的TVDI分布分析可得,在2002年10月,所有土地利用类型的旱情都加重了。从土地利用类型对TVDI的影响来看,耕地和林地在湿润和极干旱区域水分散失较强,在干旱区域保水能力较强;高草在干旱和重旱有很强的保水性;中草和低草在重旱区域,具有强保持水能力,且低草保持水的能力比中草强。更多还原

【Abstract】 Drought is the water shortage which is caused by a regional imbalance in water supply and demand. Agricultural drought is the result of inadequate soil water supply led to normal growth and fertility of vegetation ,which has been held all over the world, one of the major natural disasters. Agriculture is the primary industry of our country; agriculture drought will have a huge impact on national product. So it is more significative to explore some feasible methods to monitor drought. It is very important to Industrial production and agricultural production and people’s lives. The traditional monitor drought method of using Meteorological data is not timely and accurate. Remote sensing with macro, integrated, dynamic and rapid, can timely, accurately and comprehensively acquire the drought information. it has important practical significance to the promotion of agricultural production, food security and sustainable development in the region .Vegetation Index (NDVI) and land surface temperature (LST) Can reveal soil moisture information and reflect the state of vegetation affected by the drought through response ,which vegetation is stressed by the drought .when they are used alone ,there are both limited. TVDI index composed of two-dimensional feature space Based on vegetation index and land surface temperature is unique combination of two parameters of the physiological and ecological significance. It is not only indicative of environment when vegetation is stressed by water and heat, but also reveals the symptoms in which vegetation is coercive environment. TVDI index can effectively improve the accuracy of drought monitoring and efficiency. Xinjiang is located in arid, semi-arid climate zones, it is climate exacerbated salinization of development and hampered the development of agriculture. Therefore, agricultural drought monitoring has become one of the important topics, which need to be solved.In the first, This article based in thermal infrared remote sensing theory use ETM images combined with meteorological data to obtain the key parameters: brightness temperature, surface emissivity, atmospheric transmission rate, the average atmospheric temperature, and used the mono--window algorithm of Qin zhi hao to retrieve land surface temperature in YU TIAN .Secondly, the author uses the vegetation index - land surface temperature feature space principle as a guide to establish LST-NDVI feature space of the study area, fit the wet edge and dry edge of feature space to build TVDI index, obtain drought distribution picture of the study area and grade to reveal drought distribution information. Then, ETM+ images are selected to retrieve land use / land cover types by maximum likelihood method. finally, we analysis the relationships during TVDI index, vegetation index and land surface temperature and Discuss the relevance between TVDI index and land use / land cover types on the basis of the land use / land cover database and drought monitoring.The conclusions are as follows: the low-value of the land surface temperature in yu tian mainly concentrated in the water, the high-value concentrated in the Gobi and desert. .Land Surface temperature with the oasis for central from the inside to the outside gradually increased. The drought is also progressive severe from the central of the oasis to outlying oasis. The place near the water and with good vegetation is fewer droughts; drought and severe drought are located in the Gobi and desert; the drought in October 3, 2002 is more serious than September 13, 1999. We found that the correlation coefficient between TVDI index and land surface temperature is as high as 0.957, but the correlation coefficient between TVDI index and vegetation index is not high, when Analysising the relationship during TVDI index, land surface temperature, vegetation index. That the vegetation index reflects capacity of the drought is very limited. So it is feasible that land surface temperature predict drought. By analyzing the drought distribution of the different land-use types, we found arable land, woodland is distributed in the lower TVDI value region and need high surface moisture; the TVDI value distribution of high grass is much wider; the distribution of grass and low grass is in drought and severe drought region and do not ask for much of the water. Through the distribution analysis of TVDI of different times, ravages of a drought of all land-use types became severer in October, 2002. Conclusions can be drawn from the impact of land-use types on TVDI that: moisture loss of arable land and woodland is serious in the humid and serious arid regions, whereas it’s contrary in arid regions; water-holding capacity of High-density grassland is very strong in arid and serious drought areas; Middle-density and Low-density grassland have a strong capacity to maintain water in serious drought area, and the capacity of Low-density grassland is better.更多还原